Contents

0.1 Space Transport and Engineering Methods: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

0.1.1 Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 Part 0 - Introduction 3
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.1 Purpose of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Links to Additional Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.3 Book Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Part 1 - Science and Engineering Fundamentals 7

2.1 Science and Engineering Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2 Mathematics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.3 Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.4 Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.5 Design Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Basic Sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1 Physics as a Subject . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.2 Units and Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.3 Motion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.4 Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.5 Continued on page 2 → . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 Basic Sciences 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.1 Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.2 Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3.3 Thermodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3.4 Continued on Page 3 → . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4 Basic Sciences 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4.1 Astronomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4.2 Planetary Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.4.3 Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.5 Orbital Mechanics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.5.2 Orbits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

i
ii CONTENTS

2.5.3 Velocity Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2.5.4 Powered Flight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6 Propulsive Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.6.1 Reaction from Expelled Mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.6.2 External Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.7 Energy Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.7.1 Energy in General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.7.2 Mechanical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.7.3 Chemical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.7.4 Thermal Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.7.5 Electrical Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.7.6 Beam Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.7.7 Nuclear Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.7.8 Matter Conversion Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.8 Systems Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.8.1 Systems Engineering In General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.8.2 The System Life Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.8.3 Requirements Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.8.4 Requirements Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.8.5 Continued on page 2 → . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.9 System Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.9.1 Functional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.9.2 Requirements Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
2.9.3 System Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.9.4 Optimization and Trade Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.9.5 Synthesis and Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2.10 Engineering Tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.10.1 Engineering Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.10.2 Computer Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.10.3 Computer Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
2.10.4 Instrumentation and Test Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.11 Engineering Specialties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.11.1 Aerospace Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.11.2 Other Engineering Specialties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.12 Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
2.12.1 Organizational Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
2.12.2 Project Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.12.3 Project Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
2.12.4 Financial Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
2.13 Existing Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
2.13.1 Government Programs - International . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
CONTENTS iii

2.13.2 Government Programs - United States . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

2.13.3 Continued on page 2 → . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
2.14 Existing Programs 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
2.14.1 Government Programs - United States (continued) . . . . . . . . . . . . . . . . . . . . 71
2.14.2 Commercial Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
2.14.3 Not-for-Profit Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
2.15 Future Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.15.1 Bulk Mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.15.2 Industrial Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
2.15.3 Manufactured Items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2.15.4 Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
2.15.5 Engineered Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
2.15.6 Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
2.15.7 Science and Exploration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
2.15.8 Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
2.15.9 Entertainment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

3 Part 2 - Space Transport Methods 84

3.1 Space Transport Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.1.1 The Space Transport Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.1.2 Known Transport Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.2 Structural Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.2.1 Structures in General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.2.2 A. Static Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.2.3 B. Dynamic Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.3 Guns and Accelerators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
3.4 Guns and Accelerators 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.4.1 C. Light Gas Guns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.4.2 D. Electric Accelerators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
3.5 Combustion Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
3.5.1 A. Air-Breathing Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
3.5.2 B. Internally Fueled Engines (Rockets) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
3.6 Thermal Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3.6.1 Electro-thermal Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
3.6.2 Photo-thermal Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
3.6.3 Nuclear Thermal Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
3.7 Bulk Matter Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
3.8 Ion and Plasma Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
3.8.1 Ion Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
3.8.2 Plasma Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
3.9 High Energy Particle Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
3.10 Photon Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
iv CONTENTS

3.11 External Interaction Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128

3.11.1 Magnetic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
3.11.2 Gravity Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
3.11.3 Aerodynamic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
3.11.4 Mechanical Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
3.12 Theoretical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
3.13 Comparisons Among Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.13.1 Comparisons Among Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.13.2 Transport Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

4 Part 3 - Space Engineering Methods 134

4.1 Space Engineering Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.1.2 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.2 Design Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2.1 Technology Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2.2 Availability Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2.3 Physical Design Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2.4 Integration Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4.2.5 Human Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4.2.6 Environment Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4.3 Subsystem Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
4.3.1 Elements by Subsystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
4.4 Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
4.4.1 History and Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
4.4.2 Progression by Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
4.4.3 Long Range Exploration (Astronomy) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
4.4.4 Medium Range Exploration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
4.4.5 Short Range Exploration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
4.5 Resources 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
4.5.1 Solar System Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
4.5.2 Solar System Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
4.5.3 Galactic Matter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
4.5.4 Galactic Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
4.6 Resource Extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
4.6.1 Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
4.6.2 Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
4.7 Processing and Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
4.7.1 Production Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
4.7.2 Handling and Storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
4.7.3 Materials Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
4.7.4 Parts Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
CONTENTS v

4.8 Assembly and Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

4.8.1 Assembly Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
4.8.2 Construction Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.8.3 Outfitting Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.9 Verification and Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.10 Operation and Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
4.10.1 Operations Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.10.2 Operations Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.10.3 Maintenance Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.10.4 Maintenance Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
4.11 Recycling Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

5 Part 4 - Combined Space Systems 159

5.1 Complex Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
5.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
5.1.2 Organization and Content of Part 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
5.2 Program Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
5.2.1 Program Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5.2.2 Program Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5.2.3 Program Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
5.3 Advanced Manufacturing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
5.3.1 R&D Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5.3.2 R&D Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5.3.3 R&D Sub-Phases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
5.4 Startup Launcher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
5.4.1 Industrial Scale Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.4.2 Industrial Scale Habitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.4.3 Industrial Scale Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.4.4 Existing Launchers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.4.5 Small Multistage Rocket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
5.4.6 Augmented Rocket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
5.5 Orbital Assembly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
5.6 Hypervelocity Launcher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
5.7 Low G Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
5.8 Electric Propulsion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
5.8.1 Near Term Electric Thruster Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
5.8.2 Electric vs Chemical Thrusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
5.8.3 Comparisons Between Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
5.8.4 Electric Propulsion Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
5.8.5 Moved Text to be Merged . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
5.9 Orbital Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
5.9.1 Rationale for Orbital Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
vi CONTENTS

5.9.2 Steps Towards Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197

5.10 Orbital Mining 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
5.10.1 NEO Ore Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
5.10.2 Example Mining Missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
5.10.3 Debris Belt Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
5.10.4 Small NEO Return Mission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
5.10.5 Bulk Regolith Return Mission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
5.10.6 Large NEO Return Mission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
5.10.7 Product Return Mission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.10.8 Mining Tug Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.11 Processing Factory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.11.1 Desired Factory Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
5.11.2 Factory Design and Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
5.11.3 Processing Factory Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
5.12 Space Elevator (Skyhook) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
5.12.1 [Text still to be integrated] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
5.13 Lunar Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
5.13.1 Lunar Region Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
5.13.2 Industry Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
5.13.3 Project Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
5.13.4 Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
5.13.5 Needed Program R&D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
5.14 Interplanetary Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
5.15 Mars Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
5.15.1 Early Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
5.15.2 Long Term Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
5.16 Later Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

6 Part 5 - Design Studies 234

6.1 Design Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.1.2 Full Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.1.3 Studies in Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.1.4 Study Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.1.5 Discarded Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
6.2 Human Expansion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
6.2.1 Conceptual Design Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
6.2.2 Program and Customer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
6.2.3 Program Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
6.2.4 Program Benefits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
6.3 Human Expansion 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
6.3.1 Selection Criteria Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
CONTENTS vii

6.3.2 Program Requirements Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240

6.3.3 Design Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
6.3.4 Initial Program Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
6.4 Human Expansion 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
6.4.1 Program Requirements Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
6.5 Human Expansion 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
6.5.1 Program Evaluation Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
6.6 Human Expansion 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6.6.1 Functional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6.6.2 Current Parameter Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6.6.3 Existing Baseline Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6.6.4 New Program Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
6.6.5 Initial Program Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
6.7 Environment Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
6.7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
6.7.2 Initial Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
6.7.3 Detailed Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
6.8 Open Source Space Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
6.8.1 Background of Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
6.8.2 Requirements Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264

7 References and Sources 266

7.1 References and Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
7.1.1 Current News . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
7.1.2 Multiple Documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
7.1.3 Book Length Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
7.1.4 Websites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
7.1.5 Articles and Technical Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
7.1.6 Discussion Forums . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
7.2 App 1 - Fictional Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
7.2.1 Fictional Methods Which Are Not Supported By Known Physics or Engineering . . . . . . 270
7.2.2 Fictional Methods Which Are Supported by Known Physics or Engineering . . . . . . . . . 270
7.3 App 2 - Reference Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
7.3.1 Aerodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
7.3.2 Nuclear Rockets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

8 Text and image sources, contributors, and licenses 275

8.1 Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
8.2 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
8.3 Content license . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
0.1. SPACE TRANSPORT AND ENGINEERING METHODS: 1

0.1 Space Transport and Engineer- 9. Existing Programs - Government || page 2: Gov-
ernment (cont.), Commercial, Not-for-Profit
ing Methods:
10. Future Projects - Transport, Exploration, Mining,
An Introduction to Space Systems Engineering Industrial Capacity, Manufactured Items, Energy,
Engineered Environments, Communication, Enter-
tainment
0.1.1 Table of Contents
Part 2: Space Transport Methods

1. Structural Methods - Static, Dynamic

2. Guns and Accelerators - Mechanical, Artillery ||

page 2: Light Gas, Electric

3. Combustion Engines - Air Breathing, Internally

Fueled (Rockets)

4. Thermal Engines - Electro-thermal, Photo-

thermal, Nuclear-thermal

5. Bulk Matter Engines - Rotary, Coilgun, Railgun

Photograph from the space shuttle Atlantis (STS-135) in its final 6. Ion and Plasma Engines - Ion, Arcjet, Plasma
fly-around of the International Space Station in July 19, 2011.
Shows the ISS, Earth and moon. 7. High Energy Particle Engines - Particle Rockets,
External Particle Interaction

8. Photon Engines - Photon Sails, Photon Rockets,

Part 0: Introduction
Photon Gun

9. External Interaction Methods - Magnetic, Grav-

ity, Aerodynamic, Mechanical
Part 1: Science and Engineering Fundamentals 10. Theoretical Methods - not currently supported by
established physics
1. Basic Sciences - Physics: Units, Motion, Forces
|| page 2: Energy, Mechanics, Thermodynamics || 11. Comparisons Among Methods - Performance,
page 3: Astronomy, Planetary Science, Chemistry Status, Cost
2. Orbital Mechanics - Orbits, Velocity Map, Pow-
ered Flight, Mass Ratio, Staging Part 3: Space Engineering Methods
3. Propulsive Forces - Reaction from Expelled Mate- 1. Design Factors - Technology, Availability, Physi-
rial, External Interaction cal, Integration, Human, Environment
4. Energy Sources - Mechanical, Chemical, Thermal, 2. Subsystem Design
Electrical, Beam, Nuclear, Matter Conversion
3. Resources - Exploration Methods || page 2 - Re-
5. Systems Engineering - In General, Life Cycle, Re- source Inventory
quirements || page 2: Functional Analysis, Allo-
cation, Modeling, Optimization and Trade Studies, 4. Resource Extraction - Mining: Solid, Atmo-
Synthesis, Work Breakdown Structure, Elements by spheres, Liquids, Gas Giants, Particulates, Stars ||
Type Energy: Solar, Fission, Fusion, Orbital

6. Engineering Tools - Data, Computer Hardware, 5. Processing and Production - Production Control,
Computer Software Handling and Storage, Materials Processing, Parts
Production
7. Engineering Specialties - Aerospace, Other Spe-
cialties 6. Assembly and Construction - Assembly, Con-
struction, Outfitting
8. Organization and Economics - Organization,
Funding, Financial Analysis 7. Verification and Test
2 CONTENTS

8. Operation and Maintenance - Operations Con- 2. Environment Ranges

cepts, Tasks, Maintenance Concepts, Tasks
3. Seed Factories - Conceptual designs for self-
9. Recycling Methods - Waste Recycling, Closed expanding automated factories (in separate wiki-
Loop Life Support book).
4. Open Source Space Program
Part 4: Complex Programs

1. Program Overview - Goals, Summary, Structure References and Sources

2. Phase 0 - Research & Development - Planning, 1. Appendix 1: Fictional Methods

Process, Sub-Phases
2. Appendix 2: Reference Data
3. Phases 1 to 3 For Earth - Locations: Starter, Net-
work, Distributed, Industrial, Difficult, Extreme page traffic for this month.
4. Phase 2B Industrial Locations For Space - Pro-
duction, Habitation, Transport & Services
5. Phase 4A: Low Orbit || page 2: Orbital Assembly
6. Hypervelocity Launcher - Scaling, Launcher De-
sign, Projectile Design
7. Low G Transport - Choosing among Alternatives,
Existing and In-Development Launchers, Alterna-
tive Launchers
8. Electric Propulsion - Electric vs Chemical,
Thruster Types, Comparison Between Types
9. Orbital Mining - Rationale, Mining Steps || page 2
- Ore Types, Example Missions, Design
10. Processing Factory - Outputs, Process Research
and Development, Factory Design
11. Spaceport Network - Concept, Applications, De-
sign Parameters, Design Components, Design Issues
12. Phase 5A: Lunar Locations - Concept, Starter Kit,
Expansion Kit, Bulk Delivery
13. Interplanetary Transfer - System Concept, High
Orbit Skyhook, Inclination Station, Transfer Habi-
tats
14. Mars Development - Near Term: Phobos and
Deimos, Mars Skyhooks, Mars Surface Systems ||
Long Term Development: Magnetosphere, Green-
houses, Full Atmosphere
15. Later Projects - Main Belt Asteroids, Outer Moons
and Minor Planets, Kuiper Belt, Gas Giants, Oort
Cloud and Beyond

Part 5: Design Studies

1. Conceptual Design for Human Expansion -

Goals and Benefits || page 2: Conceptual Design
Approach || page 3: Requirements Analysis || page
4: Evaluation Criteria || page 5: Functional Analy-
sis
Chapter 1

Part 0 - Introduction

1.1 Introduction all, and the design of relatively small satellites with finite
lives as cargo for such transportation. These topics are
adequately covered by existing textbooks such as Sutton’s
1.1.1 Purpose of the Book
Rocket Propulsion Elements and Griffen and French’s
Space Vehicle Design.
Starting with more recent projects, like the International
Space Station (ISS), new design characteristics have be-
come more important. These include permanent habi-
tation, evolution and growth, sustainability, and indus-
trial capacity. Future projects will require new types of
propulsion, life support, production, and assembly. Sin-
gle projects can no longer be considered in isolation, ei-
ther. For example, the ISS depends on multiple launch
vehicles to deliver crew, supplies, and hardware for ex-
pansion. This book is intended as an introduction to
both the historical design challenges of transport and
space hardware design, and the new engineering meth-
ods needed for future projects. The online format is also
intended to address some of the limitations of traditional
printed textbooks:

• Their high cost in recent years, which is addressed

by making this an open-source Wikibook,
Earth as seen by Apollo 17.
• The inability to include other media types, such as
As this photo of the Earth from a distance shows, our audio, video, simulations, and software.
planet is located in space. So space is not a different place
to go to: we are already there. We think of it as different • The relative difficulty of updating printed books in
because most of us spend most of our lives on about 1/8th a fast moving field.
of the solid-gas interface of one planet among the billions
in the Galaxy. It is where we evolved, and where condi- • The very large amount of information in this field,
tions are reasonably habitable for us, so we think of it as which is addressed by links and a supplementary li-
normal. In reality, our comfortable planet is the excep- brary.
tion, and normal is unusable environments hostile to life
• The lack of practice projects. Learning well involves
as we know it. We can change that by technical means,
more than just reading, it should include hands-on
and are starting to do so first by placing devices in space,
work. The latter part of the book includes real
then by creating habitable environments.
projects which interested readers or teams can work
As of 2014 global space industry had already reached on.
US$ 323 billion/year of economic activity. This is ex-
pected to greatly increase in the future as our civilization
learns to use this environment and expand into it. For 1.1.2 Links to Additional Material
these and other reasons space systems engineering is a
worthy field of study within the larger context of all sci- We cannot fit the entirety of a field of engineering into
ences and engineering. Historically the main engineering a book-length introduction such as this one. Therefore
challenges were transportation that could reach space at we will include extensive hypertext (World Wide Web)

3
4 CHAPTER 1. PART 0 - INTRODUCTION

links to more detailed sources, such as the Wikipedia on- Systems Engineering. The goal of systems engineering
line encyclopedia. A Rocket Science Library is being is to optimize a complex system over its life cycle. Most
set up to house some of the more useful publicly avail- space projects are complex enough that this discipline is
able articles and books to supplement the book. We can- very useful. Complexity demands methods of tracking
not include copyrighted works without permission, but the details to keep the project as a whole a coherent ef-
we can list useful ones and where to find them in both fort. We introduce the tools that engineers use in their
the References section of this book, and in notes in the work, followed by the variety of specialty engineering
Library folders. We are aware that popular culture in- disciplines such as structural and electrical design which
correctly calls Space Systems Engineering “Rocket Sci- are used in parallel with the overall system optimization.
ence”, as it is part of engineering rather than science and Most projects require specialists in different areas, be-
not exclusively about rockets. We will cheerfully adopt cause there is too wide an area of knowledge for one per-
the popular label for our Library and hope that serious son to cover all of it.
students understand the difference.
In early concept development one person might do all
Engineers use other tools besides books to do their work, of the work on a project. As the design progresses, the
such as simulation software and spreadsheets. Wiki- amount of detail and knowledge required makes it more
books does not support including software files, so links to efficient to use teams of specialists. This in turn demands
where these items can be found are intended to be added efficient coordination within a project team. The gen-
over time. A draft paper summarizing a real proposed eral topic of project economics addresses project organi-
program can be found at: To Mars and Beyond. This zation and management, funding, and financial analysis.
program will be used for examples within the book, and The last two sections of Part 1 consider existing projects
can be used by readers for practice in design. and the areas for future projects. One needs to compare
proposed future projects to already existing ones, to see
if they give enough improvement to justify their creation.
1.1.3 Book Organization The range of possibilities for future projects helps iden-
tify goals, requirements, and growth paths.
The book is organized as a progression from science
and engineering fundamentals, to subsystems, complete
projects, and then combined systems which have linked Part 2: Transport methods
projects. We will try to emphasize underlying princi-
ples and key design parameters, as those have more last- The next part of the book considers transportation. Be-
ing value than, for example, the particulars of a current fore you can do anything else at a given space location,
launch vehicle. We include many methods and concepts you first have to get there. Historically the transportation
which are not in current use for several reasons. First, component has been much more massive than the cargo.
when starting a new project, it is a good idea to at least Solving transportation challenges became the primary fo-
briefly consider all the possible alternatives, before nar- cus of space projects. Although still important, we take
rowing down to the most relevant ones. Second, this is a a more balanced approach in this book. Improved trans-
future-oriented book. While a concept may not be useful portation methods and use of local resources will shift the
today, knowing what the state of the art is and what tech- mass ratio towards the destination, so we give that part of
nology areas to watch helps to tell when a concept will space projects equal attention.
become useful. Third, having all the ideas in front of you
Many more transport methods have been proposed (about
might spur a new combination or application that had not
75) than have actually been used to date (about 5), and
been thought of before. This has actually happened to
one of them, chemical rockets, has been used by far the
one of the authors (Eder) on more than one occasion, so
most. Part 2 lists the many known and speculative space
I can attest to its usefulness.
propulsion methods by category. After listing the avail-
able concepts, the later sections will make some compar-
Part 1: Science and Engineering Fundamentals isons, and discuss how to select the best candidates for a
given project.
All of engineering depends on underlying knowledge Part of the reason chemical rockets have been used so
from mathematics and the sciences. We therefore start much is “first mover advantage”. They were the first type
with some prerequisites which the reader should be fa- of space propulsion to get extensive development. They
miliar with, and the more important principles, formu- have the longest history, most optimization of design, and
las, and methods which apply to space systems. The most familiarity, so they continue to be used. That does
largest part of this is physics, including the motive forces not mean they are objectively the best solution for all
and energy sources for space projects, but also some key time, and all circumstances. Use of alternative meth-
ideas of the other sciences, including Astronomy, Plane- ods, such as ion engines, is becoming more common in
tary Science, and Chemistry. Following that we introduce recent years. When starting a new project, a survey of
general engineering methods, including the discipline of all possible technical choices is worth doing, before nar-
1.1. INTRODUCTION 5

rowing the list down to the best candidates, so that no can do nothing without raw materials to work on. Com-
good option is overlooked. That is one reason the list in bining these, a vehicle can transport raw materials to the
this book attempts to be comprehensive. Additionally, plant, and get fuel from the plant for later trips. Then
no single method is optimal for all locations and mission the combined system can operate indefinitely as long as a
requirements. surplus of fuel is produced from each trip. So to design a
complete program, you need to look at all the component
systems and how they interact.
Part 3: Engineering Methods
In addition to the parts of a program interacting at a given
point in time, it can also evolve over time. An analogy can
Part 3 considers what to do when you reach a location
be made to road systems on Earth. They started as foot-
in space, and how to do it. It starts by listing overall
paths, and were upgraded over time. The upgrades were
design factors which affect multiple parts of a project.
not independent of the types of vehicles that used them,
Having accounted for propulsion in Part 2, we then ex-
and the existing roads were used to deliver machinery and
plore the other subsystems that make up a complete sys-
supplies to construct the upgraded roads. So it will be
tem item. A large program or project will consist of mul-
with sensibly designed large space projects - using one
tiple system items with different purposes and functions.
set of vehicles and facilities to help build the next level of
The remainder of Part 3 covers the engineering methods
improvements.
that are specific to these functions. This includes obtain-
ing resources, turning those resources into useful materi- We give an extensive example of a complex program that
als and parts, assembly and construction, verification and both interacts with itself and grows over time. This ex-
test, operation and maintenance, and finally recycling and ample serves several purposes. One is to demonstrate
disposal. A particular method describes in general how a by example how such a program is analyzed and defined.
given task is done. Once completed, an engineering de- Where alternative options are presented, we explain how
sign then implements that method in a specific device. to choose among them. Another purpose is to serve as
an actual proposed plan to implement in real life. In that
The approach to date for most space projects has been
context you have to go past pure design questions and con-
to do all the design and construction on Earth, and then
sider cost and other external factors that apply to a real
launch pre-built and pre-supplied hardware such as com-
project. The final purpose is as a “class project” or “lab
munication satellites or interplanetary probes as complete
experiment” method of learning. Things like working in a
items. That is adequate for smaller projects, but as larger
team are best learned by doing, and using new knowledge
ones are contemplated this method becomes unreason-
in a project helps fix it in memory. Taking an unfinished
ably difficult and expensive. The International Space Sta-
part of our example to the next level of design detail can
tion is an example where it was too large to launch as one
be used for this purpose. Experience working on a real-
complete item, with supplies for it’s full operating life.
istic design can be useful if someone intends to work in
Instead it was assembled from smaller parts that each fit
this field in the future.
within a single rocket launch, and supplies are delivered
periodically. This reduced the payload per launch to a Our proposed program is likely not the best one imagin-
manageable size, but all the parts and supplies are still able, but it is intended as a good starting point. As indi-
being brought from Earth. viduals and teams make suggestions and apply engineer-
ing effort to it we hope it will improve over time, both as
As larger and more distant projects are considered, the
an actual project proposal, and as a teaching tool.
transport requirements continue to go up, and this starts
to become very expensive. This is mainly due to the deep
gravity well of the Earth, which requires a lot of energy to
climb out of. At some point it becomes more economical
for a project to extract supplies and energy, and eventually
do production, locally. So we give a lot of attention to
these tasks. Although doing them in space is new, there Part 5: Design Studies
is a long history of engineering on Earth which can serve
as a starting point.
In this part we include or link to the details of design stud-
ies for elements of the program described in Part 4, and
Part 4: Complex Programs other studies done as examples. This allows the narrative
in the earlier parts of the book to flow without interrup-
Having looked in part 3 at the elements of a single sys- tion by a mass of detail, and also serves as examples of
tem, Part 4 considers complex programs that are extended the types of studies and reports which engineers are fre-
in time or involve multiple systems. A key point to un- quently asked to work on. The full details showing the
derstand is no single system works best in isolation. For step by step assumptions and calculations are usually kept
example, a vehicle without a fuel supply can only travel as part of a program’s work history, and the results are
once, with it’s original fuel load. A fuel extraction plant distributed as reports for others to use.
6 CHAPTER 1. PART 0 - INTRODUCTION

References and Sources

We provide links to references and additional data in two

main ways. References or sources that are specific to one
method or idea are linked in the main body of the book.
In-line links are boldface and capitalized. Additional ref-
erences that cover multiple topics are included in this last
section, along with appendices for data that does not fit in
the main narrative. The first appendix, for example, lists
fictional transport methods which do not have any scien-
tific or engineering support. They are there for complete-
ness, but as there are no practical prospects to use them,
we put them in a separate section rather than the main
body of the book. Additional Appendices include refer-
ence data.
Chapter 2

Part 1 - Science and Engineering

Fundamentals

2.1 Science and Engineering Fun-

damentals
By that we mean mathematical formulas and calculations
produce results which match what we see when we look
2.1.1 Introduction at the real world. This is a very powerful circumstance,
because we can do the calculations before we look, even
The design and development of space systems is a part of before something exists, and thus predict the future. Why
engineering in general, which in turn relies on the knowl- mathematics works so well in describing reality is a philo-
edge base of mathematics and the sciences. Therefore we sophical question to which we don't have a good answer.
the first part of this book is a review of the fundamentals This was pointed out by Eugene Wigner in 1960 in an ar-
of these fields. We will pay the most attention to the parts ticle entitled The Unreasonable Effectiveness of Math-
that apply to space systems, but we by no means cover ematics in the Natural Sciences (which is also discussed
the whole of these fields of knowledge. A non-technical in a Wikipedia article). Regardless of why, it does work
reader can get a general idea of the concepts and projects in practice. That allows us, among other things, to design
presented in this book, but it is mainly aimed at people systems that will work as intended.
who want a deeper understanding of, or to actually work
with future space projects. To do that, a proper founda- The correspondence of mathematical predictions to the
tion in mathematics and the sciences is needed at a sec- real world is not just a general one. In many cases it can
ondary education (high school) graduate/first year univer- be astoundingly exact. One of the earliest examples of
sity science or engineering level. If you do not have such prediction is the motion of the Sun, Moon, and planets in
a background, there are open source textbooks available the sky. Even in ancient times people were able to pre-
online, such as those from the CK-12 Foundation, as dict where they would be in the future. Those predictions
well as video lectures from the Khan Academy, and tra- allowed knowing useful things, like when to plant crops,
ditional books and classes. because of the linkage of the Earth’s motion around the
Sun to the seasons. Nowadays we can predict the mo-
Our discussion in this book is thus at an introductory en- tion of objects in the Solar System to fractional parts per
gineering level. It is not a complete survey on any topic. million accuracy. An example of this accuracy was the
In many cases there is simply too much detail to fit it all. 2012 landing of the Curiosity rover on Mars within 2 km
In others the technical level is too advanced, and, in the of the intended location, after a trip of 566 million km.
case of some future methods, the ideas have not been fully This could not have been done without predicting both the
developed yet. Other books, articles, and materials are spacecraft trajectory and the future location of the land-
linked throughout the book, especially in the References ing point on a moving and rotating planet to 4 parts per
section, and also in our online Library. Readers are billion. Further examples of using mathematics in design
encouraged to delve deeper into any topics that interest are all around you. Every tall building and bridge relies on
them. The next few sections will give a more detailed the simple mathematical relationship that the strength be
summary of the background of mathematics and science, greater than the sum of all the loads. When you design
and how it relates to engineering and key design princi- such structures you calculate the strength, and calculate
ples for space systems. the loads, and then make sure the first is greater than the
second. Proof that this method works is that tall buildings
and bridges rarely fall down.
2.1.2 Mathematics
Like other engineering fields, space systems engineering
The importance of mathematics to science and engineer- relies on using such formulas and calculations. They are
ing can be summarized in one sentence: derived either from the sciences or practical experience

7
8 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

and measurements within engineering. We present many Other fields besides these three will also prove useful de-
of these formulas and calculations in this book. Therefore pending on the type of project. At least a basic under-
as a minimum you should understand the following math- standing of these areas of science is needed to work with
ematics topics (links are provided to introductory text- space systems, since engineering of those systems is de-
books): rived from that knowledge. Beyond the branches of sci-
ence, you should have an understanding of the Scientific
• Algebra - How to manipulate algebraic formulas Method, by which ideas are generated, experiments and
and how to obtain a numeric answer given input val- observations are made to test those ideas, and thus they
ues, the relationship of formulas and functions to are validated or rejected. Peer review, statistics, and re-
graphs, and exponents and polynomials. peatability are among the methods used to ensure obser-
vations and conclusions are reliable. Absolute truth is
• Geometry - The types of geometric shapes and an- never reached in science, merely increasing confidence in
gles, and how the dimensions of two and three di- a given explanation, which is known as a Theory. Suffi-
mensional shapes (perimeter, area, and volume) are ciently well tested ideas join the body of knowledge con-
calculated. sidered settled, but they are always subject to revision and
new ideas are constantly proposed.
• Trigonometry - Basic trigonometric functions and
graphing them, vectors, and polar coordinates.
• Probability and Statistics - The ideas of averages, 2.1.4 Engineering
random error, distributions, and regressions.
Mathematics and science are developed for their
own sake and for their ability to predict the future.
More advanced topics, such as Mathematical Analysis, Engineering then applies accumulated knowledge, from
Calculus and beyond are helpful in understanding how the sciences and from experience, towards useful ends
the formulas are derived, or in solving the more complex by designing, building, and operating systems to perform
problems in engineering. They are mostly not needed for intended functions. When the systems are complex, a
an introductory level book such as this one. method called Systems Engineering is used across an
entire project to organize and optimize the resulting de-
2.1.3 Science sign. This method can coordinate the work of thousands
of people. Systems Engineering is described in more de-
Science is a systematic enterprise that builds and orga- tail later in Part 1.
nizes knowledge in the form of testable explanations and The total of accumulated engineering knowledge is too
predictions about the universe. The predictions are of- vast for any single person to know more than a small part
ten embodied in the results of mathematical formulas and of it. Therefore engineering in general is divided into
calculations which relate to the real world. It is pursued major fields of specialization, each of which has it’s own
partly on the basis that knowledge about the Universe sat- training path. It starts with a common basis in science
isfies human interest and curiosity. It also often turns out and mathematics, then concentrates on particular areas of
that the knowledge is useful in some practical way. We do application, such as Mining, Chemical, Mechanical, and
not know in advance what knowledge will turn out to be Electrical Engineering. Working engineers often further
useful, so scientists as a group study everything. Knowl- specialize their study and experience. They, or the orga-
edge is a seamless whole, but from its history, and for nizations they work for, are called on as needed for each
the purposes of teaching and study, it is conventionally project. This is more efficient than keeping full time staff
divided into branches according to the object of study. for every possible subject area. The specialists who are
The ones most relevant to space systems include: called on also have more experience in their area from
having worked on many similar projects. Since the teams
• Physics - This is the study of the forces and interac- working on a project are not permanent, how you manage
tions of matter and energy, the results of those inter- their interaction then becomes important. Project orga-
actions, and the fundamental laws and components nization is also covered later in this part of the book.
of which things are made. Aerospace Engineering is the specialty field within which
• Astronomy - This is the study of objects and phe- space systems fall. Space systems are projects which hap-
nomena outside the Earth’s atmosphere. This is the pen to operate in the space environment in the same way
same location as all space systems operate in, so that ships and airplanes happen to operate in the water
is highly relevant. Planetary Science in particularor through the air. Although the particular environment
studies condensed objects orbiting stars. imposes differences in how things are designed, they all
rely on the same base of knowledge in subjects like me-
• Chemistry - This is the study of matter at the chanics, materials science, and thermodynamics. So a
atomic, molecular, and larger scales as far as how complex project will use engineers from many of the spe-
they react and their physical properties. cialty areas such as Mechanical, Chemical, and Electrical
2.2. BASIC SCIENCES 9

engineering, as well as Aerospace Engineers specialized design drawings. The natural environment can fluc-
in the methods and environments that apply to space. We tuate over time, and be uneven from measured av-
will identify the other specialties later in Part 1 of this erages. So all engineering designs need to account
book, but will concentrate on the methods that apply to for the uncertainties in the physical data they are
space. There are many existing books and articles about based on and production variations. One method to
the other specialties for those who are interested. do this is to introduce Design Margins above the
This book is aimed at an introductory university engi- expected conditions that are larger than the uncer-
neering level reader. If you have no prior background in tainties. How much margin to use is based on cost,
experience, and the use to which the design is put.
engineering or in space systems in particular, you may
want to start with Engineering - an Introduction by For example, a passenger airplane would generally
have higher margins than a drone with no crew, even
the CK-12 Foundation. You can get additional back-
ground from some of the book-length and website ref- though both are aircraft.
erences in the References section, the JPL Basics of
Space Flight, Glenn Research Center Beginner’s Guide
to Rockets, and Mark Prado’s Permanent and Robert 2.2 Basic Sciences
Braeunig’s Rocket and Space Technology
2.2.1 Physics as a Subject
2.1.5 Design Principles Physics is the study of how the Universe behaves in its
component parts: matter, energy, forces, motion, space,
Through training and experience, engineers develop a time, and so on. Our understanding of physics is based
sense of what will work or not, and how to optimize a on experiment and observation, and ideas developed to
design. Partly this is through broad principles that apply explain what we see. There are many ideas, but only one
in their specialty. We note a few of the more important reality. So the Scientific Method uses experiment and
ones that apply to space systems here. These and others observation to determine which ideas best match reality.
will appear throughout the book and we will try to high- Ideas are loosely graded in quality; as hypotheses, theo-
light them: ries, principles, and laws. These terms are applied based
on how firmly and widely they have been tested.
• Earth vs Space - On Earth, transport involves fric- In the sciences, no idea is considered final or absolute
tion of various kinds, and most things are moving truth. Rather, they are always subject to revision or re-
slowly in relation to each other. Therefore energy placement when confronted by new observations and ex-
and cost are proportional to distance, but not time. periments. Each new observation increases our degree of
Space is a nearly frictionless medium, and things confidence for the ideas that it supports, and many ideas
are moving at relatively high velocity with respect have been tested for so long, and in so many ways, that we
to each other. So difficulty and cost are more re- can rely on them routinely, even in engineering projects
lated to kinetic and potential energy, which governs whose failure would be catastrophic and expensive.
the paths you follow. It also depends on the time
you start, since your destination does not stay in the Some ideas in theoretical physics explore aspects of re-
same relative location, rather than absolute distance. ality that are not easily observed. They require devices
like larger and more expensive telescopes or particle ac-
celerators, or nobody has figured out a way to test the
• Non-Linearity - Many of the formulas and vari- idea yet. Theoretical ideas can be developed with noth-
ables related to space systems have values raised to a ing more than mathematics, a pencil, and pad of paper.
power or an exponential. So the difficulty of a task However, most of physics is empirical - based things we
does not have a one-to-one relation to the magni- can see, test, and use. For example, quantum mechan-
tude of the desired goal. This is called a non-linear ics is illustrated and tested by the double-slit experiment,
system. Understanding the direction and amount and special and general relativity by observing bodies in
of the non-linearity is important, as this can greatly motion through the solar system. These also have engi-
help or hinder a given task. One of many exam- neering consequences in some applications, such as in-
ples is atmospheric pressure, which decreases expo- tegrated circuit design and the timing of satellite signals
nentially with altitude, thus decreasing aerodynamic (such as GPS). Conversely, some observations, like rota-
drag proportionally. tion curves and gravitational lensing of galaxies, don't yet
have a good explanation. We call what causes the obser-
• Uncertainty and Margins - Although some values, vations Dark Matter, but we don't yet know what dark
like the orbit of a planet, are known quite accurately, matter is. So physics and the other sciences are unfinished
no physical parameter is known with absolute ac- works in progress. Many parts are well understood and
curacy. Anything built by humans will deviate by settled, but around the edges are parts still being worked
some amount from the ideal item embodied in the on.
10 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

[Move to page 3 Astronomy: In some cases, it is possi- 2.2.2 Units and Coordinates
ble to test something by looking for logical consequences
which would not be true in competing scenarios. For ex- Units
ample, measurements of the cosmological constant has
helped narrow down the theoretical possibilities regard- In order to obtain the correct results from a formula, a
ing the origins of the universe by excluding those which consistent set of units and method of measuring physi-
predict higher or lower energy levels.] cal quantities such as position in those units is necessary.
For example, adding two feet to three meters to get five
of something does not produce a meaningful result be-
Scope of this section cause the units are different. The International System
of Units (PDF file), abbreviated from the French to SI,
In this section we discuss key physical principles that re- is the preferred system of units for engineering and sci-
late to space projects. These are only a subset of physics entific work. It is also known as the Metric System be-
as a whole. We first look at them in ideal terms for the cause the base unit of length is the “Metre” (or meter in
purpose of understanding them individually. However, English). For historical reasons some values in space sys-
realistic design work has to consider less than ideal con- tems design are reported in US customary units, but these
ditions, such as friction or perturbing additional forces, should be converted to SI values. There are also units of
and also must account for uncertainties in how well we convenience, such as gravities being a multiple of Earth’s
know any measured property. This difference represents surface gravity. It is convenient to express acceleration
the difference between physical laws and practical engi- effects on humans in this way, relative to the Earth nor-
neering. mal value. It should always be recognized as a conve-
nience, and converted to SI units when doing calculations
Physical principles are usually expressed mathematically
by it’s standard ratio of 1 gravity = 9.80665 m/s2 . Phys-
as algebraic formulas and geometric relationships, with
ical quantities include both the numerical value and the
supporting explanations to provide meaning and context.
units, and units must be carried through properly when
When known numerical values with proper units are in-
doing calculations.
serted into these formulas, you can solve for an unknown
value you wish to know. The ability to calculate unknown The base SI units are not defined in terms of other units,
values is enormously useful when designing or operating but rather by a description of how to measure them from
space projects. As noted in the introduction to Part 1, the nature. In one case, the kilogram, the base is a physical
reader should have an understanding of mathematics if artifact, but that may soon be replaced by measurements.
they want to use these formulas themselves. The base units are currently the Meter for length, Kilo-
gram for mass, Second for time, Ampere for electric
current, Kelvin for thermodynamic temperature, mole
To Learn More: for amount of substance, and candela for luminous in-
tensity. Efforts are underway to define these base units
It is not our purpose here to include an entire physics text- in terms of constants of nature, but they are not com-
book, but rather a summary of the most important rela- plete. Derived SI units are the products of powers of
tionships that apply to space systems. For more detail on the base units. For example, the unit of force, called the
physics in general, you can refer to one of the following Newton, is 1 kilogram-meter times seconds−2 . Many of
sources: the derived units are named after famous physicists, but
these named units are identical to the form expressed in
base units. Multiples and sub-multiples of units are in-
• Related books in the Wikibooks Physics subject dicated by prefixes which indicate integer powers of ten
heading, ranging from −24 to +24. Among the more common are
kilo, indicating 103 or 1000, and milli- indicating 10−3
• Physics articles listed in the Wikipedia Outline of or 0.001.
physics and extensive Index of articles,

• Pages in the HyperPhysics website, Position

• Short YouTube videos from the Khan Academy, In modern physics there is no absolute or preferred ref-
erence frame in the universe. Therefore position is mea-
• Open source textbooks like the ones at the Light sured relative to a starting point known as the Origin,
and Matter or Motion Mountain websites, or which is given a value of zero. Altitude on Earth is mea-
sured relative to sea level in the direction opposite the
• A printed college textbook such as University local gravity direction. On bodies without an ocean to
Physics, though less expensive used textbooks may serve as a reference, an average ellipsoid based on the
be a better choice. shape of the planet or satellite is defined as zero altitude.
2.2. BASIC SCIENCES 11

On gas giants, which do not have a visible solid or liquid with vectors. It is somewhat different and more complex
surface, altitude is based on pressure. On near-spherical than simple algebra.
bodies, Latitude and Longitude are measured relative In accelerated motion, the velocity at any given instant is
to the points where the surface meets the axis of rotation, changing. We can define an Instantaneous Velocity at a
which are called the Poles, and a point assigned values particular time, and an Average Velocity over an interval.
of zero in both coordinates called the Zero Point. Units In a closed orbit the moving object returns to its starting
of Degrees, which are 1/360th of a circle measure the point. Thus for a full orbit, the net change in position is
location relative to the zero point. zero, and as a vector the average velocity is also zero. If
Objects in space are generally moving in relation to each you measure the total length of the orbital path and divide
other in paths defined by gravitational forces. When the by the time one orbit takes, you can obtain an average or-
paths are purely the result of gravity they are called Or- bital speed as a positive scalar value. This illustrates how
bits and are measured by six parameters called Orbital different vector and scalar values can be. Acceleration
Elements from which you can calculate position at a can also change with time. For example, the accelerating
given time. When absolute position is more useful, it force due to gravity changes as the inverse square of dis-
can be measured in three dimensions relative to an origin, tance. Thus a falling object will increase in acceleration
such as the center of the Sun, with axes defined relative as it gets closer. Under constant acceleration in a straight
to the stellar background. Alternately a radial distance line we can determine change in position or distance d
and two angles relative to a reference plane can be used. from:
Most often the reference plane is that which contains the
Earth’s orbit around the Sun, known as the Ecliptic. In a
Universe of three physical dimensions, it takes three val-
ues to define a position uniquely, either distance in three d = ∆x = 1/2 × at
2

axes, or a radius and two angles. These values are known

as the object’s Coordinates. In circular motion, where v is the velocity and r is the
radius, we can find the acceleration a from the following
formula:
2.2.3 Motion
Displacement is the change in position. It has both
v2
amount and direction, such as “three kilometers North”. a=
r
Velocity is the rate of change in position per unit time.
When stated without a direction and purely as an amount
One use for this formula is finding a required velocity for
it has a single value in units of meters per second. Where
a circular orbit from the acceleration of gravity (see un-
x is position in the direction of motion, t is time, and the
der Forces below) and the radial distance from the center
Greek letter delta (which looks like a triangle) indicates
of the body. Note there is no centrifugal force. Gravity
change in those values, then velocity v is given by the fol-
or a rotating structure provide an inwardly directed force
lowing formula:
to maintain circular motion of the object, but there is no
outward one.
∆x
v̄ =
∆t
Acceleration is the rate of change in velocity, or the sec-
2.2.4 Forces
ond derivative of position, with respect to time. Thus ac-
celeration a is: The 20th century theories of General Relativity and
Quantum Mechanics are more accurate predictors of
how objects behave in the realms of the fast and the small,
∆vx 2
d x but in many cases the simpler formulas of Classical Me-
ā = = 2 chanics are sufficiently accurate to use. Examples where
∆t dt
classical mechanics is not accurate enough include long
The horizontal bar over v and a in the respective formu- term changes in the orbit of Mercury, which being the
las indicates velocity is directional. A value such as this closest planet to the Sun, moves the fastest, and GPS nav-
when stated with both a magnitude and direction is called igation, which relies on extreme accuracy of the orbits
a Vector, while a value without a direction is called a of the satellites, and the effect of gravity on their sig-
Scalar. The direction can be given in terms of two an- nals, to determine user position. Isaac Newton formu-
gles, or the velocity can be expressed as components in lated many of the basic ideas of classical mechanics in
the three (x, y, z) axes of a reference system, but either his Principia, published in 1687. These include his three
way a total of three values are required to state a velocity laws of motion, conservation of momentum and angular
vector. Vector algebra is a method for doing calculations momentum, and the law of universal gravitation.
12 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Newton’s Three Laws formula force has units of kilogram-meter per second
squared. We call this unit a Newton (abbreviated “N”)
These are laws in the mathematical sense, which New- in the SI system of units, and is named after the scientist.
ton deduced from experiments performed by others be- The force which the Earth exerts on a falling apple is co-
fore him. They involve two opposing concepts: Forces incidentally, given stories about the scientist and falling
which tend to create motion, and Mass which tends to apples, about 1 Newton. The force of gravity on an ob-
oppose it via the property of Inertia. The relationship of ject is referred to as Weight. Most humans live where
forces to the motion they create is known as Dynamics. acceleration of gravity is within 0.2% of the same value,
Forces are vectors, having magnitude and direction, and so we often confuse weight with mass. They are propor-
multiple forces act as the vector sum of the component tional, but they have different units. On another planet,
forces. This also means single forces can be decomposed the same object would have a different weight. Weight
into components, such as vertical and horizontal compo- does not disappear when in orbit - aboard the Space Sta-
nents relative to an axis system, or perpendicular (normal) tion the force of the Earth’s gravity is only 10% less than
and parallel components relative to a surface. Decompo- on the ground. So-called “zero gravity” is more properly
sition is done when it is useful in solving a problem. The described as Free Fall. The astronauts inside the Station
three laws are: and the Station itself are both affected by the same ac-
First: Inertia - A body acted on by no net force moves celeration of gravity. So the difference between them is
with constant velocity (which may be zero) and zero ac- zero, and the astronauts do not feel their bodies pressed
celeration: against anything. On Earth what you feel is parts of your
body pressed against the ground or furniture, and inter-
nally pressing against other parts of your body. This pres-
∑
F⃗i = 0 ⇒ a = 0 sure is what you experience as “weight”.
The product of mass times velocity is called momentum.
This is contrary to common earthly experience where
It is given the symbol p since mass already uses the letter
friction acts to stop objects in motion.
m. Force also equals the change of momentum with re-
Objects moving freely in the vacuum of space demon- spect to time, since acceleration is the change in velocity
strate this Law more clearly since it is a frictionless en- with respect to time and we are just adding the multiplier
vironment. An airplane in level flight has multiple forces of mass:
acting on it (gravity, lift, thrust, and drag), but if the vec-
tor sum of all the forces is zero, it will continue moving
at the same altitude and velocity in the same direction. F⃗ = d⃗ p
From the fact that buildings typically are not accelerating dt
we can deduce there is no net force acting on them. To Third: Reaction - Single forces do not act in isolation.
put it another way, the sum of the forces is zero. Since At the most fundamental level the particles which carry
gravity acts to pull the building down, there must be an the four forces of nature act on both the emitter and ab-
equal force from the bedrock acting to hold it up. Apply- sorber of the particles. At the macroscopic level we live
ing this idea to every structural component of the build- in, where forces are the combined action of many parti-
ing is a powerful way to determine the necessary design cles, we observe the dual action as for every force there
of those components - at each point where components is an equal and opposite reaction force. Where the sub-
connect, the forces must sum to zero, therefore you can scripts indicate the force of object A on object B and
calculate the forces which a particular component must object B on object A:
withstand.
Second: Force - When a force does act on a body of
mass m, the acceleration a is related to the magnitude of Fab = −Fba
⃗ ⃗
the force F by the formula Therefore a body can never move itself by applying forces
only to itself, because the reaction force would cancel it
out. You cannot lift yourself above the ground no matter
F⃗ = m⃗a
how hard you try by applying forces to your own body. A
The arrows above the symbols indicate they are vectors, pole vaulter, however, can raise their body a considerable
meaning a quantity in a particular direction. Thus a force distance by applying force to the ground. The reaction
in a given direction produces an acceleration in the same force of the ground through the pole then acts to raise
direction. Manipulating this simple formula has very their body. Of great interest for space systems is that a
wide ranging use in space systems. Given any two of the rocket engine applies a great deal of force to expel gases
values, we can find the third. Summing across time, we in one direction, and the gas applies a reaction force in
can find total velocity change. the opposite direction, which moves the vehicle.
Since mass has units of kilograms, and acceleration has Multiplying both sides of the above formula by units of
units of meters per second squared, then by the above time, and subtracting the left side from the right side,
2.2. BASIC SCIENCES 13

we find that sum of momentum (mass times velocity) The force acts on a line between each pair of objects and
changes is always zero. This is known as the Law of the total force is simply the sum of the individual forces
Conservation of Momentum. It is referred to as a phys- accounting for the direction of each, and each is found by
ical law because it has never been observed to be violated, the formula
and conservation in the physics sense means a value which
does not change. It is found to be conserved both for lin-
ear and rotational motion. The latter is referred to as An- GM m
F =−
gular Momentum. Thus the Earth would continue to ro- r2
tate forever unless acted on by outside forces. Such forces
Since force is also mass times acceleration, we can equate
do in fact act, mainly tidal forces from the Moon. So the
them and remove mass m from both sides of the equation,
Earth’s rotation is slowing down measurably - the day is
giving the acceleration due to gravity of an object with
getting longer by 23 microseconds per year. But since
mass M as
angular momentum is conserved, slowing the Earth’s ro-
tation means the Moon increases its angular momentum.
This increases the size of its orbit by a measurable amount
GM
(3.8 cm/year) ⃗a = −
r2
When restrained from accelerating, such as when you
The Forces of Nature
stand on the surface of the Earth, you experience the
downward force as weight. Your mass does not change
There are only four fundamental forces that we know of,
according to what object you are standing on, but the ac-
responsible for all motion in the Universe. These are the
celeration does, due to the object’s different mass and ra-
gravitational, electromagnetic, weak nuclear, and strong
dius. Therefore your weight will be different on other
nuclear forces. These forces interact via gravitons, pho-
planets and satellites.
tons, W and Z bosons, and gluons respectively. For more
detail see Fundamental Interaction. The latter two are When unrestrained from accelerating, also known as free
short range forces which mostly occur within atomic nu- fall, then the parts of your body, for instance if riding in
clei, so the two that concern space projects the most are vehicle, all will accelerate at the same rate. They do not
gravity and electromagnetism. have any acceleration relative to each other, sometimes
called Zero gravity, but incorrectly since nowhere in the
Universe is there truly no gravitational field. So in the
Gravity Gravitons are the hypothetical particles which case of unrestrained acceleration “free fall” is the more
should carry the gravitational force. They are hypothet- correct term, and “zero apparent gravity” will provide a
ical because they have not yet been observed. Because better indication as to express how it appears to a human
gravitons never decay, their range is infinite, and the grav- relative to their surroundings.
itational field of any object in the Universe affects every
other object in the Universe. As a result, the total grav- The local acceleration at the Earth’s surface has a standard
itational field surrounding an object remains the same value of 9.80665 meters per second squared, but actu-
at any distance. The area of a sphere surrounding an ob- ally varies about 2% depending on location. The standard
ject is 4 × pi × the radius squared. So the gravitational value is given the symbol g, and accelerations are some-
field per unit area decreases with the square of the ra- times stated as multiples of standard Earth gravity to give
dius r. The rates of graviton production and absorption an impression of how humans would experience it, but
are both proportional to the mass of an object. Between for calculation purposes meters/second squared should be
any two objects the total gravitational force depends on used to avoid unit errors. Similarly weight should not be
the product of the two masses, since the first object emits confused with mass. Weight is in reference to the local
a number depending on it’s mass M, and then the sec- gravity field, while mass is the more correct unit to use at
ond object absorbs some of them according to it’s second any location.
mass m. The rate of graviton production and absorption
is measured as a universal constant G which applies to Electromagnetism Photons, the particles which carry
every object in the Universe, as far as we know: the electromagnetic force, behave similarly to gravitons in
that they do not decay as they travel, and obey an inverse
square field law. Where gravity is the result of mass, elec-
G = 6.67 × 10−11 N m2 /kg 2
tromagnetic force is the result of electric charge. Unlike
Gravity always acts to attract two objects to each other, gravity, charge comes in two types which we call positive
in other words reduce the distance, therefore the force is and negative. The names are arbitrary, positive charges
given a negative value. As a practical matter, since the are not larger or higher than negative ones, but they have
field falls as the square of distance, objects sufficiently the property that like charges repel each other, and unlike
far away can be ignored to the extent you need to accu- charges attract. The electromagnetic force is found by the
rately calculate the total gravitational force on an object. formula
14 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

merical) value found by multiplying two vectors (direc-

q1 q2 tional values), the direction of the applied force, and the
F = ke 2 direction of motion. Since those directions do not have
r
to be the same, the vector product can vary as the cosine
Where F is the force, k(e) is fixed value called Coulomb’s
of the angle between them, and therefore be zero or neg-
constant, q1 and q2 are the electric charges, and r is
ative. For example, if you apply a lifting force to a table,
the distance between them. Note the form of this equa-
but not enough to raise it off the floor, you do no work in
tion is similar to the one for gravitational force. When
the physics sense, even though your muscles will tell you
both of the charges are positive, or both negative, their
they are working in the biological sense. If you manage to
product is positive, and so is the force. Positive forces
lift the table, the direction of motion (up) is opposite the
act to increase the distance between charges. When the
direction of gravity (down), and therefore the work done
charges are unlike, one positive and one negative, the
on the table is negative. As odd as that sounds in conver-
product is negative, and thus the force acts to decrease
sation, the mathematics works out when solving physics
distance. Coulomb’s constant, where C is charge in units
problems.
of 1 Coulomb = 1 mole of elementary charges is
Energy is then defined as the ability to do work. It comes
in many forms which can be converted either by natural
ke = 8.987 × 109 N · m2 /C2 actions or human devices. As far as we have reliably ob-
served, total energy always remains the same, a princi-
Elementary charges are those on a single electron or pro- ple known as Conservation of Energy. An exception
ton, and are always observed as integer multiples of those to this might be Dark Energy, a hypothesis to explain
charges, never fractions. Charges are additive by simple the apparent acceleration in expansion of the Universe.
arithmetic, with negative charges canceling the fields of We do not yet know what Dark Energy is, though. It is
positive charges. Since unlike charges attract each other, a label we apply to explain certain observations, and we
they tend to annihilate if they are antiparticles or form definitely have no way to apply such energy. For practical
neutral atoms if they are protons within atomic nuclei and engineering purposes we ignore it and treat conservation
electrons. So large quantities of matter tend to have low of energy as a firm principle.
net charge. Since mass is always positive, large quantities
of matter always have large amounts of gravity. Energy is measured in SI units by Joules, named after a
19th century physicist who helped discover the relation-
Moving electric charges create a magnetic field, includ- ships of energy, work, and heat. Since energy comes in
ing the imputed spin of the charge from elementary parti- different forms, the Joule has several equivalent defini-
cles. Materials with aligned atomic spins can have a static tions. Leaving out the numerical values and only looking
magnetic field. A steady flow of electric charges is called at units it can be expressed as:
a Current, and also creates a field. Magnetic fields in turn
affect the motion of electric charges creating a force F,
where I is the current, l is the length of the wire, and B is kg · m2
the strength of the magnetic field: J= = N · m = Pa · m3 = W · s
s2
The first equivalence is to base SI units of kilograms, me-
F = Iℓ × B ters, and seconds. Where N is Newtons, Pa is Pascals, the
unit of pressure, and W is Watts, the unit of power, the
The bold face symbols indicate these are vector values, following expressions are in terms of force × distance,
having directions. The force is perpendicular to both the pressure × volume, and power × time. Note that W as
direction of the wire/current and the magnetic field. Nat- work and W as Watts mean different things, and the lat-
ural magnetic fields, such as the Earth’s, are assumed to ter is distinguished by having a quantity attached (ie 100
be caused by electric currents within the metal core. W meaning 100 Watts). Unfortunately there are more
physics concepts than letters of the alphabet, which can
be confusing at times. When a formula could lead to such
2.2.5 Continued on page 2 → confusion, write out the unit in full rather than abbreviate,
or define the symbol in words as we usually do around a
2.3 Basic Sciences 2 formula. Which of the above expressions for energy are
appropriate to use depends on the types, and which con-
versions of it, are involved in a particular situation. One
← Back to Page 1
of the forms which energy takes is as matter. Where E
is energy, m is mass, and c is the speed of light they are
2.3.1 Energy related by the famous equation

Work in the physics sense is a force applied through a

distance, or in equation terms W = Fd. It is a scalar (nu- E = mc2
2.3. BASIC SCIENCES 2 15

Since the speed of light is a large number, by definition as the inverse of distance. Where G is the gravitational
exactly 299,792,458 meters per second, and that number constant, m(1) is the mass being moved, and M(2) is the
is squared in the formula, the energy contained in a given large mass producing gravity, then the potential U is:
amount of mass is enormous. The conversion of less than
1% of mass in nuclear reactions produces sufficient en-
ergy to power stars, atomic bombs, and nuclear power m1 M2
U = −G
reactors. Objects which are moving in gravity fields are r
very common in space projects. We find it useful to define If no forces besides gravity are acting on an object then
the following two energy quantities based on their motion the sum of kinetic and potential energy is constant. Col-
and position: liding with a planet or other object involves other forces
than gravity. So an object in an elliptical orbit, which
does not collide, and is free to repeat its motion, con-
Kinetic Energy stantly exchanges potential and kinetic energy as the dis-
tance r from the center of the body it is orbiting changes.
Kinetic Energy is the energy an object possesses through It has more kinetic and less potential energy at the lowest
its motion. It can also be described as the amount of work point and so moves faster. The velocity v at any point in
required to get a body of mass to move. In mathematical the orbit can be found from:
form Work = change in Kinetic Energy, or

Semi-minor axis
Semi-major axis
W = ∆KE = K2 − K1
focus focus
This is in addition to the bound energy of matter ( mc^2
). Referring to Newton’s first law of motion, an object
will retain its kinetic energy unless acted upon by another
force such as friction or gravity. For example, objects free
of a gravitational field and in a vacuum will retain their
kinetic energy, direction, and velocity. Kinetic energy KE
is a function of mass m and velocity v according to the √ ( )
formula 2 1
v = GM −
r a
where:
KE = 21 mv 2
• r is the distance between the orbiting object and the
A Reference Frame is a non-accelerating environment. body it orbits. The body is located at one focus of
Velocity is measured relative to such a reference frame. the ellipse.
Therefore a space station in orbit, and an astronaut inside
it, may both have a large velocity relative to the center • a is half the long axis of the orbit shape, or Semi-
of the Earth. They then both have a large kinetic energy major Axis (see figure at right)
in an Earth-centered reference frame, in fact sufficient
to raise their temperature hotter than the Sun, to 7000
K. Relative to each other, however, their velocity is near 2.3.2 Mechanics
zero, so in a reference frame moving with them they have
near zero kinetic energy. Since the formula above takes Mechanics is the description of the motion of an object
the square of velocity, kinetic energy is always positive, under the influence of forces. For space projects this is
even if the velocity is negative in a given reference frame. usually the Thrust forces generated by a propulsion sys-
tem, and the influence of Gravity. The motion in a vac-
uum among massive objects like planets and the Sun is
Potential Energy called Orbital Mechanics, which we will cover in sec-
tion 1.2. When operating within an atmosphere, an ad-
Potential Energy is the difference between the energy ditional force is generated when moving. This force is
of an object in a given position and its energy in a refer- decomposed into a perpendicular component called Lift,
ence position. When work is done against a conservative and a parallel component called Drag. When two objects
field, such as gravity, then the energy of that work is con- are in contact they additionally generate a force which is
verted to potential energy. Setting the reference position decomposed into a parallel, or Friction force, and a per-
as infinitely far away, the gravitational potential is always pendicular, or Normal Force. The combination of all
negative since you must do positive work to lift an ob- forces, including less common ones not listed in this para-
ject to infinity. Since gravity varies as the inverse square graph, produce a vector sum total force on the object and
of distance, the sum of the work going to infinity varies thus an acceleration in some direction.
16 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Friction When something external acts on the vehicle it is referred

to as an accelerating force, and often specifically named.
Friction forces matter for space systems for things like The magnitude of the thrust due to expelled mass is given
rovers moving on the surface of a body and for inter- by
nal parts of motors and pumps. Since they always op-
pose motion, they require energy to overcome, and thus
a source of energy in the system. Friction at a micro- dm
T= v
scopic level is caused by interactions between electrons dt
of the two objects, thus is an electromagnetic effect. At
dm
a slightly larger level they cause temporary bonding and where T is the thrust generated (force); dt is the rate
physical obstruction due to surface roughness. At human of change of mass with respect to time (mass flow rate
scale these microscopic interactions can be summed up of exhaust); and v is the speed of the exhaust measured
as average values which are proportional to the normal relative to the vehicle.
forces and depend on the types of surfaces in contact. The
multiplier to the normal forces is called the Coefficient
Drag
of Friction, with symbol µ . The friction force, f, is then
related to the normal force, F by:
Drag is a force component generated by interaction with
a fluid medium, such as the Earth’s atmosphere. It is par-
allel to the incoming flow direction, and given by the for-
f = µ s Fn mula:
Coefficients of friction are determined experimentally,
and depend on whether the objects are moving (kinetic
friction) or not (static friction). Static friction is typi- FD = 1
2 ρ v 2 CD A
cally higher because the objects have time to form atomic
where FD is the drag force, ρ is the mass density of the
bonds and settle into the bumps of surface roughness. The
fluid, v is the velocity of the object relative to the fluid, A
coefficients also depend on type of material and whether
is the reference area, which is the projected area occupied
any gas or liquid is trapped between them. The ability of
by the vehicle in a plane perpendicular to the motion, and
skaters to move easily on ice comes from a microscopic
CD is the Drag Coefficient — a dimensionless number.
layer of water which forms due to pressure of the blades.
While most of the terms in the above formula are simple
Vertical lift from contact only requires breaking atomic
to determine, drag coefficient varies in a complex way
bonds, and not interlocking of surface roughness. Thus
based on object shape, velocity, and other parameters.
wheels and ball bearings, which vertically separate the
This is caused by complex flow conditions such as turbu-
contact surfaces, have lower Rolling Friction than slid-
lence, shock waves, heating, and even decomposition at
ing contact.
higher velocities.
When a surface moves relative to a fluid, the layer clos-
Normal Forces est to the surface is affected most by the molecules col-
liding with the surface. They are deflected by the an-
A Normal Force acts perpendicular to a surface, and has gle of the surface, roughness in the surface, or atomic
several sources. (Normal here refers to the perpendicu- forces between their respective atoms. This causes that
lar direction in geometry, not common or average.) They fluid layer to tend to move along with the surface. The
includes gravity, magnetic or electrostatic attraction, and surface layer in turn affects farther layers by collisions of
gas or liquid pressure. In reality, friction and normal the molecules. At lower velocities, this sets up a smoothly
forces are components of the total contact force. Since varying Boundary Layer near the surface. At higher ve-
motion is prevented in the perpendicular direction by the locities the deflection is violent enough to create flow vor-
existence of a solid surface, it is easier to calculate the ef- texes, where the fluid develops circular motions perpen-
fects by looking at the components separately. When per- dicular to the direction of motion, known as Turbulence.
pendicular motion is not prevented, which happens with Since it takes more energy to create the vortexes, the
liquids and gases, it becomes more complex. The field forces on the surface are higher, increasing friction or
of Fluid Dynamics is the study of these more complex drag. These effects happen both externally to a vehicle
motions, both against solid surfaces and internally within moving in an atmosphere, and internally to a gas or liquid
fluids. flowing within an engine.
The ratio of inertial forces, such as the sideways deflec-
Thrust tion, to the viscous forces caused by shearing (varying
speed) in the boundary layer is called the Reynolds Num-
Thrust is the force generated by a vehicle by expelling ber. The transition from smooth, or Laminar Flow,
reaction mass or by interacting with the environment. to turbulent flow, and the size of the vortexes, and thus
2.3. BASIC SCIENCES 2 17

the drag, is found experimentally to depend on Reynolds 2.3.3 Thermodynamics

Number. It is a Dimensionless Number, meaning all
the units in the formula cancel out when using consistent Thermodynamics is a branch of physics concerned with
units, leaving a pure number. The Reynolds Number, Re, heat and temperature and their relation to energy and
is found by: work. For space projects it becomes important for things
like the operation of rocket engines, aerodynamic heat-
ing during reentry, and thermal radiators to dissipate ex-
cess heat. Thermodynamic variables and mathematical
ρvL vL
Re = = laws are simplifications applied to bulk amounts of mate-
µ ν rial. Their underlying cause is the microscopic behavior
of a very large number of smaller particles like molecules.
where: v is the mean velocity of the object relative to Thermodynamics is often related to chemistry, because
the fluid (m/s), L is a characteristic linear dimension of chemical reactions can give off or or absorb heat. A prime
the surface (m), µ is the Dynamic Viscosity of the fluid example is combustion in chemical rocket engines, where
(Pa·s or N·s/m² or kg/(m·s)), ν is the Kinematic Viscos- the propellants (fuel and oxidizer) react to produce very
ity ( ν = µ/ρ ) (m²/s), and ρ is the density of the fluid high temperature gas. Thermodynamics is a complex
(kg/m³). The characteristic dimension is defined by con- subject, so we cannot delve into all aspects of it here. We
vention for various types of shapes, such as the diameter refer you to full textbooks like Thermodynamics and
for a sphere. Since the motions within a turbulent fluid Chemistry 2nd edition by Howard DeVoe.
are too complex to reduce to a simple formula, for early
design purposes drag coefficients are usually found from To give you a flavor for the topic, and because chemi-
tables and graphs based on Reynolds number, which in cal rockets are very important in current space projects,
turn were developed from experiment or historical data. we will describe their combustion cycle here, but it is not
a complete description and many specialized terms are
In more detailed or important design projects fluid forces used:
like lift and drag are measured for the proposed de-
sign in a wind tunnel or other experiment, or calcu- The thermodynamic cycle for a liquid rocket booster is a
lated by detailed numerical simulations, a topic known modified Brayton (jet) cycle. A one-dimensional analysis
as Computational Fluid Dynamics, or CFD. In CFD may be performed by assuming the following ideal steps.
simulations the flow is broken up into sufficiently small
volumes that the flow in each volume obeys relatively sim- 1. Fuel is injected into a combustion chamber isentrop-
ple formulas, and thus the total flow in the simulation ically either through use of pressurized fuel tanks or
can be determined reasonably accurately. Historically by a high-pressure pump, increasing the pressure to
this needed the largest available computers, and there- pc and increasing the enthalpy.
fore physical testing often proved easier. With the vast
2. Heat is added to the fuel by means of combustion.
increase in computer speed in recent decades this method
In an ideal situation, it is assumed that the pressure
has become much more practical.
remains constant during this step, but that the tem-
perature rises. Both enthalpy and entropy increase
during this step.
Lift 3. The combusted fuel expands isentropically to the
exit pressure, pe , as it goes through the nozzle into
Lift is the other force component generated by interaction the surroundings, which is at pressure p0 . Ideally pe
with a surrounding medium. It is perpendicular to the should equal p0 . During this process, the enthalpy
incoming flow direction, and given by the formula decreases from hc to he .

The thrust produced by a rocket is given by

L = 12 ρv 2 ACL
T = ṁp ve + Aexit ∗ (pe − p0 )
where L is lift force, ρ is density of the medium, v is the
velocity relative to the medium, A is planform area of the where ṁp and ve are the mass flow rate and exit velocity
shape, and CL is the lift coefficient at the desired Angle of of the propellant, Aexit is the exit area of the nozzle and
Attack (angle between the reference plane of the shape pe and p0 are the pressure at the exit point of the nozzle
and the velocity direction). Lift Coefficient depends on and the atmospheric pressure. The enthalpy represents
Mach Number, and Reynolds Number. Like drag, lift the internal energy available for work or the potential en-
coefficient is a result of complex fluid flow, and depending ergy. Thus, the energy change per unit time as the pro-
on the design need is found by looking at a table or graph, pellant moves from the combustion chamber to the nozzle
physical experiment, or CFD simulation. exit is
18 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

2.3.4 Continued on Page 3 →

1
ṁp (hc − he ) = ṁp ve2
2 2.4 Basic Sciences 3
Solving for the propellant velocity yields
← Back to Page 2

√
ve = 2(hc − he )
2.4.1 Astronomy
Let us assume that the combustion mixture of the propel-
Astronomy is the study of all objects and phenomena be-
lants is an ideal gas. The internal energy per unit mass of
yond the Earth’s atmosphere, and Planetary Science is
an ideal gas is given by
specifically the study of condensed objects orbiting stars.
Since this is where space systems function, a basic un-
derstanding of these fields is highly relevant to working
h = ĉV T on such projects. We provide only a very short introduc-
tion here. For more background than will fit in this book,
producing an equation for the propellant velocity of see Nick Strobel’s Astronomy Notes and the Wikibook
General_Astronomy. One of the key ideas to emerge
√ from these studies is the Uniformity of the Universe.
( )
√ Te As far as we can tell, the natural laws and processes that
ve = 2ĉV (Tc − Te ) = 2ĉV Tc 1 − operate here and now have always operated in the past,
Tc
are the same everywhere in the Universe, and we expect
When an ideal gas expands isentropically, a change of them to continue to be so in the future. Having learned
temperature and pressure such that the following two re- what these general principles are, we can then apply them
lations hold to specific examples as needed.

( )γ/(γ−1) Origin and Evolution of the Universe

p1 γ−1 2
= 1+ M
p 2 The Universe is the totality of existence. The origin and
history of the Universe as a whole is of great interest
T1 γ−1 2 to many people for its own sake, but only selected fea-
=1+ M
T 2 tures are relevant for space systems design. This includes
where M represents the Mach number at the location hav- that Baryonic matter (the ordinary kind of matter that
ing static pressure p and temperature T. Using these two we and the Earth are made of) started out as about 76%
equations, we can relate the temperature and pressure ra- Hydrogen, 24% Helium, and almost nothing else. Grav-
tios as ity caused the nearly uniform early Universe to develop
denser regions with emptier regions in between. The
denser regions coalesced into many Galaxies, of which
( )(γ−1)/γ the Milky Way galaxy is the one which our Sun and plan-
T1 p1
= ets are part of.
T p
Galaxies in turn form denser condensations where nu-
Thus, we can rewrite the equation for the propellant ve- clear reactions occur, which we call stars. The reactions
locity as convert lighter elements into heavier ones, increasing the
proportion by mass of Helium to about 27% and heavier
elements to about 2%. Stellar nuclear reactions release
v [ ] a great deal of energy, but this source is finite. So the
u ( )
√ u pe
(γ−1)/γ
ve = 2ĉV (Tc − Te ) = t2ĉV Tc 1 − Sun and other stars will eventually run out of fuel, and
pc the composition of the Universe will reach a stable con-
dition. Several lines of evidence indicate the current age
The final analysis step in the one-dimensional analysis is of the Universe is about 13.6 billion years, and the era of
the effects of the nozzle. The previous equation demon- stars will last about 100 trillion years. If the expansion
strates that making the ratio pe /pc as small as possible of the Universe continues to accelerate, then most of the
maximizes the propellant speed, which in turn maximizes Universe will be rendered undetectable long before all the
the thrust. The nozzle is designed to match the exit pres- nearby stars die out. Regardless of the eventual destiny
sure as close as possible to the pressure of the atmosphere of the Universe, on human time scales it will last a long
or the vacuum of space. while relatively unchanged.
2.4. BASIC SCIENCES 3 19

The Milky Way Galaxy of the Sun by about 40%. The current output is 3.846 x
1026 Watts. This will continue to increase by about 1%
The oldest stars in our home galaxy are about 13 billion per hundred million years.
years old, which indicates the galaxy started forming, at
least to the point that stars condensed, shortly after the
Universe as a whole formed. The baryonic mass is em- 2.4.2 Planetary Science
bedded in a five times more massive amount of material
which only reveals itself by its gravity. Since this mate- Solar System Evolution
rial does not form stars, it is dark, and so we call it Dark
Matter. It is poorly understood at present, and our main Planetary Formation Whatever internal motions the
concern for space projects is how it affects the motions Solar Nebula had, there was no way to dispose of angu-
of baryonic objects such as stars, planets, and molecu- lar momentum (net rotation). Therefore a small part of
lar clouds. Evidence from the composition and motions the nebula remained orbiting the proto-Sun rather than
of parts of the Milky Way indicate it formed by infall of falling in, and keeping most of the angular momentum.
gas clouds and smaller galaxies, which continues to the This region is estimated to have been 50 AU in radius,
present. The shape seems to have evolved starting with and disk shaped as the net result of rotation. The increas-
the halo and central bulge, followed by growth of the disk ing temperature of the proto-Sun created a temperature
shaped region. The baryonic mass of our galaxy is esti- gradient based on distance from the center. The inner-
mated at 200-300 billion times the Sun’s. most part was too warm for icy material to remain solid,
The Sun is in the disk region, about 27,000 light years while the outer parts were cold enough for water, ammo-
from the center of the Milky Way, orbiting at about 220 nia, and other ices. No part was cold enough for Hydro-
km/s and thus taking about 225-250 million years to com- gen and Helium to condense. Particles condensed out of
plete an orbit. We do not know the exact shape of the the nebula as the flat shape radiated heat to space and the
Sun’s orbit but it is suspected to be elliptical. Random optical thickness radially kept the outer parts from be-
motions of nearby stars are on the order of 50 km/s. ing heated by the Sun. Small particles could grow first
Over the age of the Sun these random motions amount by sticking to each other, then later by gravitational at-
to 450,000 light years, which is much more than the cir- traction. The mix of objects which formed this way are
cumference of the Sun’s orbit. This indicates the current called Planetesimals. We now have fairly good evidence
nearby stars are not the ones the Sun was born near. In of this process by observing disks around young stars.
fact, the current stars within 100 light years will be re- Gravitational attraction is a runaway process. As an ob-
placed by an entirely different set in 1 million years. ject gets larger, it can attract objects from a larger dis-
tance, thus increasing its growth rate. Larger objects also
have a potential energy well, so approaching objects will
Formation of the Sun
accelerate to impact. The impact energy eventually be-
comes large enough to melt the object. In addition, there
We observe new stars forming in denser regions of our
were more radioactive elements in the early Solar Sys-
galaxy known as Molecular clouds, and we assume our
tem than there are now, and decay of those elements con-
Sun and the rest of the Solar System formed in such a
tributed to the heating of the growing bodies. The largest
cloud, which has since dispersed. Based on radioactive
objects were able to affect the orbits of the smaller ones,
dating we estimate our Solar System to be 4.6 billion
causing them to either impact or get scattered away. This
years old, which is about 1/3 the estimated age of the Uni-
tended to clear out a region around each large object. The
verse. The presence of 1.5% heavier elements in the Sun
very largest objects had a sufficiently deep gravity well
confirms that it formed from recycled matter that had pre-
that they could collect gaseous Hydrogen and Helium,
viously been enriched by older generation stars. Loss of
forming the gas giants. Some of the scattered objects,
heat from radiation allowed gravity to collect part of the
and planetesimals which formed at the outer edges of the
original molecular cloud into a distinct object called the
nebula, have survived relatively unchanged beyond Nep-
Solar Nebula. The core of the nebula continued to con-
tune. The material from the inner Solar System which
tract, and the increased pressure caused by self-gravity
could not condense there tended to be blown outwards,
heated that core to create a Proto-Sun. Once the core
and the point where they could solidify is near Jupiter’s
of the proto-Sun reached a temperature of 12 million
orbit, which may account for the large mass of Jupiter
kelvins, hydrogen fusion could begin, and the Sun proper
and Saturn compared to the rest of the planets. The en-
was born. This collapse until ignition took around 30 mil-
tire accretion and clearing process took about 100 million
lion years.
years, with the final growth of the planets perhaps taking
Nuclear reactions in the core of the Sun have converted 10 million years.
Hydrogen to Helium, increasing the concentration there
to about 60%. Since Helium is heavier, the core has got-
ten denser and hotter, thus increasing the reaction rate Planetary Evolution The bodies which were large
of the remaining Hydrogen and the total energy output enough to melt had their denser compounds sink to the
20 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

center under their own gravity. Iron and related metals interior hot. It has an inner metallic Core solidified by
are the heaviest common materials, so they ended up in pressure, even though it is about the same temperature
the cores. Going outwards, the layers include rocky min- as the Sun’s surface (6000 K). Outside of this is a liq-
erals of different densities (a mantle and crust), then ices uid metallic outer core, and then a rocky layer called the
and an atmosphere if the body formed with these com- Mantle. The deeper parts of the Mantle have tempera-
ponents. This layered structure is what we find today in tures of 2000K or more. Although pressure keeps them
the planets and larger satellites, along with the compo- solid at these high temperatures, the rock is able to flow
sition change with distance from the Sun. These trends slowly over time in a type of thermal circulation taking
are not strict rules because random collisions and gravita- on the order of 100 million years. The least dense and
tional scattering have changed the location and make up coolest rocks form a roughly 120 km thick solid layer
of the objects since they formed, and smaller bodies have called the Lithosphere, which has a relatively high tem-
lost original atmosphere and in some cases ices. After perature gradient relative to the rest of the planet. By
the original formation era of about 100 million years, the composition the lower part of the Lithosphere is part of
larger planets continued to interact with each other chaot- the Mantle, and the outer part is less dense rock called
ically until they settled into a relatively stable arrangement the Crust.
about 3.8 billion years ago. The planetary shifts affected The internal motions of the Mantle and heat traveling
the smaller bodies, who continued to be scattered or im- outwards can cause local temperature to get higher than
pact. The evidence of this is still visible in craters and the the melting point dictated by local pressure. The molten
locations of scattered objects. rock is called Magma, and its composition can vary be-
Bodies in the Solar System continue to interact gravi- cause different minerals have different melting curves.
tationally, and impacts and scattering continue, but at a Movement of magma and bulk Mantle circulation drive
lower rate. Bodies like the Earth have active processes about 20 pieces of the Crust, called Plates to slowly move
that tend to erase craters. Smaller ones lack an atmo- across the Earth’s surface and change shape. These move-
sphere or crustal motion, and preserve them through the ments and Weathering, the mechanical and chemical
life of the Solar System. The resulting current distri- changes from the surface environment, explain the ge-
bution of matter, and the very large energy output of ography and geology we find today. These dynamic pro-
the Sun, are the main resources to work with for space cesses combine to erode most of the early Earth’s history.
projects. The current surface averages ten percent or less of the age
of the planet as a whole.
The Earth-Moon System
The Moon Early in the Earth’s history, a Mars-sized
Earth Science is the study of the Earth and its compo- body called Theia is presumed to have hit the proto-
nent parts. The study of the Earth predates detailed study Earth. Some of the debris from that impact collected by
of other planets because this is where humans started, and gravity to form the Moon. It formed much closer to the
it continues to be the best studied planet. In the context Earth than it is now, but tides act to slow the Earth’s ro-
of modern astronomy and planetary science, the Earth tation and increase the Moon’s orbit, a process that con-
is now studied as one planet among many. In the con- tinues. The Moon is smaller than the Earth, and so lost
text of human history it still has a special place because its internal heat faster, and is now mostly solid. It is too
we evolved here, and until now, none of us has left the small to retain an atmosphere because of the low escape
Earth-Moon system. Nearly all of the design, materials, velocity. So the Moon retains evidence of its early history
equipment, and operations for space projects to date has in the form of large impact basins and craters of all sizes.
actually occurred on Earth. This will continue to be true The larger basins filled with magma to create relatively
for at least the near future. Therefore some understand- flat and darker areas mistakenly named Mare (Latin for
ing of the Earth is still needed to carry out space projects. “Sea”) because, before telescopes, they were thought to
For a longer introduction to the field see: have water. The greater tides from the time the Moon
was closer to Earth slowed the Moon’s rotation so the
• CK12 Foundation - Earth Science Concepts same side now always faces us, with a little wobble. It
also caused most of the Mare to form on the near side.
The Earth formed in the same way as the rest of the large
bodies in the Solar System, mostly by collisions. De-
bris from a very large collision late in the process formed 2.4.3 Chemistry
the Moon, explaining the difference in its composition
relative to the Earth. Impacts and radioactive decay re- The science of Chemistry has historically been consid-
leased enough energy to melt the entire planet, and the ered a separate subject from physics. In a more general
high temperature likely led to loss of some of the more sense it can be considered a subset of low energy physics
volatile ices and gases. Continued radioactive decay, sup- where arrangements of atoms via atomic bonding is im-
plemented by other energy releases, has kept the Earth’s portant. We humans happen to require living conditions
2.5. ORBITAL MECHANICS 21

where atomic bonding is important, so we give that en- accurate your result needs to be. Typically detailed cal-
ergy regime more attention. In reality, something like culations are done with a computer simulation, since ob-
99% of the matter in the Universe is in the plasma state, jects in general follow orbits, and move in relation to each
where electrons are no longer bound to atoms, and inter- other. Therefore the strength and direction of their grav-
atom bonding is rare. ity varies.
The importance of chemical reactions to space projects An Orbit is the path that an object will follow when only
until now was primarily in providing high temperature affected by gravity. Orbits around uniform single bod-
gases to propel rockets. In future projects chemical reac- ies are Conic Sections, which are shapes generated by
tions for life support systems, and extraction and prepa- slicing a cone. In order of eccentricity these are circle,
ration of raw materials in space, will become much more ellipse, parabola, and hyperbola. Circular and elliptical
important. So at least a basic understanding of chemical orbits are bound to the body being orbited and will re-
principles is useful for space systems design. We refer the peat. Parabolic and hyperbolic orbits are not bound to the
reader to the following sources for more detail: body, although influenced by its gravity. They will not re-
peat. Simple orbit calculations only consider the nearest
massive body. This is suitable when that body’s attrac-
• CK-12 Foundation Chemistry pre-university level
tion is much greater than other bodies, and for short time
textbook mentioned at the start of Part 1
periods. More detailed and accurate calculations have to
consider non-uniformity of the main body, and other bod-
• Wikibooks Introductory Chemistry
ies with enough gravity to influence the accuracy of the
• Chem1 Virtual Textbook by Stephen Lower result.
First let’s consider the ideal case of a single uniform mas-
sive object being orbited. Circular orbits have a constant
velocity and distance from the center of mass of the body.
2.5 Orbital Mechanics This also means they have a constant Orbital Period, the
time to complete one revolution around the body and re-
2.5.1 Introduction turn to the starting point. The circular orbit velocity, vₒ,
for any body can be found from:
Astrodynamics or Orbital Mechanics is mainly con-
cerned with motions under gravity, either purely as a sin-
√
gle force, or in combination with forces like thrust, drag, v = (GM /r)
o
lift, light pressure, and others. As a topic it has a long
history, with the motions of the planets, Moon, and Sun Where G is the Gravitational constant (6.67 x 10−11
studied since ancient times, and with a scientific base Nm2 /s2 ), M is the Mass of the body orbited (in kg), and r
starting about 500 years ago. With the advent of human- is the radius to the center of the body orbited (in meters).
built spacecraft it has shifted from merely observing the G is a universal constant, and the mass of the Earth is es-
motions of natural bodies, to planning and executing mis- sentially constant (neglecting falling meteors, atmosphere
sions. The relevance to Space Systems engineering is, of leakage, and things we launch away from Earth), so often
course, the need to travel to a desired destination or orbit. the product GM = K = 3.986 x 10^14 m3 /s2 is used.
We will present some of the key ideas here. A more The orbital period, P of a small body orbiting a central
detailed and advanced introduction can be found in the body in a circular or elliptic orbit is
Wikibook Astrodynamics and a set of Wikipedia arti-
cles at Astrodynamics - A Compendium. A good in-
troductory printed textbook is Fundamentals of Astro- √
3
dynamics, and there is an MIT Astrodynamics open T = 2π a /K
course with downloadable materials.
Escape velocity, the velocity required to escape from a
body’s gravity to infinity, or vₑ is found by
2.5.2 Orbits
√
Gravity has no limiting distance, and therefore every ob- ve = (2GM /r)
ject in the Universe is affected by the gravity of every
other object in the Universe. For practical purposes, the Since this formula is the same as that for circular orbit,
force of gravity from most of the Universe cancels out, except by a factor of 2 in the square root term, escape
since there is about the same amount of material in all velocity is the square root of 2 (1.414+) times circular
directions. What remains are more massive and nearby orbit velocity. Elliptical orbits will have a velocity at the
objects, whose gravity is large enough to matter for a nearest point to the body, or periapsis, in between that
given calculation. Which objects matter depends on how of circular and escape.
22 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Orbital Elements

Orbital Elements are the parameters required to fully L4

describe the location and orientation of an orbit, the shape
of the orbit, and the position of an orbiting object at a
given time. They are described relative to the major body
the object orbits. These elements change over time un-
der the gravitational influence of other objects, which is
called Perturbation, non-ideal shape and mass distribu- L3 L1 L2
tion of the major body, the effects of relativity, and out-
side forces like drag and light pressure. The more impor-
tant elements include:
Axes - Periodic orbits are generally ellipses. An ellipse
has a major and minor axis, which are the longest and L5
shortest distances across the center of the ellipse. These
axes are perpendicular to each other (see section 1.1 page
2 ). Half of these axes, or the distances from center to
edge of the ellipse, are called the Semi-major and Semi- relative to the large bodies where the net forces keep the
minor axes respectively, with symbols a and b. The small body approximately in the same position relative
Semi-major axis is the value usually used to describe the to the two larger ones. They are called the Lagrangian
overall size of an orbit. Points after one of the mathematicians who discovered
Eccentricity - The foci of an ellipse are the points along them. Three of these, called L1, L2, and L3, are unstable.
the Semi-major axis such that the sum of the distances If you move slightly away from the exact point, you will
from the foci to any point on the ellipse is constant. An tend to move further away. The other two, L4 and L5,
orbit of a small body around a more massive one will have are stable. Slight movements around these points will not
the massive one located at one focus of the elliptical orbit. cause the small body to drift away, but rather orbit around
The Focal length, f, is the distance from a focus to the the points. L1, L2, and L3 are located between, behind,
center of the ellipse. The shape of the orbit is measured and opposite the second of the large bodies, respectively
by Eccentricity, e, which is defined as: (see figure). L4 and L5 are located in the same orbit as
the second large body, 60 degrees ahead and behind it. As
the largest planet, Jupiter has the largest collection of as-
teroids in it’s Lagrange points. These asteroids are called
e = f /a
the “Jupiter Trojans”, since the first few at Jupiter were
The higher the eccentricity, the narrower is the ellipse named after characters in the Trojan War. More gener-
relative to the semi-major axis, and the greater the differ- ally, objects at the stable Trojan points of other planets
ence between the nearest and farthest points of the body are also called Trojans. Since most orbits are elliptical,
from the one it is orbiting. the Lagrange points shift as the distance between the two
major bodies changes.
Periapsis and Apoapsis - The prefixes peri- and ap- re-
fer to the nearest and farthest points of an orbit from
the center of the body being orbited. Different suffixes Rotation
are used to indicate what body is being orbited, such as
perigee and apogee for the lowest and highest points of Nearly every natural body in orbit also rotates, so that the
an Earth orbit, and perihelion and aphelion for distance direction from the body’s center to a fixed surface point
from the Sun. The general symbols are q for perigee and changes. This has several effects:
Q for apogee, and can be found from the formulas:
Rotation Period - This is the time it takes the body to
complete one rotation with respect to the stars, the Sun,
or the planet if it is a satellite of one. The most obvi-
q = a − f = a(1 − e)
ous effect of the rotation period is the day-night cycle on
Q = a + f = a(1 + e) Earth. Some objects become locked into a rotational res-
onance with the parent body they orbit. This means the
f = ae rotation period is a simple fraction of the orbit period.
When the resonance is 1:1, it is called Tidally Locked,
and the Moon is the most obvious example of that. The
Lagrangian Points
result is one side always faces the Earth, with a small wob-
Given two large bodies, such as the Sun and Jupiter, and a ble.
third small body, such as an asteroid, there are five points Axial Tilt - Rotation defines an axis of rotation. The
2.5. ORBITAL MECHANICS 23

ample, is pulled by it’s gravity 6.6% more strongly than

the opposite side, because it is closer in distance. The
difference in gravity between the near and far sides is
called the Tidal Force because it is the source of ocean
tides on Earth. Those tides happen because water is free
to move towards the Moon, but continents are more re-
stricted. Tides do distort the shape of the solid part of the
Earth, but less than the oceans. Tidal forces affect other
moons and planets too.
The lesser gravity of other bodies also affects the orbit
Axial tilt of the Earth to the axis of its orbit, which is perpendic- of an object around the primary it is bound to. These
ular to the orbit plane defined by the red circle. are called Perturbations, because they perturb the orbit
caused by the strongest gravity force. On long time scales
perturbations can drastically affect an orbit. This is most
places where the axis meets the surface of the body are obvious in the case of Jupiter and comets. Long period
called poles, and midpoint of the surface between the comets are often near escape velocity, so small velocity
poles is called the Equator. On smaller bodies with ir- changes caused by the large mass of Jupiter can drasti-
regular shape, the Equator may not be well defined. On cally change their orbits. This can lead to them becom-
larger bodies which are more or less round, the Equa- ing short period comets that stay near the Sun, or being
tor has the largest distance from the rotation axis. Axial ejected entirely from the Solar System.
Tilt is the angle the body axis makes with the axis of the
body’s orbit. The rotational inertia of large bodies causes
their rotation axis to remain relatively fixed relative to the 2.5.3 Velocity Map
stars. For example, the Earth’s north pole points near the
star Polaris, but this point is tilted 23.44 degrees from the In space, physical distance does not matter as much as
axis of its orbit around the Sun. So it orbits, first one pole, velocity, since space is mostly frictionless, and what costs
then the other, points towards the Sun, causing seasonal you energy is changing velocity. This graph below shows
changes (see figure) the minimum ideal velocity relative to escape for the
Rotational Velocity - When you are on the surface of Sun’s gravity well on the horizontal axis, and for plane-
a rotating body, the circular motion about the axis pro- tary wells and some satellites and asteroids out to Jupiter
duces an acceleration which can reduce gravity. The ve- on the vertical axis. There is no absolute reference frame
locity and acceleration depend on the distance from the against which to measure velocity. We choose escape as
axis and the rotation period. For example, at the Earth’s the zero point since it has the physical meaning of “to
equator the rotation velocity is 465 m/s, which generates leave this gravity well, you must add this much velocity”.
an acceleration of 0.0338 m/s2 , or about 3% of gravity. Since you must add velocity to leave, the values are neg-
Thus the apparent weight is less at the equator than the ative. If you have more than enough velocity to leave a
poles. Large bodies, more than about 1000 km in diam- gravity well, that is called excess velocity, and is mea-
eter, have internal forces greater than the strength of the sured infinitely far away.
internal materials. Since rotation lowers gravity in some Determining Total Velocity
parts relative to others, the body flows into an ellipsoidal,
Total mission velocity is the sum of vertical and horizontal
or flattened, shape. This is called hydrostatic equilibrium.
velocity changes on the graph, both in km/s. Note that the
The rotation of any body lowers the difference between
axes are different scales. To travel from Earth to Mars,
orbit velocity and surface velocity when they are in the
for example, you first have to add velocity to climb out
same direction. In the case of Earth it is 5.9% less, but
of the Earth’s gravity well, add velocity to change orbit
in the case of some asteroids, like 4 Vesta, it can be 36%
within the Sun’s gravity well, then subtract velocity to go
lower. Very small objects which do not have structural
down Mars’ gravity well. On the graph, that means taking
flaws can even rotate faster than orbital velocity around
the vertical axis velocity change from Earth surface to the
them, producing regions where you cannot remain on the
top line, which is Solar System orbits (11.18), plus the
surface without mechanical aid.
horizontal segment to go from the Earth’s orbit to Mars’
orbit (2.3), plus the vertical change to go to the Martian
Perturbations surface (5.03). That gives a total mission velocity of 18.5
km/s, which has to be accounted for by various propulsion
Gravity forces extend to infinity. Therefore nearby large systems. To return to Earth, you then reverse the process.
bodies, such as the Moon and Sun for the Earth, also add The graph shows theoretical values (single impulse to es-
an acceleration component to the gravity of the Earth. cape). Real changes in velocity (delta-V) will be higher
This varies over time as their direction and distance because (1) maneuvers are not perfectly efficient, (2) or-
changes. The side of the Earth facing the Moon, for ex- bits are elliptical and inclined, and (3) propulsion systems
24 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

are not perfectly efficient in performing a given maneu-

ver. Various losses are measured by the difference be-
tween ideal velocity, the velocity you would reach in a
vacuum with no gravity well present, and the actual ve-
locity you reach in a given circumstance. So this chart is
not an exact method for mission planning. It is intended
to give a rough estimate as a starting point from which
more detailed planning can start.
Velocity Bands
There are two velocity regions on the vertical axis for each
planet or satellite. The lower sub-orbital region (thin
purple line) is where you have enough velocity to get off
the object, but not be in a stable orbit. Those orbits will
intersect the body again. So they can be used to travel
from point to point on the body’s surface, but not to stay
in motion for multiple orbits. The higher orbital band,
shown with a thicker line, indicates enough velocity for a
repeating orbit. The shape of the orbit matters too, but
for circular orbits the lowest point in this band is an orbit
just above the surface of the body, and the highest point
is an orbit just fast enough to escape from it’s gravity well.
Since gravity varies as the inverse square of distance, rela-
tively large velocity changes are needed for small altitude
changes near the surface of a body. Conversely, near es-
cape velocity, relatively small velocity changes can pro-
duce large changes in altitude, and at escape it produces a
theoretically infinite change. In reality there are multiple
gravity wells that overlap, so escape from Earth merely
places you in the larger Solar gravity well, and escape
from the Sun places you in the larger gravity well of the
Galaxy.
Solar Orbits
The top blue line represents orbits around the Sun away
from local gravity wells. The surface of the two largest as-
teroids, 4 Vesta (−0.35) and 1 Ceres (−0.51) are marked,
but the orbital bands for these two asteroids, and the en-
tire gravity well for most smaller asteroids, are too small
to show. Instead, the range of Solar velocities are shown
for Near Earth Objects and the Main Belt between Mars
and Jupiter. In reality the velocities of small objects in the
Solar System are spread across the entire chart. The two
marked ranges are just of particular interest. The surface
of Jupiter and the Sun, and their sub-orbital ranges are off
the scale of this chart because of their very deep gravity
wells.
2.5. ORBITAL MECHANICS 25

2.5.4 Powered Flight planet’s gravitational pull.

Boost From Rotating Body
Powered flight refers to trajectories and orbits which are
not only under the influence of gravity and other natural [To be added]
forces. Rather they are the result of natural forces plus
those created internally by an artificial system, or artifi-
Mass Ratio: Tsiolkovsky Rocket Equation
cially applied to a system from the outside. The com-
mon example of internally created forces is a chemical
For any rocket which expels part of its mass at high ve-
rocket engine. Examples of outside forces are a power-
locity to provide acceleration, the total change in velocity
ful laser directed at a lightweight sail, or a gun that uses
delta-v can be found from the exhaust velocity v(e) and
compressed gas to accelerate a projectile. The powered
the initial and final masses m(0) and m(1) by
part of a flight may last a short time, as when a rocket
launches into orbit. After that it coasts, only affected by
gravity and other natural forces. A solar-electric engine, m0
on the other hand, may operate over most or all of a flight ∆v = ve ln
m1
or mission.
The difference between the initial mass m(0) and the fi-
nal mass m(1) represents the propellant or reaction mass
Ascent Trajectories used. The ratio of the initial and final masses is called
the Mass Ratio. The final mass consists of the vehicle
Circular orbit velocity at the earth’s surface is 7910 me- hardware plus cargo mass. If the cargo mass is set to
ter/sec. At the equator the Earth rotates eastward at 465 zero, then a maximum delta-v is reached for the particu-
meters/sec. So in theory a transportation system has to lar technology, and missions that require more than this
provide the difference, or 7445 meters/sec. The Earth’s are impossible.
atmosphere causes losses that add to the theoretical ve-
locity increment for many space transportation methods.
The design problem is then to find the most efficient tra- Staging
jectory that minimizes losses.
In the case of chemical rockets, they normally fly straight A certain fraction of a vehicle’s loaded initial mass will
up initially, so as to spend the least amount of time in- be the vehicle’s own hardware. Therefore from the above
curring aerodynamic drag. The vertical velocity thus rocket equation there is a maximum velocity it can reach
achieved does not contribute to the circular orbit veloc- even with zero payload. When the required mission ve-
ity, since they are perpendicular. So an optimized ascent locity is near or above this point, dropping some of the
trajectory rather quickly pitches down from vertical to- empty vehicle hardware allows continued flight with a
wards the horizontal. Just enough climb is used to clear new mass ratio range based on the smaller hardware mass.
the atmosphere and minimize aerodynamic drag. This is known as Staging, and the components of the ve-
hicle are numbered in order of last use as first stage, sec-
The rocket consumes fuel to climb vertically and to over- ond stage, etc. Last use is mentioned because stages can
come drag, so it would achieve a higher final velocity in a operate in parallel, so the one to be dropped first gets the
drag and gravity free environment. The velocity it would lower stage number.
achieve under these conditions is called the 'ideal veloc-
ity'. It is this value that the propulsion system is designed The velocity to reach Earth orbit is approximately twice
to meet. The 'real velocity' is what the rocket actually has the exhaust velocity of the best liquid fuel mixes in use.
left after the drag and gravity effects. These are called So the rocket equation yields a mass ratio of e^2 or 7.39,
drag losses and gee losses respectively. A real rocket has and a final mass of 13.5%. This percentage is close to
to provide about 9000 meters/sec to reach orbit, so the the hardware mass of typical designs, so staging has com-
losses are about 1500 meters/sec, or a 20% penalty. monly been used with rockets going to Earth Orbit. We
desire a rocket with a number of stages that optimizes the
Boost From a Non-rotating Body economic efficiency (cost per payload unit mass). The
To go from a non-rotating body’s surface to orbit requires economic efficiency depends on a number of factors, the
that a rocket change its velocity from a rest velocity (zero) mass efficiency being only one factor.
to a velocity that will keep the payload in orbit. If our Let us assume that we desire to launch a payload of weight
rocket maintains a constant thrust during its ascent, then P. The weight of each stage in the stack is
the total velocity change is

∫ torbit ∫ torbit Wi = P w i
T D
a dt = − − g dt
0 0 m m where wi is a normalized weight for the stage. The total
where a is the acceleration, D is the drag, and g is the stack weight is thus
26 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

materials are generally called Engines or Rocket En-

( ) gines. Specific types each have a descriptive or or com-
∑
N
mon name noted later in the book where they are dis-
W =P 1+ wi
cussed. The forces generated by an expelled material are
n=1
derived from Newton’s Laws. Where F is force, dm/dt
The change of velocity per unit mass for each stage is is amount of mass expelled per time, and ve is exhaust
velocity, the force can be found by

∆vi = Ispi ln µi dm
F = ve
where µi is the ratio of the weight before the burn of dt
the ith stage to the weight after the burn of that stage. For accelerating vehicles, we call the force Thrust and
Thus, µi will always have a value greater than 1. The to- give it the symbol T. From this equation we can see the
tal change in velocity per unit mass for all the stages is possible ways to increase thrust are (1) increase the mass
then flow rate, (2) increase the exhaust velocity, or (3) some
combination of both. The available material to expel, also
∑N known as Reaction Mass or Propellant, is finite when
∆v = n=1 Ispi ln µi
coming from an internal source. So it is usually prefer-
able to increase velocity to get more total performance.
Because this type of propulsion system is self-contained,
2.6 Propulsive Forces it can operate in many environments, particularly the vac-
uum of space.
Space transport generally involves the application of
Exhaust Velocity
forces to get to a desired location. The comprehensive
listing of space transport methods found in Part 2 of this
book can be categorized in two dimensions, as shown
in the “Table of Space Transport Methods” at the start
of that section. The X (horizontal) dimension is how
the motive force is applied, and Y (vertical) dimension
is the source of the energy used. (See also Spacecraft
Propulsion at Wikipedia for another list of methods.)
This section will discuss the forces, and section 1.4 En-
ergy Sources will discuss the energy sources. Transport
methods can also be categorized by parameters like state
of development, exhaust velocity, or efficiency. Section
2.11 - Comparisons will consider these other parame-
ters. We list all known forces considered for transport,
without regard to practicality or state of development.
How to select the best option for a given transport task
will be covered later.
Propulsive forces can be divided into two large groups.
The first group develop reaction forces from internal ma-
terial which is expelled from the vehicle. The second
group are forces generated by interacting with an entity
outside the vehicle. The law of conservation of momen-
tum (i.e the sum of changes in mass times velocity for
the parts of a system is zero) requires that the force you
impart to the object you want to move is matched by an
equal and opposite force on something else. That some-
thing else can be mass you expel from the object, or an
outside entity, but in either case the combined forces must
be zero. Figure 1.3-1 - Space Shuttle solid booster exhaust emitted as a
cone of gas.

2.6.1 Reaction from Expelled Mass A parallel, unidirectional flow has all the mass expelled
in a single direction. An expanding gas is generally emit-
Vehicles which use this type of force production are gen- ted as a cone with some angular width (see Figure 1.3-1).
erally called Rockets, and the devices which expel the Molecules not moving parallel the axis of the cone only
2.6. PROPULSIVE FORCES 27

contribute the parallel component of their motion parallel well in a vacuum or a very low density non-conducting
to the reaction force. This component is found from the medium. Any appreciable outside pressure would stop
cosine of the angle of motion times the molecular veloc- the microparticles by drag or collision. Relative to ion
ity. Molecules in an expanding gas also have a range of engines, the charge to mass ratio is lower, so the same
velocities determined by their temperature and how the electrostatic voltages result in lower exhaust velocity and
flow was shaped. The average axial component over the less performance.
whole flow is called the Exhaust Velocity. A derived unit
is Specific Impulse. Where T is thrust, ṁp is propellant
flow rate, and g is standard Earth gravity it is defined as C. Gas Flow (0.1 - 10 km/s) While in theory you can
eject a liquid to obtain thrust, in almost all cases better
performance can be gotten using a gas. This is due to
T the higher average molecule velocity and ability to ex-
Isp =
ṁp g tract energy from the gas expansion. So liquid is skipped
The units for specific impulse are in seconds, and are in- among the solid-liquid-gas states of matter as a reaction
terpreted as how many seconds one unit of fuel can pro- method. However liquids are a useful as a way to store
duce one unit of thrust at 1 gravity. As an example, the reaction mass due to higher density and lower vapor pres-
high energy propellant combination of H2 +O2 produces sure. Higher pressures require more storage tank mass.
a specific impulse of about 450 seconds. Exhaust veloc- Gas flow includes ambient temperature gas such as the
ity in meters/second is is the preferred SI unit, since that Nitrogen “cold gas” thrusters used in spacesuit maneu-
is not Earth-centric by using the level of Earth gravity in vering backpacks. Cold gas thrusters are useful when
the calculation. you don't want to damage hardware with a hot or chemi-
In the list that follows, the approximate range of exhaust cally reactive exhaust plume, but they are very low perfor-
velocities is noted, and the list is generally in increasing mance (~0.5 km/s). Heating a low molecular weight gas
order. Note that what we mainly use today (combustion (i.e. Hydrogen) allows much better performance due to
gas) is among the lowest in performance. To achieve the higher average molecule velocity. At sufficiently high
higher mission velocities than the characteristic exhaust temperature (around 3000K) the Hydrogen molecules
velocity you can use large amounts of propellent, stack will start to decompose to individual atoms, further in-
multiple stages of the same method, or use multiple dif- creasing velocity. There are numerous possible methods
ferent methods. of heating the gas, including electric discharge through
the gas (arcjet), concentrated sunlight (solar-thermal),
electric filament heaters (resistojet), or heat from a nu-
A. Bulk Solids (0-10 km/s) Solid pellets or slugs are clear reactor (nuclear-thermal).
expelled via mechanical devices such as a rotary cen-
trifuge, or an electromagnetic accelerator such as the
Mass Driver Reaction Engine. The advantage is be- D. Combustion Gas (2 - 5 km/s) In this method, hot
ing able to use nearly any solid material as the reaction gas is generated by chemical reactions in the propellant.
mass. One disadvantage is the relatively low exhaust ve- For rockets the hot gas is expelled via a supersonic ex-
locity compared to Ion and Plasma engines. The extra pansion exit nozzle. That type of nozzle produces the
work in extracting a suitable fuel for the latter types is highest velocity in a narrow stream. Performance is lim-
usually much less than the gain from using 5-10 times ited by the reaction energy in the propellant, which is a
less reaction mass. Another is the creation of a debris maximum of about 15 MJ/kg for non-exotic fuel combi-
impact hazard by emitting large numbers of uncontrolled nations. This provides a maximum of 5.5 km/s exhaust
objects. A centrifuge or mass driver launching bulk mass velocity in theory, and 4.5 km/s in practice. For atmo-
from a body which is collected in orbit does not create the spheric jet engines, some of the energy in the gas is used
same impact hazard, even though it uses the same kind of to drive a turbine and bypass fan blades, which greatly in-
devices. Mechanical devices are like centrifuges are lim- creases the affected mass flow. The remainder is expelled
ited to fairly low velocity, and have not generally been by a nozzle, but the nozzle is typically a simpler geome-
considered for space transport. try. The use of external oxygen extracted from the air
flow, and bypass air for mass flow, dramatically reduces
the rate of fuel use, but also limits the flight velocity due
B. Microparticles (0 - 4 km/s) In addition to us- to drag and heating.
ing bulk solid, finely powdered solid microparticles or
droplets such as from inkjet type devices may be acceler-
ated by electrostatic forces after giving them an electrical
charge. Advantages of this method are using unprocessed
rock dust or single fluids, and enabling very small engines.
Compared to bulk solids, microparticles or droplets are
less of an impact hazard, though they may pose a con- Figure 1.3-2 - Hybrid rocket concept with liquid Oxidizer (blue)
tamination problem. Disadvantages include only working and solid fuel grain (green).
28 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

There are several types of combustion gas rockets. A be kept low enough that the engine can disperse the heat
Monopropellant has a single ingredient which is decom- gained. There is no theoretical limit to how hot a plasma
posed and heated by passing over a catalyst bed. A Bi- can be. Since the kinetic energy of the exhaust increases
Propellant has two ingredients, a fuel and an oxidizer, as the square of the velocity, there is a practical limit
which are generally mixed and burned in a Combustion based on how much energy you can supply. Plasmas can
Chamber. In a Liquid Rocket the two ingredients are be physically confined by magnetic fields, but they also
stored in liquid form in separate propellant tanks, al- emit light that escapes to their surroundings. The surface
though one or both may be converted to gas before reach- of the Sun, for example, is a 5780 K plasma. So heating
ing the combustion chamber. In a Hybrid Rocket one of of the engine components will also limit a practical de-
the ingredients is in solid form (usually the fuel), and the sign. Many plasma confinement techniques derive from
other in liquid form. In a Solid Rocket all the ingredi- fusion research, which deals with extremely hot plasmas.
ents are in a finely mixed powder which has been cast into Recent development of superconducting coils improves
a solid form. A typical solid rocket formulation has an their efficiency, so design of effective plasma engines is
oxidizer like Ammonium Perchlorate (NH4 ClO4 ), and fairly recent. Because their exhaust velocities can be ten
a complex fuel containing powdered aluminum, rubber, or more times higher than for combustion gas, plasma en-
and epoxy, which both binds all the ingredients together gines are in active development, but have not reached op-
and is part of the fuel being burned. erational use as of 2016.
There are a large number of combinations of form and
fuel mixtures for rockets, but only a few are used with
neutralizing electron gun
any frequency. Because of the high thrust-to-mass ra-
magnets
tio and ability to work in many environments, combus-
tion gas systems have been by far the most popular for
space transport so far, and almost exclusively the one for
launch to Earth orbit. Their disadvantage is that the ex- electron
haust velocity of the best propellants is about half the nec- gun

essary velocity, including various losses, to reach Earth

orbit. Solving the Tsiolkovsky Rocket Equation (section
1.2), we find the mass ratio is at least 7.4, which leaves negative grid
only 13.5% of liftoff mass for the vehicle and payload. electron positive grid

The best propellant includes liquid Hydrogen as the fuel, neutral propellant atom

positive ion
which is very low density, so keeping the vehicle mass low
enough for reasonable payload is difficult. If a denser and
Figure 1.3-4 - Electrostatic ion engine.
lower performance fuel is chosen, less structural mass is
needed for propellant tanks, but the mass ratio is higher.
This still leaves small payload mass. Until now, the so- F. Ions (2 - 200 km/s) Ions are atoms from which one
lution has mainly been to use the rocket once, which al- or more electrons have been removed. In an Ion Engine
lows lighter hardware, and discarding part of the hard- the propellant is first ionized using electron bombard-
ware once some of the propellant has been used (staging). ment or RF oscillations (Figure 1.3-4). The positive ions,
This is an expensive solution. For the future, a better ap- which represent nearly all the mass, are accelerated across
proach is to move away from chemical rockets and their a voltage gradient to high velocity. The voltage gradi-
limited fuel energy in relation to what is needed to leave ent can be a set of metal screens, or a charged plasma.
Earth. That will allow more of the vehicle mass to be To maintain overall charge balance across the vehicle, an
hardware and payload, and less to be fuel. Designs can electron gun separately emits negative charges, or the ions
then be more robust and used many times, and the cargo recombine in the charged plasma. Once outside the en-
can be a larger fraction of liftoff mass. The combination gine the combined emission is electrically neutral so that
will drastically reduce costs. the vehicle does not accumulate a net charge. Compared
to a plasma, an ion beam is generally lower density, so re-
quires lower power levels to operate and generates lower
E. Plasma (5 - 200 km/s) In this category the propel- thrust. The main difference is creating high ion velocity
lant is heated to the point that the atoms disassociate into via electric forces rather than heating. Efficiency in con-
charged components (ions and electrons), then directed verting power to thrust is similar (60-75%), as is maxi-
out of the Plasma Engine with magnetic fields. Heating mum exhaust velocity. Like plasma engines, the energy
can be accomplished by a vigorous electric discharge, in- required rises as the square of exhaust velocity. Practical
tense microwaves, laser, or internal heating in a fusion limits on the power source then limits performance. Be-
plasma. Plasmas are hot enough to melt most materi- cause ion engines can operate with less overhead in small
als, including the ones used for the rest of the engine, sizes than plasma engines, and are derived from vacuum
so they are usually contained by a magnetic field. Al- tube technology, they were developed earlier.
ternately the plasma density (and hence the thrust) can Ion engines are in use, as of 2016, on many communica-
2.6. PROPULSIVE FORCES 29

tions satellites, and a few planetary spacecraft where high Unbalanced photon emission as an incidental effect has
total mission velocity is needed. In the future we expect been noted for the Pioneer spacecraft and with respect
their use to increase, along with the larger plasma engines, to small asteroids, where it is called the Yarkovsky Ef-
because they use an order of magnitude less propellant fect. However these accelerations are very small and not
than chemical rockets. In addition, the last 20 years has intentional. For useful propulsion a very large amount of
seen the development of lightweight and efficient solar photon energy needs to be emitted in a chosen direction.
panels to provide power for them. Their disadvantage is
low thrust, so they cannot be used on their own for launch
or landing on large bodies. This can be overcome by com- Discussion
bining them with other methods.
Split Energy Source:
G. Atomic Particles (1000 km/s to near c) Atomic Many of the above listed propulsion methods have sepa-
particles include nuclei of atoms from which all electrons rate propellant mass and energy to accelerate that mass.
have been removed, single electrons, neutrons, or protons, With these methods there is a design trade between mass
or more exotic particles such as muons. Whereas an ion and energy. The equation above gives force as mass flow
engine typically uses a single voltage gradient, a parti- rate times exhaust velocity. To get a desired mission ve-
cle accelerator has multiple chambers that add successive locity change, or ∆V , you can minimize the mass used,
amounts of energy to the particles. This enables exhaust and use a lot of energy to generate a high exhaust veloc-
velocities up to near the speed of light, c , or 299,792 ity. Alternately you can minimize energy and accelerate a
km/s. Another method is direct emission of atomic par- larger propellant mass. It is a trade-off where using more
ticles from fission decay, fusion reactions, or antimatter of one requires less of the other. The kinetic energy re-
decay. These have particle velocities that are a significant quired for a given exhaust velocity is KE = mv2 /2, so it in-
fraction of c. Charged particles (generally those besides creases faster than the mass flow is reduced. Typically en-
neutrons) can be directed by magnetic or electric fields, gine efficiency is relatively constant in the normal operat-
while neutral particles must be allowed to leave in one di- ing range. So source power level, which is KE/efficiency,
rection and absorbed in the others to produce net thrust. also tends to increase as the square of exhaust velocity.
Although accelerators on Earth have reached very close Higher power levels require more power source mass, and
to c, which is the universal speed limit, no space mission represent overhead beyond the cargo you are delivering.
has yet required this high a performance. So this reaction Whether that mass increases linearly with power level or
method remains theoretical at present. not will affect the optimum exhaust velocity. In addition,
In this category is the Bussard Ramjet concept, which is the choice of completing the mission in the least time,
a kind of fusion rocket. Rather than carry all the fuel on- least propellant used, or somewhere between those ex-
board, it uses a huge electromagnetic field as a ram scoop tremes will affect the optimum exhaust velocity to use.
to collect and compress hydrogen from the interstellar Once a design has been set for a vehicle, there is addi-
medium. The field compresses the hydrogen until fusion tional optimization possible within a particular mission.
occurs, then directs the energy as rocket exhaust, acceler- The design and mission fix the quantities of propellant
ating the vessel. The difference between the incoming in- mass and mission ∆V . If the engine type allows vari-
terstellar gas pressure and outgoing exhaust provides the able exhaust velocity, it takes less total energy to “dump”
thrust. The Bussard Ramjet currently has conceptual dif- the early exhaust mass at a lower exhaust velocity Ve , and
ficulties and is very far from being used. then eject the later mass with a higher Ve , than to eject at
a constant Ve . If the goal is minimum time and you have
H. Photon Emission (1.0c exactly) Photon momen- a fixed energy source, this will be the optimum operating
tum is calculated by E/c, where E is the energy and c is profile.
the speed of light. Thrust is then P/c where P is the power For missions which use solar power as a source, the op-
of the emission. Direct emission of photons, while low timal operation is more complex. First, solar flux varies
in thrust, has the highest possible exhaust velocity. For as the inverse square of distance from the Sun. Second,
practical use, an extremely high energy source needs to solar powered orbit changes around a large body intro-
be used, such as fusion or antimatter decay. Fairly sim- duces shadowing and a stronger gravity field, which takes
ple thermal (black body) emission and reflector arrange- longer to traverse. For low thrust engines, short
√ thrust in-
ments can align the light beam to produce useful thrust. A tervals require less ∆V by up to a factor of 2 , but close
laser aligns the light very accurately, but the gain is small orbits are in shadow up to half the time and limit when
compared to a light beam with a width of a few degrees. you can generate thrust. When the local gravity accel-
The momentum contribution for an off-axis photon is co- eration is low relative to the vehicle acceleration, which
sine(a), where a is the off axis angle. For small angles that happens farther from the body, then short thrust intervals
is very close to 1.0. Like atomic particles, no space trans- do not impose a large time penalty. Deeper in a grav-
port mission has needed this high a performance, so it is ity well, they do. Lastly, for bodies with radiation belts,
a theoretical method at present. like Earth and Jupiter, time spent climbing through those
30 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

belts can damage solar arrays, other hardware, and hu- but in this group the measure is effective velocity change
man passengers. In complex cases like these, the best rather than exhaust velocity.
thrust plan is found using a numerical simulation which
divides the mission into small time steps and adjusts the
variables looking for the best result for whatever param- I. Mechanical Traction (5 km/s) This method uses
eters the mission planner wants to optimize (time, fuel mechanical forces applied to a vehicle or payload from
used, radiation exposure, or others). an external source. Applications of this method include:

Combined Energy Source:

• Using a cable or net to capture and decelerate the
Other propulsion methods listed above, such as combus- vehicle relative to a destination.
tion gas rockets, have propellants which provide both the
mass flow and energy supply to generate an exhaust ve- • An elevator to climb a tower or cable at a constant
locity, Ve . In these cases the Tsiolkovsky rocket equa- velocity.
tion shows that higher velocity is always better. Com-
• A tow cable between two vehicles or between a fixed
bustion engines have been developed with extraordinary
installation and a vehicle, for acceleration or lift.
power/mass ratios, up to 2.7 MW/kg. This enables them
to lift not only themselves, but the entire vehicle’s mass • A rotating structure or cable to provide radial or an-
against the Earth’s gravity. This makes possible launch gular acceleration.
to orbit without external assistance. Their performance
is limited, however, by the available energy in the fuel, Mechanical devices transmit forces by atomic bonds in
which in the case of Earth is about half that needed to the materials used. They are limited by the strength of
reach orbit. Atmospheric engines obtain much of the the bonds, and so the bulk material strength, to around
mass flow and energy from outside the vehicle, so they by- 3-5 km/s for existing materials. The theoretical strength
pass the limits of a purely internal propellant. This usually limits for carbon nanotubes (100 GPa tension, per Zhou,
comes at the expense of increased mass from items like Ultimate Strength of Carbon Nanotubes, Phys. Rev.
inlets, turbines, and wings. This lowers the power/mass B, v 65 p144105, 2002) and diamond (90 GPa compres-
ratio, and of course only functions while you are still in sion, per Telling Theoretical Strength and Cleavage
the atmosphere. The choice to use atmospheric engines of Diamond Phys. Rev. Lett. v84 n22 p5160, 2000),
will therefore depend on many factors beyond propellant given a density of 3500 kg/m^3 and a design margin of
energy. Atmospheric engines can have as much as 10- 2.8, produce a theoretical velocity of around 12 km/s.
20x lower fuel use for a given thrust, so that is a strong However, real materials accumulate defects, even if at an
incentive despite the complications. atomic scale. So the usable strength in practice is much
Comparing split vs combined energy sources, a separate lower than the theoretical value. A large advantage for
energy source relieves the limits on energy per propel- mechanical systems is their inherent ability for repetitive
lant mass used. This allows higher exhaust velocities use. This divides the initial cost by the number of uses
and lower propellant use, often by an order of magni- over the economic life of the system.
tude or more. The drawback is current technology for
these methods usually generates much less than one Earth
gravity, because energy sources like solar panels are be- J. Friction (2 km/s) This method uses frictional forces
low 200 W/kg, far below combustion engines. This limits against a solid surface to lower relative velocity. If your
their use for the important job of getting from the ground intent is to slow down on, for example, the Moon, you can
to Earth orbit. Once in orbit, their performance advan- have a flat paved runway or a raised rail, and simply use
tage will often make them the preferred option. This is mechanical braking against the runway or rail to stop. To
because the best chemical propellants contain about 15 accelerate to orbit, you can grasp a trailing cable or rail
MJ/kg of energy. A 175 W/kg space solar panel (state on an orbiting platform and apply friction to gain velocity.
of the art as of 2016) will produce that much energy in This method has the virtue of being simple. Space veloc-
about a day. Since the panel typically lasts 15 years, it ities, however, often represent more kinetic energy than
will produce about 5000 times more total energy over its it takes to heat and melt materials. So the friction forces
life. must be distributed in time, rate, or total amount to pre-
vent overheating. It is therefore more useful for planets or
satellites with low orbital velocities, and for small velocity
changes as part of a larger total.
2.6.2 External Interaction

The second major group of propulsive forces are applied K. Gas Pressure (6 km/s) This method uses external
by or against some external object or field, rather than ex- gas pressure differences to apply forces to a vehicle. It
pelling some material from the vehicle. Since no reaction includes all types of guns, where the gas is confined in
mass is consumed, the Tsiolkovsky rocket equation does a tube to accelerate linearly. Gas expansion is limited
not apply. We list them roughly in order of performance, by the temperature and molecular weight of the gas. Hot
2.6. PROPULSIVE FORCES 31

Hydrogen then yields the best performance, which is lim- speed of light. This is nearly twice the force from emis-
ited to around 9 km/s in the best case, and about 6 km/s sion, because the photon changes velocity by twice the
for practical designs. There are a variety of methods to speed of light from forward to backward. It is slightly less
generate hot gases. The oldest, conventional firearms, than twice because no reflector is 100% efficient. When
depend on fast combustion of a solid propellant grain. the reflection is not perpendicular, the force is reduced
Newer versions function via combustors , fuel-air detona- by off-axis cosine losses. The quantity of light (photons)
tion, particle bed heaters, or other methods to create heat is not limited by an internal power source, so can reach
and pressure in a short time. When the hot gas so created relatively higher forces than emission, but still generally
is not sufficient, a two-stage method transfers the energy low compared to other methods. A solar or light sail
by means of a piston to a lighter gas (usually Hydrogen). uses thin but very large reflectors in order to intercept the
A properly tuned piston mass can result in higher tem- maximum amount of light. To get the highest accelera-
peratures and pressures in the light gas than in the first tions, the maximum ratio of reflected energy to mass is
stage. desired. Tungsten may be a dense element, but can oper-
Various kinds of guns have been used for high velocity ate at much higher temperature close to the Sun where the
light intensity is higher. Where maximum temperature
research since about the 1960’s. Most have been used in-
doors, but at least two have seen outdoor use. They have is not required, a light alloy like Magnesium-Aluminum,
not been used for space launch yet, but that is a matter with high reflectivity, is preferred.
of size and location, rather than technology level. An ex- Given a sufficiently powerful light source, quite high ve-
treme use of gas pressure is to use a fission or fusion de- locities are possible. Using natural light sources such as
vice to heat a large amount of gas to plasma temperatures the Sun we get approximately 8.2 microNewtons/m2 at
in a chamber, which then is applied to accelerate a vehi- the Earth’s distance. The gravitational force from the Sun
cle. This concept remains theoretical because of bans on on a lightweight sail may be 0.35 times this. The Light-
testing and the tendency to destroy the chamber. ness Ratio, LR, or ratio of light pressure to gravity then
determines the maximum escape velocity by
L. Aerodynamic Forces (4 km/s) These forces include
lift, buoyancy, and drag. Wings and fan blades develop
lift by pressure difference across the upper and lower sur- Vmax = (LR − 1)Ve
faces when a fluid such as air flows across them. At higher
where V(e) is the local escape velocity. So paradoxi-
velocities, a WaveRider type inclined surface rides the
cally, to reach maximum final velocity you want to start
lower shock wave it generates to create lift. Buoyancy
as close to the Sun as possible, where local escape ve-
develops lift by having a lower density than the surround-
locity is higher. For a lightweight sail, this might be 0.2
ing fluid, such as in a balloon. Drag generates forces by
AU, limited by the melting point, and an escape veloc-
accelerating the surrounding fluid in the direction you are
ity of 100 km/s. Thus the maximum final velocity is 185
moving, thus producing a force opposite your motion. A
km/s. Any velocity less than this maximum is possible.
parachute, for example, is shaped to capture and accel-
The practical limits using artificial light sources such as a
erate the maximum amount of air, producing the maxi-
laser are unknown, since no lasers with high enough sus-
mum amount of drag to slow you down. Lifting forces
tained power to be useful for this purpose exist. In theory
invariably create drag, and tend to create more at higher
a powerful enough laser can accelerate a sail to substan-
velocities. Heating also becomes a limiting factor at high
tial fractions of the speed of light, before accounting for
velocity, tending to set a limit of about 4 km/s for aerody-
drag from the interstellar medium.
namic forces, except for re-entry systems where the heat-
ing is efficiently dissipated. Since aerodynamic forces in By tilting a sail from perpendicular to the light source,
themselves always create friction and other dissipation, off-axis forces can be generated. The net force will be
by themselves they can only slow a vehicle relative to the approximately perpendicular to the sail and allow inward
fluid. They require a source of thrust, like a jet engine, to spiral orbits by directing the force against the orbital ve-
accelerate. locity, or tilting of the orbit plane by directing the reflec-
tion perpendicular to velocity. Sub-orbital or zero veloc-
ity motion can be produced by balancing light pressure
M. Photon Reflection (185 km/s) Instead of an inter- against local gravity. These are not orbits because they
nal source, as in H. Photon Emission above, this method depend on constant forces. Rather they fall into the class
uses an external source of photons, such as a star or laser. of powered trajectories.
Normal (perpendicular) reflection produces a force of

N. Particle Deflection (100 km/s) The thin plasma

E(1 + R) and gas emitted by the Sun’s heated outer layers is called
F =
c the Solar wind. The wind has typical velocities of
where E is the incident energy on the reflector, R is the re- 400-750 km/s relative to the Sun, and extends outwards
flectivity (fraction of incident light reflected), and c is the to about 125 AU, where it encounters the interstellar
32 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

medium. The force per area due to light pressure is much gravity assists take extra time to complete, and restrict the
greater than that due to solar wind flux. A Magnetic choice of trajectories because you can't choose where the
Sail is a proposed method of spacecraft propulsion. It moon or planet you are using will be at a given time.
would use a magnetic field to deflect charged particles in Gravity fields extend to infinity, so every place in the Uni-
the plasma wind. Since the field itself is non-material, in verse has a field, and almost everywhere has a non-zero
theory it can be large enough to get useful thrust, even net field. So all space transport has to account for grav-
though the area density of the solar wind is very low. By ity forces in mission planning. Usually it is an obstacle
tilting the field, a sideways force can be generated. This to overcome, in reaching orbit or changing orbits, but in
method is limited to the velocity of the wind, and gener-
some cases, such as gravity assist, it can be used to advan-
ally to producing forces away from the Sun. An artificial tage. Gravity changes as the inverse square of distance
beam of particles could also, in theory, be used to apply
from an object. If your vehicle or object is elongated,
a force to a vehicle, which either absorbs or deflects the the Gravity Gradient, or difference in gravity between
beam. This remains theoretical because powerful enough
the lower and upper ends, can also be used as a torque to
beams are not available, and keeping them focused over stabilize your orientation. These gravity gradients exist
typical distances in space is difficult.
whether or not you use them, so they must also be taken
into account as a force to be overcome in system design.
O. Magnetic Field (20 km/s) This method gener-
ates a force by using a current carrying wire, coil, or
magnet to react against other magnetic fields (natural Discussion
or man-made). In this category fall magnetic levita-
tion or Maglev; Coilguns, which use a series of timed Compared to the expelled mass group, external interac-
coils; Railguns, which use two high current rails and a tions do not consume a finite supply of reaction mass. So
plasma short across them; and Electrodynamic Propul- these methods would be preferred, all other factors be-
sion, which reacts against a natural magnetic field. It can ing equal. Reaction mass, though, can be expelled under
be used both for net thrust and drag, or for torque forces to a wider range of circumstances in most cases, and usu-
rotate a vehicle. Magnetic flywheels are often used to ori- ally with a higher thrust/mass ratio. So it is not possible
ent or rotate spacecraft. The Sun and some planets have to state a general conclusion as to which approach is bet-
a natural magnetic field to react against. Many artificial ter. The choice would depend on a variety of detailed
satellites also use the magnetic field to rotate the satellite circumstances, including destination, trajectory, desired
to a desired orientation using a Magnetorquer. For ex- trip time, cargo mass, frequency of trips, and how far in
ample, see Galysh et al, CubeSat: Developing a Stan- the future the transport occurs. The latter affects what
dard Bus for Picosatellites. In theory there is no limit technologies are available and how ready for use they are.
to the velocity you can reach using magnetic fields. In One approach to selection is to look at the external in-
practice, the natural fields such as the strong one around teraction group first, to see if any can be applied to the
Jupiter, or the practical scale or field strength of an artifi- job at hand, then look at the group that uses up reaction
cial accelerator, limit velocity changes to around 20 km/s. mass second, since that requires overhead above whatever
cargo you are trying to deliver.

P. Gravity Field (20 km/s) Gravity forces accelerate

objects towards any nearby mass. Normally this results in
an orbit or simply falling towards an object. A Gravity
2.7 Energy Sources
Assist is purposely choosing a hyperbolic path to change
the direction, but not the total amount, of your velocity 2.7.1 Energy in General
relative to a given object. The maximum change is twice
the escape velocity of the object, when the direction is All of civilization requires energy to function, including
changed 180 degrees. To reach a desired mission desti- about 8 MJ/day (~2000 Calories) of food energy per per-
nation, usually much less than this can be done. Since a son. Space systems are no exception, They require energy
planet, for example, is moving with respect to the Sun, both for propulsion and for other systems like life support,
changing direction relative to the planet can change the computers, and materials processing. The general field
total velocity relative to the Sun. By conservation of mo- of energy is vast, involving many fields of engineering.
mentum, the planet also must change velocity, but since Energy for space projects involves an equally wide range.
it is much more massive, the velocity change is small So we can only provide an introduction, and supply refer-
enough to ignore in most cases. Within the Solar Sys- ences for further study. In this section we survey the range
tem, the achievable velocity changes are about 20 km/s, of possible energy sources for all types of space systems.
and may need multiple gravity assists to reach this level. Since the book is oriented to future space projects, we list
The major advantage of gravity assist is it does not need a number of sources that are not developed yet, but are
internal propulsion except to line up the flyby, and there- possible according to known physics. These sources can
fore saves propellant mass. A disadvantage is one or more then be used for the propulsion methods tabulated in Part
2.7. ENERGY SOURCES 33

2, and for other engineering purposes described in Part 3 on more specific aspects of energy systems. Wikipedia
and later. By making a two dimensional table of energy also has an Outline of Energy article, with many links.
sources vs propulsive forces, we can categorize all possi- Some of the concepts listed below are currently theoreti-
ble propulsion methods, and we do this at the start of Part cal, so they are not well covered in reference books about
2 of the book. We have not yet developed a similar table current energy use or engineering. Information about
to neatly categorize the other systems in a space project. them will mostly be found in research reports and sci-
Proximate vs Ultimate Energy entific/technical papers.

The principle of Conservation of Energy states that en-

ergy can neither be created nor destroyed, merely trans- 2.7.2 Mechanical Sources
formed from one kind to another. So all energy for
a project must come from a preexisting source. For a Mechanical Energy includes energy stored by virtue of
given project you can distinguish a proximate energy previous work, as in compressed gases, and that which ex-
source, which is in the form consumed by the project, and ists by virtue of position (potential) and motion (kinetic).
an ultimate energy source, which traces the proximate Objects in orbital motion have a combination of potential
source back through previous transformations to it’s orig- and kinetic energy.
inal form. Ultimately, all energy traces back to the cre-
ation of the Universe, but for engineering purposes we
A. Compressed Gas Although Pressure vessels are
rarely go back that far, and are generally concerned with
strictly an energy storage method, for space missions the
proximate sources.
tanks are normally pre-filled. So they operate as a prox-
For current space systems the energy is typically stored imate energy source in flight. The available energy, W,
internally as chemical energy in the case of launch vehi- stored in a pressurized tank, can be found from
cles, or uses sunlight in the case of satellites. In the fu-
ture, energy needs are likely to change, and the sources of
energy will also change. Permanent locations, such as a W = pB vB ln pA
large orbiting habitat or surface base, generally need con- pB
tinuous sources of energy to operate. Devices like batter- where B represents the high pressure and A represents
ies become unwieldy at that scale to supply power for the the low pressure, and p and v are pressure and volume re-
night part of Earth orbit or the two week Lunar night. Fu- spectively. So a 1 cubic meter tank with a high pressure
ture projects may also need much greater power levels for of 20 MPa and a low pressure of 10 Mpa would provide
tasks like processing of local materials. So the following 13.8 MJ of available energy. Compressed gas is a low
headings attempt to include all potential energy sources, density energy storage method. It is often used in space
including many not yet used, but which may become use- vehicles for tasks like cold gas thrusters and pressurizing
ful in the future space projects. We list them all so that liquid fuel tanks. Its chief advantages are simplicity, re-
designers know the full range of possibilities, from which quiring just a storage tank and a valve, and rapid release
they can then select viable options for a given task. We of the stored energy. When larger total amounts of energy
exclude sources such as human and animal power from are needed, a higher density but more complex system is
consideration here, both due to their low power levels, often preferred.
and because living things are not subject to the same kind
of engineering design as we apply to non-living system.
We also exclude sources like wind and geothermal, which B. Potential Energy Potential Energy is the ability of
mostly apply to Earth. Last, we includes some energy a system to do work by virtue of it’s position or configu-
storage methods, which are not strictly sources. Energy ration. In space projects this is usually position relative to
storage, however, is often a necessary and important part the gravity well of a massive object such as a planet. A
of system design. simple hypothetical example is a stationary space eleva-
tor cable. While raising a cargo, electricity is converted
Energy References
to potential energy of height. When lowering a cargo, the
A starting point for understanding energy sources in gen- potential energy can be extracted back to electricity. The
eral, not just as they apply to space projects, is a National formula for potential energy U was given in Section 1.1
Academies book America’s Energy Future, 2009. - Physics as
About 150 other books are available for free download in
the Energy and Conservation topic from the same site.
Encyclopedic references on the topic of energy, include m1 M2
U = −G
the Encyclopedia of Energy, Encyclopedia of Energy r
Engineering and Technology, and the Macmillan En-
The difference in energy at two radii gives the amount
cyclopedia of Energy. These are often expensive refer-
of potential energy stored or released over the distance.
ence books, so library or other sources are recommended
For small changes in radius (height) relative to the dis-
to get access. There are numerous engineering books
tance r, the potential difference can be approximated by
34 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

the average gravitational force (weight) times height. On 2.7.3 Chemical Sources
planetary surfaces, large amounts of available mass can
be used to store potential energy. On Earth this is done Chemical sources are an arrangement of atoms in a higher
with dams for hydroelectric power. On other bodies, a energy state which are converted to a lower energy state
mountain and a pile of rocks can serve the same purpose. by a Chemical Reaction, releasing the difference. Com-
Transporting the rocks up and down the mountain can bustion is the most common way to release this energy,
serve to store or release energy. accounting for over 80% of total human energy use. Bat-
teries are characterized by a reversible reaction, so that
the same device can store and release energy multiple
C. Kinetic Energy Kinetic Energy is that which an
times.
object possesses by virtue of it’s motion. Its formula was
given in Section 1.1 - Physics as
D. Fuel-Atmosphere Combustion The Earth’s atmo-
sphere, neglecting the variable amounts of water vapor,
KE = 12 mv 2 = F d contains 20.95% Oxygen (O2 ) molecules, which react
where KE is the kinetic energy, m is the object’s mass, with many other compounds to release energy. This
and v is the velocity. It is also equal to an accelerating Oxygen is the byproduct of Photosynthesis in living
force F times the distance d it is applied over. An object things. In the case of aircraft, a hydrocarbon fuel such as
in orbit has both kinetic energy in it’s orbital velocity and Kerosene is reacted with the atmospheric Oxygen in an
potential energy in it’s altitude. In an elliptical orbit, it engine. Since only the fuel is carried internally to the ve-
continuously exchanges altitude for velocity. So it also hicle, the energy released, about 43 MJ/kg, is about three
exchanges potential and kinetic energy, but the combined times as much as when both ingredients are carried inter-
total stays the same. nally, such as in a typical rocket. Large amounts of Oxy-
gen in an atmosphere is unstable, because it is so chemi-
Rotating objects such as a space station or reaction fly- cally reactive. It only exists on Earth in this form because
wheel have a form of kinetic energy in it’s motion around plants constantly produce it. So this energy source is not
an axis. Rotational energy is E(k) calculated by available on other bodies. The reverse option is available
on a body such as Titan, which has a hydrocarbon atmo-
sphere. In that case, Oxygen can be the carried ingre-
1 2
Ek = Iω dient, and burned with the surrounding atmosphere. For
2 atmospheres which are mostly CO2 (Venus and Mars),
Where ω is the Angular Velocity in units of radi- which is an end product of combustion, or bodies with no
ans/second, and I is the Moment of Inertia of the mass atmosphere at all, this energy source is not available.
about the center of rotation. The moment of inertia is
the measure of resistance to Torque, or rotational force,
applied on a spinning object. The higher the moment of E. Fuel-Oxidizer Combustion The energy source in
inertia, the slower it will spin when a given force is ap- conventional rockets is Combustion, where both the fuel
plied. Moment of inertia depends on the distribution of and oxidizer are supplied from internal sources. The in-
mass in the rotating object. The farther out a given por- gredients with the highest reaction energy, Hydrogen and
tion of the mass is, the larger the contribution. Formulas Oxygen, provide 15 MJ/kg of propellant. Although lower
for many shapes are found in the List of Moments of in Specific Energy than D. Fuel-Atmosphere Combus-
Inertia. For complex shapes, the total moment can be tion, it is not restricted to operating in the Earth’s atmo-
found by dividing it into simpler parts and summing the sphere. Liquid rocket engines also have extraordinary
individual moments. power-to-mass ratios. This enables launch trajectories
from large bodies like planets. Combustion can also be
Some examples of moment of inertia formulas are:
used as a secondary power source in Auxiliary Power
Units. Because the rate of energy release is very high,
1 2 combustion is useful when high power levels are needed.
I= mr The efficiency of combustion engines is typically 1/3 to
2
2/3, so other options may be preferred when that is an
I = mr2 important factor.
1
I = m(rexternal 2 + rinternal 2 )
2
Kinetic energy can be exchanged for other forms of en- F. Chemical Battery An Electric Battery is a device
ergy by gravitational forces, as in gravity assist maneu- which converts stored chemical energy to electricity, and
vers; or electromagnetic forces, as in many electrome- in a storage battery also reverses the reaction. Common
chanical devices. Potential energy can also be exchanged examples of chemical batteries include the Lead-Acid
for kinetic energy via the Oberth effect by expending type used in automobiles, and the Lithium-Ion type used
propellant deep in a gravity well. in many portable devices. Depending on battery type,
2.7. ENERGY SOURCES 35

they generally store less then 1 MJ/kg, considerably less ference between those and surface temperatures.
than combustion. The ability to cycle energy in and out
multiple times can outweigh the lower energy density. An
H. Concentrated Light A number of industrial tasks
example is the International Space Station, where large
require heating, which is easily done in space by con-
batteries supply power in the shadowed part of its orbit.
centrating sunlight. Examples are heating of raw mate-
A Fuel Cell is a type of battery where the reactants are rials to extract volatiles, or maintaining temperature and
stored in external tanks, rather than in a sealed battery growing ability in a Mars greenhouse. The concentra-
case. It can have high specific energy because the tanks tion ratio determines the maximum Black Body temper-
are lightweight compared to electrolyte solutions. Fuel ature that can be reached, up to the temperature of the
cells have therefore been used in space projects, such as light source. In the case of the Sun, the upper limit is
the Space Shuttle Orbiter. A Hydrogen-Oxygen fuel cell the Sun’s surface temperature, 5,775 K, less reflection
combined with an Electrolysis unit to convert the result- losses and radiation losses from the object you are heat-
ing water back to Hydrogen and Oxygen can supply en- ing. Since Tantalum-Hafnium-Carbide, the highest melt-
ergy storage with a specific in the range of 3-10 MJ/kg. ing point substance known, melts at 4200 K, concentrated
Sealed batteries are simple and reliable, and can be made sunlight should be sufficient for most industrial processes.
in very small sizes. Fuel cells have higher specific ener-
For space transport, a reaction mass can also be heated
gies, but are more complicated devices, since they need
by concentrated solar or artificial light. Lighter molecules
valves and a way to pump and store the various chemicals.
can be used than the exhaust products of chemical reac-
tions, so higher performance can be reached. Lack of
2.7.4 Thermal Sources powerful enough lasers limits their use for propulsion at
present, but sunlight is widely available in space.
Thermal Energy is the internal energy of a system due to
its temperature. It comes from the kinetic and vibration 2.7.5 Electrical Sources
energies, and the attractive potentials of the molecules or
other particles making up the system. Thermal energy Electricity is the set of phenomena associated with the
can be stored for later use, or added from an outside en- presence and flow of electric charge. Common examples
ergy source for immediate use. We sense high temper- are Electric Current, in the form of electrons moving
atures as heat, and even higher temperatures as visible in a conducting wire, and Lightning, a powerful elec-
light. Energy naturally flows from higher to lower tem- trostatic discharge through a plasma channel in a storm.
perature areas by conduction, radiation, and convection. Electricity is a very versatile energy source because it can
When a temperature difference exists, some of the ther- be converted to other forms efficiently, controlled in both
mal energy can be converted to other forms and used. An tiny and large amounts, and moved about from place to
example is a steam turbine that generates electricity from place relatively easily. There are a number of ways to
the difference between hot steam and the cooled outlet. produce and distribute electrical energy.

G. Thermal Storage Bed For locations like the Lunar I. Power Line Most space projects use electrical energy
surface, which has a long night, solar power is not ef- in some form. The parts of a project on Earth are of-
fective half the time. So storing heat in a Thermal En- ten by far the largest. These include factories to produce
ergy Storage system may be a viable option. Heat is put the vehicles and spacecraft, launch sites, and control cen-
into rocks during the daytime and extracted from them at ters. Typically they get their electricity from a network of
night to run a generator. The rock bed is enclosed in a Electric Power Transmission and Distribution lines.
container, and gas transfers heat to a turbine for genera- This is distributed to the point of use by local Electrical
tion, and from a solar collector for storage. Since the rock Wiring. What distinguishes these three is the scale of
can be obtained locally, the energy stored per mass of in- power, P, in Joules/second, or Watts, and the voltage and
stalled equipment is fairly high. Environment tempera- distance the power is moved. Most wires have electrical
tures during the Lunar night are quite low, and this can Resistance, R, which is a measure of the difficulty in car-
be enhanced by thermal shields between a radiator and rying a current, I, using a voltage V. They are related by
the ground and daytime Sun. So the temperature differ- the formulas
ence between the storage and rejection temperature, and
therefore efficiency, can be fairly high.
Some bodies, like Earth and Jupiter’s moon Io, have rela- V = IR and P =VI
tively high interior temperatures. They serve as a natural The resistance causes some of the energy to be converted
thermal storage bed by the low thermal conductivity of to heat. The amount of power converted, P, is found by
surface rock. The source of heat can be radioactive de-
cay or tides. This energy can be put to use by drilling
down to high temperature regions, and exploiting the dif- P = I 2R
36 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

When efficient delivery of energy, and not heating, is the produce electricity. The wind rotates aerodynamic blades
intended purpose, you want to minimize the resistance mounted on a central shaft, and the shaft is connected to
heating. Since resistance is a material property, by this a generator.
formula you want to use a low current, I. By the previous
formula, the useful power P = VI, so a low current implies
a high voltage. Therefore long distance lines are operated K. Magnetic Storage Some space projects, such as
at higher voltages, and voltage changes are provided by electromagnetic launch from Earth, require very high
Transformers as needed. electric power levels for short periods of time. These
power levels can exceed what is available from the power
The same principles will apply to the space portion of a grid. Magnetic storage accumulates energy over a longer
project when the distance between the point of generation period of time, then releases it quickly when needed. It
and the point of use is large. Additionally, mass is usually uses a 'Superconducting or high inductance/low resistance
a factor for space systems. So besides minimizing losses coil to store energy in a magnetic field. Superconductors
from resistance and transformer efficiency, you want to eliminate resistance heating losses, but require cryogenic
minimize the mass of the wires. The International Space refrigeration to maintain the superconducting state. A
Station is an example where generation and use are sep- large coil, cooled to lower resistance but not cryogenic,
arated by an average of about 50 meters. This is because may be sufficient for some purposes. The energy, E (in
the Station is intended as a zero gravity laboratory, but the Joules) stored in a magnetic field can be found from
solar arrays need to rotate to follow the Sun. Since they
are quite large, they are placed to the sides with rotating
joints.This option mostly applies to fixed locations rather 1 2
E= LI
than vehicles. This is not a large enough distance to re- 2
quire high voltage lines for efficiency. Examples of future L is inductance in Henries, and I is current in Amperes.
projects with longer power lines include mining water ice Storing energy in this way causes structural loads from
in shadowed craters at the Lunar poles. Solar arrays at the the field back to the coil, so the total storage amount is
crater rim may have continuous sunlight while the mining limited by the strength of the structure. Some uses in
area has none. So a transmission line can bridge the gap. space, like pulsed plasma propulsion systems, can bene-
Another example is a nuclear power source for nighttime fit from smaller magnetic storage units to produce high
power for a Mars Base. In that case the source is sepa- power pulses from a lower power steady source.
rated from the rest of the base for safety. Finally, large
orbital habitats and industrial plants may use centralized
generators that are lighter and more efficient, and need
long power lines due to the size of the facility.
All types of wires need isolation from other system ele-
ments and each other, to prevent power leakage, shorts,
arcing, and for safety. In a vacuum or non-conducting
atmosphere, which the Earth’s mostly is, isolation can be
provided by mechanical gaps and spacing of wires. When
the wires must be spaced close together, an Electrical In-
sulator can provide isolation, and combinations of spac-
ing and insulation can also be used.
Figure 1.4-1 - Best research cell efficiency since 1975
J. Electric Generator An Electric Generator con-
verts mechanical energy to electric energy. The two gen- L. Semiconductors Photovoltaic cells convert light,
eral types are a Dynamo which produces direct current, usually from the Sun, into direct current electricity using
where electrons flow in one direction; and an Alternator semiconductor materials. This technology is rapidly de-
which produces alternating current, where electrons flow veloping and has multiple materials and techniques (Fig-
in both directions in an alternating cycle. Most of the ure 1.4-1). Conversion efficiency from sunlight of the
Earth’s electric power is produced by large alternators. best research cells, using multiple layers to capture dif-
The mechanical energy enters the device via a rotating ferent wavelengths, has reached 46.0% as of 2015. Pro-
shaft, and an arrangement of magnetic fields induces a duction panels for use in space, made of multiple cells
current in coils of wire. The mechanical energy can come each, are near 30% efficiency, and the more common but
from any of a number of sources. On Earth it is usually less expensive single layer panels on Earth are typically
from high pressure steam or falling water acting on a tur- 20% or less. Note that the efficiency on Earth vs in space
bine whose shaft is connected to the generator. In the case are based on different solar intensity and spectra, because
of steam, it is created by burning fossil fuels or a nuclear the Earth’s atmosphere absorbs some wavelengths.
reactor, and more recently, from concentrated sunlight. Thermophotovoltaic devices convert infrared and vis-
A growing number of Wind Turbines are being used to ible light from any hot object into electricity. They
2.7. ENERGY SOURCES 37

use similar semiconductors as photovoltaic cells, but in only one direction, a direct current is passed on. Many
optimized for the lower temperature source. A antenna elements are combined into an array to capture
Thermoelectric Generator uses semiconductors to con- the whole of the incoming energy. Rectennas have been
vert a temperature difference into electricity. The most proposed as the receiving element of a long distance mi-
common use in space is generating power from radioiso- crowave power transmission system, such as from Earth
tope decay, in locations where solar panels are cumber- orbit to the ground. Much more solar energy is available
some, is not available all the time, or is too dim, such in space, which results in more net energy delivered on
as beyond Jupiter. An isotope like Plutonium-238 pro- the ground, despite conversion losses. The beam can also
duces decay heat on a steady basis, which thermoelec- travel from the ground to orbit to deliver power to a satel-
tric cells convert to electricity. For space applications, lite.
pure efficiency is not the only significant measure. Vari-
The length of a dipole antenna scales with the
ation with temperature, radiation exposure, and the spe- Wavelength of the incoming energy. In principle, mi-
cific power (W/kg) are also important. Given the trend of
croscopic antenna arrays can be made by the same meth-
past improvements, it is expected semiconductor devices ods used for integrated circuits. This would allow for the
will continue to improve, at least in the short term. The
direct conversion of infrared or visible light. Small scale
latest data should be checked for current performance. antennas are in an early stage of research. Their advan-
(The following are old references, and should be updated) tage for long-range transmission is in a smaller transmit-
ter for a given distance. Microwave technology, by com-
• Anonymous “Conference Record of the Nineteenth parison, is well developed, and high efficiency rectenna
IEEE Photovoltaic Specialists Conference- 1987”, conversion has already been demonstrated.
New Orleans, Louisiana, 4-8 May 1987.
• Anonymous “NASA Conference Publication 2475: 2.7.6 Beam Sources
Space Photovoltaic Research and Technology 1986:
High Efficiency, Space Environment, and Array Entropy is a measure of disorder in a system. A di-
Technology”, Cleveland, Ohio, 7-9 October 1986. rectional beam of energy has low entropy because the
waves are highly ordered (parallel). Useful work can be
• Chubb, Donald L. “Combination Solar Photovoltaic extracted from a low entropy system. This results in in-
Heat Engine Energy Converter”, Journal of Propul- creased entropy (disorder), typically as random motion of
sion and Power, v 3 no 4 pp 365-74, July-August atoms in thermal equilibrium and random thermal emis-
1987. sion. Beams can be natural or artificial, and consist of
electromagnetic waves or particles. Energy beams can be
M. Solar-Driven Turbine/Generator In space, most used for various kinds of Beamed power propulsion, or
electrical power so far comes from photovoltaics, since for powering more stationary activities.
solar panels are lightweight and simple for small to
medium amounts of power. For large scale power, O. Sunlight At increasing distances from the Sun,
Brayton Cycle turbines have been proposed, because of around 14 million km or more, sunlight becomes highly
their potential high efficiency and low mass. The turbine directional. The source, which is 1.392 million km in
shaft then drives an electric generator. The high and low diameter, then fills a small angular part of the sky. At
temperatures for the cycle would be produced by solar the Earth’s distance it appears 0.5 degrees in width. The
concentrators and radiator panels. Stirling type engines small angle allows directional reflection as a controlled
have also been proposed for space use. Sunlight is abun- propulsion method. It also allows for concentration by
dant in space, and lightweight reflectors to concentrate it lenses or mirrors, to generate high temperatures for in-
and feed a heat engine may be lower mass than photo- dustrial or propulsion purposes. This source includes
voltaics. direct use of sunlight, while the items under electrical
(The following is an old reference, and should be updated) sources are for sunlight converted to electricity.
The center of the Sun is at about 15.7 million degrees
• Spielberg, J. I. “A Solar Powered Outer Space He- K, and has a core density of about 160,000 kg/m3 . Un-
lium Heat Engine”, Appl. Phys. Commun. vol 4 no der these conditions Hydrogen undergoes Nuclear Fu-
4 pp 279-84, 1984-1985. sion to Helium, releasing 3.846 x 1026 Watts of energy.
This energy works its way from the core to the surface,
N. Rectenna Array A rectifying antenna, or Rectenna where the temperature has fallen 2
to 5,780K. At this point
is an antenna that is used to convert electromagnetic en- the intensity is 63.1 MW/m . At greater distances the
ergy into direct current electricity. A single antenna ele- same energy flow is spread over larger spherical surfaces,
ment can be a Dipole, with a Diode connected across the reaching 1362 W/m^2 at the Earth.
dipole arms. The incoming electromagnetic waves induce At a distance of more than 550 AU, the Sun acts as a
alternating currents in the dipole. Since diodes conduct Gravitational Lens, bringing the light of other stars to a
38 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

focus. Light which passes farther from the Sun’s edge is 2.7.7 Nuclear Sources
bent less, and comes to a focus farther away. This creates
a radial focal line of concentrated light from every other Nuclear energy sources involve a change in one or more
star or light source in the sky. The same process happens types of Atomic Nucleii, with the release of net energy.
around every other star that has visible neighboring stars.The protons and neutrons in a nucleus are bound together
These Star Lines may be intense enough to be useful, by the strong nuclear force. A change in their arrange-
since they concentrate light from the whole circumference ment involves typically a million times as much energy
of a star to a point. than rearranging electrons, which is what chemical reac-
tions do. So nuclear energy sources are potentially very
powerful. Although the Sun operates by nuclear fusion,
P. Laser A Laser emits light through a process of op- we consider it a source of light energy. The fusion hap-
tical amplification by stimulated emission. The output pens in the Sun’s core, where it is not accessible, and what
of a laser is coherent and collimated, allowing it to be reaches us is blackbody radiation from the surface.
tightly focused, or travel long distances without spread-
ing out. The output can also be in a very narrow range of S. Radioactive decay Radioactive Decay is the spon-
wavelengths. Because of the narrow wavelength, it can taneous change of unstable atomic nucleii by the emission
be coupled efficiently to an absorber, or a high reflectiv- of particles or electromagnetic energy. Unstable natural
ity reflector for that specific wavelength. It can also be elements were created before the formation of the Solar
coupled to a photovoltaic device with high efficiency. As System, most likely in supernova explosions. The less un-
an energy source for propulsion it can supply higher in- stable ones, such as Uranium and Thorium, still survive
tensity light than natural sources like stars. High power after billions of years, and continue to decay at a steady
lasers have been proposed for launch from Earth, but suf- rate. Artificial radioactive materials, such as Plutonium-
ficiently high power ones to make that use practical do not 238, are created in nuclear reactors or particle accelera-
yet exist. Lower power lasers can augment natural sun- tors. They are more unstable, and thus decay faster (an
light falling on spacecraft solar arrays. Very high power 88 year half-life in the case of Pu-238). This element
lasers focused by the Sun’s gravity have been suggested to produces 500 Watts/kg of heat, when fresh, through ra-
power interstellar vehicles, but that use would far in the dioactive decay, making it a useful energy source. It has
future. been used for this purpose on a number of planetary ex-
ploration missions. Other elements with very long decay
times in their natural state are too weak to use as energy
Q. Microwave This energy source involves direct use sources.
of the microwave beam, while item N. Rectenna Array
converts it to electricity. A microwave beam can be ab- [This is an old reference and should be updated]
sorbed and converted directly to heat, or it can be used to
• Lockwood, A.; Ewell, R.; Wood, C. “Advanced
create photon pressure. Any suitable wavelength can be
High Temperature Thermo-electrics for Space
used to create a directional energy beam. However some
Power”, Proceedings of the 16th Intersociety En-
wavelengths are absorbed by the Earth’s or other atmo-
ergy Conversion Engineering Conference, v 2 pp
spheres. Shorter wavelengths can be focused more easily,
1985-1990, 1981.
since that depends on the ratio of antenna size to wave-
length. The efficiency of producing shorter wavelengths 9
is typically lower, and generators with high enough power O Fe U 16 56 235
8 C 12
U
to be useful may not be available. Microwave band equip- 238
Average binding energy per nucleon (MeV)

He 4
7
ment is developed enough to not suffer from these limi-
tations. 6
Li 7

Li 6
5

R. Neutral Particles A Particle Beam is a collimated 3 H3

He 3
stream of high energy particles to deliver energy from one 2

place to another. The concept was originally developed as 1 H

2

a weapon, but less lethal amounts of energy can be used

0 H
1

as a power source. Charged particles, such as protons in a 0 30 60 90 120 150 180

Number of nucleons in nucleus
210 240 270

Particle Accelerator repel each other, so a beam would

spread out once it leaves the confinement of the accelera- Figure 1.4-2 - Binding energy per nucleon for common isotopes,
tor. To prevent this, the charged particles are allowed to in MeV.
combine with electrons to form neutral atoms, or neutral
particles like Neutrons are used. Particle beams are in an
early state of development, and mostly for military use, T. Nuclear Fission The low natural decay rate of some
rather than energy delivery. elements can be increased by artificial means. A Nuclear
2.7. ENERGY SOURCES 39

Reactor is a device for doing this in a controlled way for Boron. What is required to achieve fusion is to bring the
the production of energy. The two main ways to do this positively charged nuclei close enough together against
are Nuclear Fission, the splitting of heavy nucleii into their electric repulsion for the nuclear forces to take over.
smaller parts, and Nuclear Fusion, the merger of lighter This requires the equivalent of millions of degrees K, or
nucleii to form a heavier one. The reason for the two particle kinetic energy in the tens of kilo electron Volts
types can be found in the Binding Energy per nuclear (keV).
particle in a nucleus (Figure 1.4-2). A higher binding en- [These are old references and should be updated]
ergy means the particles are more strongly held together
and more stable, so energy can be released in forming that
nucleus. The binding energy has a peak at Iron-56, so re- • Miley, G. H. et al “Advanced Fusion Power: A pre-
actions from either the light end (fusion) or heavy end liminary Assessment, final report 1986-1987”. Na-
(fission) towards the middle both produce energy. Fis- tional Academy of Sciences report #AD-A185903,
sion reactors are a significant source of electrical power 1987.
on Earth. In space, a few small-scale reactors have been
• Eklund, P. M. “Quark-Catalyzed Fusion-Heated
used, and work is in progress on developing larger scale
Rockets”, AIAA paper number 82-1218 presented
versions with higher power demands.
at AIAA/SAE/ASME 18th Joint Propulsion Confer-
[These are old references and should be updated] ence, Cleveland, Ohio, 21-23 June 1982.

• El Genk, M.S.; Hoover, M. D. “Space Nuclear

Power Systems 1986: Proceedings of the Third V. Muon-Catalyzed Fusion Muon-Catalyzed Fu-
Symposium”, 1987. sion is a method of catalyzing fusion reactions at tem-
peratures far below the millions of degrees K other-
• Sovie, Ronald J. “SP-100 Advanced Technology wise required. A beam of muons is directed at a deu-
Program”, NASA Technical Memorandum 89888, terium/tritium mixture, where they cause multiple fusion
1987. reactions. This heats the gas, which can then drive a gen-
erator for electricity. Although this method and more
• Bloomfield, Harvey S. “Small Space Reactor Power complicated systems based on it are sound from a physics
Systems for Unmanned Solar System Exploration standpoint, a practical system from an engineering point
Missions”, NASA Technical Memorandum 100228, of view has not been developed yet. It must be considered
December 1987. a possible future energy source at this time.
• Buden, D.; Trapp, T. J. “Space Nuclear Power Plant
Technology Development Philosophy for a Ground
W. Nuclear Explosions Unfortunately, explosive
Engineering Phase”, Proceedings of the 20th Inter-
Nuclear Weapons are all too well developed. Various
society Energy Conversion Engineering Conference
concepts have been proposed to use their high energy
vol 1 pp 358-66, 1985.
output for space projects. These include a nuclear-
powered launch device, where the explosive heats gas in
U. Artificial Nuclear Fusion Natural fusion occurs in an underground chamber. This then propels a projectile
stars, and the resulting light output has been addressed up a barrel. Another idea is to detonate small nuclear
above under beamed power sources. This item is for ar- explosions behind a space vehicle, directly pushing it
tificial energy sources. Fusion has been achieved mo- with the blast wave. These concepts are speculative at
mentarily in nuclear bombs, but steady state operation present, because there is no way to safely test them on
has proved difficult. The most researched approach uses Earth, and treaties prohibit nuclear weapons in space.
Tokamaks, which are doughnut shaped magnetic fields
which contain a hot plasma. This approach has not yet
produced a working device, although various research 2.7.8 Matter Conversion Sources
machines have been built or are under construction. A
Tokamak type power reactor would be too massive for In physics, Mass-Energy Equivalence is the idea 2
that
a reasonable propulsion system. For stationary projects mass is related to energy by the formula E = mc . Since
it would be as reasonable on another planet as on Earth. the speed of light, c, is a large number16- 299.8 million m/s,
A number of alternate intermittent and steady state fu- the square is very large: 8.9875 x 10 Joules/kg. This is
sion devices are under varying levels of research, but equal to the output of a nuclear power plant for 2.85 years
all at much lower funding than the work invested in the for each kg of mass converted to energy. In theory, total
Tokamak type devices. Some of those might yield a matter conversion provides the highest amount of energy
lightweight enough device for propulsion. per unit mass. In practice, however, this is not so easy.

All fusion reactions combine light atomic nuclei into

heavier ones. As shown in Figure 1.4-2, the largest en- X. Antimatter Antimatter is composed of antiparti-
ergy release is in the first few elements, from Hydrogen to cles, which have the same mass as normal particles, but
40 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

opposite charge and other properties. When a particle loads and other parameters by themselves, with the help
and anti-particle meet, they destroy each other and are of a calculator and some reference data. Large and com-
converted to other particles or photons, releasing large plicated projects, however, need the knowledge of multi-
amounts of energy in the process. Our universe is almost ple specialists, must satisfy multiple desired conditions
entirely made of matter. So use of this material as an en- and functions, and involve large amounts of time and
ergy source requires making it artificially. This requires money. A need then exists to coordinate the work, and
at least as much energy as is later released by the annihi- ensure the final product meets the intended goal, in the
lation. Antimatter is therefore an extreme type of energy most efficient way. Systems Engineering methods have
storage. Antimatter is made in small amounts today in been developed for this coordination task. They have be-
particle accelerators, and used for physics research. We come their own specialty field and are used in addition to
do not have practical ways to make and store it in large the other engineering specialties. Systems Engineering
enough amounts for space projects. Conceptually, a space can be used for any type of complex project. However,
vehicle would store some amount of antimatter, then use space systems are usually complicated enough to bene-
it to produce energy for propulsion. If the storage system fit from it, and systems thinking and methodologies are
is light enough, the energy per mass would then be higher often used in this field.
than nuclear fusion or other methods.
[These are old references and should be updated]
2.8.1 Systems Engineering In General
• Hora, H.; Loeb, H. W. “Efficient Production of An- What is a System?
tihydrogen by Laser for Space Propulsion”, Z. Flug-
wiss. Weltraumforsch., v. 10 no. 6 pp 393-400,
November-December 1986.

• Forward, R. L., ed. “Mirror Matter Newsletter”, self

published, all volumes, contains extensive bibliogra-
phy.

Y. Black Hole A Black Hole is a region of Spacetime

with such strong gravity that nothing can travel from in-
side to outside it. Two forms of energy extraction are
possible for black holes. The first is infall energy, gen-
erated as material in an accretion disk around the black
hole heats by friction and emits energy. It is essentially
converting potential energy into heat. Since the gravita-
tional potential of a black hole is extreme, this can re-
Figure 1.5-1: Example system diagram showing inputs and out-
lease a lot of energy. The second is Hawking Radia-
puts to system boundary.
tion by quantum tunneling from hypothetical quantum
black holes. Black holes can form by the collapse of a
Given an identified need or desire, how does one select the
large star at the end of it’s life, or a sufficiently dense
best design to satisfy it out of the infinite number of pos-
and massive region at the center of a galaxy. Quantum
sible solutions? For a complex project, the concept of a
black holes are smaller, and hypothesized to have formed
System has proven useful. A System is defined as a func-
during the creation of the Universe. The nearest known
tionally, physically, and/or behaviorally related group of
stellar-mass black hole is 2800 light years from Earth,
regularly interacting or interdependent elements. They
and quantum black holes have not been discovered, nor is
are distinguished from the rest of the Universe by a Sys-
there a known way to make them. So use of black holes
tem Boundary (Figure 1.5-1). A system is not a physi-
for space projects is theoretical at present.
cal entity, but rather a mental construct, created because
of it’s usefulness, by drawing a line or surface around a
collection of elements. The elements have internal rela-
2.8 Systems Engineering tionships to each other and form a comprehensible whole.
The rest of the Universe outside the system is referred to
Engineering applies scientific principles and other forms as the System Environment, or simply the environment.
of knowledge to design, build, and operate systems which Flows of many types enter and leave the system as In-
perform an intended function. It is a broad discipline, puts from and Outputs to the environment by crossing
whose parts we will discuss later in Section 1.7 - En- the system boundary . The scope of a given engineering
gineering Specialties. In a simple project, such as de- task is then defined by the system boundary, what crosses
signing a bookcase for home use, a formal engineering the boundary, and what is inside. Systems may contain
process is not needed. One person can calculate the shelf smaller systems within them, which are called Subsys-
2.8. SYSTEMS ENGINEERING 41

tems. These may be nested to any level, but flows into

and out of a subsystem must appear in the parent system,
or at the top level in the environment. This rule may be
called Conservation of Flows - that flows do not appear
from or vanish into nothing. Following that rule ensures
that all the required inputs and outputs are accounted for.

Systems Engineering Method

A single person may have the time and knowledge to

do a preliminary concept or design. A complete space
project is usually too complex or would take too long for
one person to do. So the Systems Engineering method
can be used to help carry out such projects. Aerospace
projects, including space systems, are particularly suit-
able due to their complexity, and were among the main
ones for which the method was developed. The method
focuses on how projects should be designed and managed
over their entire Life Cycle, that being from initial con-
cept to final disposal. Since it applies to the whole project,
it is interdisciplinary, connecting tasks performed by Sys-
tems Engineering specialists to those of other engineering
branches. Key parts of the process include:

• Breaking down a complicated project in such a way

that the smallest pieces are simple enough for hu-
mans to design.

• Modeling the system so it can be analyzed and opti-

mized, and comparing the actual physical system to
the models.

• Control and track the information and design of the

pieces and their relationships so the total system will
do what you wanted.

Figure 1.5-2 illustrates in general the steps within the Sys-

tems Engineering process. The trend is from top to bot-
tom, but we do not show arrows connecting the steps be-
cause it is not a strict linear flow. As results are obtained
in any task, they can feed back to earlier steps in an it-
erative fashion, until a stable design solution is reached.
So these tasks can happen in parallel, and applied across
the different stages of the life cycle. The tasks can also
be applied at different levels of detail. They are started at Figure 1.5-2
a general level. Once a stable configuration is reached at
one level, it then is re-applied at lower levels until detailed
design can be done on individual elements. At all levels,
there is communication with design specialties, and with
outside entities such as the customer, suppliers, and other
scientific and engineering organizations. The steps are The systems engineering process is bounded by natural
described in more detail in sections 3 and 4 below and on and human-made constraints. Many of the human con-
page 2. It should be noted that an organization capable straints are not directly related to design in the way physi-
of designing and building complex space systems is itself cal properties of materials are. These indirect constraints
a complex system. While it is not often done, Systems include economics, laws, and safety of life and property.
Engineering methods can be applied to the organization The process is then also outward-looking, beyond the de-
itself to design and optimize how it functions, or to any sign itself. Other engineering specialties are more fo-
complex system of any type, not just space hardware. cused on the internal details of the design.
42 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Large Scale Systems

It is not required that a single organization do all the Sys-

tems Engineering tasks. Some very large systems, that
have been deliberately engineered, such as the US Inter-
state highway system, the Internet, and the US program to
land humans on the Moon, involved many entities work-
ing together. A national system of government, human
civilization, or the Earth’s biosphere can be considered
as very large systems in terms of having inputs and out- Figure 1.5-3 - Linear life cycle stages.
puts, a system boundary, and an external environment.
There is a growing understanding that such large entities
are systems composed of many smaller systems, whether 2.8.2 The System Life Cycle
designed or not. Analyzing such large entities as systems
can help with understanding how they function and de- Complex systems evolve through a Life Cycle much the
termining if corrective action is needed. Although some way living things do, from conception to disposal. The
attempts at designing governments have been made, they life cycle is divided into a number of stages where differ-
have yet to be done based on scientific and engineering ent tasks are performed (Figure 1.5-3). The design stages
principles. Climate Engineering, which is the concept (the first three boxes in the figure) can be organized in dif-
of deliberately affecting the Earth’s climate, is an exam- ferent ways depending on the nature of the system. These
ple of biosphere level engineering projects. Doing them include linear, parallel, spiral, or closed loop sequences,
deliberately, as opposed to the inadvertent side effect of or some mixture of these. The illustration shows a typ-
civilization, is still in the conceptual stage. More work ical linear sequence. A spiral process repeats stages in
has been done in the field of Economics in analyzing eco- increasing detail, while a closed loop repeats at the same
nomic systems, and sometimes attempting to design or level of detail. Beyond the design stages, the process is
influence them. more typically linear from production, through test, in-
stallation, operation, and retirement.
Life cycle stages are used for two important reasons.
First, the design process should consider all the later
Reference Sources
stages, so that the best total solution is found, rather than
optimizing for just one part of a system’s life. Second,
As a well-developed engineering specialty, there are a breaking down a system by time is another way to sim-
number of reference books, standards, and special meth- plify the design work, along with breaking it down by sub-
ods and software used by systems engineers. They are systems and components. The stages are further broken
used to understand and manage the interactions, and com- down into internal tasks which have inputs and outputs
municate the current state, of a complex project. The re- that connect them, and have decision points for when it is
mainder of Part 1 of this book summarizes parts of the time to proceed to the next stage.
systems engineering method. This includes the elements
A life cycle is a time oriented view of an entire system.
of a system, engineering tools, involvement of other de-
Other views of the same system include functional dia-
sign specialties, and economics. A given program also
grams, which show what tasks it performs and their in-
has to be understood in the context of other existing and
puts and outputs, and a work breakdown, which tabulates
future programs. All of these tools and knowledge must
be integrated properly for a new project. the elements and sub-elements which make up the sys-
tem. Which view of the system is used depends on the
For additional detail on Systems Engineering beyond design task at hand, though all the views need to be kept
what is in this book, see: current or the design process can become disjointed.

• BKCASE Project, Systems Engineering Body of Life Cycle Example

Knowledge wiki, v1.8 - 27 Mar 2017.
The names, and the task contents, of a given project’s
• DAU Press, Systems Engineering Fundamentals, stages can vary according to the needs of the project.
2001. However a somewhat standard linear flow is often used in
aerospace engineering, including space-related projects.
The stages and typical major tasks include:
• NASA, Systems Engineering Handbook, 2007.
Conceptual Design
• NASA, Systems Engineering Class Materials -
Website developed since approx. 2008. • Identifying the need - what is it you want the system
2.8. SYSTEMS ENGINEERING 43

to do? This is embodied in goals and requirements. • Element and System Test - At each assembly level,
testing that the assembly functions, then moving up
• Establish selection criteria - how do you decide one to larger assemblies to the final product.
design is better than another?
• Verification - Proving the system meets the stated
• Establish a system concept - this includes the main design requirements, by a combination of test,
functions, operation, and maintenance of the sys- demonstration, inspection, and analyses.
tem.
• Feasibility analysis - can the need be met at accept- Installation and Deployment
able cost, schedule, and other parameters.
After production, the system elements may
Preliminary Design need delivery, installation, and activation at the
location they will be used.
• Functional analysis - identify and break down the
complex system into smaller functions and their re- Operation and Maintenance
lationships, including alternate arrangements
• Operation - Using the system for the purpose it was
• Design Allocation - subdivide and assign require-
designed, in the intended environment.
ments to lower tier functions
• Support - which includes operator training, perfor-
• Formulate alternatives - develop alternate solutions
mance monitoring, and logistic support.
- what are the range of possible options?
• Maintenance - includes planned maintenance and
• System Modeling - develop mathematical models of
unplanned repair, and in-place upgrades.
the system so variations can be assessed.
• Optimization and selection - making each option as Decommissioning
good as it can be, then compare options and choose
the best. When the system has reached the end of its use-
ful life, the removal, recycling, and disposal of
• Synthesis and definition - combining the selected
system elements, and return of former sites to
options into a total design, and recording the con-
their original conditions.
figuration and requirements details

Detail Design Life Cycle Engineering

• Design - Once broken down to a low enough level, As a process that applies across the whole life cycle, Sys-
individual elements are assigned to engineering spe- tems Engineering is not just used in the initial design
cialties or design teams to complete. Design in- phase. Part of good design practice is to know when to
cludes physical hardware components and facilities, stop designing. A design can always be improved with
as well as software, operating procedures, training, more work, but at some point additional work does not
and other non-physical elements. provide enough added improvement to justify it. At that
point the design should stop, and the system progresses to
• Integration - Design elements are combined into the next stage, which is usually fabrication and assembly.
larger functional units that work together, up to the With time, the original design assumptions for a project,
system as a whole. such as the available technology level, or launch to or-
bit traffic levels, will change. The systems engineering
• Engineering Models and Prototypes - Physical par-
process can then be re-applied to see if a design change,
tial models and complete prototypes built to validate
upgrade, or even complete replacement of the system is
the design.
warranted. Even if the system was optimally designed
when first created, future events may require changes. If
Fabrication and Assembly the system was properly modeled and documented, then
monitoring of these external changes will reveal when it
• Production of components - For physical items, the is time to restart the engineering.
step where you make the parts.
• Assembly - Putting parts together into complete el- 2.8.3 Requirements Analysis
ements.
Developing a new system starts with a desire or need
Test and Verification which cannot be satisfied by existing systems. The needs
44 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

and desires are expressed by a Customer, For systems ery, that leaves 15 days to complete ground processing.
engineering purposes, the direct customer is the person The stated goal was thus 160 hours for ground process-
or entity who is paying for a project or can direct the en- ing, composed of 2 shifts (16 hours) x 10 work days over
gineering staff. For example, in the Boeing Company, two weeks, thus 14 days. This goal was expressed, but it
that is the engineering managers and general managers may not have been included in the system requirements,
of the company. The ultimate customers, which are air- and it definitely was not allocated to lower tier hardware
line passengers, cannot express their desires directly. So and tracked at the lower levels like hardware weight was.
the company management serves as a proxy to express Only after the Shuttle was already flying was it noted that
their desires as an input to the engineering process. Other
ground processing was taking too long, and efforts started
methods, such as surveys, can be used to determine the to reduce it. At that point it was too late to make any fun-
desires of the ultimate customers.
damental changes in the design, and so ground processing
The initial expression may be in the form of general ver- never got below about 800 hours, about 5 times the orig-
bal goals, system properties, levels of technical perfor- inal goal. This was a major contributor to the Shuttle
mance, and similar statements. The customer also will never reaching its intended flight rate. In order to have
have some value preferences which describe what a bet- reached their goal of 160 hours, processing time would
ter design is from their point of view. These can be have had to be allocated to sub-systems, such as landing
things like “minimum cost”, “minimum waste output”, gear or maneuvering thrusters, and then each subsystem
and “maximum efficiency”. designed to meet it’s assigned time. Conversely to pro-
The first major systems engineering step, Requirements cessing time, weight has always been a tracked parame-
Analysis, is the process of converting these general cus- ter in aerospace systems, since airplanes cannot function
tomer desires and preferences to specific measurable fea- if they are too heavy. The Space Shuttle had very de-
tures which can be used for design and evaluation. Two tailed weight targets and a tracking system by component,
main parts of this process are Requirements Definition with monthly reports. It more or less reached its design
and Measures of Effectiveness. payload, which is the 1.5% of available launch weight re-
maining after the vehicle hardware and fuel are accounted
for.
Requirements Definition This example emphasizes why desired features must not
only be stated quantitatively, but passed down for engi-
The highest level general desires are first converted to spe- neers to meet in detailed design, and tracked so you can
cific measurable features and values called System Re- tell if you are going to meet them. Measurable param-
quirements. These are later broken down into more de- eters can be a simple yes or no, for example “Does this
tailed lower level requirements, which are assigned to log- airplane design meet FAA regulations?", or it can be a
ical elements of a system called Functions to perform. numerical value, range of values, table, formula, or graph
The assignment ensures that somewhere in the system all indicating the range of acceptable values for that system
the top level goals are met. At the most detailed level a characteristic.
subset of the lower level requirements are assigned to a
single function box. This now becomes the detailed de-
sign conditions for that function. Assuming the analysis Measures of Effectiveness
has been carried to a low enough level, the detailed de-
sign of the element that performs the function can then Desired features are often in opposition. For example,
be done with a reasonable effort. higher performance and reliability often come at the ex-
The first step in requirements definition is documenting pense of higher cost. There are also alternate designs
the original desire or need of the customer in as much de- which have different amounts of each feature. Estab-
tail as they can provide. We will use as an example the lishing Measures of Effectiveness is the quantitative
Apollo program to land humans on the Moon. That was method to account for these disparate features at the level
expressed by President Kennedy as a well known goal of the whole system. Like requirements, they are derived
with a deadline. That very general statement was not from customer desires. In this case it is what features
sufficient to design the hardware. The key task is to put would be “better” when comparing one design over an-
all the requirements in forms that can be measured so that other. Since different features typically have different
you can tell when you meet them. Experience shows if a units of measurement, they need to be converted to a
desirable feature or parameter is not expressed and mea- common measuring scale. This is done by formulas that
sured, it will not happen as desired. An example of this convert each different value, such as cost or performance,
failure is the Space Shuttle Program. The original goal to a score. These scores are given relative weights based
was to fly 60 times a year. Given a fleet of 4 Orbiters, each on their importance to the customer. The weighted mea-
one had to fly 15 times a year, or one launch per 24 days. sures can then be used in a single mathematical model or
Subtracting 7 days on orbit and one day before and af- formula to determine “better” as a total score, when the
ter flight for launch preparations and post-landing recov- component values vary across different designs.
2.8. SYSTEMS ENGINEERING 45

The value scale is often in a range such as 0 to 100%, or tem goals. Historically the first two requirement types,
1 to 10, but this is arbitrary. What is more important is a performance and cost, were the primary ones considered.
definite conversion from a measurable feature to a scoring As systems have grown more complex and their outside
value, and the relative weights and method of combining interactions and side effects become better understood,
them to a total score. As an example, a value of 0% might the number of desirable features, and thus the number of
be assigned to a payload of 15 tons, and 100% to a pay- requirements, have increased. This trend is expected to
load of 45 tons, with a linear scale in between, and pay- continue in the future.
load given an importance of 30% in the total score. The Aside from the biblical Ten Commandments, require-
total scoring system becomes a mathematical model of
ments are rarely set in stone. Not all of them will be iden-
the customer’s desires for the system. Getting a customer tified at the start of a project. As a result of interaction
to define “better” in such detailed numerical form is of-
with the customer and feedback from the design process,
ten difficult, because it removes their freedom to choose they can end up modified. For example, a launch capac-
what they personally prefer in a design in spite of the en-
ity of 10 launches at 100 tons each might be specified for
gineering solution. It is necessary, though, if you really a rocket, and later analysis show that 20 launches at 50
want an optimal answer. At the least this process makes it
tons each yields lower total cost. The requirements would
obvious when the customer is over-riding the engineering then be modified to reflect that. At any given time, how-
process. ever, the current set of requirements guides the engineer-
It should always be kept in mind that a particular design ing work. Over time, requirements become more firmly
solution may not be “good enough” in terms of of it’s mea- fixed, generally from the higher to more detailed levels
sures when compared to existing systems. This can be in sequence. Changing a requirement forces rework of
found by including the existing system as one of the al- previous designs. So the cost of changing a requirement
ternative designs being scored. In that case the proper grows later in the process, and this tends to exceed any
answer is to stop development of the new system and stay benefit from the change.
with the existing ones. Often the cause is not enough per- Performance
formance improvement relative to cost, but other mea-
sures can result in a decision to stop. Performance requirements are measures of the primary
intended function of a system. Every system must have
at least one performance measure for what it does, and
2.8.4 Requirements Types often there are a number of them. As an example, the
design capacity for space transport systems is often ex-
The following subheadings list major types of system re- pressed as a Mission Model. The mission model quan-
quirements. Not all would be relevant to a given project, tifies the system performance in terms of multiple pa-
and others besides these might be important to the cus- rameters like dates, flight rate, payload dimensions and
tomer, so it is presented as a starting point for consider- mass, mission duration, destination orbit, type of cargo,
ation. Each type can include more specific requirement and maximum g-level. For a space habitat, performance
values. The types listed here are linked and somewhat might be measured in number of crew supported, lev-
overlapping. For example high reliability and high safety els of atmosphere, food supplies, and gravity, and total
generally go together. Requirements set limits on a de- living volume. An industrial system might have require-
sign, and overlaps between requirements in effect overlap ments for Throughput, in tons per day of materials pro-
the range of limits they impose. This is acceptable as long cessed, and Efficiency in terms of (theoretical energy re-
as the designer understands the range of overlaps and in- quired)/(actual energy used). The particular performance
teractions among the requirements. A particular design measures which matter will vary by system.
parameter will be governed by the strictest requirement
An example mission model for the Apollo program might
when there is such overlap. An example from civil en-
have started out as follows, with more detail added as the
gineering is that earthquake, wind, and snow loads are
project progresses. Even in this early version, it lists a
all requirements to meet in a building design, and they
number of different performance measures that the design
overlap in that all of them affect the required strength of
needs to meet:
structural elements.
Cargo characteristics:
When broken down to lower levels of the system de-
sign, the requirement types and values will become more
specific and detailed. Care is needed to maintain logi- Number of crew to the lunar surface: 2/mission
cal and numerical consistency across system levels. Re-
Maximum Stay time: 4 days/mission
quirements, or parts thereof, should not be inserted or
dropped at lower levels. Traceability is the ability to Additional science equipment: 250 kg/flight
follow the chain of requirements across the system lev-
Lunar samples returned: 100 kg/flight
els, and is maintained by documenting how they are con-
nected. This is necessary so you can prove satisfying
the lowest level details actually meets the top level sys- Mission Schedule:
46 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

First Flight: as early as possible but before Jan Compliance Requirements are set by external human
1, 1970 rules such as laws, regulations, codes, and standards. Hu-
Flight quantity: 10 to lunar surface (this was man rules often set minimum requirements in areas like
the original plan) safety. This does not prevent a system from adopting
stricter levels. Human rules are usually designed to pre-
Flight rate: 4 flights/year vent undesirable effects. For example, speed limits on
driving are intended to reduce the frequency and sever-
Performance requirements only address what a system ity of accidents. Compliance requirements exist whether
does when it is operating as intended. It does not address or not they are explicitly incorporated into the engineer-
what happens outside that context, such as ing process. It is better to incorporate them explicitly
to avoid later problems. Other requirements are set by
• In between active operation, such as the 80 days be- nature, such as the minimum altitude for a stable Earth
tween Lunar missions in the mission model above. orbit. These do not fall under compliance, but are ac-
counted for elsewhere. In the case of altitude, this might
• When the system fails, as did the Apollo 13 mission,
be a performance requirement that a rocket deliver the
• Before and after the 5 years of manned missions. payload to a 250 km high orbit.

• Interactions external to the program, such the supply Technical Risk

of technical personnel for the project, environmen- Especially in the early stages of design, the engineering
tal impact of the launches, or return of Lunar germs process may reveal gaps in knowledge, performance un-
to Earth. The last turned out to be a needless worry, certainties, resources which are not available, or other is-
but the quarantine system for returning astronauts sues which prevent selection, optimization, or synthesis
and moon rocks is not something covered in perfor- of a design. These issues can prevent progress to the next
mance measures. stage of the project, or cause a final design which does
not meet desired goals. Measures of these unknowns are
So performance alone does not encompass the entire sys- given the general name Technical Risks. For example,
tem over the entire life cycle, and other requirement types a new technology which has not been demonstrated yet,
are needed. i.e. a fusion rocket, would be rated as high risk, while a
chemical rocket, which has decades of operating history,
Cost
would be comparatively low risk. A mass budget consid-
Cost represents the net resource inputs to a project from erably below past experience or with insufficient margin
outside the system. Space projects don't use Dollars or during preliminary design would be high risk. New re-
Euro directly, but rather use them to pay for labor, ma- search, modeling, or prototyping can be done to reduce
terials, and services they do use. So cost is a measure of the risks, or the system modified to avoid them. Before
flows across the system boundary, rather than an inter- these risk reduction efforts the risks will still exist, and it
nal property of the system. Every system will consume is necessary to account for them. Otherwise you accept
some resources during its life, but funding sources are the alternate risk of the system not performing as desired
not unlimited. So cost limits are almost always consid- or even at all.
ered a requirement, whether implicitly or explicitly. Total
Not every risk will be known at the start of a project, but
cost over the entire life of the project is called Life Cycle
sound engineering practice is to identify them as early as
Cost. This can be further broken down into development,
possible, and to allow for modifying development plans
production, and operations costs, and then accounted in
when they appear. Depending on how much new technol-
much greater detail across the system elements. In addi-
ogy is included in a project, sufficient performance, time,
tion to total cost, limits can be placed on spending rates.
and cost margins should be included for unexpected prob-
This is most explicit in government agency budgets, but
lems caused by technical risks. Technical risk is gradu-
even private projects have limits on spending per year.
ally retired during the design and production of a system.
Some systems generate revenue to offset costs. When
Once a system is operating, a small uncertainty remains
revenues exceed cost, the system as a whole generates a
for things like operating life or failure rates. These are
profit in financial terms. Revenue may be delayed until
not eliminated until the end of program operations. Even
after the design and construction phases and the system
after disposal of a system, some environmental risk may
begins to operate. The peak net cost accumulated until
remain. A prime example is nuclear waste, which is a
revenues exceed expenses is described as Capital or De-
hazard long after the reactor that created it has been de-
velopment Costs. Customers generally want high per-
molished.
formance and low cost, so the ratio of performance/cost
is often a key measure for a project. Safety
Compliance Safety is the state of being protected against adverse con-
sequences to living things, or damage and destruction of
Performance and cost requirements are set by the
inanimate objects. It is an inverse measure of other risks
project’s customer with the help the engineering team.
2.8. SYSTEMS ENGINEERING 47

than those under the previous heading. So a higher safety Noise and quantum effects are natural variations which
level is measured by lower risks. Under the principle of cannot be eliminated, but a large margin above those vari-
“protecting the innocents”, hazards to a crew that volun- ations in effect reduces variability and increases quality.
teers to accept a risk can be higher than those allowed Sustainability
to the public at large. A safe system, such as a nuclear
power plant or passenger airplane, may have less than Sustainability is the capacity of a system to endure. For
one expected accident during the system life. So safety example, does a system consume a scarce resource or
often involves assessing low probability events. Require- generate a waste output that limits it’s long term use? If
ments to maintain control of a system despite failures, in- so, it is not sustainable over the long term, although it may
herent fail-safe design, design margins, backup systems, last a desired system life. A current example is hydrocar-
and redundancy can improve safety when properly imple- bon fueled rockets. If obtained from fossil sources, they
mented. are both limited in supply, and cause unwanted change to
the atmosphere. If they are produced as a biofuel, they
Reliability can be sustainable.
Reliability is the probability the system will perform its Community
intended function for a specified time period. The inverse
is probability of failure. It is related to Resilience, which These are requirements to have a positive, or at least min-
is the ability to function in the face of internal damage or imally negative, impact on the surrounding human com-
external failures. It is also related to Robustness, which munity. This might mean employing local staff, or avoid-
is the ability to function in the face of external or internal ing traffic problems during a shift change, or noise impact
variables, such as line voltage or temperature. A closely from rocket launches. Positive educational impact is an-
related measure is Availability, which is the probability other community effect.
a system can start operating at a random requested time, Environment
or the percentage of a total time interval it can be oper-
ated. A high reliability system may require multiple units Like community, these requirements relate to impact on
in place, so that at least the minimum required number the surroundings, but in this case the non-human portion.
are available at a given time. An example is passenger For space projects a key environmental requirement is to
airplanes, which require multiple engines for high relia- avoid contamination, either biological, chemical, or from
bility, in case one stops working. radiation, both forward and backward from the ends of
the mission.
Durability
Manufacturing
Durability is how long a system can perform its intended
function. It is typically measured in terms of service life. This type of requirement covers items such as how many
Service life of components, and of the system as a whole, sources are there for a given manufacturing method, or
is related to their maintenance. If not enough mainte- how exacting the tolerances are. In sum they measure
nance is performed, the life is reduced. Once the life of how easy or hard the system is to produce.
the item has been reached, it will need repair or replace- Testing
ment. Service life can be measured in terms of number of
These requirements deal with the types and quantity of
uses, operating hours, or calendar time. Durability is re-
tests required for a system. Development tests occur dur-
lated to the economics concept of a Durable Good, one
ing the design stage, qualification tests occur during ap-
that yields utility over time rather than being consumed
proval, and periodic inspection and testing may be needed
in one use. A passenger airplane has high durability, be-
during the operations stage.
cause it can operate for decades and tens of thousands of
flights. The opposite of durability is consumption. The Maintenance
fuel used in the airplane is consumed (used only once), These requirements include parameters like hours and
and is therefore not durable. costs to maintain the system in an operating state, proba-
Quality bility of system failure, and levels of spares required. A
system can fail without safety risk. For example, your
Quality is a measure of how well a system meets expec-
car may not start, which is different than the brakes fail-
tations. One aspect of quality is a measure of the lack
ing to operate. The more often that items fail to oper-
of variability or defects in the design and manufacturing
ate, the more spares are needed in stock, and the more
stages of the life cycle. Variability and defects increase
time and money to repair or replace items. So the vari-
the chance that performance will fall outside required lev-
ous maintenance requirements are linked. Maintenance
els. Another way to put it is conformance to initial spec-
requirements can be divided into preventive, which is be-
ifications. Wear or defects caused during normal opera-
fore something stops working, and corrective, which is
tion are not a quality problem unless they are unexpect-
after.
edly large. Normal wear would fall under maintenance
requirements. Another quality factor is parameters like Flexibility/Adaptability
signal-to-noise ratio and error rates in electronic devices.
48 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

This is the ability of the system to adapt to new tasks or 2.8.5 Continued on page 2 →
functions, or different performance levels for the original
tasks, than first designed for. Similar requirements come
under names like Extension, which is adding new tasks
2.9 System Elements
or functions while keeping the original ones, or Agility,
which is concerned with how fast the system can adapt. ← Back to Page 1
Reconfiguration is the ability to change the arrangement
of a system to perform a different function. An exam-
ple is the Curiosity rover, which had one physical shape 2.9.1 Functional Analysis
and software load to cover travel to and landing on Mars,
and a different arrangement of wheels, camera mast, and Once requirements have been developed to a certain level
software for surface operations. The change of state was (see page 1), the next step is Functional Analysis. En-
accomplished by a combination of mechanical design, gineering analysis in general is the breaking down of an
planned sequence, and new software upload. object, system, problem or issue into its basic elements,
to get at its essential features and their relationships to
Scalability each other, and to external elements. Analysis includes
This is the ability of the system to change in size either developing abstract models or performing calculations for
by scaling the size of a unit, or by installing more units. the component elements of a system, to help arrive at a
There is always a limit to scaling imposed by some phys- complete and optimized design. Functional analysis is a
ical constraint. If the system can be scaled to meet the breakdown on the basis of what a system does, in terms
full demand for it without reaching scaling limits, it can of the functions it performs or a sequence of operations.
be said to be scalable. Modularity is a related parameter, This is before considering how it is performed. “How” is
concerned with how separate the elements of the system a design solution, which we don't want to choose prema-
are, and how easy it is to replace them with other ele- turely. Instead we want to consider all the alternatives and
ments of the same or different types. Instances of mod- optimizations, which we do in later steps of the process.
ularity can be horizontal - at the same level of a system, Prior to selecting the best design, there may be multiple
or vertical, as in the layers of the Internet Protocol stack. functional breakdowns, at least one for each system con-
cept, and often alternate versions within a concept. The
Evolution
details of these steps and their interactions are recorded
This requirement type considers how well the system can in the form of diagrams and models, which can then be
change over time to a different type of system. It is related used for calculations and assessments.
to flexibility, which is more about changes while staying
the same kind of system. Redesign is concerned with the
difficulty and cost of making changes. Concept Identification
Usability
Although we don't want to prematurely select a design so-
Usability requirements deal with the interface between lution, we do have to generate alternate concepts for how
humans and system elements. When a system can be used the system will function. At the level of the system as a
without great amounts of planning, preparation, physical whole it is difficult to define requirements and measures
strain, or training it is said to have high usability. without an idea of how the system will work. Concept
Interoperation identification involves identifying alternate approaches to
how it will operate, and synthesizing one or more system
This is a measure of how the system fits with other sys- concepts for each potential approach. System level con-
tems. For example, a new airplane with doors that did not cepts give the general approach to design and function,
fit existing gates, or a computer network that exclusively without specifying exact values of parameters or what
uses a new protocol that nobody else uses would fail in components will be used. For the Apollo program, for ex-
this parameter. Compatibility is more concerned with ample, there was a choice of “Direct to the Moon” vs “Lu-
the direct interfaces between systems, such as the output nar Orbit Rendezvous” as mission concepts, with the lat-
of a computer video card matching the input to a moni- ter being the one actually chosen. System concepts may
tor. These features are more prominent in the information include major variables such as type of propulsion (i.e.
technology fields because of the sheer number and vari- chemical or nuclear), service life (one or multiple mis-
ety of hardware and software elements which must work sions), and supply concept (i.e. closed or open loop life
together (with varying degrees of success). support). At the least it covers what main tasks the system
Openness will perform, and how it will be operated and maintained.
Once the system concepts are established, the process of
This is the degree to which a system is composed of pro- analysis, optimization, and selection can begin to find the
prietary or secret elements compared to open, public, or best version of each competing concept, and then com-
standard elements. pare and select among the concepts to carry to the next
stage of development.
2.9. SYSTEM ELEMENTS 49

At lower levels of the design process, this step is repeated Individual functions in a diagram transform inputs into
when there are multiple possible approaches. An exam- outputs (Figure 1.5-5). The diagram typically shows
ple would be thermal protection for re-entry. An ablative functions as boxes and input/output flows as arrows con-
heat shield burns off some material each time, so check- necting the boxes running from left to right. Flows can
ing thickness and periodic replacement would be part of contain any sort of item, including information, matter,
the necessary operations. A metallic heat shield might energy, labor, etc, or a combination thereof. They may
not burn off, but suffer cracking from heating and cool- divide and combine on a diagram, but the divided flow
ing cycles, and require different types of inspection. So must sum to the contents of the undivided one. This
when listing design alternatives, it is not just ablative vs derives from the physics concept of conservation laws,
metallic that is important, but also how that choice affects where matter and energy do not arise out of nothing. Sim-
the total flow of operations in the system. ilarly the flows within a system do not arise from or vanish
into nothing, they must enter from outside or be converted
by a functional task. By following the Conservation of
Functional Diagrams Flows logic, then all the inputs and outputs of a system
will be accounted for.
Control inputs regulate the operation of a function. By
convention they are shown entering the top of a function
box. Mechanisms perform the function, but are not trans-
formed themselves, and are shown entering from the bot-
tom. A mechanism example is a stamping press, which
converts flat steel blanks to shaped stampings. Tne blanks
and stampings are the inputs and outputs, respectively.
For a complex system, the diagrams form a hierarchy,
with one box on a given level expanded to a full diagram
with multiple boxes at the next level down. Developing
the levels of diagrams is a continuing task done incremen-
tally, rather than all at once. The diagrams are a way to
record and communicate the structure and operation of a
Figure 1.5-4: Complex Functional Flow Diagram. system. They allow numerical calculations, for example
noting the time required for each step to find the total op-
Functional Diagrams are prepared during analysis to il- eration time, or summing staff required for each function
lustrate the component operations a system performs and to get total staff needed to operate the system. Functional
their inputs and outputs. They are a model of how a sys- diagrams can also be converted to mathematical simula-
tem operates in visual form. They start at the top level tions of system operations, typically with computer soft-
with external inputs and outputs which cross the system ware made for that purpose. Any amount of description
boundary, then are broken down into multiple levels of or other information may be attached to items in a dia-
detail. They are often in a step by step time sequence, but gram, by means of a unique function or flow reference
can be more complex networks of operations with deci- number. By convention, expanded lower level diagrams
sion points and loops (Figure 1.5-4). For example, for an use the same number as the parent box (i.e. 9.2), with
airplane, the main functions would be Load Passengers another period followed by another number (9.2.1, 9.2.2,
and Cargo, Taxi to Runway, Takeoff, Fly, Land, Taxi to etc.). This is not required, but it makes tracing the con-
Gate, and Unload Passengers and Cargo. Each of these nections between diagrams easier.
main functions are further divided into smaller steps, then
assigned to system elements to carry out. For exam-
ple, the landing gear might be assigned multiple functions 2.9.2 Requirements Allocation
such as “absorb landing loads” and “provide steering for
taxiing”. Those then become requirements for detailed The third major step in the systems engineering pro-
design and testing of that element. cess is Requirements Allocation. To ensure all the
top level requirements will be met, they are assigned to
one or more functions to implement. The assignment
may be the whole requirement, or by dividing it into
parts and then assigning the parts to separate functions.
Allocated requirements are documented in lower level
requirements documents or specifications that apply to
parts of a project. Traceability is being able to trace
the links between lower and higher level requirements and
the logic of how they were generated. At the lowest level,
Figure 1.5-5: Single Functional Diagram Box a subset of the requirements are assigned to a particular
50 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

hardware or software item, skilled staff, procedures, fa- A model, such as an Input/Output Model lets you ac-
cilities, interfaces, services, or other elements of the final tively see how a change in any one component, such as
system. With a complex project, software tools become a new design, impacts the system as a whole. By sum-
very useful for the requirements allocation and tracking ming the flows of the component functions in a model
process. They can help manage the mass of details, and spreadsheet (or other computer model) you can immedi-
ensure everyone on a project has the most current infor- ately find changes to the rest of the system components
mation. and the totals for the entire system. The Input/Output
Requirements allocation is not a one-time task, although model and functional diagrams both model aspects of the
same system, and may be combined within a single soft-
it is weighted towards the early stages of a project. As
design and testing make progress, they can provide feed- ware tool or database if it can represent all the details of
a system in sufficient detail. This is desirable for plotting
back and adjustment to the assigned requirements. These
changes propagate to higher levels, and by tracing back how changes in system components affect the system as
a whole.
their impacts, you can determine how they affect the top-
level goals of the project. Changes can also have sideways Functional diagrams at a basic level are maintained as
effects at the same level. For example, an increase in static drawings, and input/output models can be an ac-
weight of one part of a system may require weight-saving tual spreadsheet. Using the same numbering system and
efforts elsewhere to not affect top level performance. structure for the diagrams and models maintains the re-
lationship between them. They are both representing the
same system, just different aspects.
2.9.3 System Modeling

The next major step is to model the system and alter- 2.9.4 Optimization and Trade Studies
nate approaches to the design. Various methods are used
to model the design and configuration of the system el- Optimization and selection is done at all levels of engi-
ements. Traditional ones include two dimensional di- neering design. In the Systems Engineering process it is
agrams (blueprints) and physical scale models. These first applied at a high level to concepts before detailed
methods can help visualize a system, but are not easy design is performed. Optimization is varying parameters
to modify, derive parameters, or perform simulations. within a single concept or element in order to find the best
The trend is towards integrated software modeling, where values for those parameters. Trade Studies compare dif-
software tools model and simulate multiple aspects of the ferent concepts in order to select the best one. Different
system, or communicate from one tool to another. In soft- parameters like weight, cost, and risk cannot be directly
ware tools, a system is represented as data and mathemat- compared. So they are scored by measures of effective-
ical relationships, which makes it much easier to change, ness (see page 1). The concept or optimized parameters
optimize, and evaluate. that yield the highest score is the “best option”. In the
early stages of design, there will be larger uncertainties
in parameters like weight and performance. Finding how
Input/Output Model much effect variations in such parameters have is called
Sensitivity Analysis. Knowing those will guide which
Input-Output Models were first developed for quantita- areas to work on to reduce uncertainties.
tive understanding of the total flows in an economy. They If the difference in evaluation score between two con-
can be applied to any system, not just economic ones, for cepts is sufficiently more than their uncertainties, the
determining if all the inputs and outputs of a system add lower scoring one may safely be discarded. If the scores
up. It can be visualized as a spreadsheet with the elements are within the range of uncertainties, both should be
of a system as rows, and additional rows for items outside worked on in more detail until a clear winner emerges.
the system. Types of inputs and outputs are in columns. If the effort to reduce uncertainty is judged more than
Each component requires inputs such as power, data, fuel, the uncertainty reduction is worth, then one of the com-
etc. It also produces some kind of output. The purpose peting choices can be selected arbitrarily. Note that opti-
of a model is to see if your system as a whole has closure mization of a system as a whole may not mean optimiza-
and balance. In other words, are all the inputs matched tion of each individual part, since the parts can interact
by outputs? Are there missing components identified by in a complex way. Once the optimization and selection
missing inputs? Is the size or quantity of a particular el- is completed, the results are recorded and used to update
ement in the system the correct size/productivity? Will the system concept and current design configuration.
the system as a whole produce the desired output, and if
so how much? These are really all questions of account-
ing. Rather than counting everything in money, this type 2.9.5 Synthesis and Documentation
of spreadsheet does the accounting of each type of in-
put/output/resource/supply separately as categories. Note The last major step in the systems engineering design cy-
that human labor is one of the input types. cle is synthesis and documentation. In systems engineer-
2.9. SYSTEM ELEMENTS 51

ing, “System Synthesis” is assembling the results of com- In theory a WBS can be structured in any way you choose,
pleted analysis and studies into a coherent design. The but usually each level of division within the structure has
design for a complex system typically includes multiple a common basis. Examples include:
items of hardware, software, facilities, etc. Each separate
item is referred to as a Configuration Item, and the cur- • By location on Earth or in Space
rent state of that item’s design at any given time is called a
Configuration. Configuration Management is the task • By type of function, such as Production, Operation,
of documenting the current state of the design and anal- and Transportation
ysis work. This is necessary to coordinate the work for
• By type of element, such as Data, Software, Hard-
a complex design with many people involved. Otherwise
ware, Facilities, and Staff.
some work would be based on obsolete data or incorrect
assumptions. Other documents included in recording the • By end item product, such as a launch vehicle or lu-
work are Requirements, Specifications, Study Reports, nar base
Simulation code and results, 3D Models, and any other
data and notes created in the course of the work. All of • By Subsystem, such as structures, mechanical, or
this is kept as a base for further work, if later changes are electrical
needed, or if questions or problems come up. Design data
• By time sequence, such as Phase I, or Version 2 Up-
is also needed for later project stages like production.
grade
• By type of data such as drawings, analyses, or re-
Work Breakdown Structure (WBS) ports

The basis to use depends on what makes sense for the

project, but a consistent structure, such as all second level
divisions are by end item, makes the overall structure eas-
ier to understand and use. It is more important that ev-
eryone on the project use the same structure than exactly
how it is divided up. Maintaining the structure is often
assigned to systems engineering specialists because it is
related to the other tasks they perform. Each part of the
structure is given a number or identifying key, typically
using decimal points to distinguish levels, i.e 1,2,3, ... for
the top level, then 1.1, 1.2, 1.3, ... for the parts that make
up the next level below item 1, and so on. This is not the
only way to do such structuring, but it is commonly used
Figure 1.5-6: Example WBS Drawing. and easy to understand. The following section illustrates
some of the ways to arrange a given WBS level. It is not
A common method to document a system is to index all an exhaustive list.
the requirements, plans, drawings, analyses, reports, bud-
gets, work logs, and other data by a numbering system
called a Work Breakdown Structure, which covers all WBS by Item Type This example is for an automated
the elements of the system across it’s life cycle. In mod- factory that consists of operating data, software, and
ern projects the actual data is mostly stored electronically, hardware, facilities, and staff:
but a WBS helps organize and find particular items in the 1.0 Operating Data Components
same way classification systems for books are used to or-
ganize libraries. • 1.1 Design Standards
A WBS is a hierarchical table or drawing showing all the
parts of a complex system and their successive division • 1.2 Manuals and procedures
into smaller parts until you reach the level that specialty
engineers can do the detailed design. It gives structure to 2.0 Software Components
what would otherwise be an amorphous mass of design
work. The WBS serves as a tracking method and index, • 2.1 Design Software - A great deal of design soft-
so that people working on different parts of the project ware already exists. The specific need for advanced
tell they are talking about the same items. It also serves manufacturing is to design parts so they can fit the
as a method to collect and file the engineering data as it production capability of given machines, and supply
accumulates, assign tasks to individuals and groups, and processing and assembly instructions for individual
track progress and costs. The WBS is often derived from and collections of parts. This may required modify-
the functional analysis of the system. ing or adding to existing software.
52 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

• 2.2 Work Order Software - Takes an incoming prod- • Sensors

uct design in the form of CAD files, compares it to
the factory capabilities and inventory, and generates • Displays and Controls
a work order list of tasks for each machine, parts
and materials to order, etc. Work orders are then • Internal Environment
scheduled among the various components.
• External Environment
• 2.3 Machine Driver Software - Each type of auto-
mated machine requires specific driver software to • Crew Support
control how it operates, and to collect data back to
track progress and other purposes. • Maintenance and Repair

3.0 Hardware Components

2.10 Engineering Tools
• 3.1 Storage - Materials, parts, and assemblies need
to be stored when not actively being worked on. Space system engineers use a wide variety of tools to
do their work. The most important of these is their
• 3.2 Materials Handling - To transport items from own knowledge and experience, which we hope this book
one location to another. helps upgrade. The work includes producing designs for
a project; then recording the designs and other informa-
• 3.3 Production Machines - Turn raw materials into tion in formats that can be shared with other people, or
inventory stock or finished parts, possibly using sev- for use by computers and machines as input. For simple
eral machines for different steps. or early stage work, some reference books, a scientific
calculator, and a pad of graph paper may be all that is
• 3.4 Assembly Machines - Convert a collection of
needed. For the bulk of the work as done today they typ-
parts into a finished item. This will generally involve
ically require a variety of data sources, computer work-
one or more robots.
stations and more powerful supercomputers or networks,
• 3.5 Inspection and Observation Hardware - To test and specialized software. When the work gets past design
items and oversee operations. into R&D, prototyping, manufacturing, and test, they of-
ten use physical tools and specialized test equipment to
measure performance and collect data.
4.0 Facility Components - This includes modification of
the surroundings, controlling the factory environment us-
ing buildings, and supplying utilities, but not specifically 2.10.1 Engineering Data
producing any items.
5.0 Staff Components - Humans are not components to No engineering design can be done without input data of
be designed, but rather selected and trained for required some form. It can be determined internally, but more
skills, and then supplied in needed numbers. usually obtained from outside sources. Types of data in-
clude:

WBS by Subsystem Type This example is a typical set • Engineering Codes and Standards - These are
of subsystems for space hardware, and also lines up with documents that specify required or accepted meth-
design specialties: ods and features for a design. For example,
Building Codes embody accumulated experience
• Structures in how to design and build safe and sound build-
ings. Adoption of building codes by governments
gives them the force of law - they must be followed.
• Mechanical
Technical Standards are formal documents that es-
tablish uniform and accepted engineering criteria to
• Power and Electrical be followed. For example ASTM Standards for
the composition and strength of steel do not have
• Propulsion the force of law of themselves, but allow steel sup-
pliers and engineers to work together because both
know what is expected from a given alloy grade.
• Thermal
Standards may be incorporated by reference in laws,
regulations, or contracts. Large engineering organi-
• Data zations may develop their own internal design stan-
dards based on their experience, so that more consis-
• Communications tent results are obtained and new staff can be trained.
2.10. ENGINEERING TOOLS 53

• Handbooks, Textbooks, Monographs, and Jour- • Workstations - Today an engineering workstation

nals - Handbooks are compilations of useful infor- is merely an ordinary computer of sufficient speci-
mation in a particular engineering field. They are of- fications to run engineering software or to remotely
ten written by multiple contributors and updated pe- access higher performance clusters. The higher end
riodically. An example is the Handbook of Space ones may have two or more processor chips, each
Technology. For students, handbooks are often with 6 or more CPU cores each. They can also in-
quite expensive, so it is suggested to find them used clude up to 4 graphics or parallel compute add-on
or in a library collection. Textbooks are intended cards based on graphics technology. These are used
to teach a subject, like this Wikibook. Monograph for massively parallel calculations. Typically mul-
means “one writer”, books written by one or a few tiple large monitors are used, and relatively large
authors. They are typically about advanced topics, amounts of memory and hard drive storage. More
primary research, or original scholarship. Journals moderate workstations will have specifications simi-
are periodical publications containing shorter arti- lar to modern gaming systems, because game graph-
cles than monographs, reporting new research or re- ics and engineering computations both rely on mak-
viewing the state of the art. There are a vast num- ing large numbers of calculations. Even relatively
ber of books and journals covering every engineer- powerful workstations are not expensive relative to
ing topic, so it is impractical for an individual to an engineer’s salary (the software they run is a dif-
collect them unless for a very narrow specialty. A ferent matter), so the choice of hardware will be
good technical library can provide access to all these driven more by ability to run the needed software
works. than by cost.

• Supplier Data - One of the basic rules of efficient • Storage Servers - When working on complex
design is to not repeat a design if someone else al- projects, the amount of data involved can exceed
ready has. Many designs will incorporate parts or what can be stored on individual workstations, and
subsystems that already exist and are made by some- backups should be made in case of accidental dele-
one else. The suppliers of such items have literature tion or hardware failure. A storage server’s main job
and documentation about what they supply, and of- is store the extra data where it can be accessed by
ten will consult about the use of their products. anyone on the project team who needs it. That would
include a history of older versions of the design, and
• Online Data Sources - A huge amount of data is test and simulation data, which can be voluminous.
online these days, but the quality is variable. Since
incorrect data can lead to space systems failing,
• High Performance Clusters - Some types of en-
these data sources should be carefully selected for
gineering calculations require more speed than can
quality. Online data also changes quickly, so any
be reasonably installed in an individual workstation.
links we give would soon be out of date. A good ap-
High performance clusters, or Supercomputers as
proach is to use a search engine, and understand how
they are also called, group many computer chips
to define the search terms to get a specific result.
into racks with high speed data connections between
them. They run specialized software designed to
2.10.2 Computer Hardware make use of this hardware, and the fastest such clus-
ters represent the most powerful single computers in
Historically engineers worked at large drawing tables or existence. When the the need for high speed trans-
desks where they could produce the drawings and docu- fer between cores is not as great, the Distributed
ments that represented a design. Such methods have been Computing method can be used. This harnesses a
largely replaced by computer workstations for several network larger single computers, or the excess com-
reasons. Computer systems can communicate changes puting power of a number of workstations, either
much faster than paper-based methods. They can repre- off-shift or by using whatever extra processing abil-
sent designs in three dimensions, which was difficult on ity is not needed by the primary user of the work-
two-dimensional paper. And finally, computers can per- station.
form analysis and simulation of a design vastly better than
hand methods. At one point mainframes and engineering • Computer Networks - Networks are almost uni-
workstations were specialized and expensive equipment. versally used in modern engineering to transfer data
Today a basic workstation may be no different in hard- both within a project and with the rest of the world.
ware than an ordinary desktop computer, although more Since installing a network is like adding new util-
powerful computers are still used for intensive calcula- ities to a building, forethought should be given to
tions. Just as important as the workstation hardware is making it easy to upgrade, and putting in enough
the specialized software which runs on them, and the net- network capacity that it does not need to be up-
works which connect them to each other, to production graded too often. Networking protocols and hard-
and test areas, and the outside world. ware change constantly like most computer-related
54 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

things. Currently the most common method uses the spreadsheets (for simpler analysis), general numer-
Internet Protocol and routers. The protocol defines ical calculators, such as Mathworks MATLAB
how addresses for each destination and data packets for more complex analysis, computer algebra soft-
to be sent are constructed. Routers are the devices ware, such as Wolfram Software’s Mathematica
which look at the address on a packet, and send it or Maplesoft’s Maple for symbolic problems, and
towards the destination. There are many methods more specialized programs written for particular
of transmitting the data between locations, ranging fields. A more detailed list of Numerical Analysis
from Ethernet, to fiber, to wireless. In some cases it Software can be found on Wikipedia.
is faster and cheaper to send large amounts of data
in the form of tapes or hard drives, because of their • Simulation - This software category analyzes the
enormous storage capacity in a small package. behavior of a design with respect to time or changing
conditions. They can cover a single type of behavior,
such as mechanical stress, or multiple ones, which
2.10.3 Computer Software are called Multiphysics tools. These can do multi-
ple analyses in series from the same source model,
As mentioned above, engineers typically use specialized or in some cases a combined effects analysis all at
software to help with their work. The particular software once. A detailed list of Simulation Software can
will vary according the task being done. Software usu- also be found on Wikipedia.
ally evolves rapidly, so we will discuss it in terms of cat-
egories and give some examples. If working on an actual
project, a designer should find out what is the best soft- Software Resources
ware and most up-to-date versions available at the time.
In some cases, no existing software is completely suit- • Multiple Programs
able, and modified or completely new software would be
NASA Open Source Software - Repository of 240 soft-
needed.
ware projects.
Public Domain Aeronautical Software - Website with
Analysis and Simulation Software many downloads of programs, source code, and docu-
mentation.
Historically numerical analysis relied on manual meth-
ods with devices like slide rules and tables of perfor- Aerospace Software Tool Library - A list of links to
mance. With the advent of digital computers, special pur- commercial, government, and free software, sorted by
pose programs were written in mathematically oriented category.
languages such as FORTRAN. These performed calcula- Open Channel Foundation - Hosts nearly 300 mostly
tions much faster than by hand, but were still limited. The technical software applications, including a COSMIC
processing speed and memory capacity of early comput- Collection contributed by NASA.
ers limited the complexity of the mathematical models
and how many calculations could be performed in a rea- • Aircraft Design
sonable time. The fastest available processors in 2016,
which have evolved from mainframes to supercomputers CEASIOM - Software package for airplane design.
with many parallel cores, are up to a 30 billion times faster Download with registration.
than mainframes from 50 years earlier. Desktop work-
stations are millions times faster than 50 year old main- • Space Simulators
frames. So the mathematical models of a design can be
much more detailed and smaller time steps or more it- Space Engine 0.9.7.2 - Space simulation software.
erations of the analysis can be run. Parametric analysis
Celestia 1.6.1 - A 3D space simulator which can be used
allows varying parameters of the design or simulated con-
as a planetarium or for mission visualization.
ditions over a range of values. Since this requires multiple
runs of the calculations, they have become more feasible
- Celestia Motherlode - A collection of add-
with faster computers.
ons for Celestia.
What started as individual special purpose programs is - Celestia Wikibook - An online guide to the
evolving into integrated general purpose suites. This re- Celestia software.
duces the need for re-entry of model data. Often the data
can be used directly from the original design software, or
the analysis results can be fed back to the design program Design and Manufacturing Software
directly. For some projects, custom software may still be
needed where general purpose software is not adequate. These are the modern replacement for drafting tables.
They include 2D and 3D drafting, 3D modeling, and il-
• Numerical Analysis - This category includes lustration programs, and software to feed manufacturing
2.10. ENGINEERING TOOLS 55

data direct to factory machines or to vendors. Modern the process of using these kinds of machines, and the
graphics cards and processors allow direct visualization software category for producing commands and control-
and manipulation of the design in real or near real time. ling the machines. Wikipedia has a very large list of
As noted above, the design and analysis software cate- Computer-aided Technology companies and software
gories are becoming more integrated. Design category is projects.
also called Computer-Aided Design (CAD). When use
of computer workstations and mainframes was new, the
phrase distinguished it from the traditional drafting on pa- Software Development Software
per type of design. Today design on paper is a rarity, so
saying it is done with computers is mostly redundant. We
These are tools to help make software. Many end prod-
group the types of software below in terms of function:
ucts today require sensors, data transfer, and internal de-
drawing, modeling, and production.
cision making and control which requires custom soft-
ware to operate. Naturally enough, such software is de-
2D and 3D Drafting This category produces a set of veloped on computers, using Integrated Development
drawings, which in turn consist of a set of lines, curves, Environments (IDE) such as the Microsoft Visual Stu-
and text or attached notes. They are distinguished from dio suite. When such software runs inside hardware other
3D models by the drawing elements existing independent than computers, special test rigs and test software may be
of each other, and not forming more complex entities with required to test the target software, and how it functions
attached non-drawing properties. Nowadays only lower- with the intended hardware. For example, a surface rover
tier software such as AutoCAD LT or [Solid Edge 2D being sent to Mars is a unique item. So extensive testing
is restricted to 2D. would be done with software simulation and prototypes
before installing it on the flight unit.

3D Modeling This category defines the three dimen-

sional shape of an object in terms of a linked set of
points, lines, curves, surfaces, or volumes. In addi- Planning and Management Software
tion to the shape, a wide range of other parameters
may be associated with the object. Primitives, basic Complex projects have to track more than just the en-
shapes such as boxes, cylinders, or spheres, often are used gineering design. They have to coordinate the work
as starting points, and then various operations are per- of many people, do advance planning, track production
formed to modify or join them into more complex shapes. and costs, etc. Project Management Software is de-
Wikipedia has an extensive list of 3D Modeling Soft- signed to help with these tasks. Both project management
ware A few examples are: and documentation tasks can use general office software
suites, such as Microsoft Office, which has a compatible
• Autodesk Products - Originally developer of Auto- Project program. A given project can also use specialized
cad, a 2D drawing program, this company, through programs for accounting, scheduling, inventory tracking,
acquisitions and development of new software, now etc. Wikipedia has a very extensive list comparing var-
has a vast range of overlapping and linked products. ious Project Management Software packages. There
The tendency is to offer more integrated suites of are very many other pieces of business software avail-
compatible programs rather than individual ones. able. Strong consideration should be given to compatibil-
ity between programs, so that data may be moved easily
between them, rather than having to convert or re-enter
• Solidworks suite by Dassault Systemes. - This is a data.
high end commercial software set for design, simu-
lation, and data management.

• FreeCAD is an open source 3D modeling program. Documentation Software

This category is used to record all the data created in a

Manufacturing Software Modern factories use ex- project so it can be found, shared, updated, and used.
tensive computer control for their operation, which in
turn requires software to control the equipment. As
each factory is different, the software is often customized
for a given application. Computer Numerical Control 2.10.4 Instrumentation and Test Hard-
(CNC) is the name for the machine category controlled ware
this way. This was to distinguish it from the earlier man-
ual control of factory equipment, and the intermediate Physical instruments and test equipment can be grouped
numerical control, via stored commands, but without a into two categories: Those used in common with other
computer. Computer-aided Manufacturing (CAM) is industries, and those unique to space systems
56 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Common Instrumentation and Test Equipment • RF Chamber - Most spacecraft communicate

through radio frequencies (RF) and antennae. An-
Space systems projects use many of the same items tenna operation and links to the rest of the spacecraft
as other industries to test, measure, and inspect dur- are tested in am RF-shielded chamber and separate
ing manufacturing, assembly, and test. Amazon’s web- transmitters that simulate ground stations.
site has a large listing in that category, but there are
many other sources for instrumentation and test equip- • Thermal-Vacuum Chamber - The space environ-
ment. Categories include calibration, dimensional mea- ment is usually in vacuum. Hardware is subjected
surements, electrical, electronics, and software testing; to cold from the cosmic background near absolute
motion, speed, and forces; pressure and temperature, air- zero, and heat from the Sun, which is more intense
flow and air quality, inspection and testing, light, network above the atmosphere, or if the mission goes closer
and cables, recording and data acquisition, weight, sound, than Earth to the Sun. Since vacuum does not allow
and surface and hardness. The modern trend is to use heat transfer by conduction through the air, different
equipment that directly feeds computer data storage, so sides of a spacecraft can be hot and cold at the same
that manual recording of data isn't necessary. Common time. These conditions are tested in a large vacuum
tools, such as wrenches to remove an inspection panel, are chamber, which is provided with cooled walls and
also used, but normally those are available from produc- intense lamps to simulate the cold and heat condi-
tion areas and don't need to be specially provided. tions.

Beyond these devices, which are commonly used for

Special Test Equipment
whole spacecraft, special purpose equipment may be
needed for particular instruments. For example, the
Space hardware is typically exposed to two special en-
Chandra X-Ray Telescope needed a 300 meter vacuum
vironments. The first is launch on a rocket, followed by
tunnel to test the X-Ray optics from an optically distant
the conditions in space. To make sure the hardware will
source.
work properly, the hardware is subjected to a number of
tests to simulate these environments. These tests require
special test chambers to reproduce the conditions. Com-
monly used ones include: 2.11 Engineering Specialties
In Section 1.5 - Systems Engineering we discussed
• Acoustic Chamber - Rocket engines emit high
methods to coordinate the work across a large project.
pressure gas through a constriction, the engine
Complex space projects require more knowledge and
throat, and therefore function like a whistle or or-
experience than any single individual can have. Such
gan pipe, generating huge amounts of sound and vi-
projects need teams of people not only because there is
bration. The sound portion is tested in an acoustic
a lot of work to do, but also because each person sup-
chamber with powerful speakers, which play a noise
plies a different set of skills. For the design portion of a
spectrum matched to the launch vehicle the hard-
project, the field of engineering is divided into a number
ware will ride on.
of Branches of knowledge, with specialists who concen-
trate on wider or narrower portions of it. The specialists
• Shaker Table - The high speed flow from the rocket address different areas of design, and different areas of
engines, and air flowing past the rocket in flight is application to a given project.
turbulent, generating physical vibration in the ve-
Much of the actual work for current space projects hap-
hicle. This is distinguished from sound that trav-
pens on Earth, in places like offices, factories, launch
els through air. Vibration is simulated by a table
sites, and control centers. Building and operating those
that holds the hardware the same way it is held for
locations uses much of the same engineering knowledge
launch. The table is moved in all directions with
as any other large project on Earth. For the in-space seg-
powerful pistons, springs, and unbalanced masses,
ment of projects, many of the specialties are still relevant.
to reproduce the levels the space hardware will ex-
For example, nearly all space hardware has structural
perience.
parts, and their design is the province of mechanical en-
gineers. In the future, more production and construction
• Zero-G Deployment - Spacecraft often have solar will happen in space. This is in contrast to today, where
panels, antennas, and other items that are folded to mostly finished hardware is launched from Earth. So the
fit in the payload space of a rocket, then unfolded importance of additional fields like mining and industrial
once in space. The unfolding happens in zero grav- engineering will increase, with suitable modifications for
ity, and this is simulated by doing it sideways, with operation in space. In the farther future, extremely large
counterweights to remove the weight from the joints projects described as Astrophysical or Planetary En-
and mechanisms. gineering are possible. Examples are Terraforming,
2.11. ENGINEERING SPECIALTIES 57

making a body more Earth-like, or changing the orbit of a Dynamics

large asteroid. Such very large projects are mostly specu-
lative, except for the human-caused 43% increase in CO2 In general physics and engineering, dynamics is the evo-
in the Earth’s atmosphere. This is anti-terraforming our lution of physical processes with time. General courses
planet - making it less Earth-like than its original state. on this subject include:
Large-scale space projects are one way to correct this
problem if it becomes too severe. Very large projects • Dynamics (MIT)
like these are not yet organized as a distinct specialty, but
would include knowledge from many areas of science and • Aerospace Dynamics (MIT)
engineering.
One branch in particular, Aerospace Engineering, is con- Within aerospace engineering these particular areas of
cerned with the design and construction of vehicles and dynamics are important:
hardware that travel to and operate in air or space. We
discuss it first, because it is nearly always involved to some Aerodynamics - uses knowledge of fluid and
degree in space projects. It should be remembered that gas dynamics as applied to the interactions of
aerospace engineers are part of a team, and not the only atmospheres with primarily solid objects such
skill required. Although engineering has many branches, as vehicles. Vehicles going to space, or return-
they all rest on a common foundation of the sciences and ing from it, must traverse the Earth’s atmo-
mathematics, so there is some overlap between them. For sphere. Some transport methods also actively
example, aerospace and mechanical engineers analyze use the atmosphere, and some destinations in
loads on space vehicle structures the same way civil en- space have their own atmospheres to be navi-
gineers analyze them for ground structures. Where they gated. A combination of high velocity and at-
differ is in what materials are used, where the loads come mospheric density can create large forces and
from, and their operating environment. high temperatures. These must be accounted
for in design. Natural movements of atmo-
The remainder of this section lists other major engineer- spheres, such as wind and Gravity Waves, fall
ing branches with some relevance to space projects. We under the operating environment for aerospace
will not go into great detail about them, but a space sys- systems. Relevant online courses include:
tems designer, whether generalist or specialist, should at
least know what other areas exist besides their own. They
• Aerodynamics (MIT)
can then find detailed information on a topic, or find spe-
cialists as needed, when it is beyond their own area of • Aerodynamics of Viscous Fluids
knowledge. For those who want to learn more about a (MIT)
particular area, one place to start is the Massachusetts In- • Compressible Flow (MIT)
stitute of Technology (MIT) OpenCourseWare website,
which has an increasingly large collection of college level Astrodynamics - which is also known as
open source course material available (about 2250 so far). Orbital Mechanics, is the application of bal-
We list a number of their courses below, but they are not listics and Celestial Mechanics to practical
an exclusive source. Additional information can be found problems. Celestial mechanics is the branch of
through the links below, the References Section at the astronomy that deals with the motions of nat-
end of this book, and the huge number of books written ural objects in space. Space systems that are
on engineering topics. not using propulsion, nor interacting with at-
mospheres or magnetic fields, follow the same
motions as natural objects. The path a system
2.11.1 Aerospace Engineering follows, called a Trajectory or Flight Path,
can be set up with periods of natural motion
This is the primary field concerned with the design of under gravity, and periods of active propulsion.
systems which operate in the atmosphere and space. It The natural motion can include coasting be-
is further divided into Aeronautics, having to do with tween objects, and Gravity Assists, which are
flight and operation in an atmosphere, and Astronautics, close passes of a larger object to affect velocity
having to do with travel and operation in the space envi- and direction. Relevant online courses are:
ronment. The latter is the primary subject of this book.
An introductory course on this subject is Introduction to
• Astrodynamics (MIT)
Aerospace Engineering and Design (MIT). Aerospace
engineering work is also divided into specialties accord-
ing to how travel is accomplished, the environment that Structures and Mechanisms
systems travel through and operate in, and the internal
subsystems which make up vehicles and other space hard- Structures and mechanisms are the load bearing and me-
ware. These specialties include: chanical parts of an aerospace system. The primary
58 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

structure carries the main loads from gravity, acceler- in all cases is distributed through cabling, with fault pro-
ation, aerodynamic forces, etc. Secondary structure tection, switching, regulation, and control.
holds equipment items in position, which are lesser loads.
Mechanisms are the moving parts of the system, such as
joints and actuators. Typical mechanisms do steering for Propulsion Systems
a rocket engine, or unfolding and pointing of solar ar-
rays. Aerospace engineers can specialize in this design Propulsion Systems in general include a source of
area, but they have significant overlap with Structural power, and means of converting this power to propulsive
and Mechanical Engineers, who also deal with load- force. The purpose of these systems is to move people or
bearing structures. The specific knowledge for aerospace goods over some distance, usually as part of a Vehicle,
systems involves operating conditions like high accelera- an artificial carrier. Space Propulsion is the methods
tions, vibrations, large temperature ranges, and exposure of propulsion useful for space projects. Some of them
to vacuum. It also involves specialty materials to save only work in space. Others work in an atmosphere or on
weight. These are used for performance and program cost the surface of an object, and are then similar to methods
reasons. Underlying knowledge for this system includes used on Earth. A wide variety of space transport meth-
Materials Science, which covers the relationship be- ods are listed in Part 2 of this book. Not all of them in-
tween the structure of materials at atomic scale and their volve vehicles with propulsion systems. For the ones that
larger scale properties, and the selection of materials for do, chemical, electrical, nuclear, and other power sources
particular applications. It also includes Solid Mechan- are used. A variety of engineering specialties are there-
ics, which is the behavior of continuous solid matter un- fore involved in their design. Aerospace propulsion engi-
der external actions, such as forces, temperature changes, neers have specific experience in one or more of the space
or applied movements. Modern design of structures and transport methods, as opposed to transport methods used
mechanisms is usually performed with computer-aided on Earth. Relevant online courses include:
software, and increasingly is integrated in process. This
goes from defining the shape of parts, to analysis and sim- • Introduction to Propulsion Systems (MIT)
ulation, optimization, and then submitting design files to
computer-controlled factory equipment to produce them.
Once designed, a structural Test Article is often built to • Rocket Propulsion (MIT)
prove the physical version can withstand all the design
conditions. Relevant online courses for this subject in- • Space Propulsion (MIT)
clude:

• Structural Analysis and Design (MIT) Thermal Systems

• Computational Methods in Aerospace Engi- Thermal Systems have the function of keeping all parts
neering (MIT) of an space system within acceptable temperature ranges.
This includes passive methods that work because of their
inherent properties, such as insulation, coatings, ther-
• Plates and Shells (MIT)
mal couplers and isolators, reflectors, and radioisotope
heaters. Active methods use devices like electric heaters,
• Computational Mechanics of Materials (MIT) heat transfer fluids and radiator panels, mechanical lou-
vers, and thermoelectric coolers.
Power and Electrical Systems The air and space environment, and the operation of in-
ternal systems, can generate wide variations of tempera-
This system is concerned with the supply of power, and ture. The main natural source of heat is the Sun, which
electrical systems such as heaters and motors, with re- is nominally 36% more intense in space due to lack of
spect to aerospace systems in particular. It overlaps with atmosphere, but this varies according to distance. Sec-
Electrical Engineering, which is the more general sub- ondary heat can come from reflection or thermal emis-
ject (see below). Some space equipment needs high levels sion from nearby large objects. The main source of heat
of power for a short time, like during launch of a chemi- loss is the Cosmic Background, which is in all direc-
cal rocket. This can be produced by an Auxiliary Power tions in the sky behind individual foreground objects. It
Unit which uses a turbine and chemical fuel. For longer- consists mainly of the Cosmic Microwave and Cosmic
term needs, solar arrays and sometimes nuclear sources Infrared Backgrounds. Their combined effective tem-
are currently used. Selection of power sources for space perature is about 3 K above absolute zero, or −273 C.
projects requires understanding their operating environ- This is much colder than most space hardware, so more
ment, need for long-term operation without maintenance, heat is lost to the background than gained from it. Be-
and other special conditions. Once generated, the power sides the natural environment, operating hardware that
may be stored temporarily in devices like Batteries, and makes up a space system can generate heat or be very
2.11. ENGINEERING SPECIALTIES 59

cold. For example propulsion systems can do both - gen- accounted for in space systems design, and so specialists
erating large amounts of heat in a rocket engine, while in environment effects on systems and people are needed.
having cryogenic temperatures in the propellant tanks. The space environment includes all external factors that
Thermal control engineers analyze spacecraft environ- can affect a system - such as free fall (zero gravity) or
ments and operations to determine what temperatures low gravity; vacuum, rarefied, or different atmospheres,
will occur. If hardware plus natural heat results in an which can cause drag, erosion, or electrostatic charging,
unacceptable temperatures, then thermal control is re- and wide temperature fluctuations. Particular hazards to
quired, and these engineers help design the solutions. humans and space hardware are unique to space. They
include:
Heat transfer in general is a topic of physics, and
is addressed as a design task by mechanical engineers
(see below), but specific problems and conditions in • Meteoroids and Space Debris - Natu-
aerospace require specialized solutions. Some relevant ral and artificial objects that, due to their
online courses include: high relative velocity, can cause damage
on impact.
• Thermal Energy (MIT)
• Space Radiation - Particles of high
• Radiative Transfer (MIT) enough energy to damage people and
equipment. On the Earth’s surface
Control Systems we are sheltered by the magnetic field
and atmosphere from the naturally high
Control Systems Engineering is the specialty that ap- radiation levels that exist most other
plies Control Theory to design active systems with de- places in the Solar System. Sources in-
sired behaviors. Since most space systems have active clude Cosmic Rays, high energy par-
components, they also need control elements. These in- ticles and photons from the Sun, and
clude sensors and instruments to detect the current sta- trapped Radiation Belts around objects
tus, devices to transmit, store, and process the data thus with magnetic fields. Artificial nuclear
generated, methods to generate the appropriate response, devices used in space can increase radi-
and to transmit commands to other parts of the system. ation levels, and thicknesses of mass or
The commands are implemented by actuators, such as magnetic fields can shield from it.
a valve in a rocket engine that controls fuel flow. Typ-
ically control systems operate in a Closed Loop fash- Study of the space environment is part of the science of
ion. This is where a cycle of detecting status, process- Astronomy, especially Astrophysics and Planetary Sci-
ing data, issuing commands, and then detecting the new ence. These fields have benefited greatly from space sci-
status repeats multiple times. The manufactured parts of ence missions, which have allowed measuring the envi-
aerospace control systems are are often called Avionics, ronment directly. Environment effects engineers often
short for “aviation electronics”, even if used in space sys- come from the sciences first, then apply their knowledge
tems. Human-in-the-Loop control systems include hu- to space projects. Alternately, they start from an engi-
mans as part of their operation. For example, airplane neering specialty, then add the relevant scientific knowl-
pilots use their eyes, brains, and hands as part of the con- edge, often at a graduate level.
trol loop.
Control systems are used in everything from traffic lights Life Support Systems
to chemical plants, so the subject is taught to all kinds of
engineers. Aerospace controls specialists deal with theThe space environment is generally hostile to humans,
particular problems of air and space operations, such as
and life in general. For space projects that involve living
control of a rocket in flight, or a space system’s robot arm.
things, a Life Support System is then needed to pro-
Some relevant online aerospace courses include: vide suitable conditions. Basic structure contains an at-
mosphere, and thermal control keeps temperatures in a
• Principles of Automatic Control (MIT) habitable range. These were covered previously. Beyond
this, humans need the right atmosphere mix, water, and
• Feedback Control Systems (MIT)
food. Because of the closed environment, liquid, solid,
and gaseous wastes (including especially CO2 ) must be
• Prototyping Avionics (MIT)
removed, both from people and other system operations,
and microbes controlled. In free fall conditions air does
Space Environment not circulate by convection, so circulation fans are needed
to prevent “dead zones” where harmful concentrations of
The Space Environment has a number of significantly gases can accumulate. Biological systems can be used to
different conditions than found on Earth. These must be recycle wastes and supply food, like they do on Earth, but
60 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

this is still in early research. Most life support systems to products. Testing uses actual materials, components,
date have single-use supplies of food, and limited recy- subsystems, up to completed products. Test can be
cling of other materials. in simulated environments, like a Vacuum Chamber,
Engineering of life support systems is cross-disciplinary, or in the actual operating environment. For complete
involving both biology and mechanical systems. Spe- aerospace vehicles, the latter is called Flight Test. For
cialists therefore come from areas like Bioengineering very large and complex systems, such as the International
and Mechanical Engineering (see below). They typically Space Station, there was no way to test it as a whole. In-
learn about the specific design of life support on the job, stead its parts were tested on the ground, extensive anal-
ysis was performed, and the ability to repair and update
because there are so few examples of life support for
space that it has not developed an educational path yet. parts as needed had to suffice.
Related work is done for airplanes, high altitude climb- The Hardware and Software test planning should ideally
ing, and working underwater. Life support systems can start early in a project. Specialists in these areas are often
be large enough for a number of people, or small enough drawn from the respective design fields, and from Systems
for a single person, as in a space suit. Engineering experience.

Human Factors 2.11.2 Other Engineering Specialties

Human Factors, also is the aspect of design that takes The following are major conventional divisions of the
account of the interaction of a system and the people who engineering field into specialties, listed alphabetically.
use it. People can't be designed the way a piece of hard- Knowledge in general does not have such divisions, they
ware can, so the system design has to accommodate their are made by humans for historical and teaching purposes.
capabilities and limitations. The people may be passive, A given engineer may have knowledge and experience
as in airplane passengers, or active, as in the flight crew. that spans across multiple specialties or is concentrated
Subject areas in this field include physical, cognitive, and in a narrow area within only one. Where we listed indi-
organizational interactions. Because of the unique con- vidual courses for specialties within aerospace engineer-
ditions in space, like having to do repairs while in zero ing above, the larger engineering fields tend to have whole
gravity and a bulky space suit, or working with a large academic departments or even entire institutions devoted
support team on the ground, the various subject areas to them.
have assumed importance in aerospace. Human factors
also includes topics like how to design control inputs and
information displays for zero gravity or high acceleration, Bioengineering
and how to maintain crew training on a long mission.
Biological Engineering, or “Bioengineering” applies
The roots of human factors design extend as far back as knowledge from the biological sciences towards satisfy-
people have made tools, since the tools must fit our hands, ing human needs. This includes producing food, mate-
and the strength we have to wield them. The modern de- rials, energy, maintaining human health, and the natural
velopment draws from disciplines like psychology, engi- environment. As an engineering field it has developed
neering, biomechanics, industrial design, physiology, and rapidly since about 1960, because of the increased un-
anthropometry. A relevant online course is: derstanding of genetics, and development of tools to ma-
nipulate it. The MIT Department of Biological Engi-
• Human Factors Engineering (MIT) neering lists a number of open courses on this subject.

• Human Control of Automated Systems (MIT) Agriculture

Agricultural Engineering is the subset of Bioengineer-

Simulation and Test
ing concerned primarily with food, wood, fibers, biofuel,
medicinal, and other material products produced on land.
Aerospace systems are often complex, and therefore ex-
As a human activity, Agriculture extends back to the ori-
pensive. The environment and operating conditions in
gin of civilization. As one based on scientific knowledge
which they operate are different and more severe than
it has greatly improved in the last 200 years, and as an
those normally found on the Earth’s surface, and failures
engineering field extends back to about 1900. Growing
can be catastrophic. So to ensure basic functionality,
plants in space is at an early experimental stage. This
safety, reliability, and meeting design requirements, the
is expected to increase greatly in coming decades, since
Simulation and System Test specialties have developed.
growing food and other products, and production of oxy-
Simulation includes early mathematical and computa- gen as a by-product, would greatly reduce supplies needed
tional modeling, and physical scale models or functional from Earth.
simulators. They reproduce important aspects of a sys-
tem, but do not use the actual hardware and software Biomedical Engineering
2.11. ENGINEERING SPECIALTIES 61

Biomedical Engineering applies engineering principles delivery from orbit. As one of the larger fields, it has a
and design concepts to medicine and biology for health- range of courses, such as at MIT’s Department of Elec-
care purposes. It is a relatively new field with a heavy em- trical Engineering and Computer Science
phasis on research and development. For space projects,
it is applied to keep crew and research animals in good
condition, and also a subject of research on the effects of
space conditions. Industrial Engineering

Industrial Engineering deals with the optimization of

Chemical Engineering complex processes or systems, and improvement of pro-
ductivity, quality, efficiency, and profitability of busi-
Chemical Engineering is the application of physical and nesses. Traditionally it encompassed planning of indus-
life sciences, and applied mathematics and economics, to trial production systems (factories), but has broadened in
produce, transform, transport, and properly use chemi- scope to a wide range of complex operations. Industrial
cals, materials, energy, and useful products. It is a broad is distinguished by being concerned with the whole busi-
field with many specialized branches, and typically has a ness and supply chain, where manufacturing engineering
full set of courses at schools such as MIT’s Department (part of mechanical below) is concerned with individual
of Chemical Engineering. Historically, the relevance to production machines and tasks.
space projects has been supplying propellants for rockets,
and the alloys and other materials to build them and their
payloads. In the future, extracting fuel and other prod-
ucts from space locations will require modified versions Mechanical Engineering
of chemical plants.
Mechanical Engineering applies the principles of
Civil Engineering physics and materials science for the design, analy-
sis, manufacturing, and maintenance of mechanical sys-
Civil Engineering deals with the design, construction, tems. Specialized mechanical systems for space are listed
and maintenance of the built environment, including above under aerospace engineering. Other specialties in-
roads, bridges, canals, dams, and buildings. As an empir- clude acoustical, manufacturing, thermal, sports, vehicle,
ical practice it is as old as Civilization (both words derive power, and energy engineering. Mechanical systems such
from the Latin civitas, or city). As a scientific and techni- as turbopumps and valves are core elements of rocket
cal field it began around 1800. For space projects to date engines. Wheels, suspensions, drive motors, and robot
it is mostly related to factory locations and launch sites arms enable rovers to perform complex tasks on planet
on Earth. In the future it will be applied to construction surfaces. As a modern technical field, mechanical engi-
at locations beyond Earth. As one of the older engineer- neering has grown since about 1800, with the develop-
ing fields, it typically has a full set of courses, such as at ment of Machine Tools and engines to drive them. As
MIT’s Department of Civil and Environmental En- one of the largest engineering fields today, it has a very
gineering. Branches of civil engineering include Envi- large range of courses, such as at MIT’s Department of
ronmental, Geotechnical, Structural, Transport, and Wa- Mechanical Engineering
ter Resources Engineering. Since all the materials used
to build with must be obtained from the natural environ-
ment, it is closely related to Mining Engineering (see be-
low). Mining Engineering

Mining Engineering is concerned with extracting and

Electrical Engineering processing minerals from the natural environment. Mod-
ern civilization uses vast quantities of materials. So this
Electrical Engineering deals with the applications of field of engineering, and closely related sciences, have
electricity, electromagnetism, and electronics. It became developed to support those uses. At present, mining on
an identifiable field after about 1850. Most space sys- Earth is not directly related to space projects, as raw ma-
tems have electrical components, and the specialized as- terials are normally processed by intermediate factories
pects are noted above under aerospace engineering. More and plants before being used to build hardware. This
general electrical engineering is used today in supply- may change with the development of integrated and au-
ing power to offices, factories, and launch sites for space tomated production systems. In space itself, extracting
projects. It is also used in computer hardware, software, materials is of great interest to avoid the high energy and
and communications networks for development and op- financial cost of launch from Earth, and making available
eration of space systems. Future uses include electric new sources of materials and energy. Mining in space is
launch and space propulsion methods, and solar power still in the research and early development stage.
62 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Nuclear Engineering essary resources, which is often in the form of money.

The tools and methods of economics are therefore used
Nuclear Engineering is concerned with applications of to analyze, justify, and obtain the needed money and re-
nuclear processes like fission and fusion. These processes sources. Organizations and economics are extensively
can release very large amounts of energy. On Earth the studied at business schools, such as the over 100 courses
primary use is for power generation in Nuclear Reac- at MIT’s Sloan School of Management, and there are
tors. Some space missions today carry sources based many textbooks on these subjects.
on radioactive decay. In the future fission and possibly
fusion reactors can supply power when solar is insuffi-
cient. Nuclear engineering deals with high energy radia- 2.12.1 Organizational Structures
tion, which also is part of the natural space environment.
Although a fairly new field, dating to the 1940’s, it has de- There are a number of organizational structures that have
veloped extensive specialized knowledge, such as courses been used or are possible. Which ones are best suited to
from MIT’s Department of Nuclear Science and En- a project or program depends on the complexity, dura-
gineering. tion, and life-cycle stage of the tasks. Structures may be
changed or replaced over time as the needs change. Hu-
mans are social entities, and their individual needs, de-
Software Engineering sires, and motivations differ. This has to be considered
along with technical questions like what kind of desks and
Software Engineering is the systematic application of workstations to use. Membership in an organization is
engineering methods to the design, development, and usually not permanent, either because the project’s or the
maintenance of Software - the changeable instructions individual’s needs change. Continuity of a project has to
and data that computers use. This field is young, only be supported when this happens.
developing from about 1950, but has rapidly grown due
to the rapid advance of computer hardware and elec-
tronics in general. Software is an integral part of mod- Structural Types
ern space projects, from initial concept formulation us-
ing Productivity Software, to collection and analysis of For a given project, you can adopt an existing structural
mission data and control of remote spacecraft. Essen- type. For an ongoing program or business that structure
tially all engineers use software, and many develop soft- may already be in place. You can then modify the exist-
ware as part of their work in other specialties. However ing type to accommodate the specific needs of the project.
creating complex and reliable software, where the conse- Alternately you can design a custom organization in the
quences of errors mean loss of life or expensive systems, same way as you design a custom piece of hardware. The
requires teams of specialists dedicated to the task. Soft- choice will depend on the potential gains from customiza-
ware engineering is closely related to Computer Science, tion, versus the lower adoption cost of an existing type.
and their teaching is often combined in one set of courses, Existing structural types include:
like at MIT’s Department of Electrical Engineering
and Computer Science.
• Entrepreneurial - One or more founders or lead-
ers make decisions, and communications are direct
person-to-person. Staff are added on an ad-hoc ba-
2.12 Economics sis, and tasks are not rigidly defined. This type is
suited to new projects at an earlier stage of their
The Systems Engineering approach described in Section life cycle, where many decisions have not yet been
1.5, along with the tools and engineering specialties in made, and change is rapid.
sections 1.6 and 1.7, cover parts of how design gets
done. Although not strictly part of the design process, • Bureaucratic - This may be thought of as a machine
a large and complex space project also needs an effec- made of people. The structure has a degree of stan-
tive Organizational Structure to function. An organi- dardization, with defined roles and responsibilities,
zation allocates tasks, coordinates the work, makes deci- a hierarchical (pyramid) structure, and respect for
sions, supplies the various resources and inputs needed, merit. It is suited to larger and more complex or-
and provides support to the core design, production, de- ganizations. Since one person can't comprehend all
livery and operations tasks. Individual projects may be the details or have time to make decisions for a large
part of a larger program or ongoing business, and organi- organization, information flow and instructions are
zations must then sequence and allocate between them. compartmentalized through the pyramid structure,
Economics is the social science that studies produc- with a given person responsible for actions at their
tion, distribution, and consumption of goods and services. level, and one level up and down. Rules and stan-
Space projects are then an economic activity, and part of dards make tasks routine, so decisions are reduced
a general economy. Large projects need a source of nec- to exceptions or changes. The accumulated struc-
2.12. ECONOMICS 63

ture and rules makes this type harder to change when well known psychological cognitive bias of Illusory Su-
needed. periority, where people in general misjudge their ability
relative to others. It also partly is due to the belief that
• Networked/Consensus - In the previous two types, reaching a high position by itself demonstrates superior
decision-making is concentrated in a few leaders. knowledge and ability. Lastly, there is the fear that any
In a network or consensus structure, decisions are changes will break how the existing organization oper-
reached by multiple interested parties. Communi- ates. This is a fallacy because organizations in general
cations flows to and from any point in the network are capable of doing things no single individual can do,
as needed. This is suited to more rapid and ad- like design an entire airplane or computer operating sys-
hoc change, but increases effort to make decisions tem. Therefore as a whole they should be more capable
because more people are involved. However, in- of designing the organization itself than any single indi-
creased information inputs can lead to better deci- vidual within it. Nonetheless, these various barriers exist
sions, and wider participation can lead to better ac- and should be recognized.
ceptance and enthusiasm. If an organization is new, or being organized for a specific
project, most of these barriers cam be overcome. Then
Structures are not usually pure examples of any of these a rational application of the systems engineering process
types. For example, bureaucratic ones may have boards can find the best organization structure. This can be one
or committees which reach decisions by consensus, and of the existing types, since they exist for good reasons.
organization members may communicate through an in- But choosing a structure by design is better than choosing
formal network in addition to a formal pyramid structure. one by personal preference or random accretion, since it
The parts of a structure may be defined by specialized is more likely to produce a better result. Another way to
functions, such as production, marketing, staffing, and implement change is to use an outside organization to de-
accounting. It may also be divided by location or prod- sign the new structure, present their results, and then have
uct, where each part includes multiple functions. Again, the old organization adopt it as a whole. Approached in
mixed examples are possible for a given organization, this way, individual objections can be overcome, but it
with some parts functional and others location or product- can also induce loss of staff beyond that caused by in-
based. creased efficiency.

Designed Organizations
2.12.2 Project Management
An organization capable of designing and building com-
plex space projects is usually itself a complex entity. Project Management is the discipline of initiating, plan-
Therefore systems engineering methods can be applied to ning, executing, controlling, and closing the work of
the design of the organization itself, as well as to the space a team to reach specific goals and success criteria. A
systems they produce. In this approach, the organization Project has a defined goal and end point, whereas a
is treated as a system, where inputs like staff time, of- Program is a larger effort that may include multiple
projects. Programs may be open-ended with no defi-
fice buildings, and factory buildings produce outputs like
launch vehicles and spacecraft. By comparing the outputs nite end point, such as “explore the Solar System”. Pro-
grams use more permanent organizations such an agency,
to the inputs, you can decide if a given project is worth
doing. The internal functions and flows within the orga- research institute, or business enterprise. Individual
projects can draw from these organizations, and later re-
nization determine the relationship of inputs to outputs,
and so are subject to analysis and optimization. turn staff and other resources to apply to the next project.

Designing an organization this way is not often done for Historically, skilled and experienced individuals such
several reasons. First, organizations often have a history, as architects, engineers, and senior craftsmen led large
and existing structures and methods in place for how they projects, which were usually in construction. By the
operate. These may have developed before they started 1950’s, project management had become a distinct dis-
on space projects, and even before the systems engineer- cipline from these skilled fields, with its own tools and
ing method existed. If existing organizations and meth- techniques. There are a number of approaches to manag-
ods have worked well enough, the need to optimize or ing a project, including applying systems engineering to
update them may not be felt. Second, people are often the whole project. The rapid growth in complexity and
resistant to change, especially when that change involves scale of projects has driven the creation of new methods
re-designing or eliminating their own job. Some people to manage them, and this is an ongoing process today.
are attracted to the power or money that comes from a po- New tools like computer networks and software enable
sition within an organization, and so oppose any change some of these new methods. See Wikipedia’s Outline of
that would displace them from that position. Third, many Project Management for a more detailed overview of
people in a high position believe they know better how the field
to organize and run things. This results partly from the A Project Manager is a person given responsibility for
64 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

part or all of a project’s execution. Traditionally man- use government-owned launch sites, and partnerships be-
agers are placed on top of an organization structure, but tween sectors are common. The differences involve
that is a matter of history, not necessity. They com- sources of funds and objectives, but not as much the tech-
bined several functions such as planning, decision mak- nical issues. A rocket engine works the same way no mat-
ing, and coordination, in an era when communication was ter who pays for it. What kind of rocket engine gets built,
more difficult. Modern computer networks allow every- however, may depend on available funds and who is the
one on a project to have immediate access to all project source.
information, so the need for top-down hierarchies is re-
duced. Quite large and complex projects, like the Linux
operating system, have been developed mostly with self-
directed work. However, many space projects require Government Funding
dedicated specialists, large buildings, and other resources
that are difficult to obtain on individual initiative. Governments were the first large source of funds for space
Typical project management processes are: projects, growing from military uses like ballistic mis-
siles and spy satellites. This was followed by scientific
and earth observation missions such as weather satellites.
• Initiating - which includes identifying needs and
These uses are intended for the general public benefit, and
requirements, reviewing existing operations, finan-
so obtain their funds through taxation or appropriation.
cial analysis, stakeholder analysis, and establishing
Governments decide how to apply the funds to projects
a project charter.
by political processes. This is typically by means of an-
nual budgets, where space projects are part of more gen-
• Planning - of time, cost and resources for the eral government funding. Although theoretically for the
project tasks. This includes a breakdown of deliver- public good, in practice other factors enter into political
ables and organizing a logical sequence of the work. decision-making. Typically space projects are carried out
by a mixture of internal government staff and facilities,
• Production and Execution - consists of following and outside purchases of goods and services.
the plan, allocating staff, outside supplies, physical
space, and other resources as required.

• Monitoring and Controlling - includes observing Business Funding

production and execution to identify and correct
problems, and measure performance and variances Businesses normally operate to generate profits for their
from the plan. owners. Early space projects were high risk, and so
were mostly carried out by governments. Businesses were
• Closing - includes formal acceptance of the project hired to carry out parts of these projects because of spe-
outputs, ending contracts, reassigning resources, and cial skills or production capacity the government did not
archiving data and knowledge gained. have. Since government guaranteed payment, this low-
ered the risk and businesses were willing to participate.
Communications emerged as a profitable space business
2.12.3 Project Funding because information has no mass, and so the high cost
of space transport did not affect that business as much.
Large projects need a way to obtain all the resource in- Space industry beyond government projects is now the
puts needed for their completion. An existing organiza- dominant economic sector. In the future, additional types
tion or program may have some of these resources, like of space businesses are expected to grow as transport
land and buildings, in place already, but nearly always costs come down. Businesses obtain their funds from in-
some additional ones need to be supplied from outside vestors, retained previous profits, and finance on the open
the project. In modern economies, Money is a gener- market.
ally accepted medium of exchange, which can be traded
for all these resource types. This simplifies the resource
provision task to obtaining enough funding in monetary
form. There are a number of different sources and ways Social Funding
to obtain project funding.
For space projects, the sources can be divided into three Social funding comprises non-profit foundations, re-
main sectors - government, business, and social - which search organizations, private donations, and volunteer
relate to major parts of economies in general. We will efforts. Large terrestrial telescopes are an example of
list them individually, but a mixture can be used for projects funded by a mixture of social and other sources.
a given project. As examples, governments often hire This type of funding is at a relatively small scale com-
businesses to build parts of space projects, businesses pared to government and business.
2.12. ECONOMICS 65

2.12.4 Financial Analysis thing now than in the future, a phenomenon known
as Intertemporal Choice
Whatever the source of funding, large projects have to
justify themselves to be approved, estimate the funding
• The Interest Rate is a measure of how much more
required, optimize the use of funds, and implement the
value the present has over the future. It is the per-
acquisition and distribution of funds during the project.
centage increase per time, usually years, needed to
Various tools from economics and Accounting are used
equate a future value to a present one. In other
for these purposes. Money is often used as a common
words, how much larger a future value do you de-
measure to compare disparate resource and output types.
mand to delay a present use? Reversing the time
However it should not be the sole measure to evaluate
order, a future value is reduced by a percentage per
projects. Measures of Effectiveness, as described in Sec-
time, called the Discount Rate, to equate it to the
tion 1.5, are a way to evaluate non-monetary factors like
present.
performance and safety along with monetary ones like
cost and revenue. These kinds of measures can also be
applied at the organization level. Costs and benefits of a space project may extend for a
number of years, and costs generally happen before ben-
Finance is the science of money management. Among its
efits. Therefore both are equated to Present Value by
other uses, money is a measure of value to the end users
applying the discount rate. Which discount rate to use
of a project, and a measure of the resources a project will
depends on who is providing funding. A government gen-
consume. The concepts and formulas from Economic
erally has a lower one than a business because they assume
analysis are useful in project decision-making and op-
a longer time horizon to collect benefits.
timization. If there is too little value in relation to re-
sources, a project may never happen at all. Resources
include human labor, physical resources, and intangibles Financial Estimating
such as rights to use something. The resources to run the
organization itself is often a major part of total project Future costs and benefits are not known exactly, espe-
resources. So organizational efficiency is important to cially when a project includes developing new technol-
reaching desired program cost and schedules. ogy. Since these predictions are often key in deciding
Project Finance is obtaining money for a project from if a project happens at all, or which design options are
other than internal sources of the sponsor. This is often used, it is very important that they have as sound a basis
necessary for new organizations, and for large projects as possible. Therefore various estimating methods have
which exceed the scale of previous organization efforts. been developed to predict the financial flows for a project.
Financial estimates are developed before money is actu- They vary in detail and accuracy based on stage of the
ally required to determine if a project is worth starting, project. Regardless of which method is used, some mar-
and how much funding will be required. Actual spending gin or range should be included in any estimate to allow
during a project is tracked to compare progress to esti- for errors in estimating, or unanticipated costs. Methods
mates, in order see if additional funds are needed or if include:
there is surplus to apply elsewhere.
• Historical Costs - This is based on similar past
Life Cycle Cost Analysis projects, and uses units like number of person-
hours, drawings, or lines of software. These are
Life Cycle Cost Analysis is an economic method that multiplied by historical rates for the given param-
accounts for all costs of a project, from initial concept to eter.
final disposal. It parallels the general Systems Engineer-
ing method of considering the entire life cycle of a project • Parametric Estimates - These are formulas for an
from start to finish, but focuses on the financial aspects. entire finished item using multiple parameters such
The method applies some basic financial concepts: as mass and stress level, and factors such as “sim-
ple design” or “complex design” to modify the basic
• Time Value of Money - Human life is finite. There- values.
fore the amount of time you can expect to use re-
sources, or their equivalent in money, is shorter at
• Bottom-Up Estimates - This approach makes esti-
a future date than it is in the present. Changes be-
mates for individual parts of a project at a detailed
tween now and a future date may affect the desir-
level, then summing them to get the total cost.
ability of any given item. For example, an old com-
puter that may have been useful at one time has be-
come less so due to changes in technology. Inflation • Supplier Quotes - A supplier or sub-contractor who
can also affect the future value of money itself. For makes a definite quote on part of a project replaces
these reasons, humans attach greater value to some- an estimate with a specific number.
66 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

Large space projects typically have many cost compo- can be used in space, but that is not the field’s main goal.
nents, and use specialists for cost estimating and track- For the many such fields, which are less directly related to
ing, with tools like spreadsheets to integrate the data into
space, we will not list them here. If you need information
useful summaries. about these subjects, there are many online, library, and
NASA has a Cost Estimating Handbook developed in- academic sources which can be used as starting points to
ternally for space projects. The International Society of research them. The other fields are important, however,
Parametric Analysts has a Handbook for more general in that general improvements in science and technology
projects. affect what you can do for space projects.

2.13.1 Government Programs - Interna-

2.13 Existing Programs tional

A space systems designer should be aware of existing We will start with the major national and international
projects and programs in order to avoid duplication of agencies outside the United States, approximately in or-
effort, know the current state of the art, serve as a base- der of annual budgets. It is approximate because cur-
line technical level to improve upon, and as a compari- rency exchange rates fluctuate, and so do agency budgets
son point to justify the need for new projects. Projects in terms of their own currency.
and programs can be categorized by type of fund-
ing - government, commercial, or not-for-profit sectors;
time - past, current, and proposed future; and by type ESA
of project - science and exploration, communications,
tourism, mining, etc. There is significant overlap and in- The European Space Agency (ESA) is an intergoven-
teraction between the various sectors and projects. For mental organization of 22 member nations. It’s budget
example, government programs procure equipment and in 2015 was 4.43 billion EUR, or $5.15 billion dollars.
services from commercial entities, and contribute fund- It was established in 1975, and operates a spaceport in
ing to not-for-profit programs. Another overlap is that French Guiana, designs launch vehicles, a number of in-
launch vehicle programs provide transportation for dif- dependent spacecraft, and participates in international
ferent payload projects. projects and programs. The spaceport is located outside
Wikipedia has extensive article lists for Spaceflight in Europe for safety and launch performance reasons. Mem-
general (see also the Category, Portal, and WikiProject ber nations typically split their space program funding be-
links on that page), more specific Lists and Timelines, tween ESA and domestic projects.
and categorized Spaceflight Navigation Boxes (used in
articles to link to related articles). These can be used to
get general background on the various projects and pro- Roscosmos
grams. According to the UCS Satellite Database, as of
August, 2015, there were an estimated 1305 operational Roscosmos State Corporation is the current space sci-
spacecraft in Earth orbit. There are also approximately ence and aerospace agency for Russia. It’s budget in 2015
20 active Solar System Probes beyond Earth orbit, and was 186.5 billion RUB, or $2.4 to 3.7 billion in dol-
6 spacecraft at Sun-Earth Lagrangian Points. The lat- lar terms. The dollar number is variable because of se-
ter does not include the many natural object at various vere exchange rate fluctuations. Prior to the dissolution
Lagrangian Points. In addition to active spacecraft, there of the Soviet Union, the Soviet Space Program had a
is a large amount of human-made Space Debris, nearly long history starting in the 1930’s, with notable theoret-
all in Earth orbit. They can be considered either a haz- ical work starting before that. The program had many
ard or a resource. Finally there are all the natural objects notable firsts and continues today in areas like participat-
of the Solar System, which can be considered as destina- ing in the International Space Station (ISS). During the
tions for exploration, resources to be used, or hazards to Soviet era, military and civilian space activities were not
be dealt with, such as radiation belts and sources. separated, and were very secretive.
As of 2015, there are seventy civilian Government
Space Agencies which pursue their own space projects
and programs, or participate in some way. There are also JAXA
a number of military programs and projects, but due to
secrecy, their number and activities are less certain. The The Japan Aerospace Exploration Agency is the na-
following sections cover some of the more important pro- tional aerospace agency for Japan. It’s budget in 2013
grams, but are not exhaustive. It mainly lists those that was 211 billion JPY, or $2.0 billion dollars. It has built
are directly related to space. Other fields, for example a number of space science missions, and participated in
Materials Science, can develop new structural alloys that international projects like the ISS.
2.13. EXISTING PROGRAMS 67

2.13.2 Government Programs - United space environment), planetary science, and astrophysics.
States Finally the Human Exploration and Operations Di-
rectorate develops and operates the very well known
The US government is a major funding source and oper- manned programs, and less well known science and tech-
ator of space projects. The funding ultimately originates nology projects related to human exploration. The divi-
from the nation as a whole through taxation. Funds are sion of funding (in US dollars/year) in 2016 and major
appropriated annually through the US Congress to sev- projects include:
eral Executive branch agencies. Activities of the US gov-
ernment in general are covered through by the annual Scientific Research ($5.59 billion)
Aeronautics and Space Report of the President, is-
sued annually. Budget figures in that document may not NASA Science programs include Earth Science, Plane-
include classified military programs, so we can say the tary Science, Astrophysics, and Heliophysics project di-
government total is $30 billion or more. visions, and also satellite work NASA performs for other
government agencies (NOAA and USGS). The science
and technical parameters of astronomy and planetary sci-
NASA
ence projects funded by NASA are listed below under
not-for-profit programs for several reasons. The scientists
The National Aeronautics and Space Administration,
who are the end users of the data generated are often at
or NASA, is the primary civilian space agency of the
universities and other not-for-profit institutions, the ter-
US government. It was originally created in 1958, and
restrial projects often have multiple funding sources, and
currently is authorized under Title 51--National and
programs in the sciences are better understood as a whole
Commercial Space Programs (Public Law 111-314, 18
rather than divided by funding agency. The budget and
Dec 2010). Funding is appropriated annually by the US
organizational details are noted in this section.
Congress to operate NASA. For 2016 it is contained in
H.R. 2029 at pages 74-77. NASA is headquartered in Earth Science - This division is directed at studying the
Washington, D.C. and currently has 17,345 staff and a Earth, but the spacecraft hardware and instruments are
number of Centers and Facilities. A large part of its ap- often similar to what is needed at other locations. In
proximately $19.3 billion annual budget is spent on con- addition to the specific projects noted below, the divi-
tracted work. Wikipedia has an extensive List of NASA sion funds data archiving, organization, and delivery, ad-
Program Articles, and the NASA Missions Page lists vanced technology development, and application of the
past, current, and future missions. data to human needs.

• Earth Science Research - Scientific re-

Past Projects When the United States decided to cre-
search aimed at understanding the Earth,
ate NASA as a civilian agency in 1958, it inherited facil-
its components, interactions, changes,
ities and staff from the Department of Defense and the
and how these affect life. It includes
previously existing National Advisory Committee for
a substantial effort at high performance
Aeronautics, which dates back to 1915. The agency’s
computing for modeling and analyzing
History Office published a self-history covering 1915-
the large volume of data generated by
1990. It was originally in paper form as NASA special
satellites and other instruments.
publication SP-4406, then online as Orders of Magni-
tude, and also available as a PDF file. Numerous other • Earth Systematic Missions - This is
history publications have been produced by this office. a collection of Earth-observing research
Care should be taken to recognize bias in any internally satellite missions which collect different
written history, but factual data is likely to be correct. types of data on the Earth as a whole.
Some of the projects are jointly between
NASA and other US agencies, or other
Current Projects The Current and Recent Budgets national agencies. New satellites include
page provides data on recent and current activities from Surface Water and Ocean Topography
a financial standpoint. The 2017 Complete Budget Es- and NASA-ISRO Synthetic Aperture
timates from that page, and similar ones for past years Radar. It also includes operating nu-
under the “Previous Years’ Budgets” section, also provide merous current satellites and instruments,
detailed descriptions of on-going projects. and scientific analysis of the collected
NASA operates under three main mission directorates. data.
The Aeronautics Research Directorate operates a num- • Earth System Science Pathfinder -
ber test facilities and research projects. Some of these These are small to medium projects and
have application to space launch. The Science Mis- satellites aimed at collecting Earth data,
sion Directorate manages numerous projects and mis- using new and emerging priorities and
sions related to Earth science, heliophysics (the Sun and technology.
68 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

• Venture Class Missions - This includes and presence of life on extrasolar planets. It funds gen-
suborbital, small missions, and instru- eral research and data analysis from NASA missions, as
ments for earth science investigations. well as specific programs as follows:

Planetary Science - This division is directed at study- • Cosmic Origins - This program inves-
ing the content, origin, evolution, and potential for life tigates the evolution of the Universe and
of our Solar System and planetary systems beyond our its components. Major projects include
Sun. To get better data than observing from a distance the Hubble Space Telescope, SOFIA
(i.e. astronomy) it has a progressive strategy of flyby, or- airborne telescope, and Spitzer infrared
biting, landing, roving, and sample return for each object. telescope.
Generally astronomy is how original detection of new ob- • Physics of the Cosmos - Investigates
jects occurs, although the Near Earth Object Program how the universe behaves under extreme
within this division is directed to find close and possi- conditions. It supports two operating
bly hazardous objects down to 140 meters in size. It missions: the Fermi gamma ray tele-
funds general science research and technology develop- scope and Chandra X-ray observatory,
ment projects, plus specific groups of spacecraft and mis- along with supporting science and tech-
sions noted below. nology work.
• Exoplanet Exploration - Searches for
• Discovery Program - This includes rel- and tries to characterize planets outside
atively smaller missions with shorter de- our solar system. Much of this re-
velopment times. It includes the Dawn search happens with ground-based tele-
spacecraft studying Ceres, InSight Mars scopes such as Keck and LBT, but this
lander, and Lunar Reconnaissance Or- program includes the Kepler space tele-
biter mission. scope, which is a dedicated planet find-
• New Frontiers - This includes medium ing mission. Work on the WFIRST large
size, high priority science missions, in- space telescope is in the early stages. The
cluding the New Horizons Pluto flyby, program includes science and technology
Juno Jupiter orbiter, and OSIRIS-REx development for future instruments and
NEO sample return missions. missions.
• Astrophysics Explorer - This program
• Mars Exploration - This program seeks
funds smaller, dedicated science mis-
to understand the past, present, and fu-
sions. Current projects include the TESS
ture habitability of Mars through a se-
exoplanet survey telescope, participation
ries of incremental missions. It in-
in international projects, and continued
cludes continuing operation of the Mars
operation of some previous missions.
Odyssey and Mars Reconnaissance or-
biters, MAVEN atmosphere sampler,
and Opportunity and Curiosity surface James Webb Space Telescope - This is a 6.5 meter di-
rovers. In development is the Mars 2020 ameter red to infrared space telescope, the largest of its
Rover. kind to date. It is currently planned to launch in October
2018. The extreme cost of this project ($8.8 billion total)
• Outer Planets and Ocean Worlds - has generated congressional scrutiny and separate budget
This program is aimed at understanding tracking from the rest of the Astrophysics science budget.
the origin and evolution of the outer solar
system. It includes the current Cassini Heliophysics - This division studies the Sun, the space
mission to Saturn, development of the environment between bodies, and that surrounding plan-
Europa Mission to study that moon of ets. Like the other science divisions, it funds topical
Jupiter, scientific research, and planning science research, data analysis and archiving, and more
for future missions. specific programs noted below. Sub-orbital launches
and continuing operation of previously launched satellites
• Technology - This program develops im- are also funded by this division. This includes the two
proved technology for future planetary Voyager spacecraft, which have completed their plane-
missions. It currently includes projects in tary science work and are now primarily collecting data
electric propulsion, radioisotope genera- on the space environment.
tors, and mission management software.
• Living with a Star - Studies the interac-
Astrophysics - The areas of research for this division tion of the Sun and Earth and its effects
include the Universe as a whole, the evolution of galax- on life and society. Current projects in-
ies, stars, and planets, and the characteristics, habitability, clude the Solar Probe Plus which will fly
2.13. EXISTING PROGRAMS 69

very close to the Sun, and the joint Solar The actual work is done within NASA centers, and the
Orbiter mission with ESA. commercial and not-for-profit sectors. The OCT devel-
• Solar Terrestrial Probes - Studies the ops roadmaps to plan their work, which were reviewed in
plasma environment between the Sun, a 2012 Report by the independent National Academies,
Earth, and solar system. Projects in- and guides funding plans. In addition to fostering tech-
clude the four Magnetospheric Multi- nology transfer across NASA programs, and externally
Scale spacecraft, and continuing opera- across other agencies and society in general, specific pro-
tion of other missions. grams include:
• Heliophysics Explorers - Includes small
to medium targeted science missions. It • Small Business and Technology
includes the ICON ionospheric explorer, Transfer Program - These are com-
and operation of a number of previous petitively awarded contracts to small
missions. (<500 employee) business for innova-
tive technology. They progress from
initial merit and feasibility through
Aeronautics Research ($640 million)
commercialization.

The NASA Aeronautics Research Directorate works to • Early Stage Portfolio - Invests in about
improve aviation and high speed flight. Entry, descent, 400 basic and applied research and early
and landing through an atmosphere from space is studied technology development activities. This
elsewhere under Space Technology. In addition to general includes external research grants and in-
management, innovation, and education activity, specific novative concepts, internal innovation at
programs include: NASA centers, and Centennial Chal-
lenge prize competitions.
• Airspace Operations and Safety - Per- • Game Changing Development - Ma-
forms research and development to in- tures technologies from early stages to
crease the throughput, efficiency, and flight demonstration. It includes hard-
safety of the National Airspace System. ware topics like robotics, manufacturing,
and materials. It also includes mission
• Advanced Air Vehicles - Performs re-
topics like entry, descent and landing; fu-
search for new generations of civil air-
ture propulsion and power, and destina-
craft, including supersonic and hyper-
tion systems and instruments.
sonic technology. Note that the US De-
partment of Defense does similar work • Technology Demonstration Missions
for military purposes. - Demonstrate readiness for operational
use using prototypes and demonstration
• Integrated Aviation Systems - Focuses
units in the relevant space environments.
on experimental flight research using in-
Current examples include satellite servic-
tegrated systems, flight test ranges, and
ing, deep space optical communications
aircraft. This complements high perfor-
and atomic clock, non-toxic propellants
mance computing for analysis and sim-
for small thrusters, solar-electric propul-
ulation, and physical testing in wind tun-
sion, supersonic decelerators, composite
nels and propulsion test facilities. Testing
materials, long-term cryogenics, in-space
is coordinated with the DOD, who have
robotic manufacturing and assembly, and
similar facilities and test needs.
small spacecraft technology.
• Transformative Aero Concepts - Solic-
its and works on multi-disciplinary and
Human Exploration ($4.03 billion)
revolutionary ideas, from original con-
cepts through small scale ground and
flight testing. It also advances computa- This budget account is focused on human exploration be-
tional and experimental tools and tech- yond low Earth orbit and transport to the existing Interna-
nologies. tional Space Station (ISS). The program is based on goals
in the NASA Authorization Act of 2010.
Space Technology ($686 million) Exploration Systems Development - About three quar-
ters of exploration funding goes to three major programs
This category funds space technology development from under this heading:
very early concepts to flight demonstration, after which
it would be incorporated into operational programs. It is • Orion Multi Purpose Crew Vehicle -
managed by the Office of Chief Technologist (OCT). This is a modular launch abort system,
70 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

crew capsule, and service module de- failures, and safety and mission assur-
signed for long duration deep space mis- ance.
sions. It is currently in development • ISS Research - Funds biological and
by Lockheed Martin Corporation for physical research which benefits from the
NASA, and had an early flight test in orbital environment. Major areas of re-
2014. and uncrewed launch on the SLS search include human microgravity ef-
in Nov 2018. fects, plant and microbiology, fluid, ther-
• Space Launch System (SLS) - This is mal, and particle physics, combustion,
a heavy-lift rocket for launching pay- and materials processing. It is also a plat-
loads beyond low Earth orbit. The ini- form for Earth and space observation, al-
tial design uses used Space Shuttle pro- though not unique in this aspect. The ISS
gram SSME (RS-25) engines and mod- is also being used as a demonstration plat-
ified Shuttle Solid Rocket Boosters for form for robotics, life support, and fire
the lower stage, and RL10 second stage safety technology. The US portion of the
engines. Contractors include Boeing, Station has been designated the ISS Na-
Aerojet Rocketdyne, and Orbital ATK. tional Laboratory.
First launch is planned for Nov 2018.
• Exploration Ground Systems - Pro- Space Transportation - This program provides current
vides funding for the SLS specific parts and future transportation to Earth orbit of astronauts, sci-
of the launch site at the Kennedy Space ence experiments, supplies, maintenance hardware, pro-
Center. This work is in parallel with SLS pellants, and return of wastes. The current destination is
development. the ISS. Funding was $2.25 billion in 2015.

• ISS Crew and Cargo Transportation

Human Research Program - Conducts research and de-
- Purchases commercial transportation
velops technologies that allow humans to travel safely and
to the ISS from Orbital ATK, SpaceX,
productively in the environment of space. It was allocated
Sierra Nevada, Boeing, and Roscosmos.
$142 million funding in 2015.
Note that other ISS partners provide their
Advanced Exploration Systems - Funds high prior- own transportation and payloads.
ity projects for future human missions. Areas of work
• Commercial Crew Program - Supports
include life support, deep space habitation, advanced
development of commercial US crew
propulsion, landing systems, and resource prospecting
transport by Boeing and SpaceX. Flights
and processing. They were allocated $189 million fund-
are expected to start in 2017.
ing in 2015, but this is expected to significantly increase
in coming years.
Space and Flight Support - This includes other op-
erational activities, including the communications net-
Space Operations ($5.03 billion) work, launch services, propulsion testing, and human
space flight.
This part of the budget includes operating the Interna-
tional Space Station, the space communications network, • Space Communications and Naviga-
and launch and test operations on Earth. tion - Operates the communications net-
International Space Station (ISS) - As the name indi- work and navigation services for all
cates, this is an international program. This program cov- NASA spacecraft. It includes the Deep
ers the NASA funded portion, which amounted to $1.52 Space Network for long range communi-
billion in 2015. The program is expected to continue cation, NASA-owned Tracking and Data
through fiscal year 2022. One domestic purpose of the Relay Satellites in synchronous orbit, and
ISS is to promote commercial research and transporta- development of future laser communica-
tion in Earth orbit. Tasks within the program include: tions for higher data rates.
• Human Space Flight Operations - Pro-
• ISS Sytems Operations and Mainte- vides astronaut crew training, and health
nance - Responsible for assembling, op- and safety monitoring.
erating, and maintaining the ISS with an • Launch Services - Manages the Expend-
onboard crew of 6. This includes mission able Launch Vehicle program. Actual
planning, ground monitoring and com- cost of individual rockets is included un-
munications, spares and logistics, crew der the project or mission that requires
training, mission integration (allocating them, while contracting, and launch site
all orbital resources), handling system operations are funded here.
2.14. EXISTING PROGRAMS 2 71

• Rocket Propulsion Test - Operates four • Advance STEM education and workforce
rocket engine test facilities. pipeline.

Support Activities ($2.87 billion) Manage the agency’s people, technical capabilities,
and infrastructure.
This portion of the NASA budget covers general staff and This includes attracting a skilled workforce, innovative
support not tied to a particular program or project. It in- work environment, and necessary facilities, tools, IT, ser-
cludes public education activities, agency staff, facilities vices, and capabilities. It also includes maintaining safety
construction and maintenance, and inspector general. A general set of objectives is not the same as what will ac-
Education - Interacts with all levels of education insti- tually or likely get funded and built. Better detail for the
tutions, museums, and science centers to promote inter- next five years can be found in annual budget requests,
est and programs in science, technology, engineering, and such as the one for Fiscal Year 2017. The duration of
mathematics. current and upcoming projects often runs longer than the
5 year budgeting horizon. So individual agency offices
Safety, Security, and Mission Services - Oper- and projects usually have plans and schedules which run
ates NASA centers and headquarters, including 17,300 somewhat longer. NASA’s funding is on an annual basis,
agency staff and IT. and therefore unpredictable to some degree. The nature
Construction - Includes new construction, demolition, of research and technology development, which involves
environmental compliance, and repair of existing build- finding out new things, is also unpredictable. Therefore
ings and other facilities. detailed long range planning does not make sense. In-
stead, long range goals and directions are set, with de-
tailed plans made approximately 5 years ahead. Addi-
Future Projects The 2014 NASA Strategic Plan pro- tionally, concrete plans for future projects would imme-
vides a general description of their planned future objec- diately raise funding issues with the US Congress, who
tives. These include: would have to approve the money for those projects.
Expand the frontiers of knowledge, capability, and
opportunity in space:
2.13.3 Continued on page 2 →
• Expand human presence into the Solar
System, and to the surface of Mars. 2.14 Existing Programs 2
• Conduct research on the International
Space Station. ← Back to Page 1
• Employ U.S. commercial capabilities to
deliver cargo and crew to space. 2.14.1 Government Programs - United
• Understand the Sun and its interactions States (continued)
with Earth and the Solar System.
Department of Defense (DOD)
• Ascertain the content, origin, and evolu-
tion of the Solar System, and the poten-
The space projects of the US Department of Defense
tial for life elsewhere.
(DOD), are primarily managed by the United States Air
• Discover how the universe works, explore Force (USAF) Space Command, with some assistance
how it began and evolved, and search for from other branches and agencies within the department.
life on planets around other stars. It is difficult to get a full view of DOD space-related
• Mature crosscutting and innovative space projects, since it only makes up part of the department’s
technologies. total activity. The 2016 DOD Budget Overview reports
$7.1 billion for space investment (page 5-3). The Space
Foundation 2015 DoD Space Budget Comparison indi-
Understand the Earth, and develop technologies to
cates a total of $9.75 billion, with detailed tables. How-
improve life on our home planet
ever, much of the space-related activity is classified, and
not included in these totals. Between 2012 and 2013, re-
• Advance aeronautics research for safe ported DoD space spending dropped from $27.5 to 10.8
and sustainable aviation. billion, and this difference may represent the classified
• Advance knowledge of Earth as a system. amount.
• Optimize technology investments, open What is known publicly is the DOD procures and op-
innovation, and infuse technology for na- erates a number of satellites and supporting launch sys-
tional benefit. tems in the areas of communications, weather, nuclear
72 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

detection, mapping, and navigation. It also funds a • Electrical, Communications, and Cyber Systems
significant amount of scientific research and engineer- (ECCS), which funds electronics, communications,
ing development either directly or indirectly related to power sources, networking, and robotics.
space projects. Examples of indirect efforts include high-
speed air-breathing propulsion and radiation-hardened Mathematical and Physical Sciences Direc-
electronics. torate (MPS}

Past Projects Ballistic missiles were developed for Funds the scientific fields of astronomy, chemistry, ma-
military purposes rather than space launch. The require- terials research, mathematics, and physics. This in-
ment to deliver a bomb thousands of miles requires reach- cludes the Astronomical Sciences Division (ASD),
ing about 80-90% of Earth orbit velocity. Therefore bal- which funds research in astronomy, including planetary
listic missiles could be adapted for space launch, usually science, and contributes funding or operates a number of
by adding or enlarging the upper stage. A number of them ground-based observatories such as ALMA and LSST.
were adapted this way in the late 1950’s to 1960’s. As the Details of astronomy projects are gathered under the not-
size of space satellites grew far beyond that of nuclear for-profit section, because typically they have multiple
bombs, those launchers were repeatedly modified, to the funding sources for the instruments and researchers.
point that some of them no longer retained any original
parts except the name. Besides being used for delivery
of DOD spacecraft, the same launch sites and vehicles Department of Commerce - NOAA
have often been used for non-military launches. Ballistic
missiles are not designed to be used more than once, and The US Department of Commerce’s National Oceanic
their value is relative to the target they intend to destroy. and Atmospheric Administration (NOAA) operates
So an unfortunate side effect of using modified ballistic weather and earth observation satellites via its Satellite
missiles for space launch is they were not optimized for and Information Service. Weather satellites are located
cost. In fact, expensive throw-away hardware is the exact both in low polar and geostationary orbits, which sup-
opposite of optimized for cost. Since the previous tech- ply different levels of resolution and temporal coverage.
nology and experience was gained in that environment, NOAA also operates the JASON sea-level measurement
later space projects have had to struggle to overcome this and DSCOVR solar wind satellites. Space related fund-
history. ing was about $2.1 billion in 2014.

National Science Foundation (NSF) Department of Energy (DOE)

The National Science Foundation, or NSF, is an in- The US Department of Energy is a cabinet level depart-
dependent federal agency which funds research in all ment formed in 1977 to consolidate nuclear and other
fields of science and engineering, except medical sci- energy activities of the United States. It performs re-
ences, which have their own agency. The expected fund- search into all forms of energy and some scientific re-
ing from the 2016 Budget Request totals $7.72 billion, a search. Almost all space projects require energy to func-
$380 million increase over 2015. The space-related part tion, so some of this research is relevant. Details of de-
of their budget was about $460 million in 2014. Although partment’s current programs and funding can be found in
all knowledge is a seamless whole, and much of it applies it’s 2017 Budget Request. Space-related parts of their
to space projects, certain fields are currently more directly activities include:
related than others. The NSF offices in this category in-
clude:
Energy Efficiency and Renewable Energy
Engineering Directorate (ENG)
This office performs research into all forms of energy
Funds research in all the fields of engineering. Much of supply and energy efficiency. Total funding in 2016 was
this can apply to space projects because most fields of $2.07 billion. Programs of particular interest include:
engineering are used in such projects. Particular areas of
interest are:
• Solar Energy Techologies - Develops photovoltaic
and concentrated solar power, aiming to reduce
• Chemical, Bioengineering, Environmental, and
cost and increase efficiency. Given the high solar
Transport Systems (CBET), which funds research in
flux in the inner Solar System, many space projects
chemical, mechanical, and aerospace topics.
use this as their main power source. Note that by
• Civil, Mechanical, and Manufacturing Innovation 2016 ground solar energy has developed into a large
(CMMI), which funds topics named it it’s title plus worldwide industry, and much research and devel-
materials design. opment now occurs outside US government funding.
2.14. EXISTING PROGRAMS 2 73

• Advanced Manufacturing - Participates in the 2.14.2 Commercial Programs

National Network for Manufacturing Innova-
tion. Mining and manufacturing will be increas- Commercial space projects are part of the more general
ingly important in future space projects, because of“Aerospace and Defense” business sector. An Annual
the high leverage of using local resources rather than
Overview of the entire sector is compiled by Price Wa-
launching everything from Earth. terhouse Cooper. There is considerable overlap between
items used in space projects and those developed for other
Nuclear Energy parts of the sector, and often the same company does
both types of work. The sector as a whole had worldwide
revenue of US$ 729 billion in 2014. Note this excludes
This office performs research into Nuclear energy sources undisclosed classified or private projects, and that sup-
in particular. Total funding in 2016 was $986 million. plier sales which end up in final products in the industry
Programs of particular interest include: are double-counted.
The Satellite Industry Association is a US-based trade
• Small Modular Reactors - Since mass is a con- association for commercial space businesses. It has a
cern for space projects, power sources smaller than 2016 Industry Report which indicates the global space
terrestrial nuclear power plants are desirable. industry in 2015 was US$ 335 billion, of which the satel-
lite portion was US$ 203 billion. The latter represents
• Radiological Facilities Management - Provides 9.2% of global telecommunications revenue. The vast
Radioisotope Thermal Generators, which are used majority of satellite revenue is from services and ground
as power supplies for space missions. They are equipment, such as for satellite television. Satellite hard-
developed, built and tested at the Idaho National ware manufacturing and launch accounts for US$ 21.8
Laboratory, and several other DOE national labo- billion of the total. Since 208 satellites were launched
ratories. in 2014, that implies the average unit cost to build and
launch is US$ 105 million. For comparison, the average
Science prices of Boeing passenger aircraft range from US$ 60-
330 million.

This office funds scientific research and scientific tools By far the largest segment of commercial programs is for
to understand nature and advance energy security. To- communication services, of which consumer satellite TV
tal funding in 2016 was $5.35 billion. Much of the work is the largest single component at 2014 revenue of US$
occurs through a network of National Laboratories. Pro- 95 billion. All other satellite services amounted to US$
grams of particular interest include: 27.9 billion. This included satellite radio and broadband,
commercial communications, mobile voice and data, and
remote sensing. Unclassified satellite manufacturing av-
• Basic Energy Sciences - Includes research in ma- eraged about US$ 13.75 billion per year in recent years,
terials science, engineering, chemical, geoscience, and launch revenue averaged about US$ 5.2 billion. Both
bioscience, and large scientific user facilities. are counted in year of delivery, while actual costs usually
are distributed over several years. Ground equipment to
• Fusion Energy Sciences - Supports research to de- receive from or transmit to satellites accounted for US$
velop fusion as an energy source. Although the Sun 58 billion in 2014.
is a natural fusion reactor, solar energy is not avail-
able deep underground, or at large distances from Future commercial programs which are in research and
the Sun, so artificial fusion would be useful in these development but not yet significant in revenue include
locations. space tourism, orbital mining, and energy transmission.

• Nuclear Physics - Supports research into all types

of nuclear matter. This includes artificial and puri- 2.14.3 Not-for-Profit Programs
fied isotopes, which can have space applications.
The not-for-profit sector includes activities performed for
the general benefit of humanity, most often at univer-
ARPA-E
sities and research foundations. At present, most such
space programs are related to Astronomy, and the cross-
This agency within the Department of Energy focuses on discipline field of Planetary Science. Astronomy is
early-stage energy technologies which have high potential partly related to space systems because that is the science
and high impact, but are too early for private investment. which studies the Universe i.e. all of space outside the
Total funding in 2016 is $291 million. Nearly all space Earth. It is also partly because the Earth’s atmosphere,
projects require energy to operate, so improvements in gravity, and day/night cycle interfere with certain types of
this technology would be useful. instruments and observations, so they must be performed
74 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

in space themselves. Individual researchers can work on of telescopes is a 2.4 meter UV to infrared
their own, but the larger projects, such as new telescopes, space telescope launched in 1990 and expected
are often a mixture of private and government funding. to operate to until either equipment failure or
Planetary science is the study of objects and systems orbit decay terminates the mission, likely in the
which orbit stars. Originally it was purely a subset of as- 2018-2024 period. In addition to other sci-
tronomy and restricted to the Solar System, as no other ence, Hubble has been used to find and exam-
planetary systems were known. As better instruments ine outer solar system objects, extrasolar plan-
and close-up observations have been made, we have pro- ets, and protoplanetary disks. Hubble is mainly
funded by NASA and ESA, and scientific op-
gressed from merely determining orbit and approximate
size of planets and moons to detailed mapping and ge- erations are managed by the Space Telescope
Science Institute.
ology. So planetary science now draws heavily on the
Earth Sciences to understand the history and develop-
ment of these bodies. In recent decades Circumstellar Spitzer Space Telescope - This is a 0.85 me-
Discs and Exoplanets have been detected around other ter infrared space telescope launched in 2003.
stars, extending the study beyond our Solar System. More Its original helium supply, used to cool the in-
recently, the first few Rogue Planets have been discov- struments, ran out in 2009, so at present only
ered. These objects are too small to be stars, but not at- the shortest wavelength instruments are still op-
tached to any stellar system. They were either ejected erating. At some point equipment failure will
from star systems or originally formed as unattached ob- end the mission. Spitzer observed many solar
jects. system and extrasolar objects. It is funded by
NASA.

Astronomy Programs Kepler Mission - This is a 0.95 meter visi-

ble light space telescope operating since 2009
The US National Academies compile a Decadal Survey through at least 2016. It is designed to detect
every 10 years, laying out priorities for astronomy and planets orbiting other stars which cross in front
astrophysics. This is a good starting point to review cur- of the star (transit) and dim the star’s light.
rent and near-term programs in this field. Other science Since that only happens when the orbits are
and engineering departments at universities do research edge-on to us, Kepler can only detect a fraction
related to space, and there are some smaller foundations of the total planets in the direction it is looking.
dedicated to space research. Although much of the fund- The total number of planets can be estimated
ing for astronomy comes from government sources, we from the fraction it can see. Kepler is funded
list all of the programs here to give a better view of the and operated by NASA.
field as a whole.
Astronomy programs can be roughly sorted by location Stratospheric Observatory for Infrared As-
and wavelength of the instruments. Locations include tronomy (SOFIA) - This is a 2.5 meter in-
ground, airborne, and space. The latter two are more ex- frared airborne telescope mounted on a 747
pensive, but are used to get above interference from the aircraft which first saw operation in 2010. At
Earth’s atmosphere. The Electromagnetic Spectrum its operating altitude of 13.7 km it is above
ranges from long radio to very short gamma, and instru- most of the absorption caused by the Earth’s
ments exist to cover most or all of it. The location of the atmosphere. It will study, among other things,
instruments, which are grouped into Observatories, is the formation of stars and planets, the inter-
nowadays where the viewing is best for that device. This stellar medium, and planets and small objects
is usually different than where the funding organizations within our own solar system. It is funded 80%
or the astronomers who make use of the data are. by NASA and 20% by the German Aerospace
Center (DLR).
There are a great number of observatories in use, since
even small privately owned ones can collect useful data.
Wikipedia has a lists of Astronomical Observatories Atacama Large Millimeter Array (ALMA)
and Space Observatories, the latter meaning located in - This is an array of 66 x 12 and 7 meter radio
space (all of them look at space). A few of the more sig- telescopes that function as an interferometer to
nificant ones which are relevant to future space programs combine their signals and act as a large single
are noted below. These are mainly ones that look at our instrument up to 14 km across. It began scien-
own and nearby planetary systems. tific observations in 2011 when partially built,
and was fully operational in March 2013. It is
a general purpose radio telescope in the 0.35-
Current Projects 10 mm wavelength bands. Among the types
of observations it is making are circumstellar
Hubble Space Telescope - This most famous dust and planetary systems around other stars.
2.14. EXISTING PROGRAMS 2 75

ALMA is jointly funded by the US, Europe, Thirty Meter Telescope (TMT) - This is a
Japan and is hosted by Chile. The ALMA 30 meter near-UV to mid-infrared (0.31 to 28
Website has additional information about the μm) ground telescope with an estimated com-
project. pletion after 2022. It will be made of nearly
500 1.4 meter hexagonal mirror segments on
Gaia Observatory - This is a space satellite a single large mounting to be built at Mauna
with instruments to measure the accurate posi- Kea Observatory in Hawaii. It has multiple
tion, motion, brightness, and spectra of about a science goals, including extra-solar planet and
billion stars. It was launched in late 2013. It is Kuiper Belt measurements. It is jointly funded
likely to find many planets by the wobble they by multiple foundations, universities, and na-
cause in their parent stars. It is also expected to tional governments. The project is currently
find and accurately measure the orbits of many (2016) delayed by local protests and the con-
small objects in our Solar System. The ESA struction permit process.
sponsors this mission, and their Gaia Website
has additional details about it. Giant Magellan Telescope (GMT) - This is a
24.5 meter equivalent visible and near infrared
Radar Astronomy - A few instruments are ground telescope with an estimated operational
used to actively send out radar signals and mea- date of 2021. It is made up of seven 8.4 me-
sure the return. The timing of the return sig- ter mirrors on a single large mounting, because
nal provides extremely accurate location and that is the largest mirror size that can currently
detailed shape information. The latter comes be made in one piece. As of 2016 one of the
from timing differences from parts of the ob- mirrors has been completed and three others
ject that are at different distances, and repeat- have been cast. Construction the observatory
ing the measurements as the object rotates. was started in Chile in late 2015. Like other
Signal intensity falls as the 4th power of dis- large telescopes it has multiple science goals,
tance, so this technique has been limited to ap- including imaging of planets around nearby
proximately 500 nearby objects, mostly Near stars. It is currently funded by a group of uni-
Earth Asteroids. versities.

Future Projects Large Synoptic Survey Telescope (LSST) -

This is an 8.4 meter visible and near-infrared
James Webb Space Telescope (JWST) - This ground telescope with a 3200 megapixel cam-
is a 6.5 meter visible to mid-infrared space era which has begun construction. The LSST
telescope expected to launch in 2018. It has Construction Schedule shows a completion
a number of science objectives, including ob- date of early 2022. It is a general purpose sur-
serving extrasolar planets, brown dwarfs, and vey telescope which the LSST Science Book
outer Solar System objects among the more estimated could find 90% of Near Earth Ob-
relevant ones for space programs. The instru- jects larger than 140 meters in 12 years (ie
ments have a fairly narrow field of view, 2x4 2034) if 15% of it’s time is used for that pur-
minutes of arc or about 1% of the area of the pose (section 5.11.1).
Moon as seen from Earth. So it will do tar-
geted observations rather than wide surveys of European Extremely Large Telescope
the whole sky. (ELT) - This is a 39 meter visible and near-
infrared segmented ground telescope expected
Wide-Field Infrared Survey Telescope to be operational about 2024. This is the
(WFIRST) - This is a 2.4 meter near-infrared largest optical telescope currently planned.
space telescope proposed for launch in the Its science goals include finding extrasolar
mid-2020’s. The Science Definition Team planets. It is being built by the European
membership comes from many universities Southern Observatory, which is funded by
and several independent observatories and 16 nations and one host country (Chile), where
US government centers. They released a the telescope will be located.
Final Report in February 2015 describing the
science goals and telescope design. Among
the science goals is to perform a search for Planetary Science Programs
planetary systems in our Galaxy using grav-
itational microlensing. The program would Approximately eighty Graduate Schools around the
be primarily funded by NASA, but the data world offer degrees and perform research in planetary
would be used by astronomers worldwide. science. Some of them participate in government-funded
76 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

planetary and space observatory missions by building in- the delivery and processing in theory can be low cost. So
struments or supplying scientists. Others participate in there is an economic incentive to obtaining bulk materi-
or run ground-based observatories, perform independent als in space. Bulk materials can be used for propellants
theoretical research, or analyze data generated by the ob- and fuels, serve as radiation shielding, or treated as min-
servatories and missions. eral ores for extraction and further processing into useful
products.
NASA has termed use of local materials in space In-
2.15 Future Projects Situ Resource Utilization (ISRU). They have set up an
ISRU Project Office and are doing some early research
on methods. The ISRU name is more palatable to some
Section (1.9) described existing space programs and
members of committees who decide NASA’s budget than
projects. This section lists categories of space activities
“space mining”, which sounds too much like science fic-
and mission objectives for future programs and projects.
tion. Despite that, Title IV of Public Law 114-90, also
Parts 2 and 3 of the book will describe the transport
known as the Space Resource Exploration and Utiliza-
and engineering elements to implement these activities.
tion Act of 2015 makes provisions for private commer-
They also discuss how to design, build, and operate com-
cial recovery and ownership of space resources. Aster-
plete systems that perform the desired functions. A given
oids have negligible gravity wells, and by early 2016 over
project, location, or mission may include more than one
14,000 of them in the Near Earth Asteroid (NEA) cate-
of the items below. The multiple activities can be present
gory have been discovered. Different asteroids also have
but not otherwise interlinked, in the same way a large of-
a variety of useful chemical compositions. This combina-
fice building may house multiple unrelated tenants. In
tion has led to the most interest in Asteroid Mining rel-
that case the project may share joint services like electri-
ative to other space locations, but any location with avail-
cal power among the various activities. Alternately, the
able materials and energy is a candidate. NASA is also
multiple activities may link to each other in parallel, such
in the early stages of developing an Asteroid Redirect
as retrieving asteroid materials and refueling the tug from
Mission to bring back a boulder from an NEA to a safe
those same materials. The activities may also be con-
lunar orbit, where it can later be examined in detail.
nected in a logical sequence, where one activity follows
another, such as building new exploration or mining ve-
hicles from previously produced inventory materials. We
call any program with multiple different activities a Com- Propellants and Fuels
plex Program, and an extensive example is provided in
Part 4. Many other such complex programs are possible, Propellants are the materials consumed and ejected in
but our example is intended to be both a tutorial and a many types of propulsion systems. They often make
reasonable proposal for future activities. up a large or dominant portion of space system mass.
Propellant applications include space transport, ballistic
The activities listed below are somewhat in order of transport, aerodynamic transport, surface transport, and
idealized complexity. An actual program that imple- portable power. Due to the wide range of applications,
ments multiple items would likely choose a different or- environments for their use, and sources of raw materials,
der which makes sense from a design and schedule stand- there is also a wide range of potential propellants. Spe-
point. As civilization expands into space, eventually we cific propellants for space transport are discussed in Part
will incorporate all the industries and activities that occur 2 of this book. The other applications are listed below.
on Earth today. So this list is not comprehensive. Instead,
it includes the more significant items that are particularly Fuels are energy sources reacted with materials in the
space related. environment, such as air-breathing engines, or with an
oxidizer in the case of chemical rockets. Nuclear fuels
develop energy by internal reactions, and do not require
2.15.1 Bulk Mass another material to react with. Of particular interest for
advanced uses are fission fuels such as Uranium or Tho-
Bulk mass is matter which is undifferentiated into manu- rium, and if fusion power is developed, light isotopes for
factured components, either in the raw unprocessed state, those reactions. This is the energy released by nuclear
or processed into a refined product. The quantity of bulk reactions is very high relative to chemical ones.
matter is a main variable, as all the bulk material of a Production of propellants and fuels from local resources
given type is equivalent. The Earth has a deep gravity is called In-Situ Propellant Production. To date, all
well, which requires 62.5 MJ/kg to escape. This is cur- propellants and fuels have been supplied from Earth. Us-
rently difficult and expensive. Local sources of bulk ma- ing local resources will often prove less expensive. The
terials, that are already in space near to where you need mass overhead for delivery is reduced by using depots at
them, can often be delivered and processed for less to- intervals along a trip. If all the propellant mass is car-
tal energy. Solar energy is abundant in the inner Solar ried from the start, more of it has to undergo a velocity
System in quantity, and available nearly everywhere. So change. So you have the overhead of using some of it
2.15. FUTURE PROJECTS 77

up to move the remainder. By refilling at intervals, the of radiation, and heavy elements for others. Where mass
on-board mass is less, and so the mission overhead of is important, optimized and processed materials may be
moving it is less. The supply sources for the propellants preferred for shielding, but unprocessed bulk material is
are close to or at the depot locations, and delivery is by easier to supply.
high-efficiency electric propulsion. Therefore the supply Habitats, vehicles, and equipment can reduce radiation
overhead is also low. levels by their own mass, before adding additional shield-
Besides conventional space transport, other propellant ing. For example, large habitats with thick walls, agricul-
and fuel applications include: tural soil, water, and atmosphere may provide sufficient
shielding as is. For vehicles or lightweight habitats, addi-
• Ballistic Transport - On an undeveloped or heav- tional shielding can be added bulk propellants, water, and
ily cratered terrain, attempting to drive across the food supplies. The surface or subsurface of large bodies
surface will be difficult, and for long distances will usually has sufficient local material to provide radiation
be slow. Ballistic transport uses similar methods as shielding. The approaches to using it are building under-
space transport, but instead of going to and from or- ground, or surface construction by moving enough mate-
bit, or between orbits, is used in sub-orbital trajec- rial around and over radiation-sensitive areas. A thickness
tories to go from one point to another. The lower of 25 cm or more of loosely packed unprocessed mate-
velocities potentially allow for simpler propulsion or rial can reduce exposure to reasonable levels for humans
higher payloads. This method is especially useful on (see Miller et al, 2008). In open space, where bulk mate-
smaller bodies where the velocity requirements are rials are being processed, the unprocessed raw material,
lower. stockpiles of processed materials, and wastes after pro-
cessing (slag) can all be used for shielding with proper
• Aerodynamic Transport - On bodies with a suf- arrangement around habitats and equipment.
ficiently dense atmosphere, you can potentially use
aerodynamic lift or buoyancy and a propulsion en- Ore Delivery
gine for transport, similar to how we use aircraft on
Earth. An Ore is any natural material containing enough of a de-
sired product to be economic to extract. On Earth, crude
• Surface Transport - Solar power is long-lived, and oil, iron ore, and crushed stone are all ores used to make
can generate thousands of times the energy over it’s other products, and by mass are the largest volume trans-
life than a chemical fuel. However it does not have ported. In space, bulk ores and their products are also
a very high power to area ratio, making it cumber- likely to be a major transport item by mass. Depending on
some for faster surface vehicles, and batteries can economics and technology, bulk ore can be transported
take a significant time to recharge. As an alterna- in it’s raw state, concentrated in the desired components,
tive, a fuel cell or combustion engine can provide called Beneficiation, and then transport the concentrate,
more power in a compact device and be more suit- or processed in place to a final material which is then de-
able for such vehicles. Stationary fuel stations can livered to where it is needed. Final destinations can be
refuel a vehicle quickly when distributed within the anywhere the ore or its products are needed. Some se-
vehicle’s operating range. lected examples include:

• Portable Power - Solar power and/or batteries are • Delivery to Earth - Nearly all people and economic
suitable for small portable devices which do not need activity is on Earth, and that will continue to be true
high power levels. Higher power levels for short pe- in coming decades. A large and economically de-
riods of time may be satisfied by propellant or fuel veloped population uses a lot of material resources.
sources. Some materials may become scarce enough that it
is economically feasible to obtain them from space.
Returning high-value materials like Platinum-group
Radiation Shielding elements does not pose a significant transportation
challenge, so we will look at the other end of the cost
Varying levels of radiation occur throughout the Solar
scale.
System, including on Earth. Many locations in space have
levels that are hazardous to people, or can affect electron-
ics and materials. Natural space radiation comes from Assume you want to import large amounts of
four main sources: cosmic rays, solar wind and plasma Iron to the Earth, since high-grade iron ores
events, trapped particles in magnetic fields, and concen- are in finite supply, and M-type asteroids can
trations of radioactive elements. Human-caused radia- supply nearly pure iron-nickel-cobalt alloy al-
tion can come from nuclear and other high energy de- ready in metal form. The simplest method is
vices. The effectiveness of shielding varies by composi- to aim pieces from metallic asteroids at a se-
tion, with light elements protecting better for some types lected spot, and just collect the bits that make
78 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

sumes part of the mass to bring the rest of the mass

up to the speed of the main vehicle. The delivered
mass is used to further accelerate the main vehicle
and resupply other materials. This allows somewhat
better velocities than starting with all the fuel at the
start of the mission, since the main vehicle has less
mass to accelerate. For this to work, you need to
know where the objects are ahead of time, or trust
that their density is sufficient to find them along your
path as needed.

2.15.2 Industrial Capacity

Figure 1.10-1 - 60 ton Iron-Nickel Hoba meteorite in Namibia.
Simple bulk materials like propellants and radiation
shielding are useful, but don't satisfy all future project
it to the ground. This requires no process- goals. Space Manufacturing is the production of more
ing. Figure 1.10-1 shows a natural example of refined or complex goods in space, from raw materials
this. You can find other examples of surviving and available energy. It is distinguished from space in-
pieces in museum asteroid collections. Non- dustry on Earth, where complex goods like satellites and
metal slag and volatiles will tend to burn off rockets are made, then delivered to space ready to use.
during re-entry. Choose a size, likely around Historically, a key feature of all manufacturing is using
10-50 tons, for the pieces so that re-entry drag tools to make more tools. When applied to space manu-
will slow them down, and you don't get a big facturing, a large industrial capacity can be grown from
crater. There are plenty of places on Earth a smaller and simpler starter set. This avoids the cost of
with few people and decent access for ship- sending entire factories into space. A simple bootstrap-
ping. The market for steel is about 1.5 bil- ping example is making pressure vessels from metallic
lion tons/year. The challenge is delivering it to asteroid material, which are then used in chemical pro-
Earth for around $1/kg or less. You likely will cessing of ores.
need to redirect and chop up a megaton (60 me- Some production will be for Final Goods in the eco-
ter) metallic asteroid or more, but the yield is nomic sense. These are end products not used in fur-
worth at least $1 billion ( 1 million tons steel ther production. An example is an orbiting greenhouse
@ $1000/ton), which may be enough to cover that produces food for people in space. The remainder of
operating costs for an efficient operation. production is either Capital Goods, lasting items used
in further production, or Intermediate Goods, which
• Delivery to Space Locations - Because of the cur- are partly finished items between raw materials and final
rent high cost of launch to orbit, material in a desired products. When the industrial capacity is intended for
space location is worth much more than most mate- more than final products, and some of the capital goods
rials on Earth. It needs to be moved from where it are made in space, rather than delivered from Earth, then
naturally occurs to where it is needed. The Earth’s the production and growth sequence must be optimized
gravity well requires a lot of energy to climb out of, for factors like design cost, growth schedule, and initial
and to date launch systems use inefficient rockets to mass launched from Earth.
do it with. Materials which are already nearby in
space can use efficient electric thrusters for trans-
port. On the surface of bodies, some materials can Seed Factories
be obtained locally. Transport to and from bodies
smaller than Earth can use mechanical and electric A “seed factory” is a starter set of equipment which is
methods. These all are potentially much less expen- intended to grow to a mature industrial capacity. This
sive solutions once they are set up. is by analogy to a plant seed, which grows to a mature
plant. Some portion of the factory output is directed at
• Oort/Rogue Object Delivery - In the distant fu- self-growth, and the remainder to desired final products
ture, a large interstellar mission may require a lot to be used. An early NASA-funded study of the concept
of propellants and other supplies, due to the high is reported in: Freitas and Gilbreath, eds. Advanced Au-
velocity and long duration of the trip. In this con- tomation for Space Missions, NASA Conference Pub-
cept, several comets or unbound rogue objects, or lication 2255, 1982. Computer, automation, and com-
parts thereof, are intercepted by a propulsion unit munications technology was not good enough in 1980 for
that comes from the main vehicle, or is sent ahead the intended use on the Moon, so NASA did not pur-
from the launch point. The propulsion unit con- sue the idea further. The concept was limited to “replica-
2.15. FUTURE PROJECTS 79

tion”, making exact copies of the starter set until enough method. To recover large amounts of material
total capacity was reached. The study also assumed the from inner Solar System asteroids, Iron-Nickel al-
seed factory made 100% of its own parts, was 100% au- loy found in the metallic type can be rolled into foil,
tomated, and could only be used in space. and then used to make solar sails. If what you want
The current seed factory concept includes two other is the metal, then it sails itself to where you want
methods of growth: “diversification”, which is making it. If you want some other asteroid material, larger
new items not in the original starter set, and “scaling”, amounts of sail area can be used for a cargo tug. To
making items of different sizes than what you start with. make the sails, you need the functions of a rolling
mill - a way to heat the material and a way to force
The assumptions of 100% self-production, 100% au-
tomation, and only used in space are removed. Therefore it between two rollers to make thin sheets.
a seed factory may start out making only a percentage
of its own items, with the remainder supplied from else- Drawbacks to Iron-Nickel sails are their higher
where, and use people directly or by remote control to mass compared to light alloys like aluminum-
do some tasks. This greatly simplifies the design of the magnesium, and their reflectivity is lower in the
starter set. Over time, the growing production capacity natural state. Solar sails are also somewhat lim-
can make more of its own items, and need less supplied ited in the directions they can apply thrust. Ad-
from outside. The idea of self-expanding production us- vantages are the raw material is readily avail-
ing local energy and materials applies everywhere, not able in large quantities from the asteroids them-
just in space. It is a complex subject, and this wikibook selves, and it does not need a lot of processing
is about space systems, so a separate book was started to make into a usable form. An Aluminum or
for it. Aluminum-Silicon alloy coating can be added
to increase reflectivity if desired.
Self-expanding production can provide a large amount
of leverage in terms of mass launched from Earth to fi-
• Glass - Besides the obvious use as windows for
nal products and missions you can carry out. We there-
greenhouses, glass can be used for fiber optic cable,
fore consider it an important concept, and will reference
and for inert reaction vessels, including those which
it extensively. Because space locations are not uniform
concentrated sunlight is sent through. The chemi-
in energy or material resources, the seed equipment and
cally simplest glass is Fused Quartz, which is pure
what it makes will likely be distributed in a trade network
Silicon Dioxide (SiO2 ). Silicon and Oxygen make
where components are optimized for location. The net-
up over 60% of Lunar soil, and about 50% of stony-
work will include some parts and materials that still are
type asteroids, so the components are very common.
supplied from Earth. The various parts of the network
However they are usually bound in silicate minerals
may be owned and operated by different entities, leading
with other elements, and require chemical or ther-
to a self-sustaining space economy.
mal separation.

2.15.3 Manufactured Items • Brick and Concrete - Brick and concrete can be
used for lower-strength construction, such as radi-
Once you have set up an industrial capacity, you want to ation shielding, thermal shelters, and landing pads,
use it to make useful products. There are as many pos- when air leakage is not an issue. Conventional brick
sible products to make in space as there are on Earth. is made by heating a mixture of sand and clay until
Which ones make sense to make at a given location de- the particles partly melt and bond together, a pro-
pends on mass, complexity of production, and economic cess called Sintering. Building elements can be
value. The following items have been suggested for rela- made the same way in space, provided a sufficient
tive ease of production and significant mass savings: source of heat and right type of ingredients can be
found. Its chief advantage is simplicity. Concrete
• Structural Materials - A variety of structural ma- is a class of artificial stone made from varying size
terials can be made from local materials in space, re- crushed stone, called Aggregates, and a binder ma-
ducing the amount of material that has to be brought terial to hold them together. On Earth the most com-
from Earth. An example is Iron-Nickel shapes like mon binder is Portland Cement, a mixture of shale
columns or plates, made from metallic type aster- and limestone heated to high temperature and then
oids. Another is cast or sintered rock, made from ground to a fine powder. Many other binders are
Lunar or Mars surface material, or stony-type aster- possible, and some of them would be useful in space.
oids. They would be melted with solar or microwave The usefulness of concrete is based on it’s relatively
furnaces. A third example are high strength Carbon low cost, and the ability to be cast at room temper-
or Basalt fibers from respective local sources. ature in a variety of shapes which then harden.

• Solar Sails from Metallic Asteroids - This is a • Chemical Products - This includes plastics, chemi-
combination of structural material and transport cal reagents, lubricants, and many other items made
80 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

by chemical processing. Lunar rocks are high in • Solar-Thermal - Some future uses require heating
metal oxides, such as Silicon, Iron, Aluminum, rather than electricity, such as an industrial casting
Magnesium, and sometimes Titanium. These met- furnace. Concentrating reflectors can achieve this
als are useful for structures, solar cells, and electron- rather easily. Thermal storage using local bulk mass
ics. Converting the oxides to metals and separating can bridge night-time power needs. The material is
the elements is called Extractive Metallurgy. Pro- heated during the day, and the stored heat used to
cesses include physical ones like crushing and grind- operate a generator at night. This avoids needing
ing to separate mineral grains, and magnetic sep- large batteries for higher power demands.
aration. They also include chemical ones like liq-
uid solutions, thermal ones using high temperatures,
• Nuclear Sources - Radioisotope decay heat cou-
and electrical ones like electrolysis. The chemical
pled to thermoelectric generators has been used on
solutions and electrolysis require suitable reagents.
a small scale for scientific missions to the outer So-
These include compounds with non-metallic ele-
lar System, or when daytime sunlight and batteries
ments like Calcium, Potassium, Sodium, Phospho-
are insufficient on a planetary surface. A number
rous, and Sulfur. These elements can be found in
of small reactors using thermoelectric or thermionic
some types of asteroids. Plastics and lubricants are
generators have been flown. For future applications
typically Carbon compounds, which are also found
like higher-power propulsion or surface habitats, re-
in asteroids.
actors with heat-engine generators, which are more
efficient, have been proposed. If fusion rectors are
• Biological Products - This of course includes food, developed, they would be very useful in space, since
but also non-food items like wood, and chemical hydrogen is widely available as you get farther from
outputs of micro-organisms. The oldest example of the Sun.
the latter is alcohol from yeast, but modern biotech-
nology can produce a wide variety of items. Grow-
ing food typically requires water, Carbon Dioxide, • Beamed Power - Civilization on Earth has a large
fertilizers with Nitrogen, Phosphorus, and Potas- and growing demand for energy, but fossil fuels are
sium, and trace elements. Some asteroids have wa- unsustainable. Ground-based solar panels have be-
ter or hydrated minerals. The Moon is deficient in come popular in recent years, but orbital locations
these, because it formed in a molten state and has can provide seven times as much energy per panel
a low escape velocity, and they were baked out and on average. This is due to night, weather, and atmo-
lost. Mars and Venus have atmospheres with high spheric absorption on the ground. A large Space-
percentages of Carbon Dioxide, and Mars has Ni- Based Solar Power plant can send power to the
trogen. Asteroids are a source for the elements in ground using an efficient microwave beam. Advan-
fertilizers, and the trace elements. tages of orbital solar power are nearly 100% op-
erating time, and lack of Carbon emissions or nu-
clear risks. A disadvantage is the size of the col-
2.15.4 Energy lector on the ground is governed by the transmis-
sion wavelength and distance of the orbital station,
Uses for energy in space are as ubiquitous as they are on so there is a minimum size for it to function effi-
Earth. Solar Panels have bee used on satellites from the ciently. This can be counteracted to some extent by
earliest days, since they are modular, light-weight, reli- using shorter wavelengths or lower orbits. To be fea-
able, and produce more energy over their life than bat- sible for Earth, the total system (orbital and ground
teries or fuel cells. Power levels have varied from a few collector) needs to be less than 7 times as expen-
Watts to approximately 100 KW on the Space Station. sive as solar panels on Earth, otherwise using ter-
Future energy needs include larger amounts for propul- restrial panels would be less expensive. Alternate
sion and to operate industrial systems. Habitats, commu- uses would be to beam power to a Lunar surface base
nications, and scientific equipment can be large energy from orbit to supplement nighttime power.
consumers, and finally a major future use is exporting en-
ergy to Earth or other locations in space. Types of future At current launch costs, it makes economic
energy production include: sense to beam power *up* to space by swap-
ping the transmitter and collector locations, as
• Solar-Electric - This includes existing photovoltaic power in orbit is worth more than power on
solar panels, and solar thermal power systems. the ground. In the form of visible light or
The latter concentrate sunlight onto a heat en- microwaves this would supplement on-board
gine/generator combination. Solar flux is adequate power obtained from sunlight. For orbital tugs
in the inner Solar System, but large lightweight re- at low altitude, the supplement is especially
flectors may be used in the outer Solar System to useful as the Earth’s shadow covers 40% of typ-
improve power density. ical low orbits.
2.15. FUTURE PROJECTS 81

Laser power transmission is a future possibil- or large non-rotating ones, could support planetary-
ity. Beam generation is less efficient, but it scale populations. Surface habitats can range from
can be focused more easily over long distances individual pressurized modules, to permanent bases
due to the shorter wavelength. Uses range from and cities. At the limit, an entire moon or planet
powering launch vehicles from the ground, to can be converted to habitable conditions, which is
interstellar missions using the Sun as a gravi- known as Terraforming. Supporting planetary-
tational lens. Given a suitable atmosphere, for scale populations won't be needed for a long time,
example Carbon Dioxide rich ones like Venus so most current work is aimed at the lower end of
and Mars, the atmosphere might be used as the size scale.
a lasing medium to generate powerful beams.
This concept has not been explored in detail as Since the environment parameters that humans and agri-
far as is known. culture prefer is fairly narrow, habitats will have similar
functions regardless of size. These functions include:
2.15.5 Engineered Environments
• Atmosphere Maintenance -
All of space, and many parts of the Earth, have conditions • Temperature Control -
not suited to humans, or higher life in general. On Earth
we apply the various fields of engineering and technology • Artificial Gravity -
to modify the natural environment in specific locations, • Lighting -
such as buildings, ships, and aircraft. Common modified
conditions include temperature, protection from weather, • Radiation Level -
and pressure (in the case of aircraft and submarines). In • Food Supply -
space, we must modify additional conditions like atmo-
• Water Supply -
spheric composition (or entire lack of atmosphere), grav-
ity level for long term stays, radiation levels, and other pa- • Waste Disposal -
rameters. In some circumstances the environment would
be set up for plants (high CO2 ratio), or machines, rather Artificial habitats are much less mass intensive
than humans. To date, engineered space environments than natural ones. For example, the Space Sta-
have involved short-term travel in vehicles, even shorter tion uses roughly 100 tons to support each per-
times in space suits, and up to a year and up to 6 or so son, while the Earth uses about 5 trillion tons.
people in space stations. So habitats constructed in the Solar System can
Future projects may include larger populations and longer either support vastly larger numbers of peo-
stay times, construction of habitats in space rather than ple, allow vastly larger living space and energy
launching already-built units, and production and recy- use for current population levels, or only use a
cling of basic needs like air, food, and water. Reasons in- small fraction of available resources.
clude longer-duration exploration and science, commer-
cial and industrial activities, and the desire to live in in- • Closed Life Support - All space projects that in-
teresting and unique locations. Future space environment clude people require a Life Support System of
projects include: some kind. To date, these systems have been “open”
in the sense of needing outside supplies of oxygen
• Space Habitats - Humans evolved on Earth, so it and food. “Closed” systems recycle part or all of the
is the only place we know of where we can survive, materials used to sustain life, therefore the amount
even for a short time, without the help of technol- of stored or newly delivered supplies can be reduced.
ogy. Parts of the Earth, such as much of Antarctica, If coupled with local extraction of needed materials,
are lethal even in a short time without at least cloth- outside supplies can be eliminated entirely. Water,
ing. A Habitat in nature has the right conditions for air, and food are the principal items that would be
particular species to live. Artificial habitats, such as recycled. Closed systems can be artificial, using ma-
homes and greenhouses, are purposely built artifacts chines and chemical processes, or ecological, using
which provide the right conditions. Space habitats living things. For food at least, people have a prefer-
provide these conditions, but are located in space, ence for naturally grown items, and plants naturally
rather than on Earth. They are distinguished from produce Oxygen, so this tends to result in mostly
space vehicles and stations by long-term occupancy ecological systems. Closed life support can be in-
and size. Theoretically a space habitat could be tegrated with human living space, such as a habitat
sized for a single person, but needed technical skills dome with both living quarters and farm areas. Al-
and psychology probably set a lower limit at about ternately, a greenhouse might be optimized for plant
6-8 people. At the other extreme, linked assem- growth conditions, including high CO2 levels, and
blies or close formations of rotating orbital habitats, people use breathing equipment and remote control.
82 CHAPTER 2. PART 1 - SCIENCE AND ENGINEERING FUNDAMENTALS

• Habitat Construction - Launch from Earth in- fragments thereof. Below about 2000 km altitude,
volves passing through the atmosphere, where large this debris is denser than natural small meteoroids.
size increases drag, and very large objects are not Transporting nuclear wastes and very hazardous bi-
mechanically suited to fit on launch vehicles with di- ological materials from Earth to safer locations in
ameters of a few meters. The Space Station was space has been studied, but is currently too expen-
therefore assembled from many smaller prefabri- sive. Debris impacts can damage active satellites,
cated components over time. Even larger future where the latter are expensive for the same reason,
projects may use multiple orbital construction meth- so there is a stronger economic case for removing
ods. Folded structures, such as solar panels, have the debris.
been extensively used for decades. To contain an at-
mosphere, rigid pressure vessels have been used in Asteroids and comets are known to hit large
the past. Inflatable structures, which are collapsed bodies everywhere in the Solar System, includ-
for launch, are currently being demonstrated. In the ing Earth. Evidence includes impact craters on
future, large habitats may be assembled from com- every solid body, a comet impacting Jupiter,
ponents launched in compact form, or manufactured and current falls on Earth. A future project
locally in space. is diverting asteroids and comets which are
on dangerous paths. Extensive searches have
• Recycled Vehicles - A conventional rocket takes the been ongoing for Earth-approaching objects,
final stage, along with the payload, into orbit. By re- and their discovery rate is increasing as better
fueling the stage, or by converting the stage tanks telescopes are built for this purpose. Lunar im-
and structures to another use (such as an occupied pacts are often neglected, but more mass can
pressurized module), some payload weight and vol- be tossed off the Moon, because it’s smaller,
ume is saved. For example, the Skylab space station to end up in Earth’s larger gravity well nearby.
was made from a converted Saturn V 3rd stage. The You be killed just as well by a 1 ton Lunar frag-
conversion can be done before launch, as in the case ment as by a megaton asteroid, but the deaths
of Skylab, preparations for later conversion can be are more distributed in time and space. Meth-
installed, such as connections for refueling and at- ods to move or destroy dangerous objects is at
taching later hardware, or the stage can be entirely an early experimental stage.
unmodified, and all the changed performed in orbit.
A number of studies were done on re-using Space Long period comets are undetectable with cur-
Shuttle external tanks for purposes like pressurized rent technology until they get within about 10
living space or raw aluminum for orbital manufac- AU of the Sun. If one was headed towards
turing, but these did not progress to actual projects. Earth, there is not time to arrange a shift in
Re-fueling of upper stages has also been studied, but it’s orbit, so the only reasonable way to deal
not tested. with it is to use an interceptor with one or more
large nuclear bombs to fragment or destroy it.
Comet trajectories are hard to predict because
2.15.6 Transport they have natural rocket thrusters in the form of
gas jets. Future projects may place search tele-
Space transport began before reaching orbit, in the form scopes farther out, to find such comets earlier.
of ballistic missiles. It continues to be needed today, for They may also station interceptors farther out
delivery of new equipment and facilities, cargo, and raw into the Solar System, so fragments have more
materials. The primary transport method used to date time to disperse. If more time is available, dan-
is chemical rockets, both for launch from Earth, and in- gerous comets can be diverted using their own
space missions. More recently, air-breathing propulsion material as propellant, or another, smaller, nat-
using carrier aircraft, ion-electric, aerobraking, and grav- ural object can be diverted to collide with it.
ity assists have been used. Solar sails, electrodynamic,
and higher speed air-breathing engines are at an experi- • Interstellar Transporter - The energy to transmit
mental stage. Existing and future transport methods are the description of an object to another star, even
more fully described in Part 2 of this book. These meth- at an atom by atom level, is about a million times
ods will continue to be needed as long as space projects less than the energy to physically move the object
exist. Some future transport applications include: from one star to another. In the far future, after
a first probe sets up a receiving/replication station
• Hazard Removal - Artificial and natural hazards at the other star, other objects are more efficiently
exist in space and on Earth, and transportation meth- scanned, transmitted, and reconstructed at the re-
ods can remove them or place them in safe or- ceiving end. Using atomic scale technology (such as
bits. Types of artificial hazards include Space De- scanning tunneling microscopes) it may be possible
bris from old satellites, empty rocket stages, and to eventually scan and send people this way. The
2.15. FUTURE PROJECTS 83

subjective time to travel at the speed of light is then communications has been by microwave radio. Some ex-
zero, although the actual transit time is still governed periments have been done with laser transmission, which
by the speed of light. can provide much higher bandwidth, especially for loca-
tions beyond Earth orbit.

2.15.7 Science and Exploration • Gravity Lens Relay - Massive objects like stars
bend light via gravity. If you travel a sufficient dis-
Science and exploration have been the main motivations tance from the star, at least 550 AU for the Sun, that
for missions beyond Earth orbit, once the Space Race of light reaches a focus. A far future project would
the 1960’s wound down. They have also been important be to use the star as a giant lens, to focus com-
motivations in Earth orbit, but no longer the dominant munications over interstellar distances. For optimal
ones. Most work in these areas is paid for by government signal relay, you would use such gravity lenses be-
agencies, so their pace is limited by available budgets. tween pairs of stars, and boost the signal at each
Humans have not traveled past Low Earth Orbit since star before passing it on to the next. Such a net-
the Apollo program, but numerous Solar System Probes work could link the entire galaxy, although the speed
have visited all the discovered major planets, some aster- of light presents a significant obstacle to practical
oids and comets, and five have left or are on trajectories use. Gravity lenses can also be used for astron-
to leave our home system entirely. omy, which requires the least effort as you only need
Future human exploration of the Moon and Mars has equipment around the Sun, and for power beaming
long been discussed, but has mostly been held up by to distant spacecraft. The latter requires much more
lack of sufficient funding. Current development of the effort than communications, since the power levels
large Space Launch System (SLS) rocket, lower cost are much higher.
commercial vehicles, and better technologies, including
use of in-space resources, is starting to change the sit-
uation. With limited funding, divisive arguments about 2.15.9 Entertainment
“Moon vs Mars” and “Humans vs Robotic” have stalled
program planning. In reality, these are false choices. This category includes activities like Space Tourism,
With better technologies, multiple destinations are af- Zero-Gravity Sports, and other activities purely for en-
fordable. Robots can prepare the way for and assist hu- tertainment. Tourism on Earth represents about 3% of
mans. Leveraging commercial systems brings economies worldwide spending, and involves over a billion annual
of scale which agency budgets can't reach on their own. travel arrivals. There is every reason to expect a lot of
The proper way to think of it is science and exploration people will travel to space for entertainment, if only the
pave the way for future commercial and industrial activ- extreme cost and significant risk were reduced. Spec-
ity, and commerce and industry make the science and ex- tator sports in zero gravity may come sooner than gen-
ploration more affordable. Part 4 of this book discusses eral travel, because top athlete income is already in the
same range as astronaut launch cost to orbit. Racing or
both types of projects in the context of an integrated long-
term program. prize competitions are a possibility, since events like the
America’s Cup already draw huge budgets and involve
There are many planned astronomical instruments and in- high technology. Such competitions would also have the
terplanetary probes, and many more have been proposed, useful purpose of promoting improvements in space tech-
but are too far in time to be funded yet. Only a few hu- nology, as the Google Lunar X Prixe is demonstrating.
man missions beyond Low Earth Orbit are in develop-
ment or detailed planning. Detailed plans for Lunar or
Mars missions have not yet been prepared, although tech-
nology work is in progress.

2.15.8 Communication
The use of space for communications is well developed.
It is the primary purpose of 52% of the 1261 operational
satellites as of the end of 2014. Essentially all other satel-
lites have some communications functions. If costs were
lower, use of space communications would increase in
areas like satellite broadband. This isn't used much to-
day because of higher cost and technical limitations com-
pared to ground services. A future large network of low-
orbit satellites could overcome some of the limitations
and provide world-wide coverage. To date, most satellite
Chapter 3

Part 2 - Space Transport Methods

3.1 Space Transport Methods • The strength and other properties of the materials
used to build the rocket, which determines the mass
In Part 1 we discussed the basics of the relevant sci- of the vehicle hardware.
ences, including propulsive forces (Section 1.3) and en-
ergy sources (Section 1.4). These two items can be com-
bined into one or more space transport methods. In Part 2
we will discuss the numerous transport methods, includ-
ing those in current use, in various stages of development,
It happens that the combination of these three factors pro-
and proposed or theoretical ones. Before looking at indi-
duced a small, or even negative payload in an ideal one
vidual methods, we will briefly look at why reaching space
piece, long life rocket. The total launch mass of such a
has been so difficult and expensive in the past, and what
rocket consists of 3 main parts: Propellant, Vehicle, and
can be done in general to change that situation. We also
Payload. Payload mass is what is available after account-
present some reference tables listing the various methods.
ing for the first two, and in 20th century designs that re-
mainder could be negative, or at best a few percent. For
example, a single stage LO2/LH2 vehicle with a mission
3.1.1 The Space Transport Challenge velocity of 9000 m/s and an exhaust velocity of 4500
m/s might be 86.5% fuel, 10-15% vehicle hardware de-
An ideal transport system has an infinite life and mini- pending on operating life, and therefore −1.5 to +3.5%
mal operating cost when delivering a desired cargo. Real payload. The lower hardware weight was associated with
systems never reach that goal, and the designer’s job is more flimsy single use construction, and the higher weight
to approach it as closely as possible given the level of with thicker long life design. LO2/LH2 was the highest
technology and other project constraints like funding and energy fuel combination. The large ratio of rocket to pay-
schedule. For air and space transport, the desired cargo is load weight led to high launch cost measured in $/kg.
called a Payload. For transport purposes it is measured
in absolute mass (kg) or Mass Fraction, which is the per- Various compromises had to be made from the ideal con-
centage of payload mass to total vehicle mass including cept of a vehicle you could just refuel and launch again
payload. like an airplane. One was to make the hardware last just
a single flight, thus allowing lighter structures than ones
built to last many flights. Another is to drop parts of the
The Historical Problem vehicle during flight, producing a Multistage Rocket. As
fuel is used up less thrust is required to maintain acceler-
ation, so fewer or smaller engines are needed. Tanks also
The difficult job of the mid-to-late 20th century rocket
get emptied, so you can drop the excess engines and tanks
designer was to find the best compromise between high
once you don't need them any more. The remainder of the
cost and small payload when going to Earth orbit. This
vehicle starts anew to accelerate towards orbit, but has the
compromise was forced by three factors which conspired
benefit of the velocity gained with the previous stage, and
to produce a small payload fraction, and therefore high
less vehicle weight. Low service life and staging are are
relative cost:
expensive - you have to replace or re-assemble the hard-
ware, but were necessary given the state of technology.
• The mass and size of the Earth, which determines The designer had to find the best balance between high
the velocity needed to reach orbit, cost due to small payload, or high cost due to discarding
or rebuilding a complete rocket. Single-use, multi-stage
• The energy contained in chemical rocket propel- rockets are still the most common method used to reach
lants, which determines how much of them you Earth orbit, and so space transport to date has been very
need, and expensive.

84
3.1. SPACE TRANSPORT METHODS 85

21st Century Solutions to get from the ground into this orbit range. Travel be-
yond LEO has not been as constrained to chemical rock-
The mass and size of the Earth and the energy in chemi- ets as initial launch. That is because for vertical rocket
cal propellants are fixed, and structural materials improve launch you need to be able to accelerate the whole vehi-
fairly slowly. Significant improvements in cost can come cle at more than one gravity (9.8 m/s2 ) in order to take off
from using the latest materials, optimized fuel choices, at all, and maintain a relatively high thrust to prevent hit-
and recovery and reuse of vehicle parts. But ultimately ting the ground again while still below orbital speed. One
chemical rocket designs are limited by orbit velocity and of the key features of chemical rockets is their very high
fuel energy, which are fixed values. To make more dra- thrust to mass ratio. This made them attractive despite
matic changes in cost we therefore must break one or their low absolute efficiency.
more of the 20th century’s limiting factors. For exam- Once in a stable orbit, lower thrust levels can be used to
ple: travel further, since you are not in danger of immediate
re-entry. So alternatives to chemical rockets have been
• You can build large structures that span part of examined for those missions, and some of them even put
the Earth’s gravity well. This reduces the velocity to use in recent decades. There are now more alternatives
the vehicle needs to provide. Even a small reduc- for going beyond LEO, which give better performance or
tion shifts the near-zero payload fraction into posi- wider design choices. The larger change, however, is in
tive territory. Past systems have implicitly assumed the first step of getting to LEO, which was limited to one
chemical rockets do the whole job of reaching or- relatively low performance method until recently.
bit, because it was the only type of propulsion con-
sidered useful for that job. There is no physical law
that demands using one type of propulsion for the 3.1.2 Known Transport Methods
entire job, and using multiple methods often gets
better total performance. As a starting point for designers, Figure 2.0-1 lists the
83 main space transport methods known to the book’s
• There are now a multitude of alternatives to tradi- authors. They are described in more detail in the fol-
tional chemical rockets, ranging from some in actual lowing sections and are organized into logical categories
use, to those merely theoretical as yet. Part 2 of this by type. The order listed here and later in the book is
book attempts to list all the known ones, of which by similarity of type, and not by feasibility or develop-
there are 83 so far, not including variations. Many ment status. Those factors are considered later. Many
of them have better performance, reducing the fuel of these methods also have variations in the concept or
required and increasing the net payload. application. From this list, the designer can then narrow
the choices to the relevant options for a particular project,
and eventually the final selection. By starting with all of
• Since the mid-20th Century, better structural mate-
them, you can be sure no viable option is missed.
rials and lighter weight components have emerged,
but some current rockets still don't take advantage Figure 2.0-2 places the same transport methods in a table
of them. by their numbers, since including the names makes the ta-
ble too large. Rows in the table indicate different energy
sources, which are grouped by major types. Columns in-
Although having more options is more complex, the mod-
dicate how force is applied to generate motion. They are
ern designer should consider the full range of available
grouped in two main categories: (1) expelling a material,
transport methods, and apply them where they function
which generates an opposite reaction force by Newton’s
best. The potential gain in going from short life, low pay-
Law, and (2) the vehicle interacting with outside materi-
load transport to long life, high payload designs more than
als or forces. The table is an aid to organizing and think-
justifies the extra work.
ing about the methods. Some theoretical methods are not
shown on this table, since we do not have definite de-
Beyond Low Earth Orbit signs for implementing them. Some of the methods could
span more than one box if all their possible variations are
From an orbital mechanics standpoint, Low Earth Or- considered, but we have assigned them to single boxes in
bit, (LEO) is the altitude range below 2000 km, or about this version. Empty boxes can stimulate thought about
1/3 of the Earth’s radius. In this region the Earth’s gravity possible transport methods which are still unknown, or
is more than 58% of the surface value, and orbit periods whether that particular combination truly lacks possible
are less than 50% higher than the 84 minutes of a theoret- use.
ical orbit at sea level. From a practical standpoint, LEO is These two figures represent the state of knowledge in
the altitude range of 200 to 1,000 km. This is above sig- mid-2012. There are likely some additional methods not
nificant atmospheric drag, and below the Van Allen Ra- yet known to the authors, and new ones will likely be de-
diation Belts. The historical space transport problem was vised in the future. If a serious project is contemplated,
86 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

designers should survey the relevant specialists and liter- are strong enough to perform useful space transport func-
ature to include the latest concepts. tions. A structure only needs to be built once, but can be
used multiple times. The cost per use thus goes down the
more often it is used, and the longer the structural life is.
This is a very different economic situation than single-use
rockets, the main space transport method used to date.

3.2.1 Structures in General

For space transport purposes, we want to know how use-

ful a structural material is in and of itself, rather than as
part of a vehicle with other propulsion. To find that out
we can derive performance measures from the material’s
properties. Then we compare these measures to the trans-
port job of reaching orbit from the surface, or changing
orbits. The measures are relative to a body’s gravity well
or orbit velocity.

Gravity Wells

Conceptually, a Gravity Well is related to water wells

with steep sides, which you must climb if you find your-
self at bottom, requiring energy to get out. For large bod-
Figure 2.0-1 - List of space transport methods. ies, the “depth” of a gravity well can be expressed as the
surface gravity times the radius of the body, which has
units of m2 /sec2 . This may be thought of as climbing one
radius above the surface at constant surface gravity is the
same amount of work as climbing to infinity under the ac-
tual inverse square decrease of gravity with distance. For
Earth, that is 6,378,000 meters at 9.80665 m/ss (standard
surface gravity) or 62,547,000 m2 /sec2 . This is derived
from the formulas for potential energy and gravitational
force in the Physics section:
The formula for potential energy is −GM m/r where G
is the Gravitational Constant, M is the mass of the large
body, m is the mass of the object of interest, and r is
the radius of the large body. Since gravitational force g is
GM m/r2 , the potential can be expressed as −gr . Since
Figure 2.0-2. Transport methods by Energy Source and Force generally F = ma , dividing g by the object mass gives
Application. the potential per mass as −ar , where a is the surface
acceleration of gravity and r is the radius. This is a con-
venient form for calculations, since both surface gravity
and radius are usually known for large bodies.
3.2 Structural Methods This derivation of gravity well depth assumes a non-
rotating, uniform, spherical object. Real bodies depart
We begin our review of space transport methods with the from these assumptions in varying degrees. Determining
structures category. Surviving artificial structures appear the “depth” of any point on a real body can be done by
as early as 23,000 years ago, in the form of a rough stone corrections to the simple formula if they are small. For
wall at the entrance to Theopetra Cave. The wall was fast-rotating or irregularly shaped objects a numerical in-
probably intended to block cold winds, since that time tegration of the gravity field may be required.
was near the Last Glacial Maximum. A “tower whose
top is in the heavens” is famously mentioned in Gene-
sis 9:3, although the baked brick, stone, and tar technol- Scale Length
ogy of the first Millenium BC would not have been up to
the task. It is only since the late 20th Century that struc- In physics in general, a Scale Height H is a distance over
tural materials like Carbon Fiber became available, that which quantity changes by a factor of the mathematical
3.2. STRUCTURAL METHODS 87

constant e (2.718...). It is most often applied to change strength. For example, a space structure may be degraded
in atmospheric pressure with altitude due to gravity. The by impact damage or exposure to the space environment,
variation in pressure due to gravity for gases also hap- and no longer be as strong as intended.
pens in solids. So for structural engineering purposes we
can also calculate a scale height for columns or length for
cables. A vertical column or cable with a constant hori-
Efficient Design
zontal (cross-section) area has a mass m of
Constant area columns or cables are simple to design and
calculate for small structures, but are not an efficient de-
m = DAh
sign for large ones. This is because the load is a maximum
where D is the density in kg/m3 , A is the cross sectional at only one end of a large structure - the bottom for a col-
area in square meters, and h is the height in meters. The umn and the top for a hanging cable. A constant stress,
compressive force at the bottom (for columns), or tensile rather than constant area, design makes the best use of the
force at the top (for cables), due to its own weight, is found material by using all of it at it’s safe stress limit. There-
by the usual F = ma formula, where acceleration a in this fore the cross sectional area must vary to suit the local
case is the local one due to gravity. Dividing by the area loads. This results in minimum weight and cost, since
A gives force per unit area, or pressure P as you only use as much structure as you need at any point.
For example, the support columns of skyscrapers are typ-
ically larger and thicker near the bottom because they are
supporting the whole building and its contents, while near
P = DAha/A = Dha
the top floors they are supporting much less weight above
The tensile or compressive strength of a material, S, also them.
has units of force per area. Equating them and solving For a column where the mass you are supporting is at the
for h then gives the maximum height or length, H, the top, each part of the column below the top has to support
material can sustain before failing as that weight plus the part of the column above that point.
Therefore the column has to support an increasing load
S = Dha becomes H(scale) = h = S/Da as you go down, and needs a larger area to keep the stress
per area the same. Similarly for a cable with a mass at
A taller constant width column or longer constant width the bottom, each point along the cable supports that mass,
cable will exceed the strength of the material and there- plus all the cable below that point. So the cable cross sec-
fore collapse or break. Structural limits can be reached tion area should increase to support the increased mass as
from other causes than gravity, such as centrifugal accel- you go up. The amount of increase in both columns and
eration. We therefore call the general case Scale Length cables is 1/(working length) per meter, since the working
because it is not always due to height. As examples, com- length is over how many meters the stress will increase by
mon steel has a strength of 275 MPa and a density of 100% due to the structure itself. The fractional increase
7800 kg/m^3, and Earth surface gravity is 9.81 m/s^2, is based on the sum of supported + structural mass be-
so the scale length is 3600 meters. A very strong car- yond that point. The constant fractional increase in load
bon fiber/epoxy composite column has a strength of 1300 per meter results in an exponential taper, by a factor of
Mpa and a density of 1650 kg/m^3 and so a scale length e (2.718...) per working length. In other words the cross
of 80 km. Since the Earth’s radius and equivalent grav- section Area Ratio, AR, where h is the total length, is
ity well depth is 6378 km, the gravity well to scale length
ratio for this material is 80:1.
Scale length is a theoretical value like the ultimate tensile AR = A(bottom) = eh/(H(scale)/F S)
strength at which materials fail. Real designs will always A(top)
use some value below that so loads are well below the
failure point. We can define the Working Length as the In theory there is no limit on the area growth, you can just
scale length divided by the design Factor of Safety, FS. keep making the structure thicker at the bottom or top. So
This is a ratio of ultimate stress at failure to design stress. in theory you can build a structure of any height or length
It is based on experience and detailed understanding of with any material. In reality, the exponential growth in
how materials fail. The chosen value is intended to reduce area from a real material implies a similar growth in struc-
failure probabilities to an acceptable level for the given tural mass and cost. At some point the design becomes
purpose. The Margin of Safety, MS, is the Factor of infeasible to build. Where that point is depends on what
Safety - 1. It represents the amount of added stress above the purpose of the structure is. Feasibility is especially
those caused by expected loads which the structure can significant for large bodies like the Earth, where the ratio
withstand. Both safety values are intended to account for of gravity well to working length is high. This produces a
known and unknown variations in loads, and the actual large exponent in the area ratio formula if you try to span
strength of a structural element vs it’s theoretical design the entire gravity well with a single structure.
88 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

Tip Velocity

Some large structures involve rotation about a center,

rather than being vertical in a gravity well. This still re-
sults in a tapered design, but the loading force comes from
centrifugal acceleration rather than gravity. The general
formula for centrifugal acceleration at any radius on a ro-
tating object is
Figure 2.1-1. Example material properties.

a = v 2 /r
involved in understanding, developing, and applying ma-
where v is the velocity and r is the distance from the cen- terials. It is a well-developed subject, which can be ex-
ter. The distance traveled per rotation period is 2πr , so plored in textbooks, handbooks, and online courses, such
the velocity of any point between the tip and center is pro- as from MIT’s Department of Materials Science and
portional to r. Thus the acceleration at any point varies Engineering.
linearly with radius, and the average acceleration over the
Figure 2.1-1 gives some examples of common materi-
radius is half that at the tip. The same type of constant
als used on Earth, and a few that can be used for space
stress design as for a vertical structure leads to a taper-
projects where high strength is needed. Strength is not
ing area from center to tip of a rotating one. Instead of a
the only important property for material selection. When
near-constant gravity along the length of a vertical struc-
doing a detailed design you should do a full search for
ture, we have a strongly varying acceleration in a rotating
available materials and also consider all relevant proper-
one. When a rotating structure is near a large body, we
ties before a final selection. Real designs will have a factor
have gravity forces in addition to the centrifugal ones. As
of safety, which is not included in this table, and will also
the structure rotates, the direction and strength of gravity
have structural overhead for items like connector fittings
will vary relative to the centrifugal acceleration of rota-
and coatings. The overhead can be treated as extra loads
tion. If the structure is also in orbit, it also has centrifugal
to be supported or a reduction in the useful strength of
acceleration of orbital motion, which can largely cancel
the material.
that from gravity.
The strongest materials are fibers which are strong in ten-
A rotating structure made of a given material will accu-
sion. In order to use them for compression structures,
mulate stress from the varying accelerations from center
they have to be embedded in a matrix of some other ma-
to tip. Since the acceleration grows linearly from zero at
terial to give them stability. Otherwise they would simply
the center, the average is half the tip acceleration. Where
bend like a thread. An example is carbon/epoxy, which
a(tip) x r(tip) x 0.5 equals the working length, we have
encapsulates carbon fibers with an epoxy matrix. A typi-
the same accumulated stress as a vertical structure under
cal ratio by area is 60% fiber and 40% matrix. The epoxy
constant gravity. The velocity of the tip, v(tip) is then a
is relatively strong as a plastic by itself, but most of the
characteristic value for the material. We can compare the
strength comes from the fibers. We do not list carbon
characteristic tip velocity at one scale stress to the orbit
nanotube fibers or single-crytal solids like Diamond be-
velocity for the planet or body it operates on or near. This
cause we do not yet have ways to produce them in large
gives us a measure of usefulness for rotating structures in
enough pieces or quantities for large space projects. They
that location. For example, if tip velocity = orbit velocity,
have extraordinary strength, but until they can be mass-
you can build a device on the surface which mechanically
produced, they are not yet useful for large structures.
throws payloads into orbit, or reach down from orbit at
zero ground velocity if tip velocity cancels orbit velocity.
As with vertical structures, you can build devices which
rotate faster than their characteristic velocity, but at the 3.2.2 A. Static Structures
cost of increasing taper factor, and thus increasing mass,
Structural methods are divided into two main groups,
relative to the supported load or payload. For a large body
static and dynamic. Dynamic ones are discussed in the
like the Earth, large taper factors are needed to match or-
next section. Static structures have parts which are mostly
bit velocity, but useful designs can be built with lower tip
velocities. fixed in relation to each other. The structure as a whole
may move with respect to the ground, such as the main
truss of the International Space Station. Large struc-
Material Properties tures are primarily governed in their design by the ratio
of strength to density, or specific strength. That ratio is
There are a vast number of known materials, each with converted to a scale length by dividing by the local ac-
their own structural properties, and new ones are devel- celeration. Other important properties for a large struc-
oped on a regular basis. Materials Science is the field ture include stiffness, temperature dependence of prop-
3.2. STRUCTURAL METHODS 89

erties, and resistance to decay from the surrounding en- each station low. A pipe could also serve as
vironment. a pneumatic system to transport cargo besides
Some method to travel the height or length of the gases. The depth of the gravity well will deter-
structure is often required. These methods include: mine the practicality of this method.
conventional elevator (which does not need further ex-
planation), incremental winch, linear motor or rails, and Regardless of the method used, lifting a mass against
fluid transfer: gravity or centrifugal acceleration takes energy accord-
ing to the change in potential energy = mah, where m is
mass, a is acceleration against which you are lifting, and h
Incremental winch - A hanging cable, as is
is the height. For example, a 2000 kg passenger elevator
used with conventional elevators, that spans the
climbing 10 meters per second in Earth gravity requires
entire height of a tall structure, ends up dupli-
2000 x 10 x 9.81 = 196,200 Watts.
cating the loads of the main supporting struc-
ture and would be quite massive. An incre-
mental winch has a small motor-driven trolley 1. Towers
which pulls a length of cable behind it as it
climbs up the structure. The cable is unreeled Alternate Names: Space Tower, Megastruc-
from a spool on the elevator compartment. The ture
trolley then hooks the cable to a fixed point
on the structure some reasonable distance up. Type: Potential Energy via Mechanical Trac-
The cargo elevator remains attached to the next tion
lower point on the structure during this time.
The elevator then reels in the spool like a winch
to pull itself up from one attachment point to
the next. By this method the cable length and
mass are kept relatively low. The elevator car
requires power for the winch. This can be by
rails or wires attached to the main structure, so-
lar arrays or other power source attached to the
elevator compartment, or beamed power from
an outside source.

Linear Drives - Instead of a cable, this uses

either traction or magnetic force to climb the
structure. Traction would use friction pressure
against a rail or cable, or geared drivers against
a linear toothed rail. Magnetic forces would
use coils functioning as an electric motor, but
instead of the coils being in a circle and produc-
ing rotation, as in an ordinary motor, they are
in a line producing a linear motion. Magnetic
Leviataion (MagLev) trains work this way. As
with the winch, it needs an power source such
as wires or conductive rails.

Fluid transfer - Rather than moving an el-

evator compartment, a pipe could be used
for higher volume transfer. Dr. Dana An-
drews, formerly with Boeing, suggested pump-
ing Oxygen gas generated on the Lunar surface
Figure 2.1-2. Tokyo Skytree.
up to the Lunar L2 point on a Lunar space el-
evator. A column of Oxygen at .1 atmosphere
at L2, and a temperature of 1000 K (a solar Description:
heated pipe can be used to keep the gas hot) Towers are self-supporting compression structures,
would have a pressure of 2310 atm (234 MPa) whose main purpose is not habitation. When the main
at the bottom. So a single pressurized pipe sec- purpose is to house people, we call them tall buildings, or
tion puts heavy loads on the design. A better Skyscrapers. The Eiffel Tower is probably the most fa-
approach is to have pumping stations spaced mous example, but they are also commonly used as trans-
along the elevator, to keep the pressure rise at mitting towers, such as the w:Tokyo_Skytree (Figure
90 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

2.1-2). As noted in the example material properties table are used for construction, then the standard method of
above, advanced aerospace materials have scale lengths building from bottom to top can be used. Wind loads
of many kilometers, so it is possible to build towers in are significant below 20 km altitude, where the product
that size range. Such towers can be used as a high altitude of atmospheric pressure and wind speed produces max-
astronomical platform, a launch platform for a propulsive imum Dynamic Pressure. To reduce these wind loads,
vehicle, or a support structure for an accelerator system. the structural elements can be enclosed in pivoting air-
Design - In theory a tower of unlimited height could be foils, which have a much lower Drag Coefficient than
built. At some height, though, the exponential growth of circular or triangular struts.
the base area and total structural weight and cost makes Status:
it impractical to go higher. For example, let us assume a The tallest man-made above-ground structure is the Burj
structural factor of safety of 2.5, and that a launch tower is Khalifa, in Dubai at 830 m (2723 ft or 0.51 miles ).
used many times. The tower mass might then be limited The tallest freestanding structure is the Magnolia Tension
to 100 times the rocket and equipment mass at the top. A Leg Platform, which is 1580 m ( 5200 ft) from the sea
carbon fiber/epoxy tower on Earth would then be limited floor to the top of the surface platform. The tallest build-
to about 150 km in height. ing under construction is the Jeddah Tower, in the Saudi
In a real structure, the load, the mass you are supporting Arabian city of the same name. It is estimated to be
besides the structure itself, probably won't all be at the 1000 meters (0.63 miles) high when completed. A num-
top. Design calculations then have to account for where ber of larger Tall Buildings and Structures have been
the loads are distributed along the height. Additionally, proposed but not yet started construction. Civil engineer-
for the bottom 20 kilometers or so on Earth, wind loads, ing and construction are very well developed fields. Suit-
ice build-up, and other environmental effects have to be able materials exist for multi-km tall towers, though they
accounted for. Above 20 km, ultraviolet light and atomic may require a advances in construction techniques. The
oxygen can attack certain structural materials. This is not economics of such towers is more limiting at the present
commonly a problem at low altitude, so you need a pro- time.
tective layer for the structure or choose different materi- Variations:
als.
A large tower would typically be built as a truss, like the • 1a Unguyed Tower - This type is self-supporting
Tokyo Skytree example. The space between the verti- from its base, like a Pine Tree. Like trees, such
cal elements in a truss gives it stability, but it does not towers may need an extensive underground struc-
have to be a solid structure to support most reasonable ture to distribute the weight and tipping forces like
space-related loads. Truss elements will bend if too much wind. The base diameter will typically be 5-10% of
load is placed along their length - imagine pressing on the the height to prevent bending. In the lower part of
ends of a drinking straw or spaghetti noodle. Stiffness or the tower, wind loads may require the base to spread
Elastic Modulus is a material’s resistance to this bend- more than the upper part, which only depends on
ing. To make best use of the material strength, the design buckling for its necessary width. This approach as-
is often made so that buckling (failure from bending) and sumes that most of the loads on the tower are verti-
crushing (failure from direct load) happen at the same cal, as in an elevator riding up and down the tower
time. For high strength materials this results in individ- height. Carbon-epoxy materials can support up to
ual elements roughly 20 times longer than their smallest 700 MPa in compression. The largest rocket in de-
cross section. It also results in the tower as a whole being velopment is about 2.55 million kg in mass. If we al-
roughly 20 times taller than the base width. These are low 10 million kg total for an example use like rocket
only generic values, the real ratio would be determined launch platform on top of the tower, the load is about
by structural analysis of the actual design conditions. 100 MegaNewtons (MN). This load can then be sup-
Construction - These types of towers can be built 'from ported by 1/7 of a square meter of cross section of
the top down' in order to avoid human construction work carbon-epoxy columns. This is far less than the min-
in a vacuum. In this process, the top section of the imum launch platform size of about 2500 square
tower is assembled at ground level. Hydraulic jacks then meters. So the columns can be hollow tubes, and
raise the tower up by one section length. The next sec- the tubes spaced apart from each other in an open
tion down is then installed underneath the top section. truss (Figure 2.1-2). The spaces between the vertical
The progressive jacking process is repeated for the whole structural elements can be used for other purposes.
tower height, so all the construction work takes place near
ground level. Special anchoring provisions are required to • 1b Guyed Mast - Masts are structures stabilized
stabilize the tower while being built in this fashion. Since by diagonal supports like ropes or wires, as in the
the tower is typically tapered, anchor masses, jacks, and Rigging of sailing ships. Diagonal supports can bet-
assembly cranes must gradually move outward from the ter resist sideways forces like wind. This is com-
center as the tower grows. If remote controlled robots monly done with television and radio towers because
the antenna itself is not very heavy, and so the main
3.2. STRUCTURAL METHODS 91

loads are winds on the lightweight tower structure. A

very tall structure may combine diagonal supports in
the lower 20 km, where wind is significant, and be
Space Elevator
self-supporting in the upper portion.
Counterweight
• 1c Series of Towers - An electromagnetic accelera-
tor for people and delicate cargo may be hundreds of
km long, with the upper end many km high to avoid
aerodynamic drag and heating, forming a long ramp.
To support the device you can use a series of tow-
ers of increasing height as supports, with connect- Center of mass
ing structure similar to a suspension bridge between for system
(above geostationary level)
them.

• 1d Inflatable Tower - Many materials are stronger Geostationary

in tension than compression, so concepts like the orbit altitude
ThothX Tower have been proposed, using internal
gas pressure to put the structure in tension and sup-
port loads.

Rotates with Earth,

References: remains vertical. Cable

• Krinker, M., Review of New Concepts, Ideas, and

Innovations in Space Towers, 2010.

Climber
2. Space Elevator

Alternate Names: Skyhook, Beanstalks, Ja-

cob’s Ladder, Space Bridge, Geosynchronous
Towers, Orbital Tethers Anchored at equator

Type: Potential Energy via Mechanical Trac- North Pole

tion
Earth
Description:
In 1895, astronautics pioneer Konstantin Tsiolkovsky
was inspired by the recently built Eiffel Tower. He
envisioned a tower that would reach all the way to
Geostationary Orbit, where orbit period matches the
Figure 2.1-3. Basic space elevator concept.
Earth’s daily rotation. We now refer to any similar con-
cept as a Space Elevator, because you could use an eleva-
tor rather than a rocket to reach space. Since the top end
matches local orbit velocity, you can merely let go and are than gravity, and the weight then pulls upward on the at-
in orbit. Tsiolkovsky’s concept was more a thought exper- tached cable. This version of the space elevator is often
iment than a practical design, because in 1895 no mate- used in popular illustrations, but early 21st century mate-
rials existed with anywhere near the required strength rials are still not strong enough to make this size elevator
A more modern version of the concept (Figure 2.1-3) as- practical. Smaller versions with shorter cables, however,
sumes a cable in tension rather than a tower. This is be- can result in feasible designs. We can call these Staged
cause fibers like PAN-derived carbon are currently the Elevators since they only provide part of the velocity to
highest strength available bulk materials, and bulk quan- reach orbit, in the same sense that the stages of a multi-
tities are needed to build a space elevator. The part of stage rocket do. A different transport method or multiple
the elevator below stationary orbit sees more gravity than elevator stages are needed to do the full job of reaching
centrifugal acceleration, and would fall down if unsup- orbit.
ported. Therefore this version requires a counterweight If staged elevators are not connected to the ground, they
above that orbit, where centrifugal acceleration is higher are not required to be centered on a 24 hour orbit, nor ro-
92 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

tate at the same rate the Earth does, creating a whole class can store and transfer momentum and potential energy
of possible designs. They can be located in whatever orbit to vehicles or cargo, and support objects away from the
is needed, and will appear to be in motion when viewed structure’s center of mass or under rotation. Structures
from the ground. For bodies smaller than the Earth a full in orbit will naturally tend to align vertically (Figure 2.1-
elevator is more feasible, because the lesser gravity well 4, right) because gravity forces decrease as the square of
can be spanned with existing bulk materials. If future distance. So the lower end of a vertical cable has a lower
materials about 3 times stronger than current carbon fiber potential relative to the center by more than the higher
become available, the original single-stage elevator would end sees a higher potential. The total energy of a vertical
become feasible from a structural materials standpoint. It cable is then less than that of a horizontal one, so it tends
would still be a very large construction project, and have to fall into that state. The difference in gravity (otherwise
other technical challenges to overcome. The rotation state known as tides) provides tension to keep the structure ex-
of the elevator can be vertical, which is one rotation per tended.
orbit when seen from inertial space; swinging, where it
Staged elevators can be built any size up to the practi-
varies by some angle from vertical but does not do a full cal limit of the materials used. In the lower limit of zero
rotation relative to the ground; or rotating where it does
length a space elevator reduces to a simple object in or-
rotate relative to the ground. The rotation sense can be bit. For intermediate lengths and vertical orientation, the
forward, where it is the same direction as the orbit around velocity of the bottom end is the velocity of the center
the planet, or backward where it is opposite. Normally it times bottom/center distances r from the center of the
would be forward, since that results in lower velocity at body. This simply reflects that the center and bottom
the bottom end relative to the ground, and higher veloc- travel paths of length 2πr in the same orbit time. The
ity at the upper end for injection into transfer or escape bottom’s velocity can be further reduced if you rotate the
orbits. entire cable such that the bottom end travels opposite the
For any type of space elevator, the structure can be used orbit direction relative to the center (Figure 2.1-4, left).
multiple times over it’s design life. So the construction The tip’s maximum rotation velocity relative to the cen-
cost is divided by the number of times it is used. The ter is then governed by the mass and cost of the structure
larger the structure is, to gain more performance, the growing exponentially as you make it more capable. For
higher the cost. When maintenance and operations are the Earth, and current bulk materials, it is not practical
added to the construction cost, there will be an optimum to counter all of orbit velocity with rotation, but about
size and cost for a given traffic level. Mass and cost grow half can be. On smaller bodies the entire velocity can be
exponentially with size and performance, and construc- countered by rotation, so that the bottom end momentar-
tion cost per use only decreases as the inverse of the ily comes to rest relative to the body’s surface.
number of uses. So at some point the lower cost from
The remaining velocity of the bottom end relative to the
many uses is overcome by the increasing mass and cost surface has to be provided by some other method. When
of a larger structure no matter how many times it is used.
a body rotates, like the Earth does, some of the remainder
Therefore economics is a limiting factor on large space can be supplied that way. In our planet’s case, equatorial
elevators.
rotation velocity is 465 m/s, or 5.88% of orbit velocity.
The roughly half of orbit velocity not supplied by the el-
evator’s or the Earth’s rotation is a substantial improve-
ment over needing to supply all of it. The difficulty of
building a transport system like a chemical rocket is also
non-linear with velocity. For example, if it could provide
3.5% of launch mass as payload to orbit doing the whole
job, it could increase payload to 27% if it only needs to do
half, an increase 7.7 times. By using a rocket + elevator
as a two-stage system, the sum of smaller mass exponents
is less than each exponent by itself, so the total cost and
difficulty is less using both than using either one for the
entire job.
The rocket stage does not place itself fully in orbit. It
docks temporarily at the tip of the cable, and only the
Figure 2.1-4. Staged elevators: rotating (left) and vertical (right). cargo travels further using the elevator. This is less total
energy per delivery than if the rocket stage hardware had
Gravity and Stress - Where a tower has to resist the to travel all the way to orbit. The elevator cable serves as
compression force created by gravity against the solid sur- a momentum bank to store orbital kinetic energy, which
face of a body, a space elevator needs to resist the ten- can be transferred to the payload. The kinetic energy can
sion force created by the difference in gravity between be stored by running an electric propulsion system, which
it’s parts, or from it’s own rotation. Structural elements is much more efficient than conventional rockets. Using
3.2. STRUCTURAL METHODS 93

an elevator stage lowers the overall difficulty of reaching Status:

orbit, and can lower the overall cost if well designed. Ma- The Elevator on Earth dates as early as Archimedes in
terial strength to density ratio is the critical criterion for 236 B.C. The modern safety elevator was introduced in
designing these types of transport systems. Their mass 1852 by Elisha Otis, and a descendant of his company is
is highly non-linear with strength because doubling the still the largest manufacturer of vertical transport. Ropes
strength reduces the exponent part of their mass ratio by and cables have long been used on Earth for hoisting loads
half. and stabilizing tall structures such as ship masts and trans-
Structural Dynamics: - The forces affecting a space el- mitting towers. A space elevator refers to a complete
evator design vary with time, so they are dynamic forces transportation unit. A Space Tether refers to a cable in
rather than static ones. This includes arriving and depart- space that has a number of possible uses, one of which
ing vehicles, internal movement of cargo along the struc- is the structural element of a space elevator. A number
ture, deployment of extension cables if used, thrust for of Space Tether Missions have been flown as experi-
orbit maintenance and rotation, varying gravity in an el- ments, but not yet as an operational transportation sys-
liptical orbit or if the elevator is rotating, varying tidal tem. The largest rigid structure in space to date is the
forces from the Sun, Moon, or planetary satellites, and ISS Integrated Truss Structure, which is 108.5 meters
thermal stress from going into and out of shadow as it long. Taller structures on Earth await economic reasons
orbits. Elevator designs can be truss-like, with sufficient to build them, they have not reached the limits of available
compressive structural elements to keep a stable shape materials. Space elevators require enough traffic to sim-
against these varying forces. They can also be cable- ilar orbits to justify their construction. Significant traffic
like, with primarily tension elements, and the structure exists as of 2016 to synchronous orbit, but not enough to
allowed to flex with the applied forces, or they can be a justify a synchronous-capable elevator system.
mixture of the two. Active damping for vibrations can be A Variable Gravity Research Facility has been pro-
applied with shock absorber/spring combinations or with posed in Earth orbit to study the effects of partial gravity
thrusters along the structure. Design for dynamic forces on people and growing plants. Such partial gravity lev-
is similar to the design problem of a suspension bridge, els are found on the Moon and Mars, and it is currently
which must withstand static forces from its own weight, unknown what gravity level is necessary for health and
and dynamic ones from vehicle traffic and varying winds. plant growth on long missions. Such a facility would in-
In both cases, the design must not exceed safe working clude large space structures to generate artificial gravity,
stresses at any point in the structure, under any combina- and can also serve as a test bed for tether dynamics and
tion of forces.
operations. It would be a step towards required knowl-
Because of the typically high slenderness ratio (ratio edge to build an operational space elevator system.
of length to maximum width) and varying forces noted Variations:
above, the structural dynamics will be complex and re-
quire a good theoretical understanding and likely com-
• 2a Orbital Vertical Elevator - This is the most
puter simulation. A further complexity is unlike terres-
commonly presented space elevator concept. It has a
trial skyscrapers, which are constructed empty and then
cable kept vertical in smaller versions by tidal forces,
loaded when complete and not usually changed after-
or in larger versions by sufficient cable or counter-
wards, a space elevator may grow over time while already
weight past GEO to apply tension to the part below
operating. This is likely because a large elevator can as-
GEO. Mass grows exponentially with gravity well
sist in it’s own construction by reducing the work for a
depth. Therefore a compressive tower built up from
launch system, and can help offset it’s cost by operating
the ground meeting a cable from above results in
as soon as possible. So instead of analyzing a completed
lower total mass, because it splits the structural task
building and then checking construction loads do not ex-
into two smaller exponentials. Despite that, current
ceed design loads, a growing elevator would need analysis
materials are not sufficient for a full vertical eleva-
over a continuous range of sizes.
tor on Earth. They are for smaller bodies such as the
Maintenance and Repair - Space elevators, like large Moon or Mars.
structures on Earth, are subject to environmental degra-
dation and occasional sudden damage. These include • 2b Momentum Transfer Slingshot - If a payload
atomic oxygen, electrostatic discharge, solar UV and high is released from the end of a vertical elevator, the
energy particles, trapped radiation belts, impacts from other end of it’s orbit will be changed about 7 times
natural meteoroids and human-made orbital debris, and the initial distance from the elevator center of mass.
accidental vehicle collision or pressurized system failure. This is because while attached the payload is forced
For long-term operating life and safe operation, a space to move at a different velocity than a free object
elevator has to be designed for all these causes, with a pro- would at that altitude. Once released, it then follows
gram for maintenance and repair. In case of catastrophic the free orbit defined by its release velocity. This
damage, the design should minimize risk to people and variation increases the orbit change by adding par-
property. tial rotation and dynamic extension of a cable.
94 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

A variation from vertical cables is the orbital to altitude, after which it continues to orbit by other
slingshot. This would take advantage of the methods. With an aircraft this can be a simple tow
tendency of a long object to auto-rotate from cable where one vehicle pulls another, or a cable sys-
horizontal to vertical orientation about the cen- tem which dynamically snatches and accelerates a
ter of mass due to “tidal” effects. A relatively vehicle, possibly tossing it higher than the aircraft
light-weight vehicle, launched conventionally, flies. It requires less modification of the towing air-
would dock with a much more massive “or- craft and not having to deal with combined aerody-
bital momentum bank” (largely consisting of namics. For an airship type lifter, it avoids having
discarded rocket stages left at the bank with to build a tower that height, although cargo mass is
each launch), and be hooked to a cable reel. relatively limited.
The vehicle would be pushed out to a somewhat
higher orbit, where it would fall behind the mo- References:
mentum bank, with cable being paid out at a
matching rate. After sufficient tether has been
• Pearson, J. Konstantin Tsiolkovski and the Ori-
paid out, it would be braked to a halt, putting
gin of the Space Elevator, IAF-97-IAA.2.1.09,
it under tension. The momentum bank would
48th International Astronautical Congress, Turin,
slow and fall inward, while the vehicle would
Italy, 1997.
be accelerated and fall outward. It is released at
the desired orientation and velocity to transfer • Cosmo, M. and Lorenzini, E. Tethers in Space
to a higher orbit. Unlike an “elevator” system, Handbook, 3rd ed., Smithsonian Astrophysical
the tether need not be long enough to contin- Observatory, Dec. 1997.
uously reach the ultimate orbit, as the vehicle
will be “slung” outward up to 14 times the cable • Carroll, J. Guidebook for Analysis of Tether Ap-
length. plications, for Martin Marietta Corporation, Mar,
1985.
The momentum bank loses velocity in this ma-
• Multiple Authors, Space Elevator, search of NASA
neuver, but could use highly efficient solar-
Technical Reports Server, approx 4775 items from
powered electric engines (plasma, ion, or mag-
1916 to 2016.
netic) to recover that loss over an extended pe-
riod. Multiple momentum banks can be used • Alpatov, A. et. al., Dynamics of Tethered Space
in series the achieve higher orbits or greater fi- Systems, CRC Press, Apr. 2010.
nal velocities. If the momentum bank uses an
elliptical orbit (cheaper for a rocket launched
• Rotating Elevator -
vehicle to intercept), it may be possible to in-
sert objects into near-circular orbits by slinging
at apogee. The vehicle can also take on fuel • Carley and Moravec, The Rocket/Skyhook Com-
at the momentum bank, as the empty rocket bination, L5 News, March 1983
stages already have propellant tanks that can be • Ebisch, K. E., Skyhook: Another Space Con-
re-used. The slingshot approach has a moder- struction Project, American Journal of Physics, v
ate size and velocity capacity. 50 no 5 pp 467-69, 1982.
• 2c Orbital Rotating Skyhook - This is an idea de-
vised around 1980, where a cable is kept in tension • Baracat, William A., Applications of Tethers in
by sufficient rotation rate. On smaller bodies, where Space: Workshop Proceedings Vols 1 and 2. (Pro-
the cable end can dip low enough to grab cargo and ceedings of a workshop held in Venice, Italy, Oc-
lift it to orbit, the Skyhook name is generally used. tover 15-17, 1985) NASA Conference Publication
For Earth reaching that low is difficult because of 2422, 1986.
the high tip velocity and mass needed. Instead a ve- • Anderson, J. L. “Tether Technology - Conference
hicle coming from the ground provides enough ve- Summary”, American Institute of Astronautics and
locity to meet a slower rotating tip. This version Aeronautics paper 88-0533, 1988.
is often called a Momentum Exchange Tether or
Rotovator (rotating elevator). The momentum be-
ing exchanged is between a vehicle/cargo and the 3. Aerostat Alternate Names: High altitude balloon,
and the elevator system. Again, both launch vehicle Airship, Inflatable Tower
and Skyhook mass ratios are exponential in velocity,
Type: Potential Energy by Aerodynamic forces
so splitting the job lowers the overall difficulty.
Description: This method uses lift generated by pres-
• 2d Atmospheric Elevator - For this concept an air- sure and density differences but not primarily from ve-
craft or balloon/airship uses a cable to lift an object locity such as wing lift. One approach to minimizing drag
3.2. STRUCTURAL METHODS 95

and gravity losses for a launch vehicle is to carry it aloft avoids those losses by using a lower density gas or vac-
with a high altitude balloon or airship. Research balloons uum. Lower density can be obtained by using a gas with
have carried ton-class payloads in the range of 15-30 km a lower atomic weight, such as Hydrogen, or by pumping
high, which is above the bulk of the atmosphere. An- out some or all of the gas. This is not a transport method
other approach that has been proposed is to use pressure- in and of itself, but rather a way to avoid losses.
supported structures of great height. The highest strength Status: Low pressure pipes are a common device. It has
materials are strong in tension, so an inflated structure in not been tried for space transport.
theory can support itself. Wind loads on a large pres-
surized structure are a major design issue. If a less dense Variations:
gas is used than the surrounding atmosphere, the structure
will be partially buoyant and not require the same scaling • 4a Light Gas Tunnel - One or more light gas bal-
as one that depends on compressive strength. Sufficiently loons or pipes are strung along the path of a vehi-
large structures, which would have low surface to volume cle or projectile. The gas has a lower density than
ratios, could float just from heating the interior air. air. The formula for drag is 0.5*C(d)*Rho*A*v^2,
where Rho is the density. Thus the lower density will
Status: Balloons, airships, and pressure supported struc-
lower drag. High speed travel through any gas will
tures have been in use for a number of years, and some
develop shock waves, so the size of the projectile
experiments have been done to launch a rocket from a
relative to the size of the tunnel needs to be small
balloon. They have not reached orbit yet.
enough that the shock waves will not damage the
Variations: structure.

• 3a Balloon Carrier - A device producing lift and • 4b Evacuated Tunnel - An evacuated tunnel is sup-
carrying an instrument package or launch vehicle, ported up through the atmosphere by a combination
but not a propulsion system of it’s own. They have of towers or it’s own lift from displacing air.. A
been used extensively on Earth for science, and been launch system such as an electromagnetic acceler-
proposed for other planets. ator fires a projectile up through the tunnel. Drag
losses are minimized within the tunnel, and are low
• 3b Airship - A device combining buoyant lift and in the remaining part of the atmosphere beyond the
and some combination of aerodynamic lift and for- tunnel. The top end requires some way to keep air
ward propulsion. from flowing in and filling the tunnel - such as a hatch
that remains closed until the accelerator is about to
• 3c Orbital Airship - This concept has been pro-
fire.
posed by JP Aerospace. It involves a very
lightweight airship which starts from a floating plat-
form and accelerates via solar-electric thrusters to References:
orbital velocity. It is not known if this is technically
possible. 5. Magnetically Supported Structure Alternate
Names: Startram
• 3d Geodesic Sphere - A triangulated frame sup-
porting a pressure skin can float with merely a tem- Type: Magnetic Storage by Magnetic Field
perature difference between inside and outside if Description: A static or time varying magnetic field pro-
sufficiently large. Since structure mass goes with duces a force to support a structure. For example, a se-
area, and lifting force goes with volume, if built suf- ries of large superconducting coils stacked so they repel
ficiently large it will float. each other and support a cargo. Alternately current carry-
• 3e Pressure Supported Tower - Uses lift force ing wires generate repulsion between the ground and the
from higher interior pressure to raise a structure. structure.
This can be generalized to pressurizing any struc- Status: Startram is a concept proposed using magnetic
tural element to help support it. levitation, but has not reached experimental versions yet.
Variations:
References:
References:

4. Low-Density Tunnel Alternate Names:

3.2.3 B. Dynamic Structures
Type: Kinetic Energy by Aerodynamic Forces
Description: Traveling to or from a large body with an Static structures rely on constant forces such as from the
atmosphere, such as the Earth, can produce large losses strength of materials to hold themselves up. Dynamic
from drag and heating. Aerodynamic drag has a gas structures rely on the forces generated by rapidly mov-
density factor in it’s formula. This concept reduces or ing parts to hold up the structure. The advantage of this
96 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

approach is it can support structures beyond the limits of can also called geostatites, since they are stationary
material strengths. The disadvantage is that if the ma- with respect to the ground) are kept from falling de-
chinery that controls the moving parts fails, the structure spite to their low velocity, by a pelletized or solid
falls apart. mass stream that weighs more and is moving at a
rate somewhat higher than orbital velocity. The
mass stream maintains a lower orbital altitude by
6. Fountain/Mass Driver Alternate Names: deflection from the statite constellation onto shorter
Type: arcs. In the case of a low-mass stream, the hyper-
orbital path is close to a straight line, and a higher
Description: An electromagnetic accelerator provides
number of statites is needed to divert it around the
a stream of masses moving up vertically. A series of
earth without cutting through the atmosphere (in a
coils decelerates the masses as they go up, then accel-
polygon-like pattern). In the case of a high-mass
erates them back down again, at a few times local grav-
stream it moves only slightly higher than orbital ve-
ity. When they reach bottom, the accelerator slows them
locity, but has a lot more mass than the statites.
down and throws them back up again, at a high multi-
ple of local gravity. Thus the accelerator is many times
shorter than the fountain height. The reaction of the coils Alternate Names: Orbital Ring System
to the acceleration of the fountain of masses provides a (ORS), Jacob’s Ladder, Skyhook
lifting force that can support a structure. The lifting force
is distributed along where the coils are located. This can
be along the length of a tower, or concentrated at the top, 8. Multi-Stage Space Elevator Alternate Names:
with the stream of masses in free-flight most of the way.
Type:
Status:
Description: A multi-stage space elevator has more than
Variations: one structural element, with the parts in relative motion.
References: For example, a vertically hanging cable in Earth orbit
can have a rotating cable at it’s lower end. The advan-
tage of such an arrangement is to lower the mass ratio of
7. Super-Orbital Mass Stream Alternate Names: cable to payload compared to a single cable. The mass
Launch Loop ratio of a rotating cable is approximately proportional to
exp(tip velocity squared). If two cables each supply half
Type:
the tip velocity, then the ratio becomes exp(2(tip veloc-
Description: A strip or sections of a strip are maintained ity/2)squared), which is a smaller total mass ratio. An-
at super-orbital velocities. They are constrained by mag- other feature of a multi-stage elevator is that the tip ve-
netic forces to support a structure, while being prevented locity vector of the two stages add. Since one rotates with
from leaving orbit. A vehicle rides the strip, using mag- respect to the other, the sum of the vectors changes over
netic braking against the strip’s motion to accelerate. Sev- time. Given suitable choices of tip velocities and angu-
eral concepts using super-orbital velocity structures have lar rates, one can receive and send payloads with arbitrary
been proposed. One is known as the 'launch loop'. In speed and direction up to the sum of the two vectors. The
this concept a segmented metal ribbon is accelerated to dynamics of a multi-stage elevator are complex.
more than orbital velocity at low Earth orbit. The ribbon
Status:
is restrained from rising to higher apogees by a series of
cables suspended from magnetically levitated hardware Variations:
supported by the ribbons. The ribbon is guided to ground
level in an evacuated tube, and turned 180 degrees using
• 8a Hanging/Rotating Elevator - This consists of
magnets on the ground. A vehicle going to orbit rides an
a vertical/nonrotating space elevator structure with
elevator to a station where the cable moves horizontally
a rotating second stage at one or both ends. This
at altitude. The vehicle accelerates using magnetic drag
is more suited for within a gravity well, where the
against the ribbon, then releases when it achieves orbital
gravitational gradient will stabilize the first stage.
velocity.
Status: • 8b Rotating/Rotating Elevator - This consists of
Variations: two stages, both of which are rotating, to get reduced
mass ratio for a given velocity. This is more suited
for free space application where the lack of varying
• 7a. Birch Ring - Consists of two parts. One is a
gravity across the structure will simplify the dynam-
constellation of one or more low-altitude “geosta-
ics.
tionary” satellites with a space elevator attached, at
a distance from the ground that would be LEO in a
normal orbit. This constellation of satellites (which References:
3.3. GUNS AND ACCELERATORS 97

References has to be stopped by a braking system. For moun-

tain peak locations, the drop weight runs down a set
of rails and is stopped by running into a body of wa-
3.3 Guns and Accelerators ter or running up an opposing hillside plus possibly
wheel braking. The mountain location may be pre-
This section covers Guns and Accelerators. Although the ferred because of the greater launch altitude. This
details are quite varied, what they share is a larger fixed example assumes a mountain with a solid weight
installation which provides acceleration to a smaller pro- sliding on rails:
jectile or cargo. The construction of a large fixed device
is justified if you use it many times and the maintenance
per use is moderate. We assume that a 15,000 kg cryogenic rocket using RL-
10 engines is being thrown. An acceleration of 60 m/s^2
Note: This section is continued at Guns and Acceler- is tolerable by humans for the 20 seconds required to
ators 2 due to page size. reach 1200 m/s assuming the human is in good health
and properly supported. The linear path traversed would
be 12,000 m (7.5 miles) at constant acceleration. The
A. Mechanical Accelerators tow cable pulls with 900 kN (202,000 lb) of force on the
rocket. High strength carbon fiber is available with ten-
As the name indicates, this group uses direct force from sile strengths of up to 6.9 GPa (1,000 ksi). Allowing for
the device structure. a factor of safety and a braided overwrap surrounding the
fiber bundle (to protect the carbon fibers from abrasion),
we assume a working stress of 3.45 GPa (500 ksi). The
9 Leveraged Catapult Alternate Names: cross sectional area required is then 0.00026 m^2, or a
Type: circular cable 0.0182 m (0.72 inch) in diameter.
Description: A leveraged catapult uses a relatively large With a density of 1840 kg/m^3, the cable mass is 0.48
or heavy driver to accelerate a smaller payload at several kg/m. For a 12,000 m long cable, the mass would be
gravities by mechanical means. Such devices date back 5,760 kg without allowing for taper. Since the cable has
to the Medieval period, but this is an updated version us- to be accelerated also, the leading end, closer to the drive
ing modern materials. Devices such as a multiple sheave mechanism, has to be able to apply sufficient force to ac-
pulley or a gear train convert a large force moving slowly celerate the rocket and all the cable in between the rocket
to a small force moving fast, and transmit the force along and leading end. To a first approximation, the leading end
a cable. The mechanical advantage produces more than must be 40% larger in cross section than the trailing end
one gravity of acceleration. This concept may be one of to account for the acceleration of the cable itself. The
the simplest to implement on a small scale. Despite the mean weight of the cable is then 0.576 kg/m, giving a ca-
seeming simplicity of the concept, velocities of several ble mass of 7,000 kg. Something has to be done with the
km/s are possible, which would greatly reduce the size of cable after it finishes the job of accelerating the vehicle.
a rocket needed to provide the balance of the velocity to It can be taken care of by looping the cable back from the
orbit. The performance of this concept reaches a limit drive mechanism to the starting point (where the vehicle
due to the weight, drag, and heating of the cable attached is at the start). At the completion of a launch, the loop of
to the payload and the magnitude of the driving force, cable is moving at 1200 m/s, and is gradually allowed to
which is divided by the leverage ratio to yield the force come to a stop. Some method will be required to reduce
on the payload. hoop stress on the cable as it makes turns around a pulley
at high velocity.
Status:
The return portion of the cable has to be able to accel-
Variations:
erate the part of the cable past the drive mechanism at 6
g’s. Given a 7000 kg lead section, we get a mass of 3200
• 9a Drop Weight - A falling mass is connected to kg. Thus the total mass accelerated at 6 g’s is 25,200
a vehicle by a high strength cable running over a kg, and the total accelerating force is 1.512 MN. Given a
multiple-sheave pulley, cable reels with different di- 45 degree slope on a mountain, we assume that the drop
ameters, or connected with a gearing ratio. Two weight accelerates down at 3 m/s^2. Thus the gear ra-
types of natural locations with large height differ- tio is 20:1, and the force of the drop weight on the drive
ence are possibilities - river gorges and mountain system must be 30.24 MN. Since a free-falling weight on
peaks. Locations such as the Grand Canyon and the this slope would accelerate at 7 m/s, there is a 4 N/kg re-
Columbia River gorge have lots of vertical relief for tarding force due to the drive system. Therefore the drop
the drop weight. At these locations the weight can weight must be 7,560 tons. If it is a 4 meter square block
consist of a large fabric bag filled with water from of steel (about the cross section of a railroad car), it will
the river at the bottom. The bag can be emptied weigh 124.8 tons per meter, and hence must be 60.6 me-
before hitting bottom. This reduces the weight that ters long. The rails must be 12,000 m/20 long, or 600
98 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

meters long. g’s. Aircraft top speed on deck = 300 m/s. Object
top speed in theory would be 3000 m/s. In practice
• 9b Locomotive Driver - A set of railroad locomo- would be limited by aerodynamics and cable heating
tives provides the motive force, which is multiplied (perhaps to 1500 m/s? limit is not well understood)
by a gear mechanism to a higher speed. Example:
launching a 20,000 lb References:

vehicle at 3 g’s to 1100 m/s. This requires a 20 km straight

run of track. The rail cars needed would include: 10 Rotary Sling Alternate Names: Centrifuge Cata-
pult
1 tank car for vehicle fuel Type:
1 special purpose car to carry the glider Description: In principle, this is a sling or bolo scaled up
2-3 cars with tow rope guides to keep the tow and using aerospace materials. Rotating cables can reach
rope off the ground. a significant fraction of orbit velocity on Earth, or all of
1 pulley system car orbit velocity on smaller bodies. A drive arm is driven
in rotation by methods such as an electric motor, or a jet
30 locomotives in tandem. mounted on its tip. In larger versions a cable with the
payload attached to the end is played out gradually as the
We assume the locomotive top speed is about 27 m/s, system comes up to speed. The drive arm leads the cable
therefore a 40:1 gear ratio will provide the desired speed slightly so the cable and payload see a torque that contin-
at the vehicle. Locomotive traction averages 80,000 ues to accelerate them. When the desired payload veloc-
lb/engine, or 2000 lb per engine when reduced through ity is reached, the payload releases and flies off. The cable
the gear ratio. The gear-down mechanism and launch ca- is then retracted and the drive arm slows down. When it
ble drum are mounted on a flatbed rail car. This car can stops, another payload is attached. In smaller versions the
be anchored to a foundation on either side of the rail- cargo can be connected directly to the drive arm without
road track to hold it in place when the combined pull of a cable.
the locomotives is exerted. The starting traction of 30
In a vacuum, such as on the Lunar surface, this is theoret-
locomotives is 1800 tons. Since the couplings between
ically a very efficient system, as the sling can be driven by
engines are probably not designed for this load, a set of
an electric motor and the mechanical losses can be held
steel cables on both sides of the locomotives are used to
to a low value. Some method of recovering the energy
transmit the traction force from each engine to the gear
of the arm and cable (such as by transferring it to a sec-
mechanism. The vehicle is attached to the anchored rail
ond system by using the motor as a generator), can lead
car by a high-strength cable which is 20 km long. At 3 g’s
to efficiencies over 60% in theory.
it takes this distance to accelerate to the desired speed.
On bodies with an atmosphere, such as the Earth, the sys-
Two or three rail cars are spaced out along the 20 km
tem is hindered by air drag. One method of reducing drag
with towers with a pulley wheel on top, to guide the cable
is to attach an aerodynamic shape to the cable material, so
and keep it off the ground during the initial acceleration.
as to lower drag compared to a circular cable. Another is
The vehicle has glider type wings attached that will gen-
to mount the drive arm on the top of a large tower, so the
erate lift as it gains speed, so the vehicle will climb once
cable is not moving in dense gas. A third is to generate lift
it reaches 100 m/s or so. When the vehicle reaches the
along the cable or at the payload raising its altitude. The
desired speed, the cable is released and the vehicle contin-
rapidly moving part of the cable near the payload then has
ues to climb under the glider’s lift. Eventually the glider
less drag.
drops the vehicle, which proceeds under rocket power.
Although 30 locomotives is a lot, you only have to leaseA counterweight is desirable in situations where the un-
them long enough to do a launch, after which they can balanced load of the cargo would put too much stress on
return to normal railroad use. the pivot and supporting structure. The counterweight
gets released at the same time as the cargo. Due to gyro-
A small prototype would consist of a single Locomotive
scopic effects, the rotating cable will attempt to keep it’s
driver. 1250 lb rocket @ 4 g peak. Final velocity = 700
axis of rotation fixed. When attached to a rotating body
m/s. Accel time = 17.5 sec distance = 0.5at2 = 6.1 km.
such as the Earth, the location and mounting of the cable
Engine traction = 80,000 lb average @ 25:1 gear ratio.
will have to take that into account.

• 9c Jet Driver - This is similar to the locomotive Design Example:

case, but the gear ratio is lower since the jet can This example is for a terrestrial sling with a tip velocity
reach a higher speed on a take-off run. Example: an of 1800 m/s, or about 24% of Earth orbital velocity, and
F-15 can tow 40,000 lb rope tension if near empty. a cargo of 10,000 kg including propulsion for going to
@10:1 gear ratio can accelerate 1000 lb object @ 4 orbit:
3.3. GUNS AND ACCELERATORS 99

- Assume tip acceleration is 10 gravities (100 B. Artillery

m/s2 ) for non-living cargo.
- Then r = 32,400 meters. Guns have been in use in the West since about 1300 AD.
A “space gun” is merely one large enough, and with a
- For a 10,000 kg projectile at end, cable ten- high enough muzzle velocity, to be useful for space trans-
sion is 1 MN. If carbon fiber is used with port purposes. It also needs to point sufficiently above
a 3400 MPa design stress, then cable area is the horizon for the projectile not to hit the terrain, and
1/3400 m2, or about 3 cm2 . lower the amount of air to travel through. The earliest
- Cable mass is 0.6 kg/meter, adds 60 N/meter illustration of using a cannon to reach orbit is from Isaac
at tip. The implied scale length is 16.67 km. Newton’s Principia Mathematica, where he was illustrat-
Acceleration falls linearly with radius, so ef- ing the concepts of gravity and orbits. Thus the idea
fective radius is 16.2 km for load purposes, and is as old as orbit mechanics. In the 20th century rock-
cable mass ratio is 2.8:1. Therefore cable mass ets were developed for ballistic missiles and then placing
is 18,000 kg. payloads into orbit. They were able to reach higher veloc-
- To spin up in 1 hour, we need an average ac- ities with lower maximum acceleration and infrastructure
celerating force of 0.5 m/s2 , or 5000 N. at the time. Even though guns have a much longer her-
itage, and both artillery and rockets are both still used by
- Drag counteracts the accelerating force. As- military forces, they were bypassed in the early part of
sume the cable is shaped to be 1 cm tall by 3 the Space Age.
cm deep. Since drag is proportional to velocity
squared, and velocity goes linear with radius,
drag over the entire length will be 1/3 that at 11 Solid Propellant Charge Alternate Names:
the tip. At the tip drag is 0.5 Cd rho A v2 . Cd Type:
= 0.04 for shaped airfoil. rho = 1.225 kg/m3
at sea level. A = 0.01 m2. v = 1800 m/s. Then Description: This method uses an explosive charge
tip drag = 2381 N/m, or 1/3 x 32,400 m x 2381 which vaporizes behind a projectile in a barrel. The high
N/m = 25.72 MN at sea level. gas pressure generated this way accelerates the projectile
to high velocity. Conventional artillery reaches speeds
- If we assume our drive motor can produce a of around 1000 m/s. Velocities much higher than this
peak of 4 times the average accelerating force, are difficult to reach because the barrel mass becomes
then peak drag can be at most 15,000 N. The extreme at higher pressures, and the temperature and
motor is then driven to maintain a surplus of molecular weight of the resulting gas limits the internal
5000 N force above drag. Since sea level drag speed of sound, which in turn limits muzzle velocity.
is 1715 too high, we want to go to an altitude
where density is that factor lower, or 7.14 x Status: Artillery has a long history and extensive use.
10−4 kg/m3 . This density is at 53 km alti- The High Altitude Research Probe project attached two
tude. The cable will drape in a curve due to the naval gun barrels in series and used relatively light shells
combination of radial and gravity forces, so the to reach higher muzzle velocities than conventional ar-
tower height will need to be approx 60 km. tillery.
Variations:
Status: Slings and bolos are ancient devices. So far as References:
is known, a modern one for space transport has not been
built • Newton, Isaac, Principia Mathematica.
Variations:
• Verne, Jules, From the Earth to the Moon.

• 10a Jet Driven Sling - A jet engine is

mounted to the rotating arm in order to 12 Liquid Propellant Charge Alternate Names:
get high starting torque. The launch ve- Type:
locity is increased by the ratio of cargo
radius to jet radius from the pivot point. Description: This is similar to conventional solid pro-
pellant artillery except liquid propellants are metered into
• 10b Two Stage Sling - A second rotating the chamber, then ignited. Liquid propellants have been
cable or arm is mounted on the first one studied because they produce lighter molecular weight
to reduce mass ratio of the entire system. combustion products, which leads to higher muzzle veloc-
This adds complexity, so is only an ad- ities, and because bulk liquids can be stored more com-
vantage where the mass ratio of a single pactly than shells, and require less handling equipment to
stage version becomes too extreme. load. Metering the propellant flow into the barrel low-
ers the peak pressure relative to a detonation, leading to
References: lighter barrels.
100 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

Status: • P. Kaloupis and A.P. Bruckner, “The Ram Ac-

Variations: celerator: A Chemically Driven Mass Launcher”
, AIAA Paper 88-2968, AIAA/ASME/SAE/ASEE
References: 24th Joint Propulsion Conference, July 11-13,
1988, Boston, MA. (Applications to surface-to-orbit
launching)
13 Gaseous Charge Alternate Names:
Type: • Breck W. Henderson, “Ram Accelerator Demon-
strates Potential for Hypervelocity Research, Light
Description: A mixture of fuel and oxidizer is intro-
Launch,” , Aviation Week & Space Technology,
duced into a chamber, and then ignited. This is similar
September 30, 1991, pp. 50-51.
to how automobile engines function. This method covers
projectiles directly driven by the resulting high temper- • J.W. Humphreys and T.H. Sobota, “Beyond Rock-
ature gas. Indirect drive guns are covered under Light ets: the Scramaccelerator” , Aerospace America ,
Gas Guns on the next page. The igniter is a separate de- Vol. 29, June, 1991, pp. 18-21.
vice in the basic concept. An alternate version called a
Scramjet Gun uses the projectile itself as a traveling ig-
niter. It must be moving supersonic to keep the ignition 14 Rocket Fed Gun Alternate Names:'
front from getting ahead of it, so it needs a gas injector at Type:
the start to get it moving fast enough.
Description: A rocket engine is mounted at the chamber
Status: Research prototypes. Methane/air mix was used end of a gun to produce hot gas to accelerate projectile.
as the driver for the Livermore 2 stage gas gun. The com- In a conventional gun, all the gas is formed at once as the
bustion drives a 1 ton piston, which in turn compresses charge goes off. In this concept the gas is produced by
hydrogen working gas. Research on the scramjet gun was a rocket type engine and fills the barrel with gas as the
being performed at the University of Washington under projectile runs down it. Compared to a conventional gun,
Prof. Adam Bruckner. Research gun was located in the the peak pressure is lower, so the barrel is lighter.
basement of a building there.
Status:
Variations:
Variations:
• 13a Scramjet Gun (Ram Accelerator) - References:
Fuel/oxidizer mixture present in barrel is burned
as projectile travels up barrel. If projectile shape
NOTE: Guns and Accelerators is contin-
resembles two cones base to base, as in an inside-
ued on the next page.
out scramjet, the gas is compressed between the
projectile body and barrel wall. The combustion
occurs behind the point of peak compression,
and produces more pressure on the aft body than 3.4 Guns and Accelerators 2
the compression on the fore-body. This pressure
difference provides a net force accelerating the Note: Continued from sections A and B at Guns and
projectile. Accelerators

One attraction of this concept is that a high acceleration

launch can occur without the need for the projectile to use 3.4.1 C. Light Gas Guns
on-board propellants. If the projectile has a inlet/nozzle
shape (hollow in the middle) it might continue accelerat-Light gas guns are designed to reach higher muzzle ve-
ing in the atmosphere by injecting fuel into the air-onlylocities than combustion guns. These types of guns are
incoming flow, extending the performance beyond what also called hypervelocity guns since the projectile trav-
a gun alone can do. Another attraction of this concept els at more than Mach 5 (1500 m/s at sea level). In aero-
is the simplicity of the launcher, which is a simple tubedynamics, velocities above Mach 5 are are called hyper-
capable of withstanding the internal pressure generated sonic. They reach higher velocities by using hot hydrogen
during combustion. (or sometimes helium) as the working gas. These have a
References: lower molecular weight, and therefore a higher speed of
sound than the combustion products of smokeless pow-
• A. Hertzberg, A.P. Bruckner, and D.W. Bogdanoff, der, liquid, or gaseous fuels.
“The Ram Accelerator: A New Chemical Method Guns are strongly limited by the speed of sound of the
of Accelerating Projectiles to Ultrahigh Velocities” gas they use, since pressure waves travel at the speed of
, AIAA Journal, Vol. 26, No. 2, February, 1988. sound. Therefore any gas at the back of the barrel no
(The original scramjet gun paper) longer contributes to pushing a projectile once it reaches
3.4. GUNS AND ACCELERATORS 2 101

the local speed of sound, because the pressure waves can- maximum velocity than a gas gun, so the choice will de-
not keep up. The gases near the projectile and moving at pend on construction vs operations costs of real designs.
the same speed can still provide pressure, though. So the Electromagnetic guns have a shorter history, so their ca-
efficiency of the gun falls significantly, but does not stop pability and cost at large scale are more uncertain.
entirely as the projectile goes supersonic. Light gases like Gun/Rocket Optimization - When launching from
Hydrogen do not generate high pressures and tempera- Earth, the projectile encounters the atmosphere at high
tures by themselves, as do combustion byproducts in stan- velocity, and drag tends to scale as velocity squared. Even
dard guns. Therefore some external means are required with light gas guns, it is difficult to reach full orbit veloc-
to produce the hot gas. The various methods are listed in
ity (~8 km/s), and even at that velocity the orbit will in-
the following sections. tersect the ground, because that is where it started from.
Scaling - Light gas guns have been used for hypersonic Thus any gun system needs at least some propulsion to
research for about 40 years. Examples are testing heat raise the low part of the orbit. When optimized for cost
shield materials, and meteoroid impact damage. These and efficiency, the gun velocity generally ends up being
do not require large projectiles, so a light gas gun large roughly half of orbit velocity, with the balance provided
enough for practical use in space transport has not been by a rocket on the projectile or some other method. A
built yet. There is no reason to think larger guns will not rocket that only has to provide a fraction of orbit velocity
work. Higher muzzle velocities than needed for space will be much smaller and less expensive, and repetitive
launch have already been demonstrated in research guns. smaller launches lower the size even more. For example,
It is more a matter of designing them for low cost opera- daily gun launches compared to about six conventional
tion rather than research, and simple scaling of the parts. rocket launches per year gives a factor of 60, and approx-
Large guns will in fact be more efficient than research imately 4 times lower rocket to payload mass ratio gives
guns: a total reduction of 240 in rocket size. The main cost ad-
vantage of hypervelocity guns thus comes from replacing
• The region close to the barrel wall is called the the large and expensive conventional rocket with a much
boundary layer. This is where heat is lost to the smaller one plus a reusable and relatively cheap gun.
cooler barrel, and friction is created between the Location - Due to their size, light gas guns will likely
moving gas and stationary wall. The boundary layer be built on a mountain with the correct slope, pointing
stays the same thickness regardless of gun size, so east. This minimizes construction cost and reduces air
represents a smaller fraction in larger guns. drag from the higher starting altitude. An ocean plat-
form for the gun could be aimed, but has the drawbacks
• Atmospheric drag is proportional to the area of the of more atmosphere to fly through at sea level than from
projectile, while it’s mass is proportional to area x a mountain top, and dealing with salt water corrosion.
length. The mass thus grows faster than drag, and When choosing a mountain, equatorial ones allow meet-
deceleration due to drag goes down with size. So ing a space station every orbit (approx 90 minutes), while
drag induced velocity loss becomes a smaller factor other latitudes only allow meeting once per day, when the
for larger guns. launch site rotates under the orbit plane.
Evacuated Barrel - All large guns will operate more ef-
Larger guns also tend to have lower acceleration of the ficiently if the projectile does not need to push a column
projectile. The muzzle velocity of this type of gun of air up the barrel. So some method of sealing the end
is mostly set by the speed of sound of Hydrogen, and and pumping out most of the air is generally desired. In
the practicality of flying through the atmosphere at high smaller guns something as simple as a plastic film has
speeds. So the muzzle velocity will not be greatly differ- been used, and the projectile just punches through it. In
ent in larger guns. The barrel length goes up with gun size, larger guns a flap or flaps held in place by the pressure
so the acceleration is lower to reach the same final veloc- difference of the barrel vs outside air can be forced open
ity. That means the projectile structure can be lighter, as the residual air in the barrel gets compressed by the
and a wider range of cargo can be carried. In the limit of projectile. This sort of automatic opening is preferred to
very large guns, it would be possible to carry humans, but some mechanical valve or door. A failure of the mechan-
such large guns will not be built for some time, and might ical device would lead to a spectacular collision with the
be overtaken by better alternatives. projectile.
Comparison to other Launchers - Compared to rock- Muzzle Design - It is worth considering a silencer type
ets, weight is not an issue for a terrestrial mounted gun. device at the muzzle for several reasons. In addition to
Thus their parts can be closer to industrial grade than sound reduction, it can capture the Hydrogen gas to be
aerospace grade, and should be relatively inexpensive. recycled. Otherwise escaping hot Hydrogen will imme-
Gun barrels are also simple compared to electromagnetic diately burn in air, producing a muzzle flash. When the
coil type accelerators providing the same force. Thus projectile goes from high acceleration in the barrel to de-
they should have a lower construction cost. Electro- celeration in air, if the transition is too sudden that might
magnetic devices have a higher theoretical efficiency and
102 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

cause damage to the projectile or cargo. A silencer can ity for Spacecraft Materials Evaluation”, SAMPE
bleed off some of the Hydrogen, giving a smoother transi- Quarterly , v 18 no 2 pp 28-34, 1987.
tion. The residual air in the barrel will also provide some
degree of shock buffer as it piles up ahead of the projec-
tile. Other ways to lower the shock are to put gas nozzles 16 Underwater Storage Gas Gun Alternate Names:
around the muzzle to add a stream of gas as a transition Quicklaunch
region past the muzzle, or to shape the barrel as a flaring Type:
cone at the end to taper off the acceleration. Finally, the
Description: In a gas gun on land the amount of struc-
on-board rocket on the projectile can be used to counter
tural material in the gun is governed by the tensile
the drag deceleration.
strength of the barrel and chamber. In an underwater gun,
Projectile Features - Filling the area behind the projec- an evacuated barrel is under compression by water pres-
tile with gas, known as base bleed, has been used pre- sure. The gas pressure in the gun can now be the external
viously to reduce drag and extend the range of artillery. water pressure plus the pressure the barrel wall can with-
The velocity of light gas gun projectiles is high enough stand in tension, which is up to twice as high as the land
that shaping the back end to reduce drag may not be prac- version.
tical, and so base bleed is an option. This can be com-
Other features of an underwater gun are the ability to
bined with the on-board rocket engine by running it at
store gas with very little pressure containment (the storage
low thrust merely to fill the trailing area, or at full thrust to
tank can be in equilibrium with the surrounding water),
provide acceleration even while clearing the atmosphere.
and the ability to point the gun in different directions and
The right answer will require detailed design and analysis.
elevations.
Projectile Recovery - To lower cost, you would like to
The underwater gas gun consists of a gas storage chamber
use your cargo projectiles multiple times. Fortunately,
at some depth in a fluid, in this case the ocean, a long
making them rugged enough to fire out of a gun tends to
barrel connected to a chamber at one end and held at the
also make them rugged enough to survive re-entry without
surface by a floating platform at the other end, plus some
a lot of extra work. They already need some heat shield-
supporting equipment.
ing to survive going up through the atmosphere at high
velocity when launched. You just need some more shield- Example Design In one version of this type of gun, the
ing to also handle re-entry. The same guidance system chamber is a made of structural material such as steel. An
that enables them to reach their cargo destination also can inlet pipe allows filling of the chamber with a compressed
guide it to a landing point. The terminal velocity of the gas. A valve is mounted on the inlet pipe. An outlet pipe
empty projectile after re-entry should be low enough that of larger diameter than the inlet pipe connects to the gun
the projectile can withstand landing without any landing barrel. An outlet valve is mounted on the outlet pipe. This
system. It was fired fully loaded at high gees out of the valve may be divided into two parts: a fast opening and
gun, after all. At most it might need deployable flaps or closing part, and a tight sealing part. The interior of the
fins or a small parachute to get the landing velocity low chamber is lined with insulation. The inner surface of the
enough. insulation is covered by a refractory liner, such as tung-
sten. An electrical lead is connected to a heating element
inside the chamber.
15 Pressure Tank Storage Gas Gun Alternate
Names: An inert gas such as argon fills the insulation. The inert
gas protects the chamber structure from exposure to hot
Type: hydrogen, and has a lower thermal conductivity. An inert
Description: The gas is simply stored in a pressure ves- gas fill/drain line is connected to the volume between the
sel, then adiabatically expanded in a barrel, doing work chamber wall and the liner. A pressure actuated relief
against a projectile. Heating the gas raises issues of stor- valve connects the chamber with a volume of cold gas.
age at high temperature, but not heating the gas leads to This cold gas is surrounded by a flexible membrane such
low performance. The storage tank needs to be large rel- as rubber coated fiberglass cloth.
ative to the barrel volume, otherwise the pressure drop
In operation, the gas inside the chamber, the inert gas,
during firing leads to lower performance. This leads to
and the water outside the chamber are all at substantially
higher cost than other versions. These drawbacks lead to
the same pressure. Thus the outer structural wall does not
other versions of a light gas gun to be preferred over this
have to withstand large pressure differences from inside to
simple one. outside. One part of the chamber wall is movable, as in a
Status: sliding piston, to allow variation in the chamber volume.
The gas in the chamber is preferably hot, so as to provide
Variations:
the highest muzzle velocity for the gun. When the gun is
References: operated, this gas is released into the gun barrel. In order
to preserve the small pressure difference across the wall
• Taylor, R. A. “A Space Debris Simulation Facil- of the chamber, either the chamber volume must decrease
3.4. GUNS AND ACCELERATORS 2 103

or gas from an adjacent cold gas bladder must replace the from the periphery to the center of the heat exchanger and
hot gas as it is expelled. This arrangement prevents ocean then to the barrel.
water from contacting the chamber walls or hot gas. In the
case of the sliding piston, the membrane collapses, with • Example of Gun Component Sizing: - The fol-
the gas formerly within it moving in behind the piston. lowing calculations are intended to show the steps
In the alternate case, the membrane also collapses, with for getting the initial estimates of component sizes.
the gas formerly within it moving through a valve into the It is very far from being a complete engineering de-
chamber. sign. Rather, it is the starting point for the design
The chamber has an exit valve which leads to the gun bar- process.
rel. It also has gas supply lines feeding the interior of
the chamber and the volume between the chamber walls. - Set as a design goal to match the 1990’s
These lines are connected to regulators which maintain Livermore SHARP hypersonic gun (5 kg at 3
nearly equal gas pressures, which in turn are nearly equal km/s). From kinetic energy formula (KE =
to the ocean pressure. This allows the chamber to be 0.5*M*v^2) we get 22.5 MJ projectile energy.
moved to the surface for maintenance, and to be placed - If average barrel pressure is 20.7 MPa (3000
at different depths for providing different firing pressures psi), chosen as a reasonable value for high pres-
or different gun elevations. sure pipe, then a 10 cm diam, 5 kg projec-
tile (matching SHARP parameters) will see
The muzzle of the gun is at the ocean surface, so elevation
162,600 N force, or 32,470 m/s2. To reach 3
of the gun can be achieved by changing the depth of the
km/s requires 92.4 msec, and a barrel length of
chamber end. Since the gun as a whole is floating in the
139 m. Allowing for pressure drop during op-
ocean, it can be pointed in any direction. Some means for
eration, friction, efficiency, thermal, and other
heating the gas stored in the chamber is needed, such as an
losses, assume actual barrel will be 200 m long.
electric resistance heater. At the muzzle end of the gun,
That assumption would be later replaced by a
a tube surrounds the barrel, with a substantial volume in
simulation to get a better estimate.
between the two. There are passages through the wall of
the barrel that allow the gas to diffuse into the tube rather - From simple cylinder volume formula (V = pi
than out the end of the gun, thus conserving the gas. * r^2 * h ) the volume of the barrel will be 1.57
cubic meters. From the physical properties of
At the muzzle of the gun is a valve which can rapidly open, Hydrogen as a gas, that volume would contain 4
and an ejector pump which prevents air from entering the
kg of Hydrogen at 20 MPa and 2000K temper-
barrel. In operation, the ejector pump starts before the ature. At the point the projectile leaves the bar-
gun is fired, with the valve shut. The valve is opened, then
rel, the barrel is full of gas. At standard condi-
the gun is fired. In this way, the projectile encounters only tions (273K or 0C), the speed of sound for hy-
near vacuum within the barrel, followed by air.
drogen is 1284 m/s. That goes up as the square
Status: Hypersonic guns have been in operation for root of the Kelvin temperature, so at 2000K it
about 40 years as research devices, with relatively small will be 3475 m/s, and so for this gun it will be
(5 kg or less) projectiles. Large or ocean-going guns have slightly subsonic.
not been built. - The specific heat of Hydrogen from 300 to
Variations: 2000K is about 15 kJ/kg-K. Therefore by mul-
tiplying the known temperature rise (1700K),
References:
mass of H2 (4 kg) and specific heat, we need
102 MJ of energy for heating. To this we add
the 22.5 MJ of projectile kinetic energy, which
17 Particle Bed Heated Gas Gun Alternate Names: comes at the expense of the Hydrogen heat and
Type: pressure, for a total of 124.5 MJ.
Description: Hot hydrogen gives good performance for - To heat the gas during the launch, we don't
a gun, but is hard to store. In this method it is generated want the aluminum oxide particles to drop
when needed by flowing the Hydrogen through a chamber more than 500K in operation. They otherwise
which contains refractory oxide particles. The particles would be cooling down too much to heat the
are heated slowly (roughly 1 hour time period) by some last part of the gas. At 1300 J/kg-K, need then
type of heater near the center of the chamber. This sets need about 200 kg of aluminum oxide parti-
up a temperature gradient, so the exterior of the chamber cles. That is a rough value which would be
is relatively cool, and can thus be made of ordinary steels. refined by a thermal analysis of the heat ex-
When the hydrogen flows through the chamber, the large changer.
surface area of the particles allows very high heat transfer - The density of Aluminum Oxide is 3.9 g/cc as
rates - so the heat in the chamber can be extracted in a a solid. As grains in a heat exchanger with al-
fraction of a second as the gun fires. The gas flows inward lowance for inlet and outlet volume and screens
104 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

to keep the particles in place, assume the den- The particles are retained in the reactor by spinning the
sity is 1 g/cc. Therefore heat exchanger will be bed at high velocity, and the gas flow cools the external
about 0.2 cubic meters in size. structure. The particles of nuclear fuel do not need struc-
- The storage tank for the unheated hydrogen tural strength, so can go to higher temperatures than a
can be found from the physical properties at solid core reactor, leading to higher performance. The
room temperature. The initial pressure will improvement, however, is not great over refractory metal
have to be higher than 20 MPa for the gas to heat exchangers, so the cost and political issues for a nu-
flow towards the barrel. Assume it starts at 1.5 clear device probably outweighs the benefits.
times the average barrel pressure and finishes at Status:
0.5 times. Thus 2/3 of the gas goes into the bar-
Variations:
rel, and the initial gas to be stored is then 6 kg
at 31 MPa. Under those conditions the density References:
of Hydrogen is 25.8 kg/m^3, so the tank vol-
ume is 0.23 cubic meters. Very quickly releas-
19 Electrically Heated Gun Alternate Names:
ing the gas from the storage tank turns it into a
rocket engine, so a sturdy support is needed to Type:
keep the tank in place. Description: The working fluid is heated by electric dis-
- A fast (compared to the < 0.1 sec firing time charge or microwave induction, then pushes against the
of the gun) valve will be needed to open the projectile in the barrel. The limiting factor for a light gas
tank and let the gas flow through the heat ex- gun is the speed of sound in the fluid. One way to heat
changer and barrel. Conceptually this can be the fluid to much higher temperatures is an electric dis-
like a car engine cylinder, where a spark plug charge. If a high enough temperature is reached, the fluid
sets off a detonation, which slides the valve pis- becomes a plasma. Either magnetic fields or a sheath of
ton from the closed to open position. When do- cooler gas may be used to keep the plasma away from
ing the design, you would first review existing the walls and prevent damage. The challenge for this
valve hardware to see if a suitable one exists. If method is to deliver hundreds of MegaJoules for even
not, then a custom fast-acting valve would need a very small gun (by space launch standards) in a short
to be designed. period of time. Magnetic fields are not strong enough to
contain the plasma at these pressures, so it will lose energy
Status: A small research gun of this type has been built rapidly to physical chamber walls. Therefore the heat-
at Brookhaven Natl. Lab. ing must be rapid, which in turn requires a large power
source. For versions large enough to deliver payloads to
Variations: orbit the power source needs to deliver 2 GigaJoules or
References: more, and the power source ends up dominating the total
gun cost.
Status:
18 Particle Bed Reactor Heated Gas Gun Alternate
Names: Variations:
Type: References:
Description: Hot gas is generated by flowing through a
particle bed type nuclear reactor. Gas expands against 20 Nuclear Charge Heated Gun Alternate Names:
projectile, accelerating it. Light gas guns have been op- Nuclear Cannon
erated to above orbital velocity, and 1 kg projectiles have
been accelerated to over half orbital velocity. This type Type:
of gun rapidly becomes less efficient above the speed of Description: As in the other light gas guns, the objective
sound of the gas. As a consequence the working fluid is is to use hot gas to accelerate a projectile. In this version
usually hot hydrogen. Conventional gas guns have used the heating is from a small atomic bomb in a large under-
powder charge driven pistons to compress and heat the ground chamber filled with the working gas. This concept
gas. This is not expected to be practical on the scale only makes sense in a situation where very large payloads
needed to launch useful payloads to orbit. One way to need to be launched, due to the large minimum energy
heat the gas is to pass it through a small particle bed nu- to make any nuclear device function. A large barrel leads
clear reactor. This type of reactor produces a great deal of off the chamber upward at an angle. A crossbar is set into
heat in a small volume, since the small particles of nuclear the barrel near the chamber, and the projectile is attached
fuel have a large surface/volume ratio and can efficiently to the crossbar with a bolt that is designed to fail at a pre-
transfer the heat to working fluid. This gives the benefits determined stress. This restrains the projectile until the
of nuclear power for space launch, without the drawbacks operating pressure is reached. A small atomic bomb is
of a flying reactor. suspended in the chamber and detonated to create lots of
3.4. GUNS AND ACCELERATORS 2 105

very hot hydrogen in a very short time. gas is the best working fluid, and the question is how to
There are several obvious issues with this concept. Even a generate it. In the previous method a piston is driven by
small atomic bomb delivers too much energy in too short combustion, much like an internal combustion automo-
a time to easily manage. The challenge is to keep from de- bile engine. Here a good part of the high pressure cham-
stroying the gun itself. Another issue is radiation, which ber is dispensed with by using gravity to drive a massive
can harm the cargo, and get ejected out the barrel as fall- weight. If the gun is built on the side of a mountain, as is
out. Finally are the political restrictions on using atomic usually desired, the energy for launch is stored as poten-
bombs of any purpose. tial energy in the weight. The falling weight rides on an air
or lubricated bearing and slides down the mountain to the
Status: chamber. The piston seals the back end of the chamber
Variations: and is held in place. The falling mass is designed smaller
than the piston, so it does not need an accurate fit at the
References: high velocity it impacts. The chamber then leads to a bar-
rel containing the projectile, which accelerates upward.
21 Combustion Driven Two Stage Gun Alternate One advantage of this method is the falling mass is not
Names: stopped instantly, but rather can continue compressing
Type: the gas gradually as the gun fires. The system of falling
mass + working gas + projectile can be designed as a
Description: This is a type of two-stage gas gun. A tuned coupled spring system to maximize energy transfer
cylindrical chamber contains a piston. On the back side to the projectile. Another is the relatively low cost, since
of the piston high pressure gas is generated by combus- the falling mass operates in open air, and can be made
tion. This can be gunpowder or a fuel-air mixture. On the from low strength materials such as concrete. A third
front side of the piston is the working gas, which is usually is the mass can be raised as slowly as needed, so large
Hydrogen. The Hydrogen is compressed and heated until power supplies are not needed. Lastly this method can be
a valve or seal is opened. Then the working gas acceler- added as a “supercharger” to a particle bed type heater to
ates the projectile. The compression can result in theoret- increase performance. The gas is first heated to the prac-
ically unlimited temperature and pressure. In practice it is tical limit of flowing through a heat exchanger bed, and
limited by the pressure capacity of the chamber structure, then the weight driven piston provides additional com-
but this can be made very high. Therefore muzzle veloc- pression heating. The question then becomes whether the
ities have reached above Earth orbit velocity (8 km/s) for additional performance is needed and justifies the added
small projectiles. The drawback is the chamber must be complexity of a dual heating and compression system.
several times the barrel volume to contain both the com-
bustion driver gas and Hydrogen, which increases the cost Status: As of yet this is only a concept.
for large versions. Variations:
Status: This type of light gas gun is the most common References:
that has been built. They were first constructed in the
1960’s or earlier. The largest gun of this type was the
Lawrence Livermore Laboratory SHARP gun, which was 3.4.2 D. Electric Accelerators
used to test scramjet components in the early 1990’s. It
had a 10 cm x 45 m barrel and a 30 cm x 100 m long This group uses magnetic fields to provide a force be-
chamber. tween the accelerator and a projectile. Electric acceler-
Variations: ators of moderate length require high peak power for a
short period of time. Hence inexpensive energy storage
References:
is very important for these concepts. Two places to look
• Light Gas Gun (Wikipedia article) for inexpensive energy storage are (1) Magnetic fusion
experiments, which also need short duration high power
• Aviation Week & Space Technology, July 23, 1990. supplies, and (2) Inductive energy stores. The latter falls
• “World’s Largest Light Gas Gun Nears Completion into subcategories of cooled normal conductors, and su-
at Livermore.” Aviation Week & Space Technology, perconductors. Alternately the peak power can be low-
August 10,1992. ered by making the device very long, but that then be-
comes a construction issue to build a device many kilo-
meters in size.
22 Gravity Driven Piston Alternate Names:
Type:
23 Railgun Alternate Names: Electromagentic Gun
Description: In this method a sliding or falling mass is
used to compress gas in a chamber. The gas is then ex- Type:
panded in a barrel. As in all light gas guns, hot Hydrogen Description: In this method, very high electric current
106 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

supplied by massive rails is shorted through a plasma arc 24a Quenchgun - The energy is stored in su-
induced at the back end of the projectile. The plasma is perconducting coils making up the gun. The
accelerated by the Lorentz force interaction with the mag- circulating current is quenched ahead of the
netic field produced by current. The plasma then pushes projectile either by heating the coil above the
projectile along the rails. Given a sufficiently large power transition temperature of the superconductor,
supply, it can be considered for Earth launch systems at or by raising the field using a coil on the pro-
lower accelerations than those proposed for weapon sys- jectile. In either case the current stops flowing
tems. in the gun coil. Since the coils behind the pro-
jectile are off, while those ahead are still on,
Status: This device was under intensive development for
the Strategic Defense Initiative. A large gun was built at the net force will accelerate the projectile.
Eglin AFB in Florida and used a bank of thousands of car
References:
batteries wired in parallel as a power supply. Prototype
railguns achieved high velocities, but the high currents
• Nagatomo, Makoto; Kyotani, Yoshihiro “Feasi-
and plasma produced rail erosion.
bility Study on Linear-Motor-Assisted Take-Off
Variations: (LMATO) Of Winged Launch Vehicle”, Acta As-
References: tronautica, v 15 no 11 pp 851-857, 1987.
• Kolm, H.; Mongeau, P. “Alternative Launching
• Robinson, C. A. “Defense Department Developing Medium”, IEEE Spectrum, v 19 no 4 pp 30-36,
Orbital Guns”, Aviation Week and Space Technol- 1982.
ogy, v 121 no 12 pp 69-70, 1984.
• Kolm, H. “An Electromagnetic 'Slingshot' for Space
Propulsion”, Spaceworld pp 9-14, Feb. 1978.
• Bauer, D. P. et al “Application of Electromagnetic
Accelerators to Space Propulsion” IEEE Trans.
Magnetics vol MAG-18 no 1 pp 170-5, Jan. 1982.
3.5 Combustion Engines
24 Coilgun Alternate Names: Mass Driver Launcher 3.5.1 A. Air-Breathing Engines
Type:
The next group of methods normally involve using a
Description: In this method a series of coils forming the planet’s (usually the Earth’s) atmosphere as a supply of
gun react with coil(s) on projectile magnetically, produc- oxygen to support combustion with a fuel carried on the
ing linear thrust. Popularly known as a Mass Driver, vehicle. They differ in the details of how the incoming air
it functions on the same principles as an electric motor. flow and combustion is managed. It should be noted that
The concept was originally developed in connection with some vehicle concepts, such as the National Aerospace-
launching Lunar materials for space manufacturing. The plane (NASP) of the 1990’s, or the current British Skylon
low orbital velocity and vacuum on the Moon made it fea- would integrate more than one method in a single engine.
sible. Accelerator designs with high efficiency (>90%) This is referred to as a Combined Cycle Engine. The
and high muzzle velocities (>8 km/s) have been proposed same general engine concepts could be used in a reducing
for use on Earth. This potentially leads to a transportation atmosphere, such as Hydrogen or Methane, with Oxygen
system whose operating costs consist mostly of electric- as the carried fuel, or in a sufficiently powerful nuclear
ity. Because of the need for large amounts of power to engine with any atmosphere. In the latter case the nuclear
drive the device, the cost of that power supply tends to engine is used to drive a compressor or heat the incoming
dominate overall system cost, so an electromagnetic ac- gas flow.
celerator makes sense for very high volume launch.
Fundamental to understanding the operation of air-
Like other launchers that generate high velocity near breathing engines is the concept of mass conservation.
the Earth’s surface, some method of dealing with atmo- The mass of incoming air at the Inlet does not change in
spheric drag and heating is needed. Options include heat total amount going through the engine, although pressure
shielding on the projectile, a vacuum tunnel, or installing and temperature vary. The exception is the point where
the launcher on high enough towers to avoid the atmo- fuel is added, or in some types, where condensed air flow
sphere. is removed. Heat can be added or removed by combus-
Status: Linear electric motors and magnetic levitation tion or heat exchangers, but otherwise the mass flow stays
trains operate on the same principle as a coilgun. A coil- the same. From that constant mass flow, changes in the
gun is just a higher performance version. Prototype coil- other conditions of the gas flow can be calculated, and
guns were built around 1980 and reached 1800 gravities performance derived.
acceleration, which is more than sufficient. In practice, the flow through an engine can involve many
Variations: fan blades and turbulence, so that computer simulations
3.5. COMBUSTION ENGINES 107

are not fully accurate. Real engines also operate at ex- increases in payload.
tremes of temperature and stress. Therefore engine de- Status: In common use on aircraft for aircraft propul-
velopment typically includes extensive testing on static sion. For space launch, the B-52 bomber and L-1011 air-
test stands, in wind tunnels, and in flight attached to air- craft have been used to carry the Pegasus three stage solid
craft, to determine actual performance and durability. rocket to 35,000 ft altitude. The B-52 uses 8 fanjet type
engines for propulsion. The Stratolaunch system is un-
der development, using parts from two 747 aircraft with
25 Fanjet Alternate Names: 6 high bypass fanjet engines, and carrying a large rocket.
Numerous paper studies have been made of using aircraft
Type: Gas Flow by Fuel-Atmosphere Combustion
as carriers for rocket stages.
Description: The fanjet is the standard type of jet engine
Variations:
found on passenger aircraft and military aircraft. The
original form of the engine, the turbojet, has a series of
turbine compressor stages to compress the incoming air • Carrier Aircraft - A conventional jet
flow. This is followed by a combustor where fuel is added aircraft is used to carry a separate rocket
and burned, creating a hot gas. The gas is then expanded to an altitude of roughly 9000 meters
through a turbine which is connected by a shaft to the and velocity of 240 m/s, after which the
compressor. The expanded gas emerges at high velocity rocket ignites and finishes reaching orbit.
from the back of the engine.
The modern fanjet adds a fan which is also driven by • Booster Jets - A set of military fighter
the turbine. All of the airflow goes through the fan, but engines are attached to a rocket as sep-
only a part goes into the compressor. The air which does arate strap-on boosters, or a connected
not go into the compressor is said to have 'bypassed' the booster ring. They can loft a rocket to
compressor. The 'bypass ratio' is the ratio of bypass air about 15 km altitude and 480 m/s veloc-
to combustor air. Generally higher bypass ratio engines ity, after which they either parachute land
are more fuel efficient, in units of thrust divided by fuel or do a powered vertical landing.
consumption rate, because they increase the mass flow-
ing through the engine. In general, engines that operate
References:
at higher speeds are designed with lower bypass ratios.
This is due to more combustion needed to increase thrust
against the higher inlet drag from the higher velocity. 26 Turbo-Ramjet Alternate Names:
Typical modern performance values are engine thrust-to Type: Gas Flow by Fuel-Atmosphere Combustion
weight ratios (T/W) of 6:1 for large subsonic engines,
trending towards about 10:1 for high performance mil- Description: In this method a multi-stage fan com-
itary jets. Fuel efficiency is measured in units of thrust presses the incoming air stream, which is then mixed with
divided by mass flow rate. In English units this is pounds fuel, burned and exhausted. The compressor is driven by
divided by pounds per second, or just seconds, and is a gas generator/turbine. In a fanjet, the incoming air is
termed 'specific impulse'. In SI units this is Newtons per compressed and heated by the compressor stages, then
kilogram per second, which has the units of meters per mixed with fuel and run through the turbine stages. At
second. In some propulsion systems, such as chemical higher velocities the air gets hotter in compression since
rockets, the SI unit corresponds to the actual exhaust jet it has a higher incoming kinetic energy. This leads to a
velocity. In the case of air- breathing propulsion it is not, higher turbine temperature. Eventually a turbine temper-
the velocity result is just an indicator of engine efficiency. ature limit is reached based on the material used, which
In English units the performance of subsonic engines is sets a limit to the speed of the engine. In the turbo-ramjet
about 10,000 seconds, trending to about 7000 seconds for the compressor is driven by a gas generator/turbine set in-
supersonic military engines. Fanjets and turbojets oper- stead, which use on-board propellant for their operation.
ate up to about 3.5 times the speed of sound (M=3.5). Ex- Since the gas generator is independent of the flight speed,
tra thrust can be generated with an Afterburner, which it can operate over a wider range of Mach numbers than
burns more fuel after the turbine stage of the engine. This the fanjet ( to Mach 6 vs. to Mach 3).
comes at the cost of lower fuel efficiency. Status:
Although the maximum altitude and velocity of a fanjet Variations:
is limited compared to Earth orbit velocity, the effect on
payload can be much larger. This is because compared to References:
a conventional rocket, it can avoid most of the drag, pres-
sure, and gravity losses near the ground. Conventional
rocket payload is typically a small fraction of total mass, 27 Ramjet Alternate Names:
so small reductions in losses can produce large relative Type:
108 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

Description: In this method, the incoming air stream is Status: Scramjet engine components and small scale
decelerated to subsonic velocity relative to the engine via versions have been tested with mixed success.
a shaped inlet, mixed with fuel, then accelerated again out Variations:
an exit nozzle. Conceptually it is the simplest form of jet
engine, because it has no fans or turbines. The incoming References:
air is moving at the vehicle velocity entering the engine.
After burning the fuel, the air is hotter and can expand to
a higher velocity out the nozzle. This sets up a pressure 29 Inverted Scramjet Alternate Names: Buoyant
difference that leaves a net thrust. Ramjets cannot operate Scramjet
at zero speed, but they can reach somewhat higher limits Type:
than an engine with rotating machinery (range Mach 0.5
Description: Ramjet and scramjet vehicles would prefer
to about Mach 8).
Hydrogen as a fuel because it gives higher performance,
Status: and can be used for cooling vehicle parts heated by the
Variations: high velocity in air. Unfortunately Hydrogen is also very
low density, which leads to relatively heavy vehicle struc-
ture due to large tanks. This method inverts the problem
• Air Augmented Rocket - This is a form
by using a series of balloons or a lightweight pipe sup-
of combined cycle engine. Since ramjets
ported in the atmosphere. They contain Hydrogen, and
do not function at zero speed, an inter-
the vehicle carries Oxygen in its tanks. Oxygen is about
nal rocket chamber in the engine is used
16 times denser than Hydrogen, so the tank size is much
for initial thrust. By entraining air flow,
reduced.
the thrust level can be augmented at lower
speeds. Once ramjet speeds are reached, Status: This is only a concept at present.
the engine functions in ramjet mode, us- Variations:
ing the rocket chamber as a fuel injec-
tor. At the upper end of ramjet function, References:
the engine transitions back to pure rocket
mode.
30 Laser-Thermal Jet Alternate Names:
References: Type:
Description: In this method a laser beam is focused on
• Wikipedia article: Air-augmented Rocket and absorbed by a heat exchanger on the vehicle, or cre-
ates a laser-sustained plasma. The hot gas is then ex-
hausted for thrust. By not requiring fuel, it is poten-
28 Scramjet Alternate Names: tially efficient. The drawback is it requires a very power-
Type: ful laser to be feasible even for small vehicles. Powerful
lasers are currently expensive. Another limitation is the
Description: This is similar to how a ramjet functions.
distance over which the beam can maintain focus.
The incoming air stream is compressed by shock waves,
mixed with fuel, and expanded against the engine or ve- Status: Minimal experiments have been done in a labo-
hicle body. The difference is the airstream remains su- ratory.
personic relative to the vehicle. The ramjet requirement Variations:
to slow the airstream to subsonic speed becomes ineffi-
cient at higher velocities. Even though the gas is moving References:
supersonically relative to the vehicle, the sidewise expan-
sion can act on the vehicle if the slope of the nozzle is • Myrabo, L. N. “Concept for Light-Powered
low enough. Thus the vehicle can fly faster than the ex- Flight”, AIAA paper number 82-1214 presented
haust gas moves. Scramjets may provide useful thrust up at AIAA/SAE/ASME 18th Joint Propulsion
to about Mach 15, or 60% of Earth orbital speed. Conference, Cleveland, Ohio, 21-23 June 1982.
The very high velocities lead to extreme heating of vehi-
cle parts. High compression and expansion efficiency is
needed to get a positive net thrust, since the energy added 3.5.2 B. Internally Fueled Engines (Rock-
by the fuel becomes small relative to the kinetic energy of ets)
the air flow. Thus development of working scramjets has
proved difficult. Scramjets also do not function at zero Conventional rockets function by expelling a gas at high
velocity, so some other method is needed to get to their velocity in a desired direction. By conservation of mo-
starting point. Therefore complete vehicles will need a mentum (a physical law), the remainder of the vehicle
combined engine system. will gain velocity in the opposite direction. Rockets have
3.5. COMBUSTION ENGINES 109

been the principal method of space transport to date be- their development. [2] The German government pursued
cause they only require internally stored fuel, and can thus the development of rockets to deliver explosives by sub-
operate in a vacuum. Since they propel the vehicle, the orbital trajectories from 1937 to 1945. Subsequently the
fuels are also called Propellants. Rockets can be sorted scientists and their hardware and data went to the United
into types by how many propellants they have and what States and the USSR, where they helped develop sub-
physical form the propellants are stored in. The former orbital ballistic missiles and orbital transport, both first
category includes single fuel Monopropellant, two fuel flying in 1957. The first orbital rockets were essentially
Bipropellant, and the rarely used three fuel Tripropel- identical to ballistic missiles, but have since diverged. By
lant types. The category includes gaseous, solid, liquid, 1963 Liquid Hydrogen/Liquid Oxygen propellants were
and hybrid - part solid and part liquid. in use, which is still the highest energy fuel mix com-
monly used. Liquid and solid chemical rockets are by far
Thrust is mass flow rate times exit or Exhaust Velocity.
To get the most use from a finite amount of fuel, you want the most common space transport method, and are now
built in a number of countries by both governments and
to minimize flow rate and maximize exit velocity. There-
fore the gas should be as hot as possible and have low private companies.
molecular weight. That in turn drives the choice of chem- Design - The non-fuel mass of a rocket stage can be
ical reaction and fuels to use. There are numerous fuel grouped into engines, tanks, and “other”. Rocket engines
combinations that are possible, but a relatively few that can produce 400-1000 N/kg of engine mass in thrust,
have a combination of high energy and other desirable which is many times larger than the 9.8 N/kg force of
characteristics such as density, safety, and low corrosion. gravity. For liftoff from the Earth, you want approxi-
There are several ways to get the gas hot: catalytic decom- mately 1.2-1.5 times the vehicle takeoff weight in thrust,
position, combustion, or external heating. The first two so the engine component then has a mass of about 1.3-
are grouped under chemical rockets, and the latter are 3% of the total vehicle. A large tank, such as the Shuttle
categorized by how the gas gets heated. Chemical rock- External Tank, can weigh 4% of the fuel weight, but other
ets generally use a combination of combustion chamber tanks can range up to 10% of the fuel weight. 'Other' in-
and expansion nozzle, as that is a very efficient way to cludes plumbing, parachutes, landing gear, heat shields,
direct the gas flow at high velocity. Rocket engines will guidance systems, and such non-propulsion parts. It can
function with a surrounding atmosphere, but that impedes range from 1% up to 10% of the total weight.
the gas flow, so they generate lower thrust. The lost thrust Older materials and components required 15% of the to-
can be approximated by the local exterior pressure times tal liftoff mass to be vehicle other than fuel, assuming
the area of the nozzle exit.
a single flight operating life. Modern materials require
Rockets are generally less efficient than air-breathing en- about 10% of the total mass for a many flight operating
gines in terms of momentum per fuel mass, since the lat- life. Structures tend to get heavier at the rate of 10% for
ter can use oxygen from the air as part of the fuel. This in- each factor of 10 in life. This comes from the fatigue life
creases the combustion energy per carried fuel mass. Air of materials under load cycles (flights) as a function of
breathing engines also increase the mass flowing through stress. Lower stress and longer life requires thicker and
the engine via the Nitrogen component of air, and ad- heavier structural parts to distribute the load over. So a
ditional un-combusted air flow using engine-driven fans. 100-use structure will be about 20% heavier than a one-
Both rockets and air-breathing engines involve similar de- use structure.
sign principles, as they both use combustion and hot gas
flow to get reaction forces.
31 Solid Rocket Alternate Names:
History - The earliest reference to using expelled mass
for reaction force is the Greek Archytus around 400 BC. Type:
[1]
This used external heating to generate steam. The first Description: A solid rocket consists of a high-strength
known use of internal chemical energy is in 1232 AD in casing, a nozzle, and a precast solid propellant grain
China by fire arrows using gunpowder as fuel. The idea which burns at a pre-designed rate. The grain is a mix-
may have been transmitted to Europe by the Mongols, ture of materials containing both fuel and oxidizer, so
where experiments and use for fireworks occurred in the combustion can proceed without any external action once
13-15th centuries. Experiments began in the 18th cen- it is ignited. Modern solid propellants have a formula-
tury as a transport method rather than explosive device, tion close to the following: About 15% by weight organic
although military use continued. Notably, rockets used in fuel, usually a type of rubber, about 20% by weight alu-
a battle in 1812 were recorded in the US national anthem. minum powder (which acts as a metallic fuel), and about
The use of rockets to reach space was proposed by Kon- 65% ammonium perchlorate (NH3ClO4), which is the
stantin Tsiolkovsky in 1898. Robert Goddard built exper- oxidizer. About 1-2% epoxy is added to the powders to
imental solid and liquid fuel rockets starting in 1915. A hold them together. The epoxy, being an organic mate-
book published by Hermann Oberth in 1923 influenced rial, is also part of the fuel. Solid propellants burn from
the formation of rocket societies where groups pursued the surface of the precast grain. Therefore the shape of
the grain at ignition, and shape as it burns away, deter-
110 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

mines the thrust level.

Advantages of solid rockets are short preparation time to
launch and long term storage, compared to cryogenic fu-
els like liquid Oxygen. Disadvantages include relatively
low exhaust velocity (2.6-3 km/s), and no easy way to
turn it off or control it once ignited. They are often used
as a booster first stage since the relatively dense fuel (1.35
g/cc) lowers the area of the vehicle. This is an advantage
during the first two minutes of flight, where aerodynamic
drag is important. When aluminum is used in the propel-
lant, part of the end product is aluminum oxide, which is
an excellent abrasive. Thus nozzle erosion is a significant
effect that must be accounted for, in addition to the high
temperatures.
Solid rockets are simpler in the sense of having few mov-
ing parts, but the entire motor casing that surrounds the
fuel grain must withstand the operating pressure. In liq-
uid rockets with fuel pumps, the propellant tanks see hy-
draulic and acceleration loads, which are typically lower,
and only the pumps and combustion chamber see the full
operating pressure.
Status: In common use for rocket stages, particularly the
A test firing of the Space Shuttle Main Engine, a Hydro-
strap-on booster stage.
gen/Oxygen propellant liquid engine.
Variations:
References:
nozzle, reaching about 1.5-2 times the speed of sound by
converting temperature and pressure via expansion into a
32 Hybrid Rocket Alternate Names:
directed flow. A variety of propellant combinations have
Type: been used, including mono- bi-, and even tri-propellants.
Description: The hybrid rocket consists of a solid fuel Monopropellants typically use catalytic decomposition
grain and a liquid oxidizer. One combination is rubber for heating. The most common form of liquid rocket uses
for the fuel and liquid oxygen for the oxidizer. The fuel a separate fuel and oxidizer, which are mixed and burned
is in the form of a hollow cylinder or perforated block. in the combustion chamber. Many bipropellant mixtures
The oxidizer is sprayed onto the fuel and the material is are possible, but the highest energy-to-mass ratio mix in
ignited. By not being self-supporting in combustion, the common use is from 1 part Hydrogen to 6 parts Oxygen.
fuel part can be treated as non- hazardous when being This produces mostly steam with a little leftover Hydro-
made and shipped. Only when on the launch pad and the gen, which lowers the average molecular weight and thus
oxidizer tank is filled is there a hazardous combination. increases the average molecule velocity. This propellant
With only a single liquid to handle, the hardware is rela- mix can reach about 4.7 km/s exhaust velocity under the
tively simple in design. Hybrid rockets are intermediate best conditions.
in performance compared to solid and full liquid engines. Some propellant mixes will burn on contact, and so do not
Status: require an ignition source. These are called Hypergolic
propellants. Some liquid propellants are liquid at room
Variations: temperature and can be stored for long periods in a tank.
References: These are referred to as Storable. Others, including Hy-
drogen, Methane, and Oxygen, are only liquid at very low
temperatures. These are referred to as Cryogenic.
33 Liquid Rocket Alternate Names: Status: This is the most common form of launch propul-
Type: Fuel/Oxidizer Combustion by Combustion Gas sion used to date to put things into Earth orbit.
exhaust Variations: There are multiple combinations of liquid
Description: In a liquid rocket, the propellant ingredi- engine types that are possible and have been used. They
ents are forced into a combustion chamber, where they can be sorted by what propellant combination is used,
burn, and which leads to a converging-diverging nozzle. how the propellants are delivered into the combustion
The flow becomes sonic at the narrow part of the nozzle, chamber, and how the resulting hot gas is expanded out
then continues to accelerate in the diverging part of the of it.
3.5. COMBUSTION ENGINES 111

• Variations by composition - The fol- zle, where the gas flow is on the in-
lowing tables list some oxidizers and fu- side, surrounded by structure which di-
els, and combinations thereof. It is not rects the flow. An alternate design called
an exhaustive list, and some ingredients an Aerospike nozzle inverts this arrange-
have practical issues such as storage tem- ment, with the structure on the inside in
perature, human toxicity, corrosiveness, the form of a wedge or cone, and the
or chemical instability. Rocket propel- gas flow surrounding it. The outer edge
lants by their nature contain a lot of of the gas flow is contained by the sur-
chemical energy, and that energy can rounding atmosphere. Since that auto-
cause unintended reactions. Actual en- matically adjust for pressure differences,
gine performance depends on factors it is a type of altitude compensating noz-
such as chamber and exit pressures, so the zle. The benefit to compensation is that
table values should only be used for gen- nozzle exit area in an atmosphere repre-
eral comparisons. Kerosine is a mixture sents a thrust loss. A truncated cone can
of compounds derived from petroleum, integrate better with some vehicle shapes,
and Rocket Propellant 1 (RP-1) is a stan- and the larger chamber and nozzle area
dardized type of Kerosine specified as a can lower heat flux.
rocket fuel. Therefore it does not have an
exact formula, and it is given as an ap- • Variation by cooling type - The high en-
proximate average value. It also does not ergy combustion in liquid rocket engines
have a well defined melting and boiling can exceed the melting point of most
point, which is given as a range defined structural materials. Therefore meth-
by distillation of its components. ods to prevent this are needed in all but
the smallest engines, which simply use
Oxidizers high temperature alloys and radiate the
Fuels heat away. In one method, the fuel is
run through channels in the rocket en-
Propellant Combinations gine walls to keep it from overheating.
Exhaust Velocities assume chamber pressure of 6.9 MPa This also recovers some of the energy
and exit pressure of 0.1 MPa (Earth sea level) with opti- that would otherwise be lost. A method
mum nozzle expansion. in recent development by Orbitec injects
a counterflow vortex of one propellant in-
• Variations by fuel feed - Some engine gredient along the inside of the engine
designs use a pump to feed the propel- walls, which is mixed with the other in-
lants into the combustion chamber, oth- gredient at the back and the hot gases
ers use a pressurized tank. Large engines then flow in the forward direction down
may use a combination of tank pressur- the core of the chamber. The unburned
ization to prevent cavitation of the pump ingredient protects the structure with a
inlet and pumps to reach chamber in- layer of cool gas.
let pressure. The fuel must enter the
chamber at a higher pressure than the References:
combustion pressure in a steady state en-
gine. Non-steady state pulsed ignition en-
• Cooper, Larry P. “Status of Advanced Orbital
gines are possible, but not generally used.
Transfer Propulsion”, Space Technology (Oxford),
Steady state engined deliver more con-
v 7 no 3 pp 205-16, 1987.
tinuous thrust. Pumps require a lot of
power to operate, and generally use the • Godai, Tomifumi “H-II Rocket: New Japanese
same fuel as the rocket. In Gas Gener- Launch VehicleÊ in the 1990s”, Endeavour , v 11
ator systems a portion of the fuel flow is no 3 pp 116-21, 1987.
used to create hot gas which drives a tur-
bine to run the pump. The hot gas is then • Wilhite, A. W. “Advanced Rocket Propulsion Tech-
vented. In Staged Combustion systems, nology Assessment for Future Space Transporta-
the hot gas is not completely burned, and tion”, Journal of Spacecraft and Rockets, v 19 no
is fed into the combustion chamber. This 4 pp 314-19, 1982.
is more efficient but also more complex.

• Variation by nozzle type - Most rocket 34 Gaseous Thruster Alternate Names:

engines to date use a bell-shaped noz- Type:
112 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

Description: In this method the propellant is introduced 3.6.1 Electro-thermal Engines

in gas form to the chamber. It may be a mono-propellant
(a single gas) or a bi-propellant combination. Due to high Electro-thermal methods convert externally supplied
tank mass this is usually used for small auxiliary thrusters. electric power to heating of the propellant.
By using direct pressure from the tank to make the pro-
pellant flow, it can be very simple.
36 Electric-Rail Rocket Alternate Names: Elec-
Status: trothermal Ramjet
Variations: Type: Heated Gas Flow by Power Line
References: Description: High voltage electricity supplied by rails is
shorted through a tungsten heat exchanger in the engine.
This heats Hydrogen carried by the vehicle traveling be-
tween the rails. The rails are assumed to be set up on
35 Mechanically Augmented Thruster Alternate
an incline. Performance would be the same as a nuclear
Names:
thermal rocket, about 9 km/s exhaust velocity, since both
Type: use heated Hydrogen limited by the melting point of the
Description: The velocity of the exhaust gases are in- engine components. It requires very high power levels for
creased by placing the thrusters on the end of rotating large vehicles: 44 MW / ton / g acceleration. Since power
arms. This can add 2-3 km/s to the exhaust velocity based is coming from outside the vehicle, it can generate enough
on structural limits. It requires some external energy in- thrust for launch from Earth. Since the Hydrogen fuel is
put to maintain the rotation of the arms, since the thrust only stored for a short period, insulation and boiloff are
opposes their rotation. Full electric thrusters generally not major problems. This method would compete with
have higher performance than this method, so they are other ground accelerator type systems such as gas guns
preferred. or electromagnetic coils.

Status: Concept only, since there are better options. Status: Concept only at this time.

Variations: Variations:

References: References:

• Wilbur, P. J.; Mitchell, C. E.; Shaw, B. D. “Elec-

trothermal Ramjet”, AIAA paper number 82-1216
References presented at AIAA/SAE/ASME 18th Joint Propul-
sion Conference, Cleveland, OH, 21-23 June 1982.
[1] NASA, Brief History of Rockets, web page 15 Apr 2012
37 Resistojet Alternate Names:
[2] For a history of the rocket societies, see Frank Winter Type: Heated Gas Flow by Photovoltaic Array
Prelude to the Space Age, Smithsonian Inst Press, 1983
Description: In this method sunlight generates electric-
ity, which is used to heat gas passed over or through a
heating element, often after catalytic decomposition of a
storable fuel to a lighter gas. In principle it is similar to the
3.6 Thermal Engines Electric rail rocket, but is used for smaller thrusters in or-
bit with attached solar array as power source. This limits
Strictly speaking combustion engines with expansion noz- the thrust, so it is not powerful enough to use for launch.
zles are also thermal engines since they rely on hot gases. Compared to chemical[1]
thrusters it gets about 50% better
For discussion purposes we placed that group on the pre- exhaust velocity
vious page, and on this one look at non-combustion ther- Status: This method was used to extend the operating
mal engines. It is possible in theory to combine both life of communications satellites, since they had large so-
into a single device, but this is not usually considered for lar arrays for their main job. The recent tendency is to
practical design reasons. Performance is always impor- use ion thrusters, which have even better performance.
tant, and light molecules have higher exhaust velocity at a
Variations:
given temperature. Therefore the tendency is to use Hy-
drogen if practical since it is the lightest molecule. Liquid References:
Hydrogen requires extremely cold temperatures (14K or
−435F), so storage for long periods is not practical for a • Louviere, Allen J. et al “Water-Propellant Resisto-
small tank. Larger systems have better surface to volume jets for Man- Tended Platforms”, NASA Technical
ratios and can use active cooling to keep it liquid. Memorandum 100110, 1987.
3.6. THERMAL ENGINES 113

3.6.2 Photo-thermal Engines then expelled through a nozzle. By using a powerful en-
ergy source external to the propellant, exhaust velocity of
This group uses direct electromagnetic radiation (pho- around 10 km/s can be reached with high thrust to mass.
tons) from natural or artificial sources to heat the pro- One method of doing this on Earth is with large, ground-
pellant. based lasers. Alternately the lasers can be at ground level,
and a directing mirror is located on top of a large tower.
A vacuum pipe connects the two. The extra height avoids
38 Solar-Thermal Engine Alternate Names:
atmospheric distortion and allows more distance to the
Type: Heated Gas Flow by Solar Flux horizon. Use of laser propulsion only in an upper stage
Description: In this method sunlight is concentrated by would allow smaller lasers than are required for a first
a reflector or lens, and then heats an absorber. The ab- stage system. Even so, powerful enough lasers are not
sorber transfers heat to a working fluid, usually Hydro- available yet, which limits the use of the method. Other
gen. The Hydrogen is then expanded through a nozzle. methods of supplying energy to a vehicle are likely to be
If the absorber is in the form of pipes, the exhaust veloc- less expensive.
ity is limited to about 9 km/s. If the absorber is in the Status: Concept only at present.
form of a particle bed, which does not require mechan-
Variations:
ical strength, refractory carbides can be used. Tantalum
hafnium carbide, formula Ta4 HfC5 , is hypothesized to References:
have a melting point of 4488 K (7619 F) and would set an
upper limit to the particle bed method. There are obvious • Abe, T.; Shimada, T. “Laser Assisted Propulsion
difficulties in testing that material as no other container System Experiment on Space Flyer Unit”, 38th In-
could hold it when melted. The bed is rotated to keep ternational Astronautical Federation Conference pa-
the particles from being blown out, and Hydrogen flow is per number IAF-87-298, 1987.
from the outside in, then out a nozzle. Sunlight is focused
on the inside surface which is then the hottest point. Hy- • Abe, T.; Kuriki, K. “Laser Propulsion Test Onboard
drogen dissociates above 3000K to single atoms, leading Space Station”, Space Solar Power Review vol 5 no
to exhaust velocities slightly above 10 km/s. 2 pp 121-5, 1985.

Solar concentrators can be very low mass, and use all of • Jones, L. W.; Keefer, D. R. “NASA’s Laser Propul-
the solar spectrum. So they can reach higher power levels sion Project”, Astronautics and Aeronautics, v 20 no
than electro-thermal thrusters. They are more suited to 9 pp 66-73, 1982.
main propulsion rather than orbit maintenance as electro-
thermal typically is. The direction of the Sun is usually
40 Laser Detonation-Wave Engine Alternate
not the same as the direction of thrust, and changes over
Names:
time. So solar-thermal systems need a way to point the
concentrators. One way to handle this is to roll the vehicle Type: Plasma via Laser
about the thrust axis plus pivot the concentrators about a Description: In this method the propellant is a solid
perpendicular axis. The concentrators are typically large, block with a flat bottom. A first laser pulse evaporates
so need to be assembled or unfolded in orbit. a layer of propellant. A second, larger, pulse creates a
Status: Components have been tested by the US Air plasma detonation wave, which shocks and heats the pro-
Force. Ion and plasma thrusters have become preferred pellant layer. The layer expands against the base of the
because of their 3-5 times higher exhaust velocity. solid block of remaining propellant. The pulse pattern is
repeated as soon as the plasma dissipates and the laser can
Variations:
reach the block again. Because no fuel tank overhead or
References: engine is needed, the vehicles are potentially very cheap,
but it requires a powerful laser to function. For example,
• Gartrell, C. F. “Future Solar Orbital Transfer Ve- a 10 kg vehicle accelerated at 2 g’s using a 20 km/s ex-
hicle Concept”, IEEE Transactions on Aerospace haust velocity plasma wave would require a pulsed laser
Electronic Systems, vol AES-19 no 5 pp 704-10, with an average power of 2 MW, while large industrial
1983. pulsed lasers are about 600 W [2] . Since the propellant
temperature is not limited by any container, it can be hot-
ter than other thermal methods, and so have better per-
39 Laser-Thermal Engine Alternate Names: formance.
Type: Heated Gas Flow by Laser 10 kg is a feasible size for this type of vehicle, if the
Description: A laser beam from an external source is laser can be kept focused on it till it reaches orbit ve-
passed through a window into a chamber. It is then locity. In theory this would deliver about 6 kg to orbit.
absorbed by a heat exchanger or is focused to create a For larger payloads, such as to carry passengers, the laser
laser-sustained plasma within the gas flow. Hot gas is would need to scale to GigaWatt power levels, which has
114 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

led to the common saying within space propulsion circles Description: Hydrogen is heated by flowing through a
of “there is nothing wrong with laser propulsion except critical Nuclear Reactor, then expelled through a nozzle
the lack of GigaWatt lasers”. Similar power levels are at high velocity. The low molecular weight of Hydrogen
generally required of all launch methods carrying met- gas allows a higher exhaust velocity, about 9 km/s, than
ric ton or larger payloads from Earth. For example, the combustion rockets. Advantages include high power lev-
Space Shuttle Main Engines, of which the Space Shuttle els and high total stored energy.
Orbiter used three, each had a power of 9.2 GW. Issues for Nuclear Thermal include:
Status: Some detonation experiments have been carried
out at small scale in a laboratory. • For single missions the energy of the reactor has
Variations: barely been tapped by the time the hydrogen is
consumed, leaving a now radioactive core to work
References: around if refueling for another mission.
• Radiation shielding is needed for crew and cargo. To
• Kare, J.T. “SDIO/DARPA Workshop on Laser
some degree that is mitigated by the need for shield-
Propulsion, Volume 1: Executive Summary”
ing from the natural space radiation environment.
Lawrence Livermore National Laboratory report
number DE87-003254, 1987. • Any type of nuclear device raises extensive safety
and environmental issues, even if not activated until
in orbit, and even if not really warranted for tech-
41 Microwave Thermal Engine Alternate Names: nical reasons. One way around this issue is to mine
Type: Heated Gas Flow via Microwaves and use the fissionable materials away from Earth.
For example, parts of the Lunar surface have Tho-
Description: For this method microwaves from an exter-
rium concentrations of 10 parts per million. If used
nal source are absorbed by a heat exchanger or concen-
as reactor fuel, that can provide a net energy of 350
trated by a waveguide into the engine. Hydrogen flows
MJ/kg of unprocessed Lunar soil, or about 7 times
through the engine, absorbs the energy, and then exits
the energy density of gasoline on Earth. The pro-
by a nozzle. A large phased microwave array on the
cessed fuel of course will be 100,000 times higher
ground can focus onto a rocket-sized area over a range of
energy density, but the unprocessed ore energy den-
hundreds of kilometers. Given a way to couple the mi-
sity is an indicator of the feasibility of mining it.
crowave energy to a working fluid such as Hydrogen, this
type of propulsion could provide significant launch vehi-
cle velocities. High power microwave amplifiers exist in In comparison to ion and plasma thrusters, nuclear ther-
a variety of forms with efficiencies up to 75% and power mal has about 3-5 times lower exhaust velocity, but much
levels up to one megawatt. As compared to Laser Ther- higher thrust levels. The near-instant burns relative to or-
mal the chief advantage is the availability of high power bit time for nuclear thermal vs. constant burn for electric
microwave sources at relatively low cost. A disadvantage thrusters reduces the latter’s advantage by 30%, but are
is the much larger wavelength of microwaves vs lasers, so still are 2.1-3.5 higher. The choice of method would de-
maintaining focus over a distance is harder. pend on the importance of fuel mass, which is usually
high, so today electric thrusters are usually preferred. In
comparison to solar thermal it has about the same per-
• Design Example: 10 meter diameter re- formance in terms of exhaust velocity, but higher thrust
ceiver, 5 cm wavelength, 1 km phased ar- levels.
ray, range = 200 km.
History The two major American reactor development
efforts in the 1960s were KIWI and NERVA. Together
Status: Concept only so far.
with the $328 million spent on technology development,
Variations: $90 million spent on the Nuclear Rocket Development
References: Station in Nevada, and $153 million on other test facili-
ties, almost $1.4 billion (in then-year dollars) was spent
on nuclear rocket development from 1955 to 1972 (see
3.6.3 Nuclear Thermal Engines Appendix 2:Reference Data ). Although considerable
engine testing was done, the problem of solid core fuel
This group use a nuclear reactor to heat the propellant. damage at high operating temperatures, which are desir-
They vary in the physical state of the reactor core (solid, able for performance, was not solved.
liquid, or gas). Status: Nuclear rockets reached the testing stage in the
1960’s under the NERVA program. Lack of actual need
in a mission and rising worries about anything nuclear led
42 Solid Core Nuclear Engine Alternate Names: to a halt in development. Since then only minor studies
Type: Heated Gas Flow by Fission Reactor have been done.
3.7. BULK MATTER ENGINES 115

Variations: Status: Currently a concept only.

Variations:
• LOX-Augmented Nuclear Thermal Rocket
References:
Propulsion This injects Oxygen after the Hydro-
gen is heated by the reactor core [3] . This increases
thrust by about a factor of 3, which is useful for 44 Gas Core Nuclear Alternate Names:
initial launch, then transitions to pure Hydrogen
later for higher efficiency. Exhaust velocity is Type: Heated Gas Flow by Fission Reactor
lowered by about 1/3 when adding Oxygen due Description: In this version the reactor core is hot
to the higher molecular weight of the resulting enough to be in gaseous form. The Hydrogen flow is
exhaust. seeded with an absorbent material to directly absorb the
thermal radiation from the core. The core is kept from
• Particle Bed Nuclear Engine - Although the nu- leaking out the nozzle either by a transparent container
clear rocket program was stopped a number of years (nuclear light bulb), a flow vortex, which uses the density
ago, more recent work at Brookhaven National Lab- difference between uranium and hydrogen, or by mag-
oratories on fluidized particle bed reactors warrants netic separation, which uses the ionization difference be-
their consideration for launch vehicles. The small tween the uranium and the hydrogen. Expected perfor-
particle size (.3 mm) allows high heat transfer rates mance ranges from 15-20 km/s for a quartz container, up
to the working fluid, hydrogen, and hence potentially to 30-50 km/s for a flow vortex. The latter is in the range
high thrust to weight ratios. The smaller particles of ion or plasma thrusters, but development and testing
also potentially solves the fuel damage problem, as would be as difficult as for liquid core due to the possibil-
there is less scope for cracking in a fine powder. Ex- ity of ejecting the core during tests. A solid core reactor
haust velocity is increased slightly to 10 km/s. producing electricity for an electric thruster would have
the same performance as gas core nuclear, with fewer de-
References: velopment problems.
Status: Currently a concept only.
• Thomas, Ulrich “Nuclear Ferry - Cislunar Space
Variations:
Transportation Option of the Future”, Space Tech-
nology (Oxford) v 7 no 3 pp 227-234, 1987. References:

• Holman, R.R.; Pierce, B. L. “Development of • Wikipedia article: Nuclear Lighbulb

NERVA reactor for Space Nuclear Propulsion”,
presented at AIAA/ASME/SAE/ASEE 22nd Joint
Propulsion Conference, Huntsville, Alabama, 16-18 References:
Jun 1986, AIAA paper number 86-1582, 1986.
[1] Seitzman, J. Electrothermal Thrusters, 2006.
• Thom, K. et al “Physics and Potentials of Fissioning
Plasmas for Space Power and Propulsion”, Acta As- [2] JK Lasers Pulsed Nd:YAG Lasers spec sheet
tronautica vol 3 no 7-8 pp 505-16, Jul. -Aug. 1976. [3] Next Big Future Blog Recent Nuclear Thermal Rocket
Proposals, 17 Jun 2009.
• DiStefano, E. “Space Nuclear Propulsion - Future
Applications and Technology”, 2nd Symposium on
Space Nuclear Power Systems, Albequerque, New
Mexico, 14 January 1985, pp 331-342, 1987. 3.7 Bulk Matter Engines
This group of propulsion methods expel solid particles as
43 Liquid Core Nuclear Alternate Names: reaction mass, or accelerate cargo directly.
Type: Heated Gas Flow by Fission Reactor
Description: In order to attain higher performance than 45 Rotary Engine Alternate Names:
a solid core rocket, the reactor core is raised to a high
enough temperature to become liquid. Hydrogen is bub- Type:
bled through the liquid, then exhausted out a nozzle. The Description: A one or two stage rotary mechanism me-
Hydrogen is first used to cool the reactor container, and chanically accelerates a small amount of reaction mass,
so the temperature limit is governed by that rather than then releases it. In the two stage version, top speeds of
the melting point of the core. Expected exhaust velocity ~6 km/s are possible. The advantage is to be able to use
is up to 13-15 km/s, but development and testing of this any available material as reaction mass. The disadvantage
type of engine will be difficult, as test failures can easily is the relatively low ejection velocity. There is an oppor-
squirt the core fluid out the nozzle. tunity to mechanically augment another type of thruster
116 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

by having it moving backwards relative to the vehicle cen- Type:

ter when running, thus adding up to 6 km/s to whatever Description: A railgun consists of two current-carrying
the native exhaust velocity is. The duty cycle of the aug-rails, and a projectile that creates a short circuit between
mented thruster would be limited to about 25%, because the rails. The rails create a vertical magnetic field be-
the rest of the time it would be moving in the wrong direc-
tween them, as the current goes in opposite directions
tion to add velocity. This raises the issue of many on-off in each rail. The interaction of the field with the cur-
cycles of the thruster. rent in the short circuit creates a Lorentz force forward
Status: Centrifuges and flywheels have long been in use. that accelerates a projectile. In a large version this would
Versions optimized for ejecting mass at high velocity are be used to launch vehicles in their entirety. In a smaller
still in the concept stage. version it would eject a reaction mass and function as a
Variations: propulsion engine.

References: Status: Railguns have been under development as re-

placements for conventional artillery, but are not yet op-
erational as such. For launch vehicle acceleration, a much
46 Coilgun Engine/Launcher Alternate Names: larger power supply would be needed, and this has so far
Mass Driver Reaction Engine been merely a concept
Type: Variations:
Description: This is a type of electromagnetic acceler-
ator. A carrier, or bucket, is accelerated by interaction
of magnetic fields from 'driver' coils. The carrier holds 3.8 Ion and Plasma Engines
a reaction mass/cargo, which is released. The bucket is
slowed down and reused. There is no theoretical limit to Combustion type engines are limited in temperature, and
velocity of this type of device - particle accelerators have thus gas exhaust velocity, either by the chemical energy
reached near lightspeed - but practical limits are likely to of the fuel, or the melting point of the engine materials.
be on the order of 10 km/s. As a vehicle engine it is not Ion and plasma type engines bypass these limits and reach
dramatically different than other electric thrusters in per- much higher velocities by one or more of: using external
formance, aside from the benefit of using any bulk mass energy sources, lower energy density to limit engine heat-
as propellant. As a fixed launcher for bulk materials, a ing, or using non-material containment such as magnetic
large power supply and rapid recycling of the buckets al- fields to direct the flow. These methods also tend to make
lows very high mass delivery rates to orbit velocity. Unas- ion and plasma engines heavy compared to combustion
sisted bulk material will have an orbit that intersects the engines, so they are not used for launch from high grav-
launch point, so a device in orbit will have to catch the ity bodies. Rather they operate where there is already a
cargo lots and add some velocity to them. low-g environment in orbit or away from gravity wells.
Status: This method was extensively studied as part of The ion and plasma types both involve high energy parti-
the “Space Colony” studies of the ~1970’s, and prototype cles. The distinction is one of density. In the former the
parts were built and tested. particles act individually, while in the latter they are dense
Variations: enough to require treating them as a flowing medium.
References:
3.8.1 Ion Engines
Current
Magnetic + 48 Electrostatic Ion Engine Alternate Names:

–
Type:
field
Description: Electrostatic ion thrusters generally work
by first ionizing the propellant, then using a voltage gra-
dient to accelerate the ions. They are capable of high
Projectile
exhaust velocities (30-50 km/s), but relatively low energy
Rail density and thrust in order to prevent overheating and ero-
B sion of the screens creating the voltage difference. Since
Force
I the voltage gradient is uniform between two flat screens,
Rail F
the ion beam is well directed without a nozzle. Ejecting
only positive ions would produce a net charge on the en-
Schematic of railgun operation. gine, so ion engines also have an electron gun to balance
the charges emitted.
47 Railgun Engine/Launcher Alternate Names: All ion engines require an external power source. A num-
3.8. ION AND PLASMA ENGINES 117

ber of variations for the power source have been pro- small thrust levels for small spacecraft,
posed, but only Solar-Electric has actually been used. or for steering, since lithography can be
Since ionization merely removes an electron, but does not used to create microscopic nozzles.
accelerate the ion, it represents an efficiency loss for the • 48g Colloid Thruster - Rather than indi-
engine. Therefore ion engines tend to prefer fuels with vidual ions, this method accelerates rela-
high atomic weight relative to ionization energy. Since tively massive colloidal charged particles.
most high atomic weight elements are solids, Mercury and
Xenon have often been chosen, and sometimes Argon if
operating in a cold environment or reaching a very high References:
exhaust velocity is desired.
Status: Ion thrusters have been used in recent years on Solar-Electric, 1970s
some communications satellites and interplanetary space-
craft. • Loeb, H. W. “Electric Propulsion Technology Sta-
Variations: tus and Development Plans - European Programs
(Space Vehicles)", J. Spacecraft and Rockets , vol
11 no 12 pp 821-8, Dec. 1974.
• 48a Solar-Electric Ion - Sunlight is con-
verted to electricity by a photovoltaic ar- • Mutin, J.; Tatry, B. “Electric Propulsion in the Field
ray. The electricity is used to ionize and of Space”, Acta Electron. (France) vol 17 no 4 pp
accelerate the propellant by electrostatic 357-70, Oct. 1974 (in French).
voltage.
• 48b Thermoelectric Ion - Radioactive • Byers, D. C.; Rawlin, V. K. “Critical Elements of
isotope decay produces heat. Heat is con- Electron- Bombardment Propulsion for Large Space
verted to electricity by semiconductors. Systems”, J. Spacecraft and Rockets vol 14 no 11 pp
Electricity ionizes and accelerates atoms 648-54, Nov. 1977.
in the engine.
• Parkash, D. M. “Electric Propulsion for Space Mis-
• 48c Laser-Electric Ion - A laser tuned sions”, Electr. India vol 19 no 7 pp 5-15, 15 April
to the optimum absorption wavelength of 1979.
photovoltaic cells supplies power. The
cells convert laser light to electricity,
Solar-Electric, 1980s
which is used to power the ion engine.
The ion engine then ionizes and acceler-
ates the propellant. The reason for using • Kaufman, H. R. “Performance of Large Inert-Gas
a laser is to get higher power levels than Thrusters”, AIAA paper number 81-0720 presented
natural sunlight provides. at 15th International Electric Propulsion Confer-
• 48d Microwave-Electric Ion - A mi- ence, Las Vegas, Nevada, 21-23 April 1981.
crowave receiving antenna, or Rectenna, • Clark, K. E.; Kaufman, H. B. “Aerospace Highlights
on the spacecraft converts microwaves to 1981: Electric Propulsion”, Astronautics and Aero-
electricity. Electricity is used to ionize nautics, v 19 no 12 pp 58-59, 1981.
and accelerate atoms. Microwave anten-
nas can be low mass, but the operating • Zafran, S. et al “Aerospace Highlights 1982: Elec-
distance is limited by the ability to focus tric Propulsion”, Astronautics and Aeronautics, v 20
the long wavelength. no 12 pp 71-72, 1982.
• 48e Nuclear-Electric Ion - A nuclear re-
actor generates heat, which is converted • James, E.; Ramsey, W., Sr.; Steiner, G. “Develop-
to electricity in thermoelectric or tur- ing a Scaleable Inert Gas Ion Thruster”, AIAA paper
bine/generator cycles. Electricity is used number 82-1275 presented at AIAA/SAE/ASME
to ionize propellant and accelerate it by 18th Joint Propulsion Conference, Cleveland, OH,
electrostatic voltage. This is more suited 21- 23 June 1982.
to high power applications, and outer So-
• Anon. “Ion Propulsion Engine Tests Scheduled”,
lar System locations where sunlight is
Aviation Week and Space Technology, v 116 no 26
weak.
pp 144-5, 1982.
• 48f Electrospray Thruster - An ionic
liquid is emitted from a capillary noz- • Brophy, J. R.; Wilbur, P. J. “Recent Developments
zle similar to an inkjet printer. A high in Ion Sources for Space Propulsion”, Proceedings
voltage electrode then accelerates the of the Intl. Ion Engineering Congress vol 1 pp 411-
charged droplets. This is suited to very 22, 1983.
118 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

• Bartoli, C. et al “Recent Developments in High Cur- Nuclear-Electric Ion:

rent Liquid Metal Ion Sources for Space Propul-
sion”, Vacuum vol 34 no 1-2 pp 43-6, Jan. -Feb. • Reichel, R. H. “The Air-Scooping Nuclear-Electric
1984. Propulsion Concept for Advanced Orbital Space
• Jones, R. M.; Poeschel, R. L. “Primary Space Transportation Missions”, J. British Interplanetary
Propulsion for 1995-2000 - Electrostatic Technol- Soc. vol 31 no 2 pp 62-6, Feb. 1978.
ogy Applications” AIAA/SAE/ASME 20th Joint • Hsieh, T. M.; Phillips, W. M. “An Improved
Propulsion Conference, AIAA paper number 84- Thermionic Power Conversion System for Space
1450, 1984. Propulsion”, Proceedings of the 13th Intersoci-
• Imai, R.; Kitamura, S. “Space Operation of Engi- ety Energy Conversion Engineering Conference pp
neering Test Satellite -III Ion Engine”, Proceedings 1917-1923, 1978.
of JSASS/AIAA/DGLR 17th Intl. Electric Propul- • Ray, P. K. “Solar Electric versus Nuclear Electric
sion Conf. pp 103-8, 1984. Propulsion in Geocentric Space”, Trans. Am. Nucl.
• Bartoli, C. et al “A Liquid Caesium Field Ion Source Soc. vol 39 pp 358-9, Nov.-Dec. 1981.
for Space Propulsion”, J. Phys. D vol 17 no 12 pp • Powell, J. R.; Botts, T. E.; Myrabo, L. N.
2473-83, 14 Dec. 1984. “Annular Bed Nuclear Power Source for Electric
• Voulelikas, G. D. “Electric Propulsion: A Review Thrusters”, AIAA paper number 82-1278 presented
of Future Space Propulsion Technology” Commu- at AIAA/SAE/ ASME 18th Joint Propulsion Con-
nications Research Centre, Ottawa, Ontario, report ference, Cleveland, Ohio, 21-23 June 1982.
number CRC-396, October 1985.
• Powell, J. R.; Boots, T. E. “Integrated Nuclear
• Nakamura, Y.; Kuricki, K. “Electric Propulsion Propulsion/Prime Power Systems”, AIAA paper
Test Onboard the Space Station”, Space Solar Power number 82-1215 presented at AIAA/SAE/ASME
Review vol 5 no 2 pp 213-9, 1985. 18th Joint Propulsion Conference, Cleveland, Ohio,
21-23 June 1982.
• Rawlin, Vincent K; Patterson, Michael J. “High
Power Ion Thruster Performance”, NASA Techni- • Buden, D.; Garrison, P. W. “Space Nuclear Power
cal Memorandum 100127, 1987. Systems and the Design of the Nuclear Electric
Propulsion OTV”, presented at AIAA/SAE/ASME
• Mitterauer, J. “Liquid Metal Ion Sources as 20th Joint Propulsion Conference, AIAA paper
Thrusters for Electric Space Propulsion”, J. Phys. number 84-1447, 1984.
Colloq. (France) vol 48, no C-6, pp 171-6, Nov.
1987. • Cutler, A. H. “Power Demands for Space Resource
Utilization”, Space Nuclear Power Systems 1986 pp
• Mitterauer, J. “Field Emission Electric Propulsion - 25-42.
Emission Site Distribution of Slit Emitters”, IEEE
Trans. on Plasma Sci. vol PS-15, pp 593-8, Oct.
Electrospray Thruster:
1987.
• Stuhlinger, E. et al “Solar-Electric Propulsion for a • Micromachined propulsion systems for very
Comet Nucleus Sample Return Mission” presented small satellites, EFPL Microsystems for Space
at 38th Congress of the International Astronautical Technologies Lab.
Federation, Brighton, England, 10 Ocotober 1987.

Laser-Electric Ion: 3.8.2 Plasma Engines

49 Arc Jet Engine Alternate Names:
• Maeno, K. “Advanced Scheme of CO2 Laser for
Space Propulsion”, Space Solar Power Review vol Type:
5 no 2 pp 207-11, 1985. Description: Sunlight is converted to electricity by a
photovoltaic array. The electricity is arced through a pro-
Microwave-Electic Ion: pellant stream, heating it to a plasma. The propellant is
then expanded through a nozzle. No specific protection
• Nordley, G. D.; Brown, W. C. “Space Based from heating is used in the nozzle. The arc jet method
Nuclear-Microwave Electric Propulsion”, 3rd Sym- uses relatively low energy density and thrust levels, and a
posium on Space Nuclear Power Systems, Albu- thick-walled refractory metal for the chamber and nozzle.
querque, New Mexico, 13 January 1986, pp 383-95, Exhaust velocities of about 6-8 km/s are reached with liq-
1987. uid fuels like Hydrazine or Ammonia.
3.8. ION AND PLASMA ENGINES 119

Status: Arc Jets have been used on spacecraft, and are 51 Microwave Heated Plasma Engine Alter-
commonly used on Earth as plasma torches for cutting nate Names: Electron-Cyclotron Absorption Rocket,
metals. Variable Specific Impulse Magnetoplasma Rocket
Variations: (VASIMR), Helicon plasma thruster
Type:
• Pulsed Plasma Thruster - operates by
Description: Partially ionized gas directly absorbs mi-
ablating Teflon material between two
crowaves, becoming hot, then expands through rocket
electrodes by means of an electric arc.
nozzle. To keep the hot plasma from damaging the en-
The Lorentz force accelerates the mate-
gine, a magnetic field is used for confinement, often us-
rial out as thrust.
ing superconductors for efficiency. Current versions use
• Vacuum Arc Thruster - operates by va- Argon as propellant, but other gases should function with
porizing and ionizing cathode material by tuning. It may be possible to use unprocessed rock as
means of a vacuum arc. The plasma ac- fuel.
celerates outward as thrust.
Status: VASIMR is in ground testing
• Magnetoplasmadynamic Thruster -
operates via electric arc between central Variations:
cathode and shell anode. Flowing gas be- References:
tween them is ionized and accelerated by
the Lorentz force from current and field
interactions. • Personal communication from Ad Astra Rocket
Company in reference to alternate fuels, 27 Mar
References: 2012. Extensive publications about their thruster are
linked to the main description page.
• Hardy, Terry L.; Curran, Francis M. “Low
Power DC Arcjet Operation with Hydro-
52 Fusion Heated Plasma Engine Alternate Names:
gen/Nitrogen/Ammoinia Mixtures”, NASA
Technical Memorandum 89876, 1987. Type:
• Stone, James R.; Huston, Edward S. “NASA/USAF Description: The exhaust of a pure fusion rocket is a
Arcjet Research and Technology Program”, NASA thin, extremely hot plasma which gives very high per-
Technical Memorandum 100112, 1987. formance. If higher thrust is needed, Hydrogen can be
mixed with the plasma. This increases thrust at the ex-
• Kagaya, Y. et al “Quasi-steady MPD Arc-jet for pense of performance. By varying the mixture ratio, the
Space Propulsion”, Symposium for Space Technol- performance vs thrust can be adjusted as needed during
ogy and Science, Tokyo, Japan, 19 May 1986, pp a mission.
145-154, 1986.
Status: Fusion engines await practical fusion reactors,
• Manago, Masata et al “Fast Acting Valve for MPD which are still in the research stage as of 2012.
Arcjet”, IHI Engineering Review, v 19 no 2 pp 99-
Variations:
100, April 1986.
• Pivirotto, T. J.; King, D. Q. “Thermal Arcjet Tech- • Reactor Leakage Mixed - Some of the
nology for Space Propulsion”, Chemical Propulsion fusion reaction plasma leaks past the con-
Information Agency, Laurel, Maryland, 1985. tainment fields. This may be a mix of re-
acted particles, and un-reacted fuel. The
50 Electron Beam Heated Plasma Engine Alternate leakage can be directly used as engine ex-
Names: haust, or further mixed with additional
material for higher mass flow/thrust at
Type: lower performance.
Description: A high voltage (hundreds of keV) electron • Plasma Kernel Mixed - The fusion core
beam is injected axially into a propellant flow. The elec- plasma may be intentionally seeded with
tron beam heats the flow to plasma temperatures, which non-reacting material, which gets heated
produces high specific impulse. Cool gas is injected along as part of the reaction. A certain per-
the chamber walls to provide film cooling and protect the centage is directed out of the core for
chamber from the very high temperature plasma. thrust, balanced by new fuel added to
Status: the core. The possible advantage of this
method is eliminating the mixing prob-
Variations: lem of adding mass after the reactor. Fu-
References: sion engines produce very high exhaust
120 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

velocities, so trying to mix more matter Sample Return Mission”, presented at Joint Con-
into the stream may require very large en- ference of 30th International Symposium on Space
gine components. Otherwise the stream Technology and Science, 34th International Electric
may simply be gone before it has a chance Propulsion Conference and 6th Nano-satellite Sym-
to mix. posium, July 4-10 2015

References: Further Reading

53 Antimatter-Heated Plasma Engine Alternate • Wikipedia: plasma propulsion engine

Names:
Type:
3.9 High Energy Particle Engines
Description: The exhaust of a pure antimatter rocket is
charged particles and gamma rays. This gives an extraor- Particle engines generally have very good performance
dinarily high exhaust velocity, but relatively low thrust. If because of the high velocity of the particles. The high
higher thrust is needed, hydrogen can be mixed with the velocity also requires a lot of energy, since particle ki-
plasma, at at the expense of performance. This likely re- netic energy goes as the square of the velocity.
quires a large magnetic bottle to contain the particles and
Hydrogen ions (protons) long enough to mix.
A. Particle Rockets
Status:
Variations: Particle rockets are distinguished by emitting particles
References: from an internal source, while the next section, External
Particle Interactions, covers methods that employ parti-
cles in the environment, either natural or man-made.
53 Electrodeless Lorentz Force (ELF) Thruster Al-
ternate Names:
54 Pulsed Fission Nuclear Thruster Alternate
Type: Names: Orion
Description: The ELF-160A thruster creates a high- Type: Atomic Particles by Nuclear Detonation
density, magnetized plasmoid known as a Field Re-
versed Configuration (FRC) employing a Rotating Mag- Description: A series of small atomic bombs yield de-
netic Field (RMF). The RMF driven azimuthal currents, bris/particles which push against a plate/shock absorber
coupled with the enhanced axial magnetic field gradient arrangement. The shock absorber evens out the explosion
produced by the FRC inside the flux preserving conical pulses to a steady acceleration for the vehicle. Issues with
thruster, produce a large axial JθxBr force that acceler-this method are not so much technical feasibility as risks
ates the plasmoid to high velocity. The ELF-160A is of using it near Earth and having a shipload of nuclear
completely electrodeless, the propellant is magnetically weapons, not to mention the radiation levels for crew and
isolated from the thruster body, quasi-neutral, and therecargo. Benefits are the enormous potential payload and
is zero contact between high temperature propellant and high acceleration. A possible application is moving dan-
the thruster. gerous asteroids. That would not take very many bombs,
and the pusher plate/shock absorber could be made from
Status: Active Development under Department of De- asteroidal metals, even from the asteroid you are moving.
fense contract.
Other applications might involve moving large amounts of
Variations: One variation is the ElectroMagnetic Plas- material in the outer Solar System, or starting interstellar
moid Thruster (EMPT), an electrodeless pulsed plasma trips. Far away from Earth the issue of radiation added
thruster that generates and accelerates a Field Reversed to the environment is not as significant, since solar and
Configuration (FRC) to produce thrust. cosmic radiation are already present in large amounts.
References: Status: Atomic bombs are well tested (unfortunately). A
demonstration of the pusher plate technology was done in
• Slough, J.; Kirtley, D.; Weber T. “Pulsed Plasmoid small scale with conventional explosives.
Propulsion: The ELF Thruster”, presented at the
Variations:
31st International Electric Propulsion Conference,
September 20-24 2009
54a Antimatter Catalyzed Pulsed Nuclear -
• Pancotti, A. P.; Little, J. M.; Neuhoff, J. S.; Cor- This speculative method uses antimatter trig-
nella, B. M.; Kirtley D. E.; Slough J. T. “Elecrode- gered implosion of plutonium targets. In the-
less Lorentz Force (ELF) Thruster for ISRU and ory this would lower the critical mass to ~ 2
3.9. HIGH ENERGY PARTICLE ENGINES 121

grams significantly lowering the size threshold 56 Alpha Particle Emitter Alternate Names:
for pulsed nuclear. Type: Atomic Particles by Radioactive Decay
54b Implosion Driver Pulsed Nuclear - Stan- Description: A radioactive element coats one side of
dard fission bombs are triggered by a chemi- a thin sheet which is capable of absorbing alpha parti-
cal implosion. This method substitutes a laser, cles. The particles emitted into the sheet are absorbed,
ion beam, or other external driver to collapse while the particles emitted in the opposite direction es-
the fuel load. It potentially reduces the critical cape, providing net thrust. Advantages are this is a simple
mass with a stronger collapse than chemicals device, and heating of the sheet can produce power as a
can achieve, or by using a different isotope. side effect via thermoelectric or heat engine. Disadvan-
tages are low thrust, and you cannot turn off radioactivity.
References: You would have to close two plates so they face each other
to neutralize thrust, and open them like a book to turn the
thrust on.
55 Inertial Confinement Nuclear Engine Alternate
Status: Alpha emitters are well known. Using them for
Names: Microfusion Engine
propulsion is theoretical so far.
Type: Atomic particles by Fission or Fusion Reactor
Variations: Choice of radioactive material allows a wide
Description: There are two major methods being pur- range thrust levels.
sued for fusion power. Magnetic confinement holds a hot
References:
plasma in a more or less steady state long enough for fu-
sion reactions to produce net power. Inertial methods use
lasers or particle beams to compress a fuel pellet to high 57 Fission Fragment Engine Alternate Names:
pressure and temperature such that the fusion reaction
Type: Atomic Particles by Fission Reactor
happens very fast, after which it rapidly expands. Both
methods can be applied to space propulsion. Magnetic Description: Thin wires containing fissionable material
confinement was addressed at Fusion Heated Plasma are arranged to allow the nuclear fragments from the fis-
Engine, and this method addresses Inertial confinement. sion to escape. They are allowed to decay naturally for
low thrust devices, or arranged in a nuclear critical ar-
A fuel pellet consists of a fusion core material, such as
rangement for high thrust. The fragments are aimed by
a Deuterium/Tritium mix, surrounded by a liner to opti-
electrostatic or electromagnetic fields to mostly go out the
mally absorb the laser or particle beam energy. Option-
back end of the thruster. The performance is very high
ally a fission shell surrounds the fusion core. This is sim-
because of the high speed of the fragments. The fission
ilar to the arrangement of a fusion atomic bomb. Alter-
decay tends to damage whatever the wires are made of,
nately particle beams can consist of fissionable heavy ions
so provision for replacing or reforming the wires would
or fusible light element ions, or even highly accelerated
be needed for long term use.
solid pellets of inert or reactive fuels. The end result in
all cases is a rapidly expanding cloud of highly energetic Status: Currently theoretical only.
particles. These are directed by a shaped magnetic field Variations:
or pusher plate to produce thrust. Thrust can be varied
by how often you generate the explosions.
• 57a Antimatter Sail - Antiprotons are
Status: Inertial confinement fusion is being researched directed at a sail coated with fissionable
for power generation. A chemical implosion is also the material such as Uranium. Their nega-
method by which most nuclear bombs are set off. tive charge allows them to be captured
Variations: Multiple variations are possible by combin- in the electron shell of the atoms, and
ing: then causes fission by annihilation with
one of the nuclear protons. One fission
fragment is absorbed by the sail, while
• Compression sources: laser, particle/ion beams, or the other moves away at several percent
solid pellets, of the speed of light, generating thrust.

• Type of central target: none in the case of colliding

References:
pellets, fission and/or fusion fuel ingredients,

• Type of pellet liner: none or what material, and • Dusty Plasma Based Fission Fragment Nuclear Re-
actor Clark and Sheldon, Joint Propulsion Confer-
• Thrust method: physical or magnetic nozzle ence, 2005.

References: 58 Pure Fusion Engine Alternate Names:

122 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

Type: Atomic Particles by Fusion Reactor compression, which heats the fuel to
Description: Various thermonuclear fusion reactors fusion ignition. After the fuel burn is
have been proposed. The results of a fusion reaction completed, the energy generated heats
are high energy particles which can, in principle, be the working fluid to high temperature,
harnessed for propulsion. This differs from 52 Fusion which then goes out a nozzle producing
Heated Plasma Engine in that no additional fuel is added thrust.
beyond the fusion reactants. Instead the high temperature • 58e Dense Plasma Focus - A high cur-
plasma or escaping charged particles are used directly as rent discharge in a radial arrangement
rocket exhaust. Since a break even fusion reaction has not causes the plasma to collapse to high
been achieved as of yet, this remains a theoretical con- temperature and pressure. This is be-
cept. ing developed by Lawrenceville Plasma
The actual method of containing the fusion plasma will Physics in Middlesex, New Jersey. This
influence the design of the rocket and its feasibility, in device has a reported fusion energy of
particular the mass of the device. A tokamak or stellara- 0.044 Joules vs capacitor energy of 50
tor plasma containment device would present the largest kJoules. The output vs input energy ra-
problems in siphoning off plasma for propulsion and their tio is a measure of how close the device
large weight would lead to an uninspiring thrust to mass is to practical operation.
ratio. Inertial confinement fusion using high power lasers • 58f Magnetized Target Fusion - Col-
or x-rays on small targets would probably be easier to liding plasma balls are further com-
build into a propulsion device, however the enormous size pressed by acoustic shock waves gener-
of the lasers and x-ray generators of current internal fu- ated by mechanical drivers. General Fu-
sion projects would lead to the same problem with the sion near Vancouver, Canada is research-
thrust to mass ratio. ing this method.
Status: Significant research is in progress for fusion in
general. Applications to space propulsion are theoretical. References:
Besides the large funding for magnetic and inertial con-
finement fusion, there are several alternate approaches
• Freeman, M. “Two Days to Mars with Fusion
that have lower funding levels.
Propulsion”, 21st Century Science and Technology,
Variations: vol 1, pp 26-31, Mar.-Apr. 1988.

• 58a Magnetic Confinement - Plasma in • Kammash, T.; Galbraith, D. L. “A Fusion-Driven

a chamber similar to a tokamak fusion Rocket Propulsion Scheme for Space Exploration”,
power reactor is intentionally leaked to a Trans. Am. Nucl. Soc. vol 54 pp 118-9, 1987.
magnetic nozzle.
• Mitchell, H. M.; Cooper, R. F.; Verga, R. L. “Con-
• 58b Inertial Confinement - The fuel
trolled Fusion for Space Propulsion. Report for
pellet is heated and compressed by lasers,
April 1961-June 1962”, US Air Force report num-
electron beam, or ion beam. After fus-
ber AD-408118/8/XAB, April, 1963.
ing, the resulting plasma is directed by a
magnetic nozzle.
• 58c Electrostatic Confinement - The Inertial Confinement
fusion fuel is confined by a spherical po-
tential well of order 100 kV. When the • Kammash, T.; Galbraith, D. L. “A Fusion Reactor
fuel reacts, the particles are ejected with for Space Applications”, Fusion Technology, v. 12
energy of order 2 MeV, so escape the po- no. 1 pp 11-21, July 1987.
tential well. The potential well is at the
focus of a paraboloidal shell, which re- • Orth, C. D. et al “Interplanetary Propulsion using In-
flects the fusion particles to the rear in a ertial Fusion”, report number UCRL-−95275-Rev.
narrow beam (20-30 degree width). 1: 4th Symposium on Space Nuclear Power Sys-
• 58d Plasma Mantle Confinement tems, Albequerque, New Mexico, 12 January 1987.
- The fusion fuel is contained in a
• Hyde, Roderick, “A Laser Fusion Rocket for Inter-
toroidal/poloidal current pattern, similar
planetary Propulsion” , LLNL report UCRL-88857.
to a Tokamak except all the currents are
Topics include:
in the plasma. The current pattern is
surrounded by a plasma sheath which iso-
lates the fuel from a surrounding working - Fusion Pellet design: Fuel selection. Energy
fluid. The fluid provides mechanical loss mechanisms. Pellet compression metrics.
3.9. HIGH ENERGY PARTICLE ENGINES 123

- Thrust Chamber: Magnetic nozzle. Shield- requires plasma pressures and stable confinement times
ing. Tritium breeding. Thermal modeling. Fu- larger than can be delivered by current approaches. If
sion Driver (lasers, particle beams, etc): Heat plasma pressures appropriate to burn times on the or-
rejection. der of milliseconds could be achieved in aneutronic fu-
- Vehicle Summary: Mass estimates. els, then high power densities and very compact, real-
tively clean burning engines for space and other special
- Vehicle Performance: Interstellar travel re- applications would be at hand. The PLASMAKª inno-
quired exhaust velocities at the limit of fusion’s vation will make this possible; its unique pressure effi-
capability. Interplanetary missions are lim- cient structure, exceptional stability, fluid-mechanically
ited by power/weight ratio. Trajectory mod- compressible Mantle and direct inductive MHD electric
eling. Typical mission profiles. References, power conversion advantages are described. Peak burn
including the 1978 report in JBIS, “Project densities of tens of megawats per cc give it compactness
Daedalus”, and several on ICF and driver tech- even in the multi-gigawatt electric output size. Engineer-
nology. ing advantages indicate a rapid development schedule at
very modest cost. (I strongly recommend that people take
• Bussard, Robert W., “Fusion as Electric Propul- this guy seriously. Bob Hirsch, the primary proponent of
sion”, Journal of Propulsion and Power, Vol. 6, No. the Tokamak, has recently declared Koloc’s PLASMAKª
5, Sept.-Oct. 1990. Abstract: Fusion rocket en- precursor, the spheromak, to be one of 3 promising fusion
gines are analyzed as electric propulsion systems, technologies that should be pursued rather than Tokamak.
with propulsion thrust- power-input-power ratio (the Aside from the preceeding appeal to authority, the PLAS-
thrust-power “gain” G(t)) much greater than unity. MAKª looks like it finally models ball-lightning with solid
Gain values of conventional (solar, fission) elec- MHD physics. -- Jim Bowery)
tric propulsion systems are always quite small (e.g.,
G(t)<0.8). With these, “high-thrust” interplane-
• Dense Plasma Focus
tary flight is not possible, because system acceler-
ation (a(t)) capabilities are always less than the lo-
cal gravitational acceleration. In contrast, gain val- • Lerner, Eric et al, Fusion Reactions from >150
ues 50-100 times higher are found for some fusion keV Ions, Physics of Plasmas, v 19 issue 3, March
concepts, which offer “high-thrust” flight capabil- 2012.
ity. One performance example shows a 53.3 day
(34.4 powered; 18.9 coast), one-way transit time
with 19% payload for a single-stage Earth/Mars ve- 59 Neutral Particle Beam Thruster Alternate
hicle. Another shows the potential for high acceler- Names:
ation (a(t)=0.55g(o)) flight in Earth/moon space.) Type:
Description: A high energy (order 50 MeV) particle ac-
Electrostatic Confinement
celerator generates a proton beam. This beam is neutral-
ized (combined with electrons to make neutral atoms),
• Bussard, Robert W., “The QED Engine Sys- then ejected. The exhaust is moving at a substantial frac-
tem: Direct Electric Fusion-Powered Systems tion of the speed of light, so performance is very high.
for Aerospace Flight Propulsion” by Robert W. This type of machine was explored under the US mis-
Bussard, EMC2-1190-03, available from En- sile defense program as a way of destroying missiles (with
ergy/Matter Conversion Corp., 9100 A. Center the beam). The energy required for space propulsion with
Street, Manassas, VA 22110. Summary: This is an this method exceeds the energy available from nuclear fu-
introduction to the application of Bussard’s version sion, so it only makes sense with antimatter or external
of the Farnsworth/Hirsch electrostatic confinement power sources such as a laser.
fusion technology to propulsion. 1500<Isp<5000
sec. Farnsworth/Hirsch demonstrated a 10**10 Status: Particle accelerators of this energy have existed
neutron flux with their device back in 1969. since the mid-20th century. What limits this method is
lack of a power supply, so it remains untested.
• Plasma Mantle Confinement Variations:

• Koloc, Paul M., “PLASMAKtm Star Power for En- • 59a Near Lightspeed Probe - If the particle ac-
ergy Intensive Space Applications”, Eighth ANS celerator produces very high energy protons, such
Topical Meeting on Technology of Fusion Energy, that relativistic mass increase is significant, and the
Fusion Technology , March 1989. power source is a very powerful laser located at the
origin star, and focused using the star as a gravita-
This note is 20 years old and needs updating: Aneu- tional lens, the rocket equation no longer constrains
tronic energy (fusion with little or negligible neutron flux) the final velocity. Velocities close to the speed of
124 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

light would be possible, and stopping would be pos- Description: The magsail operates by placing a large
sible by pointing the accelerator in the opposite di- superconducting loop in the solar wind or planetary ion
rection. Acceleration in this case is limited by the stream. The current loop produces a magnetic field that
power to mass ratio of the power conversion and par- deflects the ions, producing a reaction force. Because of
ticle accelerator hardware. the large area covered by the loop, which is mostly empty,
it can develop a relatively large force for a given mass. It
References: is limited in direction and strength by the local solar wind
or ion flow.

60 Antimatter Annihilation Alternate Names: Status: Magnetic field interaction with an ion stream is
well understood. Practicality of this method depends on
Type: an efficient enough design. Has not been tested as of
Description: Atoms and anti-atoms (or their constituent 2012.
particles) annihilate, producing pions, then muons, then Variations:
gamma rays. The charged particles can be acted upon by
a magnetic nozzle. The gamma rays can be absorbed by References:
a container, and the resulting heat used to supply power.
Antimatter provides the highest theoretical energy fuel
(100% matter to energy conversion), although the over- 62 External Particle Beam Alternate Names:
head involved with storing antimatter may reduce the Type:
practical efficiency to a level comparable to other propul-
Description: A remote particle beam source aims it at
sion methods.
a target vehicle. The particles are either absorbed or re-
Status: Small amounts of antimatter have been created flected to generate thrust directly, or their kinetic energy
and temporarily stored at particle accelerator labs. At is used as a power source for some other type of propul-
present, antimatter is not thought to exist naturally in sion. The major issue is keeping the beam narrow enough
large quantities. to be a useful energy transfer method.
Variations: Status: Untested as of 2012
Variations:
• Wikipedia article: Redshift Rocket
References:
References:

63 Interstellar Ramjet Alternate Names: Bussard

• Forward, Dr. Robert L. Antiproton Annihi-
Ramjet
lation Propulsion, AFRPL TR-85-034 from
the Air Force Rocket Propulsion Laboratory Type:
(AFRPL/XRX, Stop 24, Edwards Air Force Base, Description: A large funnel or inlet is used to compress
CA 93523-5000). NTIS AD-A160 734/0 Note: interstellar gas, which is then fed to a fusion reactor for
This is a technical study on making, holding, propulsion. Because of the low density of the interstellar
and using antimatter for near-term (30-50 years) medium, an extraordinarily large scoop is required to get
propulsion systems. Excellent bibliography. For- any useful thrust. Performance is limited by the exhaust
ward is the best-known proponent of antimatter. velocity of the fusion reaction to a few percent of the
This also may be available as UDR-TR-85-55 speed of light. In other words, collecting the gas causes
from the contractor, the University of Dayton drag, and at the exhaust velocity of the reactor there will
Research Institute, and DTIC AD-A160 from the be no net thrust. This makes the system an efficient de-
Defense Technical Information Center, Defense celerator by pointing the exhaust forward. In that case
Logistics Agency, Cameron Station, Alexandria, the drag and reverse thrust both act to slow the vehicle.
VA 22304-6145. And it’s also available from the Variable density of interstellar gas affects the viability of
NTIS, with yet another number. this method. By running the reactor partly off of internal
• G. D. Nordley, Application of Antimatter - Elec- fuel, this type of vehicle can be brought to high enough
tric Power to Interstellar Propulsion, Journal of velocity for the collector to start to function, and then to
the British Interplanetary Society, June 1990. go to higher velocities. By having the reactor use less fuel
than collected, it can be self-refueling.
Status: Although the concept dates back to 1960, it is
B. External Particle Interactions
untested due to a lack of practical fusion reactors.
61 Magsail Alternate Names: Variations:
Type: References:
3.10. PHOTON ENGINES 125

[D77] R. W. Bussard, “Galactic Matter and Interstellar than fuel, but that belongs in the Engineering Methods
Flight”, Astronautica Acta 6 (1960): 179 - 194. section.
[D78] A. R. Martin, “The Effects of Drag on Relativistic References:
Spacefight”, JBIS 25 (1972):643-652
[D79] N. H. Langston, “The Erosion of Interstellar Drag Further reading
Screens”, JBIS 26 (1973): 481-484.
[D80] D.P. Whitmire, “Relativistic Spaceflight and the • “So You Wanna Build A Rocket?" at Project Rho
Catalytic Nuclear Ramjet”, Acta Astronautica 2 (1975): gives quick estimates for a variety of atomic rockets
497 - 509. -- including the Bussard Ramjet, the Nuclear Light-
bulb, the Nuclear Salt Water Rocket, the magsail,
[D80] C. Powell, “Flight Dynamics of the Ram-
etc. -- and related engineering challenges -- includ-
Augmented Interstellar Rocket”, JBIS 28 (1975):553-
ing radiation sheilding, space suits, etc.
562
[D81] D.P. Whitmire and A.A. Jackson, “Laser Powered • “The Relativistic Rocket” in the Physics FAQ de-
Interstellar Ramjet”, JBIS 30 (1977):223 - 226. scribes some of the relativistic effects of traveling
close to the speed of light, and one way to calculate
[D82] G. L. Matloff and A. J. Fennelly, “Interstel- “How much fuel is needed?".
lar Applications and Limitations of Several Electro-
static/Electromagnetic Ion Collection Techniques”, JBIS
30 (1977):213-222
3.10 Photon Engines
64 Interstellar Scramjet Alternate Names: There are a variety of propulsion techniques that use ei-
Type: ther internally generated or externally supplied photons
for propulsion. Stars, of course, are a natural supply of
Description: Similar to the interstellar ramjet, the in-
photons, but a purpose-made light source of sufficient
terstellar medium is compressed to fusion density and
beam intensity can also be used. Photons, being mass-
temperature. In this concept it is only compressed lat-
less, travel at the maximum possible speed. Thus they
erally, then re-expanded against a nozzle. Incredible ve-
are theoretically the highest possible exhaust velocity. By
hicle sizes and lengths are required to reach fusion con-
the same token, being massless, they do not transfer much
ditions, but since lateral compression does not cause as
momentum.
much drag as full compression, it is not limited by the ex-
haust velocity. Thus speed may reach a substantial frac-
tion of the speed of light. A. Photon Sails
Status:
As noted above, light does not transfer much momen-
Variations:
tum. So the following methods reach useful thrust level
References: by maximizing area to mass ratio. The large area in-
creases thrust relative to mass. The resulting structure
resembles a terrestrial sail.
65 External Fuel Supply Alternate Names:
Type:
66 Solar Sail Alternate Names: Lightsail
Description: Any propulsion system which stores all it’s
fuel at the start has an exponential fuel requirement as a Type: Solar Flux by Photon Deflection
function of velocity. If added fuel is delivered by a par- Description: Starlight, for example from the Sun, re-
ticle or pellet stream, either from behind or positioned flecting off a large area sail produces a force because the
ahead, that exponential requirement is turned into a lin- momentum of the photons is reversed. The force for per-
ear one. This is because at each point the vehicle is only pendicular (normal) reflection is
accelerating itself (plus a small amount of fuel), instead
of the vehicle plus all the fuel for the entire mission. In
(1+r)(E/c) ,
the latter case you are using fuel early to accelerate fuel
for later, which causes the exponential overhead.
where r is the reflectivity of the sail, E is the incident light
Status: power, and c is the speed of light. Thus even perfectly
Variations: Lots of variations are possible in terms of absorbing surfaces see light pressure, but reflecting ones
what fuel is delivered, and if it is formed into a particle perform up to twice as well. No real surface is perfectly
stream, discrete pellets, or container tanks of fuel. Inci- reflecting, so analysis of sails needs to take into account
dentally this can also be used for other types of supplies the actual optical properties. At sail angles other than
126 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

perpendicular to the light source the net force has a di- • Marchal, C., Solar Sails and the ARSAT Satel-
rect component from the incident light, and a reflected lite - Scientific Applications and Techniques,
or diffuse component depending on the surface proper- L'Aeronautique et L'Astronautique, no 127, pp 53-
ties of the sail. For reflection the incident and reflected 7, 1987.
angles are the same amount relative to the plane of reflec-
tion. The diffuse component is light scattered by a rough • Friedman, Louis, Starsailing: Solar Sails and In-
surface into a wide range of angles. The force is reduces terstellar Travel , Wiley, New York, 1988, 146 pp.
by a cosine factor due to the spread of directions.
The absorbed component of light is eventually re-radiated 67 Laser Lightsail Alternate Names:
by blackbody radiation. If this happens equally from both Type: Laser by Photon Deflection
sides of the sail, it produces no net thrust. Coatings can
modify the emissivity of each side and therefore give a Description: A high powered laser is aimed at a target
net thrust. For high performance sails coatings add too sail. The beam of photons are reflected off the sail ma-
much mass. It ends up being better to use bare metal foil terial. Reflection of the photons reverses their momen-
of high reflectivity and just ignore the small part not re- tum vectors’ component which is normal to the sail. By
flected. Thickness so low as to be partially transparent, conservation law, the sail gains momentum. Laser sails
and micro-perforations smaller than typical light wave- can have higher performance than solar sails because the
lengths can lower weight even further without signifi- laser beam intensity is not as limited as the local star light,
cantly reducing reflectivity. Manufacturing, installation, and can be focused over longer distances. The sail can be
and operation with extremely thin foil sails will be a chal- designed to optimize reflectivity at the laser wavelength.
lenge. The best mass-to-reflectivity material for visible This method is still by the overheating of the sail and lack
light is an Aluminum/Magnesium alloy. Despite higher of sufficiently powerful lasers for useful missions.
mass, refractory metal foils like Tungsten can be used at Advanced interstellar mission concepts have proposed
much higher temperatures. Therefore they can be used very large phase plate type lenses for long range focus.
much closer to stars where light intensity is higher, and Another advanced concept is to use two sails to slow down
develop more thrust. This can be used for initial kick to a vehicle, by releasing the first sail and using that to reflect
outer system missions. The sail shape can be maintained light back to the second, smaller one.
by struts and tension wires, or by the balance of rotation
and light pressure, or some combination. Status: Not tested as of 2012 mainly due to lack of pow-
erful lasers. Megawatt class lasers are not powerful in this
At the distance of the Earth from the Sun, the incident context.
power is 1370 MW per square kilometer. This produces
about 8 Newtons/square kilometer for high-reflectivity Variations:
sails. The attractiveness of solar sails is they use no fuel, References:
and in principle the reflector can be made of local met-
als in the case of asteroids. The disadvantages are low • Forward, Robert L., Roundtrip Interstellar
thrust unless you build them very large, limited angular Travel Using Laser-Pushed Lightsails Journal
control, and inverse square dependence of thrust with dis- of Spacecraft and Rockets , vol. 21, pp. 187-95,
tance from the source star. Jan.-Feb. 1984
Status: As of 2012 a few test flights of sails have been
attempted, with varying success. Any object exposed to • “LightSail Project Documents”
sunlight will see a light pressure force, but most spacecraft • “Laser-pushed Lightsail”
do not have a low enough mass per area for the force to
be more than a minor correction to it’s motion. • Fresnel plate lens for spacecraft propulsion. Gregory
Variations: L. Matloff. “Deep Space Probes: To the Outer Solar
System and Beyond”. Section 7.5 The Fresnel Lens:
A method of improving laser collimation. p. 103.
66a Gravity Tractor - In this variation, the
sail is not attached to a cargo such as an aster- • “Laser Technologies for Starflight”; “Key Issues for
oid you are trying to move. The mutual gravity Interstellar Sails”.
of the objects anchors the sail. Reasons to do
this are not having to devise methods to attach
the sail, and allowing the sail to maneuver itself 68 Microwave Sail Alternate Names: Starwisp
with less restrictions. Type: Beamed Microwave by Photon Deflection
Description: In this method microwaves are reflected off
References: a very thin, open wire mesh. The momentum change
of microwave photons bouncing off the mesh provides
• Wikipedia article - Solar sail thrust. Because an open mesh of thin wires can have a
3.10. PHOTON ENGINES 127

very low weight, in theory this propulsion method can specific thickness of sail will be balanced at any distance.
give high accelerations. With feasible power levels, the The ratio of the star’s luminosity to it’s mass will deter-
cargo mass will still be small. A good use for such would mine the thickness. The star’s spectrum and desired sail
be delivering nanofactories, which can then build larger distance will determine what to make the sail out of. By
scale infrastructure. The beam can supply power to op- directing the light in one particular direction rather than
erate the factory at the destination. This is related to the symmetric in all directions as natural stars do, the sails
non-transport energy delivery by microwave concept. convert the star into an unbalanced light emitter, and so
Status: Generating lots of microwave power is well un- create a net thrust for the combined star + sail combina-
tion.
derstood. Keeping it focused over useful distances in
space, and building the lightweight sail material are not. Sails with a slight dihedral angle oriented perpendicular
Variations: to the star will direct the light to miss the star itself, and
will be stable in orientation. Flat sails which do not reflect
References: the light directly back to the star are slightly more efficient
since none will be absorbed by the star, but will accelerate
• Wikipedia article - Starwisp themselves sideways. So that type of sail must change it’s
orientation periodically to maintain position. Sails which
B. Photon Rockets are not flat, such as a shallow cone, can direct the light to
miss the star without drifting.
This group of methods involve emitting photons from in- To give a numerical example, a sail at the Earth’s distance
side the vehicle, rather than reflecting photons from an from the Sun produces about 8 newtons/km^2. Solar ac-
external source. Like photon reflection, photons do not celeration at that distance is 0.006 m/s^2. Therefore if
carry much momentum, so it is a low thrust set of con- your sail mass is 1330 kg/km^2, light pressure will bal-
cepts, although the exhaust velocity is the maximum pos- ance gravity. If made of aluminum, then it needs to be
sible. It can be considered an addition to other propulsion 0.5 microns thick. Foils that thick are commercially sold
methods by directional emission of radiator waste heat. today. A sail cloud reflecting 10% of the Sun’s output
will have an area of 28 x 10^15 km^2 and a mass of 37.5
69 Thermal Photon Reflector Alternate Names: x 10^15 tons (3.7% of the mass of the largest asteroid, 1
Nuclear Photonic Rocket Ceres), Such a sail cloud would produce a thrust of 225
x 10^15 Newtons, and accelerate the Sun by 3.5 meters
Type: Nuclear Reactor via Photon Emission per second per million years.
Description: A heat generating device, such as a nu- Uses for this type of engine are adjusting orbits of binary
clear reactor, is at the focus of a parabolic reflector. The or multiple star systems, escaping future supernovae, and
thermal photons are focused into a near parallel beam, generally moving stars within a galaxy or between galax-
which propels the vehicle. Another high-energy source is ies. The physics of this method are simple. The challenge
a matter-antimatter reaction, which is absorbed by a blan- is one of scale, and the extremely slow accelerations it
ket of heavy metals and converted to heat. This method is produces because stars are very massive. Note that mov-
probably not practical, since emitting photons gives less ing a star does not generally bring along objects in orbit
momentum than emitting the reaction products of the en- about the star, they would need their own propulsion.
ergy source directly. It might be worth using directional
radiators to get rid of surplus heat in addition to the reac-
Moving Planets - A more practical (relatively)
tion products. In that case it increases the total efficiency.
use for this method is moving planets, because
Status: Not tested as of 2012 of their much lower mass. The sails would
Variations: be anchored by the planet’s rather than the
star’s gravity. A larger secondary cloud of sails
References: would direct more sunlight to the anchored
sails than they would collect on their own. A
• Wikipedia Article: Nuclear photonic rocket reason to do this for the Earth is the Sun is in-
creasing it’s brightness by 10% per billion years
70 Stellar Photon Engine Alternate Names: due to increasing fusion rate in the core. Thus
Stardrive if you want to keep the Earth habitable, you
would want to move it slowly outwards. Al-
Type: Solar Flux by Photon Emission ternately, if you wanted to make Venus more
Description: This is similar to the previous Thermal habitable, you could move it into the Aster-
Photon Reflector method, except an entire star is at the oid belt, where the excess greenhouse effect
focus of a cloud of lightsails which are balanced by grav- would be an advantage rather than a detriment.
ity vs light pressure. Since both gravity and light inten- This would still likely be a slow method. Using
sity vary as the inverse square of distance from the star, a gravity assist maneuvers with large asteroids or
128 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

outer Solar System bodies would be faster. Un- • Young K. Bae Corporation, List of Articles, 2006-
like gravity assist for spacecraft, in this case the 2012. The paper on Ultrahigh Precision Forma-
object doing the flyby is massive enough to af- tion Flying has a good illustration of the concept.
fect the orbit of the planet. If a secondary body
such as a gas giant is used to absorb the orbit
changes, the flyby object could be used multi- 3.11 External Interaction Methods
ple times to shift the small planet without ex-
pending too much energy.
External interactions are transport methods which create
a force via fields or physical interaction with objects out-
Status: Very far from testing. Possibly not even used in side the vehicle itself.
science fiction stories.
Variations:
3.11.1 Magnetic Interactions
References:
72 Ionospheric Current Engine Alternate Names:
Electrodynamic Engine, Alfven Engine
71 Gamma Ray Thruster Alternate Names: redshift
Type: Electrical via Magnetic Field
rocket
Description: A current-carrying wire in a planetary
Type: Antimatter by Photon Emission
magnetic field will experience an IxB force (current cross
Description: Gamma rays produced by antimatter an- product with magnetic field, which is perpendicular to
nihilation behind a vehicle can be absorbed by a thick both). This is the same type of force created in elec-
layer of heavy metals. The momentum of the gamma ray tric motors. Currents which only flow in one direction
photons produces thrust. Like other photon engines, it is cannot be sustained, electric charges would accumulate
likely more effective to use the decay particles directly, at one end. Therefore for this method the current loop is
since they have more mass and can be directed better, closed through an Ionosphere, the portion of a body’s
but using the gamma rays as a supplement makes sense. atmosphere which is ionized and can carry a current.
Status: Currently theoretical, as antimatter is only pro- The wire accelerates in one direction (pulling a vehicle
duced in tiny amounts in particle accelerators. along), and the ionosphere accelerates in the other direc-
tion. The ionosphere motion eventually dissipates within
Variations: itself. When current is driven through the wire by a
References: power supply, this functions as a motor, generating posi-
tive thrust. When current is allowed to flow in the other
direction unopposed, it functions as a generator, produc-
C. Amplified Photon Pressure ing power which can be used on the vehicle, and creating
drag. Both power and drag can be useful in come circum-
In this approach, a reflecting cavity is used to multiply the stances, but any power generated will be at the expense
photon pressure. In one arrangement, a laser emits a par- of orbital kinetic energy.
ticular frequency, and the walls of an enclosing cavity and A current loop thruster produces more thrust/watt than
the vehicle back end are designed to have a high reflec- an ion or plasma engine. No propellant is consumed di-
tivity for that frequency. In another arrangement, there is rectly, although some material is consumed by a Plasma
no enclosing cavity, and a laser gain medium is placed in Contactor to produce a plasma that enables good elec-
front of one mirror. The vehicle carries a second mirror, trical contact with the ionosphere. Effectively this gives
and the pair of mirrors plus gain medium form a laser am- an equivalent exhaust velocity of 250 km/s. Advantages
plifier which happens to have a large vacuum gap between are this is a relatively simple device, and relatively high
the energy source and vehicle. In both cases the light will thrust for an electric thruster. Limitations of this method
bounce multiple times until finally absorbed or dispersed, are the planet must have both a substantial magnetic field
each bounce adding a momentum increment. The wave- and sufficient ion density to operate effectively, and the
length can be adapted to the physical dimensions, for ex- direction of thrust is governed by the local magnetic field
ample using microwaves instead of a laser if it creates direction. Planetary magnetic fields are often offset and
an efficient reflection. The multiple reflections will ex- tilted with respect to their poles, and in any case only
tract energy from the photons as they repeatedly bounce point in one direction at any given point in orbit. The
off the moving vehicle by redshift. If sufficient bounces thrust direction is restricted to a circle perpendicular to
at sufficient projectile velocity happen, the light would the field direction. Earth orbit up to around 1000 km,
be redshifted to a wavelength not efficiently reflected, or and some areas around Jupiter are good candidates, other
merely lose the majority of their energy, thus ending the places less so. Uses include orbit maintenance for low or-
momentum transfer. bit vehicles and Skyhooks, and apogee raising for reach-
References: ing higher orbits. This method may be combined with
3.11. EXTERNAL INTERACTION METHODS 129

other electric thrusters and share the same power source, to a direct transfer orbit, and needing to select mission
changing which thrust method is used as appropriate for dates according to when the planets are in the right po-
altitude or thrust direction. sitions. Lunar gravity assist to escape from or return to
The motion of a long wire in a magnetic field will generate Earth orbit is particularly useful, as the Moon’s orbit is
a static voltage even when not used for propulsion. Such short enough to give frequent opportunities.
voltage build-ups happen to all spacecraft, a topic known Status: Used extensively in planetary missions, often
as Spacecraft Charging, but are much larger with a long multiple times for a single spacecraft.
conducting wire. Therefore proper insulation and equip- Variations:
ment grounding is needed to avoid damage. Even without
the plasma contactor running, spark discharges to the sur- References:
rounding ionosphere can happen.
Status: An experiment called the Tethered Satellite Sys-
74 Gravity Counterweight Alternate Names:
tem was flown on the Space Shuttle in the 1980’s.
Dumb-Waiter System
Variations:
Type: Potential Energy by Gravity Field
References:
Description: In this method mass falling down a gravity
well can be an energy source to power payloads going up
• Belcher, J. W. “The Jupiter-Io Connection: an
the gravity well. Most typically this would be via a space
Alfven Engine in Space”, Science vol 238 no 4824
elevator, using the falling mass to directly lift a cargo via
pp 170-6, 9 Oct 1987.
cable, or to generate power to lift the cargo electrically.
This is most efficient when cargo is going in both direc-
3.11.2 Gravity Interactions tions and are the same mass. In that case cargo delivery
only consumes the inefficiency of the motors, which can
73 Gravity Assist Alternate Names: Planetary Flyby, be just a few percent. Another variation is to use brak-
Celestial Billiards ing energy of cargo above synchronous altitude, which
sees a centrifugal force higher than gravity, to power lift-
Type: Kinetic Energy by Gravity Field
ing cargo from the body surface to synchronous altitude.
Description: Momentum exchange between a planetary The energy for this comes at the expense of rotation of
body or large satellite and a vehicle allows changing the the body by slowing it down. Braking will induce side-
vehicle’s direction and velocity in other reference frames. ways forces on the elevator.
When considering just the vehicle and large body, a hy-
Status: Counterweights are commonly used in elevators
perbolic orbit has the same velocity on approach and de-
on Earth. Space applications have not been tried.
parture, but gravity forces change the direction. When
the large body is in turn in orbit around another larger Variations:
planet or star, the change in vehicle direction can increase References:
or decrease it’s velocity relative to the larger velocity. By
conservation of momentum, such gravity assist flybys af-
fect the object you fly past, but typically the vehicle is so
3.11.3 Aerodynamic Interactions
much smaller than the body it is flying past, so the change
in the body’s orbit is too small to measure.
Interaction with the atmosphere of large bodies which
The amount of velocity change is theoretically limited to have one, such as the Earth, can produce significant
twice the escape velocity from the body, by changing the forces. These forces are distinct from propulsion using an
vehicle direction 180 degrees. Thus larger bodies can atmosphere which was discussed under Air-Breathing
produce larger velocity changes. As a practical matter Engines, which are principally thrust forces.
gravity assist usually results in significantly less than this
because the vehicle arrives with excess velocity and the
desired final direction limits the flyby parameters. Excess 75 Aerodynamic Forces Alternate Names: Aero-
velocity reduces the time the gravity of the body can act. brake, Airfoils
More typical values are 0.5 to 1.0 times escape velocity
measured from 1.5 body radii from it’s center. The veloc- Type: Kinetic Energy by Aerodynamic Forces
ity changes are still large enough that they are extensively Description: This method uses aerodynamic interac-
used in planetary exploration missions, often using mul- tions with an atmosphere to provide forces perpendicu-
tiple flybys of different planets, or even the same planet lar to forward motion (lift), intentionally oppose forward
multiple times. The fuel cost of lining up for a gravity motion (drag) or change orbit parameters (using lift, with
assist in these cases is much lower than doing the equiva- some drag). Buoyancy, or lift by displacement without
lent velocity change directly. The disadvantage is longer forward motion, is covered by method 3 Aerostat. There
mission times consumed by the gravity assists compared is a wide range of conditions and applications for aerody-
130 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

namic forces and it is a well developed field of engineer- References:

ing.
The range of velocities is from well below the local speed 3.11.4 Mechanical Interactions
of sound, or Subsonic, to many times the speed of sound,
or Hypersonic, even beyond escape velocities. All bod- This category involves direct physical contact with a nat-
ies which are not symmetric or pointing directly parallel ural or human-made body in space.
with the velocity direction will generate some lift, and all
moving bodies generate drag. The ratio of lift to drag,
known as L/D Ratio ranges from less than 1.0 for blunt 76 Rheobrake Alternate Names: Lithobrake, Crash-
re-entry bodies to 25 or more for good subsonic Airfoil portation
shapes designed to maximize lift. Structures whose main Type: Lower Kinetic Energy by Friction
purpose is to generate lift are called Wings. All lifting
bodies induce some drag by virtue of the lift force not Description: This methods uses mechanical friction
being entirely perpendicular, as well as body drag from against a planetary surface to slow down. For example,
other parts of the vehicle besides the wings. imagine a rail made of cast basalt on the lunar surface. It
is laid level to the ground, and is shaped like the concrete
Lift forces can be used to gain altitude as part of a climb highway dividers. A vehicle wanting to land is in a low
to orbit. Drag forces would be minimized as they coun- grazing orbit. It aligns with the rail, just above it, then
teract thrust to reach orbital velocity. On return from extends some clamps over the rail. By applying clamp-
orbit, purposely designed devices to slow down include ing pressure, the vehicle can brake from lunar orbit to a
parachutes, heat shields, and High Q Aerobrakes. While stop. Obviously the brake will be dissipating a lot of heat,
already in orbit, pure drag generating Low Q aerobraking and will therefore have to be made of high temperature
can be done without special devices as long as the drag material such as graphite. Another approach is to have a
and heating are within what the structure can withstand. 'runway' which is a smoothed area on the lunar surface.
Orbital speed lift devices can change orbit direction, at The arriving vehicle slows down to below orbital speed,
the cost of some drag loss. then gravity pulls it down to the runway, and friction with
Status: Powered aircraft have operated for over 100 skids on the bottom of the vehicle slows it down.
years. Jet aircraft have been used as the carriers for rock- In order to not melt the brake, it should maximize surface
ets. They use a combination of 25 Fanjet for forward area and possibly have cooling systems like heat pipes.
thrust and wings for lift. The rail will not see such concentrated heating, so cool-
Variations: ing is not as much of a challenge. The main advantage is
not requiring fuel to land on a body. The main disadvan-
tage is the large size and mechanical accuracy required for
• 75a Parachute - Relatively low Mach
the landing rail or runway. When the velocity is greater
number drag device to come to a com-
than the speed of sound in the brake materials, irregular-
plete stop or low enough velocity for a
ities in the landing surface can create shock waves in the
terminal landing device.
device. This method is easier to implement on smaller
• 75b High Q Aerobrake - Q is dynamic bodies where the amount of orbital kinetic energy to dis-
pressure, the pressure caused by motion sipate is less.
through an atmosphere. High Q aero-
Status: Mechanical brakes are used in many vehicles on
brakes generate a lot of drag at high Mach
Earth. High performance ones are used in passenger air-
numbers. They are often designed as
craft. Use for space transport is theoretical at this point.
inflatable devices, called Ballutes (from
balloon and parachute), but extendable Variations:
flaps or panels would also fall into this
category. • 76a Rock Cloud - This involves creating
• 75c Heat Shields - They are drag de- an artificial 'atmosphere' of particles to
vices integrated into a vehicle structure, slow down against. A cloud of lunar dust
but primarily designed to dissipate the could be raised by electrostatic forces and
extreme heat of re-entry. an arriving vehicle slows by impact of the
dust particles, or deflection by charged
• 75d Low Q Aerobrake - These are of- surfaces on the vehicle.
ten re-purposed existing parts of a vehi-
cle which are used for braking. By keep- References:
ing the braking drag low, and using mul-
tiple orbits, often the braking can be done
without special components on the space- 76a Space Harpoon Description: In 2015, NASA
craft. Obviously that will take longer. proposed using a harpoon and cable to hook to a pass-
3.12. THEORETICAL METHODS 131

ing object, and use braking of the cable to accelerate and Description: Antigravity is the reduction or opposition
match velocity. Redeploying the cable and using the ob- to the normal force of gravity, which is attractive under
jects spin to sling the vehicle would add additional veloc- most conditions. One method of producing it would be
ity. with a negative mass. If such existed, the formula for
gravitational force would produce a repulsion rather than
attraction. No material with negative mass is known to
3.12 Theoretical Methods exist. By Einstein’s mass-energy relation ( E=mc2 ), neg-
ative mass would also represent negative energy. Other
methods of producing repulsion have been proposed, but
The previous sections included transport methods for
suffer a similar lack of observable support except for one
which there is a scientific or engineering consensus they
item - Dark Energy. The Universe as a whole appears
are possible to build, even if not built yet. For complete-
to be expanding at an accelerating rate. The cause of this
ness, this section includes theoretical transport methods
is hypothesized to be a cosmological constant, a pres-
for which: a consensus does not exist, there is no known
sure that exists throughout the Universe tending to ex-
method to implement it, or even contradicts established
pand spacetime. Since there is no known way to change
physics. They are sorted alphabetically, since there is no
the pressure caused by Dark Energy, which is distributed
reliable way to rank or organize these methods. Although
evenly in all directions, it is not useful as a transport
there is not a firm technical basis for these methods, to
method.
be listed here, they need at least some theoretical support
in the form of published papers or other documentation. Status: Theoretical as of 2012
Ideas which only appear in fictional works or have no the- Variations
oretical support can be found in Appendix 1: Fictional
Methods References

77 Alternate Spacetime Alternate Names: - Sub- 79 Modified Newtonian Dynamics (MOND) Alter-
space, Hyperspace, Alternate Dimensions nate Names:
Type: Theoretical Type: Theoretical
Description: - Spacetime in relativity theory is the 4 di- Description: This method assumes some violation of
mensional environment of three physical dimensions and Newton’s laws of motion are possible. Either an action
one time dimension. Travel in ordinary spacetime is lim- without an equal and opposite reaction, which produces
ited to the speed of light as far as we know. This method a reactionless thruster, or higher order terms in the
uses the idea that there is some other spacetime which can motion equations that would allow an unbalanced force.
be reached from ours. If it has different properties than While such formulas are easy to write, they do not have
ours, it could allow transport or communication faster or support from actual observations. A resonant extraction
more efficiently. If an alternate spacetime is in relative of Casimir forces from the quantum background has been
motion to ours, which is not constrained by the speed of proposed as a way to produce thrust. While the Casimir
light, then rapid travel could be possible by translating force is well observed, using it in a way that generates
to it, and then translating back at the destination. There reactionless thrust is not.
are theories that our Universe actually consists of more Status: Theoretical as of 2012
than 4 dimensions, such as string or M-theory, but the
other dimensions are compacted to the quantum scale, or Variations
unreachable. As of 2012, there is no evidence for these References
theories, although a considerable number of scientific pa-
pers have been written on the subject, and some searches
for observable effects are underway. 80 Quantum Black Hole Engine Alternate Names:
Status: Theoretical as of 2012 Type: Theoretical
Variations Description: In theory, all black holes will emit particles
as if it were a black body of a certain temperature. This
References
is known as Hawking Radiation after physicist Stephen
Drosher and Houser, Space Propulsion Device Based Hawking, who first described it. The temperature varies
on Heim’s Quantum Theory, 2004. AIAA Paper 2004- inversely with the size of the event horizon, so smaller
3700. Assumes an extension of General Relativity using black holes are hotter and emit higher energy particles.
quantized higher dimensional space. The emission of a black body changes as the 4th power of
temperature, while the area of an event horizon changes
more slowly. Therefore small black holes emit more en-
78 Antigravity Alternate Names: ergy, and ones small enough to emit useful amounts of en-
Type: Theoretical ergy are themselves particle sized, and thus called Quan-
132 CHAPTER 3. PART 2 - SPACE TRANSPORT METHODS

tum Black Holes. If new matter is added to the black Gravity Assist, where you can make use of the difference
hole at a rate sufficient to offset the emission losses, ef- in motion between two large objects.
fectively 100% conversion of matter to energy can be Status: Theoretical as of 2012.
achieved. The particles or gamma rays thus emitted are
directed for thrust or used for power generation. Black Variations
holes, quantum or otherwise, are very massive, so the References
utility of such for propulsion is questionable for anything
smaller than an asteroid sized spaceship.
83 Wormholes Alternate Names:
Although there is a good amount of theory about quan-
tum black holes, there is not a consensus that they actually Type: Theoretical
exist beyond theory. The difficulty is in how to form sub- Description: A Wormhole is a hypothetical region
stellar mass holes. Holes can be manipulated by adding a of spacetime shaped to connect two distant points. If
net charge and then using electrostatic or magnetic fields.
the connection is shorter than the non-wormhole path,
Status: Stellar and larger mass highly condensed objects traversing it would save time. Creation of wormholes in
have been observed, and are presumed to be black holes. theoretical papers usually involves black holes or Exotic
Quantum mass black holes have not been observed. Matter, matter with unusual properties such as negative
mass. While many such papers have been written about
Variations: wormholes, it is not known if the theory matches reality.
References: Therefore we do not know whether wormholes are possi-
ble or what their properties might be.
• Wikipedia article: Hawking Radiation Status: Theoretical as of 2012.
Variations
81 Tachyons Alternate Names: References
Type: Theoretical
• Kanti, Panagiota et. al. Stable Lorentzian Worm-
Description: Tachyons are hypothesized particles which holes in Dilatonic Einstein-Gauss-Bonnet Theory
travel faster than the speed of light. They would either al- posted at arXiv.org 6 Dec 2011.
low higher exhaust velocity for an engine, or by some sort
of conversion, possibly by quantum tunneling, convert an
entire vehicle into tachyons so it would travel faster than 3.13 Comparisons Among Meth-
light. Some searches for tachyons have been made, but
they have not been observed in nature. ods
Status: Theoretical as of 2012.
Transport methods can be sorted by performance, status,
Variations and cost measures. The following sections describe the
References measures and make a first attempt to rank the various
methods. Technology is constantly progressing, so any
selection of methods needs to be updated with the most
82 Warped Space Alternate Names: current data, or projections for future planning purposes.
Type: Theoretical
Description: Travel through spacetime is restricted by 3.13.1 Comparisons Among Methods
current theory to the speed of light. Spacetime itself is
not limited in this way, and in fact current Inflationary Performance Measures
Cosmology theory assumes a faster than light expansion
in the early history of the Universe. This method assumes Exhaust Velocity Any propulsion method that expels
that spacetime itself is distorted locally around a vehicle matter or energy to produce thrust obeys conservation of
in such a way that apparent travel to outside observers is momentum, which is defined as mass times velocity. The
faster than to internal passengers. An example of such exhaust momentum is equal to and opposite change in
is the Alcubierre drive, but current theory does not indi- vehicle momentum. Given a finite supply of materials
cate how to actually generate such a space warp. Note that for the exhaust, mass is fixed. Therefore higher velocity
in General Relativity theory, gravity is caused by a warp leads to higher total momentum, and thus better perfor-
of spacetime, and passengers appear to themselves to not mance for the vehicle. Exhaust velocity is measured in
accelerate, while outsiders see them in accelerated mo- meters/second. Some systems, like 72 Ionospheric Cur-
tion. The difficulty is gravitational fields are not mobile, rent Engine, do not use exhaust for momentum exchange,
being attached to the large masses which cause them, so but have a consumed supply - plasma generated for con-
their utility for space transport is limited to methods like tact to the ionosphere in this case. The effective exhaust
3.13. COMPARISONS AMONG METHODS 133

velocity is then calculated from the thrust force and mass

flow rate.

Specific Impulse Specific impulse is a derived mea-

sure traditionally used for launch from Earth, and then
extended to space. It is defined as pounds thrust per
pounds/second of fuel consumed, which simplifies to
units of seconds. When multiplied by one Earth gravity
(9.80665 m/s^2) it converts to exhaust velocity, which is
the preferred measure since it is in SI units.

Acceleration

Thrust to Weight Ratio

Status Measures

Technical Maturity

Implentation Status

Cost Measures

Development Cost

Production Cost

Operating Cost

3.13.2 Transport Optimization

Optimizing Vehicles
In past rockets, this has been done by using different type
of fuel for different stages in a rocket. In the early part
of the flight, air drag is important, so a dense fuel is pre-
ferred. A dense fuel means smaller fuel tanks, and hence
less area to create drag. Thus the Saturn V used liquid
oxygen/kerosene and the Shuttle uses solid rockets for the
first stage, both being dense fuels. Both use liquid oxy-
gen/ liquid hydrogen for the second stage. This has the
highest performance in use for a chemical rocket fuel.
The Pegasus rocket uses an aircraft to get above the bulk
of the atmosphere. A sub-sonic jet engine has about ten
times the performance of a chemical rocket, mostly be-
cause it does not have to carry oxygen to burn.
Many, many propulsion combinations are possible in get-
ting to Earth orbit and beyond. A large part of space
propulsion design is choosing which methods to use and
when to switch from one to another.
Chapter 4

Part 3 - Space Engineering Methods

4.1 Space Engineering Methods There are numerous other systems besides propulsion re-
quired for most end items. These include structures, me-
chanical, power, thermal, data, communications, sensors,
4.1.1 Introduction
and environmental protection. When humans interact
with an end item, you additionally need displays and con-
In Part 1 we described the fundamentals of physics and
trols, internal environment control, and crew support such
engineering that apply to any complex project. This in-
as furniture, food, and clothing. Items with an extended
cludes Systems Engineering, which is concerned with
life require maintenance and repair in the form of tools
managing the whole of a complex system across its en-
and spares, and supplies such as fuel.
tire life cycle, the design engineering tools and specialty
areas of knowledge, and the organization and economics
of projects. Finally we looked at what projects and pro- • Resource Exploration - This includes the meth-
grams already exist, and the categories of future projects ods of finding and characterizing resources located
which might be pursued. In Part 2 we began covering the in space, as far as their location, composition, and
particulars of space systems with the most characteristic other physical properties. We only give a summary
element, the transport methods. Since humans are start- inventory of the known resources. Full details com-
ing from the Earth, then transportation is a prerequisite prise the entire fields of space science and astron-
to doing any other tasks in space, and so we discussed omy, which are both large and ever growing.
that first. Additionally, the very large number of possible
transport methods justifies devoting a large section of the • Resource Uses - There are as many possible end
book to it. uses for the available resources in space as there are
In this Part 3 we will cover the particular design factors things to do on Earth. We list the major applications
that apply to space systems, and the remaining subsys- in terms of what we want to do first. Later sections
tem elements besides transport methods. Combinations discuss how we can do them.
of subsystem elements then form complete end items or
products which execute designed functions and missions. • Resource Extraction - Physical extraction is the
We will review these end functions and the methods avail- task of removing materials from their native loca-
able to perform them in a logical sequence by time, start- tion, which is called Mining, along with prelimi-
ing with exploration and ending with recycling. The final nary processing and transport for further processing
major part of the book with then treat combinations of and production. Since you can only extract a phys-
multiple end items and systems - how they grow, interact, ical material once from it’s source, mining is typi-
and evolve. cally mobile, where chemical processing and manu-
facturing tend to be fixed, since the equipment and
connected power supply are often massive. Energy
4.1.2 Overview extraction involves converting a primary source of
energy into more useful forms such as electricity.
• Design Factors

Design factors are those which influence the whole of a • Processing and Production - These tasks combine
design, across different subsystems. These include input a multitude of simpler operations into one or more
requirements, technology level and availability of mate- complete process flows. The process flows convert
rials and suppliers, physical design such as margins and extracted materials and energy into final bulk mate-
wear, and the limits imposed by humans as part of a sys- rials and finished components.
tem and the operating environment.
• Assembly and Construction - Assembly combines
• Subsystem Design components to create a working device or machine.

134
4.2. DESIGN FACTORS 135

Construction first prepares a location, then assem- done both across different subsystems within a design,
bles larger structures and outfits them with inter- and across different alternative design options. For ex-
nal and external systems at fixed locations or or- ample, if a solar panel power/mass ratio of 100W/kg is
bits. The distinction between assembling mobile used in one alternative, the same value needs to be used
machines and fixed construction is somewhat arbi- in other design alternatives.
trary in space, since even the largest orbital construc- The maturity or Readiness of a technology is a measure
tions can be moved. of how far it has progressed from initial idea to com-
monplace use. NASA developed a scale of Technology
• Operation and Maintenance - Once items are as- Readiness Levels (TRLs) to describe the status of a tech-
sembled or constructed, they must be operated ac- nology. The higher the TRL, the less uncertainty about
cording to their intended use, and normally will the cost and performance of the given technology. Care
require periodic maintenance to keep functioning. should be taken to understand the actual state of a tech-
The overall concepts for operations and mainte- nology, and allow appropriate risk margins.
nance should be developed during design, so that
necessary design features will be included. If a part
will need replacing, for example, a way to access it 4.2.2 Availability Factors
and remove it should be part of the design.
These involve whether a given material, component, sub-
system, human skills, or facility will be available for a
• Recycling Methods - Most engineered items even-
given project. A given design option might require com-
tually reach the end of their useful life. Many meth-
ponents which are in limited supply. Even if they would
ods and processes generate waste products, espe-
perform better, if they are not available when the project
cially living things like humans. On Earth, disposal
needs them they will cause a delay which might be unac-
of scrap and waste has been left as a casual issue, and
ceptable. Reasons for limited supply include the number
much of recycling, such as converting CO2 back to
of suppliers with the skill and capacity to make the item,
Oxygen, happens by natural processes. In space, re-
their current work backlog, upstream resources they need
cycling has to be more deliberate. In many locations
from their suppliers, import or export restrictions, possi-
there are not surplus resources that can be used, you
ble production disruptions, intellectual property, and ex-
are floating in a vacuum. You also cannot just dump
isting contractual agreements. Any item which is not ob-
wastes, as they would end up being a debris hazard.
viously in abundant supply should be verified as to avail-
Thus efficient recycling will be both necessary, and
ability, to at least identifying a supplier who could meet
less expensive than extracting and delivering new re-
the project needs. In some cases, the project itself can
sources in many cases. Thus recycling also needs to
develop the capacity to provide a given item, but that im-
be planned for and designed in from the start.
poses additional tasks to do so.

While the topics arise more or less in this order, they need
to be designed for in parallel, and will be performed in 4.2.3 Physical Design Factors
parallel, and in a connected network with transportation
elements. The following pages will discuss these topics in • Materials Selection - This involves how to choose
more detail. materials for different parts of the design. Usually
materials selection involves multiple factors beyond
the obvious ones like strength and melting point.
Strength/density ratio is important when weight mat-
4.2 Design Factors ters, which is most of the time for space systems.
Materials cost. Qualification for a material mean
This section covers factors which influence a design sufficient testing to know how well a given material
across multiple types of subsystem elements. The most will work for a given purpose. New materials may
obvious influence is the set of requirements which define not have sufficient testing.
what the system is supposed to do and the criteria used to
score how good the design is. Requirements and criteria • Design Margin
were discussed in Section 1.5.
• Design Life
• Corrosion and Fatigue
4.2.1 Technology Factors

These involve the performance levels of different tech- 4.2.4 Integration Factors
nologies, and how mature they are. A project or de-
sign should assume consistent performance values and Integration is the process of combining lower level com-
assumptions for a given technology. This needs to be ponents into a higher level element which performs a set
136 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

of functions. the production, storage, transport, and operating environ-

ments. We can divide environments in general into two
• Design Budgets groups, objects and space, since the former have distinct
local conditions.
A system will have budgets for finite items such as
component mass, maintenance time, communications,
or command inputs. These budgets apply across mul- Object Environments
tiple subsystems, and must be estimated, allocated, and
tracked. There are a wide variety of objects in the Solar System
and beyond, of which the Earth is one. The environment
• Testability conditions vary widely between them, and locally on indi-
vidual objects, so we will not try to list individual details.
• Subsystem Interactions Instead we will note the types of environment parameters
which should be considered in a design, and what effects
This is consideration of the effects different systems im- they may have. For specific locations, previous scientific
pose on each other. These may include acceleration, vi- or other data sources can be used for detailed informa-
bration, thermal, electromagnetic, radiation, and others. tion. If that is not available, new observations or visits
may be needed to get the local data.

4.2.5 Human Factors

Atmosphere If a body has an atmosphere, a number of
You cannot design humans (not yet, anyway), therefore conditions should be considered. These include:
you have to factor in human features into a design. This
includes physical factors like acceleration tolerance, and
• Static Pressure - Internal elements of the system,
mental limits such as the finite ability to learn and execute
humans for example, may require particular condi-
operational tasks.
tions. If the outside pressure is too high or too low,
then the design will need to include a pressure shell
Anthropometrics to maintain the desired level.

Humans come in a range of sizes. Therefore seats and • Dynamic Pressure - These are the forces generated
control interfaces have to accommodate different eye po- by a moving atmosphere (wind), or from moving
sitions, eye focus, arm reach, hand size, and other char- the system through the atmosphere. The structure
acteristics. Devices such as space suits either need to fit must be designed to withstand these forces in ad-
the range of users, or crew selection needs to restrict the dition to any static pressure. Lightweight or large
size range to fit the equipment. One source for data for structures may bend or shift bodily from dynamic
space projects is the NASA Man-Systems Integration pressure loads.
Standards, but this topic has gotten a lot of attention on
Earth because most systems interact with humans, and • Composition - An atmosphere may contain gases
thus have to be designed to interact with them. which react with system hardware. Examples in-
clude oxygen and water on Earth, and sulfuric
acid on Venus. Some gases are combustible (react
Training quickly) with exhausted or leaked materials from the
system. For example, a leaking oxygen tank may
Operating a complex system is not intuitive, therefore hu- combust in a methane atmosphere.
man operators must be trained for the task. It imposes a
requirement that the amount of natural skills plus train- • Dust - An atmosphere can transport dust and larger
ing fit within what a given crew is capable of. For long particles. These can abrade surfaces, or get de-
duration operation, re-training may be needed for infre- posited on them and accumulate. Dust may interfere
quent tasks. Simulators are used to do training at lower with mechanical devices and be a hazard to living
cost, and to train for hazardous conditions, such as loss things.
of an engine. So in addition to designing operating hard-
ware, training media, instructors, and simulators need to • Condensation - Under certain conditions, atmo-
be considered in the design. sphere components can condense to liquid or solid
form and then precipitate as rain and snow due to
their higher density. Very small condensed particles
4.2.6 Environment Factors may remain aloft as fog and clouds. The effects on a
design include condensation on surfaces, and accu-
Any design must be capable of withstanding the various mulation of fallen rain and snow on equipment and
environments in which it finds itself. These will include the ground.
4.2. DESIGN FACTORS 137

• Opacity - An atmosphere can reduce visibility, photosynthesis and other chemical reactions, and natural
communication, and filter incoming and outgoing lighting. When not filtered by an atmosphere, the high
light and heat due to being partially or totally opaque energy part of the spectrum (ultraviolet and above) is a
in particular wavelengths. The opacity can be hazard to humans and other biologicals, and may degrade
caused by the gases, dust, or condensation within the other materials. Lack of sunlight either from nighttime
atmosphere, and can be variable. shadowing on an object, or inside system elements, may
require artificial lighting.

Temperature The equilibrium temperature for a sys-

tem element includes solar input, heat transport from any Meteor Flux On objects without a significant atmo-
atmosphere present, and from the solid or liquid ground. sphere, designs in exposed locations should account for
Most objects rotate, so the Solar input will vary with time. the natural flux of meteors. Long term exposure causes
If their orbit is significantly elliptical, the solar input will pitting, and in rare cases larger impacts can cause more
also vary with distance. Atmospheres can transport heat severe damage.
by radiation, convection, and conduction. To the extent
an atmosphere is transparent or not present, heat can be Object Changes Object surfaces can experience tran-
lost to the very cold background temperature of the Uni- sient events such as earthquakes, vulcanism, heavy pre-
verse. Systems operating close to, or on or below the sur- cipitation and flooding, high winds, dust storms, fire, and
face of an object will get some heating from the object, others. They can also experience seasonal and long term
in addition to solar input. changes such as surface melting and terrain shifts. A sys-
A functioning system also usually generates internal heat tem should account for the frequency and severity of such
from operation of components, so the equilibrium tem- events either by design features or by financial insurance
perature is what results when internal and outside thermal against low frequency events.
flows are in balance. If this temperature is higher or lower
than desired for system operation, then components like
radiators, heaters, and thermal insulation must be added Space Environments
to bring the internal temperature into the desired range
The space environment does not have new classes of pa-
rameters which do not exist for large objects, but the de-
Gravity On Earth, gravity level is within a few percent tails of each parameter will be different:
of the standard value (9.80665 m/s2 ) for all locations.
For other objects, except for the Gas Giants, it is gen- Plasma and Atomic Species Significant atmospheres
erally lower and more variable. When the level is below are bound by gravity to massive objects, but they shade
biological or industrial process needs, then generating ar- imperceptibly into the background medium that exists
tificial gravity by rotation may be necessary. If the level between distinct objects. The upper reaches of the
is so low that traction or anchoring by weight does not Earth’s atmosphere extend past the lowest stable orbits.
work as it does on Earth, then special methods for move- Although thin, the conducting plasma and atomic species
ment and staying put may be required. When significant in this region can affect hardware. This group of parti-
gravity levels are present, the design needs to account for cles are distinct from the radiation group by having lower
structural loads caused by it, and bearing loads against the energy, and thus unable to penetrate solid objects.
object, and ability of the object to support them.

Temperature Due to the low density in the space en-

Radiation Almost all locations receive some radiation vironment it has low rates of heat conduction, and can
from a combination of radioactive decay, solar wind and therefore have very different internal temperature (de-
flares, trapped particle belts, and cosmic rays. In addition fined by particle velocity) than that of objects embedded
to natural sources, a system may contain artificial sources in it. For design purposes, system elements will mostly
such as radioisotope generators, accelerators, and reac- be affected by solar input, reflected light or shadowing
tors of various types. Humans and other biologicals, sen- by nearby objects, and heat loss to the cold cosmic back-
sitive electronics, and some instruments are affected by ground.
high radiation levels. Protection comes from distance in
the case of point sources, and shielding of various kinds
from other sources. The best type of shielding varies byGravity Since gravity operates by an inverse square
force law, it never vanishes entirely, merely decreasing in
radiation type, and other parts of the design may provide
shielding by their mass and arrangement. strength with distance. For objects in orbit, gravity forces
manifest in the shape of the orbit, which influence design
in ways like varying communications distance, sun and
Light Flux The Sun is the major light source in the So- shadow times, and travel times to reach a desired des-
lar System. It provides a source for heating and power, tination. Free orbit trajectories for the different parts
138 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

of a system are nearly identical when the hardware ele- - Main Propulsion
ments are small relative to the distance of a massive ob- - Auxiliary Propulsion
ject. Therefore the design needs to account for the lack
of forces between the elements. Large structures will see - Aerodynamics
differences in gravity, called Tides, and objects using
propulsion or rotation will see artificial forces that act like • Thermal
gravity.
- Thermal Protection
Radiation Space environments typically have higher - Thermal Control
levels of radiation than found on the Earth’s surface. The
sources include UV and particle radiation from the Sun • Data
and cosmic rays. Bodies with strong magnetic fields, such
as the Earth and Jupiter, can trap particles and create Ra- - Processors
diation Belts with particularly high levels. Radiation lev-
- Storage
els can vary significantly from short term events like Solar
flares. See the radiation heading under the previous Ob- - Network
ject Environments section for additional design factors - Software
from radiation.
• Communications
Light Flux Light flux in the space environment has
the same design influences noted previously under Object • Sensors
Environments. The differences have to do with it being
unfiltered by any atmosphere, and a different or lack of • Displays and Controls
day/night cycle, depending on location.
- Ground Software
Meteor and Debris Flux The space environment con- - Data and Communications Hardware
tains natural solid particles ranging from dust grains up - Flight Software
to whatever size distinct tracked objects are (nominally 1
meter). In addition to the natural particles, human-made - Sensors and Instruments
debris, defunct hardware, and still functioning hardware - Artificial Intelligence and Autonomous Oper-
also exist. Designs need to account for random impact of ation
small particles, and tracking and avoiding larger objects, - Trained Operators
or otherwise accounting for the risk of damage from such
impacts.
• Internal Environment

- Atmosphere Maintenance and Regeneration

4.3 Subsystem Design
- Waste Management
4.3.1 Elements by Subsystem
• External Environment
• Structures
- Debris Protection
- Primary Structure - Radiation Protection
- Pressure Vessels
- Equipment Support • Crew Support

• Mechanical - Food
- Clothing
- Positioning and Translation - Health Maintenance
- Access Hatches and Panels
• Maintenance and Repair
• Power and Electrical
- Tools and Equipment
• Propulsion - Spares and Stocks
4.4. RESOURCES 139

4.4 Resources Commercial Exploration - As uses become more lo-

cal, private, and commercial, there is more of a tendency
The energy required to move materials in space depends to keep data private rather than public. Individual gain
in large degree on the gravitational potentials that must be and advantage now comes into play, and the relative ef-
traversed. The deep gravity well of the Earth requires ex- fort of detailed exploration of one specific location is
pending a large amount of energy to climb, as evidenced greater than general mapping or detection. So while simi-
by conventional rockets. Therefore a primary way to re- lar equipment may be used for both public and private ex-
duce the difficulty and cost of space projects is to use lo- ploration, what gets done with the data is different. This
cal resources instead of bringing everything from Earth. is most evident in mining and other resource extraction
To intelligently plan to use local resources, you first need industries. Some of the information gathered for com-
to know what they are. The task of finding out what’s mercial use will end up being released publicly and adding
there is called Resource Exploration both on Earth and to the general store of knowledge. To date, almost all of
in space. This section will discuss the methods of explo- the exploration of space has been done as science.
ration, and a current inventory of known resources. The
following sections will discuss uses for the resources and
4.4.2 Progression by Distance
methods to implement them.
Exploration methods can be divided by the distance at
which they are done, and the detail which they gener-
4.4.1 History and Process ate. These generally are inversely related - less distance
generates more detail - because the instruments have a
All methods of finding resources involve first sensing the fixed sensor resolution. Greater distance from the inter-
characteristics of a location, recording the measurements, preters of the data, as happens with unmanned planetary
and then interpreting the data to determine what it can be exploration, also has an inverse relation with transmis-
used for. Historically the sensing devices were the human sion speed for the same reason. Devices like radio dishes
organs of perception (eyes, ears, skin, taste, and smell). to direct data back to Earth also have a fixed resolution,
Paper and pen, and sometimes collected samples were so greater distance makes for lower signal strength and
used to record information, and then maps and written bandwidth. So the advantage of smaller optics due to be-
accounts made by humans interpreted the data. More re- ing close to the target has to be weighed against the need
cently, instruments with better accuracy and sensitivity for more powerful transmitters and larger antennas due to
than the human senses have been developed. Some in- being far from Earth.
struments can detect properties that humans cannot sense
The following sections discuss the various types of in-
at all. Integrating instruments with computers has auto-
struments classed into Long, Medium, and Short Range.
mated the recording process, and is heading towards au-
Long range means the instrument is much closer to
tomation of the interpretation step. Humans are more in
the storage and interpretation location than the target.
a supervisory role over the instruments and computers,
Medium range means the instrument is much closer to
though some amount of direct local sensing still goes on.
the target than the storage/interpretation site, but not in
Scientific Exploration - Often measurements are made, immediate contact. Finally, short range means in imme-
such as wind speeds, for many years before a use, such diate contact with the target location and able to interact
as maps for wind turbine generators, were available. This with it.
goes to the distinction between scientific and commer-
cial data. Scientific knowledge is sometimes described
as good in and of itself. A less philosophical standpoint 4.4.3 Long Range Exploration (Astron-
is that we do not know in advance what knowledge will omy)
prove useful. Therefore we accumulate all sorts of knowl-
edge in the reasonable hope that some of it will. Space Long Range Instruments
affects the Earth in many ways, and understanding the
history of other planets helps us understand the particular Telescopes:
one we live on better. Therefore astronomy and planetary
science are expected to turn out useful in general. Prior to 1610, long range exploration was limited to
Gathering data for “science” is also the first step towards the human senses, and thus little was known besides
more detailed exploration and local use, even if we don't the brightness and position of objects in the sky. The
yet know exactly what we will use it for. Without a def- development of the telescope for astronomy by Galileo
inite use in mind, there is no benefit to keeping the data changed that situation, a change which continues to this
hidden or in duplicating effort, so science operates as a day with larger telescopes and a broadening of the wave-
public and shared enterprise. That also provides better er- length bands in which they operate. A telescope gener-
ror detection and faster progress when more people can ally consists of optics which gather and focus low inten-
review and build upon past work of others. sity photons (light, or electromagnetic waves), and a sen-
140 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

sor which then detects the concentrated photons. This al- the future the Sun may be used as a steerable gravity lens
lows detection of what would otherwise be too dim to see. by placing the observing telescope at the right location op-
Originally the detector was the human eye, followed by posite what you want to look at. The angle the Sun bends
photography starting in 1840, and by electronic sensors light requires a distance of greater than 550 AU to reach
starting in 1979. Electronic sensors are up to 50 times a focus. The attraction of using the Sun is the enormous
more sensitive than film in terms of photon efficiency. diameter, which leads to high resolution. In the interim,
A photographic plate might represent 400 megapixels of other large telescopes of increasing power will continue
image resolution, so it is only recently that arrays of sen- to be built, and existing telescopes are upgraded with bet-
sors also matched plate resolution as well as sensitivity. ter instruments and continue to be used.
Recording the data has progressed from hand written logs
and charts of what the astronomer saw, to photographic Meteorites
plates which served as both the sensor and recording
medium, to automated measurement of the plates to con- First some nomenclature, since it can be confusing: an
vert to digital form, to transfer of the electronic sensor asteroid or meteoroid are objects while still in space. A
data to computer storage. Interpretation of the data is meteor is the bright trail in the atmosphere as it heats up
now semi-automated with computer software. For exam- and melts, and a meteorite is the object after it has hit
ple, detection of Near Earth Objects is made by com- the ground. Meteorites are direct long range samples of
paring electronic images from two times. Anything that the space environment, where nature has brought them
changes between the images is a candidate NEO, and the to us, instead of us having to go there and acquire sam-
software filters known objects and variable stars, leaving ples. They provide useful comparison spectra in the lab-
a list of detections for a human to examine. Final inter- oratory, to compare to telescopic spectra from objects
pretation of astronomical data is by humans, in the form still in space. We can also analyze them directly to deter-
of maps, catalogs, technical papers, and books. mine their composition. Meteorites are more useful for
this purpose if we can record the arrival trajectory, which
Optical telescopes have grown in size, and thus sensi-
lets us guess where it started from. A drawback to mete-
tivity, migrated to better observing locations (high alti-
orites is we have no control over where or when they fall.
tude and dry, or even in space themselves), and developed
Sample return missions function in a similar way to mete-
adaptive optics to get around the blurring of the Earth’s
orites as far as bringing the material to Earth for study at
atmosphere. For other wavelengths absorbed by the at-
long range from the origin. They provide much better in-
mosphere, high altitude or space is a requirement. For
formation on the source location and better preservation
long wavelengths, such as radio bands, single instruments
by returning the samples in a container rather than mete-
do not provide much resolution. Aperture Synthesis,
orite’s open passage through the atmosphere followed by
the mathematical combination of data from widely sep-
possibly long periods of exposure on the ground before
arated telescopes, is used to provide higher resolution in
being picked up.
that case.
Wikipedia has several articles listing large instruments. Radar and Optical Ranging
Optical Reflectors are used for all the largest instru-
ments because of the difficulties with large lenses. The Some objects get close enough to use radar or laser pulses
active detector is usually an electronic Charge Coupled to determine distance, and in some cases shape. The dis-
Device, or CCD, which converts light to digital signals, tance measurements are very accurate because of the tim-
and often a spectrometer is used to sort the incoming light ing accuracy of the detectors. Shape is obtained from
by wavelength. Radio Telescopes use a number of de- time distribution of the return pulse, the parts of the ob-
signs, with steerable metal dishes of various sizes being ject further away taking slightly longer. Several pulses at
used often, and aperture synthesis by combining signals different times as the object rotates can be used to deter-
from different instruments up to the diameter of the Earth mine a three dimensional shape. Reflected pulse intensity
to get higher resolution. The active detectors are typi- varies as the inverse 4th power of distance, due to inverse
cally cryogenically cooled solid state amplifiers to reduce square law both ways. So this method is strongly limited
noise. Space Telescopes cover a number of wavelength by distance, but the use of very large radio dishes and
ranges with a variety of detector types. The ones that op- powerful transmitters has overcome that limit somewhat.
erate near visible light wavelengths are designed similar
to ground optical telescopes.
Long Range Data
Gravity bends electromagnetic waves, and natural lenses
formed by massive objects such as stars and galaxies have
The following types of information can be obtained by
been used both to observe distant objects behind the lens,
interpreting the long range instrument data:
and detect and measure the lensing object. A conven-
tional telescope is still needed near the Earth to use this Position/Orbit - Generally this is the first information
effect. Such natural gravity lenses are not steerable. In found for a newly discovered object. The sky is mostly
dark with stars and other objects showing up as bright
4.4. RESOURCES 141

spots to the sensors. Stars are slow moving relative to each in a storage device, and then transmitted back to Earth.
other on human time scales, and the slower moving ones The storage is needed because the image recording time
are referred to as the Fixed Stars, even though they are and when and how fast the transmission can be made are
not if you wait long enough. For these types of stars their often different, especially when the telescope and trans-
position in the sky is recorded in terms of latitude and mitter are both fixed to the spacecraft body. Sensors de-
right ascension projected onto a reference sphere assumed signed for different wavelength bands can be used in the
to be infinitely far away. Overlaid on the fixed stars are same instrument, and filters can be placed in front of the
objects that move in short time scales. These are stars that sensors to select specific wavelengths of interest. Farther
exhibit parallax or orbital motion, and objects within our infrared wavelengths measure thermal emission in addi-
Solar System that show orbital motion. Parallax is a small tion to reflection of sunlight and the rate of change of a
shift in apparent position of relatively close stars vs farther thermal map can indicate properties. Absorption of sun-
stars caused by the motion of the Earth around the Sun. light through an atmosphere can give it’s composition.
The width of the Earth’s orbit, which is 2 Astronomical
Units by definition, is small relative to the distance of even Non-Imaging Optical Instruments
the nearest stars (260,000 AU). Nonetheless this allows
direct determination of distance by simple trigonometry.
All farther objects require estimating distance by indirect Besides direct 2D imaging via telescope optics, several
methods. Stars in binary or higher order systems (two other types of data can be collected:
or more stars), or ones with relatively heavy planets, can
show a small motion caused by the collection of objects • Radiometer - This measures infrared brightness to
orbiting their common center of mass, rather than their determine surface temperature.
own centers.
For objects within the Solar System, their orbital velocity • Polarimeter - Measures the polarization of incoming
is sufficient to show movement against the fixed stars in light by means of polarizing plates or films
days or hours rather than months or years for stellar mo-
• Photometer - Measures the total brightness of an ob-
tions. Significant movement in a short time is the key fea-
ject to high accuracy, without necessarily making a
ture to discover a Solar System object. Otherwise there is
2D image
no way to distinguish it from the myriad stars which can
be seen with a large telescope. By taking at least three • Spectrometer - Separates incoming light by wave-
measurements of position at known times, it’s possible to length, and records the brightness in each range. A
determine the general parameters of an orbit. For discov- large amount of information about an object can be
ery these are often at short intervals, such as successive determined from its spectrum.
nights. With additional measurements, and ones spaced
out in time, which gives a longer baseline on the orbital
path, the orbit parameters can be determined quite accu- Radar Instruments
rately.
Size - The fraction of sunlight reflected from an object is Distance or altitude can be determined by the time a radar
called Albedo. Albedo times the area of the object gives or laser pulse takes to bounce off the target. Synthetic
the total amount of light reflected from it. From the orbit Aperture Radar can determine altitude from the doppler
parameters you can calculate the physical distance. The frequency change vs return time.
observed brightness and an assumed albedo then allows
an estimate of the size. On initial discovery the object Magnetometer
may occupy only a single pixel on the sensor, so measur-
ing size by the image on the detector can't be done, and
Measures the magnetic fields around a target.
estimate by brightness has to suffice.

High Energy Detectors

4.4.4 Medium Range Exploration
These instruments detect neutrons, alpha particles, ions,
Medium Range Instruments
and gamma rays, which can provide composition infor-
mation about a target. In some cases high energy natural
Telescopes/Cameras
radiation impacts the target, and secondary particles are
These operate the same way as long range telescopes on emitted which are characteristic to the materials. In other
Earth, but due to being much closer to the target they cases particles are directly emitted by the target and de-
get higher detail (measured as pixels/km or meter). Ex- tected.
cept for a few early film recorders, spacecraft have all
used electronic sensors, whose data is usually recorded Gravity Mapping
142 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

This is not an instrument in itself, but uses doppler and other, or gas chromatograph or mass spectrometer to de-
timing information from radio signals to infer motion of termine composition of volatile gases that are released.
the spacecraft caused by the gravity field of the target. Testing for biological activity from the target, or compat-
Two sensors in orbit can determine their separation very ibility with Earth biology uses similar methods.
accurately by interferometry. Finer details of the gravity
field can be inferred from changes in this distance. Kinetic Prospecting

Medium Range Data Traversing the rough terrain of a body such as the Moon is
difficult and slow. An alternative is send a lander/rover to
a high point such as a mountain or crater rim. It picks up
4.4.5 Short Range Exploration a rock of suitable mass and uses a centrifuge arm to throw
it very fast at a selected target. Then it observes the im-
Short Range Instruments pact with a telescope and spectrometer to determine the
composition of the target, and possibly other instruments
Cameras for additional data. Repeat as many times as needed for
other targets. This allows prospecting a large area with-
Cameras and telescopes are fundamentally similar de- out having to drive over all of it. Kinetic methods have
vices, consisting of optics and an electronic sensor. The been used in the Deep Impact and LCROSS missions.
primary differences are at short range, large optics are
not needed as much to get sufficient resolution or bright- Ground Sensing Instruments
ness, and targets are close enough to sometimes need fo-
cus closer than effective infinity. Two cameras separated A Seismometer is very sensitive to motions of the
by distance, or a single camera used from different posi- ground. Natural or artificial impacts or movements gen-
tions can generate stereo data, from which three dimen- erate seismic waves within the body. The timing of the
sional shapes can be inferred. When the optics are very waves as they arrive at the seismometer allows determin-
short range and magnifying, the device is called a Micro- ing the internal structure and properties of the body. A
scope. Microscopic examination can determine mineral Heat Probe inserted into the ground can measure total
types. heat flow to and from the interior, and the thermal con-
ductivity. A Gravimeter accurately measures the local
Mössbauer spectrometer acceleration of gravity, from which density of the sur-
rounding ground can be calculated.
This uses a gamma ray source, such as a radioactive iso-
tope, and measures the recoil of atoms in the target to Subsurface Examination
determine their composition.
In Core Sampling a hollow drill is driven into the ground
Alpha Particle Spectrometer to separate a cylinder of nearly undisturbed material. The
core sample is removed and examined above ground.
This uses a combined alpha particle, proton, and X-ray Drilling uses larger machines to reach depths of up to
source, such as a set of radioactive isotopes. The wave- several kilometers. The drill debris can be flushed to the
lengths and energies of the returned X-rays and particles surface and examined there, or instruments can ride be-
are characteristic of the composition, which allows you hind the drill head or be lowered afterwards.
to analyze a sample.
Short Range Data
Laser Spectometer

A medium power laser is used to vaporize a target, and 4.5 Resources 2

the resulting emission spectrum can be used to determine
composition.
Astronomy, and space exploration by governments, is a
large field of research, and there is a lot of accumulated
Chemical/Biological Sample Chambers data about objects beyond the Earth. We will not attempt
to reproduce all of that data here, but rather summarize
Samples of the target are collected and deposited in and link to more detailed information. Many of the head-
chambers, which are then subjected to various fluids and ings are links to Wikipedia or other detailed articles. His-
conditions such as heating. Various ways to analyze the torically objects have been sorted as a hierarchy of Star
result include: transmission spectra made by using a light > Planets > Moons, which is based on gravitational bind-
on one side of the chamber and a spectrograph on the ing. Here we take the approach of sorting by available
4.5. RESOURCES 2 143

resources, mainly mass, composition, and energy, instead is now roughly 60% He. Although the heavier ele-
of gravitational dependency. The point of view is what is ments are a small fraction of the Sun’s composition,
there that could be used, rather than the pure science ap- they still represent about ten times the mass of the
proach of current condition and how did it form. rest of the Solar System besides the Sun.
The inventory is organized by Solar System matter and
energy, followed by our Milky Way Galaxy matter and The Gas Giants
energy, generally in order of decreasing quantities, and
without regard to the practicality of using the resources. Gas Giants are sufficiently massive to have retained a sub-
Resources outside our galaxy are too far away to be of stantial fraction of Hydrogen and Helium, the lightest two
near-term interest. Most of our Galaxy also falls into elements. The Solar System has four, which between
far-term interest, but some of the near portions may be- them have 444.6 x Earth’s mass, or nearly all of the mass
come useful given advances in technology, so we include not included in the Sun.
it for completeness. Descriptions of large scale engineer-
Jupiter
ing projects may also involve extra-Solar resources.
• Mass: 1.899 x 1027 kg, or 317.8 x Earth’s mass.
4.5.1 Solar System Matter • Mass Balance: Jupiter’s mass balance is not avail-
able, it would come from atmospheric loss and
History: The Solar System, which includes the Sun comet and asteroid impact gain.
and all the objects gravitationally bound to it, formed ap-
proximately 4.57 billion years ago from a giant molecu- • Composition: The atmosphere is about 75% Hy-
lar cloud, likely along with other stars. Since it formed drogen, 24% Helium, and 1% heavier elements.
by gravitational collapse, the original composition of the The total composition is estimated as 71% Hydro-
cloud primarily determined the composition of the Solar gen, 24% Helium, and 5% heavier elements.
System. Some material has since been lost by strong solar
winds during early formation, and weaker solar wind to Saturn
date, and by gravitational ejection of bodies. Addition-
• Mass: 5.685 x 1026 kg, or 95.15 x Earth’s mass.
ally some Hydrogen has been converted to Helium by the
Sun. The resulting composition found today is noted in • Mass Balance: Not available.
the list of major elemental composition by object below.
Aside from the Sun, most of the elements are combined • Composition: Saturn’s atmosphere is 96.3% Hy-
into molecules and minerals which represent local mini- drogen, 3.25% Helium, and 0.45% heavier ele-
mal chemical energy. ments. The total composition is estimated to be 20-
32% heavier elements, and the remainder H and He.
The NASA Jet Propulsion Laboratory (JPL) Solar Sys-
tem Dynamics website provides information on most Uranus
known natural bodies in the Solar System. Wikipedia
also has an extensive List by Size of the larger objects. • Mass: 8.681 x 1025 kg, or 14.54 x Earth’s mass.
• Mass Balance: Not available.
The Sun • Composition: Uranus’ atmosphere is about 83%
Hydrogen, 15% Helium, and 2% Methane. The to-
• Mass: 1.981 x 10 kg, or 333,000 x Earth’s mass.
30
tal composition is estimated at 0.5-3.7 Earth masses
This is all but 0.14% of the total mass of the Solar
rocky materials, 9.3-13.5 Earth masses ices (water,
System.
ammonia, and methane), and 0.5-1.5 Earth Masses
• Mass Balance: The Sun loses 4.28 million H and He. The ices would be solids at this distance
tons/second by virtue of the mass-energy of it’s light from the Sun, but in the interior of the planet they
output, and roughly 1.4 million tons/second due to are actually hot dense fluids.
the solar wind. It gains an (unknown amount) from
comet and other object impacts. Neptune

• Composition: The visible surface of the Sun, • Mass: 1.024 x 1026 kg or 17.15 x Earth’s mass.
known as the photoshpere consists of 74.9% H,
• Mass Balance: Not available.
23.8 % He, and 1.3% heavier elements. Due to the
relatively high gravity of the Sun, denser elements • Composition: The atmosphere is about 80% Hy-
have tended to sink to the core. The bulk composi- drogen, 19% Helium, and 1.5% Methane. The total
tion is estimated at 71.1% H, 27.4% He, and 1.5% composition is estimated at 1.2 Earth masses rocky
heavier elements. From fusion of H to He, the pro- materials, 10-15 Earth masses ices, and 1-2 Earth
cess by which the Sun generates energy, the core masses H and He.
144 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

Objects With Atmospheres • Mass Balance: Loss of Oxygen ions due to the so-
lar wind is estimated at 16-35 grams/sec in a 2010
Every object of sufficient mass and temperature will have article by Fang et. al., Effect of Martian Crustal
some trapped molecules. For the purpose of this page, Magnetic Field on Atmospheric Erosion. At this
we define Atmosphere to be sufficiently dense to flow, rate the atmosphere would be eroded in 22.5 billion
rather than free molecule interactions, and is mostly non- years. Accumulation from incoming asteroids and
ionized. meteorites is not available.
Venus • Composition: The atmosphere contains 95.3%
CO2 , 2.7% Nitrogen, 1.6% Argon, and trace com-
• Mass: 4.87 x 1024 kg, or 0.815 x Earth’s mass. At- ponents. The total composition is generally a sili-
mosphere = 4.8 x 1020 kg. cate outer layer with a metallic core, with elements
by mass roughly in the order oxygen, silicon, iron,
• Mass Balance: Venus appears to be losing about magnesium, aluminum, calcium, and potassium.
16 grams/sec x atomic weight of atmosphere due to
solar wind stripping. Average accretion from comet Titan
and asteroid impacts is not available.
• Mass: 1.345 x 1023 kg, or 2.25% of Earth’s mass
• Composition: The atmosphere is about 96.5%
in total. Atmosphere = 9.05 x 1018 kg.
CO2 , 3.5% Nitrogen, with trace compounds. Be-
cause of similarity in density and total mass to the • Mass Balance: Not available, but considerations
Earth, Venus is expected to have a similar total com- are stripping by Saturn’s magnetosphere, accretion
position and structure, consisting of a rocky outer from small objects, and stripping from larger im-
portion, and a metallic core. pacts.
• Composition: The atmosphere is variable by al-
Earth
titude. Higher levels are 98.4% Nitrogen, 1.4%
Methane, and 0.2% trace compounds. Lower levels
• Mass: 5.974 x 1024 kg, or exactly 1.000 x Earth’s are 95% Nitrogen, 4.9% Methane, and 0.1% trace
mass. Atmosphere = 5.28 x 1018 kg. compounds. Total composition is roughly half sili-
cate core and half ices and likely liquid water at some
• Mass Balance: Earth loses about 3 kg/s of Hydro- depth. The surface appears to be mostly water ice,
gen and 0.05 kg/s of Helium from the upper atmo- with variable amounts of Carbon Dioxide, Methane,
sphere. A small amount of man-made objects de- Methane Hydrate, Ammonia, and Methanol by lo-
part the Earth per year. On the order of 1.25 kg/s of cation.
extraterrestrial material of all sizes (dust to dinosaur
killing asteroids) accretes. Therefore as a whole the
Triton
Earth is losing mass, but the current rate amounts
to 40 parts per billion of the total mass over the re-
maining life of the Earth. • Mass: 2.14 x 1022 kg, or 0.36% of Earth’s mass in
total. Atmosphere ~5 x1013 kg.
• Composition: The atmosphere contains 78% Ni-
• Mass Balance: Not available.
trogen, 21% Oxygen, 1% Argon, variable amounts
of water, and trace compounds. The total composi- • Composition: The thin atmosphere is mostly Ni-
tion consists of the following major elements: Iron trogen. The solid composition based on density is
(32.1%), Oxygen (30.1%), Silicon (15.1%), Mag- likely 30-45% water ice, and the remainder rocky
nesium (13.9%), Sulfur (2.9%), Nickel (1.8%), Cal- core. Surface is 55% solid Nitrogen, 15-35% water
cium (1.5%), and Aluminium (1.4%); with the re- ice, and 10-20% carbon dioxide ice.
maining 1.2% consisting of trace amounts of other
elements. Due to density segregation, the core is
95% Iron and Nickel, while the uppermost layer, the Objects Without Atmospheres
Crust, is nearly all metal oxides. By weight these are
oxides of Si, Al, Fe, Ca, Mg, Na, K, H, C, Ti, and This group includes those without significant atmo-
P. spheres. This happens whenever the combination of low
mass and temperature allows gas molecules to escape or
to be stripped off by Solar wind interaction. They are
Mars grouped by the International Astronomical Union into
one planet (Mercury), several dwarf planets, the remain-
• Mass: 6.42 x 1023 kg, or 0.107 x Earth’s mass total. ing satellites of the major planets, a very large number of
Atmosphere = 24.8 x 1015 kg. Small Solar System Bodies which are further divided
4.5. RESOURCES 2 145

into types, and a possible class of Rogue Objects. These Outer System Objects The outer system includes ob-
divisions into groups are more for convenience by orbit jects outside the orbit of Jupiter, but not the outer plan-
location than mass, composition, shape, and internal ar- ets themselves or their associated satellites. In approx-
rangement. There is a more or less continuous distri- imate order of semi-major axis (symbol a, in AU) this
bution of all of these properties, including orbits. Be- includes the Centaurs, Kuiper Belt, Scattered Disk',
cause most orbits are elliptical, and sometimes highly so, Hills Cloud, Oort Cloud, and Rogue Objects. There
a single object can cross several nominal orbit locations. is overlap between these groups and with asteroids. The
Wikipedia has a table of Solar system rounded objects distinctions are based on convenience in describing their
which includes the dwarf planets, larger satellites, and orbits and composition rather than a real difference in
dwarf planet candidates. type. They are all objects not attached to a major planet
with orbits beyond Jupiter and too small to maintain a
permanent atmosphere. All the outer system objects are
Mercury
presumed to form from the same solar disk which formed
• Mass: 3.3 x 1023 kg, or 5.5% of Earth’s mass. the planets, but were then scattered by the gas giants once
they grew massive enough. Comets II and The Solar
• Mass Balance: Not available. System Beyond Neptune are recent reviews of the state
of knowledge for some of these objects.
• Composition: Approximately 70% metallic core
and 30% silicate outer layers. The total mass of all the outer system objects is poorly
known and is an active area of study. Density and spec-
tra indicate the outer system objects as a whole consists
Dwarf Planets Dwarf planets are defined as large mostly of Hydrogen-bearing ices: Methane (CH ), Am-
4
enough to have been shaped by gravity (into an ellip- monia (NH ), and Water (OH ), which are all solids at
3 2
soid), but not so massive as to clear its neighborhood of large distances from the Sun. Comets are former outer
other objects. The category was created in 2006 when system objects whose orbit now comes closer than Jupiter
it was obvious that Pluto was part of the Kuiper Belt re- to the Sun, so that gas evaporates off their surface, thus
gion, which contains some objects larger than the former they are listed below. The remainder stay cold enough to
planet. Currently 5 dwarf planets are officially recog- be solid.
nized (1 Ceres, Pluto, Haumea, Makemake, and Eris),
and it is expected several hundred to 2000 others will be
found as observations of the region beyond Neptune im- • Centaurs
prove. Objects larger than 838 km in diameter are provi-
sionally classed as dwarf planets, but final status requires These are objects which have closest approach to the Sun
they be rounded, and most such objects have not been (perihelion) between Jupiter and Neptune, and no limit
observed that well yet. on the farthest point. Approximately 200 are known as of
2012, and are being discovered at about 15-20 per year.
• Mass: Five official dwarfs: 3.8 x 1022 kg, or 0.63% The discovered ones range in size from 200 down to 2 km
of Earth’s mass. Perhaps 10% Earth mass once all in diameter. Estimated number larger than 1 km in size
outer system objects are discovered. is 44,000. Because they cross the orbits of the gas giants,
their orbits are unstable on a scale of a few million years.
Planetary Satellites This group of objects is distin-
guished by being attached to a major planet and not • Kuiper Belt
large enough to maintain a significant atmosphere. The
Wikipedia List of Natural Satellites shows the known The Kuiper Belt includes objects in stable orbits outside
ones to date, from which Titan (Saturn VI) and Triton the orbit of Neptune (30 AU) up to 50 AU, having eccen-
(Neptune I) are excluded since they have atmospheres and tricity generally from 0 to 0.2, except for resonant objects
were listed above. It is very likely there are additional which can go to about 0.4, and with inclinations up to
small satellites orbiting the major planets which have not about 35 degrees. Pluto is now considered the largest such
been discovered yet. Total mass is around 475 x 1021 kg, object, and 1241 such Trans-Neptune Objects (TNOs),
mostly in the 4 largest moons of Jupiter and the Earth’s as they are also known, have been found since 1992. Es-
moon. Composition varies by distance from the Sun from timated number of TNOs larger than 100 km diameter is
rocky (the Moon) to icy. We group these objects into 0.1 million, >10 km possibly as high as 10 million. Es-
two classes by size: Large and Small. The large group timated mass is 0.04-0.1 x Earth’s mass, but models of
are shaped by gravity, and possibly internal heating from planet formation predict there were 30 Earth masses of
radioactive decay or tidal friction from the planet it or- such objects originally. Where the rest went in that case
bits. This group are approximately spherical, and would is unknown at present. Composition in general is poorly
be considered dwarf planets if not orbiting a planet them- known, but a combination of CH4 (methans), NH3 (am-
selves. The more numerous Small group are irregular in monia), and H2O (water) ices are expected from the small
shape and likely not differentiated by internal melting. number of density and spectral observations.
146 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

• Scattered Disk' have been definitely identified in our vicinity, but in 2011
some candidates in our galaxy were discovered through
Scattered Disk Objects (SDOs) have perihelion (closest gravitational microlensing. Since Solar System objects
approach to the Sun) beyond Neptune (ie 30 AU) with have been ejected by the gas giants, by symmetry there
higher eccentricity (0.20 to 0.94) at a given distance, and should be objects ejected from other stars which are cur-
extending to larger distances than the Kuiper Belt. Their rently near the Sun. In addition to ejected objects, objects
inclinations range up to about 40 degrees. The rise in ec- that formed separate from stars by the same mechanism
centricity with distance may be a selection effect, since are expected. Total numbers are very uncertain, with es-
we can only find objects that come closer to the Sun. timates from 2 to 100,000 such objects per star, including
Outer system objects observed from Earth dim as the our own.
4th power of distance. This is the product of solar in-
tensity, which falls as the square of distance, and angular Inner System Objects This group includes objects
area, which also falls as the square of distance. The name from the orbit of Jupiter inwards. It is listed after the
comes from their orbits being gravitationally scattered by outer system objects because this is a resource-based in-
the gas giant planets at some point in their history from ventory, and the total mass is smaller. In decreasing dis-
the lower inclination and eccentricity of the Solar neb- tance from the Sun it includes:
ula from which they formed. Their orbits are subject to
further changes from planet interactions, so are not stable
• Jupiter Trojans
over the long term. Since 1995, a total of 167 SDOs have
been found (in that table, the ones with q >30 AU). Their
total mass is estimated at 0.01-0.1 Earth Mass. There are stable Lagrange Points located 60 degrees
ahead and behind a smaller body orbiting a more massive
one. The most massive such pair in the Solar System,
• Hills Cloud
Jupiter and the Sun, has the strongest such stable regions,
extending about 30 degrees of Jupiter’s orbit each. Ob-
The Hills Cloud, also known as the Inner Oort Cloud, are jects trapped there are called Trojans because the first
objects with semi-major axes between 1,000 and 10,000 few found were named after mythical characters from the
AU. The Sun is expected to have formed in a star cluster Trojan War. As recently as 1961, only 14 Jupiter Trojans
embedded in a gas cloud. Computer simulations (2011) had been discovered, but by 2012 the Minor Planet Cen-
of such clusters indicates objects forming closer to the ter lists over 5300. There are estimated to be 0.1-0.3 mil-
Sun would be scattered out to these distances and into the lion Trojans larger than 2 km diameter. Total mass and
outer Oort Cloud with an efficiency of ~1.5%, leading to composition are poorly known at present, but can very
an estimate of ~3 Earth masses currently. The Hills Cloud roughly be estimated at 2 x 1019 kg. A relatively small
is bound strongly enough to the Sun that perturbations number of similar Trojan objects exist around other mass
from outside sources and the gas giants rarely put them pairs.
in orbits where we can detect them. Their large distance
makes them too dim to see except when they approach
• Main Belt Asteroids
closest to the Sun. Thus we have only discovered a few
objects on orbits which enter this region.
About 600,000 total asteroids are known in the Main Belt
which ranges from 1.3 AU minimum distance to Jupiter’s
• Oort Cloud orbit. The dense region of the Main Belt has a doughnut-
shaped volume of about 70 AU^3, and so an separation
The Oort Cloud is made up of objects with semi-major of 0.05 AU, or 7,300,000 km. The number of known
axis > 10,000 AU. At such great distances from the Sun, asteroids has increased by 60 times since 1980, and is
their orbits are affected by close star passes and galac- expected to continue growing, but newly found ones are
tic tides. Their origin may partially be capture of loose generally small. The total mass in this region is estimated
objects in the cluster where the Sun formed. The remain- at 3 x 1021 kg (0.05% of Earth). Total number larger than
der would be highly scattered from the solar disk during 1 km is roughly 1 million, and >100 meters ~25 million.
the Sun’s formation. Estimated mass may be 0.1-7 Earth
masses, consisting of about 6 x 1010 to 1012 objects. This • Near Earth Objects (NEOs)
estimate comes from observing the frequency of long pe-
riod comets, but this does not place strong limits on their
These are defined as all objects which have a closest ap-
number.
proach to the Sun of less than 1.3 times the Earth’s dis-
tance (1.3 AU), with the exception of the Sun itself, Mer-
• Rogue Objects cury, and Venus. Mars has an average distance of 1.52
AU, but is lower mass than the Earth. The 1.3 AU defi-
Rogue objects, also called nomads or rogue planets, are nition is thus approximately closer to the Earth than outer
not bound to the Sun and merely passing nearby. None planets from a gravity standpoint, or close enough to be
4.5. RESOURCES 2 147

interesting. This is an artificial definition from a human Halley Family

interest standpoint - there is no distinct physical group- This group of comets have orbits with periods of 20-200
ing for this class of bodies, such as there is for the Main years which are oriented so that Jupiter does not affect
Belt asteroids. NASA has a Near Earth Object Program them strongly. Their inclinations range from 0 to 180 de-
to discover and characterize them, and as of mid-2012 grees, with a slight excess below 60 degrees.
there are about 9000 known. Estimated number >1 km
size is around 1000, and >100 m is about 200,000. Long Period
The NEO category includes all object types, including These are defined by having an orbital period > 200
asteroids, comets, extinct comets (which can be hard to years, and thus a semimajor axis (a) > 34 AU. Long
distinguish from asteroids), and manufactured spacecraft. period comets strongly cluster in semi-major axes near
There is an undefined lower bound to size, which we will 10,000 AU, which makes them hard to distinguish from
assume to be 1 meter diameter, below which we refer to parabolic. This cluster of orbits led to the assumption of
them as meteoroids, dust, or particles. The population of the Oort cloud as their source. Their inclinations span the
NEOs is not permanent. Over periods of approximately full range from 0 to 180 degrees.
1-10 million years either gravity effects from planets and
larger bodies, or collisions with them, will remove them
Particulates
from the NEO orbit range. Objects smaller than approxi-
mately 1 cm are also affected by light pressure or other ef- Dividing line between distinct objects and regions or
fects more than gravity interactions, and have even shorter masses of particles better tracked as a whole.
lifetimes in those regions.
Particle Belts
Rings
Comets Comets differ from the solid objects listed pre-
viously in that they show periodic vaporization and dust Interplanetary dust Gas and solar wind
emission. This is caused by heating of their surfaces when
they get close to the Sun. Water sublimation becomes
strong at less than 2.5-3 AU, and other volatiles at other
4.5.2 Solar System Energy
distances. The lost material can create spectacular though
The Sun
not very massive tails. Since comets are in effect boiling
gases from their surface, the built up pressure can cause
Current Energy Output From fusion of hydrogen to he-
them to fragment. Their lives are limited by the amount
lium
of volatiles they start with. Defunct comets resemble as-
teroids, and distinct trails of debris in their orbits are ob- Estimate Energy Reserves
served as periodic meteor showers. Comets ultimately
originate in the Oort Cloud, and migrate to closer orbits • Hydrogen Fusion
by gravitational perturbations. • Additional Fusion Reactions
Comets are divided by their orbits into Short Period, • Stored Thermal Energy
Jupiter Family, Halley Family, and Long Period • Gravitational Collapse Energy
classes. Typical orbit changes in terms of semi-major
• Minor Energy Reserves - Spin, stratifica-
axis are about 0.001 per orbit. If there were no planets
tion, magnetic field
around the Sun, the comet orbits would tend to stay fixed.
Since Jupiter is roughly 1/1000 the mass of the Sun, it
can be thought of as a blender blade mixing up the orbits Everything Else
by that amount per pass. Therefore comet orbits tend to
randomly migrate among orbit classes. Everything else is lumped under one heading because
Jupiter Family magnitude of the Sun’s energy reserves is so much larger
than everything else combined.
Since Jupiter is by far the most massive planet in our Solar
System, it has the most influence on comets which cross Latent Heat of Formation - When massive objects like
it’s orbit. There is a noticeable cluster of comets whose planets formed within the Solar System, the collision and
maximum (aphelion) distance (Q) from the Sun is close gravity well energies, and later the stratification of the in-
to that of Jupiter (5.2 AU). The cluster roughly ranges teriors by density released a lot of energy, part of which
from 4.2 AU < Q < 11 AU, with minimum (perihelion) went to heating the interior of the object. For the larger
distance (q) ranging from 0.5 AU < q < 5.5 AU. This objects, some of that heat is still stored in their interiors.
range is generally just outside that of the main Asteroid Nuclear Fission - Objects in the Solar System incorpo-
Belt. Jupiter family comets generally have inclinations rated elements with radioactive isotopes which decay nat-
less than 35 degrees. Approximately 200 Jupiter Family urally, heating their interiors, or can be made to fission on
comets are known. purpose.
148 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

Nuclear Fusion - Just like the Sun, but on a smaller In recent years a large number of planetary systems
scale, there is potential energy in light elements that can
around other stars have been detected. The Extrasolar
be fused together. Since it takes at least 75 Jupiter masses
Planets Encyclopedia catalogs them with references to
original papers. As of 2012 there were 660 such systems
for this to happen naturally, then it must be made to hap-
pen artificially. with 837 planets detected. Methods of detection vary. In
Chemical Reactions - Materials such as fossil fuels and addition to planets which have formed, Circumstellar
undecayed plant matter can burn with oxygen in the at- Disks represent early stages of formation or incomplete
mosphere to release energy condensation to larger objects.

Orbital Kinetic Energy - The bulk of the kinetic energy

in the Solar System resides in the motion of the planets Rogue Objects
and smaller bodies. When spacecraft use a flyby to alter
their motion, they extract a little of this energy. These are objects not tied to a star, which can range
Minor Energy Reserves - from planets below the brown dwarf limit down to ejected
comets. They are currently not well understood, as only a
few candidates have been detected by gravitational lens-
4.5.3 Galactic Matter ing in 2011. Estimates of their number range from two
per main sequence star up to potentially 100,000 per star.
The Galaxy or Milky Way in capitalized form refers to The larger number depends on optimistic assumptions for
the gravitationally bound object which the Sun and Earth the mass function (number of objects vs their mass).
orbit within. In lower case, galaxy refers to the general
class of such objects.
Interstellar Medium
Summary description of the Galaxy as a whole: total
mass of the Galaxy, and baryonic vs Dark Matter.)
Dust Particles
Components sorted by Mass:

Interstellar Gas
Dark Matter

4.5.4 Galactic Energy

Stars
Total Power output
Satellite Galaxies Total Energy reserves

Clusters Fusion

Clouds
Gravitational energy

Central Black Hole

Dark Energy

Substellar Objects
Angular Momentum
Brown Dwarfs -

These are objects too small to count as stars, but above the Thermal Energy
limit for planetary bodies. The upper limit is about 80
times Jupiter’s Mass, above which hydrogen fusion can
happen and the object is considered a star. The lower Magnetic Fields
bound is about 13 times Jupiter’s mass, below which no
fusion will occur. Above this lower limit, Deuterium and
Lithium fusion can happen, but since these are much rarer High Energy Sources This includes cosmic ray flux
than Hydrogen, it limits their life and brightness. and X-Ray sources.

Planetary Systems • Wikipedia article: Ultraluminous X-ray Source

4.6. RESOURCE EXTRACTION 149

4.6 Resource Extraction gravity. Therefore equipment anchoring to keep it from

moving itself is needed, or special design to contain gravel
This section discusses how to extract resources from their and dust from traveling long distances, or even leaving
natural state. For matter and energy these are commonly the body entirely. By extrapolation from Earth construc-
called mining and energy production, but the latter is a tion equipment, local gravity below 2 m/s2 (0.2 g) will re-
misnomer. Energy cannot be created, it can only be con- quire lower equipment forces or anchoring. Blasting is the
verted from an existing source, so we prefer the term Ex- most energetic mining operation. From blasting safety
tracted. codes we estimate that gravity below 0.3 m/s2 (0.03 g)
would require safety distances of over 1 km from shrap-
nel and debris, which becomes unreasonable. That value
4.6.1 Mining will likely need updating from actual experience. For now
we will divide bodies into low, medium, and high grav-
Mining is the process of extracting, or in the case of dis- ity for mining purposes at 0.3 and 2 meters/s2 . For the
persed materials, collecting, physical materials for direct very smallest objects you do not mine them in the sense
use or further processing. A Deposit or Ore is a natu- of removing material from a larger body, but capture and
rally occurring material of sufficient size or concentration transport them to a processing location if needed.
to be mined economically. Mining techniques will vary Low-G Mining
according to what you are extracting and where you are
doing it. So this section organizes the methods by phys- Medium-G Mining
ical state and type of natural body being mined. Some High-G Mining
processing may occur at the mining site to concentrate
the ore or prepare it for transport. Concentration is called
Beneficiation, and there are a number of other processes Mining by Depth We can also divide mining tech-
which can be applied, such as crushing and sintering. niques by depth. These are surface, sub-surface or un-
derground, and deep mining. Sub-surface is anywhere an
open pit or angle of repose is not sufficient, and columns
The Challenge
or walls are needed for support. The support can be part
of the natural materials left in place, or artificial supports
Most of the baryonic mass in the Universe is inconve-
installed as part of the mine development. Deep mining
niently located in the interior of large bodies, where it
is when the gravity loads on the natural materials from
hard to get to. In fact spheres, the shape which many
what is above it start to cause shifting or collapse. This
large objects approximate, have the least surface area for
requires fully enclosed tunnels or pipes or other special
a given volume. In other words, the ratio of relatively
techniques. The strength of the surrounding material will
inaccessible material in the interior to accessible mate-
determine when you reach a deep condition, and this will
rial on the surface is a maximum. Another problem is
vary from place to place. As a guide, for rocky locations,
that useful metals, such as Iron, tend to collect in the cen-
deep can be considered more than 15km /(local acceler-
ter of large objects due to their density, where they are
ation of gravity in m/s2) . Gravity falls as you move to the
inaccessible due to surrounding layers of rock. Besides
center of a body, so aside from the few largest asteroids,
physical inaccessibility, much of the matter is in a plasma
the whole interior may not reach a deep condition from
state (e.g. in stars) or very low density (e.g. in molecu-
lack of sufficient gravity load. Conversely, on a larger
lar clouds). So the challenge for mining is accessing the
body such as the Moon, you can reach a deep condition
small fraction of currently accessible material, and devel-
within 0.5% of its radius due to higher gravity and larger
oping techniques to increase the range of such materials.
radius.
Even conventional mining on Earth literally only “scrapes
Surface Mining
the surface”. So-called deep mining and drilling typically
only reach the top 0.1% of the planet’s radius. This limits Sub-Surface Mining
our ability to obtain rare materials, or enough of common Deep Mining
materials for future large projects. Mining in space is one
method to increase the range of accessible materials, be- We will take as an example of deep mining to excavate
cause it is new accessible area with different history and the core of the second most massive asteroid, 4 Vesta, for
composition than Earth, but it should be considered a part Iron. Vesta melted in its early formation from radioactive
of the total challenge for future mining. decay, and formed an Iron core and rocky mantle. It has
the following properties:

Mining Solid Bodies

• Polar Radius = 229 km surface, 106 km
core
Mining by G-Level We can divide solid bodies into
groups by size. Small bodies are such that typical min- • Equatorial Radius = 305 km high point,
ing equipment forces and velocities are larger than local 114 km core
150 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

• Gravity = 0.329 polar, 0.1532 equatorial and dividing by the solar conversion efficiency. The solar
high point, and 0.257 core in m/s2 . The flux for the Earth averages 1360 W/m^2, and the radius
much lower equatorial gravity is due to is 6378000 meters, thus the total flux is 1.74 · 1017 J/s.
the larger radius and rapid rotation (5.34 This gives a characteristic dis-assembly time of 45.3 mil-
hour rotation period). lion years.
• Mantle Density = 3115 kg/m3
Mining Atmospheres
As an approximation we take the polar and equatorial av-
erage acceleration from surface to core times the distance. For bodies with appreciable surface pressure and a dis-
This amounts to 123,000 m x 0.293 and 191,000 m x tinct solid or liquid surface, bulk mining of the atmo-
0.205. The polar value is slightly lower, so we dig down sphere is relatively straightforward. The main require-
from the pole. A column of mantle rock of that height, ment is a pump and a storage tank. If a particular compo-
density, and acceleration sees a pressure at the bottom nent needs to be separated out it becomes more complex,
of 112.26 MPa ( 16,300 psi ). This is higher than the needing liquefaction or freezing to separate the differ-
likely strength of the rock, so an unsupported hole will ent compounds. For a low pressure atmosphere, methods
likely fail, and we are indeed in a deep mining condition. like selective ionization or mass spectrometry may work
Therefore the hole will need a lining such as steel obtained to separate particular components. For bodies without a
elsewhere, at least for the deeper parts. The core itself, distinct surface, or in cases where you don't want to land,
being made of mostly Iron, will be self-supporting until the orbital scoop mining method can be used.
too much of it has been extracted and it undergoes core
collapse. About half of the core can be extracted safely,
amounting to 23.3 million gigatons, or about 15 million
years of Earth production at current rates.

Disassembly At some point, even deep mining tech-

niques will become impractical. Many bodies have hot
interiors or liquid layers, and even without those prob-
lems, pressures would require tunnel or pipe walls too
thick to be worth installing. The only approach to reach-
ing these very deep resources is to remove all the over-
lying material first, which amounts to disassembling the
body or a large portion of it.
The energy holding a large object together due to its grav- Scoop mining ship concept
ity is called the Gravitational Binding Energy - . For a
Scoop Mining This method involves skimming the up-
uniform sphere it is found by the formula
per atmosphere of an object to collect gas, then ejecting
part of the collected mass as propellant to make up for
2 drag. The altitude to collect gas is selected based on drag
3GM
U= and heating levels for the scoop, and available thrust. The
5r
scoop is shaped as an inverse nozzle, converting a high
where G is the gravitational constant, M is the mass of velocity, low pressure stream into a low velocity (rela-
the sphere, and r is its radius. Thus to extract all the mat- tive to the scoop) high pressure gas. The collected gas is
ter from the object requires at least this amount of energy pumped into a storage tank. Since collection velocity for
input at 100% efficiency. This assumes you don't do any objects with atmospheres is typically higher than a chem-
processing, just physically remove all the material. As ical rocket exhaust velocity, the portion ejected to make
an example, for the Earth, allowing for the actual distri- up for drag will need to be at electric thruster velocities
bution of mass and density by depth, U = 2.487 · 1032 (30-50 km/s). To maintain orbit, the average thrust must
J. equal the average drag. The ratio of scoop velocity / ex-
Dis-Assembly Time - This is a characteristic time found haust velocity determines what fraction of the collected
by dividing the binding energy U, by the solar flux falling gas needs to be used for thrust. The concept will work in
on the object. There are other energy sources which could theory for any body where the exhaust velocity is suffi-
be used, but sunlight is generally available in space and so ciently higher than the orbit velocity.
gives a natural value for the time to dismantle the object Intermittent Operation - Take as an example mining
using the solar energy falling on it at 100% efficiency. the Earth’s atmosphere from orbit. At an altitude of 150
From the characteristic value, you can make estimates km, the density is 3 x 10−9 kg/m3 . A vehicle traveling at
of the actual time for a large mining project by multi- 7800 m/s relative to the equator, which is 450 m/s above
plying by the fraction of the object you intend to extract, circular orbit velocity at that altitude, will encounter 23.4
4.6. RESOURCE EXTRACTION 151

micrograms of air per square meter per second, and see higher altitude, and so not limited in size and power by
a drag of 0.18 N/m2 . The stagnation pressure if the drag limits. The scoop, pumps, smaller solar array to
scoop fully brings the airflow to a stop is 1350 Pa (0.2 power the pumps, and storage tank are lowered on cables
psi). A vacuum pump pulls this gas into a storage tank at to an optimum altitude to collect air. If the same size
higher pressure. Assume the scoop system has a mass of as the continuous operation version above has an empty
100 kg/m2 . Therefore it will lose 400 m/s of velocity in mass of 1 ton, then it needs to be at least 1 km below the
220,000 seconds (2.6 days). The orbit is elliptical, so the center of mass for the gravity gradient (tide, or differen-
collection will only be at full pressure at the lowest point, tial gravity) to be larger than the drag force. It will then
and actually take about 4 times longer. The scoop will hang at a trailing angle balancing gravity and drag. To
have collected 5.2 kg of air per square meter over this have enough difference in air pressure to be worth lower-
time. Before too much drag happens, the scoop raises ing the scoop, the cable will be much longer than 1 km.
its perigee by 50 km using an electric thruster to stop When the tank is full, the cables are reeled in and the
the collection. This requires 15 m/s of velocity. Raising tank is unloaded or swapped. This saves using propellant
apogee by another 400 m/s, and then dropping perigee by for climbing up and down from the scoop altitude, and
50 km to start the collection cycle gives a total velocity allows more powerful thrusters. Whether it is an overall
of 465 m/s required. At an exhaust velocity of 50 km/s, advantage requires a more detailed study.
performing the velocity changes will consume 0.98 kg of Terrestrial air scoops were studied in the early 1960s by
propellant, which we assume to be some of the collected R.H. Reichel at Boeing. This work looked at the opera-
air. This leaves us with a net of 4.2 kg collected. tion of a scoop in continuum flow (110 km altitude) and in
Continuous Operation - For a second example, assume a free molecule flow environment (160 km altitude). Also
a VASIMR type thruster which uses 200 kW power to examined were the power requirements for liquefaction
produce 5.7 N thrust. The solar arrays need to be about as well as radiator area requirements. The famed Michael
13 x 42 m in size, broken into 4 smaller arrays of 13x10.5 Minovitch also studied terrestrial air scoops in detail. His
m which each follow the Sun. They are arranged length- 1988 patent for a self-refueling space propulsion system
wise along the orbit direction, with the thruster in the deals with a scoop that would operate in working fluid
back. Assume the thruster works no more than 30% of environments and operating between 50 and 100 km al-
the time while mining, to allow margin to eventually raise titude.
orbit. Therefore the average drag can be 1.7 N. Assume References:
a 13x10.5 m scoop at the front, which matches the solar
array dimensions, thus a 136.5 square meter area. The
• Reichel, R. H.; Smith, T. L.; Hanford, D. R.
atmospheric density must then be 2.3 x 10−10 kg/m3 ,
“Potentialities of Air-Scooping Electrical Space
which occurs at 200 km altitude. The scoop will col-
Propulsion Systems”, presented at the ARS Elec-
lect 0.23 grams/sec, and the thruster will consume 0.034
tric Propulsion Conference, Berkeley, California,
grams/sec, leaving a net of 0.196 g/s (16.9 kg/day). Even-
March 14-16, 1962
tually whatever storage tanks you use are full, and the
mining ship increases thrust and climbs to an orbit where • Minovitch, M. A. Self-refueling Space Propulsion
it is not seeing significant drag. If the tanks hold 5 tons System And Operating Method. Patent 4,754,601.
of air, the mining ship can collect it in 300 days. The July 5 1988. Print.
question is then what the mass of the mining ship, to de-
termine a mass return ratio.
Mining Liquids
Design Concept - The illustration shows a general con-
cept for the scoop miner. The scoop is shaped as a hyper- A few bodies, the Earth being a notable one, have sur-
bolic cone and functions similarly to a turbo-molecular face liquids, and others are known or suspected of having
pump by bouncing incoming molecules off its surface. subsurface liquids. Collecting surface liquids in bulk is
Once the density is sufficiently high for the air to act as again a straightforward matter, needing a pump and stor-
a fluid instead of individual atoms, a stagnation region age tank. If the liquid is a mixture of compounds and a
will form at the narrow end, which a conventional vac- particular component is desired, then physical or chem-
uum pump can collect from. Solar arrays are mounted ical processes need to be applied to separate the desired
behind the scoop so as to not increase drag, and are piv- component.
oted to follow the Sun as the vehicle orbits. At the rear
are a storage tank, and the electric thruster powered by Drilling into a body with a submerged ocean, like Europa,
the arrays. As a safety measure, the vehicle should carry might be hard, but seismic measurements should not be.
conventional thrusters and fuel to raise orbit if the main Place one or more seismic detectors on Europa’s surface,
electric thruster fails. and then smack it really hard, like with the upper stage
that launched the probe, and look at the vibrations you
Trolling for Air - Trolling is meant in the fishing sense, get. That’s how we know about the interior of the Earth
and not the annoying Internet person sense. In this ver- and prospect for underground resources. Boundaries like
sion, the electric thruster and main solar arrays are at a ice/water tend to reflect vibration waves (ie sound).
152 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

There will likely be natural vibrations from Europa flex- Boil-Off This method involves reversing the way the
ing and moving (Europaquakes), but you can't be sure of planet formed in the first place. Planets form by collapse
them, so better to have your own source of vibration from of a gas cloud as it radiates away energy. The largest So-
a high speed impact. lar System gas giant, Jupiter, is apparently still radiating
away excess heat today, after 5 billion years. If excess
solar energy is directed at a gas giant, it will heat up and
Mining Gas Giants reverse this process. Assuming the outer layers heat up
faster than the interior, this would puff up the atmosphere
Gas Giants have no solid surface. At a particular depth into regions where orbital and escape velocities are lower.
their atmospheres gradually become supercritical fluids,
rather than a distinct layer, so different mining techniques
are needed, especially if you want to access deeper mate- Mechanical Disruption This is a brute-force method.
rials. Scoop mining has been described above, for skim- One approach involves directing a large body at high
ming the upper atmosphere. Buoyant mining equipment speed at the planet. The other approach is to collect hy-
is possible but difficult. Gas giant atmospheres typically drogen, deuterium, or helium-3 and use them to make
are mostly Hydrogen and Helium, so it is difficult to de- a very large thermonuclear device. The end result to to
sign equipment lighter on average than the lightest two throw large amounts of material into planetary orbit or
elements. Reaching orbit is also difficult due to the thick- entirely disrupt the planet. It will be difficult to get mate-
ness of the atmosphere and depth of the gravity well. rial to only go into orbit and not escape completely, which
To access deeper materials, large scale and rather dras- is why this method is not recommended for systems that
tic methods would be needed such as spin-up, boil-off, are already inhabited.
and disruption. The last is not recommended for inhab-
ited stellar systems.
Mining Particulates

Spin-Up If scoop mining is insufficient in volume, and Stellar systems and interstellar regions contain small ob-
buoyant mining is too difficult, this method increases the jects, particles, and gases in low density distributions.
typically fast rotation rate of gas giants until the equator These include:
is moving closer to orbital velocity. This makes removal
of material from the equator to orbit easier. There are a
number of techniques for increasing the rotation rate: • Asteroid rocks smaller than 6 meters diameter and
mass below 200 tons, for which collecting is a more
apt term than mining.
• Aerobraking momentum transfer: - A vehicle
accelerates while in orbit around the gas giant, then • Smaller particles such as those which create the Zo-
deposits excess velocity by braking in the upper at- diacal Light, coming from asteroid and comet de-
mosphere, adding momentum to the planet. bris, which range down to 10 micrometers in size.
• Kinetic deposition - Very high speed objects are • A flux of particles called a stellar wind emitted
directed at the equatorial region of the gas giant, from stars.
such that the average rotation rate increases. De-
pending on impact velocity, the impactor may be • The interstellar medium, consisting of very low den-
absorbed by the planet, or material may be kicked sity gas and dust grains
out as ejecta.
• Interstellar gas clouds and nebulae of higher density
• Tidal coupling - Intentionally place one or more than the general interstellar medium, but still low
sub-synchronous satellites in low orbits to raise tides density by ordinary standards.
which will accelerate the gas giant’s rotation. This
method is very slow.
The challenge for mining these types of materials is their
• Magnetic coupling - If the gas giant has a magnetic low density, forcing large collection volumes in order to
field, react against that field to spin up the core, and collect significant amounts.
eventually the rest of the planet.

• Reaction motor - A large high velocity fusion en- Mining Stars

gine, powered by Hydrogen from the gas giant’s at-
mosphere, is mounted in the upper atmosphere, and If a civilization’s material needs are large enough, it may
accelerates itself against atmospheric drag, which consider mining the stars themselves, where most of the
ends up adding to the planet’s rotation. For this to mass of a stellar system is located. The temperature of
function, the exhaust plume needs to be able to leave stars is an obvious difficulty in doing this. The following
the planet. methods are speculative at this time:
4.7. PROCESSING AND PRODUCTION 153

Artificial Red Giant Class M stars have surface tem- Sun. If used in the shadow of a large body, some stor-
peratures below the melting point of the most refractory age in the form of batteries or other devices is needed, or
compounds, and so might be mined directly. If you sur- power must shut off temporarily.
round a star with reflectors such that most of the energy
is trapped, the outer layers of the star will heat up and
expand, creating an artificial red giant. When sufficiently Thermal Energy
expanded, you can skim off material and take it elsewhere
to be used Wind Energy

Plasma Jets Artificial magnetic fields might be used to Fission Energy

create flares or jets from the star. When sufficiently far
away the material is collected and used. Fusion Energy

Orbital Energy
Stardiver A sufficiently large planet is placed in a
highly elliptical orbit which grazes a cool star. The planet The potential and kinetic energy of bodies in orbit is a
will collect an atmosphere by gravity during the grazing. potential source of energy.
In the outer part of its orbit it will cool down and can be
mined, and the process repeats. The required size of the
planet is such that it does not lose mass from evaporation 4.7 Processing and Production
during the close pass.
The Resource Extraction section discussed how to ex-
tract raw materials and energy from the available re-
4.6.2 Energy sources. This section covers how to convert those into
finished inventory and components which can be assem-
There are numerous energy sources in the environment bled into completed systems and projects.
near stars. They are listed below in approximate order of
available power and ease of extraction. There are fewer Existing industry extracts materials and energy, produces
significant sources in the dark regions between stars, but bulk supplies and parts from them, and assembles them
we will consider what is available. into finished items entirely on Earth. The methods, how-
ever, are independent of location. We just happen to have
evolved on Earth, and so that is where industry first was
Sun/Star Energy set up. In principle, the same methods used on Earth
can also be used in space or on other planets and objects.
Photovoltaic Energy Photovoltaic conversion has In practice, we specialize industry on Earth for location
been by far the most common way to extract energy where it functions best, and the same will be true for loca-
tions beyond Earth. There are some additional methods
in space. This is due to simplicity, scalability, mass,
availability, and durability. Photovoltaic cells generate which only are suited to locations beyond Earth, due to
special conditions like available energy, lack of gravity,
electricity directly without additional conversions or
devices, They come in typical units of a few watts each, or vacuum. A 1993 survey of methods and uses specifi-
cally for space, Resources of Near Earth Space repre-
which can be scaled by simply using more of them.
They are fairly low mass per power produced, and sented the state of the art at that time. Progress has been
made since then.
do not consume fuel. Except in the shadow of large
bodies they can produce power 100% of the time, and This section will list the full range of available methods,
have operating life measured in decades except in high but thought needs to go into selecting which to use in
radiation environments. a given location and circumstance for a space project.
Cells for use in space operate on the same principles as Some Earth-based methods assume gravity or air pres-
ones made for use on Earth. The operating conditions are sure in their operation. Artificial gravity or atmosphere
different, so they are somewhat modified. Sunlight is not can be provided in space locations where it is not natu-
filtered by the atmosphere, so they work with a differ- rally present, but there is overhead and complexity in do-
ent spectrum. Temperature ranges, vacuum conditions, ing so. Thus in designing a space production facility, you
and higher UV and other types of radiation all have to should consider an alternate process that does not require
be accounted for. If higher power levels are needed, and special conditions, rather than automatically adding them
launched from Earth, panels of multiple cells need to be to a traditional process that needs it.
folded for launch, and then deployed. Spacecraft point- The scope of all known production methods is too wide
ing needs are typically different than the Sun’s direction, to fit in the confines of a single book. Indeed, the entire
so the panels have to be articulated so they can follow the fields of Industrial and Chemical Engineering are devoted
154 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

to this topic. We will give a summary of the available Mechanical Processing

methods, with pointers to more detailed information. A
typical industrial process uses multiple steps in series or Mechanical processing changes the physical state but not
in a more complex flow of operations with branches or the chemical composition of the materials
loops, under some method of production control. Here
Crushing - This is breaking down a material into smaller
we list individual steps, which are called Unit Opera-
pieces by applying pressure. Milling refers to making
tions in process engineering. The range of possible com-
finer powders. In addition to standard methods of crush-
plete processes by combining these steps is quite large.
ing, kinetic impact can be used as a process in space, one
The task of the system designer is then to select the proper
that Nature has applied extensively.
set of steps and complete processes for the task at hand.
Sorting - This is the sorting of material by size or type
using gravity, acceleration, vibration, electrostatic forces,
4.7.1 Production Control or magnetic fields. Besides physical wire mesh or perfo-
rated plates to sort by size, methods like electrostatic can
Processes do not operate themselves. Under the head- sort materials by charge-to-mass ratio.
ing of Production Control fall those elements that plan Mixing -
the production, send commands on what to produce to
the system elements that actually perform the work, and
monitor the status and outputs of the operations. Control Thermal Processing
can be either manual or automated, and either local or re-
mote depending on circumstances and design. Processes Thermal Processes Evaporation
can be repetitive or continuous, producing the same out-
Condensation
put at some average rate, or singular, where a different
part is produced each time. Heat treating
Crystallization
Computer Design Drying
Heating and Cooling

Process Measurement and Control

Thermal Sources Microwave Heating

Automation and Robotics

Refining and Separation

4.7.2 Handling and Storage Filtration - Separation of solids from liquids

Distillation
Materials and parts need to be moved from process to
process, or stored between processing operations or when
complete.
Chemical Processing

Ore reduction
Parts Transport
Alloying
Parts Storage Reaction

Bulk Transfer Synthesis and polymerization

Bulk Storage
Organics Processing
Factory Environment
Agriculture and Food Growth
4.7.3 Materials Processing
Organics Conversion and Storage
Materials processing converts raw materials from their
state as delivered from mining to finished bulk materi-
als such as water or oxygen, or ready stock such as bars, 4.7.4 Parts Production
rods, or sheet.
4.8. ASSEMBLY AND CONSTRUCTION 155

Forming and Molding Coating

Molding Painting
Blowing Coating
Casting Printing
Rolling Plating
Forging Dyes

Electronics Fabrication
Subtractive Fabrication
There are three main levels of electronics fabrication.
Mechanical Machining Shearing These are making the individual components, connecting
Sawing those into circuits, usually in board form, and assembling
them into finished units.
Drilling
Milling Component Fabrication
Abrasives
Circuit Fabrication

Electrical Machining Electric Discharge Machin-

ing Unit Fabrication
Electro and photo chemical machining
Plasma arc machining 4.8 Assembly and Construction
Assembly refers to putting together parts or component
Beam and Jet Machining E-beam & ion milling systems to make a functional complete device. Con-
Laser cutting struction refers to permanent installations not intended to
move often or ever and consisting of many devices and
Abrasive water jet systems. These tasks typically consist of many simple
steps in series.

Additive Fabrication
4.8.1 Assembly Methods
Extrusion
Assembly methods can be organized by the type of oper-
Vapor deposition ator and the type of connections are made between ele-
Powder forming and sintering ments
One method similar to sintering is to spray coat reinforc-
ing fibers with droplets of molten metal to form a rein- Operator Types
forced tape. Then the tape is applied in layers to build
up the shape you want. Each layer is heated to just be- Assembly methods can be categorized by the type of op-
low the melting point then pressed to the previous layer erator that performs the assembly:
to bond it. For making shapes like cylinders this can be a
continuous winding process. • Manual Assembly - This is performed by humans
using manual dexterity and strength. It can be as-
Gluing
sisted by tools and fixtures for leverage, speed, and
Welding positioning.
Brazing • Robotic Assembly - This is performed by robots ei-
Soldering ther under human remote control or computer con-
trol.
Fiber spinning
• Automatic Assembly - This is where the devices to
Weaving perform the assembly task are built into the compo-
Sewing nents.
156 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

Mechanical Connections Cast and Block Structures

Electrical Connections These are not pressure tight structures, but installed by
methods like casting concrete or stacking blocks
Fluid Connections
Fabricated Structures
4.8.2 Construction Methods
These are from metal, wood, plastic, or other materials
Construction can be organized by analogy to Earth con- delivered as components to be assembled on site
struction, with suitable modifications

4.8.3 Outfitting Methods

Construction Planning
Outfitting involves installing and activating items within a
larger construction element once it is in place.
Logistics

Delivering construction equipment and supplies to the Sealing and Protection

construction site from other locations.
Utilities Installation

Orbital Tugs When launching multiple components Equipment Installation

from Earth that need to be collected in one place for as-
sembly, if the orbits are similar, then a single vehicle ded- Habitation Setup
icated to the job is more efficient than including a propul-
Activation
sion and navigation system on each component payload
[1]
. We call that vehicle an Orbital Tug. Besides propul-
References:
sion, navigation, and means to grab the payloads, it needs
the ability to periodically refuel. This can be done with [1] Gralla and De Weck Strategies for On-Orbit Assembly
occasional payloads being fuel tanks, which the Tug at- of Modular Spacecraft, JBIS, vol 60, p 219, 2007.
taches to itself as needed. In order not to be left stranded,
it should have at least two tank locations, and install a
fresh tank while the second one still has some fuel left.
If you have more than one Tug, they should be designed
4.9 Verification and Test
to grab each other for heavier cargos needing more total
propulsion, or in case one stops working and needs to be Verification is proving that a system element or the sys-
returned for repair. tem as a whole meets the requirements imposed on it.
Test is one method to gather data for such proof. Other
methods include analysis and inspection. Since test is an
integral part of most verification processes, they are nor-
Materials Handling
mally grouped together. Verification and test is normally
applied at multiple levels of a design, starting at the com-
Positioning items within the construction site to the point ponent level and working up to the complete system. This
of use. is to discover problems at the lowest level possible, where
it is easier to identify the cause. If lower level problems
are resolved, any problems that occur at the next level of
Docking Mechanical Docking Systems integration are likely to be from the integration of multi-
ple components itself.

Deployment Unpacking a component shipped in a

compact state to an operating configuration. 4.10 Operation and Maintenance
Prior space projects have relied on a Mission concept of
Site Work operations. That is you plan a mission ahead of time, with
a detailed timeline of all the steps you expect to perform.
Where you are doing construction on or near a natural Then the crew and ground controllers execute the plan.
body, and you need to modify the construction area by If deviations happen, corrective action is taken to return
road-building, excavation, and other methods. to the plan as much as possible. This concept works for a
4.11. RECYCLING METHODS 157

single mission in isolation with a well defined goal. Future 4.10.3 Maintenance Concepts
projects will have multiple activities in multiple locations,
with larger numbers of people involved and interactions 4.10.4 Maintenance Tasks
between activities. So planning every step ahead of time
in a linear fashion becomes difficult if not impossible. For
cost reasons, you also want to reduce the large amount of 4.11 Recycling Methods
planning and training that currently goes on relative to the
actual doing. Most locations off the Earth do not have automatic recy-
cling of waste products, and a space project will have a
finite amount of materials that have been processed be-
4.10.1 Operations Concepts yond the raw state. We define Recycling as the appli-
cation of energy and processes to convert materials from
For future projects, rather than planning everything in ad- a non-useful state back to a useful state. It shares some
vance, one approach is to develop building blocks of tasks technology with production from raw materials, except
which can then be assembled as needed to reach a goal. using materials that have already been processed previ-
As new feedback comes in from other tasks, the building ously and are not in their raw state. In fact some recycling
blocks can be adjusted or replaced as needed. Changing methods will involve feeding waste products back into a
plans may not be as efficient as carrying out a fully de- production process.
tailed plan made in advance, but the effort to make those
advance plans may be larger than the efficiency loss. The
ability to change or replace tasks, rather than a monolithic Waste Recycling
preplanned mission, also allows continuing operations to
adapt and improve over time. A goal for recycling is to achieve Closure, where all waste
products are converted back to useful states and the only
external input is energy. We expect this to be a theo-
4.10.2 Operations Tasks retical limit similar to conversion efficiency in electrical
transformers and motors. Those devices can reach 98%
Debris Removal efficiency but not 100%. Similarly, practical recycling is
expected to reach a high percentage, but not 100%. To
This task is the removal of human-made orbital debris, the extent your percent closure CL% approaches 100, the
and, if possible, natural hazardous objects. Over time, amount of new raw materials required is reduced to 100-
debris will collide with itself and decay from drag. In CL% of an open system with no recycling. So a high
the long term it will clean itself up, but in the short term percent closure can have a dramatic effect on the need
the collisions will generate more hazardous size pieces, for new raw material processing or replacement items
which is an unsatisfactory situation. Besides clean up of brought from elsewhere.
past debris, this task takes care of cleaning up future de- Some items by their nature are not amenable to recycling.
funct, lost, or destroyed hardware. There are a number A prominent example is reaction mass expelled outside a
of methods which have been proposed to do this, none of gravity well by a propulsion method. You are deliberately
which have been demonstrated yet: throwing that mass away in order to get thrust, and outside
a gravity well there is no practical way to recover it to use
• A “plasma-puffer”, promoted by Daniel Gregory. it again. Within a gravity well, such as launching from
That would use an arc discharge from a high alti- Earth using a chemical rocket, all the reaction mass is
tude balloon to puff air upwards in the path of the sub-orbital - ranging from about −1/2 orbital velocity to
junk, making it de-orbit faster. +1/2 orbital velocity depending on the vehicle velocity.
Therefore it all ends up back on Earth and can be used
• Ballistic puffer, which launches a projectile from a
again. While some reaction mass cannot be recovered,
very large gun, that releases air or some other mate-
you can deliberately choose high efficiency methods that
rial in the path of the debris, having the same effect.
lose less mass this way, and use methods such as gravity
The projectile is sub-orbital, so it does not add to the
assist that do not lose propellant mass.
space junk problem.

• Lasers - zap the front of the debris, blasting off some Closed Loop Life Support
of it, and slowing it down from the reaction.

• Drag devices - attach a large sail or conducting wire Humans are biological organisms with certain require-
to the space junk and make it slow down faster ments to continue living, including clean air, water, and
sufficient food. They also produce waste products from
• Orbital tug - goes from orbit to orbit collecting junk. the human standpoint, such as CO2 , urine, and feces. On
This requires high efficiency electric thrusters or the Earth those wastes from the human standpoint are neces-
fuel used gets absurdly high. sary inputs to plants and other organisms, and solar pow-
158 CHAPTER 4. PART 3 - SPACE ENGINEERING METHODS

ered evaporation produces clean water in the form of rain. in space of small rock particles, water, and carbon which
The natural cycles form closed loops. For space projects make up a large part of soil mass.
we have to replicate the function of those closed loops
with artificial systems, or supply air, water, and food from
external sources. The more time and the higher the num- Food Human food requirements can be measured in
ber of humans the greater the outside supply mass be- terms of total energy, commonly measured in Food Calo-
comes, and so the more desirable a closed loop system ries which are equivalent to 4184 Joules of available en-
becomes. ergy. The requirements are also measured by a wide va-
riety of specific nutrients in specific amounts. It may not
Besides inputs such as air, food, and water, humans addi- be efficient to supply all the low mass nutrients in a given
tionally need controlled lighting, temperature, pressure, situation, while supplying the main ones by mass. What-
radiation levels, and acceleration. Prepared food and fe- ever nutrients are not obtained from the closed loop sys-
ces are complex from a chemical standpoint, there are tem would have to be supplemented from outside, or sim-
secondary volatiles and shed skin produced by humans, ply done without for short missions. Human requirements
and body and clothing cleaning are desirable. Therefore can be translated to specific growing areas per person us-
a full life support system for humans is complex. We ap- ing known data on agricultural productivity. This data
ply the Systems Engineering approach of dividing it into can be modified by designs of space systems. For exam-
simpler sub-systems that each perform part of the total ple, if a plant needs 12 hours of sunlight per day, you can
functions required, and then optimize the full system as grow twice as much in the same area by using trays that
a whole. In addition to the direct functions that provide are swapped every 12 hours, and storing the other tray
life support, such as growing plants producing food, there under the illuminated one.
are also indirect functions caused by meeting the require-
ments of the plants, such as water, illumination, gravity,
and C02 concentration. The methods listed below include
meeting both direct and indirect functions.

Artificial Gravity Humans and plants are evolved in a

1 g environment. Without gravity, human bones deterio-
rate and some plants may not grow properly. With grav-
ity, water circulation and dust settling operate by familiar
methods. If local gravity from a large satellite or planet is
not sufficient, artificial acceleration can be generated by
rotation.

Illumination Photosynthetic plants need sufficient il-

lumination of the right wavelengths to grow, and humans
need sufficient light to see and navigate. Having both
evolved on Earth with the Sun as the main source of il-
lumination, both use the wavelength band that the Sun
emits the most power at, known as Visible Light. There
are slight differences in intensity, day cycle, and wave-
lengths for various plants, and for what humans need to
see properly. Natural sunlight is easily obtained at many
locations in space, so it certainly should be considered if
available. Artificial lighting is available for circumstances
where natural sunlight is not sufficient, but that usually
requires a power source to operate. For a large area of
plants that amounts to a lot of power, so natural sources
are usually preferred, even if it needs to be concentrated.

Soil Plants on Earth often grow in soil, which is broken

down rock plus microorganisms, water, nutrients, and or-
ganic matter. Hydroponics dispenses with the inert rock
component, which can save mass for space projects. If ra-
diation shielding is required, though, the inert rock com-
ponent can serve a dual function. While fully developed
soils are only known from Earth, there are many sources
Chapter 5

Part 4 - Combined Space Systems

5.1 Complex Programs Changes in technology and economics over time will also
guide changes in design. Therefore in any large or long
duration program, a multiple system approach should at
5.1.1 Introduction
least be considered to see if it gives a better result. Some
specific reasons include:
In previous parts of this book we have discussed indi-
vidual systems which carry out purposeful functions. In
• Non-Linearity - A given transportation or engi-
Part 4 we will consider more complex programs, which
neering method often has a non-linear term in it’s
involve multiple systems. Multiple systems can exist and
equations - at least quadratic if not an exponential.
interact at one point in time. They can also and grow and
For example drag goes as the square of velocity, and
evolve as a group over an extended time, with new sys-
rocket fuel required goes as an exponential function
tems replacing older ones.
of velocity. Therefore it often is more efficient to
An optimized program often results in a design with break up the total job into components because the
multiple systems. Reasons for this are described below. sum of smaller non-linear terms is less than a larger
Large and complex programs can end up with a multi- value with an applied exponent.
level structure from the program as a whole, to individu-
ally designed systems. We commonly call a set of systems • Complex Needs - Humans have complex needs, and
that exist together at one time under one management, projects we wish to accomplish typically have mul-
and have a set common goal, a Project. When the set of tiple goals. This drives design solutions which use
systems has diverse management with shared goals, we multiple materials, devices, fuels, etc. Therefore no
can call each part a program Segment. When a program single technical solution is likely to best satisfy all
exists for a long time, with significant growth, evolution, the needs.
and replacement of older systems with newer ones, we
can call the program parts Phases. Projects, segments, • Economics - A single large “all or nothing” type
phases, and other pieces are then assembled to form the monolithic system, which requires a large up-front
larger program. The names and structure are somewhat investment, often turns out to be wasteful. Aside
arbitrary, and are selected according to the needs of a par- from the non-linear effects noted above, we cannot
ticular program. It is important that the selected structure predict future technology developments, and large
cover all the work needed for the program, and that all the projects often have long development times. If you
people working on it have a shared understanding of how build in smaller steps, you have the opportunity to
the structure and its parts fit together. change direction if new developments come along,
Despite the complexity or duration of a large program, or retrofit an improvement to just the part that needs
the principles of Systems Engineering can still be applied it. Finally, with an incremental project, you can also
to optimize the overall design. When the duration is long get use out of it sooner, which can produce a higher
relative to changes in technology, society, and the en- return in the purely economic sense.
vironment, the Systems Engineering tasks may need to
done repetitively or continuously to get the most benefit. Choosing between single or multiple systems should not
be done arbitrarily ahead of time. The decision should,
if possible, be made by analysis of the alternatives and
Reasons for Multiple Systems choosing the best one. Sometimes the right choice may
not be possible for non-engineering reasons. For ex-
In general, the more complex, or more extended in loca- ample, the national civilian space program in the US is
tion, volume, traffic, or time a project is, the more likely operated as a whole by NASA, and funded by congres-
it will result in multiple systems. Different environments sional appropriation. A single agency funded from a sin-
and available resources will drive different local solutions. gle source can make it more difficult to divide or set up

159
160 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

separate systems, even if it makes sense from an engi- • In the following Design Studies section (Part 5) we
neering standpoint. The organizational structure, budget can teach by example, showing in some detail how
review process, and national politics tend to favor single the calculations and decisions are made to arrive at
and conservative solutions. Conversely, the historical di- a design.
vision of US government-funded space activities between
NASA, and the Defense, Commerce, and Energy depart- • It is intended to be a realistic program proposal,
ments can make it more difficult to combine projects and incorporating the best current technology and con-
programs, even if they would be more economical. cepts. If good enough, the proposed elements may
actually be built.

• An open-ended program using new concepts and

5.1.2 Organization and Content of Part 4 technologies allows readers to do actual useful new
work. At the same time they can gain individual de-
A major intent of Part 4 is to use the methods described sign skills and experience, and practice working in
earlier in the book to teach by example. We therefore diverse teams.
present an extended example of a long-term and com-
plex program that involves multiple terrestrial and space The program concept is partly based on work by Dani
systems. Unlike printed paper books, the example in this Eder, the original author of the Canonical List from
wikibook does not have to remain static, but rather can be which this Wikibook originated. It also includes many
developed over time. The completed portions will show new ideas not yet pursued by government programs. In
how calculations and decisions were made. There are op- its present state it is not complete or claimed to be the
portunities for individuals or teams to develop the ideas best possible program concept. Rather it is intended as a
further, and practice their design skills. Their work can starting point using recent ideas for space development,
be recorded as design studies in Part 5, or as separate doc- from which an optimal design can evolve by the contribu-
uments to be linked. The best ideas can be incorporated tions of many other people. We plan to use the Systems
back into the main discussion of the book. It is hoped this Engineering methods described in Part 1 in further de-
approach is a useful teaching method, a way for readers veloping the proposed program.
to gain real design experience, and a way to make real
progress in in future programs.
The wikibook as a whole was about a 60% complete first Part 4 Contents
draft as of Sep 2012. Part 4 is still incomplete, but is
undergoing major revision as of early 2017. A large and complex program naturally can't be described
in a single page. So the remaining sections in Part 4
provide further information about the Upgrade Program
Our Program Example in roughly chronological sequence, according to program
phases. Design studies in Part 5, and related information
elsewhere, provide more details about design choices and
Our chosen example has the overall goal of upgrading
calculations.
civilization on Earth and expanding to more difficult en-
vironments, including all areas of the Solar System and
eventually beyond. For now we will call it the Upgrade • Section 4.1: Program Overview - Summarizes the
Program (UP) for short. The program’s component program concept as it has been developed so far.
parts are not intended to be under a single centralized
control. This is both to illustrate the interactions between
Design Studies and Related Information
distributed program elements, and because it is imprac-
tical to run such a large program as a single entity in the • Section 5.1: Program Conceptual Design - This
real world. Each part of the program then needs its own is a design study showing how the conceptual design
reasons to proceed, whether they are economic or other for the Upgrade Program is developed. It is more
motivations. The program’s description and supporting extensive than a typical technical study final report.
concepts help provide these reasons, and a structure to Final reports record the results of a study. Here
inform and coordinate independent efforts. The various we also show the logic and calculations to reach the
parts would then interact with each other, and with the results, so that others may learn from and improve
outside world. The interacting parts, and the program as upon them.
a whole, are not static, but evolve over time.
We chose this example for several reasons:
5.2 Program Overview
• As a long duration and wide ranging program it al-
lows us to demonstrate design methods for many This section provides an overview of the Upgrade Pro-
types of systems and subsystems. gram (UP). Later sections of Part 4, design studies in Part
5.2. PROGRAM OVERVIEW 161

5, and related information (located elsewhere and linked) • Optimism for the future by demonstrating we are
provide more details. The program concept is incomplete not in a finite, closed world. An optimistic viewpoint
as of the start of 2017, and readers are invited to con- in turn changes how people act.
tribute to it and improve it. Conceptual and Preliminary
Design, which are the early stages of overall design and These benefits of the program accrue to civilization as a
development, have not been finished. Therefore multi- whole, although the specific projects may be funded and
ple candidates are often presented, with selection of the carried out by smaller organizations.
preferred options to be done later.

5.2.2 Program Summary

5.2.1 Program Goals
Our current civilization significantly uses only 13.5% of
Every program needs identifiable goals and objectives to
the Earth’s surface. The biosphere plus the human-built
direct the design and implementation work. Section 5.1
environment averages about 200 kg/m2 if distributed
is a study to develop these goals and objectives, and the
evenly across this area. This is a tiny fraction of the 11.71
initial design concept to meet them. The most compact
billion kg/m2 of the Earth’s total mass. The known major
statement of the program’s goals are:
planets and smaller bodies of the Solar System amount to
Upgrade civilization, starting on Earth, and progres- 447 times more than the Earth alone. In terms of thick-
sively expand to more difficult environments, includ- ness, our current civilization amounts to a 20 cm thick
ing Space beyond our planet. layer on the portion of the Earth we use. This is a very
Program objectives supporting this goal include: thin veneer compared to the bulk of the Earth. The layer
thickness assumes the contents average the density of wa-
ter (which living things approximately do) and were flat-
• Improving Life on Earth by developing better tened and distributed evenly.
technology to make material goods and to live sus-
tainably from local resources. Our civilization uses about 18 TW of energy in all forms.
Ignoring other energy sources, this is a small fraction of
• Expanding Material and Energy Resources by the 174,000 TW of sunlight that falls on the Earth, and
access to currently difficult Earth and space loca- a microscopic fraction of the 383 trillion TW the Sun
tions. produces. Problems like poverty and material scarcity
don't stem from a lack of resources. They exist because
• Increasing Biosphere Security by adapting to we hardly use the vast amounts of material and energy
more difficult environments, including future varia- resources that are there. Our program’s approach can
tions in the Earth itself, and countering undesirable then be simply explained as “using more of what’s al-
changes. ready there”. Of course, we want to do so responsibly,
and minimize side effects to the biosphere and people’s
• Reducing Hazards from Space by identifying
current lives.
what they are, followed by developing methods to
deal with them.
General Approach
• Understanding the Earth Better by observing our
home planet, its environment in space, and other
Historically, people first occupied the easiest environ-
planets and environments.
ments, and used the easiest sources of materials and en-
• Long Term Survival by dispersal to multiple loca- ergy. Our program concept is to continue this path, but
tions and acquisition of critical new resources. apply modern technology towards occupying and using
progressively more difficult environments on Earth, and
• Increasing Choice and Freedom by opening un- then in space. We can leverage existing production equip-
occupied locations to habitation. ment and automation to make more of itself, starting
from existing locations where people live. Starter sets of
• Increasing Opportunity by access to unclaimed equipment are then sent to the next harder environments,
resources and more efficient technology. where they use local energy and materials to grow. When
they have expanded enough, they can then contribute to
Benefits resulting from such a program include: starting in even harder locations. People accompany the
starter sets when the environment is not too difficult. In
• Low cost access to space, removing a current bar- the more extreme and remote locations the equipment op-
rier to its use. erates automatically or by remote control until suitable
space for people is built. In this way civilization can pro-
• Spin-off technology from doing difficult tasks, gressively upgrade itself where people already live, ex-
which can then find uses elsewhere. pand horizontally across areas we mostly don't use yet
162 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

(i.e oceans, deserts, and ice caps), and vertically down other at all levels, and the rest of civilization, via an as-
into the Earth and up into space. sortment of inputs and outputs.
We expect the program to be carried out as many sep- The first four phases are on Earth. These are:
arate projects, each being carried out for its own rea-
sons. For the purpose of engineering analysis, design, • Phase 0: Research and Development
and optimization. though, we will consider the program
• Phase 1: Starter Locations & Network
as a whole. This is similar to how airplane design has to
consider the airports and traffic control system they op- • Phase 2: Distributed and Industrial Locations
erate with, even though the airplane manufacturers, air-
• Phase 3: Difficult and Extreme Locations
lines, airports, and traffic control network are all oper-
ated by different organizations. By presenting the entire The Research & Development Phase supplies necessary
program, we hope to provide vision and inspiration for technology and designs to the later phases, and therefore
people to follow chosen parts according to their interests comes first. Phases 1, 2, and 3 are distinguished by scale
and abilities. of operations and moderate vs difficult or extreme operat-
ing environments. Different scales and environments will
lead to different specific designs, so we define phases for
each. Smaller scale equipment can be used to build larger
equipment, and gathered in numbers to make larger pro-
duction systems, hence the sequence from starter to dis-
tributed and industrial phases. The easiest places to start
are developed areas with moderate conditions, so the first
three phases begin there. Phase 3 builds on that experi-
ence and therefore comes fourth, in harder environments.
If you wonder why four out of seven phases in a book
about space systems are on Earth, keep in mind that most
current space activity occurs on the ground. For example,
six people occupy the International Space Station in or-
Figure 4.1-1 - Program Phases vs Time.
bit, but the US portion of the project alone employs about
12,000 people on Earth.
The remaining three phases are in space, which are typ-
ically harder than the most difficult locations on Earth.
This is due to lack of breathable air, radiation, tempera-
ture, and the energy and travel times to reach these loca-
tions. These phases are:

• Phase 4: Orbital Locations

• Phase 5: Planetary System Locations
• Phase 6: Interstellar Locations

Orbital Locations begin with those nearest the Earth,

Figure 4.1-2 - Sequence of program upgrade and expansion progress by distance outwards to higher orbits, then to
phases. ones around the Sun from inner to outermost. Planetary
System Locations are tied by gravity to the Moon and
The program concept is a progressive series of upgrades major planets, either in orbit around them, or on their
and expansion to new environments, with previous loca- surfaces or moons. Finally, Interstellar Locations are not
tions providing the means to open up the next ones. Ex- tied to our Sun and Solar System. The order of Phases 4,
isting environments continue to be occupied and used as 5, and 6 is according to distance and difficulty of reach-
new ones are established. Therefore we organize the pro- ing their locations, since equipment for a given location
gram into a number of phases, with staggered starting has to come from or through the previous ones.
times (Figure 4.1-1) and a sequence for which phases lead Most of the major phases are divided into more specific
to the later ones (Figure 4.1-2). Once started, a phase sub-phases: 2A, 2B, 3A, 3B, etc. The sequence of growth
continues in parallel with earlier phases, and the rest of and expansion is intended to be self-funding once started.
civilization outside the program. In turn, phases are di- Self-expanding production and other new technologies
vided into smaller projects, locations, and functions, with should be valuable enough to justify their development
engineered systems and subsystems which perform the on their own, and using those technologies on Earth can
functions. The parts of the program interact with each pay for the later steps.
5.2. PROGRAM OVERVIEW 163

Phase 0: Research and Development (R&D) the program. One example is distributed production net-
works. Traditional factories brought people and equip-
ment together in one place because it was the only way
Existing civilization has already accumulated a large
to efficiently organize the work. With modern commu-
amount of technical knowledge, designs, and equipment.
nications and transport it is possible to coordinate work
There are some new items that are needed for the various
that is distributed across many places. This has been
program phases, so the R&D phase will do the neces-
achieved for software development, but needs improve-
sary work to develop them. This includes conceptual and
ment for physical goods. Areas that need improvement
preliminary design, component research, detailed design,
include remote operations, and automatic process com-
prototype fabrication, and testing. Finished items are
pilation to support an ever-changing set of equipment.
supplied to later phases, and also to society at large when
appropriate. New knowledge and technology from this In space technology, current programs still do not fully in-
phase is also contributed back to civilization as a whole. corporate already existing knowledge and hardware, such
The R&D Phase is divided into multiple sub-phases, ac- as electric propulsion. They don't yet incorporate the
cording to the later phases for which the R&D is needed. most basic in-space production methods, although some
This includes for Phase 0 itself, in case unique facilities, research is ongoing towards it. Many new ideas languish
equipment, or processes are needed to carry out the R&D from lack of funds. TSo the R&D phase will follow a
work. dual approach. First is to make the best use of what al-
ready exists in planning a coherent program. Second is to
Seed Factories
use some of the production capacity we develop on Earth
The first major technology we have identified for Phase to build and test equipment for the new and yet-untried
0 is that of Seed Factories. Enough work has been done ideas. Many of these ideas are listed in Parts 2 and 3 of
on this concept to start a separate book. Seed factories this book. Since some of the space technology can only
are starter sets of production equipment which can up- be tested fully in space, later parts of the R&D phase will
grade and expand by making more equipment for them- involve locations in space.
selves. The current state of technology does not allow a
seed factory to be fully automated or make 100% of its
own parts, so it is not entirely Self Replicating. Instead, Phase 1: Starter Locations & Network
our goal is to reach a high and increasing level of self-
production and automation. When expanding to a new Early Phase 0 R&D, Phase 1, and Phase 2 all occur in
environment or location, you bring a starter set, and it moderate locations on Earth. We define moderate con-
sets about using local materials and energy to make parts ditions for a number of environment parameters such as
for new equipment. Like any other factory, the growing temperature, water supply, and air pressure. Each param-
set of equipment also makes useful end-products. The eter is moderate where the middle 90% of people live
farther away and more difficult the environment, the big- currently, with 5% at each extreme. If any parameter
ger the advantage of producing locally has over bringing is outside the moderate range, then the whole location is
everything from somewhere else. Even in existing loca- no longer moderate. The moderate range is intended as
tions, a starter set would be less expensive to buy than “typical” or “normal” conditions to design for, such that
a full factory. So seed factory technology has economic a single design should be able to operate there. Moder-
justification even on Earth. ate conditions are measured by the natural exterior envi-
In already developed areas, the growing factories can also ronment, and not the interiors of controlled production,
take local waste and scrap as inputs, besides local raw habitation, or transport spaces.
materials. This provides environmental benefits. In both Phase 1 begins the self-growth process and builds the first
developed and new locations they whatever inputs they
operational seed factory equipment. Early equipment in-
get and convert them to finished materials, then finished cludes conventional tools and machines that are bought
parts and complete products. Whatever items the factory or made, plus starter set machines designed in the R&D
cannot efficiently make for itself are supplied from else- phase. Starter locations are in already developed and pop-
where. Outside supplies would be a larger percentage of ulated areas which can supply whatever else is needed to
total output at first. As the factory grows, and adds more get started and operate. Phase 1 machines are typically
machines and capabilities, the percentage from elsewhere small, such as for hobby crafts and home improvement
would decrease. A small percentage of items use rare scale uses. This makes them affordable for individuals
materials not found locally, or are not economic to make. or small groups. The machines can be located in homes
These continue to be supplied from outside, even after or locally-built workshops. Besides direct use for small-
the factory has matured. All the locations within the pro- scale production in this phase, the experience will be use-
gram, and the rest of civilization, then make up a linked ful for later phases. More remote and difficult locations
trading network. will be easier to bootstrap from small starter sets.
Additional Technologies Phase 1 also develops experience in other areas. One is
Some other new or upgraded technologies are needed for creating a distributed network of individuals and small
164 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

groups, with equipment in multiple places. As a coordi- ers locations that move beyond hobby and
nated effort to produce useful products and services, it home improvement levels of effort within the
can still be considered a factory or business enterprise, network, to small business and commercial lev-
just not a traditional one in a single place. Another area els capable of providing most of an individual’s
is building the knowledge and skills of the people us- financial support, and a wider range of products
ing the equipment. Small-scale production isn't taught to and services provided to people outside the net-
most people in their general education, so along with seed work.
equipment and software, we need education and training
for the users. Phase 2B: Industrial Locations - These
We expect Phase 1 operators to mostly use their skills serve larger and more widespread markets at
and machines for themselves, and to make items for the most efficient levels of size and specializa-
each other. Commercial-scale production with signifi- tion. This scale may require outside funding for
cant sales to the public would fall to the next phase. Since land and equipment, where earlier phases could
the starter sets are designed for upgrade and expansion, be self-financed through internal growth. This
the evolution to the next phase can be natural and gradual. is because buying and developing an industrial-
Within this phase, new network nodes and new locations scale site is hard to do in small increments.
can be started with partial equipment sets sent from ex- Because sources of supply and customers are
isting ones, or purchased from outside. more widespread, transport capacity becomes
more important. Outside market forces also
become more important than the internal needs
Phase 2: Distributed and Industrial Locations of the community. One way to deal with
market forces is distributed ownership across
As numbers of people, their skills, and equipment accu- multiple industries, so that people’s work and
mulate in a given location, they can start to sell and trade equipment can be reassigned as needed.
beyond their own community. They can then get most of
their financial support from network operations, where in
Phase 1 it was more hobby and part time levels of effort. Phase 3: Other Earth Locations
To the extent people can support themselves, it relieves
conventional risks of job insecurity and displacement by Difficult and extreme locations on Earth are defined as be-
automation. In less developed regions it would mean an ing respectively more than 10 and more than 20% beyond
improved quality of life. moderate conditions in any of eleven environment and
development parameters. See Section 5.2 Environment
One way to reach these goals is to use existing machines Ranges for details on the parameters. These locations are
to make parts for larger machines. This scales the equip- still on Earth, but include remote, unpopulated, and un-
ment from home and hobby, to small business, commer- developed areas on the surface. They also include areas
cial, and industrial size. More extensive use of the equip- away from the surface, such as deep underground below
ment puts an emphasis on higher duty cycles (percentage existing cities, or at higher altitudes, which may be phys-
of the time it is in operation) and service life (total hours ically close to moderate surface locations, but harder to
of operation). Where light-duty equipment is adequate at build in. The more difficult and extreme the conditions,
a hobby level, heavy-duty equipment is preferred at indus- the more that designs have to be modified to accommo-
trial levels. Both scale and intensity of use require mod- date them. This is accomplished by either modifying the
ified designs from Phase 1, and will require continuing working equipment, or moderating the local operating en-
R&D to develop. At smaller scales it is feasible to gather vironment to a suitable range. Because of the modifica-
the full range of factory machines in one place, and make tions we divide this phase into two sub-phases, based on
a wide range of products. At larger scales, the equipment the amount of changes required. Since these locations are
and their operators are more distributed and specialized, generally undeveloped and unpopulated, starter sets can't
and serve larger markets. Location designs and produc- be obtained locally. So they would be sent from previ-
tion flows are therefore also modified to account for the ous locations. They would not typically be small scale,
changes. because local operators don't have other jobs to support
We therefore divide Phase 2 into two sub-phases based themselves, and the higher cost of transport and support-
on scale and specialization. Not everyone will upgrade ing on-site people or remote operations would make small
to larger scales and intensity of operation, so Phase 1 scale equipment uneconomic.
continues to operate, and there would be a mix of small, This phase allows us to extend civilization to the 6/7ths
medium, and large equipment. Phase 2 locations are in of the Earth’s surface we don't yet significantly use, and
the same moderate environments as Phase 1, with devel- vertically both down into the ground and oceans, and up
oped and populated areas nearby. to higher altitudes. The world’s population is expected to
continue growing in the 21st century. People in currently
Phase 2A: Distributed Locations - This cov- less-developed areas also want to reach higher levels of
5.2. PROGRAM OVERVIEW 165

development. Expansion to new areas will to some de- considered extreme. Ground and water pres-
gree be needed to satisfy all their needs. This is not so sure above 3 MPa require extra structural sup-
much a need for personal space, since on-going urbaniza- port, and are found at depths below 300 me-
tion concentrates people in compact clusters. Rather, it ters in water and 120 meters in rock. Gravity
is to access increased energy and material resources. An is never extreme on Earth, but some high back-
example project for these kinds of locations is to estab- ground radiation areas and high altitudes over
lish greenhouses and water supply systems in deserts to the magnetic poles exceed the 21 mSv level to
increase food supplies. Automation and remote control be considered extreme. Ping time over 125 ms
allows such locations to operate, even if a lot of people occurs if modern communications are absent,
don't want to live there. The sub-phases are: and travel times over 3 days if conventional
transport isn't available, and are considered ex-
Phase 3A: Difficult Earth Locations - These treme, as are residence times below 3 years 4
are defined as having at least one parameter months, and cargo transport energy above 3.5
10% or more beyond the moderate range, mea- MJ/kg.
sured linearly or logarithmically depending on
the parameter. The parameters are measured
Phase 4: Orbital Locations
before local development and upgrade. For
example, reaching deep underground is diffi-
We already explore and use space beyond the Earth to
cult before suitable tunnels and shafts are built,
some degree, but that use is limited by the difficulty and
but not afterwards. Difficult temperatures are
expense of lifting everything needed up the Earth’s deep
average winter lows below −18C or average
gravity well. The orbital locations in Phase 4 exploit the
summer highs above 42C, such as arctic or hot
energy and materials already located in space to make
desert areas. Drier deserts and wet rainforests
things locally. This saves the cost of having to bring it
have difficult levels of water supply - too little
all from Earth. Production capacity from the previous
or too much. Altitudes above 2750 meters be-
phases is used to build things like rockets and launch sites,
gin to cause problems for some people, and soil
which can deliver starter sets of equipment to space. The
strength below 0.19 MPa or ground/water pres-
starter sets then bootstrap space industry in a range of lo-
sure above 2.5 MPa are difficult to build for.
cations starting near Earth, then into “open space”, away
Energy supply, primarily from wind and solar,
from strong gravity wells. These locations share high lev-
below 125 W/m2 makes all activities requir-
els or full-time sunlight for energy, low gravity or ease
ing power difficult. Locations well below the
of moving masses, and vacuum which can enable cer-
surface are cut off from these sources. Grav-
tain production methods. These shared conditions lead
ity is nearly constant on Earth, and most places
to shared designs, so they are grouped into one phase.
are below 17 mSv/year in background radia-
However, with varying distances from Earth and the Sun,
tion, and thus not difficult in these parame-
there are increasing transportation, thermal, and power
ters. Communications round-trip (ping time)
supply difficulties, leading us to identify six sub-phases.
over 100 ms and normal one-way travel time
Each of these leads to the next in sequence, with products
for people that takes over 2.5 days are consid-
from one phase used to help deliver and set up locations
ered difficult. Average residence times below
in the next.
5 years and cargo transport energy above 2.85
MJ/kg impose extra transport burdens from Orbital production would first support existing space in-
staff turnover and difficulty of outside supply. dustry, which amounts to 1400 satellites and $334 billion
in economic activity as of 2015. As bootstrapped pro-
Phase 3B: Extreme Earth Locations - Ex- duction lowers costs, current markets should expand, and
treme locations use the same parameters, but new one become economic. So, like previous phases, this
are at least 20% beyond the moderate range, phase should be self-supporting once started.
up to the limits of technology. So average
daily lows below −23C or highs above 47C Phase 4A: Low Orbit Locations - We de-
are extreme, and only found in severe cli- fine Low Earth Orbits as extending from 160
mates, at high altitude, or deep underground. to an average of 2700 km above the surface.
Water supply below 0.12 or above 3.8 me- The lower bound is set by significant atmo-
ters/year, and air pressure below 70 kPa or spheric drag, which prevents stable orbits. The
above 120 kPa (+5500m and −1600m alti- upper bound is half the potential energy from
tude) are also extreme conditions. Soil strength the lowest orbit to Earth escape. It is stated
below 0.12 MPa, which includes water surface as an average altitude, because elliptical orbits
of no strength, are extreme conditions to build are possible, which vary constantly in altitude,
on. Therefore all of the ocean surface where but not in total energy. Most low orbits are in
it is too deep to build from the sea floor are the Earth’s shadow from 22-40% of the time,
166 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

which lowers available solar power, and typ- So material and energy resources are widely
ically requires power storage for the time in available for production and other purposes.
shadow. Temperatures and lighting are moder-
ated by the nearby Earth, and communications Phase 4D: Main Belt and Trojan Locations
and travel times are relatively short. The natu- - This region includes orbits averaging from
ral environment includes the inner part of the 1.8 to 5.2 AU from the Sun, except for the
Van Allen radiation belts, and a modest flux part within 20 million km of Jupiter. Condi-
of meteoroids. Material resources include the tions are similar to the previous phase except
upper edge of the Earth’s atmosphere, and arti- that solar flux varies from 31 to 3.7% of that
ficial space debris. Other materials would have near Earth. At the outer reaches, solar reflec-
to be imported from elsewhere. tors or other power sources such as nuclear may
be needed to maintain temperature and supply
Phase 4B: High Orbit Locations - High power. Communications times can range up to
Earth Orbits extend from 2700 km average al- 2.88 hours in the worst case, and travel take
titude to the limit of the Earth’s gravitational 12 years by the most efficient orbit. Material
dominance at about 1.5 million km. Although resources include the over 700,000 members
this is a large range in distance, it only rep- of the Main Asteroid Belt, and the Hilda and
resents the upper 25% of energy between the Jupiter Trojan groups. Their combined mass
Earth’s surface and escape. It excludes dis- is on the order of 3 billion billion tons, all of
tances within 35,000 km of the Moon and the which is accessible to sufficiently determined
Moon itself, which are assigned to Phase 5A mining operations. These materials can be sent
below. High orbits are in sunlight 78 to 100% slowly but efficiently to inner regions by means
of the time, and temperature is mostly gov- of gravity assists from the major planets. The
erned by the Sun. Communications and travel farther regions contain large amounts of wa-
times at the outermost edge can extend to 10 ter ice and other volatile compounds because
seconds ping and 7 months transit by the most of the low temperatures.
efficient route. Like lower orbits, the natural
environment includes the remainder of the ra- Phase 4E: Outer Interplanetary Locations
diation belts, and outside the magnetosphere - This region includes orbits from 5.2 to 50 AU
solar and galactic radiation sources. Meteoroid from the Sun, except for regions around Saturn,
flux is similarly modest, as it is for the rest of Uranus, and Neptune. Solar flux is quite low
the Solar system. Material resources in place in this region, from 3.7 to 0.04% of that near
consist of a smaller amount of artificial debris, Earth. Nuclear power sources or very large and
but high orbits are fairly accessible to the Moon lightweight solar reflectors will be needed to
and Near Earth Asteroids. Materials can be de- supply power and keep warm. without those,
livered from these locations, and then the high objects will naturally be at temperatures of
solar flux used for local production. −56 to −200 C, depending on color. Round-
trip communications time ranges up to 1.15
Phase 4C: Inner Interplanetary Locations days, and travel times would be up to 350 years
- This region includes orbits from as close to by the most efficient orbits. This is too long
the Sun as technically possible to an average of to make economic sense, so actual travel times
1.8 AU, where the Main Asteroid Belt starts. will depend on the transport methods available.
It excludes the four major planets Mercury, There are about 1750 known objects in the
Venus, Earth, and Mars and the gravitation- Kuiper Belt beyond Neptune (30-50 AU), and
ally bound regions around them. These orbits another 340 Centaurs, whose orbits cross one
are in sunlight 100% of the time, and solar of the gas giants. They include Pluto and sev-
flux varies from 31% to many times that near eral other dwarf planets, with a total mass in
Earth. Even the lowest levels match the best this region of 4-10% of the Earth’s. The re-
places on Earth, because there is no night or gion is rich in water and frozen gases due to
atmospheric absorption to reduce it. Temper- the low temperatures.
atures vary with amount of solar flux, requir-
ing sunshields or insulation to moderate them. Phase 4F: Scattered, Hills, and Oort Lo-
Communications time is up to 1 hour round- cations - The final orbital region is the vast
trip across this region, and travel time can be one extending from 50 to 100,000 AU from
several years by lowest energy transfer orbit. the Sun. This region shares extremely cold
There are over 16,000 known objects in this temperatures and being close to Solar escape
region as of April 2017. The largest member energy. We divide it into three parts by dis-
alone has an estimated mass of 17 trillion tons. tance - The Scattered Disk from 50 to 2000
5.2. PROGRAM OVERVIEW 167

AU, Hills Cloud from 2000 to 10,000 AU, and Conversely, Lunar Geology indicates a high
Oort Cloud from 10,000 to 100,000 AU. Our percentage of metallic oxides, making it a good
ability to detect objects in this region is poor, source of oxygen and various metals.
due to distance, and dimness from lack of sun-
light. Only about 335 objects in fixed orbits Phase 5B: Mars Locations - Mars loca-
are known, and a small number of long-period tions include the planet itself, and orbits up
comets. We expect many thousands more to 340,000 km from its center, including the
await discovery, including a possible planet the moons Phobos and Deimos. It’s orbit is ec-
size of Neptune. Total mass in this region is centric, so solar flux varies from 36 to 52.5%
not known, but may be a large multiple of the of the near-Earth reference amount. Surface
Earth’s. Travel, communications, and power- gravity on Mars varies from 3.68 to 3.74 m/s2
ing of equipment would be very difficult with (37.5% of Earth), and atmospheric pressure
current technology. So significant develop- varies from 30 to 1155 Pascals (1.14% of
ment will await future improvements in those Earth max). Mars locations would follow phase
areas. 4C: Inner Interplanetary, since you must travel
through interplanetary space to reach Mars.
Determined mining can reach depths of about
Phase 5: Planetary System Locations 20 km, providing 2.9 billion cubic km of ac-
cessible resources, about 50% more than the
Planetary system locations differ from orbital ones in be- Moon. Mars retains significant water, and has
ing tied to relatively large gravity wells, thus requiring ad- a quite varied geology.
ditional transport capacity to traverse. They also can in-
volve shadowing/night when close to or on the surface, Phase 5C: Venus and Mercury Locations -
and higher radiation from belts. The surface of large These locations include the two planets, and or-
bodies have significant gravity levels, and sometimes an bits within 600,000 and 100,000 km respec-
atmosphere. All of these conditions are different than tively. A relative lack of nearby asteroids or
the open space of orbital locations, leading to a differ- moons, higher delta-V to reach their orbits
ent phase with different designs for them. The various and land on them, and extremely hot or high
planetary systems also differ from each other, requiring pressure conditions places these locations after
designs to accommodate each. So we provide five sub- Mars. Their advantage consists of 1.9 and 4.6-
phases to cover the range of such locations. The start of 10.6 times more solar flux, providing ample en-
sub-phases for Orbital and Planetary locations overlap in ergy to extract and process resources. The lack
time, due to the need to travel through various orbits to of asteroids in this region is due to the short or-
reach the large bodies. bit times, which allow frequent flybys. A given
asteroid will then be removed by either hitting
Phase 5A: Lunar Locations - Lunar loca- a planet, shifting the orbit outwards, or shifting
tions are those within 35,000 km of the Moon’s it so close to the Sun it vaporizes. New objects
center. The Moon is relatively close in physi- are supplied by gravitational changes from far-
cal distance to Earth, but the Moon’s surface ther orbits.
requires significant additional energy to reach
due to it’s gravity well. Therefore it comes af- Phase 5D: Jupiter System Locations - The
ter Phase 4B: High Orbit Locations in terms Jupiter system includes the largest planet, 67
of difficulty. Communications and travel times known moons, four of which are very large,
are relatively short, less than some parts of and orbits within 20 million km. The larger
High Orbit. Available sunlight on the Lunar moons can perform gravity assists, making
surface is 50% of that for high orbit, and can travel between locations easier, but the great
reach 99% for the highest circum-lunar orbits. mass of Jupiter makes reaching the planet it-
Average temperatures are similar to that on self very hard. Development of Jupiter would
Earth, but areas exposed to vacuum can cy- logically follow Phase 4D: Main Belt and Tro-
cle hundreds of degrees between day and night jan, as Jupiter sits between it’s Trojan clusters,
cycles. Determined mining on the Moon can and the outermost moons are really loosely cap-
reach depths of 50 km before rock pressure tured asteroids. Progressing inwards you en-
makes it unreasonably difficult. This makes 2 counter lethal radiation levels, requiring lots of
billion cubic km of resources potentially avail- shielding for people and equipment. The large
able. Because of its formation and history, moons represent many billions of cubic km of
the Moon is depleted in Volatiles compared to resources, which would eventually make them
even nearby asteroids. These are elements and attractive sources. However solar flux of 3.3-
compounds with relatively low boiling points, 4.1% of Earth’s presents a challenge for enough
such as water and atmosphere found on Earth. power to make use of those resources.
168 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Phase 5E: Outer Gas Giant Locations - This a region of gravitational dominance, which is
includes Saturn, Uranus, and Neptune, and the 100,000 AU times the square root of the sys-
surrounding regions within 20, 12, and 12 mil- tem mass in units of our Sun’s mass. There are
lion km respectively. It includes 103 known about two dozen known planets in this region,
moons, a number of which are large, and three and two have circumstellar disks, but this data
ring systems, one of which is famously promi- is likely incomplete. Design for these locations
nent. Solar flux is 1%, 1/4%, and 1/9% of must await better information, and a lot of new
Earth’s, respectively, making alternate energy technology development.
sources very attractive. These locations follow
Phase 4E: Outer Interplanetary, as you must
travel through that region to reach the three 5.2.3 Program Structure
planets and their surroundings. This phase is
far enough in distance and needed technology To make a large and complex program comprehensible,
that a significant amount of research and de- we divide it into multiple levels of detail, with each level
velopment is needed before it can be used. dividing a given part into a reasonable number of smaller
parts. At the most detailed level, individual program el-
ements are small and simple enough to be designed and
Phase 6: Interstellar Locations implemented, without further division. For this example
of a complex program we will not carry it to that level of
The last major phase includes locations with are not tied detail. It would first require completing conceptual and
to the Sun’s gravity well. This includes open interstellar preliminary design, which is not completed yet, and there
space not tied to a specific star, and the regions around isn't space in this wikibook to include all the information.
other stellar systems. There is no reason to stop the ex- Instead, we will present the top several levels and the pro-
pansion of civilization at the borders of our Solar Sys- cess of defining their elements, with the understanding
tem, assuming we have the necessary technology and it is that future work can use similar processes at the lower
economically reasonable. However, at present we don't levels.
have full knowledge of planetary systems around even the
nearest stars, and transport technology to reach them is
currently speculative. We include this phase as a long- Level 1: Program
range program goal, but with the understanding that most
of the details will have to wait. We divide this phase into The top level is the Upgrade Program as a whole. It is
two sub-phases, to account for the different environments here that overall goals and objectives are defined, and
and activities in interstellar space vs exostellar systems. top level interactions occur with the rest of civilization.
Decisions are also made to implement parts or all of the
program vs. staying with what is already in progress and
Phase 6A: Interstellar Space Locations -
planned. Change takes more work than staying with what
This region begins 100,000 AU from the Sun,
already exists. So a new program has to be sufficiently
where it’s gravity is no longer dominant, and
better to motivate people to change. That requires de-
extends to the edge of the Universe. For de-
veloping the needed ideas and technology, and educating
sign purposes we will set an arbitrary limit of
people about why they are better. We refer to what al-
20 light-years from the Sun. If we can reach
ready exists as the Existing Baseline. For example, in
that distance, and restock/rebuild our equip-
space launch that would include rockets that are already
ment, then later projects can travel further in
in operation, and new ones already funded and in develop-
increments of 20 light years, but not require
ment. For factory production it would include currently
new designs. Contents of this region are poorly
operating factories, and the state of the art for building
known, but very low in density. The stars in
new ones. Our program is not yet developed enough to
this region make up the next sub-phase, but
recommend implementing it, but we think it has great po-
they are all in motion relative to the Sun, and
tential. This is what has motivated our work to date, and
thus their membership in the Solar Neighbor-
we hope to continue it until we can make a recommenda-
hood changes about once per 1400 years on av-
tion for or against it.
erage.

Phase 6B: Exostellar Locations - There are Level 2: Phases

currently a little over 100 known Stellar and
Brown Dwarf systems within 20 light years. Level 2 of the program includes the major program
We know much more about the stars than what phases and sub-phases described above and in later sec-
orbits them or the spaces in between, because tions of Part 4. These phases inherit parts of the top level
stars are bright and we can more easily col- goals and objectives, such that all the phases together
lect information from their light. Each star has meet all of them. Earlier phases may only meet some
5.3. ADVANCED MANUFACTURING 169

of them, or at lower levels of performance, with the in-

tent to upgrade and expand the coverage later on. Part of
the early design process is to specify what goals and per-
formance levels each phase will accomplish. Since the
phases are covered in more detail elsewhere, we won't
duplicate that information here.

Level 3: Projects

Each phase can include one or more projects. These Figure 4.1-3 - Example functional diagram.
projects are intended to accomplish a task or meet a goal.
For example, the Seed Factory Project, which has al-
ready been started, is intended to design and test proto- Eating utensils and CNC laser cutters, for example, can
type systems to prove out ideas for self-expanding pro- be more or less the same from one location to the next.
duction. This project is part of Phase 0: R&D. A hypo- So we don't have to design unique elements for every lo-
thetical “Floating Cities Project” would be part of Phase cation. Instead, we can identify the functions that need to
3, due to the difficult or extreme environment of the be performed there, then copy existing designs or modify
oceans. them as needed to satisfy the needs. Each function, like
cutting parts from stock material, has inputs like electric-
ity, and instructions on what part to make, and outputs
Level 4: Locations like the finished parts and unused scrap. These inputs and
outputs connect functions to each other, and to outside the
The program is aimed at upgrading and expanding civi- program boundary. The functions and their connections
lization for the reasons listed in the Program Goals sec- can be displayed in various ways, such as the example di-
tion above. The general approach is to build up a given agram in Figure 4.1-3.
location, adding functional elements that allow internal
growth and useful outputs. Each location helps support it- Because civilization shares common functions and these
self, interacts with previous locations, and with the rest of functions share common connections between them, we
civilization. When enough capacity has built up at current can define a Reference Architecture that applies to ex-
locations, they deliver starter elements to a new location. isting elements of current civilization, and new elements
The starter set then repeats the cycle of growth and use- of our program. This makes it easier to compare what
ful outputs. Parts and materials are delivered from pre- already exists to what is new, and saves us from having
vious locations for whatever cannot yet be made locally. to define the organization of new locations each time. A
Locations are defined by ease of internal local transport given location may only include a subset of the common
for people and bulk goods. An example is a metropoli- functions, especially when it is new. The location can
tan area around a city, where travel times are a couple of evolve over time, adding new functions. The program as
hours or less between the various parts, and a good road a whole also evolves over time, adding new functions and
network exists. Transport between locations will happen, upgrading existing ones.
but the increased time and cost will make it less frequent,
and preferably for high value, low mass items. So you
Further Levels
may ship a computer to a distant location, but you prefer
not to send a truckload of gravel that far.
Functions can be divided into more detailed ones, and
Locations in undeveloped, difficult, or extreme environ- then systems designed to perform those functions. This
ments, and in space, may be much smaller than a city at falls into the various fields of engineering. For space sys-
first. This is due to lack of transport systems like roads tems, some of the methods and concepts are described in
and airports, and large distances to the next location. In earlier parts of this book. For other systems, we refer the
developed regions, the overall size of the location is typi- reader to the enormous range of literature on the subject.
cally that of a metropolitan area, but does not include all
the contents of that area. It involves specific people and
equipment located there, who interact with the surround- 5.3 Advanced Manufacturing
ing region and other locations.

The main program goals are upgrading civilization on

Level 5: Functions Earth, and expanding it to more difficult environments,
including space. To accomplish these goals, some new
All of civilization shares common functions, such as pro- or improved technologies and methods will be needed.
tection from bad weather, supplying food and drink, fab- Once available, they can be incorporated into suitable de-
ricating parts, and assembling them into useful items. signs for their intended locations. For space locations in
170 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

particular, there has been a severe lack of production and

The R&D work can be divided into a general part that ap-
habitation capacity, with the main focus so far being on plies across the whole program, and sub-phases covering
transport and information services. This imbalance existswork for specific environments and locations. The sub-
to a lesser degree in difficult Earth environments. For ex-phases and tasks are detailed further below. Limits on our
ample, many ships (transport) cross the oceans, but rela-current knowledge, and on available project resources,
tively few things are produced there, and few people in- mean we cannot do all the R&D work in advance or all at
habit the seas. Phase 0 is therefore included in the pro-once. In some cases, a given area of R&D must be com-
gram to supply what’s needed for the later phases, and topleted successfully before following work can be done.
consciously correct unbalanced development where pos- Since we do not know in advance if we will succeed, we
sible. The major goals of Phase 0 can then be stated as: expect that R&D plans will often need to change, or fol-
low multiple paths. Therefore this phase is expected to
• (1) Identify systems and elements covering the full continue in parallel with later phases for as long as the
range of production, habitation, transport, and ser- program continues.
vices functions. Early phase production products and services can be used
to support later R&D. For example, we may demonstrate
• (2) Supply needed new technologies and methods, self-expansion of an industrial building as an R&D task,
in the form of tested and ready to use elements. and later use that building for further R&D, or as a pro-
duction area whose products finance further work. Field
• (3) Supply detailed designs for equipment and loca- experience from earlier phases can be fed back to the
tions, incorporating both existing and new elements. R&D phase to improve later designs. We also expect
progress across civilization in other technologies, so a
given design may no longer be optimal and require up-
5.3.1 R&D Planning grades.

The program as a whole is complex. The Systems En-

gineering methods (see Section 1.5) were developed to 5.3.2 R&D Process
handle such complexity, so we intend to use them to de-
velop this and later phases. Other engineering methods The research and development process is similar across
can also be used where appropriate, but the systems ap- all phases and subphases. We give the component tasks a
proach is especially useful across whole programs. This consistent numbering system so they may be coordinated
includes their interactions with the world outside the pro- and compared across the program:
gram, and the constituent parts of a program with each
other. Part of the systems process is to break down com- Task 0: Coordinate R&D - This activity includes co-
plexity into smaller parts, which are then more compre- ordinating external resource flows, tasking, planning and
hensible and easier to design. We have already started scheduling, and analyses.
this in Section 4.1 by identifying a sequence of phases
and sub-phases according to scale, type of environment, 0.1: Coordinate R&D External Flows - This
and distance. A given set of locations within a sub-phase task includes arranging and managing resource
can then share similar designs, and to a lesser degree with flows into and out of the phase, from outside
those of the major phase and program as a whole. the program and to other phases.
Civilization as a whole has common elements across all 0.2: Coordinate R&D Tasking - This activ-
of it. For example, people need protection from the envi- ity includes arranging which resources will be
ronment and food to eat no matter where they are. Heat- applied to the phase tasks.
treating alloy steel can use the same process anywhere 0.3: Coordinate R&D Planning & Schedul-
you need to do it. We can therefore define a reference ing - This activity includes developing future
architecture for these common elements of civilization, plans and schedules for the phase.
and apply it to organize the tasks of upgrade and expan-
sion. Many existing parts of civilization and component 0.4: Coordinate R&D Analyses - This ac-
technologies are good enough as-is for the upgrades and tivity examines past performance and studies
expansions we would like to do. In those cases, we don't ways to improve R&D processes.
have to change them, just use them. Other items are
deficient or undeveloped. As we identify them, we can Task 1: Conceptual Design - This includes exploring
rank them by parameters like benefit ratio, cost, diffi- new concepts and developing a reference architecture.
culty, probability of success, and time to complete. Then This is followed by a systems engineering process to reach
we can add them to our R&D plans in the best sequence. a concept level design. This includes defining the main
New technology we develop is used internally in the pro- functions and elements of the program, and how they
gram, and also supplied as a benefit to civilization at large will be operated and maintained across the stages of their
to use elsewhere. life cycle. This model is itself part of the conceptual
5.3. ADVANCED MANUFACTURING 171

design. Based on prior experience, systems engineering 1.8: Synthesize & Document Design - The
effort optimizes the program cost and schedule at ~10- outputs from this task are articles, reports, and
20% of total effort, with the systems tasks weighted to- books documenting the concepts.
wards the early part of the program. The systems en-
gineering process flow is used iteratively in later design
Task 2: Preliminary Design - Assuming the conceptual
stages. The subtasks here are a template for those flows,
design produces a sufficiently promising concept, the next
but to avoid repetitiveness they are not broken out sepa-
stage is to define the elements of the program in more de-
rately each time below.
tail. This is done in parallel with component technology
and prototype systems because otherwise size and perfor-
1.1: Explore New Concepts - This step cov- mance would be too uncertain. Multiple design alterna-
ers taking concepts, such as self-expansion and tives may exist in this stage, until competing technologies
automation, and applying them to create new and testing is far enough along to permit selection. This
products and projects. Some concepts only ap- follows same steps as conceptual design, but at greater
ply to particular phases or elements, so an ap- level of detail.
plication matrix is an output of this task.
Task 3: Build R&D Locations - This activity includes
1.2: Develop Reference Architecture - The building or acquiring use of offices, research workshops,
reference architecture is a high level design conventional production shops, and prototype test areas
used to identify technology risks and readiness needed for R&D work. The design requirements for the
level (TRL), and make early estimates of cost R&D locations comes from the previous design work and
and schedule. It is a starting point for the con- needs for technology development, and building and test-
ceptual level design. It includes program goals, ing prototypes. Consideration is also made to adapt the
an architecture description, high level inter- facilities to later phase locations.
faces, element requirements, and element de-
scriptions. Supporting data for the reference Task 4: Develop New Technology - This includes iden-
architecture includes data sources, analyses to tifying the performance needed based on the conceptual
support concept selection, and tracking from and early preliminary design, surveying the status of cur-
goals to lower elements. rent technology, ranking areas for improvement in terms
of impact, then applying effort in the most promising ar-
1.3: Identify Requirements & Measures - eas to improve performance or lower uncertainty. Some
These establish measurable features the pro- technologies are already under heavy development out-
posed design must meet, and criteria for select- side the program. So rather than duplicate that effort, we
ing among design alternatives. See Section 1.5 select areas where a limited budget can have the most im-
Requirements Analysis for details. pact, or encourage others to invest in those areas which
1.4: Perform Functional Analyses - This need the most work. Technology level work is aimed at
breaks down what the design does in terms of single processes or components.
functions it performs or a sequence of opera- Task 5: Build Prototype Elements - At some point
tions. See Section 1.5 Functional Analysis it becomes necessary validate integrated elements and
for details. demonstrate performance levels by building prototype
1.5: Allocate Requirements - This assigns the hardware. This can be simplified versions of what will
requirements from task 1.3 to functions from become final designs, scaled versions for what will be
task 1.4 to ensure they are all met somewhere larger designs, or versions that demonstrate the function-
in the design. ality, but do not use the final materials and components
because they have not yet been built. Prototype elements
1.6: Model Alternatives & Systems - There
may carry over to later phases if they work well enough,
are many possible ways to meet a given set of
or can be upgraded to final versions in some cases.
requirements. Modeling the options provides
measurable details for each. The modeling pro- Task 6: Test Prototypes - This task reduces technical
cess includes (1) Collect External Technical In- risk by demonstrating the actual performance of proto-
formation: This includes data needed for mod- type system elements. Initial testing would use the local
eling and later design, such as existing product R&D environment conditions, but later testing uses the
data, industry specialist contacts, and current full range of operating environments, either in test cham-
state of the art such as books and articles. (2) bers or by taking the equipment to suitable locations. De-
Develop Alternative Options (3) Build System ficiencies found during testing are fed back to developing
Models. new technology. Early prototypes of a given element may
1.7: Optimize & Trade-Off Alternatives - have lower performance goals, which are later increased
This includes varying parameters of a design as improved designs are developed.
option, and comparing different options, to find Task 7: Design Location Details - This activity covers
the ones that best meet the selection criteria. detailed design of specific locations, including the loca-
172 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

tions where R&D is done, and locations for later phases. it is nearly as old as toolmaking itself. What is new is op-
It uses existing technology, plus new technology devel- timizing a small starter set to bootstrap this process, ap-
oped, prototyped, and tested within the program. Be- plying modern computer software, automation, robotics,
cause of improved technology over time, goals for further and AI to the task, and combining several growth paths
expansion and upgrade of existing locations, and develop- to increase factory output:
ment of new locations, this task is expected to continue
through the program. It includes detailed design of indi- • Making identical copies of the starter equipment,
vidual elements that can be used in multiple locations or
sold as separate products. • Making larger versions of, or extensions for, the
starter elements, and

5.3.3 R&D Sub-Phases • Making new tools and machines which can do dif-
ferent tasks, and expand the range of possible out-
Sub-Phase Identification - We use the plain “Phase 0” puts.
label, with no additional letters, to identify general R&D
work which applies across the whole program. When the This technology should be worth developing on Earth
work is specific to a single phase, a letter is added, such for it’s own sake. It should make setting up new facto-
as 0A or 0B. When it applies to two or more phases the ries cheaper, especially in remote or difficult locations.
R&D work is identified with all applicable sub-phase let- With an emphasis on self-growth, they may also achieve
ters, thus 0CD or 0G-L. Since some R&D locations may high economic rates of return. Once developed on Earth,
themselves need new technology and design, the first sub- industrial-scale mature factories can build items needed
phase, 0A, applies to Phase 0 itself. The sub-phases and to reach space, such as launch sites and rocket factories.
some of the identified R&D topics are listed below. This New starter sets are then delivered to space locations, and
list is currently preliminary. The topics are listed in the the expansion process continued. The experience gained
order we identified them, rather than time order, since on Earth, and the leverage from a series of self-expanding
R&D sequence and schedule planning is a later step. factories, multiplies the savings on future space projects,
making them affordable.

Program-Wide R&D
Phase 0A - R&D for Phase 0 R&D Locations
We have identified self-expanding production, using
Seed Factories as starting points, as a general new tech- Even when no new equipment or technology is needed
nology across the program. They can be used in exist- for it, research and development needs locations for of-
ing locations across current civilization, and new ones fices, laboratories, prototype fabrication, and testing of
on Earth and in space. The general technology in- equipment for later phases. We therefore apply the R&D
cludes more specific subjects like distributed production process above to design and build these locations. When
networks, remote-controlled operation, and smart tools new and unique items are needed, such as a special test
which can operate themselves. Manufacturing in general, chamber, they are developed and built the same way as
and automation in particular, already get a lot of engi- equipment for later phases.
neering effort, so we do not need to duplicate those ef-
forts. Our R&D work will focus on the unique aspects of
Phase 0B: R&D for Phase 1 Starter & Network Lo-
self-expanding production systems, and integrating other
cations
technologies into them. The program’s goal is to estab-
lish new locations for civilization, and upgrade existing Distributed Production Networks - Traditional facto-
ones. These program efforts fit within the definition of ries and large office buildings brought equipment and
Advanced Manufacturing. Enough work has gone into people together in one place because it was the only way
the seed factory idea to start a second wikibook on the to efficiently organize the work. Modern communica-
subject. We provide a short introduction here, and refer tions and transport networks relieve the need to be in
you to the other book for detailed information. one physical place, and allow coordination of distributed
Seed Factories Introduction - All factories produce work in many places. Some prominent examples are de-
products, and some factories produce the same kind of velopment of open-source software, and Wikipedia. In a
products they use themselves. For example, steel plants modern production system, the control of the machinery
typically use some steel in their own construction. Self- can be a mix of a on-site people, remote control by peo-
expanding factories are specifically designed to use their ple, and automatic control by computers and software.
own output to grow. A seed factory is an optimized Since all the people don't have to be nearby, you can
starter set of equipment, plus plans and instructions for a operate in undeveloped, hostile, or expensive locations
chain of expansions, to reach some desired mature state. more easily, and with less of an environmental footprint.
Using tools to make more tools is not a new idea. In fact Remote operators can efficiently split and re-assign their
5.3. ADVANCED MANUFACTURING 173

work as needed between locations. It is likely that some transport, and service industries. Equipment for all these
machines and workers will still be grouped together in industries already exists, and is widely produced. The
shared locations, for efficiency or other reasons. Modern R&D for this sub-phase includes modifying their design
technology merely removes the requirement that they all so they can be made by self-expanding and distributed
have to be om one place. systems. It also includes the growth paths and methods
Some of the needed technology for distributed production to reach industrial scale from the smaller scales in earlier
already exist. The R&D tasks for this sub-phase are then phases.
to improve or fill in the parts that do not, and combine More specific areas of R&D may be identified later for in-
them into flexible distributed networks. The flexibility dustry groups or individual industries. One that we know
is needed because the program intends to constantly add of at this point is industrial transport to Low Orbit, since
new locations, and seed factory technology will expand it will be needed for the later program phases in space:
existing ones. So we cannot operate on the basis of static
networks. They have to grow and adapt along with the 3. Industrial Transport
rest of the program.
Applications to Later Phases - In this phase, a goal Launch to Low Orbits - This is placed in Phase 2B
for distributed production technology is the capability because traditional rocket factories and launch sites are
to connect and operate hobby and home improvement industrial-scale facilities on Earth. Locations for Phases
level equipment in fairly close proximity, like a single 4-6 are in space, but still interact with civilization on
metropolitan area. Later phases would need upgrades Earth. So there will be a continuing need for transport
for long-distance remote operations, such as on the Moon from Earth to orbit, and back. Obviously space programs
from Earth, or Mars from Phobos. More R&D may be already exist, and many satellites are in orbit, but their
needed later on in this technical area. Space is a partic- cost is high. Partly this is due to the transport cost it-
ularly undeveloped, hostile, and expensive location. So self, and partly due to lack of production in space. This
when you optimize your operations you will want to min- forces all equipment and supplies to come from Earth. In-
imize the on-site humans, and maximize remote control, space production is addressed in the later phases, while
and smart tools which can operate themselves. So at first this topic covers transport needs.
there will be a strong incentive for the upgraded technol- In the earlier parts of phases 4-6, transport needs to orbit
ogy. As factories, habitats, and transportation systems are will be small, and therefore using existing launch systems,
built for the later phases, people can be supported more or ones currently in development, will be the less expen-
easily on-site. So the optimum balance of local people vs sive route. As program traffic increases, the advantage of
remote and smart tools will shift. Having gained experi- new and more efficient systems will grow. The R&D in
ence with the distributed approach on Earth, using it in this sub-phase will then cover such new systems, beyond
space will not be something entirely new, but rather an those already in development elsewhere. Sections 4.4,
extension of what was learned in earlier phases. 4.6, and 4.7 present some early concepts for this R&D
work. In section 4.4 - Phase 2B Industrial Locations
we consider a small, 3 stage, fully re-used conventional
Phase 0C - R&D for Phase 2A Distributed Locations
rocket and some other alternatives for the “build our own”
option. The design is not complete enough to decide be-
R&D for this sub-phase involves design of more special-
tween make or buy yet. The intent is when traffic is suffi-
ized and larger machines than for Phase 1. These are used
cient, the start-up transportation will be augmented or re-
for small business and commercial activities, therefore
placed with larger, more efficient, and specialized launch-
would have higher duty cycles and longer operating lives.
ers. The initial cargo may consist of assembly robots and
Besides design for these conditions, another R&D topic
parts for an initial orbital platform. If we are building our
is the best growth paths from the previous phase, and ex-
own launcher we want to make it as small as practical to
pansion across a wider range of industry categories. A
keep the design and construction cost low.
third R&D area is the grouping of varied size equipment
in terms of more specialized and distributed sites across Upgraded Transport to Low Orbit - The program will
a location, and linkages between locations. All of these add upgraded transport when there is sufficient traffic
R&D areas continue in the next sub-phase to the indus- to justify the capital cost. Again, there is always the
trial scale, which uses the largest size equipment. option to use transport from outside the program, but
we consider various internal alternatives using our self-
production capacity. On Earth we use different trans-
Phase 0D - R&D for Phase 2B Industrial Locations port systems for bulk cargo than for passengers for cost
and safety reasons. One alternative is to specialize our
This sub-phase completes the sequence of growth to space transport elements for the same reasons. Section
larger and more specialized equipment, for developed lo- 4.6 - Hypervelocity Launcher presents a high accelera-
cations in moderate environments on Earth. It includes tion gun for launching bulk cargo such as propellants or
equipment for the full range of production, habitation, structural parts. Delicate cargo and humans would travel
174 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

by other methods. The launcher gives the cargo a large Phase 0G - R&D for Phase 4A Low Orbit Locations
starting velocity, so it substitutes for part of the rocket
stages. In theory it should lower cost because a fixed gun Low Earth Orbit already hosts many satellites, and as of
can be designed to fire many times, and is made from in- the start of 2017, two space stations with a total crew of
dustrial pipeline quality parts, which are much cheaper eight. However it lacks significant production capacity,
than aerospace grade parts. aside from assembly of pre-made elements at the stations.
Eight people is only one billionth of the Earth’s popu-
Section 4.7 - Low G Transport looks at methods for
lation, and no transport systems are based in low orbit.
transporting humans and cargo which cannot withstand
What transport exists is all based on Earth. So while we
the high acceleration of the hypervelocity launcher. The
have a foothold in low orbit, civilization can't be said to
choice of which to use depends on results of more detailed
have fully expanded to this region. The R&D for this
design and what other launchers area available outside the
sub-phase is then aimed at full use of low orbits, beyond
program. Some candidates to build our own systems are
current programs and activities. So far we have identified
a combined air-breathing/rocket system, or a gas accel-
the following:
erator similar to the hypervelocity launcher, but lower g
level, followed by air breathing or rocket stages. Sepa-
rate stages will be easier to develop, modify and upgrade 1. Low Orbit Production
than a single integrated vehicle, although there will be
a penalty in operations cost. A single integrated vehicle 1.2. Supply Power - Electrical power using solar panels
can be developed later once traffic will support the more and batteries is fairly well developed for low orbit. Sun-
complex design. light is available at least 60% of the time, but only special
low orbits have it all the time. So energy storage is needed
to bridge the time in the Earth’s shadow. Thermal power
using solar concentrators is an area for R&D.
Phase 0E - R&D for Phase 3A Difficult Earth Loca-
1.3. Extract Materials - Low orbit has two sources of
tions
materials besides those brought from Earth or more dis-
tant locations. The first is the upper fringes of the atmo-
Difficult and Extreme locations involve all the sizes from sphere, which can be collected by a compression scoop.
small to large that were developed for Phases 1 and 2, but The second is dead space hardware and artificial debris.
in a different environment. Therefore the existing designs Some of the gathered gases can be used as propellant for
will sometimes need modifications, and in other cases collecting the space debris, since they are in widely scat-
unique designs will be needed. The effort to set up in re- tered orbits. The space debris at the least is a hazard,
mote or hostile conditions will tend to make small scale and removal is a benefit to other space activities. But
equipment less likely, and the emphasis shift to larger it consists of aerospace-grade parts and materials, some
sizes. Example difficult environments include very cold of which may still be functional. Salvage and recycling
and hot regions, deserts and rain forests, altitudes above of these items would save having to launch comparable
2750 m, weak soils, water and ground depths of 250 and items from Earth. R&D is required to prove the gas min-
100 meters respectively, areas of low energy resource ing, collection, and reuse of old hardware is practical. It
or high natural radiation, high communication and travel would also provide some experience for later mining and
time, low stay times, and high transport energy, or com- production beyond low orbit.
binations of these conditions. Each may require R&D to
accommodate the particular circumstances. 1.4 Materials Processing & 1.5 Parts Fabrication -
Very little of either of these has been done in orbit and
in zero gravity. Extensive experiments and prototyping
are needed to find out which terrestrial methods can be
Phase 0F - R&D for Phase 3B Extreme Earth Loca- used, how they may need to be modified, and what new
tions methods can be used in the unique orbital conditions.
1.7 Low Orbit Assembly - The design of transport sys-
Extreme locations are an extension of difficult ones, but tems typically is much more expensive than a single use
farther from moderate conditions up to the limits of tech- of them. Therefore a number of deliveries on a smaller
nology. R&D would be needed to push technical lim- launch system is preferred on cost to a single delivery on
its beyond the state of the art. An example would be a very large one. This in turn drives a need for assembly
hard rock mining more than 5 km below the surface, of larger elements in orbit. Section 4.5 - Orbital As-
well below the deepest current mines. Some example sembly gives one approach, using an assembly platform
extreme environments include very cold conditions in in low orbit. At first, the platform assembles pre-made
parts of Antarctica, The open ocean surface, which has components launched from Earth. As other production
zero ground strength, great depths underwater or under- elements get added, it later shifts to assembling a mix
ground, and the most remote and inaccessible surface lo- of Earth and locally made items. The first task of the
cations. assembly platform would be to bootstrap it’s own con-
5.3. ADVANCED MANUFACTURING 175

struction. The platform is then used to assemble larger reach more distant destinations. Power, thrust, operat-
payloads, and then then later build seed elements and ve- ing life, and radiation resistance will all have to improve
hicles for new locations. Humans are kept to a minimum for these later uses, so propulsion R&D will be ongo-
in the early stages because of cost. The assembly robots ing. An early use for such engines is mining the upper
start out mostly controlled from the ground. Some expe- atmosphere for fuel, which makes the propulsion self-
rience already exists with orbital assembly of space sta- sustaining. With adequate fuel, mining the Earth’s debris
tions, and similar maintenance and repair tasks for the belt both cleans up the debris, and serves as a source of
Hubble Space Telescope. The R&D tasks here are what- raw materials, spare parts, and work repairing and refu-
ever improvements are needed beyond these levels. eling satellites that need it. New satellites delivered and
assembled at an assembly platform can also be delivered
2. Low Orbit Habitation to their destinations at low cost. One category of satel-
lites to deliver are prospector satellites to observe and re-
turn samples from candidate asteroids, to prepare for later
Variable Gravity Research Facility (VGRF) - The near mining.
Earth environment, like most of the Universe, is hostile
to human life. Therefore habitats have to be designed Chemical Propulsion - High thrust engines, such as con-
specifically to create the proper conditions. One major ventional chemical rockets, are still an attractive option
alternative for orbital habitats is gravity levels. Humans for some purposes, despite lower efficiency. These in-
need some level of gravity for long term health, but ex- clude landing on bodies with significant gravity wells, and
actly how much and for how long is not known yet. Some when velocity change or transit needs to be done quickly,
production methods work better with acceleration, and such as passing through the Earth’s radiation belts. Which
some agriculture may also turn out to work best with grav- propulsion type to use for what part of a trip will need to
ity. A VGRF type facility is needed to assess what gravity consider multiple factors including the ability to extract
levels are needed for what purpose, so that later designs fuel locally. R&D for chemical propulsion will include
can use the proper levels. adapting it to use and store propellants made in orbit, and
improving engine operating life.
Habitat Growth and Upgrade - Habitats will generally
start small and grow over time. So another research area Spaceport Network - In the long run, large numbers of
is the best growth paths for the them: in physical size, vehicles changing their orbits by consuming propellant is
from possibly zero gee to some gee level, from open food inefficient. Large scale infrastructure which reduces pro-
and air cycles to closed life support, and from hardware pellant needs would be desirable. We will refer to them
supplied from Earth to local production. The design of as Spaceports by analogy to airports. Their main func-
the habitats is likely to be complex, and we can only lay tion is transport of payloads by potential and kinetic en-
out these open questions as a starting point for further ergy change. However they will also serve as transporta-
R&D work. tion depots, with docking for multiple vehicles, habita-
tion, warehouseing, maintenance, fueling, etc. The first
concept for such infrastructure was the Space Elevator,
3. Low Orbit Transport
dating back to 1895. Unfortunately the Earth’s gravity
well is too deep for the original idea of a one-piece sta-
This category covers transport that operates within low tionary elevator to span with any known materials. This
orbits, and reaches farther destinations. Transport to low idea can work for smaller bodies, and later concepts have
orbit is covered under Industrial or Difficult Earth Loca- improved on it and are preferred for multiple reasons. A
tions, because that is where they are built and start from. network of spaceports can eventually replace much of the
Electric Propulsion - Ion and plasma engines have about propellant used in space, and increase the percentage of
5-10 times the fuel efficiency of conventional rockets, payload transported. The R&D work for such a network
and have already seen some operational use. Section 4.8 - is placed here because the first spaceport would be lo-
Electric Propulsion looks at options for propulsion mod- cated in low orbit. Section 4.11 - Space Elevator looks
ules, which can be used singly for smaller missions and at some alternative concepts for such a network.
in multiple copies for larger missions. There are several The basic transport function is accomplished by
types of such thrusters available, but they will be needed Momentum exchange between a payload and the
in some form if missions beyond Earth orbit are to be spaceport structure. Depending on direction, the payload
done economically. The higher efficiency allows bring- gains or loses energy, and the opposite happens to
ing the vehicle back and using it multiple times, a key cost
the spaceport. If traffic is balanced, or the spaceport
savings. R&D for this phase would be aimed at upgrading is anchored to a more massive body, it’s orbit is not
the propulsion to higher power levels, and enabling use of affected. Unbalanced orbit changes are corrected by an
mined propellants rather than the scarce Xenon used to- efficient propulsion method on the spaceport. To the
day. extent this replaces lower efficiency vehicle propulsion,
Electric propulsion can be used within low orbits for drag there is a net savings. Various experiments have been
makeup, and for orbit change within low orbits and to done in orbit related to this technology, but much more
176 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

work is needed, along with other technologies for a like the Moon or Mars. This view is incorrect and obso-
spaceport network and associated vehicles. lete. Open space has always been known to have a large
and constant flux of solar energy. At the Earth’s distance
from the Sun this is 1360 MW, on the order of one nu-
Phase 0H: R&D for Phase 4B High Orbit Locations clear plant, per square km, and there are 281 quadrillion
square km around the Sun. From 1990 to early 2017,
High Earth Orbits are currently used by a number of the number of known asteroids in this region has grown
remote-controlled satellite types, including communica- from 180 to 16,000, and we continue to rapidly find
tions, scientific, and navigation. They are all delivered more. There are also new ideas on how to efficiently ex-
from Earth, and local production and habitation don't yet tract large quantities of materials from the larger bodies.
exist. Transport is only that built-in to the satellites when With sources of energy and raw materials already in place
delivered. High orbit is fairly devoid of native materi- or which can be imported, widespread use of this region
als, but has a high level of solar energy, and is accessible should be possible.
from Earth, the Moon, and Near Earth Asteroids. Be-
tween current civilization on Earth, and future locations Section 4.13 - Interplanetary Development looks at
beyond it, it can serve as a useful production and transport concepts to use this space. General R&D is needed on
nexus, and later for large-scale habitation. Fully develop- Orbital Bootstrapping. This is how best to bootstrap
ing this region will require extensive R&D work. Some from early energy and materials extraction to large-scale
identified tasks include: finished locations, with a range of production, habitation,
transport, and services capacity. A future “space city” is
1. High Orbit Production not likely to be built all at once in final form, any more
than cities on Earth are. The question is then how to build
them in smaller increments. More specific R&D tasks in-
1.4 Materials Processing - This is the conversion of raw
clude:
materials to finished supplies or stock materials. In the
early stages this can be asteroid materials brought back
from Inner Interplanetary orbits to a location near the 1. Inner Interplanetary Production
Moon, such as Earth-Moon Lagrange Point 2 (EML2).
EML2 is a Lunar-synchronous location 64,500 km be- 1.3 Orbital Mining - The general rationale for mining
hind the Moon’s center. It is low energy to reach from in space, rather than bringing everything from Earth, is
interplanetary orbits, and provides full time sunlight for the Earth’s gravity well has a fixed energy cost of 31-62
power. Early products include shielding, propellants, and MJ/kg to climb, depending on orbit. Farther destinations
water. Extensive R&D is needed to identify the best lo- require even more energy to reach. The production en-
cations and processes. Alternative locations include: ergy from raw materials to finished products is less than
this, typically 10-20 MJ/kg, and production equipment
typically can process many times it’s own mass and pro-
• Air processing: aboard a low altitude scoop ship, or
duction energy during it’s operating life. It therefore takes
low orbit assembly platform.
far less energy to deliver production equipment and make
• Asteroid processing: in place at the asteroid, a high the products from local materials and energy than to de-
orbit nearer Earth, or lower orbits. liver the products from Earth. This includes the propel-
lants for space transport, which makes the delivery of the
Given the program’s prior development of self-expanding equipment easier.
production and remote operations, we assume processing Section 4.9 - Orbital Mining looks at alternatives for
is bootstrapped from a set of starter elements, delivered supplying the raw materials locally. For Inner Interplan-
by electric tug to the desired orbit. Until human habi- etary locations this involves extracting and transporting
tation can be supported locally, it would rely mostly on raw materials from Near Earth Asteroids (NEA) to a pro-
remote control and automation. Some processing opera- cessing location. NEA’s are the next easiest to reach re-
tions may not function well, or at all, in zero gravity, and sources after the atmosphere and orbital debris in low or-
others will benefit from or work uniquely in the zero grav- bits. The mass returned by a mining system and tug is
ity and vacuum conditions. So a major area for research on the order of 100 times the equipment mass per trip,
will be for which specific processing flows are to be used and the equipment life is on the order of 6 trips taking
under what conditions. 2.5 years each. A typical trip consumes 2.6% of returned
mass in propellant, but certain asteroid types contain up
to 20% easily extracted propellant. So the mining opera-
Phase 0I: R&D for Phase 4C Inner Interplanetary
tion can be self-fueling after the first trip. Mining should
Locations
drastically reduce space operation costs if that mass can
The space between the major planets and moons has tra- be put to use.
ditionally been viewed as devoid of use, to be gotten Mass return ratios of 50 or more are desirable for space
through as quickly as possible to reach the “real” locations mining in general, since these are raw materials and fur-
5.3. ADVANCED MANUFACTURING 177

ther processing is requires. A lower ratio may be accept- and repair, serve as science platforms, and serve as the
able for salvage of equipment and debris in Earth orbit, nucleus for later permanent colonies. Extensive R&D is
since they are already manufactured and even operable needed on how to build and expand such habitats, and the
equipment. These ratios look feasible from the analysis various systems they need, such as food production and
done so far. By using the Moon for gravity assist in both environmental recycling.
directions, traveling to and from Inner Interplanetary or-
bits can be done with less than Earth escape velocity.
Phases 0J to 0L: [RESERVED]
NEA orbits and composition are randomly distributed,
and we prefer to mine the easiest to reach ones at optimal The following three sub-phases do not yet have identified
positions and timing. However 82% of discovered NEAs R&D tasks. Their sub-phase headings are reserved for
are larger than 30m, and thus a minimum of 18,000 tons later use.
and usually much higher. This is too much to move as a
unit for early tugs, so R&D is needed on the best ways to • Phase 0J - R&D for Phase 4D Main Belt & Trojan
collect smaller loads of material from them. Along with Locations
the space tugs noted below, the mined material is deliv-
ered at first to High Orbit for processing. • Phase 0K - R&D for Phase 4E Outer Interplanetary
Locations
1.4 Orbital Processing - Aside from bulk radiation
shielding, some processing is usually needed to turn raw • Phase 0L - R&D for Phase 4F Scattered, Hills, &
materials into ready inventory. For items like propellants, Oort Locations
water, and oxygen they can be used at that point. Other
items need further production steps to turn them into fin- Phase 0M: R&D for Phase 5A Lunar Locations
ished products. Section 4.10 - Processing Factory looks
at the alternatives and options for this task. Materials pro- The Moon is physically near the Earth, and obvious to
cessing on Earth is a very large industry with a long indus- anyone who looks at the sky. Since 1958, over 100
try, but very little has been done in space yet. The R&D Missions have been attempted to flyby, impact, orbit,
needed for this program phase involves adapting and op- land, drive on, return samples from, or use the Moon for
timizing the processes for the unique conditions there. gravity assist. Many of these have succeeded, including
six landings with people. Despite all this activity, it is not
3. Inner Interplanetary Transport the first program phase we want to start in space. This is
due to the extra propellant needed to land on the Moon
3.1 Bulk Cargo Transport - “Space Tugs” are needed relative to orbital operations nearby. However, once pro-
to move raw materials from where they naturally occur to pellants are produced in quantity in nearby orbits, the dif-
where they can be processed, and move finished products ficulty decreases and lunar locations can be developed at
and other cargo from place to place. Tugs generally do relatively low cost.
not need human crews, and are slow but efficient. Electric
Section 4.12 - Lunar Development looks at some design
propulsion has already been developed at smaller scales,
options for the Lunar surface. With a spaceport network
but much larger units are needed for this task, and the
in place, including one in Lunar orbit, it would be possi-
tugs should be designed for refueling, so they can be used
ble to go to and from the Moon in a robust and low cost
many times.
fashion. However, this will not be available to start with.
3.3 Transport People - We want to eventually carry peo- Precursor missions can be delivered by electric tugs to
ple to open space locations and the major planets and Lunar orbit, and use conventional rockets to land. These
moons in the Inner Interplanetary region. However, ra- can explore and help select a Lunar surface locations. Re-
diation is present throughout the area from the Sun and mote controlled robots could do some preparation work,
cosmic sources. “Transfer Habitats” are a way to carry like paving landing pads by solar or microwave heating.
people safely and efficiently. They are placed in repeat- Expanding on the initial work, heavier mining and pro-
ing transfer orbits between bodies or locations, such as cessing equipment could be delivered and start to use the
between Earth and Mars. Since the habitats don't change relatively large mass and surface area available. Seed fac-
orbit once set up, they can have heavy shielding, green- tory parts and permanent habitats follow according to a
houses, and processing equipment. The raw materials logical progression. The primary question is when to start
come mainly from asteroids already in nearby orbits. Lo- using the Moon as a location, in the context of other lo-
cal production reduces payload needed from Earth, and cations and the level of technology. Specific R&D tasks
gives the crew and passengers something useful to do dur- include:
ing the trip. Small vehicles are used to get from the habi-
tats to planetary orbits at each end of the trip. Additional 1. Lunar Production
habitats can be set up on the Martian moons as way sta-
tions, and eventually other locations. All the locations The energy to reach Lunar orbit is 1.5 MJ/kg, while typ-
eventually can produce fuel, do spacecraft construction ical production energies, from raw materials to finished
178 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

products, are 10-20MJ/kg on Earth. Production energies Many of these technologies are conceptual at present, and
are likely to be similar in space. Gathering raw materials will need extensive R&D prior to use.
from the Lunar surface is fairly low energy, since repeated
impacts have pulverized the surface into a fine Regolith.
Since twice as much sunlight is available in high orbits Phases 0N to 0S: [RESERVED]
than the Lunar surface, the preference appears to be to
send many materials to orbit for further processing. Ad- Section 4.15 - Later Projects looks at some ideas for
ditionally, high orbit is a meeting point for materials from later phases. Since technology changes over time, it is
the Moon, asteroids, and Earth. Lunar surface materials not worthwhile to make too many detailed plans for far
are lower density minerals, well mixed from impacts, and into the future. Long range concepts can serve as a guide
low in volatile compounds. This is due to the Moon’s for future research, though. As the time frame gets closer,
high early temperatures and low escape velocity. Aster- ideas like these, or new ones developed in the future, can
oids usually did not get heated as much, and their denser be incorporated into updated program plans. The fol-
components are now exposed by collisions. The Moon’s lowing five sub-phases do not yet have specific identified
denser materials are trapped deep inside. So available R&D tasks. Their sub-phase headings are reserved for
materials from the Moon is different from the major as- later use.
teroid types. Some materials are rare or absent on the
Moon or asteroids, and are better brought from Earth. • Phase 0O - R&D for Phase 5C Venus & Mercury
Using all three sources allows a wider range of processes Locations
and products than from the Moon alone. More R&D is • Phase 0P - R&D for Phase 5D Jupiter System Lo-
needed to see if this is the best option, or if production cations
should be distributed in multiple locations. The best min-
ing and processing flows will also need extensive work. • Phase 0Q - R&D for Phase 5E Outer Gas Giant Lo-
cations
3. Lunar Transport • Phase 0R - R&D for Phase 6A Interstellar Space
Locations
3.1 Bulk Cargo Transport - If much of the process- • Phase 0S - R&D for Phase 6B Exostellar Locations
ing is to be done in high orbit, we will need an efficient
way to deliver bulk raw materials from the Moon. Can-
didates include centrifugal and electromagnetic catapults, 5.4 Startup Launcher
and space elevators, all of which require significant R&D.
According to the Satellite Industry Association, global
space industry as of 2015 was worth 335 billion
Phase 0N: R&D for Phase 5B Mars Locations
USD/year. Most of the activity in these existing programs
is carried out on Earth, with a relatively small amount of
Section 4.14 - Mars Development looks at concepts for equipment and people actually in space. These programs
developing the surface of Mars and it’s moons. One ap- were described earlier in Section 1.9. A portion of the
proach is to start with a habitat on Phobos. At first we Industrial Locations in Phase 2B of our program will be
use local materials from that Moon to support trips to used for upgrades or expansions to this existing space ac-
the surface. Since we don't yet know the composition of tivity. We have placed these locations in Phase 2B be-
Phobos, other materials may be needed from nearby as- cause previous space facilities have been industrial scale,
teroids. Since Mars skirts the inner edge of the Asteroid sometimes at the largest end of this scale. They have also
Belt, there are many candidates to choose from. At first been mostly located in moderate environments.
we produce propellants and crew supplies. Later we can
construct spaceport structures to exchange momentum
and reach the Martian surface more efficiently. Design Approach
We already have a number of satellites in orbit about
Mars, and landers and rovers exploring the surface. With The primary goal of our program includes expanding civ-
a propellant supply in orbit, we can start to land more ilization into space beyond our planet. The Earth is fi-
substantial equipment and build up larger facilities on the nite, and space has vast amounts of available physical
ground. These can be remote-controlled from orbit un- space, materials, and energy. They can be used to meet
til enough habitat capacity is available for full-time crew. a variety of the program objectives noted in Section 4.1.
Early missions can deliver seed factory components to Our design approach is then to identify what is needed in
start local production. With surface propellant produc- this phase, so that we can support the later Phases 4-6 in
tion, and later large ground accelerators coupled to orbital space, and then meet the main program goals and objec-
spaceports, access to Mars will be much easier in both di- tives.
rections, and large-scale development can proceed. Design issues for these locations include:
5.4. STARTUP LAUNCHER 179

• What parts of existing space activity should remain modern technology. That might need to be changed
as they are, in parallel with our new locations? with a better understanding of payload needs, but
we will use it as a starting point. Larger devices can
• Do some of the new locations belong in other be assembled from several items in orbit, but keep-
phases? ing the item size small lets you use a smaller launch
• What types of products and activities are needed in vehicle, and thus lower development cost to start.
this phase? There is also the possibility to use this as an “express
package delivery” service between larger launches
• How will our new locations interact with the rest of on other vehicles, and bring in some revenue.
our program, existing space programs, and the rest
of civilization? • Launch Rate - We assume an initial rate of about 1
launch per month, and continuing on a steady basis.
• In what sequence should the new locations and their
products be built? • Production - We assume one or more Advanced
Manufacturing type factories, as described on the
previous page, are used to build the launcher. This
5.4.1 Industrial Scale Production imposes production capabilities on the factory and
links the systems. Any materials or components that
5.4.2 Industrial Scale Habitation are not reasonable to make within the factory are
bought. The cost of the factory has to be included
This includes large scale construction which is intended when deciding which launcher to use.
to be occupied by people. Examples include large office
buildings, hotels, residential towers, retail complexes, and
entertainment venues. There are multiple possible ways to launch a small pay-
load to orbit. The conventional approach would be to de-
sign a small rocket with two or three stages. Any alterna-
5.4.3 Industrial Scale Transport tive ideas can be compared to that to see if it has a lower
expected development and operating cost.
This includes large scale transport of energy, discrete and
bulk cargo, fluids and gases, people, and data.
5.4.4 Existing Launchers
Transport to Low Orbit
5.4.5 Small Multistage Rocket
This type of transport is included here because produc-
We first consider a conventional rocket. The design pro-
tion, control, and operations for early low orbit transport
cess starts with some initial assumptions, from which we
mostly occurs in moderate environments on Earth. Some
can make an estimate of the vehicle size. We then pro-
of it may be from difficult locations on Earth, and in later
gressively add more detail and do more accurate esti-
phases involve substantial capacity in low orbit, or from
mates. This will replace our initial estimate with a series
farther destinations. They will be discussed in their re-
of better ones, and possibly force revising the assump-
spective sections later.
tions. A complete preliminary design considers all the
*****[Text still to be merged & updated]***** major components and is at the point where you would
As part of our overall space development project, we need start the detailed design and final drawings. We will not
a way to start putting payloads in orbit and doing activ- carry it that far, but intend to show enough of the process
ity in space. The choices are to use existing launchers to demonstrate how it would be done.
made by other people, build our own, or a mix of the
two. For the first choice you would look at the relevant
User’s Manuals for the existing launchers, and consult Design Assumptions
with their offices for the launchers that looks suitable. For
small hardware mass it is possible to travel as a secondary • Payload: 20 kg to 250 km orbit. We have to specify
payload along with a larger primary payload belonging to an orbit to calculate the mission velocities.
someone else. That can reduce the launch charges.
• G-Limit: 10 gravities or 100 m/s2 - This is to keep
For the self-build options we do preliminary designs, then accelerations and structural loads in the range of ex-
compare to the existing launcher choices. We need to isting space hardware designs, and so not force new
make some design assumptions to start with: designs and research for the payloads.

• Payload Mass - We will assume that 20 kg is suf- • Exhaust Velocity: 3300 m/s in vacuum - This is typ-
ficient mass for a functional hardware item using ical of a good Hydrocarbon/Oxygen propellant mix.
180 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

• All stages are re-used for cost reasons. Hardware Second Size Estimate To make a second estimate we
mass fractions are assumed to be 14, 15, and 18% need some details of the rocket thrust and drag, and there-
for the first to third stages. The upper stages would fore it’s size and shape. We assume Oxygen/Methane fuel
have higher fractions due to smaller size and increas-at 3.6:1 mixture ratio by mass. The chemistry of CH4
ing heat shielding. + 2O2 = CO2 + 2H2 O has a theoretical mass ratio of 4
Oxygen : 1 Methane. By using slightly less Oxygen some
• Launch Site: Equator at 4600 m altitude - This is of the Methane is left unburned, leaving CO or H2 in the
at Cayambe, Ecuador to take the most advantage of exhaust. This lowers the average molecular weight and
the Earth’s rotation and highest starting altitude to increases the exhaust velocity. It also ensures the com-
reduce drag and mission velocity. bustion is not Oxygen rich, which would tend to react with
surrounding materials.

Preliminary Estimate Conventional rockets are sized • Tank Sizing:

by the Rocket Equation, which determines propellant
mass ratios. A preliminary estimate of the velocity re- From our preliminary masses above, we can determine
quired can be made from experience. A second estimate tank sizes from the density of the respective fuels: Oxy-
will use a trajectory simulation that calculates fuel use, gen = 1140 kg/m3 and Methane = 423 kg/m3 :
thrust, drag, and acceleration in small time steps.
The ideal velocity to reach a 250 km orbit neglecting • From above, all stages have a final mass
losses is found from the total energy of that orbit, which of 42.4% of initial mass, therefore burn
is the sum of kinetic and potential energy. Velocity is 57.6% of initial mass of fuel. Therefore
7756 m/s, with an energy of 30.08 MJ/kg, and potential fuel masses are 406.7, 139.7, and 47.2
energy is 2.375 MJ/kg. The sum implies a velocity of kg.
8,056 m/s. The various real losses may be estimated at • With a mixture ratio of 3.6:1, the
900 m/s based on experience, giving a total ideal velocity Methane component is 1/4.6 = 21.74%
of 8956 m/s. Rotation of the Earth at the Equator is 465 by mass, and Oxygen is the remainder.
m/s, thus the rocket has to produce a net velocity of 8,491 Thus the Methane mass by stage is 88.4,
m/s. If we divide it equally into 3 stages, this gives 2830 30.4, and 10.25 kg, and the Oxygen mass
m/s per stage. Mass estimates are calculated from top to by stage is 318.3, 109.3, and 36.95 kg.
bottom as follows:
• From the densities we can calculate the
respective tank volumes. Allowing 3%
• Payload = 20 kg extra volume so that there is some pres-
surizing gas at the top of the tank and fuel
• Stage 3 final mass /initial mass = 42.4% - From margin, we obtain first stage tank vol-
rocket equation umes of 215 and 288 liters, second stage
of 74 and 98.75 liters, and third stage of
• Stage 3 hardware fraction = 18% - of entire stage 24.95 and 33.35 liters for Methane and
including payload Oxygen respectively.
• Rocket stage tanks can share a common
• Stage 3 initial mass = 20 kg / payload fraction = 20 wall between fuel and oxidizer if they are
kg / (final mass - hardware) = 81.9 kg fully sealed, and usually use an ellipsoidal
dome with a 70% height ratio to min-
• Stage 2 m(f)/m(i) = 42.4% imize structural mass. We assume the
payload has a density of 1 kg/liter, and
• Stage 2 hardware = 15% x (100-42.4%) = 8.64% - thus requires 20 liters volume. For aero-
of 2nd stage fuel only dynamic and structural reasons we want
to keep the total vehicle height at 10 times
• Stage 2 initial mass = 81.9 kg / (42.4% - 8.64%) = the base diameter or less. Each combined
242.5 kg stage tank can be modeled as two ellip-
soidal domes plus a cylinder. Applying
• Stage 1 m(f)/m(i) = 42.4% some geometry results in tank diameters
of 60, 42, and 30 cm.
• Stage 1 hardware = 14% x (100-42.4%) = 8.06% -
of 1st stage fuel only • Drag Coefficient:

• Stage 1 initial mass = 242.5 kg / (42.4% - 8.06%) = From the tank sizes we can do a preliminary layout of the
706 kg vehicle. We have to include a forward payload fairing and
5.5. ORBITAL ASSEMBLY 181

varied until the desired payload mass and orbit is reached.

5.4.6 Augmented Rocket

An alternative to a conventional rocket is to augment the

rocket stages in some way. The extra cost and complexity
of these methods has to be compared to any performance
and weight improvements they generate. These methods
aft engine sections for each stage to get the total height of include:
the vehicle. For this design we assume an aerospike type
engine with platelet injectors for each stage, which gives
a total vehicle height of about 6 meters. The layout shown • Ejector Rocket - This is a low grade augmentation
here is not intended as a design drawing. It is a schematic by entraining air flow with the rocket exhaust. It in-
sketch to estimate size and shape of the cylinder and cone creases thrust in the first stage by increasing mass
sections, from which the drag can be estimated. The lay- flow.
out grid lines are at 25 cm spacing.
• Carrier Aircraft
At our assumed launch altitude of 4600 m, air density
is 0.769 kg/m3 , velocity averages 120 m/s in the sub- • Balloon Launch
sonic region, the rocket length is 6 meters, and the ref-
erence viscosity of air is 18.27 x 10−6 Pa-s. Therefore
• Jet Boost
the Reynolds number, Re, averages 30.3 million, but it
will change with altitude and velocity. From reference
• Gas Accelerator
data as a function of velocity and Re, the skin friction co-
efficient, C , will vary from about 0.0032 at low velocity,
to 0.00245 at 120 m/s, to 0.00215 at 240 m/s. This is ad-
justed by a correction factor based on the shape of the 5.5 Orbital Assembly
rocket, which in this case is 1.085, and the wetted area to
cross section ratio, which is ~8.3/0.283 = 29.3. Thus the
total drag will vary from 0.102 to 0.078 to 0.068 at the Given a conventional method to get cargo to orbit, and
given speeds, based on cross section area. possibly a Hypervelocity Launcher (previous section),
the next step is an Assembly Station to collect cargo
Drag coefficients in the transonic and supersonic velocity from the high gee launcher, and more sensitive cargo and
ranges behaves differently. humans from conventional rockets, and build larger sys-
If the vehicle had a base area exclusive of the nozzle, we tems there for further space projects. Assembly lowers
would need to add base drag. In the case of a function- the required size of individual launches, and thus the up-
ing rocket, the exhaust fills the base and there is no low front development cost.
pressure area to create a net force by pressure difference
relative to the front. If the vehicle flies other than directly
pointing in the direction of motion, there will be an addi- History
tional component of drag due to lift, but for this estimate
we assume a zero-lift trajectory for simplicity. Prior manned space stations, especially the current Inter-
national Space Station, have used orbital assembly as an
• Trajectory: engineering method. Particular features for assembly in-
clude, first of all, designing the parts to be assembled.
Mechanical docking devices include guides to align the
The launch trajectory cannot be determined by a sim- parts, latches and powered bolts to fasten them firmly,
ple formula or graph, because the thrust, drag, and mass and electrical and other connectors which automatically
of the vehicle are all varying continuously. Therefore a join when the parts are brought together. The heavy
simulation must be done in small time steps so that the duty physical tasks of assembly are carried out by a rail-
above parameters are nearly constant within each step. If mounted robotic manipulator arm, normally controlled
the average values within each step are close to correct, by an on-board operator. Lighter duty tasks are done by
then the total trajectory will be nearly correct. This is too humans in pressurized suits. The experience from that
many calculations to do by hand, so a computer program project is a good starting point, but it has been about 20
or spreadsheet is used. years since the Space Station design was set. Comput-
The simulation will take as inputs variable vehicle masses ers and communications have advanced a great deal since
and a Trajectory Profile, which is how the vehicle tilts then. Additionally, design for continuous growth requires
vs time and varies thrust or does staging. The inputs are a different design philosophy.
182 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Design Approach to other structures. The docking port may be as simple

as a magnet to attract another payload, and then some
The approach in this example uses smaller modular com- bolt or clamp to secure it. The propulsion unit does all
ponents than in the past. In the early construction stages the moving around to line up with the payload. Docking
these are assembled with remote controlled/automated other payloads will automatically connect power and data
robotic arms into larger units. Once sufficient facilities lines. For a first launch, it may be feasible to launch an
are in place, human crew can be added. The robotic electronics unit and a partly fueled thruster unit as a single
work uses experience from the Advanced Manufactur- cargo. Otherwise a larger capacity launch system is used.
ing step. The Assembly Station starts very basic, and
gradually extends it’s capability by adding more modu-
lar parts, and later by manufacturing items locally, rather Fuel Tanks The growing assembly station will use fuel
than just assembling delivered cargo. An important part to meet each cargo as it reaches orbit, and also to make up
of the design is to use standardized modular components. for drag losses from the thin atmosphere that exists at any
That way new parts can be added in any arrangement and low orbit altitude. Therefore it will need periodic fueling.
they will still fit, and new designs are not needed for each
new job. Additionally, use of standard parts makes it eas-
ier to stock spares. Electronics Unit This will contain some smaller solar
arrays for power, some computer systems, batteries, one
Parts like truss elements, which are naturally strong, can or more cameras and GPS units for navigation, and radio
be packed for gun launch, then assembled into a complete or laser communications.
truss. Pressurized modules with rigid walls would not fit
into a gun-launch cargo, but inflatable modules might. Al-
ternately conical or dome sections can be nested and then Robot Arms The next couple of items would be robot
assembled into complete modules. If a vacuum welder arms to give the propulsion unit the ability to do more
or laminated tape winder were available, module assem- complex tasks controlled from the ground. Items like
bly from smaller pieces would be possible. So each com- robotic arms would be subject to a design trade-off. They
ponent needs to be looked at to find the best method of would have to be made very rugged for gun delivery,
delivering it, and it will likely end up a mix of launch versus a lighter weight version launched by conventional
methods. rockets. Arms could be made as segments with one or two
The output of the Assembly Station would be commer- joints, which are connected in series to make more flexi-
cial items like spacecraft or sections of spacecraft, and ble units, and have replaceable tool/manipulator ends for
also internal production that would extend the range of different tasks. The arms are designed as double-ended,
later steps. For example, the Assembly Station could as- so that either end can attach to a base or tool, and have a
semble a mining tug from parts, which then goes to collect split joint to go from one shaft to two or more “fingers”
materials from a Near Earth Asteroid. The Station could or “arms”.
“reproduce” itself in a sense, by splitting off or assem-
bling a subset which can then go off and be the seed for
Robot Attachments These are tools that attach to the
construction in a new location. Initial growth is by simply
arms, and a rail car unit to move the arm from place to
adding more modules of a given size. Later growth can
place.
be by using larger launch systems from Earth when the
economics justify it, or by producing larger components
for later “generations” of construction. Structural Base This is a set of truss elements that
can be assembled into larger arbitrary structures to which
other parts of the growing assembly station will be at-
Component Types
tached. One approach is a ball and stick truss, with
hubs at the intersections that have fittings at 90 and 45 de-
The following is a list of parts for start up of an Assem-
gree angles. These are connected with struts of standard
bly Station. It will require a lot more detailed analysis
lengths to form the framework. The base truss might have
and design to reach a final list, but this will illustrate the
a spacing of 1 meter, with adapters to scale up or down
types of modules that would go into such a design. First
to other grid sizes as needed. Filler plates would span a
launch may be by another launch system, to enable a com-
truss bay to add rigidity or provide container spaces or
plete functioning system to get delivered as a unit. Later
additional mounting locations. The plates can be either
launches can be smaller elements as additions delivered
perforated or solid as needed.
part by part.
The basic structural system includes rails for moving
robot arms and other items from place to place. The rails
Thuster Unit One of the first items to deliver to orbit would extend a short distance from the hubs, with smooth
will be a small chemical propulsion unit. It will include joints to allow continuous motion. Either curved or pivot-
tanks, fuel and small thrusters, and a way to dock firmly ing sections would enable changing the plane of motion.
5.6. HYPERVELOCITY LAUNCHER 183

Utility Grid The concept here is to have a redundant this initial launcher is at the other end of the size scale,
and modular utility system with different services (power, and thus relatively high g-level.
data, fuel lines) added as needed. One approach is to use Conventional rockets are used in parallel with this
a truss column as the utility carrier, and install supportlauncher for the balance of delicate cargo and humans
brackets to hold the various lines, with insulation or me-which cannot withstand the high acceleration. The partic-
teoroid blankets on the sides. That allows for easy accessular type of launcher selected is 17 Particle Bed Heated
for additions or repair. Gas Gun, which is within current technology, has the low-
est development cost to start delivering cargo, and a con-
siderable cost advantage over current rockets. If conven-
Power Units There is another trade-off to do here for
tional launch costs get low enough, though, this step may
photovoltaic arrays, which are not suited to gun launch,
be eliminated.
but are lightweight, versus something like a Brayton gen-
erator, which in theory can be rugged. For low orbit, the
power units would need some sort of storage, i.e. batter- Launcher Scaling
ies, since sunlight is only available 60% of the time. To
start with, simply attaching PV arrays to your structural The largest known hypersonic light gas gun was the
base will provide a power supply. Lawrence Livermore SHARP Gun in the early 1990’s,
which reached 3 km/s with a 5 kg projectile, for a ki-
netic energy of 22.5 MJ. The largest hypervelocity gun
Electric Thruster Unit For more extended missions
launcher was it’s namesake predecessor, the High Alti-
that require more fuel, Ion or Plasma thrusters are added,
tude Research Project (HARP) in the 1960’s. That gun
which are more efficient than chemical thrusters.
was made by welding two 16 inch battleship guns in se-
ries, and was able to fire 250 kg at 2300 m/s using a gun-
cotton charge. This had a kinetic energy of 660 MJ, and
Enclosed Modules Some equipment, and humans,
could have put measurable payload into orbit given suit-
benefit from not being in vacuum. Other tasks benefit
able propulsion on the projectile. The largest known gun
from temperature control, or keeping debris contained.
of any kind by energy was the Gustav type German rail-
For those sorts of requirements an enclosed module is
way siege guns of World War II. It fired a 4800 kg pro-
needed. For early use, an inflatable module may be suit-
jectile at 820 m/s, for an energy of 1.6 GJ.
able. Finished modules are not suited to gun launch, but
a fiber-reinforced aluminum tape could be launched as For development purposes, it is not good to make too
a spool, then formed around a mandrel to create larger large a jump in scale if you are going beyond past experi-
shapes. Concentrated sunlight and pressure rollers can ence. Since the SHARP gun was the largest of that type
braze/solidify layers of tape until sufficient thickness is ever built, we will assume a prototype of about 5 times
built up. That way the small cargo volume of the gun the energy at 100 MJ (12.5 kg projectile at 4 km/s). This
projectiles could be used to fabricate larger items. Once would be followed by a gun large enough to deliver a use-
sufficient habitable volume and supplies are in place, then ful payload to orbit, which we will assume is 10 kg to a
humans can start to work on the Assembly Station, but the 250 km altitude circular orbit. Beyond that, larger guns
initial construction will all be done via remote control. would be sized by expected cargo traffic. Detailed anal-
ysis may change these numbers, but we at least need a
starting point to design to. Please note all the following
calculations are only preliminary to show how the scaling
5.6 Hypervelocity Launcher is performed. For more detailed work, refer to a manual
such as Interior Ballistics of Guns, which accounts for
After establishing an Advanced Manufacturing capabil- more of the real world factors, or do a computer simula-
ity, the next step is a low cost launcher, mainly for bulk tion.
materials. Examples of bulk materials are fuel, water,
oxygen, structural components, and even frozen food.
If there is enough cost advantage in this launcher, you Prototype Gun The location of the prototype gun is
can purposely design parts for the higher acceleration. not critical. It’s main purpose is to reduce the unknowns
For example, electronics can survive high g’s if they are before designing the orbital gun. For convenience we will
mounted properly, but most commercial units are de- assume the White Sands Missile Range as the landing
signed for lowest assembly cost, not high g’s. In general point of the projectiles, since it is large, empty, and des-
the acceleration of this type of launcher goes down with ignated for missile testing. The gun will then be placed on
size. For example, compared to the US M777 field ar- a mountain slope a suitable distance away, avoiding any
tillery, it may have 5 times higher muzzle velocity, but population centers under the projectile path. Any other
150 times longer barrel, thus the acceleration is lower. combination of launch and landing points meeting similar
In the limit of the largest practical gun the acceleration conditions will work. We will further assume intermedi-
would be low enough for humans (60 m/s^2 or 6 g’s), but ate physical dimensions and characteristics between the
184 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

SHARP and Orbital guns, unless there is some technical • Projectile Range = 54 km - If the barrel slope is 12
reason to choose otherwise. degrees, then the projectile is rising at 830 m/s as
Design Inputs: it leaves the barrel. The density of the Earth’s at-
mosphere decreases with altitude, but the equivalent
thickness at constant pressure, known as the scale
• Projectile Mass = 12.5 kg height is about 7.5 km. At a 12 degree elevation,
the total path through the atmosphere then is equiv-
• Muzzle Velocity = 4000 m/s alent to 37.5 km. We can find drag on the projectile
from the formula F(D)= 0.5* CD * rho * A * v^2.
Initial Assumptions: CD is about 0.15 for a conical hypersonic projec-
tile. Rho is air density, which for a starting altitude
• Barrel Length = ~200 m - This is a geometric mean of 2000 m is about 0.95 kg/m^3. Area and initial
between the 50 m SHARP barrel, and estimated 800 velocity of the projectile are given above.
m orbital gun. May be updated by later calculations.
The initial air drag is about 14,000 N, giving a negative
• Barrel Elevation = 12 degrees - This is based on typ- acceleration of 1119 m/s^2. This is a significant fraction
ical mountain slopes north of White Sands of the initial velocity per second, so the actual trajectory
needs to be found by numerical integration (ie a spread-
• Projectile L/D = 8 - This is the length-to-diameter sheet) using small time intervals so the changes in velocity
ratio based on a cylinder shape. The actual shape and drag per time interval, and thus the errors, are small.
will be to minimize drag, so conical at the front. You When this is done the projectile range is found to be 54
want to minimize area to lower drag and barrel size, km, after a 63 second flight, and reaching a peak alti-
so a long and skinny projectile, but not so skinny that tude of 7350 m above sea level. The impact velocity with
bending becomes an issue. 8 is a reasonable starting no landing devices is about 400 m/s. The small size of
point before structural analysis. the projectile and low gun elevation means air drag has
a severe effect on it’s path. This is not a problem for the
• Projectile Density = 1 g/cm^3 - This is the density prototype gun, since we are mainly testing the gun. A
of water, a likely filler of test projectiles, and similar range of 54 km means the projectile can land within the
to the density of LOX/Kerosine, a likely fuel for an White Sands Missile Range and not endanger the public.
orbital gun projectile.
Orbital Gun The prototype gun was mainly concerned
Derived Values: with demonstrating the components of the gun function
properly. The orbital gun demonstrates a larger version,
• Projectile Diameter = 12.5 cm. - Cylinder volume is and additionally a functioning projectile that can deliver
pi*r^2*h, and we have assumed that h = 8D = 16r. a small payload to orbit. The gun energy of 1.18 GJ is
Thus v = 16*pi*r^3, and we have v=12,500 cm^3 about 12 times larger than the prototype.
from the mass and density. Solving for r gives 6.29
Design Inputs:
cm, and we round the diameter of 2r to 12.5 cm.
The projectile diameter is also the barrel diameter,
if you add a small tolerance for a sliding fit. • Net Payload to Orbit = 10 kg to 250 km

• Projectile Acceleration = 40,000 m/s^2 (4,000 g’s) • Barrel Elevation = 23 degrees - As noted below
- This compares to the peak acceleration of 640,000 under Location, Nevada Cayambe is the best loca-
m/s^2 of the SHARP gun and 60,000 m/s^2 for con- tion, and we use the actual slope of the west side of
ventional artillery. Muzzle velocity is sqrt(2*a*d) the mountain between 4200 and 4600 m elevation.
where a is the average acceleration and d is the bar- Above that altitude is a glacier, so we try to stay be-
rel length. Solving for a gives the quoted value. Be- low that.
cause gun efficiency falls off at high velocity, we as-
sume the *peak* acceleration is 25% higher than the
• Projectile L/D = 8 - Use the same value as for the
average acceleration, thus 50,000 m/s^2.
prototype gun. An actual projectile design will be
needed for for a better estimate.
• Peak Pressure = 51 MPa (7,400 psi) - Barrel area is
0.01223 square meters, and gas pressure has to pro-
duce a force of F = m * a = 12.5 kg * 50,000 m/s^2 = • Projectile Density = 1 g/cc - Use same value as for
625,000 Newtons (N). Pressure is Force/Area. Note the prototype gun.
this is much lower than the SHARP gun peak pres-
sure of 400 MPa. Initial Assumptions:
5.6. HYPERVELOCITY LAUNCHER 185

• Barrel Length = 800 m - This is an initial guess at A mass of 82.2 kg subjected to 15,835 m/s^2 peak accel-
a reasonable number. To determine the real length eration requires a force of 1.302 MN. Graphite compos-
requires a detailed enough design in a form you can ite can be assumed to have a strength of 600 MN/m^2,
vary barrel length, see how the changes affect the with a density of 1.82 g/cc. For the given load we need
rest of the gun system, and then find the optimum 1.302/600 square meters of structure = 21.7 square cm.
value. This approach is called variation of parame- The forward parts of the structure only have to support
ters or system optimization, but we need a lot more what is ahead of that point, so we assume the structure
design detail to attempt it. For now we pick a rea- averages 65% of the area over the length of the body.
sonable starting point. Thus the total structure will be 21.7 x 65% x 188 = 2651
cc, with a mass of 4.8 kg. This is only 5.87% of the total
• Muzzle Velocity = 4500 m/s - This is a reasonable mass of the vehicle, so our 15% assumption for the total
guess based on past gun launcher work. It will also empty mass is reasonable.
be subject to optimization later. To get a better estimate of the actual velocity for the
rocket, we have to make a better drag loss estimate. Using
Derived Values a spreadsheet trajectory calculator, we find the projectile
will fall to 2613 m/s after drag losses have fallen 99%,
• Projectile Acceleration = 12,650 m/s^2 (1290 g’s) - giving an estimated loss of 1900 m/s. Since this is higher
Found by the same method a for the prototype gun than our original estimate, we recalculate the projectile
above. Again, this is the average number, so peak mass and try again several time until we get a consistent
acceleration is estimated to be 15,835 m/s^2 (1615 answer. This is called converging to a solution. After
g’s). Note this is about three times lower than the 5 iterations we can estimate the final result is a projectile
prototype gun, mainly because of the longer barrel. mass of 122.5 kg, and a drag loss of 1725 m/s.

• Projectile Diameter = 27 cm - this is found from the

• Peak Pressure = 33.88 MPa ( 4910 psi ) - Found
mass and formulas above. Projectile and barrel areas
by the same method as the prototype gun. Note the
are both 0.05725 m2 .
peak pressure is lower by about 1/3 relative to the
prototype gun, mainly because of the longer barrel.
• Projectile Range = 615 km - This is the distance
the projectile will travel if the rocket does not ig-
• Projectile Mass = 122.5 kg - With a known muz- nite. This is found by following the trajectory cal-
zle velocity and slope, we can try various projectile culator until the altitude reaches ground level again.
masses (in the next several paragraphs) to find the When choosing a launch site, be aware of the impact
one that gives us 10 kg of payload to orbit. The pro- point of a failed ignition. In this case, the impact
jectile mass can be divided into three main parts: point would be near Chiribiquete National Park in
payload, fuel, and empty vehicle. The latter includes Colombia, reached after a 4 minute flight, at a ter-
all the components like guidance electronics, besides minal velocity of 1500 m/s if the projectile reaches
fuel tank and payload support. The on-board rocket the ground intact, which is not assured with a full
is assumed to have an exhaust velocity of 3.3 km/s, load of fuel and aerodynamic heating, If the rocket
typical of a good LOX/Kerosine engine in vacuum. engine ignites and then stops before reaching orbit,
the impact point will be on the Equator somewhere
• Drag Loss = 1725 m/s - As a first approximation, east of the ballistic range, with reduced amount of
assume that drag losses equal 1200 m/s, the rota- fuel.
tion of the Earth at the equator is 465 m/s, and the
sea level orbit velocity plus energy of 250 km al-
titude to be provided by the rocket + gun is 8065 Orbital Gun with 2 Stage Projectile This is an alter-
m/s. If the net velocity after drag loss is 3,300 m/s, nate concept to deliver the same payload as the previous
that leaves 4300 m/s for the rocket. From the rocket design. It uses a 2 stage rocket to see if it improves the
equation, the net mass after the rocket burn is 27.2% overall size. The gun inherently has about 50% higher
of the total. We make a first estimate of the vehicle “exhaust velocity” than the rocket engines, so we divide
empty hardware mass of 15% based on past rocket the total velocity into 3.5 parts to equalize the “difficulty”
designs. Then the net payload is 27.2% - 15% = for each stage. From the previous design, we have a to-
12.2%, and the original mass is 10 kg/12.2% = 82.2 tal mission velocity, including drag losses of 9325 m/s,
kg. Assume the projectile density is 1.0 g/cc. Then therefore the gun would do (1.5/3.5)* 9325 = about 4000
it’s volume is 82.2 liters (0.0822 m^3), and can be m/s. Each rocket stage then has to perform about 2660
approximated by a cylinder 23.5 cm in diameter and m/s.
188 cm long. Since the projectile structure is sub- Using our previous assumptions of 15% empty hardware
jected to a known acceleration, we can estimate the weight and 3.3 km/s exhaust velocity, we can calculate
structure mass from the load it sees. the stage masses as follows:
186 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

• Mass Ratios/stage = 2.241 - found from rocket equa- • Barrel Length = 1600 m - This is set by the distance
tion. This implies final mass = 1/mass ratio = from the bottom of the mountain slope to the glacier
44.63% of start mass. line. For this size gun it is likely not worth the extra
difficulty building through the ice layer.
• Stage 2 payload = 10 kg = (44.63% final mass - 15%
empty weight) = 29.63% stage 2 start mass.
• Barrel Elevation = 23 degrees as previous size.
• Stage 2 start mass = 33.75 kg - From 10 kg / 29.63%.
• Stage 1 final mass = 44.63% start mass, Stage 1 Derived Values:
empty weight = (15% x 57.2% fuel used) = 8.31%
start mass, therefore stage 2 start mass = 36.32% • Projectile Dimensions = 57.5 cm diameter x 460 cm
stage 1 start mass. long - Calculated from mass and density as in previ-
ous sizes.
• Stage 1 start mass = 93 kg = 33.75 kg / 36.32%. This
is significantly smaller than the previous mass, and
the gun velocity is assumed to be 500 m/s lower, so • Projectile Acceleration = 7810 m/s^2 ( 795 g’s ) -
we recalculate drag losses using the ballistic calcula- Again found by same calculation as previous sizes.
tor and converge to a solution. We end up with 1655 Allowing for 25% peak increase, this gives 9765
m/s drag loss, and can thus lower the gun velocity by m/s^2 peak acceleration (just under 1000 g’s). We
70 m/s to 3930 m/s. want lower acceleration for larger projectiles both to
keep the barrel pressure reasonable, and less load on
• Average acceleration is found from a = v2 /2d = the projectile structure with the larger mass.
(3930 m)2 / 2 * 800m = 9653 m/s2 ( 984 g’s ). Peak
acceleration is 1.25 a = 12,066 m/s2 . • Peak Pressure = 45.1 MPa ( 6545 psi ) - This is
slightly higher than the previous size. Note that in
• Projectile diameter is found as above to be 24.5 cm, a large gun with a pressure drop as it fires, only the
and barrel area is 0.0473 m2 . bottom end will see the peak pressure. The muz-
• Peak pressure is found from P = (mass x accelera- zle end can use a lower strength pipe. Using high
tion)/area = (93 kg x 12066 m/s2 )/0.0473 = 23.72 strength steel for the barrel, the peak pressure re-
MPa. quires a barrel wall thickness of around 6 cm, which
is reasonable.
The combination of lower barrel area and lower pressure
leads to a total barrel mass of 57.85% of the previous ver- • Empty Vehicle = 180 kg - The projectile is suffi-
sion. We assume other parts of the gun will scale along ciently larger that we should check the empty weight
with the barrel. The projectile mass is 76% of the previ- rather than assume the previous percent fraction.
ous version. Whether the added complexity of a second The peak acceleration force is 11.7 MN, giving a
stage outweighs the system size reduction is not known. structural area of 195 square cm at the base, and a
total structure volume of 58,300 cc. This comes to
106 kg mass, so our assumption of 15% for total
Operational Gun The operational gun is designed to empty vehicle (180 kg) is still reasonable.
deliver paying cargo to orbit. The actual size will be set by
how much customer traffic is expected, but for discussion
purposes we will assume a 1200 kg projectile launched at • Drag Loss = 1000 m/s - This is found by using the
5 km/s. The location is the same mountain as the previous trajectory calculator to the point you are above 99%
gun. The kinetic energy of this gun, 15 GJ, would exceed of the atmosphere ( 34.5 km above sea level )
the largest previous gun of any kind by nearly 10 times.
Design Inputs: • Payload = 180 kg - The velocity to reach orbit is
8065 m/s. Subtracting the rotation of the Earth (
463 m/s ) and the remaining velocity after drag loss (
• Projectile Mass = 1200 kg - Set by assumption. 4000 m/s ) gives 3602 m/s to be added by the projec-
tile. The rocket equation gives a remaining weight
• Projectile L/D = 8 - As in previous sizes. of 360 kg. Subtracting the empty vehicle leaves 180
kg ( 400 lb ) of net payload.
• Projectile Density = 1 g/cc - As in previous sizes.
Note that the projectile mass increased by about 10 times,
• Muzzle Velocity = 5000 m/s - also set by assump- while the payload increased 18 times over the previous
tion. This is towards the upper range for light gas size. This is due to higher muzzle velocity from the gun,
guns. and lower relative drag losses from the larger projectile.
5.6. HYPERVELOCITY LAUNCHER 187

Large Gun This gun is towards the upper end of length above 99% of the atmosphere ( 34.5 km above sea
that can be built on Cayembe’s slope. The diameter is set level ). Horizontal velocity component is 4,160 m/s
to 1.2 meters. This assumes there is enough bulk cargo at the peak of the ballistic arc, assuming the rocket
traffic to justify the larger gun size. stage does not fire, and altitude will be 194 km at
Design Inputs: this point.

• Projectile L/D = 8 - As in previous sizes. • Payload = 180 kg - For this size gun we assume a
Skyhook from a later step is available. Skyhook tip
velocity relative to the Earth’s center is 5074 m/s.
• Projectile Density = 1 g/cc - As in previous sizes. Subtracting the rotation of the Earth ( 465 m/s )
and the remaining projectile horizontal velocity af-
• Muzzle Velocity = 5000 m/s - As in previous size. ter drag loss ( 4160 m/s ) gives 449 m/s to be added
by the projectile. The rocket equation gives a re-
• Barrel Length = 3200 m - This is the upper end of maining weight of 9330 kg. Subtracting the empty
length that can be fit to the mountain slope. The bot- vehicle leaves 7,705 kg ( 17,000 lb ) of net payload.
tom end is at 3960 m elevation, and the upper end is
at 5300 m, giving a 1340 m rise. The mountain slope Note that the projectile mass increased by about 10 times,
is not as constant at the ends because we are max- while the payload increased 43 times over the previous
imizing length, so some of the barrel will need to size. The additional 4 times gain beyond projectile size
be supported above ground level. Also this extends is due to the Skyhook. With a slight improvement in gun
above the glacier line, so the barrel will either have velocity or Skyhook velocity or orbit altitude, the projec-
to be supported above the ice, or protected from ice tile would not have to add any velocity, only do maneu-
movements. vering to meet the Skyhook landing platform. In that case
the net payload would rise another 20% to 9,225 kg.
• Barrel Elevation = 24.75 degrees - Found from sim-
ple trigonometry of arcsin(rise/barrel length).
Launcher Design
Derived Values: With the scaling of the launcher and projectile com-
pleted, you can now do a preliminary design. Preliminary
• Projectile Dimensions = 1.2 m diameter x 9.6 m long design stops just before you start the final drawings and
- From design inputs above. calculations, and is in enough detail that you can do cost
estimates, and tell if any parts require new research.
• Projectile Mass = 10,850 kg - from dimensions and
density above.
Location As noted above, the location for a sub-orbital
prototype is not critical. For a full orbital gun, the best
• Projectile Acceleration = 3905 m/s^2 ( 400 g’s ) av- location on Earth is probably Nevada Cayambe, the 3rd
erage - Again found by same calculation as previous highest mountain in Equador, and highest point in the
sizes. Allowing for 25% peak increase, this gives world on the Equator at 0 deg N, 78 deg W, 5790 m
4880 m/s^2 peak acceleration (just under 500 g’s). above sea level. It has a nice slope (23 degrees eleva-
By keeping the muzzle velocity the same, but dou- tion) pointing east just below the snow line, and you can
bling the barrel length, the accelerations are halved. position a gun so the terrain slope past the muzzle tends
to decrease. This is obviously critical so projectiles don't
• Peak Pressure = 46.8 MPa ( 6790 psi ) - This is about hit the mountain itself. For now there is a glacier on top
the same as the previous size. of this mountain, even though it sits on the Equator, so
the uppermost 1200m of the mountain would be difficult
• Empty Vehicle = 1625 kg - We again check the to build on. Launching from 4600 m saves you from go-
empty weight rather than assume the previous per- ing through 46% the atmosphere, which significantly cuts
cent fraction. The peak acceleration force is 52.95 drag losses and heating, and an equatorial space station as
MN, giving a structural area of 882.5 square cm at a destination will pass overhead every 90 minutes or so.
the base, and a total structure volume of 423.5 liters. Any launch site off the equator will be limited to one or
This comes to 771 kg mass, so our assumption of two times a day to launch. Other places will work as a
15% for total empty vehicle (1625 kg) is still rea- launch site, just somewhat less efficiently.
sonable. Because of it’s rotation, the Earth is slightly fatter at the
equator. So even though there are taller mountains mea-
• Drag Loss = 460 m/s total velocity - This is found sured from sea level, equatorial ones are “higher” in terms
by using the trajectory calculator to the point you are of reaching orbit, and get more benefit from the Earth’s
188 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

rotation. The latter can be found from the Equatorial Breech This part of the gun is where projectiles are
radius (6378.1 km) plus launch altitude (4.6 km), di- loaded, fueled if needed, and the main reaction force of
vided by the time it takes to rotate once, which is called the gun is passed into the ground.
a sidereal day. This is 86,164 seconds, thus the rotation
amounts to 465 m/s at this location, or 5.77% of the total
velocity to reach orbit. Storage Tanks The Hydrogen needed to fire the gun
is stored at room temperature in ordinary high pressure
Other considerations for a launch site are distance to pop- tanks. The storage pressure in the tanks is higher than the
ulated areas, since the gun is very loud, avoidance of operating pressure in the barrel so that the gas will flow
avalanche zones, and optimum slope. To reach orbit you correctly, but the tank volume will be smaller than the
want the maximum kinetic plus altitude energy remaining barrel because room temperature gas takes less volume
after drag effects. Some barrel elevation will be optimum than the same amount of hot gas.
for this, and finding a mountain with a matching slope
will minimize construction cost. The alternatives of tun-
neling or building a support structure for the barrel would Low Density Tunnel (Optional)
be much more expensive. To find the optimum elevation
for a given gun design, use the trajectory calculator and • See also: Low Density Tunnel
input different angles to find the best one.
In theory you could eliminate the drag and heating of
Barrel The barrel for this type of launcher is basically climbing through the atmosphere by using a vacuum tun-
a large high pressure pipe. The most similar industrial nel beyond the end of the barrel. This would be larger
item is a natural gas pipeline. For cost reasons, the ma- diameter than the barrel, to eliminate friction, and con-
jority of the barrel strength will likely be high strength tinue upward as high as drag savings and cost dictate. In
steel. Since the inside will be exposed to hot hydrogen practice, you can get 93% of the effect of a vacuum tun-
gas for a short period, and some amount of wear from nel with a hydrogen tunnel. Drag is proportional to gas
projectile friction, it may need a liner or coating. The density, and hydrogen gas is 93% less dense than air. It
need for that will be determined by materials and ther- also tends to float, so that simplifies holding the tunnel
mal analyses, which are a standard part of mechanical in the air above the end of your launch mountain. Also,
engineering. The gas and projectile do not produce great hydrogen is your gun propellant, so it just mixes with the
force on the barrel lengthwise, the reaction force to the hydrogen in the tunnel, and it becomes fairly easy to pump
gun firing will mostly be at the back end, which is called back for the next launch.
the breech in gun terminology. Even so, the barrel will You build a large enough tunnel so the hypersonic shock
likely be made of bolted sections, which allows removal waves of the projectile won't destroy it. That also gives
for maintenance, mounting and alignment to the ground, more lifting volume. How far you build the tunnel, or
and sliding fit sections for expansion from general weather whether to use it at all comes down to cost. It can be
changes and heating from firing. added later to a basic gun to hold down initial cost. With
The upper end of the barrel will likely have a flap system such a tunnel in place, you can fire at higher velocities as
to keep most of the air out of the barrel. The flaps are well as lower the overall losses.
pushed open by the remaining air piled up ahead of the
projectile. A “silencer” type device may be needed at the
Projectile Design
muzzle, either to actually lower the acoustic levels, reduce
the muzzle flash from the hot hydrogen burning when it
This example of the projectile design is for the operational
meets air, or to capture the Hydrogen gas to be used again.
gun with a 1200 kg projectile. The other sizes will be
Finally, a shaped nozzle or steam ejector at the muzzle
smaller and simpler versions will lower loads, so this is a
may be needed to ease the change from high acceleration
“worst case” design challenge to solve.
in the barrel to negative acceleration from air drag once
outside it. These options should be analyzed and tested
with the prototype gun. Structure

Heat Exchanger The heat exchanger stores heat from a Propulsion

convenient source (probably gas burners or electric heat-
ing elements), then transfers it quickly to the Hydro-
gen gas when the gun fires. Aluminum Oxide particles, Thermal and Re-entry The projectile needs to survive
commonly used for sandpaper, are used as the storage three periods of high heating: (1) Within the barrel from
medium. They can withstand high temperature, and in hot Hydrogen gas pushing it, (2) While flying up through
grain form can transfer heat quickly due to the large sur- the atmosphere at high speed, and (3) during re-entry so
face area. you can recover and use it again.
5.7. LOW G TRANSPORT 189

The first heating source comes from behind. Hypersonic If the launch site is on the Equator, then the landing
guns often use a Sabot, a disposable structure to help fit point will also be on the Equator so long as the projectile
the projectile to the barrel and provide a better seal for propulsion is always aligned east-west. In this example,
the gas pressure. Using a sabot in this design can protect the landing point can be the relatively unpopulated areas
the projectile from heating either by including insulation, about 80 km east of the Andean mountain chain, which
or thermal inertia, which is simply that solid objects take makes return of the empty projectile to the launch site
time to heat up. If that time is longer than it takes to relatively easy. Then it is a matter of timing the re-entry
leave the barrel, then extra insulation is not needed. The burn of the on-board rocket to set the landing point. The
alternative is to use the rocket nozzle at the back of the size and weight of the empty projectile can be handled by
projectile, which already needs to withstand high heating. a small truck.
In that case it also needs to withstand the high acceleration
force from the gas pressure.
Navigation
The second and third heating sources come from the
front. Flying up through the atmosphere at high veloc-
ity causes the higher peak heating rate, but re-entering Cargo Deployment We assume the cargo being deliv-
from the even higher orbital velocity gives a higher total ered has the same average density as the rest of the ve-
heating, although at a lower rate. The common method hicle, 1 g/cc. Allowing for the projectile structure, the
to deal with high heating rates is an ablative heat shield, cargo diameter is set to 50 cm. Given a mass of 180 kg,
which decomposes and generates a protective gas layer. then the length is 92 cm. Depending how the structure
To keep the projectile pointed in the right direction both needs to be designed, the cargo opening can either be a
flying up and during re-entry may require fins or control side hatch, or the end of the projectile can hinge open
surfaces, which would require heat shielding on them. and the cargo exits that way. Two possibilities for deliv-
ering the cargo are (1) without any station or depot, and
Thermal protection is measured in terms of the amount of (2) with the assistance of a station or depot. On the first
energy it has to dissipate. While flying up through the at- case, the projectile only needs to maneuver to the accu-
mosphere, the projectile goes from 5 to 4 km/s via drag, racy of the desired payload orbit. In the second case, it
thus losing 5.4 GJ of kinetic energy. During re-entry, needs to maneuver close enough for the station or depot
when the projectile is empty, it dissipates 5.2 GJ. In the to rendezvous. The final docking can be done either by
former, the heat pulse starts at the highest rate, decaying the projectile or station. It likely will be more efficient
exponentially as the air pressure goes down with altitude to put the rendezvous system once on the station, rather
with a time constant of about 4 seconds. In the latter, the than launching it each time with each projectile. Given
heat pulse starts slowly as the projectile encounters the the size of the projectile, simply grasping it with a clamp
thin upper atmosphere at high velocity, reaches a peak around the middle may be the easiest way to dock.
at lower, thicker air density but still fairly high velocity,
and then tapers off as velocity falls faster than pressure If the cargo is liquid, an integral (built in) tank will prob-
rises. The exact duration will have to be found by trajec- ably save weight and deliver more mass. Then you will
tory simulation, but is measured in minutes rather than have two projectile designs, for liquid and dry cargo. The
seconds. projectile can be designed with replaceable cargo mod-
ules for liquid, dense, or normal cargo. The weight sav-
The projectile structure is designed to withstand 800 g’s at ings for custom modules will have to be compared to the
launch, and therefore in theory should be able to handle design and operations cost penalty for having multiple de-
a similar deceleration at landing. In reality, empty fuel signs.
tanks can buckle more easily than full ones (think of a
full vs empty soda bottle for example), but the projectile
is still a sturdy device. Terminal velocity is the veloc-
ity a falling object will reach where drag equals gravity,
5.7 Low G Transport
and so velocity stops changing. For our empty projectile,
we can calculate terminal velocity assuming rear control Not every type of cargo can withstand the high acceler-
surfaces add 50% to the total area and have a drag coef- ation of the Hypervelocity Launcher. In particular, hu-
ficient of 0.4. Since the projectile has an empty mass of mans are limited to about 3-6 gravities. So until this
180 kg, gravity produces 1764 N downward force. The step of the combined system, we used whatever existing
terminal velocity is then about 150 m/s. Assuming the launchers were available for people and delicate cargo.
empty vehicle can withstand 2000 m/s^2 (200 g’s), it can With the availability of the Skyhook we now consider the
reasonably sustain an impact velocity of around 20 m/s alternatives for this task and how to choose among them.
using crushable structures. Therefore it needs a device
like a parachute or much larger control surfaces to create Choosing among Alternatives
drag. Forcing the projectile to land sideways instead of
nose-first will increase drag area and drag coefficient and There are already existing rockets for launching cargo and
lower terminal velocity. people into space, and new ones are under development.
190 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

We assume this will continue to be true in the future. that can take off and land like a normal airplane. The en-
In engineering design, parts are subjected to a Make or gine is built around a hybrid approach, it functions both
Buy Analysis to determine whether to make a part in- as a normal air-breathing engine (jet) and a rocket (in the
ternally, or buy it from someone else. This method also high atmosphere). This setup is intended to reduce the
applies to the human transport job. If building our own amount of oxidizer propellant required to send cargo into
launcher is sufficiently better on cost, technical risk, and space as to cut costs.
other parameters, then we do so. If better alternatives An European Space Agency (ESA) design evaluation
are available from others, then simply buy the launch ser- commissioned by the UK Space Agency (UKSA) and
vices. Comparing all the alternatives is called a Trade
concluded in May 2011 stating that “ESA has not iden-
Study. First you choose the parameters to use to compare tified any critical topics that would prevent a successful
with, and a scoring system to convert diverse parameters
development of the engine,”.
into a common scale. You then make the best estimates
for each alternative, and choose the one that scores best. As of April 2012 the funding of the project was mostly
When doing a trade study, it is important to use the same from private investors 85%, and funding is still being
assumptions, such as material strengths, for all the alter- sought to complete the project. The Reaction Engines
natives. Ltd Skylon Users’ Manual (Rev 1, Sep 2009) gives ex-
tensive detail about the vehicle and it’s engine.
Technology and what alternate systems are available
changes over time. There is also uncertainty in the engi-
neering estimates before design and construction is com- Alternative Launchers
pleted. Input assumptions like tons of cargo per year
can vary over time. Thus for a complex system, a single These are potential launchers to add to our combined sys-
point comparison is not sufficient. A Sensitivity Anal- tem. They are not in any particular order. A considerable
ysis looks at variations of parameters and assumptions amount of design work will be needed before reliable es-
ahead of time to see how it affects the final choice. This timates can be made, and then comparison among them
can be done efficiently with a mathematical or computer and existing launchers. So for now this is merely a list
model of the system. Later on, when one of the condi- with some rough estimates.
tions just mentioned changes, the trade study should be
repeated to see if the previous answer is still valid.
Jet Boost The Stratolaunch system currently in devel-
Existing and In-Development Launchers opment uses a subsonic carrier aircraft. The Jet boost
launcher uses military fighter engines to reach supersonic
As of 2012, the following launchers are specifically de- speeds and higher altitudes. Both systems share the idea
signed to transport humans, which requires a pressurized of using air-breathing engines for the early part of the
environment and other design features. Other existing flight, which are 4-20 times as efficient as rocket engines.
and under-development launchers can deliver cargo, and They also avoid using rocket engines in the least effi-
some of those could be adapted to carrying humans. cient part of their operating range: going vertically, which
causes gravity loss, and through dense air where you have
drag and engine pressure loss. Jet boost dispenses with
Soyuz most of the carrier aircraft by using vertical launch and
landing. Using wings allows getting more mass off the
ground, but they also limit operating altitude. Less hard-
Shenzhou
ware to develop should lower the development cost. The
engines are mounted to a Booster Ring, which in turn
Crew Transportation System This is a NASA funded carries the rocket stage. The booster ring lifts the rocket
project with multiple private sector contracts to develop to around 15 km altitude and 480 m/s (Mach 1.6) veloc-
components and ultimately a functioning transport sys- ity. The rocket ignites and continues it’s flight from there,
tem. As of April 2012, proposals for the next stage of while the booster ring returns to a vertical landing at the
development were being reviewed by NASA. launch site.
Early Version
Space Launch System For human transport, the minimum capacity is 1 person.
Extrapolating from the SpaceX Dragon capsule mass,
Stratolaunch which carries up to 7 people, we estimate total mass to
orbit as 1,500 kg, of which 750 kg is passenger and life
support, or uncrewed low g cargo. In an early version the
Skylon A project from Reaction Engines Limited Skyhook would not be present and the launcher is used
(REL) with Alan Bond directing the efforts. The Sky- to deliver the first components for orbital assembly. Air-
lon spaceplane is designed as a single-stage-to-orbit craft, breathing boosters function better with more air, so un-
5.7. LOW G TRANSPORT 191

like an all-rocket system, they prefer to launch at low al- and keep other values as above. The mass ratio is then
titude. We assume a sea-level equatorial launch site. For 4.2, leaving 23.8% of start mass after rocket burn. Net
a 200 km altitude circular orbit a delta V of 7,900 m/s cargo mass is 12.8% of rocket initial mass. With a 20
is required from 15 km, including potential and kinetic ton rocket stage, that provides 2.5 tons cargo to the Sky-
energy. The Earth’s rotation contributes 465 m/s, and hook, or about 3 human passengers. If larger payloads
gravity, drag, and pressure losses are assumed to be 200 are desired, then the booster ring would need more than
m/s from that starting altitude. Therefore the net velocity 3 jet engines. A reasonable limit would be 8 jet engines,
for the rocket stages is 7635 m/s. which can lift up to 56 tons of rocket stage, and deliver
7.15 tons of cargo.
We assume a re-used two stage chemical rocket with ex-
haust velocity of 3350 m/s, similar to the SpaceX Merlin
1C extended nozzle engine. Since ignition of the rocket is Gun Boosted Ramjet Ramjets are mechanically sim-
at altitude, we optimize it for vacuum thrust, which is ef- ple compared to turbine type jet engines, so potentially
fectively the operating condition after the first 20 seconds low cost. The drawback is they do not function at low
of operation. We increase the Falcon inert mass from velocity, so for this alternative we assume a low accelera-
6.5% of stage mass to 11% of stage mass to account for tion gun is used to reach sufficient velocity for the ramjet
heat shield and other stage recovery hardware so it can be to operate. At higher velocity, ramjets lose performance,
used again. Each stage is assigned 50% of the required so the vehicle will use rocket power to finish the mission.
velocity, so the calculations are as follows:
The gun location is assumed to be on a mountain slope
with a barrel length of 6 km, and the ends at 3200 and
• Stage 2 delta-V = 3817 m/s. Mass ratio = 3.125, so 4200 m elevation, such as the SW slope of Cayambe,
final mass = 32% of start mass. Stage inert = 11% Ecuador. Acceleration is limited to 6 g’s (60 m/s2 ) for hu-
x 68% of start mass fuel consumed = 7.5% of start man passengers, so the muzzle velocity is 850 m/s (Mach
mass. Thus payload = 24.5% of start mass, and also 2.8). An uncomplicated ramjet will operate roughly over
equal to 1500 kg from above. Therefore Stage 2 start a 2:1 velocity range. Beyond that requires more com-
mass = 6,122 kg. pensation in inlet shape and combustion conditions, so
we assume the maximum velocity will be 1700 m/s. Av-
• Stage 1 delta-V = 3818 m/s. Mass ratio = 3.126, so
erage equivalent exhaust velocity is about 14 km/s over
final mass = 32% of start mass. Stage inert = 11%
this range, using hydrocarbon fuel. We will assume sin-
x 68% of start mass = 7.5% of start mass. Thus
gle stage to orbit and do calculations purely on theoretical
Stage 2 + Payload (what the first stage has to carry) =
performance for now.
24.5% of start mass, and also = 6,122 kg, thus Start
mass = 24,989 kg, which we round up to 25,000 kg. Single Passenger Scaling
For a single passenger minimal system, we again assume
• A modern fighter engine such as the PW F-135 gen-
a 1500 kg capsule with 750 kg of delivered human + life
erates 191 kN thrust on full afterburner at sea level.
support, or low g cargo. Calculations are as follows:
For performance reasons, we want to take off at 2.0
gravities, thus the allowed mass is 9.74 tons per en-
gine. The engine itself (1700 kg) , fuel (450 kg), • Rocket mass: 12,500 kg - The rocket stage needs
and booster ring hardware (590 kg) has an estimated to supply 5,900 m/s net, which implies a mass ra-
mass of 2.74 tons . Thus each engine can lift 7 tons tio of 5.88, or 17% final mass. With 11% harware
of rocket stages and payload, and we need 4 engines mass, we end up with 6% payload. Our initial rocket
for the 25 ton rocket with some margin. mass is therefore payload in kg/payload in percent =
12,500 kg, or about 1/3 lighter than the Jet Boost
concept.
The net payload to orbit of 3% of the rocket initial mass
is not remarkable, but the ability to recover and use all • Ramjet thrust: 400 kN - At an average climb rate of
the stages repeatedly is. Liftoff mass of the booster ring 210 m/s, we want the ramjet to gain 850 m/s velocity
+ rocket is 36 tons, about an order of magnitude smaller over 40 seconds, or a little over 20 m/s2 . Therefore
than the Falcon 9 vehicle + Dragon capsule, and it should the ramjet thrust needs to be 250 kN for accelera-
therefore be proportionally less expensive to develop. If tion. Drag is roughly estimated at 150 kN, so to-
not too much low-g cargo needs to be delivered to orbit, or tal engine thrust is estimated at 400 kN (90,000 lb)
if other launch systems reach comparable operating costs, A very rough estimate of engine size would be 1.0
then this system may not be justified. Buying launch ca- m2 in area. Since this is less than human passenger
pacity from someone else would be less total cost. capsule size (1.6 m seated), the passenger size will
Advanced Version govern barrel diameter.

For an advanced version, we assume the Skyhook is in • Ramjet mass: 3950 kg - Ramjet Thrust to engine
place and reduces the required velocity rocket to 4,810 mass ratio averages about 20:1, thus the engine will
m/s. For this version we assume a single rocket stage, have a mass of around 2000 kg. Fuel required is
192 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

about 1150 kg, and remaining ramjet related parts Low Acceleration Guns You can launch people and
about 800 kg. So total ramjet stage would be 3,950 delicate cargo with a gas pressure type accelerator if you
kg. lower the g forces sufficiently. That forces the barrel
length to be as long as possible, so we need to look at ge-
• Total mass: 16,500 kg - By adding the rocket and ography to select a location. Two good locations present
ramjet stages, or about 40% less than the jet boost. themselves, although others may be possible.
It should be emphasized that these are preliminary
calculations.
Island of Hawaii
• Gun pressure: 500 kPa - A 1.6 m barrel accelerating
16,500 kg at 60 m/s2 requires a total force of 990 Hawaii is the best location on Earth as far as a large con-
kN. Dividing by the barrel area gives a pressure of stant slope mountain, requiring minimal grading and sup-
492 kPa ( 71 psi ). This is not expected to be a port for the barrel, and so lower construction cost. An
difficult challenge from a technical standpoint. More equatorial site would be preferred to meet up with the
of a challenge will be installing 6 km of pipe on a Skyhook, but let us first look at Hawaii. It is a shield vol-
mountain. cano and cooling lava flows at a constant slope. Therefore
you have a nearly perfect ramp on the west side of the is-
Small Prototype Scaling land pointing up to the east to build on about 22 km long.
You could get as much as 100 km if you extend down
To build a small scale demonstrator for this concept, let us into the ocean or add support towers on the eastern slope,
assume a payload of 20 kg to orbit, with a higher allowed but that would be more expensive than building at ground
acceleration of 10 g’s, and a two stage rocket. The higher level. For a 100 km long version at 6 g’s the muzzle ve-
acceleration allows us to reach 900 m/s over a shorter bar- locity could be as high as 3,460 m/s, but we will use 20
rel length of 4 km, and the ramjet function up to 900 m/s. km for this example.
The net velocity for the rocket stages is then 5,835 m/s,
or 2918 m/s each. For a smaller size we assume slightly Design Scaling - Assume a 20 km long pipe x 10 m
lower exhaust velocity (3300 m/s) and higher hardware diameter, pushing a 500 ton single stage multiple use
fraction (15%). Mass ratio for each stage is 2.42. Weights rocket. The vehicle will not fill the whole pipe, it is shaped
are calculated as follows: for aerodynamics, and rides on a sled and pusher plate that
fits the pipe. It works out the pressure in the barrel needs
to be 2 atmospheres (200kPa, 30 psi) to give you 3 g’s
• Stage 2 final mass = 1/mass fraction = 41.3%
acceleration, safe for most humans (general public) and
• Stage 2 payload mass = final mass - hardware = satellite parts. Muzzle velocity is 1100 m/s (Mach 3.6),
41.3% - 15% = 26.3% = 20 kg (by assumption) which is not a huge fraction of orbit velocity, but a nice
running start before you light up your on-board rocket.
• Stage 2 initial mass = 20 kg / 26.3% = 75 kg Given those starting conditions, a non-cryogenic rocket
should have a payload of around 35 tons, which along with
• Stage 1 final mass = 41.3% (same velocity as 2nd a 10 meter maximum diameter should be plenty for any
stage) cargo or people you want to launch. This is the upper end
• Stage 1 fuel used = 1 - final mass = 58.7 % of what you might want to build in terms of barrel diame-
ter. For higher mass vehicles, you just need higher oper-
• Stage 1 hardware weight = 15% x fuel used = 8.8% ating pressure in the barrel. A first low-g cargo launcher
can be a lot smaller than 10 meters, and increased in per-
• Total Stage 1 = fuel + hardware = 67.5% formance by adding length or going to larger barrels over
time. Hawaii is about 20 degrees N latitude, so a launch
• Stage 2 then = 32.5% of launch weight.
from there would not be able to reach an equatorial Sky-
• Total mass = Stage 2 / 32.5% = 231 kg hook, but it would deliver more passengers and cargo than
an unaided rocket.
At 20 m/s2 acceleration, the ramjet needs to provide
about 5000 N thrust ( 1100 lb ), which only requires Cayembe, Ecuador
roughly 1/80 square meters engine area. The rocket
stages can be represented by a cone 0.5 meters in diame- Cayembe is the name of both a city and large mountain
ter and 3.5 meters tall with a density of 1, so the engine is about 50 km north-east of Quito, Ecuador. We previ-
small relative to the rocket stage diameter. Ramjet mass ously discussed a hypervelocity launcher on the side of
would be around 25 kg, and fuel used about 15 kg. To- the mountain. For transporting people, the barrel will
tal launch mass would then be 271 kg. Allow 29 kg for need to be much longer for lower acceleration, and extend
carrier/sabot to fit the barrel, and we have an accelerated west somewhat past the town. For this version we assume
mass of 300 kg. At 100 m/s2 , the acceleration force then a trained crew rather than general public. With pressure
is 30 kN, and the required pressure is 152 kPa ( 22 psi ). suits, conforming seats, +x acceleration (forward facing
5.7. LOW G TRANSPORT 193

Rocket Performance - The net velocity required for the

rocket is found from the Skyhook tip velocity relative to
the Earth’s center (5074 m/s), less the Earth’s rotation at
the equator (−465 m/s) and gun velocity (−1560 m/s)
plus drag loss (+124 m/s) and other losses and maneu-
vering which we make an estimate for (+200 m/s). This
comes to 3,373 m/s net. The SpaceX Merlin engine has
an exhaust velocity of 2980 m/s. 4-6 engines will prob-
ably be required for sufficient thrust. The rocket equa-
tion gives the rocket mass after reaching the Skyhook as
32.2% of initial mass. Allowing 10% for the vehicle itself
gives 22.2% payload, or 89 tons. This is a large passenger
and cargo capacity, with a correspondingly large Skyhook
to support the arrival mass. A first version would likely
20 km Barrel Fit to Terrain be smaller.
Knowing the area of the barrel and the rocket vehicle
mass and acceleration, we can calculate the required pres-
seats), and crew in good condition you can safely use 6 sure as 1.91 MPa ( 277 psi ) for the 4 meter gun and 1.22
g’s, and thus get a muzzle velocity of 1560 m/s. That’s MPa ( 177 psi ) for a 2.5 meter gun. The challenge will
Mach 5.2, or 20% of orbit velocity. The Skyhook has not be barrel pressure, but filling it fast enough when the
been available since the previous step in the combined projectile is moving rapidly. The length will likely re-
system example, which subtracts another 2400 m/s from quire tanks and valves space out along the barrel. The
the rocket stage requirement. muzzle velocity will likely require a heated gas to fill the
pipe, but exactly which gas will be left for detailed analy-
The geography of Ecuador is not a smooth slope like sis. Large gas accelerators have reached above twice the
Hawaii. We assume the barrel is 20.25 km long, but muzzle velocity, so it is more a matter of lowest cost than
curved upward with a segmented radius that keeps cen- feasibility.
trifugal acceleration at or below 12 m/s^2. That will be
felt by passengers as a vertical acceleration (head to toe). Spaceport Growth - We had previously built an opera-
The barrel will need to be supported on towers or use tun- tional Hypervelocity Gun on the mountain with a muzzle
nels as needed to fit the terrain, and the curvature roughly velocity of 5000 m/s and an unaided payload to orbit of
fits the geography, which is flat initially, rising to a moun- 180 kg. With the Skyhook in place, we can calculate the
tain at the end. The ends are at 2778 m elevation south new payload as follows:
west of the town, and 5731 m at the top of the moun-
tain, with initial and final slopes of 1.4 and 12.4 degrees • The Skyhook’s tip velocity relative to Earth’s center
caused by the curved barrel The gentle curvature keeps is 5074 m/s. Earth’s rotation deducts 465 m/s. Drag
the vertical acceleration low relative to the forward accel- loss is 1000 m/s from the initial 5000 m/s. Trajec-
eration. The higher slope at the muzzle end also allows tory elevation of 23 degrees means the horizontal
faster climb through the atmosphere and less drag loss. component, which is all that counts for getting to
These assumptions may be changed with more detailed orbit, is cos(23 deg) = 0.9205 x 4000 m/s after drag
analysis. We assume the rocket stage is 4 x 32 meters in = 3682 m/s. We allow an extra 200 m/s for ma-
size, and closely fits the barrel, with a mass of 400 tons at neuvering and other unaccounted losses. So the net
launch. delta-V of the rocket becomes 1127 m/s.
Drag - With a drag coefficient of 0.2, the rocket stage • Using the same exhaust velocity as the SpaceX Mer-
will see 1.93 MN of drag at the muzzle, producing −4.82 lin engine (2980 m/s) but at 1/60th the thrust level,
m/s^2 deceleration if the rocket does not ignite immedi- we get a final mass of 68.5% x 1200 kg start mass =
ately. The climb rate of sin(12 deg) x 1560 m/s = 335 822 kg. With the same empty vehicle mass of 180
m/s. The equivalent thickness of the atmosphere is called kg as the version before the Skyhook, we now have
the scale height (8640 m vertically) over which the pres- 642 kg payload, or about 3.5 times as much.
sure drops by a factor of e (2.718...). An exponential
pressure decay per scale height over many km is how the Going from 642 kg payload with a 60 cm caliber (barrel
real atmospheric pressure changes, but it can be approxi- diameter) gun to 89 tons with the 4 meter caliber human
mated as the muzzle pressure for one vertical scale height accelerator is a factor of 139 times larger. Since the Sky-
and then dropping to zero. 8640 m scale height / 335 m/s hook has to be enlarged for the larger delivery mass, a
vertical velocity = 25.76 sec. Multiplied by the decelera- program of gradual improvement will be needed. The
tion the total drag loss can be estimated at 124 m/s. This launchers will add barrel length and move to larger diam-
value will change depending when the rocket is started, eters in steps, and use part of their cargo to deliver Sky-
since drag is a function of velocity. hook cable and other materials, so that later deliveries
194 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

with more payload can be handled. If orbital mining can 5.8 Electric Propulsion
supply sufficiently strong materials, they can be used, but
otherwise they can come from Earth. A smaller version of Electric propulsion typically has about ten times the fuel
the human accelerator than the one above could use a 2.5 efficiency of chemical rockets. Thus they turn an expo-
x 20 m size rocket vehicle with a mass of 100 tons. Using nential fuel requirement (fuel to push more fuel) into a
similar calculations, we end up with 20 tons net cargo for nearly linear one for most Solar System missions. The
it. At some point the low-g accelerator would be too small timing of this step would be in parallel or soon after
for seated human passengers, probably around 1.6 meter Orbital Assembly is started.
diameter, but they can still be used for sensitive cargo.
Bulk non-sensitive cargo will always have a cost advan-
tage because the higher muzzle velocity lets you deliver
3 times more payload as a percentage of rocket vehicle 5.8.1 Near Term Electric Thruster Types
weight, so it makes sense to keep both types of launch-
ers. There are several kinds of electric thrusters that are good
candidates for near term use. The selection here is based
Depending on traffic needs, you may want to keep smaller on state of development and usefulness:
launchers operating in parallel with their larger replace-
ments. In theory you could launch every time the Sky-
hook passes over in its orbit, which is every 100 minutes, • 49 Electrostatic Ion - This knocks electrons off
but barrel cool down or other needs may prevent firing of gas atoms making them charged, which is called
a given gun that often, so having several may be useful. “ionized”. Once charged, they can be accelerated by
At the upper bound, delivering 89 tons per launch x 14.4 metal screens with a large voltage difference. Ion
orbits per day x 300 days per year (allowing some main- thrusters are used on some communications satel-
tenance time) yields an astounding 384,480 tons/year to lites, and the Dawn spacecraft currently exploring
orbit. This compares to the ~1,000 ton/year capacity of the asteroids Vesta and Ceres.
current and near-term launchers worldwide.
Cost - At this point, cost has not been estimated to any de- • 51 Microwave Heated Plasma - This type uses mi-
gree of accuracy. The Falcon 9 rocket has a total mass of crowave frequency heaters to heat the fuel. This is
333.4 tons and a payload to low orbit of 10.45 tons. So the the same principle as a microwave oven, but much
ratio of rocket mass besides payload to payload is 30.9 to more intense. Above a certain temperature the
1. The bulk cargo launcher has a non payload mass of 558 heated atoms in the fuel will knock electrons off
kg vs 642 kg payload, or a ratio of 0.87 to 1. This 35.5 each other, turning it into a mixture of ions and
to 1 advantage should lower costs significantly, but not in electrons, which is called a Plasma. The plasma
that exact ratio. The gun and Skyhook are large instal- is contained and directed by magnetic fields. You
lations relative to the rocket stage, and their cost per use need to do that because plasma is so hot it will
will depend on how many times they are used. The Fal- melt anything it contacts, or cool itself down too
con 9 hardware is not currently reused, while the rocket much. In fact on Earth plasma is used as an effi-
stage is intended to be used multiple times. De-orbiting cient way to cut through metal. A version of this
from the Skyhook is 63% of the unaided velocity from thruster is currently being developed on the ground,
orbit, and thus (0.63)^2 = 39.5% of the kinetic energy to and will soon fly for testing on the Space Station.
dissipate. This makes the heat shield easier to design, and It’s full name is Variable Specific Impulse Magne-
the stage is pretty rugged in design, since it needs to be toplasma Rocket, which is mercifully abbreviated
fired out of a gun at high acceleration. So in principle it to VASIMR. As a category they are called plasma
should be able to be recovered and used again. thrusters.
In the absence of more detailed estimates, for now we will
adopt the 35.5 times reduction in rocket size per payload • 72 Ionospheric Current - This operates like a mo-
and apply it to the $54 million/10,450 kg = 5,167 $/kg tor, running a current in a wire in a magnetic field.
Falcon 9 cost, to get a first estimate of 146 $/kg. To com- The return path for the current is the ionosphere.
pare to some popular consumer items, the iPad 3 64 GB This method is limited to places with suitable mag-
costs 583 $/kg including packaging, and a Toyota Camry netic field and ionosphere density, but low Earth
is about 15 $/kg, although neither is designed to survive orbit fortunately is such a location. The attraction
high-g launch. is it does not require direct fuel use, only a little
leaked plasma to make electrical contact with the
ionosphere. The equivalent exhaust velocity as if it
were a fuel-using engine is 250 km/s. Since low or-
bit is the first place we want to use, developing this
type of thruster is a high priority. Note it is not as
fully developed as the other types.
5.8. ELECTRIC PROPULSION 195

5.8.2 Electric vs Chemical Thrusters tio is significantly less than the local gravity acceleration.
They can be used indirectly via space cable/elevator type
All rockets work by tossing mass in one direction, and systems. Chemical engines can reach vehicle thrust-to-
by Newton’s Law (for every action, there is an equal and mass ratios well above Earth gravity, which is one reason
opposite reaction), the rest of the rocket gets pushed in they have been the primary way to launch things to date.
the other direction. The faster you toss the mass, the more Both Ion and Microwave Plasma thrusters have exhaust
push (momentum) you get out of it. Conventional rockets velocities in the range of 20 to 50 km/s, so are 4 to 10
burn fuel in a chamber then let it expand out a supersonic times more fuel efficient than conventional rockets. Like
nozzle to get it going as fast as possible. The shape of electric devices on Earth, they are rated by how much
the nozzle is governed by the physics of expanding gases, power they use. The Dawn spacecraft has a 10 kW set of
which is why they all look more or less the same. How solar panels, and the VASIMR thruster in development is
fast it can get is limited by how hot the gas is and it’s rated at 200 kW. Generally ion thrusters will maintain ef-
molecular weight. The best combination used today is ficiency at lower power levels than plasma type thrusters
burning Hydrogen and Oxygen in a ratio of 1:6 by weight. because ion flow does not have to be restrained by con-
This produces mostly steam with a bit of Hydrogen left tainment fields, while plasma requires a field to keep it
over to lower the average molecular weight. How fast the separated from the solid hardware. At small sizes the
gas is going is technically called exhaust velocity, and is plasma volume vs total engine volume becomes small and
limited to about 4.5 km/s for this fuel type. efficiency drops.
Electric thrusters are not limited by the energy produced For efficiency reasons, ion thrusters prefer high atomic
by burning the fuel. They feed energy to the fuel from an weight fuels. The energy to ionize an atom is roughly con-
external source, thus can get much higher exhaust veloc- stant across the Periodic Table, but does not contribute
ity. This gives you more push from given amount of fuel. to thrust in this engine type. Thus using high weight fuels
Since you have a finite amount of fuel to use this is more lowers the portion of total power used for ionization rel-
efficient in direct proportion to the increase in exhaust ve- ative to acceleration. Typically Xenon is used as a fuel.
locity. By analogy to automobiles, you are getting better Plasma thrusters can use most fuel types since their goal is
“gas mileage”. to make the plasma extremely hot, on the order of a mil-
The extremely high fuel efficiency is the key to why this lion degrees. By tuning the microwave generators, most
type of thruster is important. If you are doing a lot of atoms and molecules will absorb the energy. A key ad-
moving about in space the fuel savings outweigh (literally) vantage of this is fuels like Oxygen or water are common
the mass and cost of the power supply by a large margin. in asteroids, so electric thrusters can be refueled locally,
Conventional rockets only need a fairly lightweight fuel rather than having to bring all the fuel from Earth.
tank, but burn a lot more fuel. One drawback to electric
thrusters when transporting humans is their relatively low
thrust. This makes the trip times longer. There are var- 5.8.4 Electric Propulsion Applications
ious ways to work around that drawback. For example,
passage through the Earth’s radiation belts slowly would The following early missions can be performed starting
be unacceptable radiation exposure. So you can transport with relatively small thrust levels, and working up to more
your main vehicle by electric thruster, taking weeks, and ambitious missions.
then deliver a crew in a small capsule taking hours once
the main vehicle is outside the radiation belts. Atmosphere Mining This mission involves collecting
air from the edge of the Earth’s atmosphere for fuel and
breathing. We start with a 50 kW solar array and a
5.8.3 Comparisons Between Types VASIMR type thruster which can generate 2 Newtons
thrust at 40% efficiency and 20 km/s exhaust velocity.
All electric thruster types need an external power supply The solar arrays are assumed to use modern multi-layer
since the fuel is not self-heating as in chemical engines. cells with 30% efficiency, and have a power to mass ratio
The most common power supply used in space are photo- of 100 W/kg. The array will thus mass 500 kg, and we as-
voltaic panels. Those can get unwieldy at power levels of sume operates 30% of the time by intermittent use. The
hundreds of kW or more, and their power output per area electric thruster can then produce an average thrust of 0.6
drops as the inverse square of distance from the Sun. So Newtons. At 200 km altitude, each square meter of col-
for some past and future missions a nuclear power source lector generates 0.0129 Newtons of drag, so the total col-
is preferred. Smaller size nuclear generators are based lector allowed area is 46 square meters to match the aver-
on isotope decay, and larger ones are full nuclear reac- age thrust. This will collect 0.08 g/s, and the thruster con-
tors. Any type of nuclear device brings both technical sumes 0.03 g/s, leaving a net of 0.05 g/s. This amounts to
and political complications. 4.32 kg/day, or 3.15 times the solar array mass per year.
Electric thrusters cannot be used directly for launch or Later expansion would take the same thruster module to
landing on large objects, because their thrust-to-mass ra- 200 kw power level and 5.7 N thrust at 50 km/s exhaust
196 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

velocity and 60% operating time. The operating time is 5.8.5 Moved Text to be Merged
limited by the 40% of the orbit in the Earth’s shadow.
Thus average thrust is 3.42 N, and collection rate is 0.456 [MOVE to Phase 4B High Orbit] Further expansion of
g/s. The thruster uses 0.114 g/s, leaving a net of 0.342 production may leads to power satellites, which beam en-
g/s. This is 29.5 kg/day or 10,785 kg/year, or 5.4 times ergy to Earth for 24-hour power. If this can be done
the array mass per year. With a 15 year service life for economically, it would likely be the largest export mar-
the arrays, they can supply 75 times their mass in total. ket to Earth. Solar-thermal with storage works in sunny
An electrodynamic thruster to make up for drag might climates on Earth, but many people don't live in such cli-
improve on this even further. mates. Solar flux in space is 10 times higher than low sun
For human transport, where speed is important going climates on the ground. Energy delivered from orbit may
through the radiation belt, the collected air can be sepa- prove cheaper overall, despite the extra cost of building
rated for Oxygen, and mixed with added Hydrogen from in space. This is especially true if most of the materials
Earth in a chemical thruster. Alternately a lower exhaust for the satellites and their production equipment can be
velocity, higher thrust electric thruster could be used, sac- sourced from space, and the production is highly auto-
rificing fuel efficiency for fast transit. There are several mated.
plasma and arc jet thrusters that could do that job.

Orbital Cleanup and Maintenance Earth orbit has 5.9 Orbital Mining
accumulated debris from spacecraft explosions and colli-
sions, and there are a number of non-functional satellites This step involves mining small asteroids in orbits close
which only need a single part repaired or new fuel to func- to the Earth for raw materials. These are called Near
tion again. This mission involves using a range of electric Earth Objects or NEOs. Although the Moon is phys-
thruster vehicle sizes to collect the debris, repair or refuel ically closer, it has a significant gravity well, so it is not
satellites on location, or bring them to the orbital platform the best choice for a first mining location. Small aster-
for maintenance. Debris mass ranges down to centimeter oids have essentially no gravity well, so efficient Electric
or less in size, so it would be inefficient to send a large Propulsion can do the job of getting close, and then leav-
vehicle to collect it. Alternately, satellites can range up ing. Both the Moon and NEO’s have much smaller grav-
to several tons in mass. Therefore we select electric ve- ity wells than the Earth, so from an energy standpoint it
hicle sizes to match the size of what is being collected or is much easier to fetch materials from them. That ad-
moved. For debris collection, several pieces in similar or- vantage grows when the location for mining and eventual
bits can be collected in one trip to minimize fuel use and use of the materials are closer in energy terms. Once you
mission time. The fuel for these cleanup missions comes start building installations on the Lunar surface, doing lo-
from the atmosphere mining. Depending on what the tar- cal mining there will be energy efficient, but that will be
get objects are, we perform one or more of the following a later step.
tasks:

• Collect orbital debris and either deliver it to a low 5.9.1 Rationale for Orbital Mining
enough orbit that it will decay and burn up quickly,
or feed the debris into a processing unit to extract
Past plans for space exploration or development have of-
useful materials from.
ten selected the Moon or Mars as destinations. This has
• Return non-working satellite hardware to the orbital to do with them being large and obvious, and having lots
platform to be salvaged for working parts. of land area (we are a territorial species). That does not
mean they are the correct early locations from an engi-
• Repair non-working satellites at the orbital platform neering standpoint. The Earth has a large gravity well
with salvaged or new parts. that is difficult to climb out of, so it should be evident that
• Repair, refuel, or attach a new propulsion unit to ex- you don't want to go down another deep gravity well right
isting satellite at their current location. away, especially if you have to bring all your fuel from
Earth. This question can be viewed in energy terms. The
• Transport new cargo to higher orbits. total energy to provide a kilogram of space hardware in
orbit includes the energy to mine the raw materials, the
The tasks above are approximately in order of size and energy to refine and fabricate the raw materials into fin-
difficulty. Before salvaging used satellites, you would ished parts, and the energy to transport all the matter in-
need to get permission from their original owners. The volved to their final destinations. In the case of starting
legal regime for broken debris pieces is unclear. If from Earth, the transport energy is by far the largest one,
they are considered a menace to navigation, they might and increases the farther you go. Mining materials close
be removed without permission, or the original owners to where you need them can reduce the transport energy,
charged for cleanup. and thus lower total energy needed.
5.9. ORBITAL MINING 197

The Solar System is populated with a large number of times. The motion against the background allows cal-
small objects that do not have deep gravity wells. The culating the orbit. The brightness of the pixel together
ones in orbit around planets, like Phobos and Deimos with the distance from the orbit calculation lets us esti-
are called Satellites. The ones not in orbit around a planet mate size, based on an assumed surface color. At first we
are called Minor Planets, and are named by location as don't know the actual color of the NEO, and so the size
Asteroids for the ones at or closer than Jupiter’s orbit, remains uncertain until better data is collected.
and Distant Minor Planets for the ones beyond that. Orbits of NEOs are grouped by size and shape into
The major groups are further subdivided by orbit loca- classes. All orbits are ellipses, and half the long axis,
tion. The ones which come close to Earth (NEOs) are
called the semi-major axis with symbol “a”, are scaled to
what we are concerned with in this step, since the energy the Earth’s, which is given the value of 1.00 astronomical
to reach them from Earth is low. There is not a great
units (AU). By definition NEO’s have a < 1.3. In other
distinction in composition between smaller satellites and words, their orbit is no more than 30% larger than the
Minor Planets, just in where they happen to be located.
Earth’s, or significantly closer than Mars is (a = 1.523
Even if the mission velocity to reach a particular NEO is AU). Peri- and aphelion are closest and farthest points
somewhat higher than to reach the Lunar surface, and in from the Sun respectively, in a given elliptical orbit. The
many cases it is not, it can all be performed with electric major NEO classes by orbit size are mostly named after
propulsion. Electric thrusters are 6-10 times more fuel characteristic members of the class. These are:
efficient than high thrust chemical rocket engines required
for landing on the Moon. So total fuel use is less, often • Apollo - a > 1.0 AU and crosses the Earth’s orbit
by a significant margin. Once Lunar infrastructure is in (perihelion < 1.02 AU)
place, particularly to produce fuel, the Moon will be more
accessible. But that will be covered in a later step. For • Aten - a < 1.0AU and crosses the Earth’s orbit
now we start with what is easiest. (aphelion < 1.0167 AU)
• Amor - Perihelion from 1.02 to 1.3 AU = always
outside the Earth’s orbit
5.9.2 Steps Towards Mining
• Inner Earth Objects - Aphelion < 0.983 AU = al-
Currently (2012) we don't know enough about NEOs to ways inside the Earth’s orbit
properly plan mining them. So before actually doing the
mining, we need to do prospecting to find out what is Sizes and Masses:
there in detail, and find out the best method of collect-
ing the materials. A few missions (see List of Aster- As of 10 Dec 2012, 9377 NEOs have been discovered,
oid Spacecraft on Wikipedia) have gotten close enough and about 900 new ones are being found per year. Their
to asteroids for detailed observations, but so far we have sizes can be approximated as a number N larger than di-
not visited specific ones that are easy to return materi- ameter D (in km) by the formula
als from. Some NEOs are large enough to observe de-
tails from Earth (NEO Physical Properties - Lupishko,
2002), or are small but have passed close to the Earth,
N = 1090 × D−1.95
giving a short window for high resolution observation.
The accumulated data so far are not sufficient, so addi- or roughly the total number is proportional to the inverse
tional observations and close up prospecting missions are square of size. Total mass of all the NEOs together is
needed. roughly 60 trillion tons, but this value is far from exact.
The largest known is 1036 Ganymed, which is about 32
Current Knowledge km in diameter. A small number of NEO’s have had
density estimates made, which range about 1.5-2.67 g/cc.
Although we don't yet know enough about NEOs to mine Comparing that to the most common meteorites, ordinary
them, we know a great deal more than nothing, and the chondrites at 3.0-3.8 g/cc indicates 30-50% of their orig-
data are accumulating rapidly. inal volume is low density volatiles, which get burned off
when reaching Earth, or is empty space. In the latter case,
Orbits: an NEO is better described as a pile of rocks than a solid
object. One gram/cubic centimeter is also one metric ton
The first things we discover about NEOs are generally per cubic meter, so the mass in tons can be found from
their location in terms of orbital elements, and an esti- the density times volume in cubic meters.
mate of size based on brightness. The survey telescopes
to detect these types of objects are not the largest ones Rotation and Shape:
available, and they are working to their detection limits.
Thus when an NEO is found, generally it is from single- Rotation rates can be determined from variation in bright-
pixel images against the stellar background at different ness, even when they only show up as a single pixel on a
198 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

CCD detector. Therefore rotation rates of a considerable cay and separated out into layers by density. Later col-
number of NEO’s is known. They range from very low lisions broke some of these up, and distributed pieces of
to 11 rotations per day, with larger numbers at the slower different composition in different orbits. We expect the
rates. About 60% are 5 rotations per day or less. The bulk composition of NEOs to generally vary within the
shape of the brightness variation curve gives an indica- limits set by the original Solar nebula from which they
tion of the overall shape of the NEO. Without a detailed formed. However we cannot tell by their current orbits
color map of the surface, we cannot distinguish variations what that composition is because the orbits have been
due to shape from variations due to color (one part being scrambled too much. We have to look at each individ-
lighter or darker than another part), but the indication is ual object to determine what it is made of.
NEOs vary from round to about 3:1 maximum to mini-
mum dimensions. Morphology:

Composition: Morphology refers to the mechanical condition of a body.

All objects in the Solar System that do not have atmo-
The human eye is sensitive to three wavelengths of light, spheres or surface renewal processes (crustal plates or
from which we get our sense of color and can determine vulcanism) are heavily cratered from random impacts of
a great deal about objects around us. Similarly, scientific other objects. Close-up observations of asteroids, in-
instruments like large telescopes with spectrographs or cluding NEOs, shows they are no different. Therefore
color filters are sensitive to light of different wavelengths. their surface includes a Regolith (from the Latin mean-
A plot of brightness vs wavelength is called a Spectrum. ing “blanket of stones”), a layer of impact debris and dust
Spectra from objects in space can be compared to those consisting of a mix of the original asteroid and whatever
of meteorites and pure minerals. We can make a good crashed into it. Over time, they may also have picked up
guess at the composition of an NEO by this sort of com- dust and small rocks from the space they travel through
parison, given that enough light is collected to make the via gravity or electrostatic forces, provided the material
plot. An object too small and returning too little light for started out at low enough velocity to not just make more
the available telescopes cannot be analyzed in this way, craters. High velocity impacts throw debris above the es-
but it works for many of the larger NEOs, or small ones cape velocity of the object, thus adding to the popula-
that happen to pass close to Earth. tion of smaller asteroids, rocks, and dust. From radar and
thermal observations, the regolith layer is estimated to be
Asteroid and meteorite composition is grouped into on the order of a meter in thickness, with larger rocks
classes, and observed NEO spectra are about 62% S- sticking out, but this is highly variable by object and lo-
complex class, 20% X-complex, 12% C-complex, and cation on a given object. Internal structure of NEOs is so
6% other. The ones we observe are biased by the meth- far poorly known, but they are expected to fall into three
ods we use to observe them. A simple example is very general classes. These classes have strong differences in
dark objects are harder to spot in the first place, or to how they would be mined:
measure spectra from, so the actual composition mix is
different. Obtaining spectra for the whole NEO popu-
• Monolithic (single piece) fragments of larger ob-
lation is ongoing (see Lazzarin et al, SINEO: Spectro-
jects, having strength typical of Earth rock.
graphic Investigation of Near Earth Objects, 2004).
As of 2008 around 2% had measured spectra. Observed • Rock piles of smaller objects held together by grav-
spectra only tell us what the visible surface is like. The ity, with no strength between the pieces, but strong
surface is affected by exposure to solar and cosmic radi- pieces individually
ation, and from impacts by other objects, so it is not a
firm guide to the bulk properties of an NEO throughout • Very porous comet remains with essentially no
it’s volume. strength.
The average life of an NEO is estimated to be 10 mil-
lion years. They either crash onto one of the large bodies Observing Programs
in the inner Solar System, or gravity effects change their
orbit so they no longer fit the NEO category. Since the There are ongoing scientific observation programs di-
Solar System is 450 times older than this, there must be rected at NEOs, so a mining program can build on that
a constant source of new NEOs. This is mainly from the growing base of knowledge. The International Astro-
main Asteroid belt and extinct comets. Extinct comets nomical Union’s Minor Planet Center has extensive data
are ones which have evaporated away all their volatile on known NEO’s and current observing programs, from
components and now are just the rocky remains. Comets which most of the information in the previous section is
and different parts of the main Asteroid belt formed at derived. As of 10 Dec 2012, 859 NEOs larger than 1
different distances from the Sun, thus at different tem- km have been detected, which is estimated to be 90% of
peratures. Therefore they collected different materials. the total population in that size range. Discovery of these
Large enough asteroids became heated by radioactive de- large NEOs is now less than 20/year. A further 100,000
5.9. ORBITAL MINING 199

are expected in the 100m to 1000m size range. Only useful if you wanted to reach a certain percentage cover-
about 8% of those have been found so far (see Large Syn- age by a specific date.
optic Survey Telescope Science: NEO Threat). Even the
smallest and least dense in that range (100m x 1.5 g/cc)
would have a mass of about 800,000 tons, which is large Prospector Missions
compared to the largest objects placed in space (the ISS at
about 450 tons), and a significant mass to mine. Thus the Only a limited amount of composition data can be gath-
total NEO resource population is very large, and we have ered by ground-based telescopes. For more detailed
yet to even locate much of it yet. Beyond the NEO pop- information, getting the instruments closer is essential.
ulation, there are expected to be many millions of other These can be called prospecting missions in the mining
minor planets in the Solar System, but accessing those is sense. There are several types:
harder in energy terms, and is left for a later step.
Meteorites:
At the current NEO discovery rate of 900 per year it
would take about a century to find most of the remain-
These are former NEOs which got so near they conve-
ing population larger than 100 meters. Since their or-
niently crashed into the Earth. This makes them easy
bits are randomly distributed, and so are the positions
to collect relative to objects still in space. They are di-
within their orbits, opportunities for exploration or min-
rect samples of the the asteroid population, but they are
ing missions increase about linearly with the number of
not unmodified samples. Entry through the Earth’s atmo-
known objects. This is currently about 10% per year. If
sphere, impact, and weathering (if they have been on the
the cost is not too great it would be worth increasing the
surface for a long time) have all changed them from their
number and size of telescopes searching for NEOs to in-
pre-impact state. Still, a great deal of useful knowledge
crease the discovery rate. Looking from Earth, we only
can be gained from them because we can apply all the
see the smaller ones if they get close enough to show up
available scientific instruments to examine them. Projects
on telescope instruments. It may be worthwhile to send
to gather meteorites contribute to the general knowledge
dedicated spacecraft to different parts of the NEO orbit
pool for future mining, and are inexpensive relative to
range, as proposed by the B612 Foundation. It would
space missions. Even more helpful is tracking incom-
be especially useful to search the orbits easiest to reach
ing meteorites with telescopes and radar, and then find-
from Earth up close, to more easily find those particular
ing them on the ground, since we can then associate them
ones. Even though only a fraction of all NEOs have been
with a specific source orbit. The mix of material types
found, a good number are easy to reach (for example see
for meteorites is different than NEOs because re-entry
Elvis et al, Ultra-Low Delta-v Objects section 9, 2011).
and weathering affects some types more than others, but
There will be other NEOs whose orbits are very elliptical we can adjust for this once we know enough.
or inclined, which will not be good early mining candi-
dates but will be an Earth impact hazard. For those type, Remote Missions:
a different search strategy is needed. A full sky search
will find both types, but one looking for especially easy To get better data on NEOs in their current locations we
to reach NEOs will be looking in specific parts of the must go to the source. The size and weight of spacecraft
sky near the plane of the Earth’s orbit. There is nothing instruments is severely limited compared to what we have
preventing a telescope from doing both, it’s just a matter available on Earth, therefore we want to get as close as
of what part of it’s observing time is used for each type possible to use their limited sensitivity. For NEOs that
of search. The recently started Large Synoptic Survey would be to fly nearby, go into close orbit, or land on the
Telescope project, which is much larger than past aster- object. The ongoing Dawn mission to the largest Main
oid search telescopes, can efficiently do both mining and Belt asteroids, (4) Vesta and (1) Ceres, is an example of
hazard surveys. this mission type. Dawn goes into orbit around the object,
As noted above, only a few NEOs have been observed and observes it remotely with cameras and other instru-
with large enough telescopes to get their spectra, and thus ments, then sends the data to Earth. Close observation
a start at determining what they are made of. A dedicated of a reasonable sample of typical objects could lead to
observing program with larger telescopes will be needed extrapolation of other object’s characteristics based on
to fill in the spectral data for a large percentage of NEOs, measurements from Earth, particularly their spectrum in
compared to the 1-2 m diameter class telescopes have various wavelengths. A single spacecraft with sufficient
been used so far for discovery. The LSST mentioned pre- fuel and electric thrusters could visit multiple objects, like
viously is sufficiently large for this purpose, as are other Dawn is doing.
existing and planned telescopes, but observing time on
large telescopes is in high demand. The LSST is one of Sample Return Missions:
these large telescopes, and only 15% of it’s time may be
dedicated to asteroid searches. So a dedicated or semi- A more ambitious mission type would collect a sample
dedicated telescope for gathering spectra would be very or samples from one or more NEOs, and return them to
200 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Earth for analysis. This could be either directly via a re- get 14.9 km/s as the additional velocity we can take for an
entry capsule, or indirectly by delivering it to the Space asteroid mission and use the same total fuel as landing on
Station, after which the samples are sent down to Earth. the Moon. We add this to the 4 km/s in both cases to get
Sample returns allow using the full range of Earth instru- 18.9 km/s total mission velocity. From the table we can
ments. One spacecraft, depending on design, could re- see that very few, in fact only 87 out of 8986, or less than
turn multiple samples from a single object, samples from 1% of NEOs require more fuel to reach than the Moon
different objects, or fly multiple missions to different des- given these assumptions.
tinations, with re-fueling at Earth each time. For direct The best asteroid candidates need about 11.5% fuel mass
sample missions the spacecraft either has to land, send
to final mass, while Lunar landings need 70.75%. So un-
an impactor to throw a cloud of material to be collected der the best circumstances, asteroids are 6 times easier to
from orbit, or use some kind of scoop or mining bucket
reach in fuel terms. If we look at the low velocity NEOs
from a distance. Each approach would need a different that require less than 5.3 km/s ideal velocity to reach, the
design. Landing on a small object is not like landing on
fuel required is under 16.2% and the advantage is at least
a large body like the Moon. The gravity levels are so low 4.3 times over a Lunar landing. Additional data on low
that staying in place may require anchoring. Otherwise
velocity NEOs can be found in Elvis et. al. (referenced
just using a robotic arm to scoop up a soil sample might above), although that paper neglects electric propulsion.
lever the rest of the lander off the surface.

Size and Mass The size of most NEOs is poorly deter-

Candidate Selection
mined by telescope observation from Earth because their
image is less than 1 pixel in the attached camera. For
We need to select which asteroids to investigate, and then
now we usually estimate size from brightness. This is still
later mine. Our initial criteria will be based on how easy
somewhat uncertain because brightness is the product of
they are to reach, size, and composition. Mission veloc-
physical size and reflectivity (albedo). Most NEOs are
ity determines how much fuel will be used per trip, so
various shades of dark gray to black, and the darker the
lower is better. Size determines total amount of mass that
color, the less light reaches the telescope. Until we can
can be mined, so larger is better. Composition of NEOs
make more detailed observations, we use a range of esti-
varies, so preference on that basis depends what we need
mated albedo to produce a size estimate. To further make
the materials for.
our mass estimates uncertain, we usually do not know
the detailed shapes or density. The latter can range from
Mission Velocity Starting with the first criterion, The about 1300 to 7800 kg/m^3 depending on how solid the
table on the right lists the number of known NEOs by object is and what it is made of. We shall assume a can-
velocity range from low Earth orbit, based on the JPL didate NEO should be at least 1 million tons in mass to
Table of Low Velocity Asteroids. The table presented be worth exploring in detail and setting up mining. If the
here is as of mid-2012. The JPL source data will change unknown albedo is assumed to be at the higher end of the
over time as new NEOs are discovered. Note that 5.9 range, at a given brightness it will be smaller. If we also
km/s is the velocity required to reach the Lunar surface assume the lower end of the density range we can make
from Low Earth Orbit, purely in velocity terms. So on a minimum mass estimate based on brightness, as shown
that basis many of these are easier to reach than the Moon. in the last column of the table on the right.
Asteroid missions can be done entirely with efficient elec-
tric thrusters. So even ones that require higher actual Current Candidate NEOs
velocity can be done with less fuel than landing on the
Moon. To find out how much less, we assume only the The following list of candidate NEOs is drawn from the
Low Lunar Orbit to Lunar Surface part (1.9 km/s) has to same JPL table noted above. The selection criteria are:
be done with high thrust chemical rockets. They are re-
quired for this portion because once you go below orbital • Electric propulsion delta-v from LEO < 7.5 km/s.
velocity you will impact if you do not land quickly. In The table values are for high thrust delta-v, which is
contrast electric thrusters need a 41% higher total mis- lower by a factor of 1.414 (square root of 2), but the
sion velocity, because they use continuous spiral thrust- ratio is constant, so can be used for selection. Actual
ing paths rather than the short burns of chemical rockets orbital mining missions will likely start from a high
have. Their exhaust velocity is 50 km/s rather than the orbit, so the actual mission velocities will be lower.
4.5 km/s at best for chemical, requiring 11.1 times less The difference will be a constant value representing
fuel per velocity increment. The net advantage after con- the delta-V from LEO to your actual starting point,
sidering thrust profile is 7.85 times less fuel per velocity so the relative order of candidates will stay the same.
increment for electric. The first 4 km/s are assumed to
be the same in both cases, using electric propulsion to go • Absolute magnitude (H) < 22.0. This gives a prob-
from Low Earth Orbit to Low Lunar Orbit or Earth es- able size of 110-240 meters and a probably mass
cape. If we multiply the Lunar landing part by 7.85 we greater than 900,000 tons.
5.10. ORBITAL MINING 2 201

Composition has not been used to narrow the selection thousands of times higher than on Earth, so almost any
because not many of the asteroids have had their spectrakind of mass has economic value. Raw unprocessed rock
taken and none have been visited. Therefore this list only
can be used for radiation shielding, so even that has con-
represents the state of knowledge as of 2012. As addi- siderable value. So we consider all NEOs to be ores, but
tional objects are discovered, more details about known they still have differing compositions which makes them
ones accumulated, and needs for particular materials to suited for different uses. As noted on Page 1, we don't yet
extract are developed, the best candidates will change. have very good data on the bulk composition and distri-
The columns are: bution of materials for most Near Earth Objects, but we
can make a list of general categories based on observa-
• Provisional name - This is year of discovery and a tions of the larger main belt asteroids, meteorites, and the
serial number within the year small fraction of NEOs with observed spectra. The main
categories in the widely used Tholen Asteroid Spectral
• Delta-v in km/s - As noted, the values are for high Types are further subdivided into distinct subtypes. The
thrust missions. main types are:
• Relative Delta-v - These are the previous delta-V
value compared to velocities to reach the Lunar and
C-group These represent 75% of asteroids in general,
Martian surfaces, assuming aerobraking at Mars.
and an undetermined but probably similar fraction of
• H - Absolute visual magnitude. This is the bright- NEOs. These have a similar composition to the Sun
ness of the object at a standard viewing distance of and the original Solar nebula from which the Solar Sys-
1 AU and fully lit by the Sun. The actual bright- tem formed, minus Hydrogen, Helium, and some volatile
ness from Earth varies constantly as their positions compounds. The missing components are gases at the
change. temperature and pressure of NEOs, and so have evapo-
rated away. Unlike the Gas Giant planets, NEOs have
• a - Semi-major axis in AU. This is half the long axis no gravity well to speak of to trap these light compo-
of the elliptical orbit of the object. nents. The spectra of these objects are similar to the
Carbonaceous Chondrite type meteorites, which are as-
• e - Eccentricity. This is the (difference in closest and
sumed to come from the same source bodies. Specific
farthest distance from the Sun)/(sum of closest and
types within the C-group are enriched or depleted of spe-
farthest distance = major axis). It is a measure of the
cific components. Typically they consist of a physical mix
shape of the orbit and ranges from 0 for a circular
of grains of different composition, including a significant
orbit to 1 for a parabolic orbit which just reaches
Iron-Nickel component (3-20% depending on subtype),
solar escape velocity.
mineral silicates and oxides, sulfides, water (up to 22%),
• i - Inclination in degrees. This is the tilt of the orbit and organic (Carbon) compounds from which the group
plane with reference to the Earth’s orbit. gets it’s name. C-group objects are very dark, with an
albedo (reflectivity in sunlight) of 0.03 to 0.10, so they
• Notes - The permanent object number and name are are hard to spot initially. Initial size estimates tend to un-
noted if they have been assigned. They are not im- derestimate their size, since those are based on average
mediately assigned on discovery because multiple albedos, which are assumed to be 0.10.
observers might detect the same object, it may be
human-made (which have their own numbering), or
it may not be in a permanent orbit. 1999 RQ36 is S-type This represents about 17% of asteroids in gen-
the planned target of the Osiris-Rex mission. eral, and get their name after their Stony makeup. They
contain mostly iron and magnesium silicates, but without
Continue to page 2 the carbon in the C type. They originate from the inner
Main Belt asteroids, where they are dominant at <2.2 AU,
common at <3 AU, and rare farther out.
5.10 Orbital Mining 2
Return to page 1 X-Group These include the high metallic fraction as-
teroids, ranging from 50 to 100% metals, with the re-
mainder as stony inclusions. Some of them are nickel-
5.10.1 NEO Ore Types iron alloy, thought to come from the metallic core of
larger asteroids which were later destroyed. The remain-
An ore is a raw material whose composition is of eco- der are thought to have crystallized without separating in
nomic value. On Earth, we generally class as ores those bulk from smaller asteroid bodies or being exposed to
source rocks which are particularly high in one compo- high temperatures. Therefore the metallic parts have dif-
nent. In space, the current cost of almost everything is ferent compositions.
202 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Other Types There are a number of other groups, the The debris belt is currently about ten times the hazard of
A, D, T, Q, R, and V-types, which are present in small natural meteoroids, so cleaning up this trash would greatly
numbers, but do not form a coherent large group by com- reduce the hazards to functioning spacecraft, for which
position. the operators (or their insurance companies) should be
willing to pay.

5.10.2 Example Mining Missions Some non-functional spacecraft only have a single failed
part, or ran out of fuel, and could be made to operate
There are approximately 600,000 known minor planets of again. In that case they can repaired and placed back in
all types throughout the Solar System. Near Earth Ob- service. Others could be salvaged for parts, or their ma-
jects (NEOs) are the subset of about 9,300 known minor terials recycled. Anything not useable would be disposed
planets which approach within 1.3 times the Earth’s dis- of by re-entry. Such a salvage and repair business could
tance from the Sun (1.3 AU). This is substantially closer pay for itself.
than Mars’ orbit (1.52 AU) and the Main Belt asteroids
(mostly 2.1-3.3 AU), and therefore easier to reach from
5.10.4 Small NEO Return Mission
Earth.
Newly discovered minor planets get identified by year Purpose - Return an entire small asteroid to near Earth as
and a serial number within the year, such as 2011 AG5. a demonstration, for scientific examination, and for pro-
When their orbits are better determined, they get a per- totyping processing methods.
manent serial number and possibly a name, such as (1) Description - The Keck Institute for Space Studies per-
Ceres or (4) Vesta. A 2011 study by Landau and Strange formed a study in 2011-2012 in some detail of return-
at the Jet Propulsion Laboratory (Near Earth Asteroids ing an entire small NEO about 7 meters in diameter and
Accessible to Human Exploration AAS 11-446) tab- 250-1300 tons in mass (see Asteroid Retrieval Feasi-
ulates a number of such with particularly easy to reach bility Study or Alternate Source). The NEO would be
orbits. Their estimated masses range from 50 tons to 6 bagged in it’s entirety by an inflatable structure that is then
billion tons (for (175706) 1996 FG3) so in most cases cinched down for transport. Due to the low gravity of
bringing back the the entire object is not practical for such a small object, this method will prevent loss of any
an early mining mission. Instead, we will look at several dust and rocks from its surface. Small objects are hard to
mining mission types with different difficulties and time characterize from the Earth, which accounts for the large
frames when you would consider doing them: range in mass estimate. If a small flyby or orbiter mission
is sent ahead of time, the mass and other characteristics
• Mining the Earth’s Debris Belt can be better determined. This will likely be required be-
• Return an entire small (~500 ton) NEO, fore doing detailed design of the mining tug.
• Gather 1000 tons of unprocessed regolith Assumptions - As for any technical study, the assump-
from the surface of a larger object, tions made affect the results. Major assumptions include:
• Extracting fuel onsite and pushing the en-
tire object towards Earth, and • The starting point is Low Earth Orbit and the des-
tination is high Lunar orbit for the returned NEO,
• Extracting finished materials and only re-
after which it would be examined scientifically and
turning those to Earth.
processed for materials. Lunar gravity assist is used
in both directions in the mission. This mission re-
5.10.3 Debris Belt Mining quires a total velocity of 9.56 km/s for a particular
NEO selected to model the design around.
This would both be a demonstration of mining tech- • It is performed as a complete mission with a sin-
niques, such as capturing a non-cooperating target, and gle launch on an Atlas V 551-class launcher and a
also be useful by removing hazards from Earth orbit. The single spacecraft. This limits the total spacecraft
Debris Belt is the accumulated defunct spacecraft and mass at launch, including having all the propellant
pieces of spacecraft from accidental explosions and col- pre-loaded. It also limits the solar array size to what
lisions. By combining atmosphere mining for fuel, and can be deployed automatically from a single payload
electric propulsion, sufficient velocity capacity would be volume.
available to rendezvous with the scattered objects and ei-
ther return them to an Assembly Platform for recycling, • The propulsion system uses 40 kW net at end of
or to a low enough orbit where drag will quickly cause life (plus 1.2 kW for other electrical needs), and has
them to re-enter. For efficiency, multiple objects in sim- an estimated mass not counting propellant tanks of
ilar orbits would be collected on a single trip, and several 1,370 kg, thus a power/mass ratio of 33.25 kg/kW.
collector ships of different sizes would be used. It would The power system is 60% efficient, and assumes
not be efficient to send a large ship after a small object. the solar arrays lose 20% of their original efficiency
5.10. ORBITAL MINING 2 203

traveling outward from the Earth’s radiation belts. Orbit. The radiation protection uses cover glass over
It uses 5 x 10 kW ion thrusters with an exhaust ve- the solar cells, which absorbs the radiation before it
locity of 30 km/s. The Xenon propellant has a tank can damage the active layers in the cells. The cover
mass fraction of 4.25% above the propellant mass. glass adds mass and slightly lowers efficiency by ab-
sorbing and reflecting some of the sunlight. Lower
Results - Overall spacecraft start mass is estimated at mass and more efficient arrays are used for transit
14.7 to 18.8 tons for returning 250 to 1300 tons, thus pro- from the processing plant to the NEO’s and back.
viding a 17 to 69 NEO return mass/spacecraft mass ratio. The high efficiency arrays are folded and protected
Fuel consumed is from to 8.8 to 12.9 tons, giving an NEO by radiation shielding during delivery to HEO. Un-
mass/fuel mass ratio of 28 to 101. This mission was de- protected arrays can lose 20% of their efficiency
signed as a single return mission, but if sufficient propel- passing through the radiation belt under their own
lant can be extracted from an NEO, ranging from 1-4% power, so using a hardened array for that job makes
of its mass, future missions can be self-sustaining on fuel. a significant difference.
With 40 kW power for propulsion, the total mission time
ranges from 4.0 to 10.2 years depending on asteroid mass. • The Small NEO Return Mission captured an en-
The NEO mass return rate is then 62.5 to 127.5 tons per tire small NEO with minimal disturbance to pre-
year. If higher return rates are needed, larger solar pan- serve science. This mission assumes gathering 1000
els and higher power thrusters would be needed, scaling tons of surface rocks and dust from a larger NEO
approximately linearly with mass return rate. At some for easier processing once returned. Larger NEOs
power level solar panels will get unreasonably large and are easier to detect and determine properties from
you would need multiple spacecraft. Current Space Sta- Earth, thus giving more mission candidates. This
tion array size, if replaced with modern cells, would be will lower the mission velocities slightly by having
around 300 kW, and a practical limit might be 2 MW for more to choose from, and allow a better schedule
a truss with multiple arrays and thrusters. Large arrays for later missions.
would break the single launch with no orbital assembly
assumption. For higher power levels than what is practi- • Higher exhaust velocity is used to save propellant,
cal for solar, a nuclear reactor would be the likely power and higher power is used for faster transit times. For
source. a production mining operation efficiency and annual
mass delivered are important. Propellant is assumed
to be Oxygen since that can be extracted in large
5.10.5 Bulk Regolith Return Mission quantities in all but metallic type NEOs. This makes
the mining operation self-sustaining on fuel. We as-
Purpose - Return a maximum of 1000 tons of bulk NEO sume the Ad Astra VASIMR plasma thruster and
surface material (regolith) per trip to feed a processing solar arrays have a combined mass of 10 kg/kW,
plant in High Earth Orbit (HEO). exhaust velocity of 50 km/s, and Oxygen propel-
lant tank mass fraction of 10%. (source: Ad As-
Description - HEO is from 10 Earth radii (above the ra-
tra Rocket Company A Survey of Missions using
diation belts) to the limits of the Earth-Moon system. Be-
VASIMR for Flexible Space Exploration, 2010).
cause gravity decreases as the inverse square of distance,
all HEO locations have similar velocity requirements and • The first trip may fetch a smaller load, such as 200
are near Earth escape. An HEO location minimizes the tons, to reduce the original fuel mass required from
total round-trip velocity for each asteroid mission. The Earth, and make the second trip sooner to bootstrap
Mining Tug is assumed to make repeat trips using part of fuel production. As much as half this mass is Oxy-
the extracted propellant from the previous trip to refuel gen of varying extraction difficulty, which is more
with. Asteroid orbits close to the Earth’s are low energy, than enough to fuel full loads on later trips.
but also low difference in orbit period. Therefore they
come close to the Earth at long intervals, and are not good
candidates for successive trips. Thus different destination Calculations
asteroids are chosen for each trip based on orbital oppor-
tunities. Major parts of the Mining Tug, like the solar Trajectory:
arrays and thrusters, are repaired or replaced as needed
at the processing plant to keep the Tug functioning. TheFor mission planning in general, we work backwards from
tug is therefore designed for orbital assembly and refuel-
the destination to find out what we need to start with. We
ing, both to reduce initial component launch masses fromassume you can use Lunar Gravity Assist in both direc-
Earth and to enable maintenance/upgrades. tions to help depart and return to Earth orbit. This mis-
Assumptions sion does not need to return to Low Earth Orbit (LEO),
but some final products will. Aerobrake passes through
• A radiation-hardened solar array is used to transit the upper atmosphere can be used for smaller cargo re-
through the Van Allen radiation belts to High Earth turns to LEO and save fuel. There are several detailed
204 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

trajectory simulation programs to calculate exact trajec- moving in their respective orbits, so their distance from
tories to a known destination and back. For now, we don't Earth constantly changes, and thus so does the mission
know which NEO we are going to or when. In fact, it may path. Each thruster uses 9.85 kg of fuel per day. With 5
not even be discovered yet. For now propulsion will be thrusters that becomes 49.25 kg/day at full power. From
calculated based on generic propulsive velocity required the total amount of fuel used we can calculate the engine
of 4 km/s outbound and 1 km/s inbound. run times as 110 and 465 days, and the total trip will be
About 1% of known NEOs have an ideal velocity to reach that plus whatever coasting time is needed due to orbit
from LEO of 4.5 km/s or less (see previous page). It is positions, and time at the NEO to do the mining. Very
roughly we allow 200 days coast time and 100 days min-
assumed that future discoveries will maintain the propor-
tion of low velocity candidates. Of those candidates, a ing time, and thus a total trip time of 1.125 to 2.1 years.
subset will have suitable size, composition, and orbit tim- The original assumption of 1000 tons returned is not a
ing. Starting from HEO ( > 64,000 km radius ), velocity fixed requirement. Within reason that can be larger or
to reach Lunar flyby will be about 1.2 km/s, and Solar or- smaller, as long as the main components of the mining
bit change to reach the NEO will be about 2.8 km/s. We ship scale linearly. Plasma thrusters are not as efficient at
choose an NEO which will make a close pass to the Earth lower power levels, so below about 80 kW it will make
at the time we want to return from it. That way we can sense to use ion thrusters. For reliability, one or two
make a relatively small velocity change to set up an Earth spare engines should be added above the number needed
or Moon gravity assist to help with the return trip. Since for propulsion. Large solar arrays by their nature have
the mass will be much larger on the return leg, optimizing enough duplicated parts to be reliable.
for this part reduces total propellant required. The return
velocity is then estimated at 1 km/s.
5.10.6 Large NEO Return Mission
Propellant:
'Purpose - Return a whole larger NEO to Earth Orbit for
A 200 kW plasma thruster is in development, so if we later mining. This would be second-generation mining
have 5 of them plus a 1 MW solar array, the estimated with substantial markets and infrastructure required.
hardware mass is 10 tons. The plasma thruster has an Description - We will use the example of 2011 AG5, an
exhaust velocity of 50 km/s, so from the rocket equation NEO expected to pass within 300,000 km of the Earth in
we can calculate the mass ratio is 1.02 to produce 1 km/s 2040. The objective is to shift it’s orbit enough to do a
velocity change on the return trip. Mass values will be gravity flyby in 2040, to set up for capture into the Earth-
given for both 200 and 1000 tons returned respectively Moon system later. Once captured, it is then mined for
in parentheses as (200,1000) tons. Given an end mass materials. Given an estimated mass of 4 million tons,
of (210,1010) tons (vehicle hardware plus returned re- electric thrusters using solar arrays do not seem feasible.
golith), the fuel for this part of the mission is (4.24,20.4) Several approaches are possible:
tons. On the outgoing leg of the mission we assume 4
km/s is required. The mass ratio is thus 1.083 against an
• Use a powerful nuclear reactor or solar concentrat-
empty vehicle mass of 10 tons, plus (4.24,20.4) tons fuel
ing thermal-electric generator to supply power to
for the return trip, so (1.19,2.5) tons of fuel are needed
larger plasma thrusters. The fuel comes from the as-
for this part. Thus total fuel needed is (5.43, 22.9) tons.
teroid itself, which requires an extraction plant on-
The overall return ratio is 200 or 1000 tons regolith vs site. Such a large operation might require setting
5.43 or 22.9 tons fuel, or 37 to 43:1. This is a very attrac- up a habitat and crew on the asteroid to operate and
tive ratio as long as we can extract a reasonable fraction control it. For a 500 m/s propulsive change, 40,000
of the regolith mass as useful products. In particular, if tons of fuel would be required. If we allow 5 years
we can extract at least 23 tons of Oxygen, or 11.5% of to make the velocity change, then 8,000 tons a year
the first trip’s returned mass, the later full mining mis- are needed.
sions become self-sustaining on fuel, and the return ratio
over the life of the mining vehicle goes up dramatically. • Put a container around the NEO so material is not
If the vehicle can make 5-10 trips before major hardware lost, add or build a pusher plate/shock absorber unit
replacement, the net mass returned after fuel used will be made from asteroid material, and use one or more
4100 to 9000 tons against a launch mass of 15.5 tons, for small nuclear devices to make the velocity change.
an overall return ratio of 264 to 580. The energy to change the asteroid velocity by 500
The particular numbers above will change according to m/s is 125 kJ/kg, thus the total energy needed is
which NEO you select, and which start and end dates are 500 TeraJoules. This is equivalent to 120 kiloTons
used for a mission. At present the known collection of of TNT in energy, plus an efficiency factor. If the
NEOs is growing by 10% a year, and this is expected pusher plate is 25% efficient in capturing the explo-
to accelerate as larger telescopes come on line, so there sive energy, then 500 kT of devices are needed, di-
are more to choose from over time. NEOs are always vided into however many units are needed to match
5.11. PROCESSING FACTORY 205

the shock absorber and container strength. Use of 5.11.1 Desired Factory Outputs
nuclear devices presents obvious hazards, as does
any asteroid material return when the impact energy The output of the factory can be grouped into the follow-
from objects approaching Earth is 15 times their ing categories by type and complexity. The ones listed
mass in TNT. are expected to be needed in large amounts, and not too
difficult to extract, but a good deal of research and engi-
neering will be needed before you can produce a defini-
tive list of what to make, and in what order.
5.10.7 Product Return Mission

Purpose - Set up a processing plant on the NEO, and only Bulk Supplies
return finished items rather than bulk regolith.
• Shielding and counterweights - This requires min-
Description - The farther from Earth the final user is, the imal processing of mined ores. Shielding is needed
more it makes sense to process the NEO material from a for most locations beyond Low Earth Orbit due to
nearby asteroid, rather than returning to Earth orbit first, natural radiation levels, thermal variations, and if
and then send out again to the final location. For multiple man-made reactors are in use. Counterweights are
destinations, this would imply multiple processing plants, needed for some artificial gravity and space elevator
so they would need to be smaller and more efficient to designs. For these uses, sifting ores for composition
operate than a centralized plant somewhere near Earth/ and density and packing into containers, or com-
pressing/sintering into uniform blocks may be all
that is needed. The leftovers after extracting other
materials, known as tailings or slag in industrial ore
5.10.8 Mining Tug Design extraction, can be used as shielding, as can unpro-
cessed inventory.
(this section is preliminary)
Small asteroids are typically rotating. So the mining con- • Oxygen - The obvious use is for breathing, but
cept is to enter synchronous orbit around the body, and oxidation is useful for other chemical processes,
using some method like sending down a scoop on a cable especially high thrust chemical rockets. Electric
to haul materials up to the tug. This avoids the issues of thrusters can be designed to operate with Oxygen
trying to land on a moving target, and the relatively low as the fuel, greatly reducing the amounts required
thrust-to-mass of a loaded tug, which might have diffi- for a given mission. Oxygen can be produced from
culty getting off even a small asteroid. Another reason to a variety of ores via thermal roasting or electrolysis
avoid landing is losing power from being on the night side methods.
of the asteroid. In theory landing at a pole of the aster-
oid simplifies the moving target problem, but that restricts • Water - Again, water has an obvious use for hu-
your choice of mining locations. mans, and for plants in a habitat. It also serves as
Sufficiently small objects would not have much rotation a convenient way to store O2 and H2 propellants,
velocity, but have the problem of staying attached if you needing just electrolysis to separate into it’s compo-
land. We do not know enough about surface cohesion nents. Water is also a good shielding material. It also
yet to design anchor systems. It may be necessary to run can be used in suitably designed electric thrusters.
cables entirely around the object in order to stay anchored. Some C-type asteroids contain water, and merely re-
quire mild heating to extract it.

Building Materials
5.11 Processing Factory
• Iron and Steel - A basic construction material.
The Orbital Assembly step used pre-made components About 6% of meteorites, and presumably a similar
supplied from Earth. This step adds the capability to con- fraction of NEOs, are metallic. They consist of pri-
vert raw materials gathered in the Orbital Mining step marily an Iron-Nickel alloy, with 5-10% of the lat-
into useful inventory and supplies. These either feed into ter, and a small amount of Cobalt. The remainder
more assembly, or are sold to others, for example fuel is rocky components. Iron-Nickel as-is should be a
and oxygen sold to other projects. The processing factory ductile structural metal. With the addition of a small
will not be able to produce 100% of every item needed. amount of carbon (up to 4%) it should become a rea-
Rather, the goal is to produce as much as possible with the sonable steel or cast iron alloy. The more numerous
least initial equipment so as to reduce the amount that has C-type asteroids contain carbon, so obtaining that
to be launched from Earth. should not be hard.
206 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

• Other Metals - Magnesium and Silicon are com- run roughly like the following example for extracting oxy-
mon elements in the rocky portions of asteroids. gen from NEO regolith:
These are not present as native metal like iron, but
rather as oxides such as the mineral Olivine. There- • Simulate the processes via mathematical and com-
fore to extract the metal requires removing the Oxy- puter models
gen, a process called Reduction. If Oxygen is be-
ing extracted for it’s own sake from rocky material, • Demonstrate extraction with prototype hardware
then some metal reduction will occur as a side effect. and simulated regolith made from similar minerals
Other elements may be present in useful amounts,
but it requires more exploration to determine quan- • Optionally demonstrate with meteorite material on
tities. Earth

• Demonstrate extraction with a prototype in Earth or-

• Glass - For observation and habitat windows, and bit using simulated regolith
industrial processes that need concentrated light in
an enclosed space. Silicon dioxide, or quartz, is an • Demonstrate extraction with real NEO mined mate-
excellent transparent material, and silicon and oxy- rials
gen are abundant in rocky asteroid material. Vir-
tually any regolith can be solar melted into a glassy • With experience from prototype experiments, de-
obsidian, which is black due to the included magne- sign production equipment
sium and iron, but can be reliably created with very
little infrastructure.
Process Selection The many industrial processes
used on Earth are candidates to use in space, along with
• Fibers - Fibers are useful for their high tensile some that can uniquely be done in space. The selection of
strength, for cables or reinforcing. The two likely processes to use needs to take account of the differences
types to be made from NEO sources are fiberglass, of the space environment:
which can be made from melted rock, and carbon
fibers, which can be made from the organic compo-
• Lack of gravity unless you design for it by rotation
nent of C-type asteroids.
• Easy access to vacuum and plentiful full spectrum
sunlight
Habitat Supplies
• Relatively high cost of supplies and materials from
• Soil and nutrients - Although hydroponics can be Earth
used to grow things without soil, soil can serve a
dual use as a radiation shield as both mineral grains • Relative difficulty to dispose of waste heat
and water in the soil are effective shielding materi-
als. Research and testing is required to see if NEO-
Suitable Industrial Processes
derived soils are will make a good growing medium
and don't contain hazardous materials. Some exper-
Many large scale industrial and mining processes on
iments were done with Lunar soils from the Apollo
Earth, such as ball milling and gravity separation, assume
program. Research is also required for what ad-
the presence of gravity, and would require large motors
ditives in the form of fertilizers and nutrients, and
to produce equivalent centrifugal effects in space. Froth
what seeding of soil with micro-organisms would be
flotation, very common in mineral ore beneficiation, re-
needed for a productive soil. After all that, soil-
quires both gravity and large amounts of volatile water.
based plants need to be compared to hydroponics
and aeroponics to get the overall best answer for food In space, solar thermal processing is likely the simplest
production and recycling of organics. starting point for material processing, because solar re-
flectors require no moving parts. Clear candidates in-
clude roasting volatiles such as oxygen and water from
5.11.2 Factory Design and Development ores. Many high-end industrial processes on Earth use
vacuum furnaces based on electromagnetic induction or
Process Development electric arc, heating mechanisms which a solar furnace
may be able to augment or directly replace. Due to the
There is a great deal of experience with industrial pro- high temperatures achievable in the vacuum of space, ore
cesses on Earth, but not much for in space or at other beneficiation could be performed by vacuum distillation,
locations. Before planning major projects, then, a ra- successively extracting metal oxides with increasing boil-
tional development program should be done to gain the ing points or differing relative volatility, such as silicon
needed experience. A progression of experiments would dioxide (boils at 3220K).
5.11. PROCESSING FACTORY 207

Direct electrolytic processing is also promising, using material from various NEOs to get the required compo-
electrons from photovoltaic panels to directly drive chem- sition mix.
ical reactions. The electrolysis of water into hydrogen
and oxygen is the simplest example, but many others are
possible. On Earth, the most common industrial applica- Growth Plan
tion of electrolytic reduction is in Hall–Héroult cells for
the production of aluminum metal from aluminum oxide, A fundamental design feature of a space-based factory is
which liberates oxygen gas. Aluminium is only a few per- it does not need to have a single fixed design like a car
cent of most asteroids, but direct electrolytic reduction or a computer monitor on Earth. Rather, it can evolve
of iron, silicon, and magnesium has been demonstrated, and grow by several methods. One is to add new process-
and electrolytic reduction of magnesium chloride domi- ing flows with new types of equipment. This expands the
nates terrestrial industrial production. Electrolysis is typ- range of outputs it can produce. Starting with fewer types
ically performed in either a conductive liquid such as a of equipment is partly for cost reasons - it would be too
molten salt; or in a solid oxide (solid oxide electrolysis expensive to launch every kind of equipment at the start,
or SOE), which can be a variety of electroceramics, fre- and partly because all the various processing methods will
quently yttria-stabilized zirconia (YSZ). not be adapted for space right away. In the longer term,
new technology will come along which was not avail-
For separating elements from ore, a variety of schemes able at the start. Another growth path is increasing scale
have been proposed. Electrophoresis separates consitu- by adding more copies or larger versions of the existing
tent particles by size, and is commonly used in DNA equipment types. With a larger range of outputs and the
analysis. Electromagnetic isotope separation in a calutron ability to assemble items developed previously, the fac-
separates charged particles by mass, by ionizing the atoms tory complex will be able to grow itself mostly from in-
in a near vacuum, and using a magnetic field to exploit ternal production rather than deliveries from Earth.
the mass dependence of a charged particle’s curved flight
path. Very much like an ion drive, this process only pro- Likely early components for the factory will include:
duces milliamp ion currents due to the space charge ef-
fect, and requires high voltages and near-vacuum. • Various machine tools for making parts
Iron-Based Path • Smelter for extracting metals
Native iron is available both on the Moon as impact de- • Thermal refinery for extracting liquids and gases
bris from asteroids, and in metallic asteroids. Thus an
iron-based starter factory is worth considering. On Earth, • Some robots for remote control, so it can operate
iron/steel is obviously a large part of the existing econ- before the people get there and prepare things.
omy, especially in making manufactured items. Consider
using metal scrap as a material in a starter factory on
Factory Locations
Earth. That will translate to space directly, where, for ex-
ample, using trees as a starter material for wood construc-
Objects in space are inherently mobile. The industrial
tion would not. Iron is also available in oxidized minerals,
capacity does not need to stay as a single factory in a sin-
but that requires more chemical processing than already
gle location. As the processing capabilities and the set of
reduced native metals.
users develop, you can expand from the initial factory lo-
An iron-based path would start by concentrating on items cation, likely in Earth orbit, to multiple locations with op-
made from iron/steel. This would include structures, timized capabilities for each location. Eventually you end
tanks and piping for chemical processing, reflectors for up with a network of mining stations, ore processing sta-
solar power (with an aluminum coating), electric gener- tions, assembly stations, and supply depots/habitats, with
ators to use the concentrated light, and, of course, ma- some locations doing multiple tasks. A number of ques-
chine tools for making more items with. Once the initial tions need to be answered about where to put the indus-
production is established, other types of products would trial elements:
gradually be added.
Consumables Path • Should it start with a single location or multiple loca-
tions? It can split over time if the facility duplicates
This path starts with extracting consumable supplies such itself or builds specialized units for different loca-
as Oxygen and water. tions.
Full Use Path
• A location close to Earth minimizes cost of deliv-
This path starts with the assumption that you want to use ering the fraction of supplies and equipment that
as much of the NEO starter material as possible, so you needs to come from there. It also allows for real-
include whatever equipment is needed to do that. Since time control from the ground, either by computers
NEO composition varies by object, you would either need or humans. On the other hand, a higher orbit has
to know which object you are going to use, or bring back more exposure to sunlight, and is less delta velocity
208 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

to deliver raw materials from NEOs. So where is the Satellite Salvage Send larger tugs with robotic capabil-
optimum location(s) for a given process? ity to salvage satellites which have run out of fuel or oth-
erwise can be repaired or bring back to have useful parts
• You can have a strategy of starting in low orbit at the extracted. This will develop orbital maintenance skills,
start, when more material is coming from Earth, and and allow sale of repaired items.
then move to a higher orbit once a higher percent-
age of NEO material is used. Eventually you might
High Orbit Station When sufficient fuel has been col-
move the processing factory all the way to the NEO,
lected, move the processing plant or (part of it) to high
and only ship finished items back.
orbit location. For our example wee choose a 13.66 day
elliptical orbit for ease of access to the Moon and NEO
orbits. This is half the Moon’s orbital period. From
Power Sources the orbit period formula, this gives a semi-major axis of
241,400 km. We place the high point near the average
With the exception of living plants, most processes re- distance of the Moon, but 90 degrees offset in position,
quire a power source. In near-Earth space sunlight is so the Moon will not change the orbit significantly. The
abundant and higher intensity than on Earth due to lack orbit then becomes approximately 132,800 x 350,000 km
of atmospheric absorption and weather, and is available a in radius. This orbit is chosen to for access to other lo-
higher percentage of the time. Photovoltaic panels have cations. To depart for Near Earth Asteroids, vehicles
been the primary method to generate power because they will add velocity so that they encounter the Moon and
are relatively light, have no moving parts, and can be get a gravity assist. To deliver items to lower orbit, they
made in whatever reasonable size you want. For very high subtract velocity until they can aerobrake. To reach the
power levels the panel area becomes large and thus a de- Moon itself, They do a mild gravity assist to raise the orbit
sign challenge. If you have a source of building materials, to near Lunar, then use propulsion to enter Lunar orbit.
and the factory is not moving about much, then heavier
generator types can be considered that use concentrated A radiation-hardened solar array will be needed for the
sunlight. That includes concentrated photovoltaic, where orbital tug to get through the radiation belt. Lightweight
the panels come from Earth, and the reflectors to concen- arrays can be used beyond that region. They are deliv-
trate the light are made locally. ered folded and protected from radiation to maximize
their later performance. The High Orbit Station will serve
multiple purposes: laboratory/research park, processing
plant, and assembly and repair.
Thermal Sources

Many industrial processes require high temperatures. So-

lar furnaces can reach temperatures only limited by the 5.12 Space Elevator (Skyhook)
surface temperature of the Sun, which is sufficient for
many processes. For the few that require higher temper- Space Elevators have been a theoretical transportation
atures, electric driven devices can be used. It is also fairly method since 1895. The original idea is impractical to
easy to reach cold temperatures, by blocking sunlight and build. This step adds a much more practical design as
exposing items to the cosmic background at 2.7K as a a transport hub for getting from one orbit to another
heat sink. quickly and efficiently. Initial construction can use ma-
terials from Earth, but in larger sizes or locations beyond
Earth orbit using local materials is assumed.
5.11.3 Processing Factory Example
Skyhook Concept
In this example we start with Earth Orbit mining, and ex-
pand outward in steps. The popular concept of a space elevator is based on the
original design proposed by Tsiolkovsky in the late 19th
century. It involves a single tower/cable extending all the
Air Collection We use orbital scoop mining deployed way past Geosynchronous (24 hour) Earth Orbit (GEO).
from our previously built assembly station to collect air If the center of mass is at GEO and matches the Earth’s
from the upper atmosphere and use it for fuel and breath- daily rotation it will appear to hang motionless relative to
ing supplies. the ground. Getting to space in theory then becomes an
elevator ride. There are several problems with this sim-
plistic design:
Debris Collection Using fuel obtained in previous step,
dispatch small electric tugs to collect orbital debris for raw • The depth of the Earth’s gravity well (6378 g-km)
materials, then feed it to the processing unit. exceeds the scale length of the best available mate-
5.12. SPACE ELEVATOR (SKYHOOK) 209

rials (350 km for carbon fiber) by 18 times, which to velocity, it makes sense to split the job between
then requires a structure-to-payload mass ratio of 65 both, because the sum of two exponents, for exam-
million to 1. This would require more carbon fiber ple e^2 + e^2 = 14.8 is less than a single exponent of
than the world makes to lift a reasonable cargo mass, the combined powers, i.e e^4 = 54.6. The optimum
and would never be economical because it would division of work between the Skyhook and vehicle
take too long to transport sufficient payload to jus- coming from Earth will depend on technical details
tify the massive cost of that much structure. and costs, but a simple division of half to each results
in a theoretical Skyhook mass ratio of 60:1. A real
• It is of no use for delivering cargo to low orbits. Re- design will be heavier, but 60:1 is a feasible start-
lease points somewhat below GEO result in elliptical ing point, where 65 million is not. Reaching half of
orbits with a low perigee, but lower circular orbits orbit velocity for a single stage rocket with a life of
cannot be reached. It also is of no use transporting many flights is quite feasible.
cargo from the ground when partially built.
• Assuming the tip is at 1 gravity, a rotating Skyhook
• Even with a magnetically levitated elevator car run- with a tip velocity of 30-50% of orbit velocity has
ning at 300 km/hr, it will still take 5 days to deliver a radius of 500-1400 km. The center point needs
one payload to GEO, and you can only deliver one to be that altitude plus enough that the tip does not
payload at a time. dip into thick atmosphere and create drag (100-200
km). Releasing from the center of the Skyhook at
• A single cable catastrophically fails when hit by nat- 600-1600 km altitude allows access to low Earth or-
ural or man-made debris. A cable over 35,000 km bits.
long has a lot of area exposed to such hits.
• A partially built Skyhook can still function because
The Skyhook concept addresses all these problems. In- the remainder of the velocity is provided by the
stead of a static cable that stays over a fixed location, it can Earth vehicle. During construction the velocity split
be either a rotating cable in a moving orbit, much like two is more towards the Earth vehicle. This reduces pay-
spokes of an imaginary wheel rolling around the Equator, load mass, but it can still deliver some. In particular,
or a non-rotating cable in a moving orbit that maintains a if part of your payload is more Skyhook structure,
vertical orientation relative to the parent body. For more that payload pays for itself in increased payload on
information on this go to Space exploration technologies. later trips. This is a literal version of “lifting yourself
by your bootstraps”.
• The non-rotating orbiting Skyhook is a much shorter
version of the planetary surface to geostationary or- • The same fast elevator car at 300 km/hr can reach
bit Space Elevator that does not reach down to the the center in 2 to 5.3 hours. If your destination is
surface of the parent body, is much lighter in mass, high orbit or Earth escape, you do not have to ride
can be affordably built with existing materials and the elevator at all. You wait half a rotation of the
technology, and in its mature form, is cost compet- Skyhook and let go, at which point you are going
itive with what is thought to be realistically achiev- at orbit velocity plus tip velocity. To imagine this,
able using a Space Elevator. It works by starting think of the top point of a bicycle wheel. It moves
from a relatively low altitude orbit and hanging a faster than the center relative to the ground. A half
cable down to just above the Earth’s atmosphere. rotation takes only 10 to 20 minutes.
Since the lower end of the cable is moving at less
than orbital velocity for its altitude, a launch vehi- • Space debris cannot be eliminated. Even if all the
cle flying to the bottom of the Skyhook can carry a man-made junk in Earth orbit is eliminated, the
larger payload then it could carry on its own. When natural flux of meteors will continue. Therefore
the cable is long enough, Single Stage to Skyhook the Skyhook design has to take that into account.
flight with a reusable launch vehicle becomes possi- The most practical way to do that is to use mul-
ble at a price that is affordable to just about anyone. tiple redundant cables to distribute the load such
that cutting one or two is not catastrophic. The
• A full orbit velocity rotating Skyhook reduces the cables should be spaced far enough apart that any
structural requirement to about 2868 g-km, because single object will only hit one or two strands. The
only the tip sees the full Earth’s gravity. The center strands should also be cross-connected periodically
is in orbit and thus has zero acceleration load. This to distribute the load around a break. Repairing a
immediately reduces the theoretical mass ratio from break then becomes replacing a short segment of
65 million to 3,620:1. There is still an exponential one strand. Since you have the ability to install seg-
relation of mass to tip velocity. Since conventional ments during original construction, you are able to
rockets also have an exponential relation of mass replace segments as a maintenance job.
210 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Skyhook Applications theoretical gravity well to carbon fiber scale length ratio
is 18, so it is much easier to build a Lunar Skyhook on a
The Moon and Mars have smaller gravity wells than the relative basis. Another way to say this is the Lunar orbit
Earth, by ratio of 22 and 5 respectively, so Skyhooks with velocity of 1680 m/s is less than the 2400 m/s tip velocity
the same materials can do more of the transportation job assumed below for the Earth Skyhook. That provides all
relative to Earth. But in this step-by-step combined sys- of the velocity between the Lunar surface and orbit, while
tem example, getting off the Earth comes first. We will the Earth orbit version only provides 1/3. Since escape
discuss the other locations here, but the actual construc- velocity is 1.414 times circular orbit velocity, and a full
tion will be delayed until easy transport to those locations Lunar Skyhook is capable of releasing cargo at 2.0 times
is needed. orbital velocity, it can handle cargo well beyond escape
velocity. By climbing to a chosen radius from the center,
and timing when you let go, you can get a wide variety of
Lunar Skyhooks There are two systems that are fea- orbits.
sible because of the small gravity well of the Moon. The Besides using it as transport to and from the Moon, there
first is a catapult system to launch bulk materials off the is also an opportunity for work crews to use the Skyhook
Moon. The catapult uses a rotating cable driven by an at 1.0 gravity as a rest location, because we don't know
electric motor to throw payloads directly into Lunar or- the long term effects of 0.16 gravity on the human body.
bit, where they are picked up by a collector system. The A lunar surface alternative is to use centrifuges to get 1.0
second is an orbiting Skyhook which can deposit and pick gravity or whatever level is needed. A full Lunar Skyhook
up cargoes at zero velocity close to the surface. would have a radius of 283 km and a rotation period of
Catapults - Basalt fibers are similar to fiberglass in that 17 min 40 sec at 1.0 gravity. Since Lunar orbit period
they are an extruded mineral. They have a strength of is 108 minutes or longer depending on altitude, the Sky-
4800 MPa, or 80% of carbon fiber, and a density of 2.7 hook will make 6 or more rotations per orbit. If the orbit
g/cc, or 50% higher than than carbon. Thus the scale is equatorial, that allows it to service multiple locations
length of Basalt fiber is 178 km, or about half that of car- around the Lunar equator, and transport cargo between
bon fiber. The dark areas (Maria or seas) of the Moon are those points at orbital speed at no cost.
covered in basalt lava [1] , so there is an a very large sup- It is an open question if an equatorial orbit is best. A po-
ply of raw materials. If a Lunar catapult delivers basalt lar orbit would let the Skyhook reach any point on the
to an orbital processing factory, or spools of fiber already Lunar surface, but generally only twice a month. The
spun on the Moon, it would be possible to build a Lu- Moon rotates very slowly, so the benefit of the rotation to-
nar Skyhook out of local materials. The choice of Lu- wards the orbit velocity is only 4.6 m/s, 1% of the Earth’s
nar basalt for a Skyhook would have to be compared to contribution. A polar orbit can be arranged as a Sun-
the higher performance carbon fiber brought from Earth synchronous orbit, where the orbit plane always is in
or made from NEO carbon. Certainly for Lunar surface sunlight, while an equatorial orbit is in shadow about 40%
construction it would have the advantage of being very of the time. Thus the solar arrays that power the Sky-
local. hook are more effective in the polar orbit. You can have
Catapults could also be built on the Earth or Mars, but for both Skyhooks in orbit around the Moon, as long as you
Earth it would need to be placed above the atmosphere arrange their orbits to never intersect, such as by using
on a tall tower to get significant velocity. It is proba- different altitudes. In that case you might want to make
bly not the best method when compared to the alterna- the g-forces at the tips higher so the radius is smaller, and
tives. Mars is much smaller, has less atmosphere, and move humans quickly up the cable to a more comfortable
very tall volcanoes that a catapult can be placed on. So g-level.
it is worth considering placing a catapult there to deliver In any Lunar Skyhook, a lander vehicle will need some
materials to orbit. Any catapult (Lunar or other) will need propulsion because the Moon is not a perfect sphere. So
a significant power supply for the motor. In order to not the tips need to stay high enough to miss any high points
waste the rotation energy when stopping to load the next of the terrain, and some maneuvering is needed for an
cargo, it makes sense to consider two catapults, and use accurate landing. If you have two Skyhooks at different
one as a generator to supply power to the other. Motor- altitudes, the vehicle will need more fuel to land and take
generator efficiencies can be above 90%, so most of the off.
energy could be recycled.
Skyhook - The Moon’s gravity well is equal to 287 km at
1.0 Earth gravity. Thus even for the lower strength basalt Asteroid Skyhooks The largest asteroid, 1 Ceres, is
fibers, the gravity well is only 1.6 times the scale length of 487 km in radius at the equator, with a day length (ro-
178 km. For the higher strength carbon fibers with 361 tation period) of 9.074 hours. Therefore the equator is
km scale length the ratio is 0.80. Note that scale length is moving at 94 m/s. Orbit velocity is estimated at 360 m/s.
based on breaking strength, actual designs will use lower The exact number will be found when the Dawn space-
loads and have overhead above a bare cable. For Earth the craft arrives at Ceres in 2015 (it is in orbit around the
5.12. SPACE ELEVATOR (SKYHOOK) 211

2nd largest asteroid, Vesta, as of 2012). A Skyhook thus Design Inputs Tip Velocity = 2400 m/s
needs only the difference of 266 m/s in order to land and Tip Acceleration = 10 m/s^2 - Earth surface gravity is
pick up cargo and then toss them at more than escape ve- 9.80665 m/s^2. We use 10 for simplicity. That provides
locity. The radius in this case at 1 g works out to 7.25 km. normal gravity for any humans on the Skyhook.
This is small enough that it could be built near the Earth,
and then transported whole to Ceres. Setting it up in orbit
would allow mining of the largest asteroid with easy ac-
cess. A synchronous space elevator would be longer and Derived Values Skyhook Radius = 576 km - This is
not provide a 1 gravity environment, but could be used found by solving the centrifugal acceleration formula ( a
to launch cargo into transfer orbits away from Ceres or = v^2/r ) for the radius.
capture incoming cargo. Rotation Period = 25 minutes - We know the circum-
For small asteroids, a Skyhook isn’t necessary for surface ference of a circle by 2 x pi x r, or 3619 km in this case.
access. Even low efficiency chemical rockets do not use Dividing by tip velocity gives the time. For convenience
much fuel to land, and you can just mechanically throw to reach the Skyhook from a launch site, the numbers can
stuff into orbit or escape. be adjusted so the period is an even fraction of the orbit
time, i.e. 100 minute orbit with 25 minute rotation time.
That way the landing platform will be in the same place
each time relative to the launch site.
Mars Skyhooks Pavonis Mons is one of the large
mountains on Mars. Since it is located on the Equator, Orbit Altitude = 750 km - If the tips of the Skyhook
it is an ideal location for some kind of transport system. reach deeply into the Earth’s atmosphere, that will cause
Candidates include a centrifuge launcher like with the drag and heating, and eventually cause the Skyhook to fall
Moon, or a linear accelerator. The higher mass of Mars down. By placing the tips at least 175 km altitude, then
makes it more difficult than for the Moon, but a ground- the center must be that plus the radius high. The exact
based transport system can still do most or all of the job height will be a trade off between less drag, and ease to
of reaching orbit velocity. Similarly, Mars orbit velocity reach from the ground.
of 3.6 km/s is within reach of a Skyhook, and there are Orbit Velocity = 7474 m/s - Found from the formula be-
two convenient former asteroids (Phobos and Deimos) low where G is the Gravitational constant, M is the mass
as a source of building materials. A Martian Skyhook of the Earth, and r is the orbit radius, which is the Earth’s
would likely be placed in low Mars orbit, with the ability radius plus the orbit altitude:
to transfer down to the surface and up to Phobos, Deimos, √
or escape orbits. vo ≈ GM r

Launch Vehicle Payload = 13% - A good chemical

Split Systems A catapult can be used in combination rocket would have an exhaust velocity of 4.5 km/s and
with a Skyhook to enable higher velocity missions with empty weight of 10%. Without a Skyhook, the total ve-
lower total mass ratios. Bodies as small as the Moon do locity required is about 9 km/s, which results in a payload
not require very large mass ratios o reach orbit, so doing fraction of 3.5%. Subtracting the 2.4 km/s provided by
a split system will not gain much at the cost of the extra the Skyhook results in a payload of 13%, or 3.7 times
complexity. Conversely the Earth has a fairly dense at- higher. The exact numbers will vary depending on the
mosphere, so a high velocity centrifuge would see a lot launch vehicle design, but that gives an idea of the pay-
of drag unless placed on a very tall structure. The best load improvement the Skyhook can provide, and thus part
location for a split system turns out to be Mars, partic- of the reason to build it. The greatest advantage of a
ularly with its tall mountains that are in near vacuum at Skyhook is not from the increased payload it provides,
their peaks. By dividing the velocity between two sys- but using some of the increase to increase the fatigue life
tems, it becomes 1.8 km/s each, which can be reached of the vehicle, which is highly non-linear - typically ten
with existing materials and conservative mass ratios. times higher for a 10% addition in structure. Airplanes
and rockets cost about the same per kg to build. This
is not surprising since both are built by aerospace com-
Design Parameters panies out of the same materials. The vast difference
in transport cost is due to airplanes flying about 20,000
The Earth orbit Skyhook does not have a fixed design as times during their service life, and rockets usually only
noted above, but rather grows over time. We also do not flying once. By taking some of the payload increase from
as yet know what an optimum size will be for given cir- a Skyhook and applying it to giving the launch vehicle a
cumstances. A concrete example, however, lets us ex- long operating life, the operating cost will be vastly re-
amine the feasibility and understand what is needed for duced.
the various parts. We will assume a tip velocity of 2400 Payback Time = 1 to 76 launches (average of 43) - If we
m/s for this example, or roughly 1/3 of orbit velocity, and remove the last 100 m/s of tip velocity, the launch vehicle
derive the other characteristics. payload falls to 12.56% from 13.07%. So the incremental
212 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

benefit of the last 100 m/s is 0.51% of the vehicle mass. load mass increase stays the same. On the other hand,
Assume we use Torayca T1000G carbon fiber as our main if traffic rates go up, do does the cost benefit. No mat-
cable material. It has a tensile strength of 6370 MPa and ter what numbers are used, the exponential growth of the
a density of 1.8 g/cc. We allow 40% overhead mass above Skyhook mass with tip velocity will eventually limit it’s
the bare fiber for a finished cable system, and a 2.0 factor size for economic reasons. That is when the incremental
of safety. Thus the working strength is reduced to 2275 cost of making the Skyhook larger becomes more than
MPa, and at a tip acceleration of 10 m/s, the working the incremental payload increase is worth. We emphasize
length becomes 126.4 km. that limit can change over time, however, as new materi-
als become available, different methods of cable delivery
The 2400 m/s Skyhook has a radius of 576 km, and the
acceleration varies from 0 at the center to 10 m/s at the are used, and traffic rates change. The Skyhook/launch
vehicle interaction is a good example of why combined
tip, so effective length is half, or 288 km. The cable
mass ratio is then e ^ (288/126.4) = 9.762:1. Subtract- systems have to be looked at in their entirety, and not as
single technologies or methods.
ing the payload mass gives a theoretical cable mass of
8.762. Since this calculation is only for one arm of the
Skyhook, we double it to 17.524. The payload of the Design Components
Skyhook is the launch vehicle payload + the empty ve-
hicle structure without fuel (10% of launch weight), for Structure The main structural component of the Sky-
a total of 23.07% of launch weight. The Skyhook cable hook will be the tension strands. In addition there will be
mass is then 404.3% of the launch vehicle weight. Doing secondary structure holding the strands in position, and
the same calculations for the 2300 m/s Skyhook, we have for the landing platform, propulsion system, habitats, and
a radius of 529 km, effective length of 264.5 km, mass other items attached to the structure. Cables are not a sta-
ratio of 8.106, and cable mass of 16.211 times the vehicle ble structure unless they are in tension, so for the central
arrival mass of 22.56% = 365.7%. The incremental cable portion of the Skyhook before it is set rotating, we as-
mass is then 404.3 - 365.7 = 38.6%. Since we gain 0.51% sume a core rigid truss structure. The initial total radius is
in payload from this incremental addition to the Skyhook, 2500 m, counting core plus cables. This allows 1.0 grav-
it pays for itself in payload mass in 76 launches if we use ity acceleration at the tips with a rotation period of 100
the launch vehicle to deliver the added cable. If we get seconds. The latter number is chosen so that humans are
the extra cable from another source, such as our hyper- not dis-oriented by rapid rotation. Initial habitats would
velocity gun, or from NEO carbon, the payback could be be placed at the 2500 m radius. Strands are installed both
much faster. lengthwise and in parallel to expand the Skyhook.
Doing the same type of calculation over the whole Sky- Damage Tolerant Design - Man-made satellites and or-
hook, we have a gain in payload from 3.5 to 13%, or 9.5% bital debris are the largest impact hazard for the structure.
of liftoff mass. The Skyhook cable mass is 404.3% of For now we assume no cleanup of Earth orbit, though that
liftoff mass, so the payback time in increased payload is is desirable. Doing so would reduce the risks by about
43 launches. The first 300 m/s increment of the Skyhook ten times. If we assume there is a 0.5 mm protective
increases payload from 3.53% to 4.46%, a gain of 0.93% sheath around the strand core, then objects smaller than
of liftoff mass. It would have a radius of 9 km, and a ca- half that thickness will merely make a crater and not pen-
ble mass of 7.25% times arrival mass, or 0.98% of launch etrate the core. If we assume a large Skyhook supports
mass. Thus the payback time is 1 launch, and the cable fits 1000 tons of payload at 1.0 gravity, with an 8:1 mass ra-
in roughly 1/4 of a payload on it’s delivery flight. So the tio, we have a total load of 10 MN. Given the 2275 MPa
first part of the Skyhook has an immediate payback and working strength, that requires a total area of 44 cm^2
is very desirable. Note that a Skyhook massing less than of cable. Assume each strand is 2 cm in diameter. Then
the arrival mass would not have enough orbital energy to we need 14 load carrying strands, and some number of
give the vehicle on arrival. It would need to be attached extra strands for damage tolerance, which we will use 7
to a larger “ballast” mass such as an assembly platform, for now, giving 21 total strands.
bulk mined materials, or collected space debris.
Any debris object larger than 1/3 the strand diameter will
Mass payback is not the same as cost payback, but if we likely cause enough damage that it fails. Based on a 1995
assume they are for now, at a relatively low rate of one orbital debris assessment [2] there were about 1 million
launch/month, payback takes about 3.6 years, which is objects that size and larger. That produces an impact flux
reasonable from an economic standpoint. Since the ca- of about 10 ^ −4 per square meter per year. If we want
ble mass grows exponentially with tip velocity, the earlier a 1% chance of strand failure per year, then we are al-
parts will pay back faster, and growth beyond this point lowed 100 square meters of area. Since the diameter is
will take longer at a fixed launch rate. A real payback assumed to be 2 cm, then the allowed length is 5000 me-
analysis will have to take into account real cost instead ters. We place cross-connecting rings at that interval to
of mass ratios, and actual launch rates. If less expen- distribute load around a failed strand. At each ring, there
sive launch systems from Earth are developed, then the are 7 points where 3 strands each fan out. Since by de-
cost benefit of a Skyhook goes down, even if the pay- sign two are required to handle the load, the failure of the
5.12. SPACE ELEVATOR (SKYHOOK) 213

third by debris impact does not cause any reduction in the an open question, but since landing at 1 gravity has been
Skyhook’s total load carrying ability. solved multiple times on Earth, it should be solvable for
We do not want a single object to impact too many strands this task.
at one time. In the worst case, the largest object in orbit After delivering it’s payload, the vehicle just needs to let
besides the Skyhook, currently the Space Station, should go or be pushed off the platform at the right time. At
not hit more than our reserve of 7 extra strands. In real- it’s low point, the platform is sub-orbital, so the vehicle
ity, 99% of damaging debris is smaller than 30 cm, and will automatically re-enter. The vehicle will be moving
the Space Station is under active control, so it should not at 4,600 m/s relative to the Equator, compared to 7,400
ever hit the Skyhook, but we are looking at a worst case. m/s for a rocket without a Skyhook trying to re-enter.
Since the Station is about 120m wide, if the strands are Therefore the vehicle has to dissipate 39% as much ki-
arranged in a circle, and there are 21 total, a 120m ob- netic energy, which makes the heat shield design much
ject should not intersect more than 120 degrees of the easier.
perimeter. Doing a little geometry yields 120m = 50% Low Gravity Platforms - Low- or Zero-gravity is desir-
of the radius, thus the diameter of the circle is about 480 able for some tasks in Space. You can place platforms
meters, and the 7 attachment points are spaced about 200 or pressurized habitats at chosen distances from the Sky-
meters apart. A truss spans between each attach point, hook center and get any value between 0 and 1 gravity.
making a 7-sided ring. If a given strand is damaged, then For true zero-g, you would need to de-spin the structure
it is simply replaced by the same construction method the so it does not rotate along with the rest of the Skyhook.
Skyhook was built in the first place. As long as strands Likely this will be a structure extending along the axis
can be replaced at least as fast as they are damaged, the of rotation like the axle of a wheel, where the Skyhook
Skyhook can be maintained indefinitely. cables would be spokes of the wheel.
The above calculations are an example. For a real design,
you would find the optimum strand diameter and count,
rather than just assume 2 cm and 21. The real debris pop- Electric Propulsion Due to fundamental conservation
of energy, transporting more payloads up than down
ulation is not all in orbits that could intersect the Skyhook.
For example, the Space Station is about 400 km altitude, causes the Skyhook to lose orbital energy and eventually
and could only intersect with the bottom 225 km of the re-enter if not corrected. We use the highly efficient elec-
Skyhook when it is in the vertical position. Some efforts tric thrusters developed in an earlier step to maintain or-
may be made to clean up orbital debris. But even if not, bit. In effect, the electric thrusters substitute for the lower
we can make a reasonable design that can withstand worst efficiency chemical rocket engines on the launch vehicle.
case damage and reduce expected damage to a 1% per Electric type engines have too low a thrust to reach orbit
year maintenance job. by themselves, but by attaching them to a Skyhook, we
can add orbital energy gradually, and then give that to the
Landing Platform - This functions somewhat like the
payload in a short time. The Skyhook becomes a very ef-
deck of an aircraft carrier, in that it is a mobile plat-
ficient battery for storing orbital energy, about 25 times
form which vehicles land and take off from. We assumed
the energy/mass of Lithium batteries on Earth.
above that the Skyhook structure supports 1000 tons of
load. This includes everything besides the main structure, Power Rquirements - For each kg of payload we are
including the landing platform. Arriving vehicles would placing in orbit, we are changing the velocity from 5074
have a smaller mass that is added to the load temporarily. to 7474 m/s. This requires adding 15 MJ of energy. Since
Unlike zero-g docking, which is done slowly, the landing there are 31.5 million seconds in a year, then that means
platform is rotating at 1 gravity, so landings will be sim- for each kg/year of payload, we need 0.477 watts of deliv-
ilar speed to landing on Earth. The size of the platform ered orbital energy. In Earth orbit we are not in sunlight
will be governed by the accuracy of the vehicle naviga- 100% of the time, so solar panels would need to be larger
tion. The design can be a horizontal platform, or some- to average this power level, and electric thrusters are not
thing like a latching hook or arresting cable, such as used 100% efficient, either. Using reasonable numbers of
on aircraft carriers. In that case the Skyhook name be- 60% sunlight time and 65% thruster efficiency, we get our
comes a literal description. An alternate method is have solar panels need a peak output of 1.22 Watts/kg/year. If
a vertical capture net. It’s as wide and tall as needed to we want to deliver 1000 tons/year, then the power supply
make a good target. The vehicle has redundant capture needs to be 1.22 MW. Nuclear power is excluded from
latches deployed ahead of it, and arrives slightly faster so consideration for political and safety reasons so close
it runs into the net, and snags multiple cables. to Earth. Existing solar cells, allowing for 100% over-
head, have a mass of 1.68 kg/m^2[3] , and an efficiency
It is assumed the landing will be automated for uncrewed
of 29.5% vs a Solar constant of 1360 W/m^2. Therefore
cargo delivery, with radar, lidar, and other aids to getting
they produce 238 W/kg of output and their mass is 5.1
within the landing target area. The landing platform is
tons.
made a multiple of the navigation accuracy in size to have
a high probability of hitting the target. The best design is Thruster Type - For low Earth orbit, there are three
types available with near-term technology: Ion, Plasma,
214 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

or Electrodynamic. Electrodynamic uses less “fuel”, but elevator. Still, reducing the work for a launch vehicle
is not as well developed. Ion is well developed, but does by 30-50% brings dramatic cost reductions. For smaller
not scale up to the high power levels as well. We will bodies such as the Moon or Mars, a Skyhook could span
assume Plasma thrusters, but development of Electrody- the whole gravity well.
namic should be pursued, and all three types considered As a large transport infrastructure project, similar to a
as candidates. Current plasma thrusters in development bridge or airport on Earth, the Skyhook is built when traf-
are designed for 200 kW continuous power, so 6 units fic demands it and not before, and then expanded incre-
plus some number of spares would be needed for the Sky- mentally. The materials for the Skyhooks, such as car-
hook design. The estimated mass of the thrusters is 3
bon fiber, may come from orbital mining and process-
tons. ing. In that case their construction would not require large
Fuel Requirement - With 1000 tons/year of cargo de- amounts of mass to be launched from Earth. Even if all
livery to which 2.4 km/s of velocity is added, we need the mass has to be brought from Earth, the potential for
to expel 48 tons/year of thruster propellant at 50 km/s to improved payload justifies at least more analysis to see if
maintain the Skyhook orbit. This can either come along it is feasible.
with the cargo, serving as 4.8% overhead, or if materials
are being extracted from nearby asteroids can come from [1] Gunn, Dr. B.M.Lunar Basalts and Anorthosites
them. The latter is preferred since it’s more efficent for
[2] National Academies Press Orbital Debris: A Technical
the launch vehicle. Assessment, 1995
[3] Emcore Space Solar Cell Products
Design Issues

Space Environment Radiation Environment 5.13 Lunar Development

Plasma Environment - The ionosphere can cause charge
build up. The Moon is obvious to anyone who has looked up at
the sky. The same reasons for expanding civilization to
other locations apply to it too. They include access to new
5.12.1 [Text still to be integrated] energy and raw material resources for upgrading civiliza-
tion, and to meet the other objectives listed at the start of
multiple smaller orbiting systems, known as Rotovators Section 4.1.
can perform most of the same task, and use current ma- However, we do not immediately start developing the
terials. Each rotovator has a smaller task, so the re- Moon for several reasons. First, we must travel through
quired material strength is lower. We also take advantage the Low and High Orbit regions to get there and navigate
of orbital mechanics to transit between them, which re- the Moon’s gravity field to reach closer orbits or the sur-
quires no materials at all. Various space elevator concepts face. This requires more energy and new transport equip-
have been theorized, and small-scale experiments flown ment. The Lunar surface has different conditions than the
in space. Significant R&D is needed to bring this tech-
orbital region around it, or the orbital regions around the
nology to a ready state. When traffic volume is not large, Earth. So surface equipment also needs new designs. De-
and much of it is restricted to low orbits, the savings from
veloping the Moon is made easier with support from the
a space elevator are not large enough to justify their con- previous orbital locations. So we delay the start of Lu-
struction. We therefore place it in this sub-phase, where
nar development until after Phases 4A and 4B have been
reaching high orbits gives them greater advantage. El- started, and necessary research and development (R&D)
evator research can be combined with a variable grav-
is completed in Phase 0M. The Moon’s surface is roughly
ity research facility, as both can use rotating structures. the size of Africa and Australia combined, so it is much
An eventual skyhook network can provide fast velocity too large to be developed all at once. Once started, Phase
changes for people and cargo around gravity wells, while 5A would continue in parallel with previous phases, with
electric tugs can perform slower transfers between them, progressive upgrades and expansions over time.
and to new locations beyond the existing network.
Concept exploration for Lunar locations begins with de-
Skyhooks scribing the characteristics of the region and an indus-
A space elevator system in the form of a rotating Sky- try survey to identify possible future activities. Moti-
hook would allow using highly efficient electric thrusters vations, economics, technology, and prior developments
in place of low performance chemical rockets for much of elsewhere will drive which of these activities get started
the transport job in gravity wells or between orbits. The and when. Combining this information we can define spe-
first one could be built in Low Earth Orbit, and then oth- cific Lunar projects, place them in approximate time or-
ers in higher orbits and around other bodies. The Earth’s der, and link them to other phases and projects. In the
gravity well is too deep to fully span with current materi- course of doing this, we will identify needed R&D from
als, so the low orbit Skyhook is not a full ground-to-orbit Phase 0.
5.13. LUNAR DEVELOPMENT 215

5.13.1 Lunar Region Features of 1361 W/m2 is the same. That sunlight is partly blocked
in lower Lunar orbits, and blocked 50% or more on the
The Lunar region includes the Moon itself, and orbits surface on a monthly cycle. There is no natural atmo-
with average distances (semi-major axes) within 35,000 sphere to moderate temperature changes on the surface.
km of the Moon’s center. At this distance the Moon’s Objects exposed to space and direct sunlight can vary
gravity exceeds the Earth’s by 50% and is the dominant from 100-400K at the equator, and as low as 30K in shad-
local force. Lunar orbits in general are unstable due to owed polar craters. The surface is covered by particles
the influence of the Earth and Sun. The Moon also has separated by vacuum, and is an excellent insulator. Soil
mass concentrations from past impacts that create an un- temperatures more than 30 cm deep are not expected to
even gravity field. Particular orbits, where these effects vary more than a few degrees from average, despite two
are minimized, can be stable for long periods. Other- week intervals of daylight and night. The average soil
wise propulsion must be used to correct them, or the or- temperature is about 240K at the equator (Vasavada et
bit allowed to evolve over time. Since the Moon orbits the al. 2012), and colder with increasing latitude and lower
Earth, the Lunar region is embedded in the larger High sun angles. Heat flow from the hotter interior has been
Orbit region around our planet. measured at 0.008-0.03 W/m2 , depending on location.
The concentration of radioactive elements in the crust and
good insulation properties leads to a somewhat uncertain
thermal gradient of 10-50 K/km of depth.
Temperatures in orbit will mainly depend on the percent-
age of time in sunlight vs in the Moon’s shadow, and how
much light is reflected or emitted from the Moon at a
given distance. The Earth appears about 2 angular de-
grees in size from the Lunar region, and thus fills less than
0.01% of the sky, and makes a small thermal contribu-
tion. The remainder of the sky is the cosmic background
at near absolute zero (2.7K), which serves as a radiative
heat sink. Since orbits can be elliptical and change over
time, the resulting thermal environment can vary signif-
icantly. The albedo, emissivity, and thermal properties
Figure 4.12-1 - Lunar topography referenced to average radius of an orbiting object will determine how much heat is
(a 1737.4 km sphere). gained and lost from environment conditions, and what
equilibrium temperatures they will reach.
As noted above, horizontal surface area of the Moon is
quite large, 37.93 million km2 or about one quarter of Atmosphere and Water - As noted under temperature,
the Earth’s land area. This does not include the added the Moon has no natural atmosphere. If pressurized envi-
exposed area from large-scale topography (Figure 4.12- ronments are needed for people or equipment, the gases
1) and numerous craters and other small-scale features. need to be produced locally, or imported. About 50 kg/s
See USGS Topographic Map 3316 of the Moon, 2015 of solar wind particles flow through the region at high ve-
for a more detailed one. locity. Since this is spread over an area of 3.85 billion
km2 it would be difficult to collect in useful amounts. An
The Moon orbits the center of the Earth-Moon system ev- estimated 600 million tons of water ice has been discov-
ery 27.3 days with respect to the stars (inertial frame), and ered in permanently shadowed craters near the poles. The
29.5 days with respect to the Sun, which sets the length remainder of the Moon’s surface has 10-1000 ppm chem-
of the Lunar day. The Moon makes one rotation per or- ically bound to minerals. The dry and airless state of the
bit period, and therefore keeps approximately the same Moon is due to the low escape velocity, coupled with ex-
side facing Earth. It is not exact because the Moon’s orbit tended periods of early heating from impacts, tides, and
is not circular, it has a slight residual pendulum motion, radioactive decay.
and the Earth’s diameter is 1/30 of the Moon’s distance.
That size allows varying views of the Moon, depending Ground Loads - Ground strength for surface structures
on where you are. Orbit durations around the Moon vary and transport is adequate to excellent. The surface con-
from 108 minutes close to the surface, to 6.8 days at the sists of broken rocks and dust of various sizes, which
upper edge of the region. The main environment and re- compacts a few cm but then can support heavy loads. Ex-
source features of the region are as follows: cavation, mining, or drilling reach the limits of overly-
ing rock strength at ~38 km. Depths below this require
support structures. Much of the shallower depths consist
Environment Parameters of fractured material, and will also need support against
movement and collapse.
Temperature - The Moon has the same average distance Gravity Level - Surface gravity averages 1.625 m/s^2, or
as the Earth is from the Sun, so the incoming Solar Flux
216 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

average.
Transport Energy - Reaching the Lunar region from the
Earth’s surface involves first getting to a high orbit that
intercepts the Moon’s location, then entering a stable orbit
or landing on the surface. The first of these theoretically
requires 62 MJ/kg, and entering orbit or landing requires
up to 2.8 MJ/kg. These are ideal values for potential and
kinetic energy. The actual expended energy depends on
the details and efficiency of the transportation methods.
The transport energies from Earth are high compared to
the 10-20 MJ embodied energy of typical products or 15
Figure 4.12-2 - Lunar surface gravity map. Near side on left, far
MJ/kg of high energy chemical propellants. This favors
side on right.
local production where possible.
Escape velocity from the Lunar surface is 2380 m/s, or
21% of Earth’s. Therefore escape energy is only 4.5% of
1/6th of Earths, with a total variation of 0.0253 m/s^2 ac-
Earth’s. Low orbit velocity is 1680 m/s, so the difference
cording to location (Figure 4.12-2). Free-fall conditions
to escape is an added 700 m/s. Orbit velocity at the upper
in orbit produce no effective acceleration. Structural sup-
edge of the region is 375 m/s, and escape an additional
port needed against gravity is therefore much lower than
155 m/s. These velocities and the associated energy are
for Earth. The gravity levels required for long-term hu-
relatively low. Internal regional transport and reaching
man health and plant and animal growth are not yet de-
the High Orbit and Inner Interplanetary regions are then
termined, but are likely to be more than natural surface
fairly easy from this region.
or free-fall levels. Artificial gravity on both the surface
and in orbit can be generated by rotation, otherwise stay
times may be limited. Centrifugal forces from rotation Available Resources
will impose significant design loads on structures.
Radiation Level - Unprotected radiation levels on the Energy Supply - The Sun provides a nearly constant flux
2
surface and and in orbit are in the range of 0.1-1 mSv/day, of 1361 W/m at 1 AU. The Earth’s orbit is slightly el-
which could reach lifetime astronaut exposure limits in a liptical, and the Moon’s orbit brings it closer to and far-
few years. A single powerful solar flare can deliver lethal ther from the Sun than the Earth. So the available energy
amounts of radiation, although most are much weaker. in the Lunar region varies somewhat from this value. In
Radiation also causes long term damage and electronic full sunlight this energy is sufficient to process about 2
2
faults in equipment. The radiation mainly comes from the tons/m /year of raw materials. Collecting the energy with
Sun and cosmic background. A meter or more of surface solar panels or concentrating reflectors may only require
2
material or building underground can provide safe radi- 2 kg/m of equipment. So the primary energy return can
ation levels on the Lunar surface. Bulk mass of various be very high. Most of the Moon’s surface is exposed to
kinds can provide shielding in orbit. the Sun 50% or less of the time, depending on local to-
pography. Higher latitudes get less sunlight per unit of
Communication and Travel Times - Round-trip (ping) surface area due to lower sun angles. Orbits are exposed
communication times from Earth to the Lunar region to sunlight 50% or more of the time, generally increasing
varies from 2.2 to 2.94 seconds, depending on where in with altitude. Dark periods are relatively short in orbit.
the region, and the Moon’s distance in it’s orbit. This in- Localized areas on the Moon contain up to 10 parts per
cludes satellite relay time if you are communicating with million of Thorium and Uranium, which may be useful
areas behind the Moon that can't be seen directly from for energy production.
Earth. Travel time from Earth is nominally 3-4 days for
people, by direct transfer orbit. Cargo delivery by electric Dark periods on the surface last half a Lunar day, or
tug is much more efficient, but also much slower. Slow two weeks. Since direct sunlight isn't available for this
transits without shielding or other mitigation would ex- time, alternate energy solutions are needed. Possible ap-
pose people to high levels of radiation while traversing proaches include:
the Van Allen belts.
Stay Time - Average stay times for people affect trans- • Ceasing Operations - High power operations can
port needs and the types of habitation required. Only nine simply be stopped when sunlight isn't available, and
trips with people have been made so far to Lunar loca- minimal support provided conventional sources like
tions, and those lasted only a few days each. These trips batteries. One of the following options can be com-
were made nearly 50 years ago. This data is too sparse bined with reduced nighttime operations, but not
and too old to establish an average for design purposes. stopping them entirely.
Future projects will need to define stay times based on
internal needs, until long-term habitation establishes an • Thermal Storage - This uses the vast amount of
5.13. LUNAR DEVELOPMENT 217

rocks and dust as the storage medium, and the nat- 5.13.2 Industry Survey
ural vacuum as insulation. The material is heated
during the day by sunlight, and the heat used to gen-
erate power during the Lunar night. We can develop a list of possible Lunar activities by look-
ing at all Earth industry categories, and identifying ones
that can potentially operate there in the future. To these
• Nuclear Power - The surface is already bathed in we can add any activities that are unique to the Lunar
high levels of natural radiation, so nuclear sources region. Existing industries are classified for statistical
are not as much an issue as on Earth. A variety of purposes by the North American Industry Classifica-
nuclear designs are possible, with different power tion System (NAICS). We will use their numbering sys-
levels, and fueled from Earth or local sources. tem and sequence for our survey, and insert unique Lunar
items where most appropriate.
11 - Agriculture: Local agriculture will be useful to the
• Beamed Power - Lunar orbits receive sunlight when extent that people and other living things in the region
the surface does not. Energy can be directed to need food, and it avoids transport from Earth or other
the surface using simple reflectors, microwaves, or space locations. Plants can also recycle waste products
lasers, in order of smaller target size vs distance. from people while producing food and other useful prod-
ucts. Filtered sunlight is widely available in Lunar orbit,
but the long Lunar night on the surface, and lack of sun-
• Transmission Lines - East-West transmission lines light underground, would require artificial lighting. Car-
can deliver power from areas in sunlight to those that bon and water are rare in Lunar regolith, and may need
are dark. In the worst case at the Equator the lines to be imported from other locations.
would have to be 2700 km long to provide continu- 21 - Mining: The Moon’s surface is covered by an av-
ous power. This is 1/4 of the Lunar circumference, erage of 10 million tons/km2 of already broken up rocks
with one line in each direction. These are very long and dust, and 1.36GW/km2 of daytime solar energy is
distances, and would only make sense when large- available to power the mining and processing operations.
scale development is present. Shadowed craters get Mining appears feasible for local use, and to supply loca-
no sunlight, but the crater rim may get adequate tions which take less effort to reach than from the Earth.
power levels. A shorter transmission line can deliver Delivery of mined products to the Earth itself would
that power to where it is needed. be limited. The mineral oxides making up most of the
Moon’s surface are also widely available on Earth at low
cost. There may be some rare materials that will be worth
Materials Supply - Lunar orbits are essentially devoid of transporting that far, especially if very low cost transport
raw materials, so projects located there must import what is developed.
they need. The Moon’s surface has a somewhat variable,
22 - Utilities: All activities in the Lunar region will need
and reasonably well understood, Geology. This is the re-
sult of a number of lander and orbital missions, some of local power, which at first can be supplied by equipment
which returned samples for analysis. Lunar meteorites such as solar panels and batteries brought from Earth.
have also thrown to Earth by impacts. Broadly, the sur- When higher power levels are needed, simple systems like
face composition is oxide minerals of silicon, iron, cal- concentrating reflectors and thermal storage can be made
cium, aluminum, and magnesium, in order of elemen- locally. As local manufacturing develops, more complex
tal concentration, with 3-4% other elements. There are power systems can be built. Our civilization consumes a
few volatile (low boiling point) compounds left on the large and growing amount of energy, and solar energy is
Moon. It formed in a molten state, suffered many high available in enormous quantities in the High Orbit region.
energy impacts, and is too small to keep an atmosphere, The Lunar region is likely too far to efficiently generate
so these compounds mostly escaped. Since the Moon re- and transmit energy to Earth, but can be a source of ma-
tains no atmosphere, it does not slow incoming objects, terials and products to do so in the High Orbit region.
and weathering does not occur as on Earth. This results 23 - Construction: The Earth is large enough to support
in a heavily cratered and broken up surface, but fairly un- its human population from a purely physical space stand-
modified otherwise. The surface is covered by a 2-8 me- point for the foreseeable future. At the density of Man-
ter thick Regolith made from a mix of the original crust hattan, a population of 10 billion would fit on 0.25% of
and impacting asteroids, which have been thrown around the world’s land area. Even though more space is not nec-
many times during crater formation. Because of the bro- essary, it would be desirable for a number of reasons. The
ken nature, the regolith is easily collected and moved. Lunar region can supply some of that physical space. Ac-
Since the layer is global on the Moon, large quantities are tivities in the Lunar region that involve people and other
available. See the Lunar and Planetary Institute’s Lunar living things will require habitats to support them, since
Sourcebook (1991) for more details about the Moon’s the natural environment cannot. Large habitats cannot
geology. be delivered in a single piece, so on-site construction is
218 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

needed for them. Beyond habitats, other constructed fa- icant operations begin there. Networking, communica-
cilities which are too large to deliver as single pieces will tions, and information technology for internal use within
need construction capacity. the Lunar region, and to Earth and other regions, will be
31-33 - Manufacturing: The range of possible manu- extensive from the start. The equipment used is complex
factured products in the Lunar region is vast. The first to and fairly low mass. So it is likely to continue to come
be made are likely to be those that use local resources, from Earth.
displace significant mass otherwise brought from Earth, 52 - Finance and Insurance This will most likely be sup-
and are fairly simple. Examples include bulk shielding, plied from Earth, since ownership rights, contracts, and
thermal storage mass for nighttime power, basic chemi- money are all non-material relationships and can be trans-
cals for propulsion and life support, and mineral products acted remotely.
and metals for construction. Once starter sets of produc- 53 - Real Estate, Rental, and Leases The Outer Space
tion equipment are delivered to the region, they can begin Treaty prohibits claiming celestial bodies but allows
to bootstrap their own expansion, and increase the range peaceful uses in space. International agreements for the
of products made locally. Transport energies to and from Space Station, and orbital slots and frequencies for com-
nearby regions are low, so a wider range of material in- munications satellites have set precedents for ownership
puts and product exports for local manufacturing may be and use of equipment in space. What is not settled is
feasible. territorial rights to less than entire bodies. For example,
42 - Wholesale Trade: Individual operations in the re- if someone builds a mining operation on the Moon, or a
gion will trade with each other and with other regions, colony in Lunar orbit, will they own a territory around it in
according to the economic principle of comparative ad- the sense of land rights on Earth. A reasonable answer is
vantage. The Moon has large amounts of easily extracted exclusion of others for technical and safety reasons. Thus,
raw materials. So those materials, or products made from someone else cannot build so as to shadow your solar col-
them, are likely to flow to Lunar and Earth orbits where lectors or damage your equipment with rocket exhaust.
they are lacking, and where transport energy is lower than Until the legal questions are settled, we cannot say what
from Earth. Ownership and control of trade may be from the scope of this industry category will be. Like Finance
Earth rather than local. Communicating with the ground and Insurance, whatever the scope is, it will likely be han-
is fairly easy from this region. dled from Earth for similar reasons.
44-45 - Retail Trade: We do not expect significant re- 54 - Professional, Scientific, and Technical: Some re-
tail trade to develop in the region for a considerable time.search and development, especially scientific, is best car-
Early populations in the Lunar region will be there for ried out locally in the Lunar region. Such activity began
scientific and basic industry purposes. Their employers in the 1960’s, to a large degree for political reasons, and
will likely provide for their basic needs, and even bud- continues to the present. Scientific and increasingly com-
get for some optional items like entertainment media. As mercial activity is expected to continue in the future in the
permanent habitation develops and a non-working popu- Lunar region. Most of the people involved in space ac-
lation accumulates, there will be more time and resources tivity so far have been on Earth, building the equipment
that can be applied to personal choices, and non-essential and operating it remotely. This is likely to continue for
specialties. Retail trade may then develop. some time, but increasing numbers of people will visit
48-49 - Transportation and Warehousing Transport and work in the Lunar region. They will be concentrated
is required from previously developed regions in order in scientific and technical fields at first.
to start any activity in this region. Such transport is ac- 55-56 - Management and Organizational Support:
counted for in earlier phases from their starting locations Business management and administration will likely be
on Earth or in Earth Orbits. Transport within the Lu- mostly handled from Earth. Support activities like orbital
nar region and to farther destinations is accounted for debris collection and radiation remediation are tasks for
in this phase. At first the transport equipment is all Earth orbits. These in turn may be supplied from the Lu-
made elsewhere, but over time propellants, structures, nar region.
and more complex items can be made locally. Transport 61 - Education: This will likely be done in advance or
includes infrastructure such as spaceports and surface ve- remotely from Earth at first. Local education will be
hicle hangars. Warehousing includes all types of storage, mostly limited to training for industrial operations. Gen-
which will start with bulk items like unprocessed rock and eral childhood and higher education will be deferred until
basic chemicals. permanent habitation develops with populations of young
51 - Information Transmitting information through people to teach.
space requires no mass and little energy. So many satel- 62 - Health and Social Services: Health monitoring and
lite systems already exist in Earth orbit purely for com- first aid capacity are needed for people in the Lunar re-
munications, and nearly all have communications systems gion from the start. At first, this will be by training crews
to support their primary purpose. Communications net- themselves, and remote monitoring and advice. As the
works will be extended to the Lunar region once signif- number of people grows in the region, more dedicated
5.13. LUNAR DEVELOPMENT 219

equipment and health specialists can be supplied. Telep- 5.13.3 Project Drivers
resence, artificial reality, and haptic robots are not cur-
rently good enough to do health care remotely, and may Which of the activities identified above make sense for
not ever be so from Earth, due to speed of light delays. the Lunar region, and when, is a function of a num-
Automated health care carried out locally may be pos- ber of factors. They include project motivations, their
sible. These depend on future R&D, so for now health economics, available technology, and prior developments
care concepts would involve existing methods. Nursing elsewhere. These drivers will change over time, affecting
and residential care is assumed to be provided on Earth of these activities get started and when.
at first, by returning people there. As population grows in
the region, local facilities may be established. We assume
social assistance will be done remotely or not be needed, Motivations
as basics like habitation and food will be provided to ev-
eryone in the region, or they will be returned to Earth. Economics
71 - Arts, Entertainment, and Recreation: Entertain-
What projects are feasible, and when, is partly driven by
ment will start out as remote delivery or software for
available technology and support from previous phases,
people in the region. Active recreation would start with
but also by the economics of the projects themselves.
necessary exercise for health maintenance. Early Lunar
Economic principles like comparative advantage and re-
landing sites may be preserved as historical locations, and
turns on investment will still apply in the Lunar region.
unique sports may develop in the Lunar region, such as
The advantages of the Moon include relative closeness to
rover racing or low-g gymnastics, but these are specula-
Earth and already populated Earth Orbits, low energy to
tive at present.
reach those orbits (including aerobraking to reach lower
72 - Accommodations and Food: Living space, food, orbits), large quantities of raw materials and solar energy,
and drink must always be provided for people in the re- and a position high in the Earth’s gravity field, making
gion. At first these will be provided by sponsoring orga- access to more distant regions fairly easy. The Moon can
nizations and self-operated. As local capacity grows and also be used for gravity assists for both departing and ar-
people establish long-term residences, there will be room riving transport.
for temporary travel accommodations, tourism, and spe-
The Moon is deficient in volatile compounds because of
cialty food and drink locations. Space tourism to Low
its early history of high temperatures and low escape ve-
Orbit has already happened, and proposed to the Lunar
locity. So while large amounts of materials are available
region, but it is a very limited market due to extreme cost.
in the Lunar region, they are not a complete source for all
It does demonstrate there is an interest once costs become
possible needs. A fully developed economy would there-
more reasonable.
fore supplement local materials from other space loca-
81 - Other Services: This category covers miscellaneous tions, such as Near Earth Asteroids, or from lower orbits
activities not covered elsewhere. Repair and maintenance or Earth. Near Earth Asteroids are preferred for bulk ma-
is highly desired for the Lunar region from the start. This terials because of the low energy to reach and return from
is because the equipment is either expensive to replace or them. Lower orbits and Earth are preferred for higher
life-critical. Personal services will start out self-provided value and finished goods because of the more established
until populations are larger. Private and civic organiza- industries there. The need for outside supplies then drives
tions are not needed at first, or self-organized. the need for transport and trade.
92 - Public Administration: At first, most of this will be Earth already maintains a complete civilization, and phys-
handled from Earth. Fire and public safety will start out ical products from the Lunar region are not likely to com-
self-provided, and develop as specialties as populations pete with Earthly ones on cost. Exceptions include scien-
grow. Environment quality and monitoring would be de- tific and collectible samples, where their value specifi-
signed into Lunar systems as necessities. Publicly funded cally comes from their Lunar origin. Information is low
civil and national security space activities have been the mass and fairly easy to deliver, so may have significant
only ones carried out so far in the region, and they are economic value when coming from the Moon. Energy is
likely to continue to be important. Government budgets a large market on Earth, and if it can be delivered eco-
for these activities are finite, while business and private nomically would be a spur to space industry in general,
ones are not limited in this way, so we expect the latter and for exports from the Lunar region.
to dominate eventually. For the sake of efficiency, public
and private projects should be integrated so as to support
each other. Technology

Prior Development

As of 2016, launch costs to the Lunar region were about

$55 million/ton, which severely limits what kinds of
220 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

projects are feasible. The raw wholesale energy cost is reaching orbit. Once Low Orbit projects are built
only $1000/ton, which shows there is much room for im- in Phases 4A, slower but higher performance al-
provement. A number of new transport systems are in ternatives can be used, and transport infrastructure
development. The SLS may eventually reach a cost of $1 built, from Low Orbit to the Lunar region. Even
billion/flight for 55 tons to the region using all chemical with chemical propulsion, system cost can be low-
propulsion, or $18 million/ton. The SLS and an efficient ered compared to the current practice of single-use
electric tug from low orbit could deliver about 105 tons transport. This is possible if a cheaper propellant
for $1.2 billion, or $11.5 million/ton. The Falcon Heavy source is available or the hardware can be used mul-
can lift 64 tons to Earth orbit, and with an electric tug can tiple times. Multi-use orbital transport vehicles are
deliver 50 tons to the Lunar region for about $200 mil- generally referred to as Space Tugs by analogy to
lion, or $4 million/ton. These are improvements, but still tugboats, which push ships and barges around.
very high. To the extent that transport costs are this high,
it favors high value/low mass activities, and those where • Missions to the Lunar Region - There have been
economic returns do not matter, such as publicly funded 64 successful public Missions to the Moon or
research and exploration. involving the Lunar region since 1959. Notable
among them were the Apollo missions which carried
• Activities that avoid/leverage transport mass, rela- people. About 20 more public and private probes
tive value over time allows new activities. are under development, intended to launch between
2017 and the 2020s. Two short-term crewed mis-
sions are also in development. A dozen more probes
5.13.4 Projects and crewed missions are currently proposed but not
yet in development. The NASA Deep Space Gate-
Phase 5A of our Upgrade Program would include a num-
way is a crew-tended station planned for the 2020’s
ber of future projects in the Lunar region. There are al-
in a halo orbit near the Moon, and the European
ready numerous current and near-term projects involv-
Space Agency has proposed a Lunar surface base
ing the region. We note them first, and their existence
by the 2030s. Both may involve international and
should be accounted for in future plans. Future projects
private participation. We expect publicly funded
are grouped by type, then in approximate start time order.
projects in the region to continue and evolve, but be
They are expected to overlap once started, and interact
limited by available funding.
with other program phases and projects and the rest of
civilization.
Future Projects:
Current and Near-Term Projects:
Lunar Orbit Production Lunar orbits require less en-
• Transport from Earth - Direct transport from ergy to reach than the surface, and are easy to reach from
Earth is being worked on by NASA through their the High Earth Orbit region because the Lunar region is
Space Launch System (SLS) and Orion Space- embedded in it. So we expect orbital production to pre-
craft projects. The SLS provides the capacity to cede surface production. Early production can be an out-
deliver payloads to the Lunar region (and other des- growth of public projects such as the Deep Space Gate-
tinations), and Orion for carrying people. Other way. These projects will begin with assembly of pre-
Chemical Rockets in varying stages of operation made components brought from Earth. Local production
and development can potentially deliver payloads can then be added incrementally. Production can start
to the region, but only the larger ones can deliver with pre-made tools and equipment, then bootstrap its
useful-sized payloads. The Scaled Composites own expansion by using the seed factory approach. The
Stratolaunch carrier aircraft is part of a hybrid tur- starter set is used to make some of the parts for expan-
bofan/chemical rocket transport system and can po- sion, with the remainder imported. Over time, more can
tentially support Lunar missions. The combined cy- be made locally, to the limit of what makes economic
cle SABRE air/rocket engine for the Reaction En- sense. Production outputs can be used locally in orbit, de-
gines Skylon spaceplane is in early stages of devel- livered to the surface, exported to existing markets in the
opment. Skylon is intended to carry enough payload High and Low Orbit regions, and exported to farther des-
(15 tons) it could support Lunar missions. tinations. We do not expect much export to Earth from
the region because civilization is well developed there and
• Transport from Low Orbit - Chemical rocket per- costs are likely lower.
formance is limited by the available energy of the
propellant. It is used in the Earth to Low Orbit • Materials and Energy Sources - There are no sig-
transport segment because it provides high enough nificant materials sources in Lunar orbit, so they
thrust to overcome the Earth’s gravity and prevent must be imported. Despite physical distance, Near
the trajectory from intersecting the surface before Earth Asteroids are an easier source of materials at
5.13. LUNAR DEVELOPMENT 221

first. This is because electric propulsion can be used, produced in Lunar orbit, then delivered to their final
which is much more efficient, and the Moon can be location for construction and installation.
used for gravity assists, reducing propulsive delta-V.
Neither is available for early transport to and from
the surface. Asteroids also have more varied com- Lunar Surface Production Surface factories can pro-
positions than the Lunar surface, allowing a wider duce some finished items, but the relatively limited vari-
range of products. When transport infrastructure ation of Lunar ores means more materials would have to
such as catapults and skyhooks become available be imported to support the factories. This favors export
(see Lunar Transport below), the lower cost, higher of unfinished goods and more finished imports, result-
volume, and shorter delivery time from the Lunar ing in two-way trade. The preferred locations for surface
surface becomes an advantage, shifting the balance production depends on availability of energy and mate-
to more local materials supply. Some materials are rial sources, local conditions, and the intended destina-
rare in both asteroid and Lunar ores, and would need tions and uses for the products. For example, a scientific
to be supplied from Earth. Orbital locations have outpost may not want industrial operations nearby if they
more solar energy available on a more consistent ba- cause unwanted disturbances. Possible products include:
sis. Since the embodied energy in products is typi-
cally larger than the kinetic energy to move materi- • Minimally Processed Regolith - Early Lunar min-
als from the Lunar surface to orbit, production can ing and construction would be in support of near-
generally be faster in orbit. term public scientific and exploration projects such
as a surface station. This would include clearing
• Materials Processing - Some products, like bulk and leveling sites and access paths; then excava-
radiation shielding, don't require further processing, tion, placement, and covering of station elements
just delivery. But most require conversion of raw with unprocessed local regolith for radiation, ther-
materials to finished materials by mechanical, ther- mal, and debris protection. Debris protection is
mal, chemical, electrical, or other methods. Pro- both from natural meteoroids, and materials thrown
cessing can start with the easiest and simplest meth- by lander exhaust. Covering may be by simple
ods. An example is water extraction from Chon- loose piling of material over a support structure, and
drite type asteroid rock, which only requires heating not require processing beyond sorting for rock size.
a container using concentrated sunlight, and a shad- Paving and blocks may be produced by melting the
owed condenser to collect the evolved vapor. Metal- soil with concentrated sunlight.
lic asteroids are dominated by native Iron, Nickel,
and Cobalt, which make up more than 95% of their • Water Extraction - The Moon is devoid of known
mass. However individual samples vary in composi- liquid water, but Lunar Water is known to exist in
tion, and may have undesirable trace elements, lack two main forms. The first is chemically bound as
desirable alloying elements, or contain rocky inclu- hydrates and hydroxide minerals, for which higher
sions. The Stony-iron group have higher percent- temperatures are required to extract it than the lo-
ages of silicates and less native metal. Processing cal environment produces. The concentrations are
is therefore generally needed to produce uniform al- ~10-1000 parts per million. The second is water
loys in desired sizes and shapes. ice trapped in permanently shadowed craters near
the poles, where temperatures are low enough to
• Orbital Fabrication - There are many known pro- preserve it. Asteroids can contain up to 20% wa-
cesses to convert ready materials into finished parts. ter, and Hydrogen can be combined with abundant
Which ones are suitable for Lunar orbit and in what mineral oxygen to produce 9 times its mass in wa-
sequence to develop them requires a lot of R&D. ter. So import is an alternative to the low concen-
trations and limited locations of local sources. In
• Orbital Assembly and Construction - This begins the long term, water is abundant beyond the mid-
with prefabricated elements delivered from Earth Asteroid Belt Frost Line, where temperatures were
and assembled in orbit, such as space stations and low enough to retain solid ice. If low cost transport
larger and more powerful satellites. As local pro- is available, they would be a preferred source over
duction develops, the orbital products are less con- the limited Lunar sources.
strained by launch mass and cost from Earth, so they
can use simpler and heavier designs. The Lunar orbit • Chemical Processing and Metallurgy - Lunar soil
region is a useful location to combine raw materials is about 40% Oxygen, so any of several reduction
supply, energy supply, and parts delivery from Earth processes can be used to extract it at a smaller scale.
or lower orbits. It may therefore develop as a ma- Oxygen is useful as a propellant, and for life support.
jor assembly and construction site. Large projects More extensive mining and processing can supply
like space colonies may be too massive to move once finished materials such as refined metals and basalt
constructed. In that case, smaller elements may be fiber for local use or for export. Early markets will
222 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

likely be in previously developed regions, and more Lunar Orbit Transport These projects include trans-
solar energy is available in higher Lunar orbits. So port based in the Lunar orbit region.
we expect most materials by mass will be sent to
Lunar orbit. The Moon’s surface is not uniform in • Reusable Landers - Early landings on the Moon
composition (see Lawrence et. al. 1998), with do not have the support of much infrastructure, so
high concentrations of Iron, Potassium, Phospho- require high thrust chemical rockets to navigate the
rus, Rare Earth, and Thorium mostly in the Oceanus Lunar gravity field. For transport down to the sur-
Procellarum region. These are the result of frac- face, Carbonaceous type asteroids contain up to
tional crystallization as the Moon cooled. 20% carbon compounds and water. This can be
reformed chemically to hydrocarbons and oxygen,
• Helium-3 Mining - Mining of Helium-3 from the which is a common rocket fuel combination. They
Moon has often been suggested for terrestrial power, would be produced at a high orbit location where
because it has low radiation by-products in fusion there is full-time sunlight for power, then an electric
reactions. There are three problems with this idea. tug delivers them to low Lunar orbit for efficiency.
The first is that Deuterium/Helium-3 fusion is about A low Lunar orbit fuel station then fills a reusable
ten times harder than Deuterium/Tritium, and we lander which delivers people and early cargo to the
don't yet know how to do the latter at all, much less surface. Oxygen is the most common element on the
economically. The second is the abundance of this Moon, and a number of ways of extracting it from
isotope is 15 parts per billion or less on the Moon. oxide minerals have been studied. An oxygen plant
The abundance is 1000 times or more higher in to refuel for the return trip can reduce the round-
Uranus and Neptune’s atmospheres, requiring pro- trip mass ratio from 2.9 (LOX/CH4) to 1.96. This
portionally less mining and processing to extract a increases payload per trip from ~25% to ~40%.
given amount. Although the outer planets are much
farther than the Moon, if we need He-3, then fu- • Orbital Tugs - These are needed for efficient trans-
sion would be solved, and it can power ships to reach port within Lunar orbits, and to and from these or-
to reach them. The third problem is relative energy bits to other regions. Transport to the High and Low
content. Pure He-3 can supply 200 TJ/kg, but at orbit regions can be relatively low energy. In some
Lunar concentrations the mined soil can only supply cases the Moon can be used for gravity assists to
up to 3 MJ/kg in reaction energy. Silicon and Alu- lower orbits, and aerobraking used to lower apogee.
minum make up 28% of typical Lunar soil. If 10% Gravity assists can also be used to escape from the
of the soil is used to make solar cells and structure Earth-Moon system, but upwards transport gener-
for power satellites, and the satellites produce 100 ally requires more propulsion. Gravity and drag do
W/kg, then the net power per mined kg of soil is 10 not help in changing orbits around the Moon, but to-
W/kg. A nominal life for such cells is 15 years in tal velocity requirements are fairly low. We expect
orbit, during which they would produce 4.75 GJ/kg most of the propulsion will be electric.
of energy. So if you wish to mine the Moon for en-
ergy supply, you can generate 1500 times as much • Lunar Skyhooks - Lunar basalt and Carbon from
via solar power than He-3 fusion. asteroids carbon can be turned into high strength
fibers to build a much larger type of orbiting cen-
trifuge known as a “Skyhook”. For a tip velocity
As noted above under Energy Supply, some regions of
equal to orbit velocity, the tip acceleration is much
the Lunar surface contain ~10 parts per million Uranium
lower at a larger size. If it is set to 1 Earth gravity,
and Thorium. These elements were concentrated in the
a crew can live comfortably in orbit, and make trips
crust by differential processes as the Moon cooled from
to the Lunar surface with very low fuel use. The
a molten state. The mined ore has an energy content of
tip velocity cancels orbit velocity at the low point,
about 800 MJ/kg, or hundreds of times the He-3 energy
so a lander can be dropped off at low altitude. It
content. So even for Lunar power, He-3 is not the most
would not be dropped directly on the surface be-
efficient source. Solar is even higher energy content, but
cause of the variable gravity field and Lunar moun-
to span the long Lunar night, shadowed craters, or for
tains and crater walls. The lander therefore uses a
portable power, nuclear may be useful.
small amount of fuel to cover the last 10-20 km to
and from the surface. A “frozen” orbit at 86 degrees
• Surface Fabrication and Assembly - TBD. inclination minimizes orbit perturbations and allows
access to nearly all of the Lunar surface. However
it limits the times when a particular location can be
Lunar Orbit Habitation accessed. An equatorial orbit would have more fre-
quent access but more limited surface coverage. Un-
like Earth, the Moon rotates so slowly that the ben-
Lunar Surface Habitation efit of rotation velocity is negligible.
5.13. LUNAR DEVELOPMENT 223

A large orbital Skyhook only makes economic A linear accelerator requires high peak power,
sense if the traffic to the Lunar surface is large because the acceleration must be completed
enough, so it would not be built right away. over the relatively short length of the device,
There is a one-time cost in mass to build a rather than a slowly accelerating centrifuge. It
Skyhook, but afterwards it saves most of the is therefore more suited to high volume de-
required lander propellant. A Skyhook can livery, where the large power supply is not a
be built incrementally, and provide partial sav- penalty. Comparing linear vs centrifugal sys-
ings before completion, which affects the eco- tems, it is not currently known which is pre-
nomics. At the other end of the rotation, the ferred for a specific delivery task.
centrifuge is moving faster than Lunar escape
velocity, and can therefore send vehicles to a
large range of orbits by choosing the radius and Lunar Orbit Services
time of release. Catching and releasing vehi-
cles affects the orbit of the centrifuge, but if
traffic is balanced in direction it is a temporary • Satellite Maintenance and Refueling Support -
change. If traffic is more in one direction than Most existing satellites are in the Low or High Or-
the other, the difference can be made up by bit regions, because most of civilization is on Earth
electric propulsion. Since low gravity is known and that is what their services support. This situ-
to be harmful, an orbital centrifuge can allow ation should continue for some time. This project
crew to mostly live in normal gravity, and op- covers satellite support from the Lunar region to
erate equipment by remote control with short other regions, as well as internally within the region.
ping times. At the same time, the centrifuge With a few notable exceptions (the ISS and Hub-
can provide easy access to the surface when ble Space Telescope), most space projects have been
needed. So it is not purely a transport system. single-use, because maintenance and refueling were
too difficult or expensive. However replacing entire
satellites when they break or run out of fuel is also
Lunar Surface Transport These projects include very expensive, due to the high cost of new hardware
transport based on the Lunar Surface. and transport to space. So a maintenance and refu-
eling capability is desirable. Raw materials from the
• Surface Vehicles - These begin with rovers for sci- Moon and Near Earth asteroids, combined with Lu-
ence and prospecting purposes remote-controlled nar region energy, can support this capability, along
from Earth. They progress to mining and construc- with projects in the other regions and support from
tion vehicles which are still remote-controlled or Earth. Support products include propellant supply,
smart enough to self-operate. Once people start to both for chemical and electric propulsion. Propel-
be located on the surface regularly, we add unpres- lants are used to refuel satellites, remove orbital de-
surized and pressurized crew transport, and eventu- bris, and either transport maintenance equipment
ally mobile habitation for extended stays. the satellites or bring the satellites to a maintenance
location. They also include materials supply, includ-
• Lunar Catapults - These are systems based on the ing radiation shielding for people and equipment to
Lunar surface, to accelerate bulk materials to orbital support maintenance and bulk material for radiation
velocity. They can be centrifugal or linear designs. belt depletion.
An electric centrifuge can deliver 1000 times the so-
lar array mass per year to Lunar orbit. This consists
of an electric motor driving a rotor with a long and
Lunar Surface Services
short arm that are balanced. During the day the so-
lar panels feed power directly to the centrifuge until
the tip of the long arm is moving at slightly above
Lunar orbit velocity. The payload is released and 5.13.5 Needed Program R&D
coasts to a collection station in low Lunar orbit. At
the same time, a counterweight is released from the Many of the future projects are not ready to start today,
short arm, which hits a hill behind the centrifuge. and will need research and development first. R&D to
The reason for the counterweight is an unbalanced support Phase 5A projects is assigned to Phase 0M. Some
rotor would produce huge forces on the centrifuge of that work may be carried out on Earth. Other parts
structure. Whether the payload is sintered blocks may require using the Low or High Orbit regions, or need
of unprocessed surface material (regolith), or first to be performed in Lunar orbit or the surface. The ones
separated to, for example, metals, is yet to be deter- which cannot be performed on Earth will require suitable
mined. An electric tug takes loads from the collec- transport to be available first. R&D and projects from
tion station to a higher orbit where further process- earlier phases may also have to have been started or com-
ing and production is done. pleted.
224 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

=====[Text to be Merged]===== general way: a Starter Kit, an Expansion Kit, and bulk
delivery of materials from the Moon to Lunar Orbit.
Low gravity is known to be harmful, so long-term habi- These are only some of the industrial infrastructure for
tats on the Lunar surface may require centrifuges to cre- the Moon, and does not include other end items you might
ate artificial gravity. An example would be a large habitat want to build there.
dome for spaciousness, and a centrifuge built around the
rim for living quarters. Residents would spend enough
time in the centrifuge to maintain health, but could work Starter Kit
and enjoy the low gravity the rest of the time. We have es-
sentially no data on how much gravity is enough between The starter kit includes the equipment to mine and pro-
zero and 1.0. We know the body deteriorates over time cess Lunar iron, and fabricate and assemble items from
in zero gravity. So as a worst case, people would need to it. This gives an immediate return on hardware delivered
spend most of their time in a one gee centrifuge of some to the Moon.
type, but this subject needs more research.
We developed our orbital infrastructure in earlier steps Iron Mining Iron-bearing meteorites have impacted
to avoid the overhead of a large gravity well. With that the Lunar surface since Since it’s origin. From Apollo
in place we can now go to and from the Lunar surface mission rock samples we know around 0.5% of the bro-
relatively easily. The advantages of using the Moon are ken material making up the Lunar surface, called the re-
that it is nearby and has very large available resources. So golith, is bits of iron [2] . It is generally small particles
once the cost and difficulty of reaching it are reduced, we attached to other minerals by melt glass formed during
want to develop it as the next step. the impacts. This is in addition to 5-15% iron oxides in
minerals, but metallic iron does not have to be chemi-
cally processed, which avoids complexity in a starter kit.
Development Concept
All you need to extract the iron-bearing fraction is a mag-
net. So you send a solar-powered rover around the nearby
We assume that thorough prospecting has been done via
lunar surface, sifting out the smaller rock particles, run-
orbiting spacecraft and surface rovers before a surface lo-
ning them past a magnet, and saving the part the magnet
cation is selected for mining and development. The Moon
attracts. Then carry that back to a central location and
has a different origin than the Near Earth Objects and has
dump the collected iron enriched dust. Repeat as many
undergone melting and separation, longer exposure time
times as needed.
to the space environment, and extensive mixing from im-
pacts. So its composition will be different than the NEO
population. Certain locations are special, such as craters Iron Extraction The next machine has a large concen-
that could contain water ice or places in constant sunlight, trating dish to focus sunlight, and some manipulator arms
both in the polar area. to smooth the ground and make slight depressions and
We deliver a combination of robotic, automated, and re- grooves for molds. Spread your iron-enriched ore over
mote controlled machines to the Lunar surface ahead of them and focus the sunlight to melt it. This is essentially
large numbers of humans to start preparing the site. Once sand casting except the metal is melted in place rather
mining, chemical processing, and parts production are than separately in a furnace and then poured into a mold.
operating, and you have a stockpile built up, then you Iron is relatively dense, and will sink to the bottom of the
can set up habitats and start having people there in larger mold, while the glassy and mineral parts will rise as a slag.
numbers. You continue to use the machines in parallel It will end up as basic shapes like plates and bars, with
to humans, but now they can be controlled locally rather slag stuck to it on both sides from melting. To prevent
than remotely. Taking this kind of approach will lever- the mold from melting, you can optionally select miner-
age the limited early human population. There might be als with high melting points as the mold material. The
as many as 50 to 100 machines per person. raw stock will need to be sand-blasted to clean off the
slag. Fortunately there is no shortage of mineral grains
We don't know for sure what the best starter kit is for
on the Lunar Surface to do that with. You now have an
Lunar development, but as an example we will assume an
inventory of iron stock to use for all types of construction
Iron-based economy. Whatever items cannot be made lo-
and parts making.
cally at any given time are brought from elsewhere, with
an increasing amount of local production as the facilities
develop. Metzger et. al. [1] have modeled bootstrapping Part Fabrication Machine tools such as lathes and
industry on the Moon, and found 12 tons might be suf- milling machines are designed to make metal parts from
ficient for a starter kit. Under a fairly wide range of as- metal stock. Modern ones are computer controlled, and
sumptions, that starter kit could grow to a much larger are themselves mostly made out of metal. Thus an ini-
installation. tial set of machines supplied with raw stock can mostly
Below we discuss three items for Lunar development in a make more machines. Given plans for a wider variety of
5.13. LUNAR DEVELOPMENT 225

machines, the initial set can increase in size in complex- Bulk Delivery
ity. For example a forge press or rolling mill is useful for
making some types of metal parts, but those are likely too A catapult system was mentioned earlier in the Space El-
heavy to bring to the Moon to start with. If you have a evator section as a method to deliver bulk Lunar basalt.
supply of raw materials, you can build them locally. This might be used to make high strength fibers for the
Lunar Skyhook. With the Skyhook in place a bulk deliv-
ery method might not be needed any longer, but we will
Assembly Standard assembly methods like bolts and leave that choice to later analysis. Other bulk materials
screws can use robotic equipment. For some items that produced on the Moon could be delivered this way, but as
need to be pressure-tight processes like welding are pre- a starting point for design, the calculations below assume
ferred. basalt is the cargo.
Welding - One process that may work on the Moon, since Delivery Orbit - Any object thrown into an orbit from
it is in vacuum, is to use articulating mirrors to direct the surface will intersect the launch point one orbit later.
a beam of concentrated sunlight at various angles. That Therefore your cargo needs a propulsion unit to raise the
could heat and weld parts to each other to make larger orbit, or you need something already in orbit to gather the
structures like habitats. If that does not work, then stan- cargo and change its orbit. One device in orbit is prob-
dard methods like vacuum plasma arc welding can be ably less expensive than one on every cargo, so for dis-
used. cussion we will assume a small Skyhook with a catcher
device hanging down some number of kilometers from
Expansion Kit the main core. The offset allows the catapult to throw
slightly slower than orbit velocity to match trajectory with
Power and Heating Electric Power - Solar power is the catcher. Missed catches will land somewhere else on
abundant on the Lunar surface. While initial power de- the Moon instead of coming back to the launch site and
vices will be brought from elsewhere, we can rapidly ex- doing damage. An equatorial location allows launching
pand it using our metal supply. Since mass is not an issue at about 108 minute intervals.
for a stationary facility, you can deliver closed cycle gen- Tip Velocity = 1680 m/s - We don't yet know how much
erators based on a thermal cycle, and use concentrated less than orbit velocity we will need, so for simplicity we
sunlight to power it. The mirrors can be made locally assume the catapult reaches full orbit velocity.
of sheet metal with an evaporated coating of aluminum
Cargo Mass = 20 kg - Using basalt fiber as the main ca-
to make them more reflective. Later generators can be
ble material, the Lunar Skyhook would have a nominal
made entirely locally, but in the early stages making just
mass ratio of 20:1. If we want to deliver 10 ton payloads
reflectors is fairly simple. To supply power during the Lu-
to the Lunar surface, that requires 200 tons of fiber. As-
nar night, you can take a container filled with Lunar rock
sume we lost 1/3 of the basalt to processing, so the starter
and a little gas for heat transfer, and surround that with
mass is 300 tons of raw basalt. We allow 10% extra time
more rock and insulation. Vacuum serves as a good in-
for maintenance and missed catches, so the average in-
sulator by itself. During the day you heat the container
terval is 2 hours per cargo. If the catapult launches 20 kg
with concentrated sunlight, and at night run the thermal
at a time, that will be 43.5 tons per year if it only oper-
generator from the stored heat.
ates during the Lunar day, so about 6.9 years to launch
Process Heating - Many industrial processes need heat- enough cargo for the Skyhook construction. Larger ver-
ing in some form. For this you can also use concentrated sions would work proportionately faster.
sunlight without an electrical generator.
Containers - Bulk Lunar rock will not hold together
when tossed at high velocity. Two options are iron con-
Chemical Processing Unprocessed Lunar rock can be tainers from Lunar rock, and Kevlar bags which are de-
used for things like shielding, and with melting can be livered and recycled.
cast into blocks for roads or landing pads, but for other
elements and compounds some chemical processing will Iron Containers:
be needed, such as for reducing oxide minerals to metals
and oxygen. Some elements are rare in Lunar rock and
would have to be brought from elsewhere. These would be made from local metallic iron, which
could be used in orbit as another building material. Iron
also gives the opportunity to fine tune the trajectory af-
Shelter Some items need protection from daily temper- ter the catapult throws the container using magnetic coils.
ature variations, radiation, meteorites, and human-caused Together with the catcher mechanically positioning itself
hazards such as lander rocket exhaust. Most of that can during the 30-40 minute ballistic trajectory of the cargo,
be met with a structural metal shelter covered with suffi- that should ensure a high percentage of catches. Lunar
cient Lunar regolith either in natural form or sintered into regolith has a density of about 3 g/cc, therefore a con-
blocks. tainer holding 20 kg will have a volume of 7 liters. A
226 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

cone 13 cm in radius and 42 cm tall has about this vol- additional 42% in energy in conversion losses. So the to-
ume. The reason for a cone shape is so there can be a tal average power supply needs to be 11.5 kW, and each
single attach point to the catapult. We restrict the con- motor-generator needs to transfer 30 kW when operating
tainer mass to 10% of the cargo, or 2 kg. The load will ( 40 hp ). An off the shelf 40 hp electric motor has a mass
be constant across the cylinder height, so thickness will of 250 kg. Given the Lunar night, we need to double the
vary to maintain the stress, and the bottom will be domed solar array peak power to 23 KW. Modern arrays have a
for optimum strength, so we can model the structure as power level of about 100 W/kg, so the solar array mass
a 50 cm tall cylinder with an area of 4000 cm^2. With would be 230 kg.
iron density of 7.8 g/cc, that gives a wall thickness of 0.64
System Mass Estimate - Besides the bare rotating arm
mm. The cross section area is 5 cm^2, and with a work- and power supply, there will need to be anchor structures,
ing stress of 125 MPa, allowed load is 62 kN.
robots to collect the raw materials and put it in containers,
the fine tuning magnets downrange, and communications
Kevlar Containers: and other support equipment. The total for this is un-
known until more design work is done. For now, we will
Kevlar is a high strength material used on Earth. Bags assume each of the two centrifuges will mass 1.25 tons
made of Kevlar cloth with a capacity for 7 liters each total and total support equipment, including power, will
would be delivered to the Lunar surface. A typical cloth be an additional 2.5 tons, for a total of 5 tons. Since the
( 5 oz/yd^2, 650 lb/in strength ) of similar dimensions to system can launch 168 times per lunar day, that means it
the iron container would have a breaking strength of 21 can deliver 3.36 tons. Mass payback therefore takes 1.5
kN, and a working strength of half that. We assume a 6 lunar days (1.5 Earth months).
layer bag (or thicker cloth) to get the required strength,
with steel wire mesh woven in for magnetic steering. Es-
timated mass is 0.45 kg/bag, or 2.25% of the cargo mass.
Without an initial Skyhook the bags would need to be Lunar Catapult 1750 m/s is the velocity needed to
delivered with a lander, along with the rest of the cata- reasonably clear Lunar obstructions. 1680 m/s is min-
pult system. For 300 tons delivered, that would require imum theoretical orbit velocity, but the Moon is physi-
15,000 bags with a mass of 6.75 tons. cally and gravitationally lumpy, so you need to throw a
Radius = 1000 m - A force of 62 kN on a 22 kg loaded little faster. Assuming a 50 meter radius throwing arm,
container implies a centrifugal acceleration of 2820 m/^2 the tensile stress along the arm at that tip velocity is 156
( 288 g’s ). With a tip velocity of 1680 m/s, the radius is g-km. High-strength carbon fiber has an ultimate strength
then 1.0 km, and it makes 0.267 rotations/s ( 16 rpm ). of up to 386 g-km.
For accuracy and stability, the structure will need to be Centrifugal forces are higher at the hub than the tip.
something like a truss frame, so it will support itself when
That’s because at the tip you are only supporting the pay-
not rotating. Added tension cables take up the accelera- load, and at the hub you are supporting the payload + arm.
tion load at full speed since cables can withstand higher Therefore the arm will be thicker at the hub to handle
stress. the increased forces. We don't want to work at ultimate
strength, and allow a design factor of safety. I will as-
Arm Mass = 92 kg - We used a working length of 126.4
g-km for carbon fiber in the Space Elevator section. We sume 2.4 as a reasonable factor. This reduces our work-
ing strength to 161 g-km. The arm taper then works out
have an average of 144 g’s over 1 km in the centrifuge,
which produces a mass ratio of 3.1, and thus a cable mass to 2.8:1 from center to tip. This is a reasonable number.
of 2.1 times the loaded container. The centrifuge should The circumference of rotation is 2 pi x 50m = 314 meters.
be balanced so it does not move during operation, so a to- So at a tip velocity of 1750 m/s implies 5.57 rotations per
tal of 4.2 times the container mass of 22 kg or 92 kg. The second or 334 rpm. This is quite a reasonable number
balance arm should be shorter, and release un-contained for an electric motor geared down by a factor of 10, since
counterweight rock at the same time as the loaded con- they commonly run at 3600 rpm.
tainer, but in the opposite direction, and aimed to hit a You are correct that throwing something into orbit this
hill or travel far away ballistically. way will bring it back to the launch point one orbit later.
Power = 11.5 kW - A 22 kg container moving at 1680 m/s There are several ways to avoid collision with the launcher
has a kinetic energy of 31 MJ. If the catapult is operated or terrain. First, you can choose a local high point for
once per 108 minutes when operating, it needs to supply the catapult. Second, you can include a small cold gas
4.8 kW on average plus losses. This is doubled due to the thruster that fires at the high point of the orbit, and raises
counterweight rock. Additional energy is needed to spin perilune so you don't hit anything. The gas can come from
up and down the rotating arm. If we use two catapults, baking lunar rock in a vacuum (which are both abundant
the energy of one can be transferred to the other rather on the Moon). Some of the mineral oxides will decom-
than wasting it as braking friction. The rotating arm has a pose at high temperature, releasing Oxygen. Third, you
mass of 4.2 times the thrown mass, and motor-generator can have a “catcher” or chase vehicle in orbit to grab the
conversions can be about 90% efficient. Thus we lose an payload and adjust the orbit.
5.14. INTERPLANETARY TRANSFER 227

To keep the catapult physically small and low power, you System Concept
would throw many smaller payloads into orbit, then gather
them into a larger load for a cargo tug. The tug then hauls The system concept has the following main parts:
the cargo to a high orbit, such as Earth-Moon L2, a stable
point on the far side of the Moon. This location is in • Use free flying electric thruster powered vehicles to
sunlight all the time, so you have more energy to turn your reach new locations that have useful resources.
raw materials into useful products. The payload is raw
or selected Lunar rock. So it doesn't care that you are • Set up seed factories in each location to build up in-
subjecting it to 6000 g’s during launch. dustrial capacity, including more ships and seed fac-
tories for the next location.
Assume our tug makes a trip every 100 days, and hauls
100 tons of cargo. We therefore have to launch 1 ton • Produce fuel, life support supplies, and habitats at
per day. Since low Lunar orbits take 1.8 hours, we can each location so they can be occupied permanently.
launch 13.3 times a day to have the payloads arrive at a
common collection point. Therefore each payload would • Build more Skyhooks to provide fast velocity
be 75 kg. The kinetic energy of the payload is then 115 changes, but get the benefit of electric propulsion
MJ, and the catapult requires 35 kW of average power at efficiency.
50% efficiency. This is half the output of one of the Space
Station’s four main solar array wings, so is a reasonable High Earth Orbit Skyhook
power level. The Tesla Model S base model uses a 270
kW electric motor, so a 35 kW motor can be quite small. In a previous step we defined a Low Earth Orbit (LEO)
Since you are in Lunar night half the time, the solution is Skyhook with a tip velocity of 2400 m/s. Since it’s orbit
to build two catapults that operate during the day. This velocity is 7474 m/s, at the top of it’s rotation it could
makes them more efficient, since it is hard to load the release a cargo at a total of 9,874 m/s. At an altitude
payload unless the centrifuge is stopped. The flywheel en- of 1,226 km, or a radius of 7,604 km from the Earth’s
ergy in the arms just after launch can be transferred to thecenter, with a Standard gravitational parameter of 398.6 x
other catapult, rather than just dissipating it by braking. 10^12, we get an escape velocity of 10,239 m/s. Thus our
Regenerative braking, as this is called, is also common in LEO Skyhook is only 365 m/s short of reaching escape
electric cars. velocity. The LEO Skyhook can thus deliver cargo to an
elliptical transfer orbit with a semi-major axis of the orbit
Modern space solar arrays plus mounting and tracking on
of 54,278 km, and thus a high point of 100,952 km.
the Lunar surface would produce ~100 W/kg. 70 kW
would then require 700 kg of arrays. The catapult struc- A circular orbit at that altitude has a velocity of 1,987
ture is likely to be 20 times the cargo mass, or 1500 kg. m/s, and the transfer orbit arrives at 743 m/s. The differ-
We will need a mining robot to gather raw materials, and ence is 1,244 m/s. A tip velocity of 1,500 m/s or more
equipment to weigh and package the payloads. NASA has would allow injection to Mars and Main Asteroid Belt
already tested thermal extraction of oxygen from rock on transfer orbits, and any closer transfer orbits to the Moon
Hawaii, using volcanic rock similar to the Moon’s. A total and low inclination Near Earth Asteroids. A High Earth
mission mass allowance is then ~10 tons delivered to the Orbit (HEO) Skyhook can therefore serve as a launch
Lunar surface. Since we deliver 1 ton per day, and nom- platform to any desired inner Solar System destination
inal operating life for space hardware is 15 years, we de- orbit by selecting the radius, and thus velocity, and the
liver a total of 5,500 tons, or 550 times the system mass. time, which gives the direction, of release. This location
is outside of the Earth’s radiation belts, but it is also un-
protected by the Earth’s magnetosphere from solar and
References cosmic radiation. So human habitats will need radiation
shielding. Such a high orbit is relatively easy to reach
[1] Preprint obtained from author, Apr 2012, to appear in
from the Moon or NEO’s, so bulk matter for shielding
Journal of Aerospace Engineering
will likely be brought from one of those sources.
[2] Morris, Origins and size distribution of metallic iron
The construction sequence would start with fetching Near
particles in the Lunar Regolith, 1980
Earth Orbit asteroid material and placing it in High orbit,
and delivering equipment from Earth. Once a processing
plant, factory, and habitat are set up, carbon from car-
5.14 Interplanetary Transfer bonaceous asteroids is used to make carbon fiber for the
Skyhook. Initial velocity capacity would not be as high,
With a network of systems developed around Earth and and so more vehicle propulsion would be needed, but as
the Moon, the next step is to extend the network out to the Skyhook grows, it can reach a wider range of orbits.
Mars and the Main Asteroid Belt. A similar process of Momentum changes are not free, so a substantial power
setting up seed factories and adding facilities is followed supply and thruster set will be needed at the Skyhook.
as before. But since those do not have to be carried along with the
228 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

vehicles who are getting their orbit changed, the propul- from whichever asteroids are already in the closest orbits.
sion can be as large and heavy as needed. Depending on size, the NEO would be mined for materi-
als, or if small, moved entire to the desired transfer orbit.
Optionally a transfer habitat would have a Skyhook at-
Inclination Stations
tached to it to enable added delta-V for arriving or de-
parting vehicles. This also provides an artificial gravity
Near Earth Objects have a limited range of Solar orbit ve-
environment for the habitat. If that is not provided, then
locities in the ecliptic plane by definition. They also have
part of the habitat would be rotating to create artificial
a range of orbit inclinations, which results in a velocity
gravity. Whether to use a Skyhook or not will take more
component when crossing the ecliptic. The Inclination
detailed analysis of the complete transportation network.
Station is a second Skyhook located in the vicinity of
Earth, such as at one of the Lunar Lagrange points, and When the habitats are not ferrying passengers to Mars,
oriented perpendicular to the ecliptic plane. Therefore they are exploiting Near Habitat Asteroids. Just like
it can deliver cargo to and from inclined orbits to reach there is a set of asteroids whose orbits are “Near Earth”,
NEOs with less fuel and mission time, while the first HEO and so easy to reach for mining purposes, there will be a
Skyhook operates in the ecliptic plane to reach Mars and different set which are close to any given Transfer Habi-
the main Asteroid Belt. Additionally, flybys of planets tat orbit. So you can busy yourself producing fuel, set-
can be used to further change orbit inclinations and reach ting up manufacturing, and eventually have a space city
other groups of asteroids. The Station itself will react to there, just one that happens to get close to Mars period-
the average of cargo orbit changes. It will therefore need ically. When it does, you drop off humans and accumu-
some propulsion to maintain position, but that will use lated hardware at Phobos, where you are also building up
less fuel than each vehicle doing it’s own orbit maneuvers facilities, and thence onward to Mars itself.
separately.
Given sufficient traffic, it may make sense to have other
Inclination Stations set up at different tilts to generate dif-
ferent combinations of ecliptic plane and inclination ve-
locity changes. The velocity difference between the var-
ious high orbit Skyhooks will be very low, and release Orbit Characteristics Since we are starting from
from less than their full radius will be sufficient to get be- Earth, we want the Habitat to pass by Earth on a regu-
tween them. Since trip times between them will be short, lar basis. If we set the orbit period to be 1.50 years, it
the first one can have the bulk of the habitat and produc- will do so every second orbit. The orbit will be an el-
tion facilities. The later ones can serve mostly as transit lipse, and the long axis will be 2.62 AU. If the near point
hubs, and not be as built up. of the orbit is at Earth (1.00 AU), then the far point will
be 1.62 AU, which is slightly past the average distance of
Mars (1.52 AU). A velocity change of 3,340 m/s beyond
Transfer Habitats Earth orbit is required to reach this orbit. This comes
from a combination of the HEO Skyhook, Lunar gravity
Transfer orbits to NEOs or Mars will require trip times assist, propulsion on the transfer vehicle carrying crew
measured in months. For human passengers there is the and cargo between points, and possibly a Skyhook at the
risk of exposure to radiation, and also a need for food Transfer Habitat. The Habitat will align with Mars once
and life support. If you expect to make multiple trips, every 7.5 years, and a velocity change of 4,440 m/s is
it makes sense to have habitats permanently in transfer needed at that end to match orbit. Again this would use a
orbits to those destinations. Then the mass of the shield- combination of propulsion, including Mars gravity assist.
ing and greenhouses does not matter that much, as they On early trips the transfer vehicle would need to do more
are not moving once set up. The passengers would use a work, but later a Mars orbit Skyhook can take up more
small vehicle between the HEO Skyhook and the Trans- of the velocity changes.
fer Habitat when it passes near Earth, then ride in the Since once per 7.5 years is not very often, you can place
habitat until it is near the destination, and then again use multiple habitats in a given orbit track, and use multi-
a small vehicle for arrival. Since only the passengers and ple orbit tracks spaced equally around the Earth’s orbit at
cargo need to change velocity the total mass transferred their near points. This will give more frequent opportuni-
per mission is greatly reduced. ties to go back and forth from Earth to Mars. For 80% of
All objects in the Solar System are in motion relative to their orbit cycle the habitats would would be doing mining
each other, and transfer orbits only line up properly when, and construction around the habitat, accessing asteroids
for example, the Earth and Mars are in the right relative in nearby orbits. The other 20% of the time they would
positions. Thus a network of multiple Transfer Habitats also be carrying passengers and cargo for the trips to and
in different orbits will be needed to deliver passengers from Mars, when they happen to line up. There may be a
and cargo to the right destinations at the right time. The better arrangement of orbits and habitats to increase the
source materials to build the Habitats would come mostly fraction of time they can be used as a ferry service.
5.15. MARS DEVELOPMENT 229

5.15 Mars Development orbit (LMO) to Phobos orbit, and whatever suborbital ve-
locity to Mars that produces. The other set would be ca-
Entire books have been written about getting to and de- pable of doing a full velocity transfer to the Martian sur-
veloping Mars. The ones I will refer to as Flags and face. The first could grow into the second over time, but
Footprints exploration missions are similar the Apollo we will look at them as specific design points.
missions to the Moon. They go there, do a bit of explor-
ing, and return, without a supporting infrastructure and LMO to Phobos Skyhooks Orbit Mechanics - The
leaving no improvements at the destination. These are product of a planet’s mass M and the universal gravita-
particularly expensive and wasteful. tional constant G is called the Standard Gravitational
The combined systems discussed in the previous sections Parameter or µ . For Mars the value is µ = GM = 42.837
are a different route to getting to Mars. They build ca- x 10^12. This value is useful for calculating circular orbit
pabilities step by step, each one preparing for the next, velocities by the formula
and generally using machines to prepare the way for hu-
mans. In this section we will discuss the last steps to get √
to Mars, building on the prior technology from our com- GM
bined system. Mars is the most nearly Earth-like planet vo ≈
r
we know of, so we will also mention some ideas for long
Knowing the average radius of Phobos from the center
term development. They would get done, if ever, much
of Mars is 9,377 km, putting that into the formula in me-
later, when Mars and the Solar System are more fully de-
veloped. ters (9,377,000 since we must use all SI units and not
multiples thereof), we can determine the orbit velocity of
Phobos is 2137 m/s. Since Mars has an equatorial radius
5.15.1 Early Development of 3,396 km, then Phobos is 5,981 km from the surface.
For elliptical transfer orbits, where r is the current radius
The early development would start with the Martian from the body center, and a is the semi-major axis, or
moons Phobos and Deimos, since they are easiest to half the long axis of the ellipse, the velocity is
reach, and progress down from there.

( )
Phobos and Deimos 2 1
2
v =µ −
r a
Conveniently Mars already has two fairly large asteroids
in orbit (Phobos and Deimos), giving us a ready supply If we want to transfer from Phobos to 400 km above the
of materials. Phobos alone has a mass of about 10,000 Mars surface ( r = 3796 km ), then the velocities at the
gigatons, which is more than we would need for a very high and low points of the transfer orbit, and at the low
long time. The first step in development is to set up a orbit can be calculate as follows:
base of operations on one of them. This would do min-
ing, processing, and construction of habitats. The com- • Transfer High point: r = 9,377,000 m ; a
position of the Martian moons is not entirely clear yet[1] , = half of high + low altitudes = 6,586,000
but they appear to be similar to CM type Chondrite as- m ; from formula v 2 = 2,632,300
teroids and meteorites. If any particular materials are not m^2/s^2, and so v = 1622 m/s.
found in these satellites, they can be fetched from Near
Mars Objects, otherwise known as the Main Asteroid • Transfer Low point: r = 3,796,000 m ;
Belt, which starts just outside Mars’ orbit. The velocity a is the same as previous = 6,586,000 m
to reach them from the Martian satellites is fairly low. ; therefore v 2 = = 16,065,000 m^2/s^2
and v = 4008 m/s.
Chondrites are known to contain a large amount of car-
• 400 km Circular: r = a = 3,796,000 m ;
bon, so the main goal of the Phobos base would be to
v = 3,359 m/s.
provide cable and other materials for a Mars Skyhook.
Secondary goals are to provide fuel for landers and hu-
man control of surface operations before humans can be The velocity difference from Phobos to the transfer or-
supported on the surface. The short distance of the Mars bit is 515 m/s, and from the transfer orbit to 400 km
satellites relative to Earth allow for real-time control. circular is 649 m/s. These velocities are relatively small
compared to the Lunar or Earth orbit Skyhooks we have
looked at previously, so they would be relatively low mass
Mars Skyhooks ratio. Assuming the tips are at 1 gravity, the Phobos Sky-
hook would have a radius of 26.5 km, and the LMO one
There are two sizes of Martian Skyhooks that make sense would be 42 km. Phobos is tidally locked to Mars, al-
to look at. One set is capable of reaching from low Mars ways keeping one face to the planet, but it is not locked
230 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

in rotation about the Mars-pointing axis, and tidal varia- gravity loss for a total velocity of 1145 m/s. This might
tions from the sun and slight orbit eccentricity cause it to be reduced if the Skyhook reached a lower altitude, but
wobble. Thus the Phobos Skyhook should probably not drag must be carefully looked at so the Skyhook is not
be attached to Phobos directly, but placed nearby. in danger of de-orbiting itself. Without a Skyhook we
Both Skyhooks will have low mass ratios because of their found in the previous section we needed 5.4 units of fuel
low tip velocities. Therefore they would shift their own for each unit of payload returned from Mars surface to
orbits by a large amount when transferring cargo. The Mars escape. Escape velocity is 1.41 times circular orbit
solution is to anchor both of them with a sufficient amount velocity, and this Skyhook has a release velocity of 1.92
times orbit velocity at the highest point, so well above
of ballast mass from Phobos at their center points. The
LMO Skyhook can drop cargo at 3,359 - 649 = 2,710 escape velocity. Thus with this system the mass ratio be-
comes 1.387, and fuel used is 28% of takeoff mass, or
m/s. Mars’ equatorial rotation is 241 m/s, so the relative
velocity to the surface will be 2,469 m/s. To reach Mars 45% of cargo mass. Therefore it uses 12 times less fuel
than without a Skyhook.
escape from Phobos requires adding 884 m/s. Since our
Phobos Skyhook can add 514 m/s, that leaves 370 m/s to An alternate approach that eliminates launch fuel use en-
be done by other means. tirely is to build an 8.67 km tall tower with an electro-
A “Flags and Footprints” type mission would require no magnetic2 or gas accelerator that operates at 6 gravities
propulsion in theory to land on Mars, if it uses all aero- (60 m/s ). This results in 1020 m/s vertical muzzle ve-
dynamic braking. It would need 5 km/s of propulsion locity.
to reach Mars escape velocity on the return trip. With a The working length we previously used for Carbon fiber
base at Phobos and the transfer Skyhooks, no propulsion is 126.4 g-km. Acceleration in the Skyhook varies
is needed is needed to land either. The LMO Skyhook smoothly from 0 at the center to 1.0 gravities at the tip.
drops the vehicle about 20% below orbit velocity and re- With an average of 0.5 gravities times a radius of 846 km
entry will be automatic. The return propulsion would re- the requirement is 423 g-km. Therefore the mass ratio of
quire about half as much velocity to reach orbit - 2.47 the cable is 28.4 times it’s load per arm, or 56.8 for both
km/s. With a permanent habitat at Phobos, there is no arms. That is heavy enough that it may not need ballast
need to escape from Mars, but if you choose to do so, mass at the center to keep from shifting it’s orbit too much
it would require 2.84 km/s. Assuming a chemical rocket in operation. It will still need propulsion to maintain orbit
for both with an exhaust velocity of 3,500 m/s, the mass when traffic going up and down are not in balance. Since
ratio will be 4.17 without the Skyhooks, giving a cargo of the Skyhook saves 4.95 units of fuel on the Martian sur-
14% if the vehicle hardware is 10%. With the Skyhooks, face for each unit of return cargo, in theory it pays for
the mass ratio is 2.02, and landed cargo is 39.4%. So the itself in fuel savings in under 12 flights. In practice it will
Skyhooks allow 2.8 times as much payload per trip. take a detailed design to find out what the system mass
and payback times in terms of mass and cost will be.

LMO to Surface Skyhook A full orbit to ground Sky-

hook will likely not make economic sense until traffic
Mars Surface Systems
grows to a higher level, but let us take a look at a possible
design. Start by assuming a 1000 km high orbit. That
will have a radius from the center of Mars of 4,396,000 Construction Earth-moving equipment will be needed
m. From the above formula we calculate the orbit veloc- for a number of purposes. The Mars surface is not pro-
ity is 3122 m/s. Subtracting 241 m/s for the rotation of tected from radiation like the Earth is, so long term habi-
Mars gives a relative velocity of 2881 m/s. If the tip is at 1 tats would need to be protected by a layer of soil. Landing
gravity centrifugal acceleration, then the Skyhook radius areas will need to be flattened, and protective berms built
will be 846 km, and the tip will become motionless when around them so exhaust plumes don't sandblast nearby
the Skyhook is vertical 154 km above Mars mean surface equipment. Once cargoes are delivered to the surface,
level. That should be high enough to avoid significant they will need to be moved, lifted, and assembled, so de-
atmosphere friction. The Mars Global Surveyor space- vices to do those tasks will be needed. Most of the site
craft used a no-drag holding orbit at around 175 km low- preparation will likely be done by remote controlled ma-
est point, and active aerobraking between 120 and 135 chines.
km. It did so when moving between circular and escape
orbit velocities of 3,500 to 5,000 m/s. So the Skyhook
with near zero velocity at the lowest point should not see Power Supply Solar panels are a viable power supply
much drag. on the Martian surface. It is rarely cloudy, aside from
Pavonis Mons is a 14 km tall mountain on the Martian dust storms, and the atmosphere is thin, which partially
Equator. So a vehicle wanting to reach the Skyhook from compensates for the greater distance from the Sun. When
there would need to climb 140 km vertically. This re- larger amounts of power are needed, then radio-isotope
quires a vertical velocity of 1,020 m/s plus about 12% or reactor devices can be added.
5.15. MARS DEVELOPMENT 231

On-Site Propellant Manufacture Producing propel- protects the atmosphere from being slowly stripped off as
lant on the surface of Mars has been studied extensively, solar wind and other particles hit the upper atmosphere.
since it lowers the mass brought from Earth for a “Flags Short of stirring up the planet’s core, there may be some
and Footprints” mission. If we already have a robust other ways to generate a field. The practicality of any of
orbital mining and processing capability and Skyhooks them is yet to be determined:
in place to deliver cargo, there may not be much ben-
efit in early production of fuel locally vs delivery. The
economics of doing so will need to be examined. For • Run one or more superconducting cables around
portable power, such as in moving vehicles, and for rocket lines of latitude, which, like any current-carrying
propellant to reach a Skyhook on a return mission, an wires, will generate a field
Oxygen/Methane fuel mix is a reasonable combination.
Once sufficient need for fuel exists, producing it locally • Place some number of iron-nickel asteroids in Mars
will make more sense orbit and magnetize them, and point their fields in
the same direction.
Linear Accelerator Pavonis Mons, which is located
on the Martian Equator, has a slope about 175 km long, • Mars is red because there is a lot of iron oxide on
which rises about 6.5 km. If large amounts of cargo need it’s surface. Extract the iron, and magnetize it. You
to be delivered from Mars, a gas or electromagnetic accel- might be able to use the iron for other purposes at
erator can be used here. If the full slope is used, orbital the same time as it being a magnet.
velocity can be reached with human-tolerable accelera-
tions (3.6 gravities). This would not be an early system,
since sufficient traffic is needed to justify such a large in- Magnets to make the magnetic field have fewer ways to
stallation. Another option is a centrifugal catapult on top break than superconductors, but if the superconductors
of the mountain for early cargo launch. work 99% of the time the other 1% doesn't make much
difference to long term atmosphere loss. Some leakage of
It is quite feasible to build a rotating space elevator the atmosphere will still happen because Mars is a smaller
(Rotovator) in orbit, coupled to a linear accelerator on planet than the Earth, so it is easier for atoms to escape.
Pavonis Mons ( http://upload.wikimedia.org/wikipedia/
commons/1/12/Pavonis_mons_topo.jpg ). You have 60-
120 km of ramp space, and no atmosphere to speak of,
so at 3 g’s and 60 km you can reach half of Mars orbit Greenhouses
velocity, and the Rotovator provide the rest.
If you want to eliminate leakage, and bring up the pres-
sure to breathable levels without importing a planet’s
5.15.2 Long Term Development worth of atmosphere, you can use greenhouse domes. If
you really need the space, you can extend the domes to
Often the phrase “Terraforming Mars” has been used in cover the entire planet bit by bit. To create Earth sea level
the past. This is not a good phrase because it means pressure on Mars, a pressure balanced dome would con-
“Make Mars like Earth”. Because of orbit and mass dif- sist of 10 meter thick quartz, glass, or equivalent, which
ferences, we cannot make Mars just like Earth, nor do I you extract from Mars surface material. Lighter domes
think that should be the goal. I prefer the word “human- tend to float up, as the internal pressure is higher than the
ize”, meaning making it more suitable for humans. It may surrounding air. In that case they need to be tied down
also mean modifying humans to better suit the Mars en- so they don't float away. A very large or planetary dome
vironment (like the lower gravity). Large scale changes doesn't need much to hold it up, just some towers or ca-
to Mars should be delayed till after we have a firm idea if bles to keep it from moving sideways.
there is any native life on the planet, and even then done
with due consideration and forethought. They should also You can design the clear material like armored glass to be
be delayed until there are enough people on Mars to jus- resist damage, and ten meters of anything is pretty hard
tify the large-scale projects. So what follows is more to to break. But anything can be broken, so a lot of thought
answer what is possible from a technical point of view, needs to go into how to deal with damage. As a green-
and less to say “I urge you to do all these”. house, you can take advantage of the “greenhouse effect”,
which is the trapping of infrared heat radiated back from
the ground. You can specifically select the glass type or
Magnetosphere add coatings to trap infrared. You also want to block So-
lar UV radiation, which is not blocked by the Martian at-
Mars lacks a strong magnetosphere - a magnetic field mosphere. On Earth the greenhouse effect is a problem,
around the planet that traps and diverts charged particles since we don't want the planet warmer than it already is.
from space. The Earth has one due to the magnetic field On Mars it’s a solution, since it’s too cold for us there at
generated by our planet’s core. A strong magnetosphere the moment.
232 CHAPTER 5. PART 4 - COMBINED SPACE SYSTEMS

Full Atmosphere For smaller asteroids staying on the surface will be more
of a problem than getting on and off. For example, the
If you find living under a dome objectionable, you would 35th largest asteroid by diameter is 9 Metis, which has an
need to provide a full atmosphere at a breathable level. equatorial radius of 170 km and a mass of 1.47 x 10^19
That is a very big job because planets are large. On Mars kg. This gives a surface gravity of 0.034 m/s^2 (0.34%
you need to provide 25 tons of atmosphere for each and of Earth). The rotation period is 5.08 hours, which give
every square meter of the planet, or 3.6 million billion a rotation velocity of 58.4 m/s and a centrifugal acceler-
tons total. That’s to provide one Earth atmosphere pres- ation at the equator of 0.020 m/s^2. So the net appar-
sure. If you are satisfied with less oxygen (similar to ent gravity is only 0.014 m/s. Indeed, the orbital velocity
mountains on Earth), and a different mix for the rest of is 76.0 m/s, so it only takes 17.6 m/s ( 39 mph ) added
the air, you can get by with somewhat less. Despite it’s velocity to reach orbit. Therefore humans or low speed
distance, the easiest place to get enough nitrogen might machinery can toss things into orbit, and a firm anchoring
be the Kuiper belt, which is outside Neptune’s orbit and method will be needed to not do it accidentally.
which Pluto is a part of. You could use a “reverse gravity
assist” from Neptune to drop the material into the inner
Outer Moons and Minor Planets
solar system. Nitrogen is rather scarce in the inner solar
system, and getting it from anyplace with a deep gravity
Kuiper Belt
well (like Earth) takes a lot of work. Some of the outer
moons might have enough ammonia (NH3).
A few points about the dwarf planet 136108 Haumea:
Haumea is massive enough to be in hydrostatic equilib-
rium, and therefore classed as a dwarf planet. However,
References
the short rotation period (3.9155 hours) means it is not
round, but rather ellipsoidal, with a long axis about twice
[1] Rivkin et al MRO/CRISM Observations of Phobos that of the short axis. Circular orbit speed at the long ends
and Deimos 2009
is ~527 m/s, while the tips themselves rotate at ~428 m/s.
So only ~99 m/s velocity change is needed to land or take
off from it, one of the lowest numbers for a large Solar
5.16 Later Projects System object. If Haumea retains any sort of atmosphere,
it would tend to be wedding-band shaped around the short
axis. Gravity would vary significantly from the long ends
It is too early to make definite plans for development be-
to the short axis.
yond Mars, but we can list some of the options in this last
step. These are approximately in order of distance and A challenge for the Kuiper Belt and farther locations is
difficulty, but there is no requirement to do them in any supplying enough solar energy to operate. Civilization
particular order. on Earth consumes about 2.7 kW/person, and we would
expect a higher number for space locations, both due to
higher standard of living, and the need to artificially do
Main Belt Asteroids some things handled by natural processes on Earth. Let us
assume 20 kW/person is needed, system mass is double
Development of the Main Belt between Mars and Jupiter that for the ISS, or 150 tons/person, and half is devoted
is an extension of development in the Near Earth and to solar collection. If magnesium-aluminum reflectors 1
Earth/Mars transfer regions. Because all three regions micron thick are used to concentrate sunlight, they will
2
have a range of eccentricities and inclinations there is no have a mass of 2.4 tons/km . So we are allowed a maxi-
2
distinct boundary between them, but we arbitrarily assign mum of 31.25 km of reflectors/person. For a net power
the area between Mars and Jupiter as the Main Belt, and of 20kW at 1/3 efficiency, we need 60 kW of sunlight.
Jupiter’s orbit and beyond as the Outer Solar System re- At Earth, solar flux is 1.366 kW/m^2, thus needing 44
2
gion. m . Since we are allowed 711,500 times this area, and
solar flux falls as the inverse-square of distance, we can
The largest Main Belt object is the dwarf planet 1 Ceres.
sufficient energy out to 843.5 AU. Beyond this distance,
Equatorial orbit velocity is 359 m/s, and equatorial rota-
operations would limited to low power situations, or re-
tion velocity is 94 m/s. Thus to reach orbit requires 265
quire other sources, like nuclear or beamed energy.
m/s net. This velocity can be reached by a mild steel cen-
trifuge and easily by any advanced material. Therefore
cargo delivery from any other Main Belt asteroid, all of Gas Giants
which are smaller than Ceres, does not require any rocket
propulsion. A 1.0 gravity Skyhook would be 7 km in ra- Depleting Jupiter’s Radiation Belts - Placing sheets of
dius and allow crew and equipment to be landed and take material in the dense parts of the radiation belts may in-
off at low acceleration, and a cost of 0.5% of net mass tercept enough of the particles to lower radiation levels
flow in reaction mass to maintain orbit. there.
5.16. LATER PROJECTS 233

Oort Cloud and Beyond

Slow Interstellar People assume that a “starship” will

be a metal can with big engines on the back. Imag-
ine colonizing a long period comet, one of the ones that
came from the Oort cloud, and is heading back out there.
Comets are made of a mix of ices (water, methane, am-
monia, CO2, etc) and rocky materials. If there is not
enough metals, get one of the metallic asteroids to match
orbits with it. Then build your colony out of the materials
there. Comets range in size up to 50 km in diameter, so
there is plenty of stuff to build with.
The Oort cloud is many times the distance of the Earth
from the Sun, and the velocity needed to get the comet to
leave the Sun and head for another star is very small. All
the ices have some amount of hydrogen, and thus deu-
terium, which means if you know how to build fusion re-
actors, you have power for a long long time. It will be a
long trip, but you have a whole city worth of space to play
in, with occasional side trips to other comets in the Oort
cloud.
There are an estimated trillion comets out there, some
will be along your route, more or less. The average spac-
ing is something like 6 AU, about the distance to Jupiter.
So you can in theory seed other comets as you pass by
with new colonies. If some people feel like it, they could
head back to the Sun, the velocities are low enough to do
that.
The requirements for this kind of slow star travel are fu-
sion power, and knowing how to build permanent habitats
in space.

Starlines Gravitational lensing occurs around every

massive object. In fact, measuring the bending of light
during a solar eclipse 100 years ago was the first proof
of relativity theory. For the Sun, the light bent from all
sides comes to a focus at distances greater than ~540 AU.
The focus is not to a point, but rather a radial line. This is
because photons that miss the edge of the Sun by a larger
distance are bent less, and thus focus farther away. Every
star in the sky therefore produces a line of focused light
on the other side of the Sun, and thus we refer to them
as Starlines. Every other star also produces a pattern of
starlines surrounding it, forming a network of lines filling
interstellar space. These lines may prove useful for power
and propulsion for distant missions.
Chapter 6

Part 5 - Design Studies

6.1 Design Studies 6.1.3 Studies in Work

These are studies which have been started, but have not
6.1.1 Introduction reached completed or full status. They are given their own
sections, with as many pages as needed by their length,
In the previous part of the book we give an extended ex- and approximate completion status is noted here.
ample of a program for Human Expansion. The narra-
tive there is intended to show why complex programs ex-
• Conceptual Design for Human Expansion (Sta-
ist, how they are organized and designed, and how they
tus: approx 20%) - This study carries through a
evolve. For that purpose, a summary of design studies
first stage analysis and design of the program as
and decisions based on them is sufficient. In this fifth
a whole. The principal purposes are (1) to deter-
part of the book we include the design studies in their full
mine if such a program is desirable over existing
detail. This is to show by example how the analyses, cal-
programs, and (2) establish one or more baseline
culations, and decisions are done. We also include incom-
designs for the next stage of work if the program
plete work-in-progress studies. Readers are encouraged
is found desirable. Because this study looks at the
individually or in teams to add to and improve these stud-
whole program, it is the first one to be worked on.
ies. This gains skill and experience applying the methods
described earlier in the book, and in working in teams,
• Seed Factory Concept Development (Status: ap-
which is how most real projects are carried out. Contribu-
prox 10%) - This study attempts to formulate con-
tions to studies can also serve as resume items when look-
cepts for a “Seed Factory”, a starter set of equip-
ing for paid work. which are too detailed or too incom-
ment that both outputs more equipment to expand
plete to include in the main narrative of the book. Any
itself and also useful products. It is part of a sepa-
added results from these studies will be incorporated into
rate Wikibook on that topic.
earlier parts of the book. As an electronic book, it can
be constantly improved, rather than the somewhat static
condition of paper textbooks. 6.1.4 Study Ideas
The list of studies below is organized first by level of study
completion, then by expected order of use within the pro- These are ideas for future studies which have not been
gram. started yet. They do not have their own section yet, but
are merely described here.

6.1.2 Full Studies Starting Location Trade

We use the term “full” rather than “completed” because The two nearest objects in velocity terms beyond Earth
no engineering work is every truly complete. As time orbit are the Moon and Near Earth Objects (NEOs). If
passes, assumptions made in the study will be affected by you want to obtain materials or set up production, the
real-world changes, and new technology and ideas will get question is which to start with? The options are Moon
developed. As a consequence, the study could be updated first, NEO first, or both in parallel.
or revised to reflect these changes. For an actual engi-
neering project, however, you must reach a conclusion or Composition
make a decision, and proceed to more detailed design or The composition of the Moon and NEOs is different. The
production. This section includes studies which are suf- Moon does have some Carbon in it, around 100 parts
ficiently complete that such conclusions or decisions can per million. Chondrite asteroids are approx 0.4% carbon
be made from them. At present (Oct 2012) none of the (4000 ppm). So given 40 time richer “ore”, and the abil-
studies have reached that point. ity to run your processing plant 100% of the time vs 50%

234
6.2. HUMAN EXPANSION 235

on the Moon (lunar night), you likely want to mine an as- The full details of our reasoning, calculations, and deci-
teroid to get the carbon to *build* a space elevator. Like sions are too long to include in our discussion here. We
for terrestrial mining, location of the mine and richness will summarize them here, and refer readers to the full
of the ore determine where you want to mine. Human Expansion Design Study for the details.
Access The general process flow in Conceptual Design starts with
general description and goals, which are noted in the next
section. Following that we perform an analysis step to
6.1.5 Discarded Studies develop these goals into more detailed and quantitative
statements. At present this has been done for the program
Some studies will turn out to be unnecessary, merged with requirements and evaluation criteria. The functional anal-
other studies, or otherwise determined to not be useful. ysis to define more detailed program elements has been
They are listed here to have a record of work done previ- started, and later parts of the Conceptual Design process
ously. This is different than a completed design study for are incomplete. Candidate designs to fulfill the program
an option not chosen. In that case the study was useful in elements are collated in the remainder of Part 4. They
making the choice among alternatives. should be considered as pieces which might fit into the
overall program, but have not yet been linked, or com-
No discarded studies are in this category yet.
pleted individually.

6.2 Human Expansion 6.2.2 Program and Customer

We begin by identifying who is the Customer for our
6.2.1 Conceptual Design Stage program and what their needs are. Any proposed pro-
gram involving many people needs sufficient motivation
The first stage in the life cycle of a well-designed pro- for why to do it instead of the alternatives of doing noth-
gram is Conceptual Design. The goals of this stage are to ing, or continuing with what is being done now. If you are
identify the needs to be satisfied, the selection criteria to building a system entirely by yourself, then “I want to do
be used, an initial concept for the major system elements, it” is enough, but for larger systems you typically have to
and how they will be built, operated, maintained, and dis- convince end customers and developers of its value. The
posed. The alternatives to building a new system are to motivations can be couched in terms of general goals, or
do nothing or to continue with existing projects and pro- financial benefits, or simple efficiency, or even “I have
grams. We therefore assess our system concept against a gun, do this or else” for totalitarian societies, but they
those alternatives and decide whether to go further or to must be there in some form. When the end customer is
stop. We do this using the selection criteria previously unaware of a program’s value, they need to be informed
chosen. about it, preferably not at the point of a gun. Thus how
For our example program we give the motivation in terms to format and deliver the results needs to be considered
of general goals first, with potential reasons to change along with the technical results.
from current space programs. Not every goal will have Customer vs Designer - Customers can be internal or
the same importance to every person or organization. As external. The former case is when the same organization
long as a sufficient number of goals apply to them, they is both the user and designer of a system. The latter case
would have a reason to support the project. Following the is the more common one for large systems, where design
goals we describe our design approach in general terms. specialists develop systems for another entity than them-
The rest of Part 4 will go into more detail of the design selves. For our future program then we must ask who is
process and results. setting the program goals and is thus the customer, and
[text inserted from elsewhere to be merged] who is doing the design work? For now both are actu-
ally the editors of this page, but we will assume a non-
Conceptual Design profit foundation will continue the initial design work us-
Conceptual Design is the first stage in the life cycle of a ing open-source methods. As a foundation they would set
program or system. We determine needs and goals to be goals for the benefit of civilization as a whole, and pub-
satisfied, what criteria to use in selecting among designs, lish the results of their work publicly. Assuming good
and a design concept for the major parts of the program, enough results are reached, then fund raising for later re-
including how they will be built, operated, maintained, search, start-up of commercial entities, or promotion of
and disposed at end of life. We compare our new pro- government funding would be used for later stages. The
gram to existing programs using the same selection crite- ultimate customer is therefore civilization as a whole, and
ria. If the new program is “better” by these measures, we the designers are part of an open-source non-profit foun-
recommend going forward to the next design stage. If it dation. At present this is the Wikimedia Foundation ed-
is not better, then we stop and wait for circumstances or itors of this open source book, but that may transition to
technology to change, or try to devise alternate concepts. a specialized organization later.
236 CHAPTER 6. PART 5 - DESIGN STUDIES

Customer Acceptance - Some people or groups will be currently in orbit, all of us live on the
negatively affected by a new project, in particular those Earth, it’s a good idea to understand it
associated with older existing projects. Other people sim- better.
ply do not like change, or are averse to the risks of new or
untried methods. Yet others have a preference or aversion • Reduce Hazards from Space - There are
to particular designs separate from their technical merits. hazards in space such as solar flares and
Finally, some methods are simply new and unfamiliar and asteroids which can affect us here on the
not considered for that reason. A system designer has to ground. In order to prepare for or pre-
understand these human elements and respond to them, vent these hazards, we first have to know
and not assume that the best technical answer will be ac- their magnitude and characteristics, fol-
cepted just because it is the best technical answer. lowed by developing methods to prevent
Program Constraints - Beyond technical design, there or deal with the hazards.
are also outside factors, such as available funding or re-
strictions on technology transfer, which affect the course • Increase Biosphere Security - There is
of a program. These outside factors need to considered only one biosphere right now that we all
in addition to the technical and human ones. depend on, and it is prone to natural
and man-made variations (ice ages, CO2
changes, volcanoes). Even the Interna-
6.2.3 Program Goals tional Space Station relies on food and
other supplies which come from Earth.
Human civilization desires, as the sum of individual de- Observing from space helps us under-
sires, to survive over the long term, with lower risks in- stand how the Earth varies better. In the
dividually, and to flourish materially and socially. There- long terms we should also set up backup
fore we state the following individual program goals to biospheres for survival reasons in case
satisfy these desires. We group them into those that ben- something catastrophic happens to the
efit civilization in general, and then those that apply to only one we have.
individuals or groups of smaller size. By the end of the
conceptual design stage, these will be refined to final goals • Expand Material and Energy Resources
and specific numerical objectives. - Civilizations require materials and en-
ergy to function, and there are vast ma-
Civilizational Goals terial and energy resources beyond the
Earth. We could move heavy industry
Our earthly civilization should be motivated to visit, de- and population off the Earth to reduce
velop, and expand into space for a number of reasons. our impact to the planet.
These reasons can stated in terms of individual goals. A
future program or programs ideally can satisfy more than • Long Term Survival - In the long term
one of these goals at once: the Earth is doomed as the Sun continues
to get hotter over its life, and eventually
turn into a red giant and swallow it. Long
• Improve Life on Earth - Attempting new
before then, it will become a poor place
and challenging things tends to discover
for life because the Sun increases it’s lu-
new knowledge and develop new tech-
minosity over time, eventually overheat-
nology. Such developments can first
ing the planet. Our choices are planetary
be applied on Earth to improve our life
scale engineering to deal with the even-
here, particularly in the areas of en-
tual heat, or moving elsewhere. Either
ergy, resources, automated production,
way involves space in some way.
and closed ecologies. With experience
gained on Earth we can then use these de-
velopments in space. Localized Goals

• Understand the Earth Better - We learn In addition to goals which apply to civilization as a whole,
more about how the Earth works by look- there are motivations which apply to particular individuals
ing at it’s current environment in space, or groups. These more localized goals can overlap with
because the Earth is not isolated in the the more general ones listed above:
Universe. We also learn from looking at
other examples of planets and environ- • Increased Choice and Freedom - Free-
ments and how they evolve. Except for dom of choice is restricted in many ways
the one part per billion of human kind by an occupied and relatively crowded
6.2. HUMAN EXPANSION 237

Earth. You cannot just set out and start This can only be done with a sufficiently detailed design
your own community, with your own that considers all of the cost elements.
rules for living, because all the land area
of the Earth is already claimed by some • Existing program costs are much higher than what
government and functions under their is possible:
rules. Dangerous or dirty experiments,
whatever their scientific value, have to With all the many billions of dollars spent by governments
be restricted, and some experiments, like and private industry on space projects, we have only man-
terraforming, simply cannot be done on aged to get about 1000 useful satellites and a half dozen
the Earth. At least to start with, space people working in space. This is because past programs
is unoccupied and will not be crowded had launch cost measured in their weight in precious met-
for a long time, so earthly restrictions als. For example, the Atlas V is listed as delivering 20
are lifted. Space will impose it’s own tons to Earth orbit for US $110 million. The cost of
restrictions due to environment and re- $5,500/kg comes to $171 per troy ounce, vs a mid-2012
sources, but they will at least be different price of $1600 for gold. The new Falcon 9 commer-
restrictions. This widens the total range cial rocket reduces that somewhat to 13.15 tons for US
of available choices. $54 million, or $128/oz ($4,100/kg), and the upcoming
Falcon Heavy is quoted by the manufacturer at 53 tons
• Increased Opportunity - Unlike Earth, for US $83 million, or $49 per ounce ($1,566/kg). This
where almost everything is already compares to an average of $30/oz for silver in the first
claimed by someone, there are many half of 2012.
unclaimed resources in space. Lack of The raw energy cost to get to Earth orbit, which is about
acquisition cost from a previous owner 33 MJ or 9.25 kWh/kg, works out to about $1/kg at typ-
means a wider opportunity for gaining ical retail cost for electricity. This is slightly higher than
wealth, but of course not a certainty of the price of Carbon Steel as of 2012, or roughly the re-
it. The ratio of available space resources tail price of potatoes per kg. Efficient shipping to or-
per person will start very high, partly bit should in theory run several times raw energy cost,
because there will not be very many peo- or roughly the cost of cheese or priority mail packages.
ple at first, but also because the absolute The ratio of upcoming launch costs and theoretical cost
amount of matter and energy available is still about 400:1, indicating there is much room for im-
in space exceeds that on Earth by a wide provement. As long as launch costs are measured in their
margin. This high relative availability of weight in precious metals, rather than cheese, it should
resources creates new opportunities for be evident not much will be done in space. One billionth
those who wish to exercise them. of the world’s population being in orbit counts as “not
much”.
6.2.4 Program Benefits
• High costs are driven by launch technology:
In theory the above goals could be met by existing space
Conventional rockets, which carry all their own fuel, can-
programs. So in addition to stating our goals and the rea-
not get really cheap and efficient because the energy to
sons behind them, we have to demonstrate why a new sys-
reach Earth orbit is about twice the energy in chemical
tem is better than just continuing what is being done now.
rocket fuel. Therefore you have to burn a lot of fuel to
This is the intended result at the end of the Conceptual
get a smaller amount of fuel halfway to orbit, which in
Design stage. Where we are now, which is the start of
turn you burn to put an even smaller amount of cargo
that stage, we can only list the potential benefits identi-
into orbit. So after subtracting fuel, fuel tanks, and rocket
fied so far. Additional ones may be identified later, and
engines, about 3% of what sits on the launch pad is the
some of those listed below may not prove favorable once
delivered cargo. Even worse is that the hardware is ex-
the concepts are developed further.
pensive aerospace-grade hardware and is often used only
once and tossed away. By comparison, airplanes use the
New Systems Can Lower Cost Dramatically: same kind of aerospace grade hardware, and so cost about
the same $/kg to build. But being more efficient (they get
If we stay with existing programs, we will get existing oxygen from the air, rather than from a tank) they carry a
planned costs. The planned costs include future projected higher percentage of cargo, and the airplane is used many
cost reductions of systems already in development. A new thousands of times, thus making it dramatically cheaper
system has the potential to reach even lower cost. The per trip.
following argument establishes the potential gain is very
large. Therefore new systems should be examined to find • Launch cost and lack of resource use drives high
out if they can actually reach some part of that potential. total cost:
238 CHAPTER 6. PART 5 - DESIGN STUDIES

The high cost of everything else in space is partly because development effort, and thus more than average opportu-
of the high cost of getting to orbit, and partly because ev- nity for it. There have been attempts to quantify spinoff
erything you need in space is currently brought from the benefits specifically from NASA programs. While we
ground. Since shipping is so high, designers spend a lot of cannot put an exact value on spin-off technology, we can
effort making the cargo lighter by spending more time re- make an allowance for some amount as part of the reason
fining the details, and by using more exotic, but stronger, to pursue new systems over current systems.
materials. That extra effort makes the cargo more ex- When space systems are intentionally developed with an
pensive also. Since you optimally spend money making it eye to Earth applications it becomes less of a spin-off and
lighter to the point where each kg saved costs the same as
more a matter of design intent. This is referred to as dual
launch cost, cargo cost tends to be proportional to launch use or multi-use technology. For example, solar cells
cost. If you lower the cost of launch, then it will also tend
were first used on a significant scale to power satellites,
to make whatever you are carrying cheaper (and heavier). but now there major application is on Earth.
The other important cost factor is the practice of bringing
everything from Earth, and thus paying shipping on all of • Expanded Markets:
it. If you could get some of needed supplies from space
and recycle more, the amount that has to be shipped to
Significantly lowered cost to do space projects can expand
do a project or mission could be cut drastically. There
existing markets for things like communications satel-
are plenty of energy and material resources in space if
lites, or open new markets such as asteroid mining for
we learn to use them, and for many of them the shipping
rare metals. Increased supply and lower cost for these
energy is dramatically lower than from Earth. The combi-
metals can prompt new uses for them which are not eco-
nation of launch cost, cargo cost, and bringing it all from
nomic at present. Therefore in addition to the direct ben-
Earth results in very high total project cost.
efits of lower cost, some estimate should be made for the
secondary effect of expanded or new markets in the total
• A potential improvement of 20,000 times is possi- value of a new system.
ble:
• Optimism for the Future:
Even an ideally efficient launch system will still have costs
for development and operation, therefore will not reach Fear of loss is twice as strong an emotion as opportunity
raw energy cost. We will assume that 4 times that cost, for gain. If the Earth is seen as a finite, closed, zero-sum
or $4/kg is a practical lower limit. This is 400 times lower world then fear of the future can dominate. In that view
than the quoted price of the Falcon Heavy. Past studies people might take less risk and try to hold on to what they
have estimated that 98% of space systems could be made have. This perception is not reality - from an engineer-
of local resources, and the remaining 2% would be rare ing point of view the Earth is open in energy and mass
or difficult to make, and thus more economical to trans- flows. However, perception can affect actions even if in-
port from Earth. This gives a 50 times reduction in the correct. An understandable path to opening the resources
amount of cargo that needs delivery. Combining the cost of the Solar System can change a pessimistic world-view
and traffic improvements gives a 20,000 times total po- to one with hope. There are other ways to change nega-
tential improvement. That level of improvement will not tive perceptions of the future, such as the development of
be reached soon, and perhaps never, but the potential is so affordable, clean energy sources. Opening space, though,
large it definitely justifies some engineering effort. How directly refutes the idea of a closed Earth. A more opti-
much is worth spending to get an improvement can be mistic world might be less fearful of new technology and
judged by the US $290 billion/year currently spent world- more willing to invest in long term projects.
wide on space-related projects.

Continued on page 2
Developing New Systems Has Side Benefits:

• Technology Spin-off: 6.3 Human Expansion 2

Technology developed for one purpose or project often back to page 1
finds uses in other areas, a process called Spin-off, or
technology transfer. One example is biochips for medical
testing which use the technology developed for electronic 6.3.1 Selection Criteria Approach
microcircuits. Although there are many other such ex-
amples, they are unpredictable in individual cases. Space Having set the initial goals we want to achieve, we now
projects are not unique in generating spin-off technology, want to start developing a design that best meets them,
but they typically have a large amount of research and and evaluate how it compares to existing systems. One
6.3. HUMAN EXPANSION 2 239

part of this is turning the vague term “best” into an ob- Number of Criteria Any parameter which has value
jective measurement method. Complex systems such as to the end user/customer can potentially be used for de-
these will have many parameters which can be used to sign selection. Thus they can cover the full range of goals,
compare alternatives. These parameters will have dif- requirements, and design features. In practice, some are
ferent units of measurement. Since different features of more important than others, or only are relevant at certain
a design, like cost and performance, cannot be directly levels of the design. A very large set of criteria is diffi-
compared, we convert them to a unified numerical scale, cult to apply because you have to evaluate all of them for
and can then choose whichever option has a higher score every design option. So the selection criteria are usually
on that scale. The conversion formulas represent a math- limited to the more important ones. Whichever criteria
ematical model of what is desirable to the Customer. are chosen, there should be a clear description showing
In other words, it is an objective model of “best”. This the relationship to customer desires and how the scoring
model will evolve through the conceptual design process. formula is derived. We will give a couple of examples of
At this point we can only start to work on it. cost criteria to show how this is done. At this point they
are not final selection parameters, merely examples.

Example of Cost Criteria We have assumed that civ-

Subjectivity of Criteria The modeling and evaluation ilization as a whole is who would be paying for the pro-
is an objective process. The choice of what parameters gram development as a whole and benefit from the results.
are to be measured and their relative importance, how- Thus they are the ultimate customer from the Systems
ever, is subjective. It derives from human needs and de- Engineering standpoint. One major reason for choosing
sires which are not set objectively. In addition, humans a new design over current ones is cost, which is a measure
often do not know what they would want because they are of the resources input to the program. So we will use de-
uninformed on a subject. In our present program, the ul- velopment and operating cost as our example criteria. In
timate customer is civilization as a whole. Most people both cases, lower is better.
do not have enough interest or information to know what
parameters they would value. Therefore the immediate • Affordable Development
customer must act as a proxy who expresses their prefer-
ences for them. The immediate customers would be the
Existing space programs are already quite substantial, in-
ones in direct contact with the designers. For now that
dicating they are valued by society. Worldwide, gov-
would be contributors to a research foundation, and the
ernment space agency budgets total US $33 billion as
program designers themselves. At this stage they cannot
of 2011. The Satellite Industry Association estimates
educate and poll all of civilization to determine what they
global space industry in 2011 of US $290 billion. Indus-
would want. Instead they must estimate what they would
try commonly sells to government programs, so to avoid
want if they were well informed and asked their opinion.
double counting we only include 30% of government bud-
As an example of this proxy method, the ultimate cus-
gets as unique. This leads to a total of $300 billion/year
tomers for a smartphone are the people who will eventu-
for all space-related activity. Present value converts an
ally use them. Since you don't even know who those peo-
annual flow of funds to an equivalent single amount. The
ple will be, you cannot ask them what their preferences
current Price/Earnings ratio of the S&P 500 stock index
are. Therefore the management, marketing department,
is 12, implying a present value for current space programs
and engineering department act as a proxy for those cus-
of US $3.6 trillion.
tomers and make their best effort to determine what their
preferences would be. A new or modified program should not greatly exceed
what people apparently are willing to spend. Therefore
we take the net development cost of a new combined sys-
tem divided by the present value of current space pro-
Types of Criteria Some criteria are a clear yes/no or grams as a measure of affordability. Small values are
pass/fail type. If the design fails to meet such a criterion, good, indicating not much extra cost. Negative values
it has no value in the eyes of the customer. This type are are even better, indicating cost savings. Some net cost
not modeled or scored, since the score if not met would is acceptable if the benefits exceed the cost. Net devel-
be zero. Instead they are included as well defined pro- opment cost here is the discounted present development
gram requirements that must be met, and other parame- cost of the system, including revenue generated. Thus it
ters allowed to vary as needed to meet the “hard” (fixed is the amount of money you would need today to finance
and unchanging) requirements. Usually this type repre- all the future development at the maximum point, after
sents fundamental program goals and objectives. Other accounting for any positive income that can cover later
parameters, such as cost, have an incremental level of de- costs.
sirability - each increment of lower cost is more desirable. We do not have any absolute measure to convert net de-
Such parameters can be scored on a sliding scale from de- velopment cost to an evaluation score. In the absence of
sirable to undesirable. one, we will assume civilization is willing to spend the
240 CHAPTER 6. PART 5 - DESIGN STUDIES

same fraction of total output on space programs. Since a ratio of 148:1, has a natural log of 5 and is scored at
world real GDP grows about 3% annually, we will set 100%. Other cost reductions are scored at 20% x ln(cost
an increase of net development cost based on the present reduction).
value multiplier of 12 as setting an equivalent time hori-
zon of 12 years. This gives 12x3% or 36%, as a nominal
value, to which we assign a midpoint score of 50%. We 6.3.2 Program Requirements Approach
will assign 0% net development cost a score of 75%, and
36% decrease in cost will score 100%. A 108% cost in-
The next part of developing the design is defining require-
crease will score 0%, and the scale will extend beyond 0
ments. Program requirements will state how the general
and 100%. The score values are arbitrary. What mat-
program goals will be met in terms of measurable fea-
ters is how they relate to the scores of the other selection
tures, parameters, and values. They will be documented
criteria. This implicitly defines exchange ratios between
at the end of the conceptual design stage. At this point
criteria, like “a 1% increase in thrust is worth $5 million
(the start of conceptual design) we can start by identify-
in development cost”.
ing categories of requirements under each of the program
A better scoring of development cost could be obtained goals. The requirements analysis step will then examine
later by surveying people after explaining what the pro- and select individual requirements from them.
gram benefits would be. We can also do a sensitivity anal-
ysis on the scoring. This is calculating how the results of
the conceptual design change as a result of changing the Improving Life on Earth
selection criteria or their scoring. If you get the same de-
sign result for a wide range of scoring criteria, the design This program goal was listed first because, for the time
is said to be Robust or Insensitive to changes. This is period we can reasonably plan a program, the majority
desirable. If small changes cause very different design of humans will live on Earth. The quality of our life here
results it is likely not an optimal design, or more work is therefore of high importance. Historically, and in the
needs to be done to distinguish alternatives or reduce un- near future, almost all of the people working on space
certainty. programs, the offices and factories, and sources of mate-
rials will also be on Earth. If new methods and technolo-
• Low Recurring Costs gies are developed to reach challenging space mission re-
quirements, they can feed back to the toolbox civilization
The previous measure reflects maximum net develop- has in general to work with, and thus improve life here.
ment cost at any point in time. A good conceptual de- An example of where this has already happened is in the
sign would also have a low ongoing operating cost. The Systems Engineering methods developed to manage com-
major current recurring cost for space projects is launch plex aerospace projects. These methods can be applied to
to Earth orbit. Total launch cost is the product of how all forms of complex projects on Earth.
many kg you need to launch and cost/kg of the transport To relate this general goal to particular requirements, we
system. Improvements such as stronger materials or us- can look at the impact of new space technologies on qual-
ing space resources reduces the required kg for a given ity of life measures already in place. An example would
project. A number of component costs make up the re- be high efficiency closed loop food production. This
curring cost/kg, but the biggest improvements will come could lead to improved food security or lower farm wastes
from different design and technology used for transport when applied on Earth. Thus during conceptual design
to orbit. we should look for methods and technologies with Earth
For this example we will measure cost reduction rela- applications, determine what quality of life impact they
tive to current (2012) values as the baseline. Current could have, and give preference to those with the greater
space hardware and propellant masses for typical mis- potential applications and impact. A fusion rocket would
sions determine the total kg required. New designs and then be preferred over a regolith engine because function-
technology result in different masses for the missions. ing fusion reactors can also help solve energy security on
The mass improvement ratio is then current mass/new Earth. A device that uses raw rock as reaction mass for
mass. For launch cost we will assume the quoted cost of propulsion does not have that kind of Earth application
US $1566/kg for the Falcon Heavy as the baseline. For (that we know of).
new transport systems the marginal operating cost to Low Quality of life measures include physical measures such
Earth Orbit after paying for development is used. The as GDP, life expectancy, and pollution levels, as well
launch cost ratio is then ($1566/kg)/(new launch cost). as social measures like education, leisure time, and civil
Total cost reduction is then the product of (mass improve- rights. One step in establishing program requirements
ment ratio) x (launch cost ratio). We will assume equal will then be to select the appropriate measures. The next
percentage reduction in cost has constant customer value, step would be to set how much of our program should
and therefore convert the logarithm of total cost reduction have potential impact on these measures, and what level
to a score. Half the maximum potential improvement, of improvement is desired.
6.3. HUMAN EXPANSION 2 241

Understanding the Earth away from the Earth, active control of biosphere param-
eters, and establishing artificial biospheres in space or on
Understanding our home planet is mainly a scientific en- other planets.
terprise. It includes gathering data, developing theories
and models, and then testing those ideas against reality.
After you have gained a better understanding, then decid- Expanding Resources
ing what to do with that knowledge is a social and polit-
ical issue outside this program’s scope. It may lead back All civilizations require resources to function. This group
to new projects such as orbital sunshades to reduce tem- of program requirements can include identifying scarce
peratures on Earth, but the choice to start such projects physical, material, and energy resources and setting quan-
should be based on sound knowledge. This group of re- tity and cost goals for increasing their availability. Physi-
quirements will address gaining that knowledge. cal resources include quality living, growing, and working
space in terms of dimension and environment. Material
We want to gather data about the Earth and other plan- resources are all the raw matter and specialty compounds
etary systems. The reason for the latter is other planets and equipment needed. Energy resources are needed in
are natural experiments in what happens under other cir- various forms for different purposes. All of these are in-
cumstances. We cannot gain knowledge by experiment- terlinked in terms of flows of resources for different tasks.
ing with the one planet we live on, but we can test our
theories and models by looking at other plants. Gather-
ing data has dimensions of time and space. We want to Long Term Survival
know both the current properties of planets and their en-
vironment in three dimensions, and the history of those Requirements in this area would include examining long
properties across time. We can then set program require- term resource depletion. For example, continued use of
ments in terms of increased detail in both. An example nuclear fission for power would eventually deplete the
would be “Map all Solar System objects to a resolution Earth of Uranium and Thorium. If that were a criti-
of 1 kilometer or better.” A large pile of data is not very cal part of keeping civilization operating, then eventual
useful by itself, so sufficient resources to organize it and collapse would occur. There are alternative sources of
develop theories is also needed, after which a new cycle power, so this particular example is not fatal, but it shows
of data collection to test the theories would happen. Pro- the idea of looking for items that would run out in the
gram requirements would then include the scientific staff long term. The Sun also gets 1% brighter per 100 million
to make use of the data and guide later cycles of data years from stellar evolution, and the Earth permanently
collection, and presenting the results of the accumulated loses about 3 kg/s of Hydrogen from water dissociation,
knowledge in a form that educates people on choices they equivalent to 850,000 tons/year of lost water. These long
may need to make. term changes will eventually make the Earth uninhabit-
able. So program requirements can be set to either coun-
teract depletion and changes to the Earth, or enable mov-
Reducing Space Hazards ing elsewhere within the timescale of the problem becom-
ing critical.
This class of program requirements includes identifying
hazards from space, followed by preparing for or prevent-
Increased Choice
ing the damage from them. Known hazards include as-
teroids and comets, solar flares, and stellar explosions. Existence on Earth under our current civilization imposes
Types of requirements can include the ability to deflect natural and human-made restrictions. Measures in this
hazardous objects, and limiting damage from events like area look at lifting or eliminating these restrictions. Some
flares that cannot be prevented. examples are freedom of location - on Earth you are re-
stricted by national governments from living anywhere
you want. Another is freedom of gravity - you cannot
Increasing Biosphere Security choose to live under a different gravity level right now.
A single uncontrolled biosphere is inherently insecure
from natural and human-made variations. Since nearly all Increased Opportunity
of us currently depend on this one biosphere, we want to
set program requirements that improve the security level. Most of the Earth (the good parts at least) are already
The requirements groups of understanding the Earth and claimed by someone. By making new unclaimed or
reducing space hazards help in reaching this goal. This under-used areas accessible, that increases opportunities
group goes further to counteracting undesirable variations for people who want to start something new without first
and adopting the idea of backups from computer technol- having to pay off previous owners. Measures in this area
ogy. Specific requirement might include providing safe would include increased area or resources which are made
storage of biological samples and testing of biohazards available.
242 CHAPTER 6. PART 5 - DESIGN STUDIES

6.3.3 Design Approach ploited. Therefore we will try to design for using
them to leverage what we bring from Earth.
Having set up a way to measure how good the design is
and establish requirement, we next need an approach to
• Build multipurpose facilities - One-time missions,
formulate the System Concept, a high level description
as have often been designed to date, tend to not leave
of what the combined space system is and how it works.
anything useful behind. Thus the next mission is ex-
There is no single magic bullet (or magic rocket) that
actly as hard and expensive as the last one. There-
can meet all the program requirements by itself. If there
fore we will try to design facilities that can be used
were, someone would have used it by now, or at least be
multiple times or on a permanent basis. An exam-
pursuing it seriously. Therefore we take the approach of
ple would be landing a solar array on the Moon that
leveraging multiple good ideas, which allows the savings
is used to recharge a rover vehicle quickly, and then
to multiply together. This will result in a complex pro-
later can also be used to power an extraction plant.
gram of multiple systems, which need to be combined
This makes more sense than having a solar array at-
for best results.
tached to the rover, and then another solar array for
To meet the cost criteria of affordable development we the plant.
do not build everything at once. Instead the ideas get ap-
plied in incremental projects and systems which build on
• Use diverse modular designs - Monolithic, or sin-
each other. This allows some return from the early parts
gle piece, designs require replacing the entire item
to help pay for the later parts. The early parts are smaller
if your needs change or you have an upgraded tech-
in scale than what comes later, which further reduces ini-
nology you can use. For long term and complex
tial development. This will result in a program which is
projects you are not able to predict all the changes
extended in time.
and upgrades that might be required. Therefore we
will try to use modular designs where possible to
Multiple Ideas To develop our system concept we will make it easier to change things. Modular designs
use the following ideas: can also start smaller and be added to in steps, re-
ducing initial costs and the size of transport systems.
• Use less of or eliminate conventional rockets - It is not required to use one method to do every-
They have been in use for 50 years and had a lot of thing. On Earth we use industrial delivery systems
engineering development and optimization. There- like pipelines to deliver large quantities of goods
fore using another conventional rocket is unlikely to cheaply, and reserve more expensive and safe meth-
bring much improvement, and other projects are al- ods to deliver people. Specialization of this sort is
ready attempting to do so. Instead we will try to use acceptable when it makes cost and technical sense.
some of the other hundred or so transport methods
and variations identified in Part 2. This gives us the Using the above ideas does not mean using them blindly
possibility of greatly improving on the performance where they are not appropriate. It means incorporating
and cost limits of chemical rockets which are im- them where it makes sense, in an optimized amount. Past
posed by their chemistry. programs have tended to not use them enough, or at all.
This has led to high cost and limited performance, which
• Design for re-use, repair, and recycling - It should our new program tries to correct.
be evident that these features will reduce hardware
and supply cost, and yet many launch vehicles and
satellites are used once and disposed of. For hu- Incremental Projects Rather than attempt to do every-
man crews on, for example, the Space Station, oxy- thing at once, we will take the approach of designing and
gen and food supplies are similarly used once and building our space program elements in progressive incre-
disposed of. Some space hardware is designed ments. The increments will establish new locations and
for maintenance and repair, but much of it is not. improve measurable parameters in several dimensions as
Therefore we will try to incorporate these multiple they get added:
use and long life features to get more use out of hard-
ware and supplies. • Working environment range - Starting with tem-
perate Earth locations provide production, habita-
• Use the material and energy resources of space - tion, transport, and other system elements that work
Again, it should be evident that bringing everything in the given environment. Then extend their work-
from Earth is a limiting factor, and the farther you ing range to hotter, colder, wetter, drier, and higher
go in space, the higher the cost of doing so. Solar and lower pressures. Later extend the range of
power is so overwhelmingly useful that it has been working environments to space locations with the
used by almost every space project, but other ma- additional variables of gravity and radiation levels
terial and energy resources have not yet been ex- and increased range of temperature and pressure.
6.3. HUMAN EXPANSION 2 243

• Time and energy range - Time has components of deserts, oceans, high altitude, cold, or underground. This
communication time, travel time, and stay time. En- expands the range of environments and the distances
ergy has components of potential and kinetic energy for remote control. Widening the range of conditions
to reach a given location. New locations will in- where people can live and work meets many of the pro-
crease the range of communications and control dis- gram requirements, and the new locations should be self-
tance, and require longer travel and stay times. They supporting physically and economically.
will also require greater energy changes to reach. After sufficient locations on Earth are established, the
production capacity is used to build transport to orbit, and
• Performance levels - These are measures like cargo the remote operations and other technologies are further
capacity, industrial output, efficiency and closure, developed for space locations and the more difficult en-
and how many humans are supported. They are des- vironments of vacuum, temperature, radiation, and lack
ignated as performance requirements for particular of gravity.
program elements and systems. Each location starts
To enable growth in new locations we design and build
with a given performance set, and it is improved in
the following kinds of hardware:
stages to higher levels.

Seed Factories
Increments which are far away in time or require param-
eters far beyond current experience become uncertain to
Developing a new location has always involved bringing a
design for. New technology may get developed in the in-
starter set of knowledge and tools. Historically that meant
terim times, and meeting untried parameter levels may be
bringing animals, seeds, axes and hammers, and whatever
difficult. Therefore at some point it is no longer useful to
else the technology of the time required to start building.
plan future items in detail. Instead, the options for these
For our future program, we want to use the best methods
items can be laid out, and a plan to develop needed tech-
that modern technology allows. Our starter set should use
nology and reduce uncertainties put in place. Therefore
automation and remote operation, and be able to not just
program engineering work is not done once and finished,
make a fixed set of products, but make more equipment
but is a continuing effort. At any point in time the pro-
for itself to widen the range of outputs. So our concept is
gram will have a baseline documenting current status and
to use Seed Factories to establish production capacity at
future plans, but that baseline will get updated on a regu-
each new location. Unlike conventional factories which
lar basis.
only produce a given set of products, a seed factory uses
part of its production to make more equipment for itself.
6.3.4 Initial Program Concept So over time it is able to make a wider range of prod-
ucts and use a wider range of local resources. The initial
Based on the above discussion we will describe a starting seed equipment, plus supplies and necessary components
point for the program concept. We emphasize that this is it cannot make yet, are delivered from the previous loca-
only a starting point, the final concept is the endpoint of tion. Over time, the production capacity will need fewer
the conceptual design stage. supplies and parts, and be able to make more items lo-
cally.

Program Description Besides self-expansion, the factory output is partly items

needed to live and work in the particular location. In
space that would include items such as parts and supplies
Our basic concept for the program is to expand the range for habitats, vehicles, mining equipment, water, and oxy-
of human civilization to new locations while meeting the gen. The remainder of the output would be commercial
above program goals and requirements. Any kind of civ- items to trade for needed outside supplies. On Earth that
ilization seems to require the ability to produce food and could be any kind of product with a market. In orbit, an
other physical items, to provide shelter, and to move peo- example would be building large communications satel-
ple and items from place to place. There may be other lites. Vehicles would carry copies of the seed equipment
basic requirements, but we will start with these. There- on to the next location to start growth there. Since habi-
fore for each new location we set up and then expand in tats for humans are part of the production output, at first
stages the production, habitation, and transport elements. you may not have the ability to support many of them
We start from existing civilization as it is, start a new lo- at a new location. Therefore the seed factories are de-
cation, and once it is sufficiently developed, travel to and signed for a mix of automated/robotic, remote control,
repeat the process at the next one. and direct human operation to minimize the latter. The
The first new locations will be easy ones on Earth, in first seed factories are built in various locations on Earth.
temperate climates. New technology such as remote op- When they have sufficiently expanded they use their in-
erations, automation, and resource extraction are first dustrial capacity to build launch vehicles and space hard-
demonstrated there. Once built up, we transport new ware. This is used to establish assembly and processing
elements to more difficult locations on Earth, such as equipment in Earth Orbit. This evolves into a full seed
244 CHAPTER 6. PART 5 - DESIGN STUDIES

factory, and from that grows to other locations in space. We have set our starting point for the conceptual design
stage by stating goals and potential benefits, initial selec-
Cyclic Systems tion criteria, classes of requirements, a design approach,
and initial program concept. We now apply the Systems
A linear system might dig up a resource to be used as Engineering method from section 1.5 to further develop
fertilizer, use it once, and then it gets removed in the the design in increasing detail. The reader should bear
harvested crops and runoff. In a crowded Earth, or in in mind that a book such as this is a linear presentation
space locations, where getting new supplies is hard, re- of sections and pages, but the actual method is iterative,
sources for once-through linear designs become expen- with the results of later steps feeding back to update ear-
sive and unsustainable. Therefore our concept includes lier results. We will try to indicate when these differences
using cyclic systems which take old items and send them between practice and description happen.
back to earlier production stages for repair, re-use, recy-
cling, or reprocessing to new items. Because transport is
an overhead cost, we prefer to do the cycling locally, but 6.4.1 Program Requirements Analysis
if it turns out to be more optimal to do the tasks at other
locations we will do so. Until now we have stated what the program is intended
to accomplish in very general terms. The requirements
High Efficiency Transport analysis step takes these general statements and develops
them into more specific and measurable features at the
program level. In later stages of design these will be ana-
New locations start with deliveries from previous loca-
lyzed in more detail and assigned to separate systems and
tions, and will continue to need transport to deliver new
lower level items as System and Design Requirements
equipment, supplies, and people. Once sufficient produc-
which their designs attempt to meet. In the process we
tion is built up, delivery of finished products back to pre-
map the general requirements to the specific ones to make
vious locations is needed. In addition, return of people
sure all of them will be met. The minimum number of
and used items back to previous locations is needed be-
necessary requirements should be defined at each level
fore setting up permanent habitats and fully cycling sys-
of a program’s development. Excess requirements limit
tems. Thus transport is a necessary function, and we pre-
design options and impose extra costs to design for and
fer to do that as efficiently as possible.
verify.
On Earth, machines like internal combustion engines are
It should be noted that setting initial requirements does
particularly inefficient, so we will look for more efficient
not mean they are physically possible, economically feasi-
replacements. In space transport, conventional rockets
ble, or a better choice than what exists now. They merely
are even more inefficient reaching orbit than gasoline
establish design targets, and later work will establish if
engines. Although chemical rockets are very efficient
they are the correct ones.
(~80%) as heat engines in themselves, when considered
as part of a total transport vehicle, most of the work goes
into accelerating fuel which is later burned and discarded. Requirements Sources
Therefore our concept is to replace conventional rockets
as much as possible with a variety of higher efficiency First we identify the sources or inputs to the requirements
transport methods. This includes several options for ini- analysis. These should come from outside the program
tial launch to orbit, and primarily electric thrusters once itself:
in orbit. Chemical rockets will not be entirely eliminated,
especially at first, but are gradually replaced as the pro- • Program Goals and Benefits - These were stated
gram evolves. Electric thrusters are about five to ten times on page 1 of this section, and come from unmet de-
more efficient in fuel use, and combined with the ability sires of civilization as a whole, as best we can deter-
to extract fuel locally require dramatically less fuel from mine them.
Earth, so we will try to use them heavily in the design.
Like all rockets, chemical ones use fuel exponentially as • Systems Engineering Experience - Past engineer-
a function of velocity. So even replacing part of their use ing experience in developing complex programs has
will significantly reduce total launch mass. So our incre- identified requirement types that often turn out to be
mental approach still yields large gains even with smaller necessary. We can use topical literature to identify
early steps. ones relevant to this program.
continue to page 3 • External Constraints - These are limits imposed
by nature or from human causes such as legal re-
quirements.
6.4 Human Expansion 3
In addition to the external requirements, we have ones
back to page 2 identified or derived internally from the program goals
6.4. HUMAN EXPANSION 3 245

and benefits, design approach, or initial program concept, these remote locations, they can more easily study them,
which are on page 2. and science may be a major “occupation” for the resi-
dents. The data requirement (2.5) aids this goal explic-
itly. The category of “Reducing Space Hazards” is cov-
Analysis Process ered by the Population Risk (6.2) requirement, but we are
not yet defining specific risks and reduction targets. This
Goals and Benefits - We examined each of the program requirement also partly addresses moving hazardous re-
goals and benefits as stated on page 1, and attempted to search from the “Increasing Biosphere Security” heading.
write individual statements of what the program will ac- Active measures to increase biosphere security are cov-
complish in terms of objectives, performance, schedule, ered under Survivability (7.2). Measures under the “Ex-
cost, technical risk, safety, sustainability, and openness. panding Resources” category includes physical, which is
The headings for these categories come from the list of dimensional in terms of locations, area, volume, and envi-
requirement types in Section 1.5. Different programs ronmental range. This is covered by the primary program
would result in a different set of such statements. We goal (1.1). Material and energy resources are covered by
tried to make the statements specific and have a measur- the Resources requirement (2.6). From “Long Term Sur-
able feature or parameter. In some cases, this is not pos- vival” we added to the Survivability requirement (7.2) to
sible yet and will have to wait until later in the design design for depletion of critical resources. The “Increased
process. Placeholders are used for the incomplete state- Choice” heading is covered by the similarly named re-
ments. The result of this examination and a discussion of quirement (1.3). The primary goal (1.1) of widening the
how each statement is derived becomes part of the initial range of civilization addresses the “Increased Opportu-
draft of the Program Requirements in the next section. nity” category in the sense of access to unclaimed and
under-used locations. The Openness requirements (8.)
Systems Engineering Experience - We next considered maintain access to the technology and locations for peo-
general categories of requirements, such as those listed in ple outside the program, giving them increased opportu-
Section 1.5, to see what might apply to this program. The nity.
program is wide scope and long term at this level of anal-
ysis. Therefore many requirements that apply at more Design Approach - Our next step was to look at our de-
detailed levels are not relevant yet. One item that came sign approach to see if that drives any program require-
up is the “Life Cycle”, ie when would the program end? ments. Normally design progresses from requirements to
We included in the discussion under 1.2 that the endpoint solution, but this is a check to see if our approach feeds
is reaching the stated populations with the ability to con- back in the other direction. The idea to reduce conven-
tinue supporting them afterwards. Compliance to laws, tional rocket use is not a program level requirement, it is
regulations, codes, and standards is too specific to apply at a design solution for the space part of the program. The
this level, as are durability and quality features. By it’s na- program requirement to lower Earth launch cost (4.3) is
ture this program is intended to have a positive impact on what will likely drive the use of that idea. This require-
community and the environment. At more detailed levels ment will also drive using reuse/repair/recycling in terms
these may become separate requirements. Manufactur- of longer vehicle life. Improved technology (2.3) and sur-
ing, Test, and Maintenance also apply at detailed levels. plus resources (2.6) along with other requirements also
Flexibility, Scalability, and Evolution are expected to ap- drive reuse/repair/recycling, as that idea is a general solu-
ply at the location or system level. Interoperation is met tion to design optimization. For the space part of our pro-
at the program level by virtue of the locations being a part gram, using the resources of space is another general so-
of human civilization as a whole. At more detailed levels lution to optimizing the design. The resource requirement
specific requirements may emerge for this category. The (2.6) to have a surplus output will drive using this idea.
end result is not many new requirements were identified Building multipurpose facilities is another optimization
at the program level from this step. vs single purpose design for each task. The program scale
requirements (1.2) for permanence and size will likely
External Constraints - For now, we consider the re- drive using this idea. Along with growth (2.2) it will drive
quirement to function in more difficult environments and using the idea of modular design. The design approach
distant locations as a sufficient identification of external of incrementally increasing environment range, distance,
constraints. For a specific location, such as the Lunar L2 and performance has already been incorporated into mul-
point, the physical environment and legal regime will be- tiple requirements. Thus we find the requirements will
come inputs to the more detailed requirements. tend to use the ideas in our design approach, but the ideas
Internal Program Requirements - Next we look at the do not force new requirements.
classes of requirements identified on page 2 to see if any Program Concept - Similarly we check our initial pro-
of these apply. From the “Improving Life on Earth” head- gram concept from page 2 to see if any new requirements
ing we define the requirement (2.4) for improved qual- are identified:
ity of life relative to current levels. The “Understanding
the Earth” goal is indirectly enabled by inhabiting more
difficult environments. When people are able to live in • Program Description: We identified a need for
246 CHAPTER 6. PART 5 - DESIGN STUDIES

production, habitation, and transport as basic to civ- Discussion: Our program concept is to develop new lo-
ilization, but these will be developed as lower level cations starting with the easiest, but with improved tech-
functions for each location rather than top level re- nologies in areas like energy, automation, and recycling.
quirements for the program. The top level is cov- This should result in lower costs and improved quality of
ered by the requirement to support a certain number life measures. This first requirement is just restating the
of people (1.2). Set up and expansion of these func- first part of the concept as a definite objective. It is quan-
tions is covered by (2.2) Growth. The main program tified and refined by all the following requirements. Note
goal (1.1) is to move to more difficult and remote that new locations include difficult ones on Earth as well
locations. This will drive remote operations and au- as in space.
tomation, because humans can't live there until habi-
tats are in place. Resource extraction is explicitly • 1.2 Program Scale - Expansion shall be demon-
included as a requirement (2.6). Specific Earth or strated by permanently supporting at least 95,000
Space location are not identified as program require- humans total among new Earth locations and at least
ments, but are implicit in number of locations (2.1). 2,000 humans per new space location.

• Seed Factories: The idea of using factory output Discussion: A target size is needed to know when you
to expand itself is a good solution to address the have reached the program goal, and to determine the
Growth (2.2) and Improved Technology (2.3) re- size of hardware designs. They are intended to be large
quirements, but does not impose new requirements enough to prove the new locations are permanently occu-
at the program level. Technical details of such fac- pied and the new technologies function reliably. They are
tories and what they produce would be lower level end-point targets for this program, but nothing would pre-
design and requirements. vent further expansion afterwards, or in parallel by other
programs. Life cycle analysis ends at this point, but the
• Cyclic Systems: The idea of using cyclic flows as locations are designed to continue operating. The initial
much as possible is also a good solution to meet numbers of people will be much smaller, possibly zero
many of the program requirements, such as 7. Sus- if robotic and remote operation is the best way to begin
tainability and 4. Cost, but we do not see new top building in space. As a starting point the numbers were
level requirements from this idea. arbitrarily chosen to be the square- and cube-root of to-
tal human population in 2050. Later sizing analyses will
• High Efficiency Transport: Again, this is a good
likely change them, but something is needed for the first
idea for cost and efficiency reasons. We have set
round of design work.
an Earth Launch Cost requirement (4.3) which will
direct how hard we work to include this idea. The Demonstration of the new locations and meeting all the
challenging cost target will tend to force reduction other requirements supports the benefit of optimism for
of chemical rocket use as a solution, but this does the future. It provides an understandable path to an grow-
not impose new top level requirements. ing, open, and improved life, rather than a finite, closed,
and zero-sum world.
For now, this completes our analysis process. As with
• 1.3 Choice - Specific locations and their internal
any engineering design, more detailed work later on may
organization, function, and operation shall be cho-
cause an update to these results.
sen by program participants and location residents
within the limits of design constraints.
Initial Draft of Program Requirements
Discussion: This requirement is included to implement
the goal of increased choice and freedom. Therefore
The following requirements list is a first draft drawn from
choices such as where to build a new location are not
the process described in the previous section. The indi-
imposed from above, but made internally. Participants
vidual requirements are sorted by category and numbered
are people like the designers, who may not end up living
for ease of tracking, with a discussion of how they were
in the new location, and residents are the ones that do.
developed. Further conceptual design work will add to
Constraints are limits on a design imposed by nature, like
and refine this list to produce a final Program Require-
level of sunlight at a location, or by humans, such as a law
ments Document, one of the outputs of the Conceptual
which prevents using fission reactors for power at a given
Design stage of a program.
location.

1. Objectives 2. Performance

• 1.1 Program Goal - The program shall expand hu- • 2.1 Number of Locations - The design shall maxi-
man civilization to a series of new locations with in- mize the number of new locations, where new is de-
creasingly difficult environments and distance. fined by at least a 10% increase in an environment
6.4. HUMAN EXPANSION 3 247

parameter or distance measured in time or energy the current average. The exact measures to compare to
terms. will need to be defined later in the concept development.
The 10% threshold is a notional goal, representing “better
Discussion: Requirements can be a fixed number or than most” conditions.
a variable parameter. Here it is a variable denoted by
“maximize” without a specific number. A requirement • 2.5 Data - The program shall collect and dissemi-
which can increase without limit results in an unbalanced nate [TBD] data about the Earth’s environment, sur-
or infeasible design. So desirable parameters like “more rounding space, and objects therein.
locations” are balanced against the others by using mea-
sures of effectiveness and scoring across the entire pro- Discussion: This requirement supports the program goal
gram. The second part of the requirement establishes of understanding the Earth better by looking at its envi-
how much difference is enough to be new in terms of lo- ronment, and the history and evolution of other planets
cation. and objects. We don't know at this stage of the analysis
what kinds of data will be useful, at what level of detail,
• 2.2 Growth - Each location shall increase the ca- or at what cost, so we use a placeholder value. Existing
pacity for production, habitation, and transport in a astronomy and planetary science program are already do-
progressive manner. ing this task, so the final requirements will be in terms of
what capabilities beyond that the program will add. In
Discussion: Our initial program concept identifies these addition, some of the data collection is needed to support
as basic functions of civilization. Setting a requirement future expansion to new locations in space.
to grow rather than meet a given level all at once allows
incremental and modular design. The growth rate and • 2.6 Resources - The program shall output a life cy-
target capacities for each function are not specified at this cle surplus of at least 100% of internal material and
point. Later modeling and optimization will determine energy resource needs.
what is feasible, and specific values will then be applied
to selected locations. Discussion: - This supports the program goal of ex-
panding the material and energy resources of civilization.
• 2.3 Improved Technology - Locations shall in- The program will consume resources internally to oper-
crease the levels of self-production, cyclic flows, and ate, and will not produce a surplus immediately, so this
autonomy in a progressive manner. requirement sets a net output requirement for the total life
of the program back to civilization.
Discussion: The program concept is not just to expand
to new locations, but to improve technology levels. This 3. Time
requirement sets the main categories of improvement. It
is in terms of all locations as a group, since one location • 3.1 Completion Time - The expansion to a new lo-
may be dedicated to a single task like mining or growing cation shall be completed before expected progress
food. Different locations then trade outputs. The choice in technology indicates a re-design is required.
of what to do at each location and how much transport is
needed between them is a design optimization to be set- Discussion: If the program is too slow to develop a new
tled later. Self-production is the amount done internally location, then re-design to account for new technology
rather than obtained from outside sources. Cyclic flows would be indicated before you are finished, and therefore
are materials which get recycled and reprocessed, rather previous design work is wasted. This is separate from
than new inputs and waste outputs. Cyclic flows do not economic or other reasons which might indicate a partic-
count new production for internal growth or external de- ular schedule. In case of multiple requirements driving
livery. Autonomy measures levels of automation and in- one parameter, the strictest one is used.
ternal control so that a location does not require as much
outside support or human labor to keep operating. The • 3.2 Operating Life - Locations shall be designed
numeric levels for these parameters will be set later by for an indefinite life, with maintenance, repair, and
what is feasible. replacement.

• 2.4 Improved Quality of Life - Completed loca- Discussion: Expansion of civilization is an open ended
tions shall provide an improved physical and social process, so the intent is to set up permanent new locations.
quality of life relative to the upper 10% of Earth civ-
Anything made by humans eventually wears out, breaks,
ilization. or could be replaced by something better. To reach an
indefinite life, then we need to provide for ongoing main-
Discussion: This makes specific the goal of improving tenance, repair of failed items, and eventual replacement
life on Earth by demonstrating quality of life well above or upgrade.
248 CHAPTER 6. PART 5 - DESIGN STUDIES

4. Cost The combination of improved technology (2.3), surplus

resources (2.5) and this requirement supports the pro-
• 4.1 Total Development Cost - The total develop- gram benefit of expanded markets.
ment cost for new technology and hardware designs
shall be less than 50 times the unit cost on Earth, and
5 times the unit cost in space of the hardware. 5. Technical Risk

Discussion: There are only a small number of govern- • 5.1 Risk Allowances - Program designs shall in-
ments and businesses capable and motivated to do large clude allowances for uncertainties and unknowns in
space projects. A reduced scale, especially in the ear- knowledge, performance, failure rates, and other
lier stages, means more possible entities that could per- technical parameters. New designs with higher risk
form the projects, and thus more chances it actually gets can be included in program plans, but a process shall
done. Large entities are not excluded from these kind be included to resolve the risk, and an alternate de-
of projects, they just should not the only ones able to do sign with lower risk maintained until resolved.
them. This is one reason to limit the total development
cost. Another is to generate an economic rate of return Discussion: Note that technical risk comes from design
so the program can pay for itself (or at least pay part of unknowns, and is distinct from accident risk. The pro-
its way). We set reasonable values of 50 and 5 times the gram goal is to develop new locations with more diffi-
unit cost of hardware for Earth and space based on an cult environments using new technology. The nature of
estimate of the number of locations the hardware will be pushing boundaries like this means that a reasonably large
used, and the desire to keep the development cost reason- level of technical risk will be encountered at first. This
able in relation to unit production cost. does not mean taking risk for it’s own sake - there should
be sufficient potential gain to justify the higher risk. We
• 4.2 New Location Cost - The peak net project cost adopt an allowance method to manage the technical risk,
for a new location shall be less than 50% of the ex- meaning we estimate the uncertainties and include them
pected long term net output. in our design. Rather than passively carry the risk, we
have added requirements to reduce the risk by methods
Discussion: For a new program to be justified over do- like testing, and carry a design alternative as a backup if
ing nothing or continuing with existing programs it must the new technology turns out not to work as desired. The
show some benefit or improvement. This requirement alternative to managing risk it to be safe and conservative,
sets a minimum benefit in purely economic terms that the but that does not lead to progress.
life cycle output be worth twice the peak project cost.
There are other types of benefits than monetary ones,
which would be covered by other requirements, and more 6. Safety
detailed cost goals will be worked out in later stages of
design. • 6.1 New Location Risk - New locations shall pro-
gressively lower internal risks to life and property,
• 4.3 Earth Launch Cost - The program shall pro- with a goal of significantly lower risk than the gen-
gressively lower the Earth launch cost component of eral population.
total system cost, with a goal of $0.08/kg of total
system mass. Discussion: Expanding to new and difficult environ-
ments will inherently involve some risk. It is acceptable
Discussion: One of the potential program benefits is dra- to have higher than average risk at first. Many people
matically lower costs for space projects. We include this willingly accept such risks if they know about them. This
as a requirement in order to drive the design to explicitly requirement sets a goal that once a location is fully de-
attempt to reach it. This only addresses the space trans- veloped, it should be relatively safe for the people and
port part of the program, Earth surface transport for ter- property internal to the location.
restrial locations is already well optimized, and any pro-
gram requirement to lower it would be separate and a • 6.2 Population Risk - The program shall signifi-
much smaller goal. Assuming local resources are used in cantly reduce natural and human-made risks to the
space, a reasonable long term goal is for 2% of total mass general population, including external risks created
to come from Earth. Therefore the goal for the launched by the program.
mass would $4/kg. The product of the two factors results
in Earth launch cost, so the component requirements are Discussion: There already exist natural risks such as so-
inversely related. This is a long term goal, and will not lar flares and asteroid impacts, and human-made risks
be reached in early parts of the program. Total system from orbital debris and launch accidents. These are haz-
mass includes consumed mass like propellants, so propel- ards to society in general. To meet the program goal of
lant production may reach very low percentage launched reducing hazards from space we set this requirement to
from Earth fairly easily. reduce them. The program itself will create some new
6.5. HUMAN EXPANSION 4 249

hazards and risks to people and places outside itself. The Discussion: 8.1 and 8.2 both address the goal of in-
goal is to reduce the total, including the newly created creased opportunity. Opening general technology and
ones. methods allows others outside the program to make use
of them. Specific designs and artifacts can be proprietary
because of the necessity to cover costs. Barriers to en-
7. Sustainability try reduce opportunity, so allowing later arrivals to travel
through or access unused resources minimizes those bar-
• 7.1 Biosphere Security - The program shall in- riers. An example of unreasonable claims would be an
crease biosphere security by establishing alternate exclusion zone extending 1000km from a high orbit plat-
biospheres and long term storage of biological ma- form. Some exclusion zone is reasonable for safety and
terials. to prevent sun shading, but not that large.
contine to page 4
Discussion: All life on Earth and human civilization cur-
rently depends on the single natural biosphere. This is a
potential single point of failure from natural or human-
made causes. Having a backup in the form of addi- 6.5 Human Expansion 4
tional functioning biospheres and storage increases se-
curity. The habitation requirements under Performance back to page 3
could be met entirely mechanically, so this requirement
adds biological features. We cannot reasonably duplicate
the entire Earth, so the scope of this requirement will be 6.5.1 Program Evaluation Criteria
set by what is feasible.
As noted on page 2, we want an objective method to
• 7.2 Survivability - The program shall design for the choose among design alternatives, and to evaluate this
long term survival of life and humanity from changes program or its parts against existing programs. For a
to the Earth which will render it uninhabitable and complex program such as this we take multiple measur-
depletion of critical resources. able features or parameters and convert them to a com-
mon scale. The features chosen are the ones of value
to the ultimate customer, which here is civilization as a
Discussion: It is well known that the Sun will get hotter
whole, and the conversion formulas are according to their
and eventually become a red giant. There may be other
relative importance and desirability. Since we cannot ask
long term changes, such as to the Earth’s orbit. This re-
everyone in the world what they want, we have to act as a
quirement sets in place design features that will enable re-
proxy for them and make our best estimates of what they
acting to those long term changes. Because they are far in
would want if they were well informed on this topic. We
the future, it may not require much action in the near term
can draw on outside information to help with this process.
beyond understanding what the nature of those changes
Each design alternative will have different actual feature
will be. Critical resources are ones necessary for civiliza-
values, and result in different scores when converted to
tion to function and have limited sources and will run out
the common scale. Evaluation of the alternatives then
at some point. Part of the design is to consider what these
amounts to adding all the scores and seeing which one has
are and look for alternatives or additional sources.
the highest total. For this step of the Conceptual Design
we can only establish what our criteria should be. De-
8. Openness vising alternatives and selecting among them will come
later.
• 8.1 Open Design - Technology and design meth-
ods developed within the program shall be open for Identifying Candidate Criteria
others to use. Specific instances of a design and pro-
duced items may be proprietary. Every possible measurable feature or specification could
be used as a point of comparison, but this is unfeasible
Discussion: The combination of developing Earth loca- for two reasons: (1) the time and complexity to evaluate
tions first(1.2 and 2.1 above), improved technology (2.3) all of them for every alternative, and (2) many features
and this requirement are aimed at the benefit of technol- are simply undefined at a high level of a program. So
ogy spin-off/Earth applications. Since they are used on we restrict our candidates to those which are the most
Earth first, and made open, there should be substantial important, and those which can at least be estimated at a
transfer outside the program. high level. Any feature which is absolutely required for
the program is not a point of comparison, since all the
• 8.2 Access - Development of a new location shall valid alternatives must include it. What that leaves us with
not prevent reasonable access for transit or to unused are parameters which are variable in some way, so that
resources. an alternative can do better or worse by that measure. An
250 CHAPTER 6. PART 5 - DESIGN STUDIES

example of a good measure is cost. Every alternative has • 2.3 Improved Technology - Like growth, technol-
an associated cost, which is almost infinitely variable, and ogy levels are required to increase progressively, but
so useful for comparisons. Most people will agree that without specified values. We can define ranges for
lower cost is better, although they may differ by how much this parameter and weigh it against other criteria.
better. The general agreement that cost is important and
which direction is better allows setting up a formula to • 2.4 Improved Quality of Life - This sets a lower
convert particular cost ranges to score values. threshold of top 10% of Earth civilization, but no
upper bound. The specific physical and social mea-
We already (page 3) developed our program requirements sures will make good criteria.
based on program goals. By designating them as require-
ments we have indicated they are important, so those are • 2.5 Data - The goal is to collect and disseminate an
the first places we should look for measurable criteria. unknown amount of data. This is inherently a vari-
Beyond that, we will also look at the same headings we able parameter, so we will include it as a candidate.
used for the requirements analysis process, and consider
civilization needs and desires generally, as we can find • 2.6 Resources - We have a requirement of 100%
them from outside sources. surplus of resource needs. Since a design might fall
short or exceed this level, it is a good candidate.

Program Requirements As a note, performance levels of a complex program are

typically a rich area for finding evaluation criteria, since
Referring to the Program Requirements on page 3, we they tend to be variable with design choices.
identify our first set of candidate criteria as follows:
3.0 Schedule:
1.0 Objectives:
• 3.1 Completion Time - The goal here is to complete
• 1.1 Program Goal - This sets the fundamental goal a location before technology renders it obsolete and
of the program, which must be met. With no room a redesign or replacement is indicated. We can ana-
for variation it provides no candidate criteria. lyze this and set a specific time range for completion
as a criterion.
• 1.2 Program Scale - This sets a requirement of
number of people on Earth and in space to be sup- 4.0 Cost:
ported. The intent is to demonstrate permanent oc-
cupation and use of a location, but this can be done
with a different number of people than specified. • 4.1 Total Development Cost - Like performance,
This makes a good candidate. cost is another rich area for evaluation criteria, since
cost is a barrier in the physics and social sense to a
• 1.3 Choice - At this level it merely requires choice program getting approval. The ratio of performance
by participants and residents, but does not specify to cost is a particularly popular measure of effective-
how much. Since we could define degrees of choice, ness. This is a good candidate.
this can be a candidate at this level or lower levels.
• 4.2 New Location Cost - This sets the net project
Criteria applied only at lower levels are either an ab-
cost less than half the long term net output. This is
solute requirement at the program level, or summed
an explicit effectiveness measure, and makes a par-
somehow to a higher level result.
ticularly good candidate criterion.

2.0 Performance: • 4.3 Earth Launch Cost - This is another progres-

sive improvement goal with a variable value and an
aggressive target, so it makes a good candidate cri-
• 2.1 Number of Locations - The requirement is to terion. At lower levels it can be divided into com-
“maximize”, but does not state a specific number. ponent measures for launch cost/kg and percentage
This is an excellent candidate because it has an in- mass from Earth.
herent range. An unbounded number of locations
would tend to result in an unbounded cost, so this
criterion would need to be balanced against cost cri- 5.0 Technical Risk
teria.
• 5.1 Risk Allowances - The requirement is to include
• 2.2 Growth - The requirement is to increase capac- allowances for technical risk in the design. Since
ity progressively, but without a number specified. the margins for unknown design factors is already
This is another good evaluation criterion candidate. included in performance and cost estimates, it would
We can define measures for absolute capacity, in- be double-counting to also evaluate those margins
crement size, and growth rates. directly as a criterion. A smaller weighted criterion
6.5. HUMAN EXPANSION 4 251

would be reasonable for the size of the margin, as • Internal Requirements Classes - We review the
less uncertainty in the design would be considered paragraphs on page 2, and note that “Improved
better than more uncertainty. Quality of Life” needs more specific measures. For
“Understanding the Earth” we note that data require-
6.0 Safety ments should trace to useful results which can be
acted upon, though how to measure that is problem-
• 6.1 New Location Risk - This is internal risk to the atic. Under “Biosphere Security”, ability to counter-
program contents and people. As a variable param- act changes can be a variable measure. Under “Ex-
eter, it is a good selection criterion. panding Resources” we can measure the increase in
economically viable available resources, including
• 6.2 Population Risk - This is external risks out- physical space.
side the program including natural hazards already
in place. Again, it is a variable parameter, and thus • Design Approach - From the list on page 2, we
a good candidate. can identify percentage of: closed cycle, local re-
sources, self production, and reduced human and re-
mote control inputs as variable components of the
7.0 Sustainability
improved technology criterion (2.3). For number
of locations (2.1) we can list temperature, rainfall,
• 7.1 Biosphere Security - Security in and of itself pressure, gravity, and radiation levels as environ-
is not a measurable parameter, but diversity of loca- ment parameters to expand upon. Time parameters
tions and species in alternate locations is a candidate. include communications, travel, and stay times. En-
• 7.2 Survivability - This is a requirement for long ergy parameters include potential of location, and
term survival. A measure would be the rate of com- available flux from natural sources. Another pos-
pensation vs expected time for a problem to become sible measure is how many parameters and by what
critical. amount is their range increased per location, or max-
imum total increase. For all these criteria, existing
8.0 Openness state of civilization is the baseline level.
• Program Concept - This does not appear to have
• 8.1 Open Design - The program will either be open new criteria.
or not, so this is not a good candidate. We will as-
sume that all design alternatives will be open.
General Needs and Desires
• 8.2 Access - This is another fixed requirement for
reasonable access, which will be included in any de- Here are some ideas about general human/civilization
sign option, therefore not an evaluation measure. needs and desires. The idea is that since civilization as a
whole is the “customer” for this program, it is their needs
This gives us a good starting list of candidate criteria. In and desires for which we should be designing, and there-
early conceptual design many of the values will be unde- fore include these types of items in the evaluation criteria.
termined for a given option. When not enough informa- The following items are drawn from online search, and
tion is available to make a clear choice, the proper course are not yet backed rigorously or empirically. Until better
is to keep multiple options until you have enough infor- defined, we are not using them as evaluation criteria.
mation to decide.
• Maslow’s Hierarchy of Needs - This is the psy-
chological theory originated in 1954 that people ad-
Additional Candidate Sources dress more basic needs before higher level ones.
The general levels are (1) Physiological, (2) Safety,
• Program Goals and Benefits - Looking back at (3) Love/Belonging, (4) Esteem, and (5) Self-
these on page 1, we find they are well represented Actualization. This theory has been criticized. A
by the requirements and don't immediately present recent paper by Tay and Diener may shed some
new candidate criteria. empirical light on this topic.
• Systems Engineering Experience - Reviewing the • States of Being - The urge to be part of some-
Systems Engineering section (1.5 of this book) does thing larger than oneself, engagement with an
not present new program level criteria. Many of the item or activity, improved personal productivity or
requirements categories will apply at lower levels. life, positive being or enjoyment, personal well-
being/meaning/fulfillment.
• External Constraints - Limits imposed by the nat-
ural environment or human rules are not variable pa- • Categories of needs and desires - Men-
rameters, they must be met. Therefore they are not tal/intellectual, emotional, and physical: see
candidates to compare options. Applied Empathy
252 CHAPTER 6. PART 5 - DESIGN STUDIES

• List of emotional needs - The need for, or need not every requirement has corresponding evaluation cri-
to be: accepted, accepting,accomplished, acknowl- teria, there are gaps in the numbering
edged, admired, alive, amused, appreciated, ap-
preciative, approved of, attention, capable, chal- • 1.2 Program Scale - Studies from animal popula-
lenged, clear (not confused), competent, confident, tions indicate below 100 members, long term via-
developed, educated, empowered, focused, for- bility is threatened. Human social research gives
given, forgiving, free, fulfilled, grown or growing, Dunbar’s Number for cohesive groups, ranging
happy, heard, helped, helpful, important, in con- from 30 up to about 2500 for various types of
trol, included, independent, interested, knowledge- groups. We therefore set 100 people per location as
able, listened to, loved, needed, noticed, open, opti- a minimum goal, gaining a score of 0%, with each
mistic, privacy, productive, protected, proud, reas- factor of e above that gaining 25% in score. Thus
sured, recognized, relaxed, respected, safe, satisfied, 5460 people per location yields a 100% score. Sim-
secure, significant, successful, supported, treated ilarly, we set the total for all locations on Earth at
fairly, understanding, understood, useful, valued, 5000 for a score of 0%, and 273,000 to 100%. We
and worthy. Everyone feels these needs in different have a preference for a diversity of locations rather
amounts. than one big one, since the overall goal is to expand
to a series of multiple more difficult ones. Therefore
• Look at empirical behavior to find out what people we will weight the total population goal 1.5 times as
really want, vs what they say they want when inter- heavily as the per location goal.
viewed. For example, working hours as income goes
up between countries, over historical time, or within
• 2.1 Number of locations - Humans already ex-
economic groups could be extrapolated to find out
ist in a range of environments. We will define the
what people would do if work was not required.
range where 90% of the population already lives as
the baseline, and increments of at least 10% be-
Selecting and Weighting Candidates yond that in at least one parameter as “new”. Al-
though 5% of the population already lives at each ex-
Now that we have established a set of candidates, prelimi- treme, when combined with other requirements for
nary as it is, the next step is to choose the most important improved technology, it leads to new designs. Pa-
ones, and establish relative scoring weights and conver- rameters which can fall to zero, like pressure or tem-
sion formulas. The weight is how much a given criterion perature have linear increments down to zero. With
contributes to the total score of a given design. The more unbounded values the 10% increments are com-
important that feature or parameter is, the more weight pounded ratios. Note that environment parameters
we give it. The conversion formula takes the feature are for the external environment, not the internal liv-
or parameter and converts it to a relative score, usually ing and working conditions. The following parame-
with a nominal 0 to 100% range, or 0.00 to 1.00 value. ters have been identified so far:
The range is arbitrary, as long as it is consistent within a
project. We will use the 0 to 100% range. The total score Environment:
is obtained from the weight x score for each component Environment temperature - range of daily high
criterion, then summing all the resulting products. and low across seasons in Kelvin (K).
We want to narrow the list to the most important ones be- Water supply - annual rainfall + running wa-
cause it takes too much time and effort to evaluate many ter/ice/air moisture flow in meters.
criteria for every design option. One way to do this is Atmosphere pressure - average value at the lo-
by making some of them fixed requirements, which all cation, in kiloPascals (kPa).
options must meet. Another is to simply drop the can-
didate as not important enough at the top program level. Ground pressure - foundation design load in
Whatever list of criteria is developed through this pro- megaPascals (MPa) or exterior water or rock
cess, they should be reviewed by the rest of the program pressure for below-surface construction.
participants and customer as best is possible. Setting cri- Energy supply - Flux from natural sources in
teria and importance is inherently subjective, so it is very W/m2 .
important to get agreement and consensus from the right Gravity level - only applies to space, in
group of subjective humans that these are the right set to meters/sec2 .
use in evaluating the design.
Radiation dose - measured by human biological
effect in Sievert/year.
Candidate Discussion The following discussion gives
our reasoning for the conversion from program parameter Time and Distance: These are
or feature to a score. We use the same numbering as for measured from 5th %ile nearest
the requirements to make them easier to compare. Since population on Earth.
6.5. HUMAN EXPANSION 4 253

Ping Time - minimum round trip communica- improve life outside itself. We give each doubling
tion delay, in seconds. of output a 25% increase in score, thus 200% out-
Travel Time - one way normal travel time for put (100% surplus) scores 25% and 15 times surplus
humans. (16 times output) scores 100%. We are choosing a
local measure of resources relative to the program
Stay time - average per person stay time per lo- rather than global availability. If civilization as a
cation in years. Increments count linearly from whole wants to increase its resources, it can copy
zero. the programs examples.
Transport energy - total potential, kinetic, and
frictional energy to reach the location for most • 3.1 Completion Time - This seems to duplicate 2.2
efficient cargo method, in megaJoules/kg. Growth as far as setting an overall time to reach a
final size. For the present we will not list it as a sep-
The two criteria we derive are the actual number of dis- arate criterion.
tinct new locations the program establishes, and the total
number of range steps expanded to by all the parameters • 4.1 Total Development Cost - For terrestrial loca-
combined. The first is measured directly at 1% per loca- tions we can set one-time (non-recurring) develop-
tion, and the second at 0.5% per range step. ment cost in the range of 10 to 100 times the unit
cost, on the principle that multiple copies of loca-
• 2.2 Growth - The fastest growing national GDPs are tions will eventually be built, and the one-time cost
about 10% per year, so we will give that a score of will be distributed. For space locations we expect
75% here. Average GDP growth is about 4%, so we fewer copies, and that some key technologies will
will adopt 5% as a score of 25%. We will use the av- have been previously developed for Earth. Thus a
erage of industrial output, resident population, and range of 1 to 10 times unit cost is more reasonable.
transport capacity to measure “output” of a location, Since lower cost is better, we invert the develop-
and growth rates relative to the final size of the lo- ment/unit cost ratio and multiply by 1000 and 100%
cation. So the minimum and maximum rates of 2.5 respectively to get a score scaled to 100%. For dif-
and 12.5% imply 40 and 8 years respectively to grow ficulty of the location beyond temperate, we allow
to final size, and sets minimum initial size as 1/40 of 10% new development cost for any environment pa-
final size. rameter step above those previously used. For resi-
dent capacity we scale by ln(actual/nominal) size of
• 2.3 Improved Technology - These are measured at 75 people, to account for larger or smaller elements.
the program level across all locations. The percent-
age of local resource use, self production, and cyclic • 4.2 New Location Cost - This is the explicit
mass flow all scale directly from 0 to 100%. Au- unit cost per new location relative to total output.
tomation is measured in reduction of human labor As written, the requirement overlaps with 2.6 Re-
hours per output relative to current technology. Au- sources, so we instead use an absolute cost per per-
tonomy is the percentage of internal human labor son, with the US total capital per person set to 50%
and control at the locations relative to the total re- score. Each factor of 2 up or down adjusts the
quired for the locations to function. These are both score by 25%. For space locations we allow twice
scored directly in percent. as much capital cost per person. We allow for the
difficulty of the location by adding 10% linearly for
• 2.4 Quality of Life - We will set a nominal GDP per each environment parameter step beyond the tem-
person of $60,000 in 2012 as the 25% score level. perate range.
This is about the average for the top 10% of world
population. There are many other potential quality • 4.3 Earth Launch Cost - This cost is the transport
of life criteria, but for simplicity we will use just to LEO component for space locations in $/kg of to-
this one for now. The highest single country GDP tal system mass. This includes mass obtained locally
is Monaco at about $180,000, so we will set that to in space. Because both actual transport to orbit and
100%, and scale linearly. use of local mass (currently 0%) can be greatly im-
proved, we use a steep scoring function. A current
• 2.5 Data - At this point of the conceptual design we baseline of $1600/kg gets a score of 0%, and each
do not know enough about data to set it as a criterion, factor of 10 reduction will score 20%. Thus the re-
therefore we put it aside for now. quirement goal of $0.08/kg will score 86%.

• 2.6 Resources - Our nominal requirement is 100% • 5.1 Risk Allowance - Less variation is better, so
surplus, and more is better. No surplus we assign we score this on an inverted scale. A technical
a score of 0%, since it is not meeting the desire to risk margin due to design uncertainty of 50% would
254 CHAPTER 6. PART 5 - DESIGN STUDIES

score zero, and a margin of 0% (which is only Weighting Discussion Next we discuss our reasoning
reached with completed and tested designs) would for the relative weights of the criteria. We will use a to-
score 100%. More advanced technology may give tal weight of 100 points for all the criteria together. Our
better potential performance, but with more uncer- weighting is subjective, based on human opinion as to the
tainty. This can be reduced by development and importance of design features and parameters. Design al-
testing, but that is in the future. Present uncertainty ternatives themselves are objective. So other people re-
is used to evaluate this criterion. viewing the program choices can simply change the scor-
ing and weighting to fit their own opinions on what mat-
• 6.1 New Location Risk - The goal is significantly ters. The design alternatives can remain the same, but
lower internal risk, although higher risks are accept- a different set of choices would result from the changed
able as a temporary measure while setting up. We evaluation scores.
scale this as equal to current general population risk Most people, all other factors being equal, prefer to get
gets a 50% score, and each doubling of risk lowers more results relative to the cost or effort expended. Since
the score by 25%, each halving of risk increases the the ratio of performance to cost can be equally affected
score by 25%. by increasing performance or lowering cost, the relative
weight of these criteria groups is often set to be equal,
• 6.2 Population Risk - These are risks to the ex- and a large part of the total weight. In our list above, the
ternal general population from program or natural performance type criteria are from 1.2 through 2.6, and
causes. Because the whole world is affected, it is the cost type criteria are 4.1 through 4.3. The remain-
difficult for one program to have much effect, so we ing criteria fall into the technical risk, safety, and sustain-
give this a narrow scoring function. The program ability categories. Historically, large complex programs
acts more as a demuonstration that the risk reduction tended to focus on cost and performance, and put rela-
is possible, and civilization can exert itself if desired tively small weight on other factors. With feelings of a fi-
to do more. Each 5% reduction to existing popula- nite and connected Earth, longer lives, and greater wealth
tion risk is worth 25% score. We base no change and standards of living, people feel there is more to lose
to population risk at 0% score since any increase in and so place more importance on the possible negatives of
total risk is generally considered unacceptable for a a project. We expect this trend to continue in the future,
new program. and this will be a long term program, so we will assign
30 points to this group, and divide the other 70 points
equally between performance and cost.
• 7.1 Biosphere Security - Maintaining biospheres
outside their natural environmental range increases Performance Group (35 points)
security by having backups and the ability to sur-
vive transient disruptions. Zoo breeding populations Scale (1.2) and number of locations (2.1) are the main
of endangered species and seed banks are examples motivations for a program of human expansion, so we will
of existing programs of this type. It is difficult to assign each 7.5 points. There does not seem to be strong
say how much of this activity is enough, so we will reason to give growth, improved technology, quality of
somewhat arbitrarily score total number of species life, or resources (2.2-2.4, 2.6) more or less weight among
x locations. For each factor of 10 increase start- themselves, so we assign 5 points to each.
ing with 10 we will add 20% to the score. Thus
100,000 species in 10 locations is 1 million total Cost Group (35 points)
species-locations, and would score 100%.
Development (4.1) and new location (4.2) cost seem
equally important, but Earth launch cost only applies to
• 7.2 Survivability - Like population risk, a single
space locations, so we give it half the weight. Therefore
program cannot guarantee long term survival by it-
the weights are 14, 14, and 7 points each.
self, so we set a narrow scoring function. For each
5% compensation for long term change and resource
Technical Risk, Safety, and Sustainability
depletion the program reaches during its life it gets a
Group (30 points)
25% score. For a change like the Earth overheating
due to the Sun, which might take millions of years,
We subjectively rate location (6.1) and population (6.2)
only the change which occurs during the program
risk more highly than the remaining factors, and thus give
duration (perhaps 50 years) is being compensated
them 7.5 points each. Technical risk (5.1), biosphere se-
for. For critical resources, only those without which
curity (7.1), and survivability (7.2) then get 5 points each.
civilization cannot function are considered. Ones
with reasonable alternatives are not critical. Com-
pensation can be by active measures, like shading Resulting Evaluation Criteria From the above dis-
the Earth from overheating, or by alternatives like cussion, we can now make a table of the resulting eval-
moving to other planets. uation criteria to apply to our design options. Note that
6.6. HUMAN EXPANSION 5 255

in some cases scores can go outside 0 to 100% range if the Environment Ranges
parameter is outside the expected range. Like all parts of
the conceptual design, this may get revised by later work. The Temperate range is defined as where 90% of peo-
ple currently live, and more difficult and extreme envi-
continue to page 5
ronments are respectively > 10% and 20% beyond that.
Analysis of where people live vs environment conditions
is being developed separately in the Sec 5.2 - Environ-
6.6 Human Expansion 5 ment Ranges Design Study. This is being done in two
steps due to the mass of detail in mapping out where peo-
ple live. The following table lists initial estimates for the
back to page 4 boundaries of the Temperate range, and the inner bound-
ary values for the difficult and extreme ranges. When the
second part of the Environment Range study is complete,
6.6.1 Functional Analysis these will be updated.

In a conceptual design such as this we want to establish

6.6.3 Existing Baseline Program
one or more program concepts. The concepts describe
in a general way how the program will operate. Then we
Our existing baseline can be defined by looking at world
begin a process of breaking down this general descrip-
development trends and existing space programs to the
tion into more detailed steps called Functions. Functions
point that uncertainties in items like GDP or new technol-
generally transform a set of inputs into a set of outputs.
ogy are larger than 50% of our estimates. At that point
These inputs and outputs are called Flows, and may in-
the future is too uncertain to project reasonably, therefore
clude any kind of entity: humans, data, hardware, energy.
we stop. Future updates to this kind of conceptual design
Functions do not define how the transformations happen,
study, done in 5 or 10 years time, can then project forward
just that they occur. The how is embodied in a design
to a new uncertainty horizon. Other program concepts
solution or alternative, which will come later.
will also have a time horizon bounded by uncertainties,
Human civilization already exists, and is expanding in and should explicitly identify what those limits are. In ex-
measures of population, energy use, and GDP. There are amining world development and existing space programs,
already space programs that also exist. So the first pro- we do not need to examine their every detail, but only the
gram concept, which we will call the Existing Baseline is parts that affect the scoring in the evaluation criteria we
to simply continue current activities without adding any- have chosen.
thing new. Any other program concepts will need to score
better than the baseline to justify going forward with
them. We described an initial program concept based World Development Trends
on a particular design approach on page 2. To do a thor-
ough job, we should examine other design approaches to Existing Space Programs
see if they lead to promising concepts. Given that a sin-
gle author is contributing this section at present, this will Evaluation Score
need to be limited in scope to examining approaches al-
ready proposed by others. There is an opportunity here We take our previous table of evaluation criteria, and ap-
for other individuals or teams to add to this work with ad- ply them to the current baseline:
ditional approaches and concepts. The important thing is
to compare alternatives on a fair basis, using similar goals,
evaluation criteria, technology levels, and cost estimating 6.6.4 New Program Concept
methods.
Our new program concept was first stated in a general way
on page 2. The basic approach was to enable expansion by
developing advanced technologies, then build a series of
6.6.2 Current Parameter Status new locations in more difficult environments, and within
each location increase in size and technical performance.
Some of our criteria for comparing programs are scored The obvious functional breakdown is then first by level of
relative to features of existing civilization, so as a starting technology, which defines what is possible to build, then
point we will identify the current status of these parame- the locations, defined by a set of environment conditions.
ters. The initial version of this functional analysis is being As better technology becomes available, the locations are
written in late 2012, so we will use the start of 2013 as upgraded to targeted size and performance levels. A “lo-
“current”. If the entire conceptual design process takes cation” is a general environment, such as tropical ocean
a long time, the information should be updated to a new or Low Earth Orbit. Selection of specific sites is left to
current date. 2015 is suggested for the next update. later. To determine how many phases of upgrades are
256 CHAPTER 6. PART 5 - DESIGN STUDIES

needed, we will optimize based on what can be reached is sufficient improvement possible, periodically feeding
in a first phase, and what size steps are reasonable. As a new improvements to later locations and phases.
goal, we think a phase should provide at least a 10 point Prototype Scaling
increase in evaluation score, to be significant enough to
implement. It is less expensive to build smaller prototypes to try out
new technologies and demonstrate performance, so we
The primary functions within a location include produc- will establish a series of mass and linear scales derived
tion, habitation, and transport capacities. These allow it
from the “full scale” level of supporting 75 people/year.
to support people at the location, and interact with the rest Not everything can be prototyped this way, but it will be
of civilization. We will develop functional flow diagrams used where effective. Habitation is an obvious example
to model these elements and the flows that connect them where scaling in all dimensions is not feasible, although
to each other and outside the program. scaling in area is possible. The scale steps get smaller
as size goes up due to increasing cost. Consistent scal-
ing steps helps ensure different items will work together
Expansion Phases
when integrated. Initial sizes of prototypes and which
scale steps to use will be determined for each technology.
Our partial evaluation score for existing civilization is There is an opportunity to sell copies of the smaller scale
about 75 points, so we will attempt to formulate a loca- elements, or do initial production and operations with
tion concept that yields a total score of 85 for Phase I. them, as ways of generating income for further growth.
Later will will adjust and optimize this concept. For now
we just want a single example to start with. Later phases The scales are as follows:
will try to increase in steps of 10 to 105 points in Phase
III. We don't expect to be able to plan beyond Phase III at • 1/10 Scale - This is 10% linear, or 0.001 mass and
this early date, since implementing that far in the future volume, or support 0.075 people/year capacity.
involves too much uncertainty.
• 1/5 Scale - This is 20% linear, or 0.008 mass and
volume, or support 0.60 people/year capacity.
Develop Technology (Phase 0) Before we implement • 1/3 Scale - This is 1/3 linear, or 0.037 mass and
the first phase, we have to develop sufficient technology volume, or support 2.75 people/year capacity.
to reach the desired program goals for that phase. This
includes doing conceptual and preliminary design, devel- • 1/2 Scale - This is 50% linear, or 0.125 mass and
oping new technologies, and building prototype systems volume, or support about 9-10 people/year capacity.
to demonstrate performance. This preliminary phase we • 3/4 Scale - This is 75% linear, or 0.422 mass and
will call Phase 0. Assuming this phase shows enough volume, or support about 30 people/year capacity.
gains over the current baseline, we then proceed to the
next phase. • Full Scale - This is 100% linear, or 1.000 mass and
volume, or support 75 people/year capacity.
Conceptual Design
Preliminary Design Later growth may require larger prototypes, and multi-
New Technologies ple copies of a given size may always be used for more
capacity.
It is not necessary to duplicate work for technologies
which already have large efforts in progress. Thus elec- An example of scaling is reducing a 36 inch x 21 ft ca-
tronics, for example, is not an area we would put much pacity, 30 hp electric commercial sawmill down to 0.5 hp
effort into. It is already a major industry with high lev-electric (a factor of 60 in power), and reducing log capac-
els of funding for technology development. Instead we ity by 2.5 in each axis (a ratio of 15.625 in log volume),
will apply efforts to areas specific to the program, and giving a total reduction of 937.5, roughly the 0.001 mass
and volume scaling for the smallest size. In US terms the
which are not getting sufficient attention. The selection of
which technologies to work on first will depend on rank- log capacity is 15” diameter by 8 ft long or 12” diameter
by 12 ft long. The smaller motor is suitable for slower
ing their relative potential impact, development difficulty,
and timing of the need for it. production rates, and the smaller size is suitable for home
use. 600 lb log weights can be handled by a single person
As a new technology reaches a sufficient level of improve- using leverage at each end singly.
ment, it will migrate from this task to prototyping. If
it performs well enough, it will then move to one of the
later implementation phases. If more progress appears Develop Phase 1 Locations The following program
possible, a given technology goes back into the devel- parameters result by working backwards from an 85%
opment cycle, and the amount of effort based again on score for each evaluation criterion, or 10% higher than
ranking vs other technologies. Thus Phase 0 does not the baseline. They are goals, the actual values will be
end once Phase I starts, but continues as long as there found at the end of the conceptual design.
6.6. HUMAN EXPANSION 5 257

• 1.2 Program scale: Average population/location = people and grows by 75 per year to 660. Annual popula-
3000, total population = 150,000. Number of loca- tion growth increases by 11% per year, so added locations
tions = 50. We will assume an inverse size distri- are started when enough margin over first location exists,
bution where size for the nth largest location is 1/n until all 50 locations are built, nominally ~50 years.
times the largest location. For 50 locations the to-Technology development is assumed to continue for 6.5
tal is 4.5 times the largest. This comes from Zipf’smore years to initiate Phase II, and 7 more years to initi-
Law, an empirical observation for city sizes. This ate Phase III, so later sites will use upgraded technology,
gives a smallest location of 660 and a largest of and older sites retro-fitted to improved levels. Budget for
33,000.
prototype work is $66 million, resulting in ~36 resident
• 2.1 Number of Locations: Actual count of loca- capacity.
tions is 50, giving 50% score. We compensate by
increasing range of environments to 240 steps, giv-
ing 120% score. Later Phases At this point it is too early to try and de-
fine later phases in much detail, aside from setting evalu-
• 2.2 Growth: 11% per year, giving 9 year time to ation scores of 95 and 105%.
completion, and 75 people as minimum location de-
sign size. [Following text saved from Section 4.1:]

• 2.3 Improved Technology: 85% direct values for

local resources, finished products, recycled fraction, Level 2: Phases
automation, and autonomy.
The program phases are defined for now by aiming for
• 2.4 Quality of Life: Equivalent GDP (counting in-
a 10 point increase in evaluation score per phase. Early
ternal production) = $156,000 US
estimates of the existing baseline indicate a score of 20
• 2.6 Resources: 10.5 x internal materials and energy points, so Phase I would aim for a score of 30, Phase
over life cycle, or 950% surplus. II for 40 points, and Phase III for 50 points. Once the
design alternatives are better understood, and what per-
• 4.1 Total Development Cost: Allow 11.7x unit formance is feasible, the number and spacing of phases
cost on Earth, 1.2x in space. Allows $890,000 de- may be changed later.
velopment/person for temperate location, +10% per
Numerical Goals
environment step compounded for more difficult lo-
cations, and ln(size) for increased size of same en-
vironment.
Phase I Phase I aims at a 10 point improvement over
• 4.2 New Location Cost: Per person = $76,000 the existing baseline. The baseline is estimated to score
Earth, Space = $152,000. 20 points, thus this phase aims at 30 points. The exact
program parameters to reach this score depend on many
• 4.3 Earth Launch Cost: $23/kg including space lower level choices still to be made, and technology still
resources factor. Nominal split is $150/kg actual to be proven. We know that this phase will involve some
launch cost and 15% non-space resources factor. number of Earth and Near Space locations, and some
• 5.1 Technical Risk: 7.5% technical uncertainty level of improved technology.

• 6.1 New Location Risk: Allow 38% casualty risk

for new locations Phase II Phase II would aim at a further 10 point in-
crease in program score. This will likely require more
• 6.2 Population Risk: 17% reduction to population
technology development, and because of the time span
risks
since the preliminary phase, some redesign and upgrade
• 7.1 Biosphere Security: 178,000 species x loca- of Phase I elements will likely be needed.
tions maintained outside natural range

• 7.2 Survivability: 17% compensation for critical Phase III Phase III is currently aimed at a program
risks score of 50 points. Because it is further out in time and
difficulty, this phase is left as a more preliminary concept,
Schedule more to guide the direction of the earlier phases. Unan-
We assume a notional schedule which ramps up gradu- ticipated new technology is likely to affect designs this far
ally. Assuming a “zeroth” location for prototyping we in the future. Phases after this one (IV+) are therefore re-
allow 6.5 years for Phase 0 to develop the technologies, served for future design work.
after which the first Earth location starts at a scale of 75 [End saved text]
258 CHAPTER 6. PART 5 - DESIGN STUDIES

Locations List compounded increments. Signif-

icant tropical areas fall into the
This list is tentative pending completion of the environ- Non-Temperate range from rainfall
ment ranges study. alone, and some areas near major
rivers reach the highest values.
• Temperate Earth - Our first environ-
ment is within the middle 90% range - Low Water Locations - Nom-
of current civilization. The reason is to inally these are areas that get
first develop improved technology, such less then 0.25 meters (10 inches)
as seed factories and cyclic flows, where of rainfall or other fresh water
it is easiest to do the work, and where it sources/year. The driest location on
will have the widest immediate applica- Earth, the Atacama Desert in Chile
tion. gets as little as 0.001 meters/year,
so there are 10 steps in dryness be-
low the Temperate range, in 0.025
• Non-Temperate Earth - This group has
m/yr increments. Significant parts
one or more parameters outside the Tem-
of the Sahara Desert fall into the
perate range, but not reaching the Diffi-
lowest step.
cult level. They are named by whichever
parameter is most out of range. Many
combinations are possible. • Difficult Earth Locations - This group
of locations push one or more environ-
ment parameters more than 10% beyond
- Hot Locations - These are where the temperate range. Based on our en-
summer daytime highs exceed vironment parameters, we can start to
310K (37C). Death Valley, Cal- identify such locations, and then combine
ifornia, generally considered the ones where multiple parameters can be
hottest place on Earth, reaches addressed at once. All the ranges are
47C for average summer daily based on what the upper and lower 5%
high, so no place on the surface of current population live with:
reaches the difficult level of 341K
(68C) except for volcanic areas.
• Extreme Earth Locations - This group
The TauTona gold mine in South
includes as many locations as needed to
Africa, which reaches 4 km below
push environment parameters to the lim-
the Earth’s surface, has a rock face
its of practicality. Some parameters may
temperature of 60C, so it also does
have no practical use beyond a limit-
not reach the difficult level.
ing value even if conditions exist beyond
them.
- Cold Locations - These are
where winter nighttime lows are • Near Earth Space - These start with the
below 260K (−13C). Vostok Sta- lowest useful Earth orbits at about 200
tion, Antarctica, is considered the km altitude and extend upwards to 10%
coldest place on Earth, and reaches beyond Earth escape energy.
an average winter low of 201K
(−72C). This is below the extreme
• Distant Space
low threshold of 208K (−65C), so
the Earth includes Non-Temperate,
Difficult, and Extreme Cold Loca- Functions
tions.
Figure 5.1-1 is a very preliminary diagram showing the
- High Water Locations - Nomi- functional elements and flows for a generic location.
nally these are areas that get more Functions must trace back to at least one program goal or
than 2.5 meters (100 inches) of requirement, otherwise they are unnecessary. This can be
rainfall or other fresh water sources, either a direct reference to a source, or an indirect deriva-
such as river flow, per year. Since tion by analysis. For the elements in this diagram at least
the maximum water supply reaches one source each that justifies their inclusion are:
about 10 meters/year, there are
about 14 steps above the Tem- • Diagram as a whole - Multiple locations will make
perate range on Earth, in 10% up the total program, so the entire diagram is an el-
6.6. HUMAN EXPANSION 5 259

be done in different ways, and often is to develop alter-

nate designs. A good understanding of the nature of the
system is very helpful in developing the lower level func-
tions, which in turn may require specialist knowledge.
As a start, we can create one list of lower level functions
drawn from past experience. They will likely strongly in-
teract with each other, with many flows between the sub-
functions, so a flow diagram may be too complicated to
use. We will consider a table or spreadsheet instead. We
also identify categories of inputs to and outputs from the
Figure 5.1-1 - Generalized development model. location as a whole. A location is connected physically
to its environment, and interacts with other locations and
civilization as a whole, so flows crossing the location logi-
ement in a higher level diagram. The need for loca-
cal boundary are part of the analysis. The inputs and out-
tions comes from 1.1 Program Goal - "...a series of
puts will then later be divided among the more detailed
new locations...”.
functions.
• Provide Production Capacity - comes from 2.3
Improved Technology - "...increase the levels of self
Location Inputs
production...”

• Provide Habitation Capacity - comes from 1.2 • Energy Sources

Program Scale - "...permanently supporting at least
95,000 humans total among new Earth locations and • Food Sources
at least 2,000 humans per new space location.” The
• Water Sources
physical support to live in a given location we will
call Habitation. • Parts and Materials Supply
• Provide Transport Capacity - comes indirectly • Tools and Machines Supply
from 1.1 Program goal "...expand human civiliza-
tion...”. The new locations are not cut off from exist- • Land Inputs
ing civilization or each other, therefore they need ca-
pacity to transport people and supplies in, and prod- • Human Inputs
ucts out. It comes more directly from "...a series of
new locations...”, since the mere existence of new • Money Inputs
locations requires transport to set them up.
• Information Inputs
Flows into and out of the diagram, and between func-
tions can contain any kind of hardware, software, data, Location Outputs
or people. Later analysis will define exactly what each
flow contains, but they must follow the rule that flows are
• Surplus Energy
conserved. This comes from the physical fact that items
do not appear from or disappear into nothing. Thus di- • Surplus Food
viding or combining flows must sum to the same totals
on both sides, and so must inputs and outputs to a func- • Surplus Water
tion (although a function may convert the types of flows).
Conservation of flows ensures that all inputs and outputs • Surplus Parts and Materials
of a system are considered and accounted for. In this pre-
liminary diagram we merely identify some of the major • Surplus Tools and Machines
flows.
• Surplus Land
To continue the functional analysis, we break down the
three top-level functions into lower level elements. Par- • Surplus Humans
titioning functions into more detailed ones creates logi-
cal boundaries inside a larger system. This then identi- • Money Outputs
fies flows which cross the internal boundaries, and cre-
ates simpler elements to design. The lower level func- • Information Outputs
tions should have an internal coherence or relatedness.
Since the partitioning is logical, and not physical, it can • Waste Outputs
260 CHAPTER 6. PART 5 - DESIGN STUDIES

Production Functions • 1.1 Program Goal - The program shall expand hu-
man civilization to a series of new locations with in-
• Control Location creasingly difficult environments and distance.

• Supply Power • 1.2 Program Scale - Expansion shall be demon-

strated by permanently supporting at least 95,000
• Extract Materials humans total among new Earth locations and at least
• Process Materials 2,000 humans per new space location.

• Fabricate Parts • 1.3 Choice - Specific locations and their internal

organization, function, and operation shall be cho-
• Store Inventory sen by program participants and location residents
within the limits of design constraints.
• Assemble Elements
• Grow Organics 2. Performance

• 2.1 Number of Locations - The design shall maxi-

Habitation Functions
mize the number of new locations, where new is de-
fined by at least a 10% increase in an environment
• Protect From External Environment
parameter or distance measured in time or energy
• Control Internal Environment terms.

• Provide Food • 2.2 Growth - Each location shall increase the ca-
pacity for production, habitation, and transport in a
• Maintain Health progressive manner.
• Provide Personal Items • 2.3 Improved Technology - Locations shall in-
crease the levels of self-production, cyclic flows, and
• Provide Information
autonomy in a progressive manner.
• 2.4 Improved Quality of Life - Completed loca-
Transport Functions
tions shall provide an improved physical and social
quality of life relative to the upper 10% of Earth civ-
• Deliver Bulk Cargo
ilization.
• Deliver Delicate Cargo
• 2.5 Data - The program shall collect and dissemi-
• Deliver Humans nate [TBD] data about the Earth’s environment, sur-
rounding space, and objects therein.
[Parked Content from Section 4.1]
• 2.6 Resources - The program shall output a life cy-
cle surplus of at least 100% of internal material and
6.6.5 Initial Program Analysis energy resource needs.

Program Requirements 3. Schedule

The initial set of program requirements were developed • 3.1 Completion Time - The expansion to a new lo-
by carefully looking at the program goals and benefits, cation shall be completed before expected progress
general systems engineering experience, and natural and in technology indicates a re-design is required.
human constraints. We divided the general goals into
more specific statements with measurable parameters. 4. Cost
We also looked at our ideas for approaching the design,
and the first list of elements to be included in the pro-
• 4.1 Total Development Cost - The total program
gram, to see if they yield any requirements. We com-
development cost for new technology and hardware
bined and formalized the resulting statements to create
designs shall be less than 50 times the unit cost on
a first draft of the top level requirements. Some of the
Earth, and 5 times the unit cost in space of the hard-
numerical values are arbitrary, but we need to set some-
ware.
thing as a starting point, which can be adjusted later as
the design evolves. • 4.2 New Location Cost - The peak net project cost
for a new location shall be less than 50% of the ex-
1. Objectives pected long term net output.
6.7. ENVIRONMENT RANGES 261

• 4.3 Earth Launch Cost - The program shall pro- these parameters by again carefully looking at all the work
gressively lower the Earth launch cost component of done so far, and selecting the ones most important at the
total system cost, with a goal of $0.08/kg of total program level. After selection, we then scale and adjust
system mass. their relative importance to each other so that a score can
be determined for each design option or variation. Our
5. Technical Risk resulting criteria and how they are scored is as follows:

• 5.1 Risk Allowances - Program designs shall in-

clude allowances for uncertainties and unknowns in 6.7 Environment Ranges
knowledge, performance, failure rates, and other
technical parameters. New designs with higher risk
can be included in program plans, but a process shall
be included to resolve the risk, and an alternate de-
sign with lower risk maintained until resolved. 6.7.1 Introduction

6. Safety This design study is part of the Human Expansion (HE)

program conceptual design. One of the HE program
goals is to expand civilization into more difficult loca-
• 6.1 New Location Risk - New locations shall pro-
tions, while at the same time improving levels of tech-
gressively lower internal risks to life and property,
nology and quality of life. The goal in this study is to
with a goal of significantly lower risk than the gen-
establish location parameter ranges for where people live
eral population.
now, and then from that derive what difficult and extreme
• 6.2 Population Risk - The program shall signifi- ranges would be for future locations. We set the mid-
cantly reduce natural and human-made risks to the dle 90% of where people live now as the existing range
general population, including external risks created for each parameter, with 5% living at each extreme. The
by the program. middle 90% is described as Temperate, and levels more
than 10 and 20% beyond the Temperate range we call
7. Sustainability Difficult and Extreme respectively.
The study will be performed in two parts. The first part
• 7.1 Biosphere Security - The program shall in- will make make initial estimates so that other parts of
crease biosphere security by establishing alternate the HE program design can start to use the information.
biospheres and long term storage of biological ma- The second part will survey where people live in more
terials. detail and generate improved values. The HE program as
a whole is using the start of 2013 as a baseline to refer-
• 7.2 Survivability - The program shall design for the ence against. The US Census Bureau projects popula-
long term survival of life and humanity from changes tion at the start of 2013 to be 7,062 million, thus 5% is
to the Earth which will render it uninhabitable and 353 million. The more detailed population analysis will
depletion of critical resources. draw from world statistical data.

8. Openness
6.7.2 Initial Estimates
• 8.1 Open Design - Technology and design meth-
Seven environment parameters and four time and dis-
ods developed within the program shall be open for
tance parameters were chosen in the HE Requirements
others to use. Specific instances of a design and pro-
Analysis to determine what makes a difficult or extreme
duced items may be proprietary.
location. In the list of the parameters that follows, we
• 8.2 Access - Development of a new location shall first define what the parameter is, then discuss how we
not prevent reasonable access for transit or to unused make our estimate for the 90% Temperate Range. We
resources. also discuss what the practical limits on Earth are for the
Extreme ranges. Once the practical limits on Earth are
reached, or they exceed what is found in space, more dif-
Evaluation Criteria ficult locations in space are used.

Setting discrete program requirements like the ones listed

above are unlikely to be the optimum values, and do not Environment Temperature
help in choosing among design alternatives. For those
purposes we choose parameters to measure our evolving Definition - On Earth, this is the winter daily air tempera-
design and guide it to the preferred result. We identify ture lows, and summer daily highs in Kelvin and (Celsius).
262 CHAPTER 6. PART 5 - DESIGN STUDIES

For space locations this is the equilibrium temperature of Estimate - On Earth nearly everyone lives in a fairly nar-
a 50% gray body with one side facing the Sun and the row range of pressures. We will assume for now the nom-
opposite side facing away. inal range is from 80 to 100 kPa (sea level to 2000 m
Estimate - On inspection of a world map, many more altitude).
people live in far north locations than far south. North- Limits - On Earth the lowest surface pressure is pre-
ern Europe, Russia, parts of Tibet, Northern China, and sumed to be at the highest point, Mt. Everest, at +8,850
Canada may supply 350 million cold climate residents. m elevation. The highest pressure is presumed to be at
Many people live in equatorial/hot climates, but the ex- the bottom of the LaRonde Gold mine in Quebec, which
tremes are seen in dry climates where water vapor does reaches 3 km below a surface elevation of 300 m, for a
not moderate daily temperature ranges. We make a rough net height of −2,700 m. The pressure range is thus from
estimate of 260 K (−13 C) as average January daily lows 31.5 to 138 kPa. To go beyond these limits would require
and 310 K (37 C) as average July daily highs as the limitsartificial structures, or ambient pressure underwater loca-
of the Temperate range. tions. In space, local pressure can range from complete
Limits - Ten percent, or 25-30K, beyond the temperate vacuum to many times Earth values on Venus or the Gas
limits is a significant range for human climate, so Earth Giants.
locations are likely to to have only one or two very cold
and no hot difficult surface climates. Deep underground
might reach extreme hot levels in certain locations.
Ground Pressure

Water Supply
Definition - On bodies with gravity this is the foundation
design load in MPa or exterior water or rock pressure for
Definition - This is fresh water supply from rain and
below surface locations. In both cases this amounts to
snow fall, flows from rivers and ice, and air moisture
a boundary pressure on location structures. Low range
in meters depth/location area/year. Since the density of
would be surface water structures or swamps, and high
fresh water is 1 ton/m3 , this is equivalent to tons/m2 .
range would be tall buildings or below surface locations.
Salt water does not count as fresh water supply, although
evaporation does. Underground aquifers that represents Estimate - We make a rough estimate of the Temperate
flows from rain or water transport count, but not amounts range by referring to allowable soil bearing pressures from
merely drawn from storage, since that is not sustainable. soft clay up to hard bedrock. Dropping extreme values,
Many people live near rivers, which supply high fresh wa- these give a range from to 0.25 to 2.0 MPa.
ter flows. We count available river flow distributed over Limits - For low range limit, we can assume a shallow
the local land area, so a high range location above that boat or raft foundation with a depth of 1 meter or less.
may be difficult to find. Dry locations with low water This gives a load of 0.01 MPa. For high range limits,
flow are common. assuming a floor + dead load of 1000 kg/m2 per floor,
Estimate - We use values of 0.25 and 2.5 meters on Earth a 100 story skyscraper might impose a bearing pressure
as the Temperate range based on rainfall definitions of of 1 MPa on the foundation base. Practical limits on tall
desert and wet climates. This needs more support from construction have more to do with economics than with
climate data. structural materials, so taller buildings are possible.
Limits - On Earth, the lowest rainfall location, the Ata- Since the depths of the ocean and deep underground can
cama Desert receives an average of 0.001 meters of rain- reach very extreme pressures and temperatures, we set
fall/year. We can get an upper estimate by assuming the a practical limit based on the energy to reach Earth or-
Amazon River flow of 200,000 cubic meters/second is bit (31.3 MJ/kg) compared to the column of mass which
allocated to the nearest 10% of the drainage basin to the must be displaced to access such deep locations. For a
main river and tributaries. Given a total area of 7 million 1 kg volume of ocean, which has a mass of ~1 kg, dis-
km2 , that gives a water supply of 9 meters per year. The placing a column 800 m high requires raising 8000 kg x
highest rainfall locations give about 11 meters/year. So 400 m avg height x 9.8 m/s2 = 31.3 MJ. For continen-
we will adopt 10 meters/year as the extreme upper value. tal crust, which has a density of 2.7kg/liter, a 1 kg vol-
Space locations would range from essentially zero in the ume then is 7.2 x 7.2 cm. Each meter of rock above this
inner Solar System, to moderate when significant num- area then is then 13.925 kg. A column 675 m deep then
bers of ice or hydrated bodies pass within reach, to very masses 9400 kg, and raising it an average of 337.5 m also
high for ice-covered satellites or bodies. yields 31.3 MJ/kg. Therefore we will consider the practi-
cal limits to be ocean depths to 800 meters and continen-
tal depths to 675 m, after which we start including space
Atmosphere Pressure locations. Other parameters may drive us to space before
depth does. The corresponding water and rock pressures
Definition - The average location gas pressure in kPa. are 8 and 18 MPa respectively.
6.7. ENVIRONMENT RANGES 263

Energy Supply Earth Orbit locations range from 80-160 mSv/yr, and
locations above this can reach much higher levels from
Definition - Flux from natural sources in W/m2 . On trapped particle belts and Solar particle events.
Earth this will mostly be from solar and wind, with water
flow and geothermal in some locations.
Ping Time
Estimate - On Earth, average solar flux ranges from 100
to 300 W/m2 after accounting for night, clouds, and sun Definition - Minimum round trip communication delay
angle. Wind power in the US ranges from below 100 to to next nearest 5% of human population, in seconds. De-
over 1000 W/m2 . Hydroelectric, tidal and sea currents, pending on location, this may be by wired or wireless
and geothermal are localized, so we will ignore them for methods.
this first estimate. Combining wind and solar we get a
Estimate - In theory all of the Earth is within 0.13 sec-
rough estimate of 150 to 900 W/m2 as the temperate
ond ping time from any other point, but actual communi-
range. In space, solar flux is 1366 W/m2 at the Earth’s
cations systems in place increase this value. We assume
distance, times the percentage time in sunlight.
at least voice/cellphone bandwidth for communications.
Limits - Any significant water or ground depth will have There are now more fixed and mobile telephone connec-
near zero energy supply. Peak Earth values might be at tions than there are people, so again, in theory, everyone
high altitudes where a combination of high winds and has access to one. From Slovakia, around 5% of the world
increased sunlight would supply high power levels. In population is within about 600 km. Theoretical ping time
space, peak energy level will vary dramatically with dis- is then 8 milliseconds (ms). The distance from Tahiti to
tance from the Sun. major population centers is roughly 6600 km. Therefore
best case ping time is 44 ms. Given actual routing, and
fiber-optic speeds, this is likely to be more like 100 ms.
Gravity Level
Limits - The worst case on Earth is no worse than loca-
2 tions like Tahiti, which require submarine cable, or satel-
Definition - Local gravity level in meters/s . Space lo-
lite relay through the Iridium network. Thus the limits
cations near large objects still has a gravity level, even
for Earth are the same as for the Temperate range. Space
though orbits may create free-fall conditions with low rel-
locations can range to 45,000 seconds from Earth to the
ative forces between system elements.
Kuiper Belt.
Estimate - The Earth’s surface, where almost everyone
lives, is entirely within 10% of the 9.80665 m/s2 standard
value. Travel Time
Limits - Values more than 10% outside the Temperate Definition - Maximum one way normal travel time for
range require extremely tall structures ( > 320 km tall ), or humans, in hours or days, from the nearest 5% of human
orbital locations. Thus this parameter is effectively fixed population.
for Earth locations. Solar system locations can range from
zero to about 200 m/s2 very close to the Sun. Estimate - In dense areas such as Europe the maximum
direct distance to reach 5% of the world’s population is
about 600 km. Allowing for actual highway routing and
Radiation Dose travel speed, we can estimate 8 hours to reach any point in
that radius. Ninety percent of the world’s land area can
Definition - Human radiation exposure in an unprotected be reached within 48 hours according to the EU Joint
state from background radiation, in milliSievert/year. It Research Centre map of accessibility.
should be noted that in many locations humans cannot Limits - The same accessibility map places parts of the
survive in an unprotected state, but for consistency the Tibetan plateau at 20 days travel to a city of any size.
exposure level is measured that way. In addition to back- Point Nemo, the farthest point in the world from land, is
ground radiation, humans also get significant exposure about 4000 km by sea from the nearest airports. There-
from medical and other human-made sources, but that is fore it is about 10 days travel by chartered ship. Cur-
not location-specific, so is not counted as part of this lo- rent travel time to space is measured in months to years,
cation parameter. mostly because trips carrying humans are rare. The ac-
Estimate - Most of the Earth’s population lives in a rel- tual transit time from the launch site is 2-3 days
atively low radiation environment, but higher altitudes or
high natural radioactivity areas exist. Data derived from a
Stay Time
UN report indicate the typical range is from 1-13 mSv/yr.
Limits - The city of Ramsar in northern Iran has back- Definition - Average per person stay time per location, in
ground levels up to 135 mSv/yr due to underground con- years. People are assumed to stay in the same location if
centrations of Uranium and mobile decay products. Low they live and sleep in the same place on a regular basis.
264 CHAPTER 6. PART 5 - DESIGN STUDIES

Estimate - In the US, in 2010-2011, 3.9% of the popu- Coldest Populate Areas
lation moved to a different county in one year. We will The following table of cold locations starts at the poles
take a county to represent a “location”. Therefore the av- and highest elevations and works down until we reach 353
erage stay time is 25 years. We assume there are signifi- million population.
cant parts of the world where the average person does not
move far from their place of birth, thus the average stay
time in one location is their life expectancy of about 70
years.
Limits - The fastest growing counties in the US in 2011 Coldest Locations
had a growth rate of 10%, and thus an average stay time of
These are locations where few or no people live, with ex-
7 years if both growth and normal mobility are counted.
treme environment conditions:
Stay times in space are currently much shorter than this
because there are no permanent residents, and time is lim-
ited by zero gravity and radiation exposure. We invert the
time scale for space, and set a minimum stay time of 10%
of Earth, or 0.7 years, as the initial range. At the other
extreme, stay times are bounded by life expectancy, so
are no higher than 70 years. 6.8 Open Source Space Program
This is a conceptual design stage study of the need for,
Transport Energy
justification, and design of an Open-Source Space Pro-
Definition - Maximum total energy to reach a location gram. The Systems Engineering process can be applied
from nearest 5% of population, by most efficient method, to any complex system, including the design of organi-
in MJ/kg. Includes kinetic, potential, and friction energy zations which in turn design space projects. The study
cost. both serves as a tutorial example for such design studies
in general, and a useful work product. Even a negative
Estimate - For Earth, since most of the surface is near the result, that open source space programs are not needed
same potential, this is mostly frictional losses for rail or or justified, is useful in eliminating a poor option if that
ocean shipping. Rail transport consumes 330 BTU/ton- turns out to be the result.
mile, or 216 J/kg-km. Assuming a total transport distance
of 1000 km for dense areas, including partial road travel Status: the study is approx. 3% complete as of 03 August
from nearest rail point, we get about 0.215 MJ/kg. Ships 2012.
consume 375 J/kg-km. Eastern Australia is about 6000
km by ship from the nearest 5% of the worlds population,
giving 2.25 MJ/kg. 6.8.1 Background of Study
Limits - Assuming a mule pulling a cart equal to its body
Why even consider open source? Historically space
mass for 20 days to reach inaccessible areas in Tibet, we
projects have been carried out by the government and
get 1.8 MJ/kg transport energy. If we then add another
commercial sectors, and data have been kept private ei-
2000 km by rail to reach 5% of the world’s population,
ther for national security or business competition rea-
this adds another 0.43 MJ/kg, for a total of 2.23 MJ/kg.
sons. However this results in duplication of effort, and
Assuming transport distance of 7000 km from population
is therefore inherently inefficient. Scientific fields, where
centers to Tahiti, this gives 2.6 MJ/kg, which is the higher
data is published openly, and open-source software, are
value. Low Earth orbit requires 31 MJ/kg, so there is a
good examples of non-duplication of effort. The ques-
large step function from Earth to space locations.
tion then becomes can those methods be applied to space
projects? The level of technology and size of economies
6.7.3 Detailed Estimates restricted early space projects to the largest governments
at first. As technologies have improved and the general
Atmospheric Temperature size of economies grown, the relative scale required has
gone down. Open projects tend to be relatively small, so
We selected the seasonal daily highs and lows for this pa- a second question is whether space projects have reached
rameter as the “normal extremes” that people are adapted a scale that open source projects can execute them?
to, even though particular days can exceed this range. We
start from the hot and cold ends of where people live, and
count until we reach 5% in each. We count by countries Past Work On This Topic
when they are small and consist of one climate region, or
by sections when they are large. If the population is con- 6.8.2 Requirements Analysis
centrated in one part, we use climate data for that area.
6.8. OPEN SOURCE SPACE PROGRAM 265

Defining the “Customer”

The customer in the Systems Engineering process is the

entity which has an unmet need or desire which a new sys-
tem might satisfy. In this case the customer is assumed
to be the set of humans who want to make use of the
resources of space, but which cannot because it is too
difficult or expensive at the moment. There are existing
government and commercial space projects, but the gov-
ernment ones at least have shown a remarkable lack of
progress in making access to space easier and less expen-
sive. At this point the individuals who make up the cus-
tomer set are mostly unidentified, so we cannot poll them
to find out exactly what requirements they might have.
Instead we must start with proxies to represent what they
might want, and later incorporate their real requirements
if we can identify them to ask.

Establish Project Requirements

Since the unmet need is easier and less expensive access

to space to make use of its resources, we must ask easier
than what? We will set the baseline as the current path of
already existing space programs. If an open-source ap-
proach can improve on that path, it would be worth pur-
suing.

Establishing Project Measures

Normally for a complex task, there will be multiple pa-

rameters that measure “goodness” in different ways. In
order to compare and select among alternatives these var-
ious measures are converted to a common scale.
Chapter 7

References and Sources

7.1 References and Sources good examples and encourage looking for good sources
in your fields of interest.
A system designer should know the current state of General News
knowledge in topics relevant to their work. There are sev-
eral reasons. One is to not repeat work already done by • Google News - General news aggregator. This
someone else. Another is to stimulate new ideas and im- means they do not write original stories, but rather
provements. In addition to the references listed below link to stories from multiple other sources. Search-
and elsewhere in the book, it is very useful to know how able and customizable, it is a good starting point to
to find additional information. Categories of information find other regular sources. Story selection for the
include: front page is automated, but the full database in-
dexes several thousand news sites.
• Current News (Magazines, Newspapers, Blogs)
• Reddit Technology, Science, and Space Topics -
• Archival Publications (Journals, Preprint Archives) Reader submitted daily links divided by topic, with
active discussions. There are many topic headings,
• Books but the three listed are the largest related to space.
It is another news aggregator, but human-submitted
• Online Data (Web and other protocols) and filtered by votes of the readers.
• Technical Reports
General Space
• Product Data
• Space.com - General news website for anything re-
• Discussion Forums lated to space.

Once information is located you should record where and General Technology
how you found it, to save having to find it again. There are
a number of ways to do that, depending on type of media: • Next Big Future - Blog for all kinds of new and
building a personal library in paper or electronic form, future science and technology. Often covers space
bookmarking in web browsers, etc. As long as you have topics.
the ability to find the data again, the particular methods
can be left to personal preference. General Science

• EurekAlert - Current science and technology press

7.1.1 Current News releases. Often a day earlier and unmodified by
journalists.
As of 2012, obtaining current news via paper-based
sources is nearly obsolete. Internet-based electronic de-
livery is faster and less expensive. There are innumer- Specific Industries
able online websites, magazines, and blogs covering every
topic relevant to space systems. One distinction between • Aviation Week - Aerospace industry news, includ-
sites is original reporting vs re-posting of news from else- ing space projects.
where. You should also consider the quality and experi- • Automation World - Manufacturing automation.
ence of the writers and any bias they might have. We can-
not list all possible news sources, and any such list would • EE TImes - Electronics engineering. This link is to
rapidly become obsolete. Therefore we will list a few the news summary.

266
7.1. REFERENCES AND SOURCES 267

7.1.2 Multiple Documents • Defense Technical Information Center - Public

access to the unrestricted subset of approximately 2
The following sites index, search, or link to multiple doc- million US DOD and other technical documents.
uments:
• US Patent and Trademark Office - Search and
download patent data. Patents often contain very
Library Indexes Two steps in finding the current state useful technical data.
of knowledge are (1) finding what works exist on a given
topic, and (2) locating a copy of the work to look at. Li- • arXiv.org - Open access to over 750,000 elec-
brary indexes help with both steps. tronic papers in Physics, Mathematics, Computer
Science, Quantitative Biology, Quantitative Finance
• Library of Congress Online Catalog - This is an and Statistics. Operated by Cornell University Li-
index to the largest single library in the world, in- brary.
cluding books and periodicals. Basic search is by
title, author, and subject. Items are shelved and in- • NATO Research and Technology - FTP server of
dexed by subject, using a cataloging Call Number NATO technical and conference reports. Some have
system. If you use the “Call Number Browse” type relevance to space projects. It is recommended to
of search, you can look through all the entries under find titles and document numbers by other search
that subject, which is useful for doing general re- methods, since the ones here are stored only by their
search on a subject. The LC Classification Outline document numbers.
has a summary of the call number system. For space
projects, the headings Q (Science) and T (Technol- • National Space Society Library - Online library
ogy) are the most relevant, although other sections with multiple space references.
can be useful depending on the topic.
• Defense Acquisition University ACC Practice
• WorldCat - This is an online database of the collec- Center - Multiple documents about how the US
tions of over 10,000 libraries worldwide. If a work government manages projects, including engineer-
is not available nearby or online, a process of Inter- ing methods in general, and space projects as a topic.
Library Loan can be used from a local library to
borrow it from another library.
Reading Lists
Online Search In addition to well-known search en-
gines such as Google and Bing, there are more special- • Encyclopedia Astronautica Recommended
ized searches such as: Books

• Google Scholar - for searching scholarly works

7.1.3 Book Length Sources
• Google Books - for searching the text of books in
general Other Wikibooks
• Open Directory Specialized Search Engine List
- Links to many other search websites. • Astrodynamics

• High School Earth Science/Early Space Exploration

Online Repositories Repositories contain multiple and High School Earth Science/Recent Space Ex-
online documents which usually can be downloaded. In ploration describes how far we've come --
some cases they only index the document online and you
then have to request or find a physical copy. • Colonizing Outer Space and Conplanet speculates
on how far we may go, and what living on other plan-
• Rocket Science Library - A collection of open ets may be like after we get there.
source, public domain, and online works compiled
in parallel with the development of this book. A list
of additional works is included for ones where copy- General Space Reference
right restrictions prevent including the document it-
self. • Jet Propulsion Laboratory, Basics of Space Flight,
2012.
• NASA Technical Reports Server - Access to
approximately 1 million documents in NASA • USAF Air Command and Staff College, Space
databases. Primer, 2nd Ed., 2009.
268 CHAPTER 7. REFERENCES AND SOURCES

General Space Systems Design • Koelle, Heinz Hermann, ed., Handbook of Astro-
nautical Engineering, New York, McGraw-Hill,
• Griffen, Michael D. and French, James R., Space 1961. - This is an excellent comprehensive refer-
Vehicle Design, 2nd ed., AIAA Education Series, ence handbook representing the state of the art as
2004. of 1961. Has a chapter covering nuclear, electric,
and solar-thermal propulsion.
• Hammond, Walter E., Design Methodologies for
Space Transportation Systems, AIAA Education
Series, 2001. • Merrill, Grayson, ed. Principles of Guided Mis-
sile Design, series, Van Nostrand, 1958. - This is
• Ley, Wittmann, and Hallmann, eds. Handbook of a series of at least 9 volumes with different authors.
Space Technology, Wiley, 2009. Although written in the Cold War nuclear missile
era, the technology is relevant to space projects.
Likely much of it will need updating due to being
Propulsion
written 50 years ago:
• Glenn Research Center, Beginner’s Guide to
Rockets - website designed for young students and • Guidance - Locke, Arthur et. al.
teachers. • Aerodynamics, Propulsion, Structures,
and Design Practice - Bonney, E.A, Zu-
• Hunter, Maxwell, Thrust into Space, Holt, Rhine- crow, M.J., and Besserer, C.W.
hard, and Winston, 1966. - An introductory text-
• Operations Research, Armament,
book on space propulsion.
Launching - Merrill, G., Goldberg, H.,
and Helmholz, R.H.
• Sutton, George P. and Biblarz, Oscar, Rocket
Propulsion Elements, 8th ed., Wiley, 2010, ISBN- • Missile Engineering Handbook -
13: 978-0470080245. - This has been considered Besserer, C.W.
the standard reference for classical rocket propul- • Space Flight - Ehricke, Krafft A.
sion. It is mostly concerned with solid, liquid, and
• Systems Engineering - Jerger, J.J.
hybrid chemical rockets, and has one chapter on
electric propulsion. • Range Testing - Freitag, R.F.
• Airborne Radar - Povejsil, D.J., Raven,
• Bolonkin, Alexander “Review of new ideas, inno- R.S., and Waterman, P.J.
vation of non-rocket propulsion systems for Space
• Automatic Flight Control - Povejsil, D.J.,
Launch and Flight” 3 parts, 2010, The Internet
Kelly, A.J., Mathews, C.W., and Mc-
Archive:
Court, A.W.
Part 1, Part 2, Part 3
Resource Use
• Mallove, Eugene F. and Matloff, Gregory L., The
Starflight Handbook: A Pioneer’s Guide To Inter- • Faughnan, Barbara (ed.); Maryniak, Gregg (ed.)
stellar Travel, 1989. - A semi-technical introduction “Space Manufacturing 5: Engineering with Lunar
to interstellar flight. and Asteroidal Materials”, proceedings of the 7th
Princeton/AIAA/SSI Conference, Princeton, New
• Morgenthaler, G. W.; Tobiska, W. K. “Aerospace Jersey, 8-11 May 1985.
Century XXI: Space Flight Technologies”, Proceed-
ings of the 33rd Annual AAS International Confer-
ence, Boulder, Colorado, 26-29 Oct. 1986. Pub- 7.1.4 Websites
lished as Advances in the Astronautical Sciences, vol
64, pt 2, 1987. This category includes online sites where the data is in
the site itself, rather than the site being a link to other
documents.
Early Works Although technology has changed dras-
tically since these were written, physics has not, and they
serve as a guide to what technical problems need to be General Space Websites
solved in any design:
• Permanent - Organization dedicated to space de-
• Bureau of Naval Personnel, Principles of Guided velopment. Their website has extensive discussion
Missiles and Nuclear Weapons, 1959. on relevant topics.
7.1. REFERENCES AND SOURCES 269

• Rocket and Space Technology - Personal website • Forward, R. L. “Advanced Space Propulsion Study
with extensive coverage of the basics of space flight, - Antiproton and Beamed Power Propulsion”, Fi-
hardware, and missions. nal Report, 1 May 1986 - 30 Jun 1987, Hughes
Research Laboratories, report AFAL-TR-87-070,
• Yarchive - An archive of space related posts from 1987.DefenseTechnical Information Center #AD-
the USENET discussion system, covering many top- A189 218. National Technical Information Ser-
ics. vice # AD-A189 218/1 PC A10/MF A01 [Quote:
, goes into detail on beamed power systems includ-
• Centauri Dreams - Long-running blog about deep
ing " 1) pellet, microwave, and laser beamed power
space exploration.
systems for intersteller transport; 2) a design for
a near-relativistic laser-pushed lightsail using near-
Real Time Data term laser technology; 3) a survey of laser thermal
propulsion, tether transportation systems, antiproton
• Real Time Satellite Tracking - Provides track- annihilation propulsion, exotic applications of so-
ing simulation of Earth satellite orbits based on lar sails, and laser-pushed interstellar lightsails; 4)
official tracking data. the status of antiproton annihilation propulsion as
of 1986; and 5) the prospects for obtaining antimat-
• Eyes on the Solar System - 3D Simulation of So- ter ions heavier than antiprotons.” Again, there is an
lar System spacecraft and planetary bodies be- extensive bibliography.]
yond Earth orbit, including current, and past
and future positions. • Forward, R. L. “Exotic Propulsion in the 21st Cen-
tury”, in Aerospace Century XXI (see Morgenthaler
and Tobiska, under Propulsion above).
Life Support
• Forward, Dr. Robert L., Future Magic., Avon,
• USDA Nutrient Data Laboratory - Includes a 1988. ISBN 0-380- 89814-4. “Nontechnical dis-
food and nutrition database, from which food re- cussion of tethers, antimatter, gravity control, and
quirements for space projects can be determined. even further-out topics.”

More References to be sorted These should eventu-

7.1.5 Articles and Technical Reports
ally end up under the appropriate section of the book:
Multiple Propulsion Concepts
• Kunz, K. E. “Orbit Transfer Propulsion and Large
• Diesposti, R. S.; Pelouch, J. J. “Performance and Space Systems”, J. Spacecraft and Rockets vol 17
Economic Comparison of Externally Energized vs no 6 pp 495-500, Nov.-Dec. 1980.
Chemically Energized Space Propulsion”, AIAA
• Poeschel, R. L. “Comparison of Electric Propulsion
paper number 81-0703 presented at 15th Interna-
Technologies”, AIAA paper number 82-1243 pre-
tional Electric Propulsion Conference, Las Vegas,
sented at AIAA/SAE/ASME 18th Joint Propulsion
Nevada, 21-23 June 1981.
Conference, Cleveland, Ohio, 21- 23 June 1982.
• Andrews, Dr. Dana G, ed. Advanced Propulsion
• Jones, R. M.; Kaplan, D. I.; Nock, K. T.
Systems Concepts for Orbital Tansfer: Final Re-
“Electric Propulsion Systems for Space Stations”
port, Boeing document D180- 26680 produced un-
AIAA/SAE/ASME 19th Joint Propulsion Confer-
der NASA contract NAS8-33935.
ence, AIAA paper number 83-1208, 1983.
• Forward, Robert “Alternate Propulsion Energy • Jones, R. M. “Space Supertankers: Electric Propul-
Sources”, AFPRL TR-83-067. (NTIS AD-B088 sion Systems for the Transportation of Extrater-
771/1) Dec 1983, 138p. Keywords: Propul- restrial Resources” AIAA/SAE/ASME 20th Joint
sion energy, metastable helium, free-radical hydro- Propulsion Conference, AIAA paper number 84-
gen, solar-pumped plasmas, antiproton annihilia- 1323, 1984.
tion, ionospheric lasers, solar sails, perforated sails,
microwave sails, quantum fluctuations, antimatter • Phillips, P. G.; Redd, B. “Propulsion Options
rockets. for Manned Missions to the Moon and Mars”, in
Aerospace Century XXI (see Morgenthaler and To-
• Wang, S.-Y.; Staiger P. J. “Primary Propul- biska, under Propulsion above).
sion of Electro-Thermal, Ion and Chemical Sys-
tems for Space Based Radar Orbit Transfer”, • Matloff, G. L. “Electric Propulsion and Interstellar
AIAA/SAE/ASME/ASEE 21st Joint Propulsion Flight”, 19th International Electric Propulsion Con-
Conference, AIAA paper number 85-1477, 1985. ference, Colorado Springs, Colorado, 11 May 1987.
270 CHAPTER 7. REFERENCES AND SOURCES

• Korobeinikov, V. P. “On the Use of Solar Energy for attached directly to the skin, from which
the Acceleration of Bodies to Cosmic Velocities”, they grow.
Acta Astronautica, v 15 no 11 p 937-40, November
1987. 1657
• Kerrebrock, J. L “Report of the National Com-
mission on Space- One Commissioner’s View”, in • de Bergerac, Cyrano, A Voyage to the
Aerospace Century XXI (see Morgenthaler and To- Moon, 1657. Uses glass containers of
biska, under Propulsion above). dew for lift when the Sun falls on them
in a first attempt. Dew is just condensed
• Harvego, E. A.; Sulmeisters, T. K. “A Compari- water, and exposing it to sunlight in na-
son of Propulsion Systems for Potential Space Mis- ture simply evaporates it. In a closed
sion Applications”, ASME Winter Meeting, Boston, container it would evaporate or simply sit
Massachusetts, 13 December 1987, 1987. there as a liquid, depending how much
• Byers, David C.; Wasel, Robert A. “NASA Electric was there. In a later chapter ranks of fire-
Propulsion Program”, NASA Technical Memoran- crackers are used, which are not imagi-
dum 89856, May 1987. nary, but merely too low in energy.

• Lorrey, M; High Density Fuels, web page, March 18th and 19th Centuries
2006.
Cavorite
• Licht, S; Solar Driven Synthesis, in Advanced Ma-
terials, 2011.
20th and 21st Centuries
7.1.6 Discussion Forums Stutterwarp Drive
Although online forums are informal, the collective
knowledge of the participants is sometimes impressive. 7.2.2 Fictional Methods Which Are Sup-
The discussion history can be searched or new questions
asked to make use of it.
ported by Known Physics or Engi-
neering

This section includes methods which have appeared in fic-

7.2 App 1 - Fictional Methods tional works, but which are possible based on currently
known science and technology.
7.2.1 Fictional Methods Which Are Not
Supported By Known Physics or En-
gineering 7.3 App 2 - Reference Data
This section includes fictional methods which do not have
any support from physics or engineering. Those that do
7.3.1 Aerodynamics
have some supporting theory are listed in Part 2, Section
• US Air Force Stability and Control Data Com-
10 Theoretical Methods
pendium - Starting in the 1920’s, the US Air Force
collected data for predicting the aerodynamic be-
Pre-Industrial Age havior of aircraft. This link is to the 1978 text
version of the compiled data, known as the USAF
c. 1st Century DATCOM, which is a 3134 page, 113 MB .pdf file.
Since then, versions of the data and formulas have
• Pseudo-Apollodorus, Bibliotheca, been compiled into software modules, which can be
manuscript approximately 1st century. run by themselves, or linked into more comprehen-
Contains the story of Daedalus and sive airplane design software.
Icarus, who use wings made of feathers
and wax. This fails on engineering due to • US Standard Atmosphere 1976 - This book
the weakness of wax and power/weight gives information and tables describing composi-
ratio of humans. Feathers obviously tion, pressure, temperature, and other properties of
work for birds, but they weigh much the Earth’s atmosphere under “standard” conditions.
less than people do, and the feathers are The actual atmosphere varies from these standard
7.3. APP 2 - REFERENCE DATA 271

conditions due to weather, the solar cycle, and secu- “The major obstacle to testing at NTS will be the reduced
lar changes such as increasing CO2 concentration. levels of radioactive debris which are allowed to transport
The reference data is still useful for doing calcu- into the public domain. The levels are more stringent than
lations, and formulas based on this data have been those present during the NERVA program. The current
incorporated into software programs and modules exposure limits of 150 m Rem to civilian personnel may
within programs. restrict the tests of the NTR to low power levels and mass
flows in the reactor... A simple solution to this problem
may be to utilize one of the Pacific Ocean Islands owned
7.3.2 Nuclear Rockets by the United States -- namely Johnston Island... (an) eco-
logical desert of ocean surrounds the area due to the stag-
Previous US Nuclear Rocket Program nation of the return of the Japanese current...”

Solid Core NTR Development McDonnell Douglas ANL Cermet Program

has a long association with nuclear propulsion, begin- From 1961 through 1967 Argonne National Labora-
ning with nuclear aircraft propulsion studies in the mid-
tory conducted a development program for nuclear
1950s.[1] . Additional work includes NERVA nuclear ve- rocket fuel elements independently of the Rover/NERVA
hicle interplanetary flight studies, and solid core vehicle
program.[4] Although no engines were fabricated or
integration studies under contract to NASA in the 1960s.tested, extensive work was conducted on testing ceramic
In 1971 studies were conducted on a Shuttle-launched fuel configurations. Thermal shock tests suggested that
Nuclear Shuttle System, which used the PEEWEE reac- these cermet fuels could have substantial tolerance to the
tor. effects of nuclear excursions. This program was termi-
In 1972 LANL undertook a study of a small nuclear en- nated in 1967, prior to its completion.
gine using NERVA technology and based on the PEE- Evaluation
WEE reactor design, that could be used in the newly de-
Two major problems were identified during the
signed NASA Space Shuttle.[2] This engine would have
Rover/NERVA program:
weighed 2,555 Kg, using composite UC-ZrC-C fuel, with
other parameters including the following: Core disassembly due to vibration, accompanied by
cracking of the fuel matrix and loss of material into the
• Total Operating Period hours 2 1 propellant flow;

• Operating Cycles 20 3 Loss of fuel matrix uranium and carbon due to coating
erosion and cracking, and through diffusion through the
• Specific Impulse seconds 860 875 coating.
• Thrust kN 71.6 73.0 The first problem was resolved by changed designs which
reduced vibration and matrix cracking.
• Hydrogen Flow kg/sec 8.5 8.5
However, the second problem, of fuel element corrosion,
• Thermal Power MWt 354 367 proved less tractable:[4]
“Corrosion was most pronounced in the mid-range re-
Various other NERVA-derived propulsion concepts were
gion, about a third of the distance from the cold end
considered, and LANL evaluated a range of mission ap-
of the fuel element. Fuel operating temperatures were
plications. For instance, it was concluded that a nuclear-
lower here than the fabrication temperatures, hence ther-
propelled orbital transfer vehicle could significantly re-
mal stresses were higher than at the hot end. Also, the
duce propellant launch requirements, based on mission
neutron flux was highest in this region...”
models that envisioned from 145 to 872 Orbital Transfer
Vehicle flights. The authors concluded that developments “No fuel element geometry or fuel material ever totally
in non-nuclear nozzle and turbo-pump technology in the solved the NERVA fuel element degradation problem.
Space Shuttle Main Engine program have enhanced the Mass loss of both uranium and carbon continued to limit
viability of NERVA derived engines. service life by causing significant perturbation to core
neutronics during the tests. Crack development in the
LANL mission analyses have identified significant reduc-
fuel element coating was never completely eliminated....
tions in Initial Mass in Low Earth Orbit (IMLEO) through
Non-nuclear testing of coated fuel elements revealed an
the use of NERVA technology:[3]
Arrhenius relationship between diffusion and tempera-
The costs of developing a flight-ready nuclear rocket was ture. For every 205 K increase in temperature (in the
estimated to be $4-5 billion, which included the cost of range 2400 to 2700 K), the mass loss increased by a fac-
rebuilding the 1960s NERVA capability, which was esti- tor of ten... resulting in loss of 20% of total uranium in
mated to include: approximately 5 hours of testing at 2870 K.”
This analysis also considered issues associated with test- A number of other program management lessons
ing, noting that:
272 CHAPTER 7. REFERENCES AND SOURCES

emerged from this period. One analysis notes that:[4] process, resulting in a thicker NbC coat with im-
“One overriding lesson from the NERVA program is that proved adherence. Some core damage occurred
fuel and core development should not be tied simply to during the 5 minute test in October 1959, which
a series of engine tests which require expensive nuclear reached power levels of 100 MWt, with some fuel
operation. Definitive techniques for fuel evaluation in elements showing blistering and corrosion. Gener-
loops or in non-nuclear heated devices should be devel- ally this reactor test was considered successful.
oped early and used throughout the program...”
KIWI B1A Using the same UO2 fuel as KIWI A3, the
fuel element design was changed to a 7 channel con-
Solid Core NTR Tests The initial series of nuclear figuration 66 cm long, with niobium carbide coat-
rocket tests was conducted by Los Alamos National Lab- ings. The December 1961 test, which was intended
oratory, under the KIWI program, which eventually cost to reach 1100 MWt, only reached 300 MWt, and
$177 million (in then year dollars). was terminated after 30 seconds due to a fire caused
Graphite was chosen as the internal structural material for by a hydrogen leak in the reactor exhaust nozzle.
these nuclear reactors for several reasons. It has excellent
strength at high temperatures and its strength actually in- KIWI B1B This September 1962 test, which was es-
creases at higher temperatures. In addition, in contrast to sentially a repeat of the KIWI B1A test, achieved
metals which are strong neutron absorbers, graphite acts a power level of 900 MWT, but was terminated
as a moderator, reducing the amount of enriched uranium within a few seconds when several fuel elements
required in the core. The great disadvantage of graphite, were ejected from the reactor exhaust nozzle.
however, is that it quickly erodes in the presence of hot
hydrogen. While this erosion cannot be eliminated, tech- KIWI B2 Design configuration not tested.
niques are available to reduce erosion to acceptable levels
over the operating lifetime of the reactor, which is mea- KIWI B3 Design configuration not tested.
sured in minutes to hours.
KIWI B4A This reactor incorporated substantial re-
Westinghouse was the prime contractor for the reactor
designs based on the failure of the KIWI B1B con-
component of the NERVA program, while Aerojet was
figuration. The fuel elements were fully extruded
the contractor for engine elements such as pumps and
hexagonal graphite blocks 1.32 meters long and 19
nozzles. The NERVA program resulted in an investment
millimetres in diameter, with 19 cooling channels,
of $662 million (then-year dollars) for development and
[4] each 2.3 millimetres in diameter. However, the test
testing of flight engine prototypes.
run in November 1962 was terminated when bright
The first nuclear rocket test, conducted in July 1959, used flashes in the exhaust stream indicated vibration in-
uncoated UO2 plates as fuel elements. This test reached a duced damage to the core was leading to its disinte-
maximum temperature of 2683 K and a power level of 70 gration.
MWt. Vibrations during operations produced significant
structural damage in the reactor core. KIWI B4B Design configuration not tested.
The first nuclear reactor tested, KIWI-A, successfully
demonstrated the principle of nuclear rockets, but it used KIWI B4C Design configuration not tested.
unclad fuel plates that were not representative of later
tests. KIWI B4D Although modifications in the design of this
reactor eliminated the vibration which had marred
KIWI A test This test, conducted in July 1959, incor- previous tests, the May 1964 test was terminated af-
porated significant modifications in the core design ter about 60 seconds at full power due to the rupture
used in KIWI A. The fuel consisted of short cylin- of a nozzle cooling tube.
drical Uranium Oxide elements in graphite modules,
with four axial channels coated with Niobium Car- KIWI B4E This reactor was the first to use coated Ura-
bide using chemical vapor deposition. The reactor nium Carbide (UC2) fuel, in place of the Uranium
ran for 6 minutes at power levels as high as 85 MWt. Oxide (UO2) fuel previously used. To avoid oxida-
tion of the Carbide fuel, it the uranium fuel parti-
KIWI-A prime tested in 1960, replaced the fuel plates cles were coated with a 25 micron layer of pyrolytic
with NbC plated graphite modules with 4 micron graphite which exuded water vapor. This pyrolytic
diameter UO2 particles embedded in the graphite carbon layer subsequently was also used to enhance
matrix. However, some structural damage occurred fission product retention, though this was not ini-
in this improved design during its 6 minute test. tially its purpose. This reactor was operated for 12
minutes, including 8 at full power. The duration of
KIWI A3 The subsequent KIWI-A3 reactor used a the test was limited by the available liquid hydrogen
higher temperature Chemical Vapor Deposition storage capability.
7.3. APP 2 - REFERENCE DATA 273

KIWI TNT This KIWI-B type reactor was deliberately in 1972. The goals of this experiment included test-
destroyed on January 1965 by subjecting it to a fast ing of a novel coated particle fuel, using a graphite
excursion. This test was intended to confirm theo- fuel matrix with a coefficient of thermal expansion
retical models of transient behaviour. closely matched to that of the coating, to reduce
thermal stress and cracking.
Phoebus 1A This first test of a new class of reactors on
June 1965 included over 10 minutes of operations NRX-A1
at 1090 MWt, with an exhaust temperature of 2370
K. NRX-A2 This September 1964 engine test included 5
minutes of operation at half to full power levels. Du-
Phoebus 1B This February 1967 test built on the previ- ration of the test was limited by available hydrogen
ous Phoebus 1A test, reaching power levels of 1500 storage capacity. At full power of 1100 MWt the
MWt for 30 minutes, with an additional 15 minutes engine demonstrated a specific impulse of 760 sec-
at lower power levels. onds.

Phoebus 2A The most powerful nuclear reactor of any NRX-A3 This April 1965 test operated for 8 minutes,
type ever constructed, with a design power level of including 3.5 minutes at full power. The first test
5,000 MWt. Operations in June 1968 were limited was terminated by a spurious trip from the turbine
to 4,000 MWt dur to premature overheating of of overspeed circuit. The reactor was restarted on May
aluminium segments of pressure vessel clamps. At 1965, and operated at full power for 13 minutes, and
total of 12.5 minutes of operations at temperatures subsequently restarted for low to medium power op-
of up to 2310 K included intermediate power level erations for an additional 45 minutes. In total, 66
operations and reactor restart. minutes of operations were accumulated, including
16.5 minutes at full power.
PEWEE This small reactor was intended to be a reac-
tor testbed, incorporating Zirconium Carbide coat- NRX-EST This engine operated for a total of 110 min-
ings on some fuel elements, instead of the Niobium utes, including 28 minutes at full power of 1100-
Carbide used on Phoebus. With a peak operating 1200 MWt, on five different days in February 1966.
power of 503 MWt at 2550 K, it achieved new lev-
els of core power density (2340 MWt/m3 average NRX-A5 This 1100 MWt engine was operated in June
and 5200 MWt/m3 peak), demonstrating a specific 1966 for 30 minutes at full power, with the test dura-
impulse of 845 seconds. tion limited by available hydrogen storage capacity.

Nuclear Furnace 1 The final phase of NERVA fuel de-

NRX-A6 This 1100 MWt engine was operated in De-
velopment was the Nuclear Furnace (NF-1) reac-
cember 1967 for 60 minutes at full power, exceed-
tor, a heterogeneous water-moderated beryllium-
ing the NERVA design goal.
reflected reactor for high-temperature nuclear test-
ing of fuel elements and other components. 47 of
XECF
the 49 fuel elements used Uranium Carbide and Zir-
conium Carbide Carbon composite fuel, while the
remaining 2 fuel elements used Uranium-Zirconium XE' This 1100 MWt engine was a prototype engine,
Carbide. Tests in 1972 of composite fuel elements the first to operated in a downward firing position.
with various carbide contents, thermal expansion It accumulated a total of 28 start cycles in March
coefficients, and thermal stress resistance demon- 1968 for a total of 115 minutes of operations. Test
strated that minimizing the mismatch in thermal stand coolant water storage capacity limited each
expansion coefficient between the fuel and coating full power test to about 10 minutes.
would reduce coating cracking and carbon erosion.
With a peak operating power of 44 MWt at 2500 K, [1] Haloulakos, V.E., et al. Nuclear Propulsion: Past, Present,
it achieved new levels of core power density (4500 and Future, Fifth Symposium on Space Nuclear Power Sys-
tems, Albuquerque 11-14 January 1988. pp. 329-332.
to 5000 MWt/m3).
[2] Bohl, R.J., and Boudreau, J.E. (January 1987). Direct Nu-
The Nuclear Furnace test facility included provisions for clear Propulsion: A White Paper. Los Alamos National
remote controlled replacement of core elements, as well Laboratory.
as a reactor effluent scrubber system to remove radioac- [3] Howe, Steven (10-14 June 1985). Assessment of the
tive contaminants from the propellant exhaust. Advantages and Feasibility of a Nuclear Rocket for a
Manned Mars Mission, Manned Mars Mission Workshop.
Nuclear Furnace 2 This reactor was built but not Huntsville, AL: Marshall Space Flight Center. Preprint
tested, due to the cancellation of all work in this area LA-UR-85-2442.
274 CHAPTER 7. REFERENCES AND SOURCES

[4] Horman, F.J., et al (4-6 September 1991). Parti-

cle Fuels Technology for Nuclear Thermal Propulsion,
AIAA/NASA/OAI Conference on Advanced SEI Technolo-
gies. Cleveland, Ohio. Paper AIAA 91-3457.
Chapter 8

Text and image sources, contributors, and

licenses

8.1 Text
• Space Transport and Engineering Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_Methods?
oldid=3226984 Contributors: DavidCary, Panic2k4, Iamunknown, Whiteknight, Danielravennest, Cshank, CatastrophicToad, Jguk, Her-
bythyme, AdRiley, Remi, Dallas1278, QuiteUnusual, Savant-fou, Adrignola and Anonymous: 16
• Space Transport and Engineering Methods/Introduction Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Introduction?oldid=3209649 Contributors: Danielravennest, Jguk, JorisvS, Adrignola and Anonymous: 4
• Space Transport and Engineering Methods/Fundamentals Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Fundamentals?oldid=3170731 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Physics Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Physics?oldid=3033212 Contributors: DavidCary, Panic2k4, Danielravennest, Dleather~enwikibooks, Jguk, AdRiley, Dal-
las1278, QuiteUnusual, Savant-fou, Adrignola, Sluffs, MDodson, Aerospace586, Zhoulikan, Lsparrish and Anonymous: 10
• Space Transport and Engineering Methods/Physics2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Physics2?oldid=3036694 Contributors: Danielravennest
• Space Transport and Engineering Methods/Physics3 Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Physics3?oldid=3036867 Contributors: Danielravennest
• Space Transport and Engineering Methods/Orbital Mechanics Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Orbital_Mechanics?oldid=3037108 Contributors: Danielravennest and Anonymous: 1
• Space Transport and Engineering Methods/Forces Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Forces?oldid=3039623 Contributors: DavidCary, Danielravennest, Jguk, Xania and Adrignola
• Space Transport and Engineering Methods/Energy Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Energy?oldid=3041348 Contributors: Danielravennest, Jguk, Adrignola, Apteva, Maskredd, Teapeat and Anonymous: 1
• Space Transport and Engineering Methods/Methodologies Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Methodologies?oldid=3203475 Contributors: Panic2k4, Danielravennest, Jguk, Webaware, Adrignola and Anonymous: 1
• Space Transport and Engineering Methods/System Elements Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/System_Elements?oldid=3042865 Contributors: Danielravennest
• Space Transport and Engineering Methods/Engineering Tools Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Engineering_Tools?oldid=3044076 Contributors: Danielravennest
• Space Transport and Engineering Methods/Engineering Specialties Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Engineering_Specialties?oldid=3047246 Contributors: Danielravennest
• Space Transport and Engineering Methods/Economics Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Economics?oldid=3049610 Contributors: Danielravennest
• Space Transport and Engineering Methods/Existing Programs Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Existing_Programs?oldid=3063863 Contributors: Danielravennest
• Space Transport and Engineering Methods/Existing Programs2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Existing_Programs2?oldid=3218155 Contributors: Danielravennest
• Space Transport and Engineering Methods/Resource Uses Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Resource_Uses?oldid=3068493 Contributors: Danielravennest and Anonymous: 1
• Space Transport and Engineering Methods/Transport Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Transport_Methods?oldid=3070068 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Structural Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Structural_Methods?oldid=3138793 Contributors: Danielravennest, Jguk, Remi0o, Mike.lifeguard, Adrignola,
Lsparrish and Anonymous: 5

275
276 CHAPTER 8. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• Space Transport and Engineering Methods/Guns and Accelerators Source: https://en.wikibooks.org/wiki/Space_Transport_and_

Engineering_Methods/Guns_and_Accelerators?oldid=2476827 Contributors: Danielravennest, Jguk, Herbythyme, Adrignola and Anony-
mous: 2
• Space Transport and Engineering Methods/Guns and Accelerators2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Guns_and_Accelerators2?oldid=2532844 Contributors: Danielravennest
• Space Transport and Engineering Methods/Combustion Engines Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combustion_Engines?oldid=2476829 Contributors: DavidCary, Panic2k4, Danielravennest, Jguk, Herbythyme,
AdRiley, Dallas1278, QuiteUnusual, Savant-fou, Adrignola, MDodson and Anonymous: 2
• Space Transport and Engineering Methods/Thermal Engines Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Thermal_Engines?oldid=2476831 Contributors: DavidCary, Danielravennest, Jguk, Adrignola and Anonymous: 1
• Space Transport and Engineering Methods/Bulk Matter Engines Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Bulk_Matter_Engines?oldid=2532846 Contributors: Danielravennest, Jguk and Adrignola
• Space Transport and Engineering Methods/Ion and Plasma Engines Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Ion_and_Plasma_Engines?oldid=3075375 Contributors: DavidCary, Panic2k4, Danielravennest, Cshank, Jguk,
Remi0o, AdRiley, Darius Clark, Dallas1278, Avian, QuiteUnusual, Adrignola, HethrirBot, Atcovi and Anonymous: 5
• Space Transport and Engineering Methods/High Energy Particle Engines Source: https://en.wikibooks.org/wiki/Space_Transport_
and_Engineering_Methods/High_Energy_Particle_Engines?oldid=3142012 Contributors: DavidCary, Danielravennest, Cshank, Jguk,
AdRiley, Dallas1278, QuiteUnusual, Adrignola, MDodson and Anonymous: 6
• Space Transport and Engineering Methods/Photon Engines Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Photon_Engines?oldid=2697875 Contributors: DavidCary, Danielravennest, Jguk and Adrignola
• Space Transport and Engineering Methods/External Interaction Methods Source: https://en.wikibooks.org/wiki/Space_Transport_
and_Engineering_Methods/External_Interaction_Methods?oldid=2990007 Contributors: Danielravennest, Jguk and Adrignola
• Space Transport and Engineering Methods/Theoretical Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Theoretical_Methods?oldid=2537626 Contributors: Danielravennest and QuiteUnusual
• Space Transport and Engineering Methods/Comparisons Among Methods Source: https://en.wikibooks.org/wiki/Space_Transport_
and_Engineering_Methods/Comparisons_Among_Methods?oldid=2537628 Contributors: Danielravennest, Jguk and Adrignola
• Space Transport and Engineering Methods/Engineering Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Engineering_Methods?oldid=2401947 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Design Factors Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Design_Factors?oldid=3117268 Contributors: Danielravennest and JackBot
• Space Transport and Engineering Methods/Subsystem Design Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Subsystem_Design?oldid=3117269 Contributors: Danielravennest and JackBot
• Space Transport and Engineering Methods/Resource Exploration Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Resource_Exploration?oldid=2418155 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Resource Exploration2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Resource_Exploration2?oldid=3220346 Contributors: Danielravennest and Anonymous: 2
• Space Transport and Engineering Methods/Resource Extraction Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Resource_Extraction?oldid=3082883 Contributors: Danielravennest, QUBot and Anonymous: 2
• Space Transport and Engineering Methods/Production Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Production_Methods?oldid=2539791 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Assembly Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Assembly?oldid=2537633 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Test Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_Methods/
Test?oldid=2439618 Contributors: Danielravennest and QuiteUnusual
• Space Transport and Engineering Methods/Operations Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Operations?oldid=2539792 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Recycling Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Recycling_Methods?oldid=2539794 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Complex Programs Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Complex_Programs?oldid=3201257 Contributors: Danielravennest and JackBot
• Space Transport and Engineering Methods/Combined System Overview Source: https://en.wikibooks.org/wiki/Space_Transport_
and_Engineering_Methods/Combined_System_Overview?oldid=3217572 Contributors: Danielravennest and JackPotte
• Space Transport and Engineering Methods/Advanced Manufacturing Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Advanced_Manufacturing?oldid=3206633 Contributors: Danielravennest, QUBot and Anonymous: 1
• Space Transport and Engineering Methods/Startup Launcher Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Startup_Launcher?oldid=3218174 Contributors: Danielravennest and Anonymous: 1
• Space Transport and Engineering Methods/Orbital Assembly Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Orbital_Assembly?oldid=2539796 Contributors: Danielravennest, Sluffs and QUBot
• Space Transport and Engineering Methods/Hypervelocity Launcher Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Hypervelocity_Launcher?oldid=2539797 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Human Transport Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Human_Transport?oldid=2539798 Contributors: Panic2k4, Danielravennest and QUBot
8.2. IMAGES 277

• Space Transport and Engineering Methods/Electric Propulsion Source: https://en.wikibooks.org/wiki/Space_Transport_and_

Engineering_Methods/Electric_Propulsion?oldid=3236529 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Orbital Mining Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Orbital_Mining?oldid=2458219 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Orbital Mining2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Orbital_Mining2?oldid=2458518 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Processing Factory Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Processing_Factory?oldid=2993850 Contributors: Danielravennest, Olawlor and QUBot
• Space Transport and Engineering Methods/Space Elevator Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Space_Elevator?oldid=3206177 Contributors: Danielravennest, QUBot, TomCraver, Skyhook1, King John2 and Anonymous: 2
• Space Transport and Engineering Methods/Lunar Development Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Lunar_Development?oldid=3240105 Contributors: Danielravennest, QUBot, Camaxtli~enwikibooks and Anony-
mous: 1
• Space Transport and Engineering Methods/Interplanetary Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Interplanetary?oldid=2539977 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Mars Development Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Mars_Development?oldid=3190324 Contributors: Panic2k4, Danielravennest and QUBot
• Space Transport and Engineering Methods/Later Projects Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Later_Projects?oldid=3220032 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Design Studies Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/Design_Studies?oldid=2506230 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Combined Systems Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combined_Systems?oldid=3171301 Contributors: Panic2k4, Danielravennest and QuiteUnusual
• Space Transport and Engineering Methods/Combined Systems2 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combined_Systems2?oldid=2420218 Contributors: Danielravennest
• Space Transport and Engineering Methods/Combined Systems3 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combined_Systems3?oldid=2501746 Contributors: Danielravennest
• Space Transport and Engineering Methods/Combined Systems4 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combined_Systems4?oldid=2543819 Contributors: Danielravennest and Anonymous: 1
• Space Transport and Engineering Methods/Combined Systems5 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Combined_Systems5?oldid=3172481 Contributors: Danielravennest and Anonymous: 2
• Space Transport and Engineering Methods/Environment Ranges Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Environment_Ranges?oldid=2439012 Contributors: Danielravennest
• Space Transport and Engineering Methods/Design Studies 3 Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Design_Studies_3?oldid=2539982 Contributors: Danielravennest
• Space Transport and Engineering Methods/References Source: https://en.wikibooks.org/wiki/Space_Transport_and_Engineering_
Methods/References?oldid=2498674 Contributors: Danielravennest, Jguk and Adrignola
• Space Transport and Engineering Methods/Fictional Methods Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Fictional_Methods?oldid=2539984 Contributors: Danielravennest and QUBot
• Space Transport and Engineering Methods/Reference Data Source: https://en.wikibooks.org/wiki/Space_Transport_and_
Engineering_Methods/Reference_Data?oldid=2539986 Contributors: Danielravennest, Cshank, Jguk, AdRiley, Dallas1278, Quite-
Unusual, Savant-fou, QUBot and Anonymous: 5

8.2 Images
• File:50_percents.svg Source: https://upload.wikimedia.org/wikipedia/commons/6/62/50_percents.svg License: CC0 Contributors: File:
50%.svg Original artist: Ftiercel
• File:An_image_describing_the_semi-major_and_semi-minor_axis_of_ellipse.svg Source: https://upload.wikimedia.org/wikipedia/
commons/7/76/An_image_describing_the_semi-major_and_semi-minor_axis_of_ellipse.svg License: CC BY-SA 4.0 Contributors: Own
work Original artist: Sae1962
• File:Best_Research-Cell_Efficiencies.png Source: https://upload.wikimedia.org/wikipedia/commons/3/35/Best_Research-Cell_
Efficiencies.png License: Public domain Contributors: National Renewable Energy Laboratory (NREL), Golden, CO − United States
Department of Energy Original artist: National Renewable Energy Laboratory (NREL)
• File:Binding_energy_curve_-_common_isotopes.svg Source: https://upload.wikimedia.org/wikipedia/commons/5/53/Binding_
energy_curve_-_common_isotopes.svg License: Public domain Contributors: ? Original artist: ?
• File:Cayembe_Ecuador_Launch_System_Fit_to_Terrain.PNG Source: https://upload.wikimedia.org/wikipedia/commons/0/
09/Cayembe_Ecuador_Launch_System_Fit_to_Terrain.PNG License: CC BY-SA 3.0 Contributors: Own work Original artist:
Danielravennest
• File:E0.0_Combined_System_Environment.png Source: https://upload.wikimedia.org/wikipedia/commons/4/48/E0.0_Combined_
System_Environment.png License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Electrostatic_ion_thruster-en.svg Source: https://upload.wikimedia.org/wikipedia/commons/f/f2/Electrostatic_ion_thruster-en.
svg License: CC BY-SA 3.0 Contributors: Own work; self made in Inkscape. Based on File:Ion_engine.gif. Original artist: Oona Räisänen
278 CHAPTER 8. TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

• File:Functional_Flow_Block_Diagram_Format.jpg Source: https://upload.wikimedia.org/wikipedia/commons/e/e9/Functional_

Flow_Block_Diagram_Format.jpg License: Public domain Contributors: ? Original artist: ?
• File:General_Development_Model.PNG Source: https://upload.wikimedia.org/wikipedia/commons/4/41/General_Development_
Model.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Generalized_System_Engineering_Steps.PNG Source: https://upload.wikimedia.org/wikipedia/commons/d/d2/Generalized_
System_Engineering_Steps.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Hoba_Meteorite_sire.jpg Source: https://upload.wikimedia.org/wikipedia/commons/1/1b/Hoba_Meteorite_sire.jpg License: CC
BY 2.0 Contributors: Flickr: Hoba Meteorite site Original artist: Eugen Zibiso
• File:Hybrids_big.png Source: https://upload.wikimedia.org/wikipedia/commons/f/f3/Hybrids_big.png License: Public domain Contrib-
utors: Own work Original artist: Jonny.dyer
• File:IDEF_Box_Format.jpg Source: https://upload.wikimedia.org/wikipedia/commons/9/96/IDEF_Box_Format.jpg License: Public
domain Contributors: <a data-x-rel='nofollow' class='external text' href='http://www.dau.mil/pubs/pdf/SEFGuide%2001-01.pdf'>Systems
Engineering Fundamentals.</a> Defense Acquisition University Press, 2001 Original artist: Defense Acquisition University
• File:Lagrange_points_simple.svg Source: https://upload.wikimedia.org/wikipedia/commons/a/a5/Lagrange_points_simple.svg License:
CC BY 3.0 Contributors: File:Lagrange_points2.svg Original artist: Xander89
• File:Materials_Scale_Height_and_Tip_Velocity.PNG Source: https://upload.wikimedia.org/wikipedia/commons/d/d4/Materials_
Scale_Height_and_Tip_Velocity.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:MoonTopoLOLA.png Source: https://upload.wikimedia.org/wikipedia/commons/b/b0/MoonTopoLOLA.png License: CC BY 3.0
Contributors: Own work Original artist: Mark A. Wieczorek
• File:Moon_gravity_acceleration_map_LGM2011.jpg Source: https://upload.wikimedia.org/wikipedia/commons/2/24/Moon_
gravity_acceleration_map_LGM2011.jpg License: CC BY 3.0 Contributors: Own work Original artist: Geodesy2000
• File:North_season.jpg Source: https://upload.wikimedia.org/wikipedia/commons/8/8b/North_season.jpg License: CC0 Contributors:
Own work Original artist: Tauʻolunga
• File:Phase_Starts_vs_Time.png Source: https://upload.wikimedia.org/wikipedia/commons/b/b9/Phase_Starts_vs_Time.png License:
CC BY-SA 4.0 Contributors: Own work Original artist: Danielravennest
• File:Program_Phases.png Source: https://upload.wikimedia.org/wikipedia/commons/d/dc/Program_Phases.png License: CC BY-SA
4.0 Contributors: Own work Original artist: Danielravennest
• File:Railgun-1.svg Source: https://upload.wikimedia.org/wikipedia/commons/9/9a/Railgun-1.svg License: CC-BY-SA-3.0 Contributors:
en:Image:Railgun-1.png Original artist: Traced by User:Stannered from a PNG by en:User:DrBob
• File:STS-044_shuttle.jpg Source: https://upload.wikimedia.org/wikipedia/commons/6/67/STS-044_shuttle.jpg License: Public domain
Contributors: JSC Digital Image Collection Original artist: NASA
• File:STS-135_final_flyaround_of_ISS_1.jpg Source: https://upload.wikimedia.org/wikipedia/commons/3/37/STS-135_final_
flyaround_of_ISS_1.jpg License: Public domain Contributors: STS-135 Shuttle Mission Imagery direct link Original artist: NASA
• File:Scoop_Miner_Concept.png Source: https://upload.wikimedia.org/wikipedia/commons/e/ec/Scoop_Miner_Concept.png License:
CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Shuttle_Main_Engine_Test_Firing.jpg Source: https://upload.wikimedia.org/wikipedia/commons/b/b1/Shuttle_Main_Engine_
Test_Firing.jpg License: Public domain Contributors: http://dayton.hq.nasa.gov/IMAGES/LARGE/GPN-2000-000543.jpg
Original artist: NASA
• File:Skyhooks.gif Source: https://upload.wikimedia.org/wikipedia/commons/e/e4/Skyhooks.gif License: Public domain Contributors:
Own work Original artist: Wolfkeeper
• File:Solar_System_Velocity_Map.png Source: https://upload.wikimedia.org/wikipedia/commons/4/42/Solar_System_Velocity_Map.
png License: CC BY-SA 3.0 Contributors: Own work Original artist: Danielravennest
• File:Space_elevator_structural_diagram--corrected_for_scale+CM+etc.svg Source: https://upload.wikimedia.org/wikipedia/
commons/8/89/Space_elevator_structural_diagram--corrected_for_scale%2BCM%2Betc.svg License: CC BY-SA 1.0 Contributors: It is
a modification of en:File:Space_elevator_structural_diagram--corrected_for_scale+CM+etc--regenerated_as_SVG.svg Original artist:
Skyway and User:Booyabazooka
• File:Startup_Launcher_Preliminary_Layout.PNG Source: https://upload.wikimedia.org/wikipedia/commons/a/a3/Startup_
Launcher_Preliminary_Layout.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Systems_Engineering_Linear_Life_Cycle_Stages.PNG Source: https://upload.wikimedia.org/wikipedia/commons/9/95/
Systems_Engineering_Linear_Life_Cycle_Stages.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Table_of_Space_Transport_Methods.PNG Source: https://upload.wikimedia.org/wikipedia/commons/e/e5/Table_of_Space_
Transport_Methods.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:Table_of_Space_Transport_Methods_2.PNG Source: https://upload.wikimedia.org/wikipedia/commons/0/06/Table_of_Space_
Transport_Methods_2.PNG License: CC BY 3.0 Contributors: Own work Original artist: Danielravennest
• File:The_Earth_seen_from_Apollo_17.jpg Source: https://upload.wikimedia.org/wikipedia/commons/9/97/The_Earth_seen_from_
Apollo_17.jpg License: Public domain Contributors: http://web.archive.org/web/20160112123725/http://grin.hq.nasa.gov/ABSTRACTS/
GPN-2000-001138.html (image link); see also https://www.nasa.gov/multimedia/imagegallery/image_feature_329.html Original artist:
NASA/Apollo 17 crew; taken by either Harrison Schmitt or Ron Evans
• File:Tokyo_Sky_Tree_2012.JPG Source: https://upload.wikimedia.org/wikipedia/commons/3/3e/Tokyo_Sky_Tree_2012.JPG License:
CC BY-SA 3.0 Contributors: Own work Original artist: Kakidai
• File:Work_Breakdown_Structure_of_Aircraft_System.jpg Source: https://upload.wikimedia.org/wikipedia/commons/e/eb/Work_
Breakdown_Structure_of_Aircraft_System.jpg License: Public domain Contributors: ? Original artist: ?

8.3 Content license

• Creative Commons Attribution-Share Alike 3.0