DEGREE PROJECT

Preliminary Design of a 30 kN
Methane-Oxygen-powered
Electric-Pump-fed
Liquid Rocket Propulsion System

Vikramjeet Das

Space Engineering, master’s level

2023

Luleå University of Technology

Department of Computer Science, Electrical and Space Engineering
 Preliminary Design of a 30 kN
Methane-Oxygen-powered
Electric-Pump-fed
Liquid Rocket Propulsion System
 PRELIMINARY DESIGN OF A 30 KN
METHANE-OXYGEN-POWERED
ELECTRIC-PUMP-FED
LIQUID ROCKET PROPULSION SYSTEM

VIKRAMJEET DAS

Master in Space Engineering

Luleå University of Technology

Supervisors: Dr Élcio Jeronimo de Oliveira

R. Murugesan (PhD)
Examiner: Dr Victoria Barabash
Vikramjeet Das: Preliminary Design of a 30 kN Methane-Oxygen-powered
Electric-Pump-fed Liquid Rocket Propulsion System
May 2023

A thesis submitted in partial fulfilment of the requirements for

the degree of Master in Space Engineering
at the Division of Space Technology
Department of Computer Science, Electrical and Space Engineering
Luleå University of Technology
Kiruna, Sweden

Correspondence with the author may be directed to

[email protected]
An electronic version of this thesis is available at
www.ltu.se/publications

Cover image: The methane-oxygen-powered JD-1 rocket engine from i-Space.

Keywords: rocket engine, liquid rocket, methalox, electropump, LRPS, LPRE

URN:
OAI:
DiVA, id:
For Cyrus boi
Contents

ABSTRACT xiii

ACKNOWLEDGEMENTS xv

LIST OF FIGURES xvii

LIST OF TABLES xix

LIST OF ABBREVIATIONS xxi

LIST OF SYMBOLS xxiii

1 INTRODUCTION 1
1.1 Design Objective. . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Project Motivation . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 FUNDAMENTALS OF ROCKET PROPULSION 13

2.1 Ideal Rocket Propulsion System . . . . . . . . . . . . . . . . . 13
2.2 Performance Parameters. . . . . . . . . . . . . . . . . . . . . 14
2.2.1 Thrust . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.2 Specific Impulse . . . . . . . . . . . . . . . . . . . . . . 15
2.2.3 Effective Exhaust Velocity . . . . . . . . . . . . . . . . . 16
2.2.4 Characteristic Velocity . . . . . . . . . . . . . . . . . . . 17
2.2.5 Thrust Coefficient . . . . . . . . . . . . . . . . . . . . . 17
2.2.6 Average Exhaust Velocity . . . . . . . . . . . . . . . . . 18
2.2.7 Propellant Mass Flow Rate . . . . . . . . . . . . . . . . 19
x

2.2.8 Density Impulse . . . . . . . . . . . . . . . . . . . . . . 19

2.3 Quasi 1-D Flow Parameters . . . . . . . . . . . . . . . . . . . 20
2.3.1 Isentropic Flow Condition . . . . . . . . . . . . . . . . . 20
2.3.2 Nozzle Throat Conditions . . . . . . . . . . . . . . . . . 20
2.4 Real Rocket Propulsion System . . . . . . . . . . . . . . . . . 21
2.5 Engine Nomenclature . . . . . . . . . . . . . . . . . . . . . . 23

3 LIQUID PROPELLANTS 25
3.1 Propellant Characteristics . . . . . . . . . . . . . . . . . . . . 25
3.1.1 Important Aspects & Desirable Features. . . . . . . . . . 26
3.1.2 Classification of Propellants . . . . . . . . . . . . . . . . 26
3.1.3 Propellant Selection Criteria . . . . . . . . . . . . . . . . 27
3.2 Candidate Propellants . . . . . . . . . . . . . . . . . . . . . . 28
3.2.1 Liquid Oxygen (LOX) . . . . . . . . . . . . . . . . . . . 28
3.2.2 Liquid Hydrogen (LH2) . . . . . . . . . . . . . . . . . . 29
3.2.3 Rocket Propellant 1 (RP-1) . . . . . . . . . . . . . . . . 30
3.2.4 Liquid Methane (LCH4). . . . . . . . . . . . . . . . . . 31
3.3 Propellant Performance Analysis . . . . . . . . . . . . . . . . 33
3.4 Mixture Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.4.1 Liquid Oxygen & Liquid Methane . . . . . . . . . . . . . 35
3.4.2 Liquid Oxygen & Liquid Hydrogen . . . . . . . . . . . . 39
3.4.3 Liquid Oxygen & Rocket Propellant 1 . . . . . . . . . . . 40
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 PROPELLANT FEED SYSTEMS 45

4.1 Types of Feed Systems . . . . . . . . . . . . . . . . . . . . . . 45
4.1.1 Pressure Feed Systems . . . . . . . . . . . . . . . . . . . 46
4.1.2 Turbopump Feed Systems . . . . . . . . . . . . . . . . . 47
4.1.3 Electropump Feed Systems . . . . . . . . . . . . . . . . 50
4.2 Feed System Selection Criteria . . . . . . . . . . . . . . . . . 52
4.2.1 Optimum Pressure Ratio . . . . . . . . . . . . . . . . . 53
4.2.2 Combustion Chamber Pressure . . . . . . . . . . . . . . 54
4.3 Suitability of Feed Systems . . . . . . . . . . . . . . . . . . . 56
4.3.1 Pressure Feed System . . . . . . . . . . . . . . . . . . . 56
4.3.2 Turbopump Feed System . . . . . . . . . . . . . . . . . 57
4.3.3 Electropump Feed System . . . . . . . . . . . . . . . . . 62
4.3.4 Analysis of the Feed System Mass . . . . . . . . . . . . . 66
4.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
 xi

5 DESIGN OF THE THRUST CHAMBER 73

5.1 Design Requirements . . . . . . . . . . . . . . . . . . . . . . 73
5.2 Design Parameters. . . . . . . . . . . . . . . . . . . . . . . . 74
5.2.1 Expansion Area Ratios. . . . . . . . . . . . . . . . . . . 74
5.2.2 Performance Parameters . . . . . . . . . . . . . . . . . . 76
5.2.3 Performance Correction Factors . . . . . . . . . . . . . . 79
5.3 Design of the Nozzle . . . . . . . . . . . . . . . . . . . . . . 79
5.3.1 Ideal Nozzle . . . . . . . . . . . . . . . . . . . . . . . . 80
5.3.2 Truncated Ideal Contoured (TIC) Nozzle . . . . . . . . . 85
5.3.3 Thrust Optimised Contoured (TOC) Nozzle . . . . . . . 87
5.3.4 Thrust Optimised Parabolic (TOP) Nozzle . . . . . . . . 88
5.4 Design of the Combustion Chamber . . . . . . . . . . . . . . 91
5.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

6 THRUST CHAMBER DESIGN SPECIFICS 95

6.1 Thermal Management . . . . . . . . . . . . . . . . . . . . . . 95
6.1.1 Booster-Stage Thrust Chamber . . . . . . . . . . . . . . 98
6.1.2 Upper-Stage Thrust Chamber . . . . . . . . . . . . . . . 104
6.2 Injector Design . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.2.1 Injection Elements . . . . . . . . . . . . . . . . . . . . . 106
6.2.2 Design of Injection Elements. . . . . . . . . . . . . . . . 108
6.3 Launch Vehicle Specifics. . . . . . . . . . . . . . . . . . . . . 111
6.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

7 CONCLUSION 119
7.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
7.2 Limitations & Future Work . . . . . . . . . . . . . . . . . . . 121

A CEA COMPUTATION 125

B MATLAB SCRIPT 127

B.1 Thrust Chamber Parameters. . . . . . . . . . . . . . . . . . . 127
B.2 Engine Cycle Parameters . . . . . . . . . . . . . . . . . . . . 139
B.3 Vehicle Mass Parameters . . . . . . . . . . . . . . . . . . . . 144

BIBLIOGRAPHY 147
Abstract

The design of a liquid rocket propulsion system, unlike that of a standalone system,
is intertwined with the overall development of a number of associated systems
and is influenced by a multitude of conditions and considerations: from the re-
quirements needed to accomplish the mission to the rationalizations involved
behind the development of each rocket system and/or component. In my thesis,
the preliminary design of a “new generation” 30 kN rocket engine driven by an
electric pump feed system and running on liquid methane and liquid oxygen is
performed. The propulsion system would be employed on a hypothetical small-lift
orbital-class twin-stage rocket to deliver a light payload of about 200 kg into a
circular 500 km LEO. Such topics as the selection of bipropellant combinations, the
feasibility of electric pump feed systems, design methodologies for thrust cham-
bers, for nozzles in particular, management of the high thermal energy and the
selection of compatible wall materials, as well as the design of an injector have
been looked comprehensively into.

It is realized that methalox is indeed better than both hydrolox (with regard to
density impulse) and kerolox (in terms of specific impulse). Besides, a suite of
attractive characteristics makes the bipropellant a combination of choice to power
rockets of the future. Yet more notably, an electric-pump-fed engine cycle is, under
the right circumstances of engine operation, established to outperform both the
pressure feed system and the turbopump feed system. With constant advancement
in battery technologies, improvement of both power density and energy density
to achieve much higher performance is but a matter of time. The adoption of a
propulsion system such as ours for a mission objective as outlined above, therefore,
is not just viable but unquestionably realistic.
xiv

Two thrust chamber versions—a sea-level variant for the booster stage and a
vacuum-optimized variant for the upper stage—are developed for our rocket.
And both the nozzles employ a TOP “thrust optimised parabolic” contour; also,
the booster stage comprises a cluster of 9 engines in a parallel burn arrangement.
Concerning thermal management, the entirety of the booster-stage thrust chamber
implements regenerative cooling (using Inconel 625), whereas the aft of the upper-
stage nozzle section implements radiative cooling (with Niobium C-103). Further,
the injector faceplate (also of Inconel 625) comprises two concentric patterns of
unlike impingement doublet sets: with 80 pairs on the outer ring and 40 pairs
on the inner ring. With rational assumptions, our hypothetical launch vehicle
is deemed to have a mass of roughly 17200 kg (200 kg of which is the payload)
and a delta-v of approximately 9600 m/s—quite within the desirable range of
specifications for small-lift orbital-class twin-stage rockets of today.
Acknowledgements

As I stand on the precipice of completion of this master’s thesis, I find myself

brimming with an overwhelming sense of gratitude—of awe—from the deepest
depths of my being. This degree project draws the curtains on my studies in
Space Science & Technology for the Space Engineering programme at LTU. The
culmination of this intellectual odyssey would not have been possible without the
unwavering support, encouragement and inspiration by the wonderful individuals
who have shaped the trajectory of my scholarly journey and touched my life in
extraordinary ways.

First, I extend my deepest gratitude to Élcio for agreeing to take me under his wing.
His expertise, wisdom and guidance coupled with his kind and patient personality
have been instrumental in influencing my work. I now understand quite well that
“it is not so simple” to design a rocket engine; and yet we have somehow done it
together. To Ramakrishnan, you have been the beacon of light when I needed help
the most and my catalyst without whom I would perhaps not be standing here today.
Your invaluable mentorship, genuine passion for knowledge and relentless pursuit
of excellence have shaped not only my research but also my character. I express
my heartfelt appreciation to Victoria Barabash for believing in me throughout my
studies. I am forever indebted to you for the immeasurable impact you have had
on both my academic and my personal growth. I am immensely thankful to all
the professors at LTU Rymdcampus, to Maria Winnebäck and to Anette Snällfot-
Brändström for nurturing my intellectual curiosity, imparting their knowledge and
offering their care during my time at Kiruna.

Throughout my studies, I have met some amazing souls who I share a lot of
fond memories with: Lea, Aakash, Matthias, Adam, Roger, Selvamuthukumaran,
xvi

Anand, Meltem, Stevan, Akila, Franziska, Markus, Max, Sergio, Ricardo, the list
keeps going. Thanks for providing me all those happy distractions. Attman, you
have been my awesome friend ever since I have known you. Nabbu and Subbu,
thank you for being there anytime and everytime I needed you. Roshni, thank you
for making my world your own and for everything that you have done for me. I
will forever cherish everything about you! Thanks, Anuswiya, for having my back
at all times, for loving, understanding and encouraging me so much, no matter
what. You are the absolute best!
Sushree, thanks for being the nicest sibling I could ever get, and for bringing in
good and happy vibes all the time. I am unfathomably thankful to my beloved
parents for all their blessings, love, sacrifices, support, care and concern, and for
making me the person that I am today. Lastly, Carl Sagan, you are—and always
will be—my greatest inspiration on my ephemeral voyage through the cosmos
aboard “the pale blue dot, the only home we’ve ever known”.
Figures

1.1. A render of the V-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2. SpaceX Falcon 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3. Rocket Lab Electron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4. Typical flight profile of a launch vehicle . . . . . . . . . . . . . . . . . 6
1.5. Electron ready to launch on the Virginia Is For Launch Lovers mission 7
1.6. Four commercial small satellites by Axelspace . . . . . . . . . . . . . 8
1.7. Overview of Rocket Lab Electron . . . . . . . . . . . . . . . . . . . . . 10

2.1. Major components/sections of a rocket engine . . . . . . . . . . . . . 24

3.1. Optimum Mixture Ratio for Liquid Oxygen and Liquid Methane . . 36
3.2. Specific Impulse vs Mixture Ratio . . . . . . . . . . . . . . . . . . . . . 37
3.3. Temperature vs Mixture Ratio . . . . . . . . . . . . . . . . . . . . . . . 37
3.4. Characteristic Velocity & Molecular Mass vs Mixture Ratio . . . . . . 38
3.5. Specific Impulse & Temperature vs Mixture Ratio . . . . . . . . . . . 38
3.6. Optimum Mixture Ratio for Liquid Oxygen and Liquid Hydrogen . . 39
3.8. Optimum Mixture Ratio for Liquid Oxygen and Liquid Hydrogen . . 40
3.7. Specific Impulse & Temperature & Molecular Mass vs Mixture Ratio
for LOX/LH2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.9. Specific Impulse & Temperature & Molecular Mass vs Mixture Ratio
for LOX/RP-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.10. Specific Impulse vs Mixture Ratio for all three bipropellants considered 42
3.11. The relative volume of respective fuels against the oxidizer volume . 42

4.1. Schematic of a pressure feed system . . . . . . . . . . . . . . . . . . . 47

4.2. Schematics of turbopump feed systems . . . . . . . . . . . . . . . . . 49
xviii

4.3. Schematic of an electropump feed system . . . . . . . . . . . . . . . . 51

4.4. Thrust Coefficient & Specific Impulse vs Pressure Ratio . . . . . . . . 55
4.5. Expansion Area Ratio & Exit Pressure vs Pressure Ratio . . . . . . . . 55
4.6. Variation of feed system mass with engine burn time . . . . . . . . . 69
4.7. Variation of feed system mass ratio with engine burn time . . . . . . 69

5.1. Variation of combustion parameters with chamber pressure . . . . . 77

5.2. Local characteristic curves of a streamline . . . . . . . . . . . . . . . . 81
5.3. Design schematic of a supersonic nozzle contour using the method of
characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.4. Nozzle contours for our thrust chamber variants using the method of
characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.5. Illustration of TIC nozzle constraints . . . . . . . . . . . . . . . . . . . 86
5.6. TOC Nozzle Schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.7. Schematic of a TOP nozzle . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.8. The two angles of a parabolic nozzle . . . . . . . . . . . . . . . . . . . 89
5.9. Combustor configuration layout . . . . . . . . . . . . . . . . . . . . . 92
5.10. Thrust chamber cross-sectional profiles . . . . . . . . . . . . . . . . . 93

6.1. Variation of nozzle parameters with Mach number . . . . . . . . . . . 96

6.2. Variation of nozzle parameters with pressure . . . . . . . . . . . . . . 97
6.3. Schematic diagram of the coolant channel . . . . . . . . . . . . . . . . 99
6.4. Configuration of the cooling jacket . . . . . . . . . . . . . . . . . . . . 101
6.5. Heat flux distribution on the inner thrust chamber wall . . . . . . . . 102
6.6. Temperature distribution on the thrust chamber inner wall surfaces . 102
6.7. Variation of both temperature and pressure for the coolant . . . . . . 103
6.8. Schematic diagram of the upper-stage thrust chamber . . . . . . . . . 104
6.9. Temperature distribution on the upper-stage thrust chamber wall sur-
faces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.10. Typical configuration of an unlike doublet injector . . . . . . . . . . . 109
6.11. Injector configuration layout . . . . . . . . . . . . . . . . . . . . . . . . 111
6.12. Impingement angles and resultant jet momentum . . . . . . . . . . . 112
6.13. Booster stage engine cluster configuration . . . . . . . . . . . . . . . . 116
Tables

2.1. Convention used for our rocket engine . . . . . . . . . . . . . . . . . . 24

4.1. Preliminary parameters for thrust chamber . . . . . . . . . . . . . . . 56

4.2. Specifications for feed system mass estimation . . . . . . . . . . . . . 68

6.1. Material properties of Inconel 625 . . . . . . . . . . . . . . . . . . . . 100

6.2. Cooling jacket parameters . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.3. Parameters for the injector design . . . . . . . . . . . . . . . . . . . . . 112
6.4. Preliminary mass and delta-v values of our launch vehicle . . . . . . 115
Abbreviations

1-D One Dimensional

2-D Two Dimensional
APS Ascent Propulsion System
CEA Chemical Equilibrium with Applications
CFD Computational Fluid Dynamics
CTIC Compressed Truncated Ideal Contour
Cubesat Cube Satellite
DC Direct Current
DPS Descent Propulsion System
Electropump Electric Pump
EPFS Electropump Feed System
ESA European Space Agency
FEA Finite Element Analysis
GG Gas Generator
Hydrolox (Liquid) Hydrogen & Liquid Oxygen
IRFNA Inhibited Red Fuming Nitric Acid
JP-4 Jet Propellant 4
Kerolox Kerosene & Liquid Oxygen,
LCH4 Liquid Methane
LEO Low Earth Orbit
LH2 Liquid Hydrogen
LiPo Lithium Polymer
LNG Liquefied Natural Gas
LOX Liquid Oxygen
xxii

LPRE Liquid Propellant Rocket Engine

LRPS Liquid Rocket Propulsion System
Methalox (Liquid) Methane & Liquid Oxygen
MMH Monomethylhydrazine
NASA National Aeronautics and Space Administration
NTO Dinitrogen Tetroxide
OMS Orbital Maneuvering System
PMSM Permanent Magnet Synchronous Motor
RP-1 Rocket Propellant 1
RPA Rocket Propulsion Analysis
SPS Service Propulsion System
SSO Sun Synchronous Orbit
TC Thrust Chamber
TIC Truncated Ideal Contour
TOC Thrust Optimised Contour
TOP Thrust Optimised Parabola
TPFS Turbopump Feed System
UDMH Unsymmetrical Dimethylhydrazine
WF Working Fluid
Symbols

𝐴 Cross-sectional area
𝑎 Sonic velocity
𝑐 Effective exhaust velocity
𝑐∗ Characteristic velocity
𝑐𝑃 Heat capacity (constant pressure)
𝑐𝑉 Heat capacity (constant volume)
𝐶 Characteristic curve
𝐶𝑑 Discharge coefficient
𝐶𝑓 Thrust coefficient
𝑑 Density
𝐷 Diameter
𝑒 Energy
𝐹 Thrust
ℎ Fluid head
𝐼 Impulse
𝐼𝑠 Specific impulse
𝐼𝑠𝑣𝑎𝑐 Specific impulse (vacuum)
𝐼𝑠𝑑 Density impulse
𝐾 Constant
𝑘 Specific heat ratio
𝐿 Length
𝐿∗ Characteristic length
𝑀 Mach number
𝑀 Mixing factor
xxiv

𝑚 Mass
𝑚̇ Propellant mass flow rate
𝑁 Number of engine clusters
𝑂/𝐹 Propellant mass mixture ratio
𝑝 Power
𝑃 Pressure
𝛥𝑃 Head rise/Pressure drop
𝑅 Gas constant
𝑅 Radius
𝑆 Surface area
𝑇 Temperature
𝑡 Time
𝑣 Average velocity
𝛥𝑣 Velocity increment
𝑉 Volume
𝑤̇ Propellant weight flow rate
𝑋 Variable
MR Mass ratio
𝔐 Effective molecular mass
𝛿 Density
𝜖 Nozzle expansion area ratio
𝜖𝑐 Nozzle contraction area ratio
𝜂 Efficiency
𝜂 Lagrange multiplier
𝜅 Safety factor/margin
𝜇 Mach angle
𝜈 Prandtl–Meyer function
𝜙 Velocity potential
𝜌 Mass density
𝜎 Yield strength
𝜃 Angle
𝜀 Structural mass fraction
𝜁 Correction factor
𝜁 Propellant mass fraction
 xxv

SUBSCRIPTS

1st First “booster” stage

2nd Second “upper” stage
𝑏 Burn duration
𝑏𝑎𝑡 Battery pack
𝑐 Convergent section
𝑐𝑐 Combustion chamber
𝑐ℎ𝑎𝑟 Characteristic curve
𝑑 Divergent section
𝑒 Energy
𝑒 Nozzle exit
𝑒𝑥𝑡 External conditions
𝐸𝑃𝐹𝑆 Electropump feed system
𝑓 Fuel
𝑔𝑔 Gas generator
𝑖 Inflection
𝑖 Nozzle inlet
𝑖𝑛𝑗 Injector
𝑖𝑛𝑛 Inner
𝑖𝑛𝑣 Inverter
𝐿𝑉 Launch vehicle
𝑚 Material
𝑚𝑜𝑡 Electric motor
𝑛 Nozzle
𝑜 Oxidizer
𝑜𝑟𝑖 Orifice
𝑜𝑢𝑡 Outer
𝑝 Propellant
𝑝 Pump
𝑝𝑟𝑜 Stage propellant
𝑝𝑟𝑠 Pressurant
𝑝𝑢 Pump assembly
𝑠 Stay duration
𝑡 Nozzle throat
𝑡 Tank
𝑡𝑐 Thrust chamber
𝑡𝑝 Turbopump assembly
xxvi

𝑡𝑢 Turbine assembly
𝑇𝑃𝐹𝑆 Turbopump feed system
𝑢 Ullage
𝑤 Wall

CONSTANTS

𝑔0 Standard gravity of Earth (= 9.80665 m/s2 )

𝑅∗ Universal gas constant (= 8.314462618 J/K/mol)
1
Introduction

We began as wanderers, and we are wanderers still. We have lingered long

enough on the shores of the cosmic ocean. We are ready at last to set sail for
the stars.
– Carl Sagan, Cosmos

“Exploration is in our nature,” remarks Carl Sagan [1]. There have been humans
for more than twenty thousand centuries, but only in the past century have we
embarked on our cosmic voyage. We have sent dozens of ships to more than
seventy worlds, and five spacecraft bound for the stars. Our present ships that
ply the Keplerian trajectories to the planetary islands are “the harbingers, the
vanguards of future human expeditions” to those unknown worlds. These voyages
of exploration and discovery—emblematic of the epoch of sailing-ship voyages of
exploration and discovery—are the latest in a long series that have characterized
and distinguished human history.

Our essence of space exploration is not just as a scientific endeavour but as a

fundamental biological necessity to explore and discover, to foster our sense of
curiosity and wonder. And the instruments in enabling our zest for visiting those
distant other worlds across the sea of space are rockets.

The rocket began, like the gunpowder that first powered it, in China where it
was used for ceremonial and aesthetic purposes. Imported to Europe around the
fourteenth century, it was applied to warfare, discussed in the late nineteenth
century as a means of transportation to the planets by the Russian schoolteacher
Konstantin Tsiolkovsky, and first developed seriously for high altitude flight by the
American scientist Robert Goddard. The German V-2 military rocket (figure 1.1)
2

of World War II employed virtually all of Goddard’s innovations and culminated

in 1948 in the two-stage launching of the V-2/WAC Corporal combination to the
then-unprecedented altitude of 400 kilometres.
In the 1950’s, engineering advances organized
by Sergei Korolov in the Soviet Union and Wern-
her von Braun in the United States, funded as
delivery systems for weapons of mass destruc-
tion, led to the first artificial satellites. The pace
of progress has continued to be brisk: crewed
orbital flight; humans orbiting, then landing on
the moon; and uncrewed spacecraft outward
bound throughout the solar system. Many other
nations have now launched spacecraft, includ-
ing France, Japan, Germany, India and China,
the society that invented the rocket in the first
place.
A rocket, working in accordance with New-
ton’s third law of motion, obtains forward
acceleration—thus generating thrust—from the
rearward expulsion of matter at high veloci-
ties using its propulsion system. Based on the
Figure 1.1.: A render of the V-2
Credit: C. Stanley [2] type of their energy source, propulsion systems
may broadly be classified into thermal, chem-
ical, electric and nuclear propulsion. In a chemical rocket propulsion system,
the exhaust matter (also called reaction mass or working fluid) is generated by
exothermic chemical reactions of the propellant(s), and is imparted with very high
momentum through a supersonic nozzle. Depending on the nature of propellants
used, chemical propulsion may comprise solid propellants, liquid propellants or
both; liquid propulsion may further be classified into monopropellant and bipropel-
lant systems. Now, bipropellant liquid rockets carry their own liquid oxidizer and
liquid fuel which undergo a reduction–oxidation combustion process inside the
combustion chamber of a thrust chamber, and have very high chemical-to-kinetic
energy conversion efficiencies.

1.1 DESIGN OBJECTIVE

The design of a rocket propulsion system, unlike that of a standalone system,
is intertwined with the overall development of a number of associated vehicle
 3

systems and is influenced by a multitude of conditions and considerations: from

the requirements needed to accomplish the mission to the rationalizations involved
behind the development of each system and/or component. But perhaps the
initial steps, and often the defining criteria, toward the preliminary design of a
liquid propellant rocket engine stem from the objectives of the mission and the
performance specifications of the system.
In a nutshell, a typical mission for our hypothetical small-lift orbital-class rocket
would be to deliver a light payload into LEO. The two-stage launch vehicle is
required to have a delta-v budget of 9500 m/s and a payload capacity of 200 kg.
And we are required to preliminarily develop a “new generation” propulsion
system that can fly such a mission.
In essence, the principal objectives of this project are to:
▪ Design a liquid rocket propulsion system for both the booster stage and the
upper stage of a launch vehicle. The two rocket engines are obligated to be
as similar as pragmatically possible, closely mirroring the design scheme of
Rutherford on the Rocket Lab Electron or of Merlin on the SpaceX Falcon 9.
▪ Design a propellant feed system for the engine (after conducting a compara-
tive study between turbopump systems and electropump systems).
▪ Design an injector assembly, together with its associated components such
as propellant manifolds and injection elements, for the engine.

After some scrutiny into thrust metrics from extant rocket engines used on similar
launch vehicles, we decide to target for a rated thrust of 30 kN from each engine.
In addition, the outer diameter of our launch vehicle is capped at 1.25 m. While a
preliminary study of a liquid propellant rocket engine can only include so much
comprehensive analysis of every system and/or component the propulsion system
integrates with, this project intends to cover the overall development of as many
associated systems and/or components as theoretically applicable.

1.2 PROJECT MOTIVATION

But why? Why ought we to take up this project? What exactly is the societal
significance behind studying small launchers?
On 28 September 2008, Falcon 1 by SpaceX lifted off from an unremarkable island
in the Pacific and became the first privately-developed liquid-propelled launcher to
4

achieve orbit around the Earth. With the flights of Falcon 9 (figure 1.2) a few years
later, SpaceX demonstrated the viability of
reusable rockets, diminished the cost of access-
ing space, and ushered in a new era of space
exploration in the process [4].

The space age, however, was not always so pri-

vatization savvy [5] [6]. Ever since the V-2 back
in 1944 became the first artificial object to touch
space, rockets have primarily been built and op-
erated by government space agencies. With the
United States and the Soviet Union mired in the
“space race” years later to achieve spaceflight su-
periority and be the first to conquer outer space
through competitive technological dominance,
there was little incentive to count cost as a factor
for space exploration; national interest was all
that mattered, since neither nation wanted to
fall behind the other. The space race, which had
its origins in the ballistic missile-based nuclear
arms race between the two nations, started ef-
fectively with the launch of Sputnik 1 into an el-
liptical orbit around the Earth in 1957, climaxed
perhaps with the landing of humans on the sur-
face of the Moon with the Apollo 11 mission
in 1969, and, owing partly to declining politi-
cal support, ended eventually with the flight of
the collaborative Apollo–Soyuz crewed mission
between the two superpowers in 1975.

A consequence of the space race was the involve-

ment of private organizations to build hardware
for the government space agencies. Besides, a
steady shift in paradigm from militaristic and
political impetus to scientific and technological
Figure 1.2.: SpaceX Falcon 9 progression over due course of time did war-
Credit: SpaceX [3] rant the need to offer low launch prices to de-
liver payloads into orbit. In the US, a law was
subsequently passed that encouraged private partners to operate their hardware in
 5

space as well as to conduct their own launch services, thereby paving the way for
the establishment of a billion-dollar commercial spaceflight industry [5]. Over on
the Soviet side, design bureaus affiliated with the state spearheaded the research
and development of pioneering space technologies. Likewise, the ESA created Ari-
anespace to undertake spacecraft launches for and promote joint space exploration
between the member European nations. This ultimately resulted in a myriad of
firsts with regard to the privatization of spaceflight—from the first commercial
satellite to the first launch vehicle.

With the emergence of commercial spacecraft (mostly through communications

satellites), the cost of accessing space began to
dominate the market competition for the first
time [4]. A number of launchers of varying ca-
pabilities have therefore been developed across
the globe to cater to the ever-increasing market
demands. The partially reusable Space Shut-
tle by NASA is one such launcher system. The
price of launching a payload to LEO has, owing
to the competitiveness, declined substantially
over time: from $54500 per kilogram with the
Space Shuttle during its operation to $2720 per
kilogram with the Falcon 9 today [8]. And de-
creasing costs of launch services, and of space
hardware in general, have enticed new enter-
prises into this market; further, companies in
a variety of industries have begun leveraging
space accessibility and satellite technology to
drive innovation and efficiency in their earth-
bound products and services.

In more recent times, space agencies together

with commercial consortia have begun the pri-
vatization of transporting cargo and crew to the
ISS as well as launching missions to the Moon.
Through cost reduction and breakthrough inno- Figure 1.3.: Rocket Lab Electron
vation, these new offerings have democratized Credit: Rocket Lab
space exploration and knocked down the bar- [7]
riers to space [9]—from reusable rockets and
commercial space tourism to lunar habitats and constellations of small satellites.
6

Figure 1.4.: A typical flight profile for an orbital-class twin-stage expendable launch
vehicle configuration.
Image credit: John Gardi [10]

On 21 January 2018, Electron (figure 1.3) by Rocket Lab deployed three CubeSats
into orbit and helped establish an expanding market for the commercial use of
space. Rocket Lab categorically targeted the SmallSat industry by launching small
satellites—which hitherto had only been launched as secondary payloads—as
primary payloads into tailored orbits at flexible mission schedules [4].

Now, the secondary payload paradigm does not provide the specificity necessary
for many sophisticated small satellites that require quite unique mission designs
[5] [6]. To remedy this, while spurred by the emergence of such companies as
SpaceX and Rocket Lab, a number of companies are currently developing (or have
only recently developed) small-lift twin-stage launchers to cater to the increasingly
specific demands of the small satellite market. With the small satellite economy
skyrocketing, Virgin Orbit has flown LauncherOne to deploy ten CubeSats into
orbit, albeit from air. Rocket 3.3 from Astra has launched a test payload for the
United States Space Force. Alpha from Firefly Aerospace has, as of now, performed
its first partially successful orbital launch. Needless to mention are the numerous
launchers in various stages of their development cycles, viz. Spectrum from Isar
 7

Figure 1.5.: Electron ready on the pad to launch on the Virginia Is For Launch
Lovers mission for one of its commercial customers.
Image credit: Rocket Lab [7]

Aerospace, RFA One by Rocket Factory Augsburg, Prime from Orbex, Miura 5 by
PLD Space, Skyrora XL from Skyrora, Terran 1 by Relativity Space, to name some
[6]. Not only does such a fierce competition amongst the companies dwindle the
cost of accessing space, but more significantly it drives innovation at quite unprece-
dented a pace—by developing new technologies and pushing the boundaries of
what is considered conventional.

A breakthrough in inventing that innovation is employing advanced propulsion

technologies to design a “new generation” liquid rocket propulsion system. Using
the combination of liquid oxygen (LOX) and liquid methane (LCH4) would be
one instance. Another instance would be utilizing an electric pump feed system to
8

Figure 1.6.: Four commercial small satellites by Axelspace called GRUS awaiting
shipment for launch, with each in the 200 kilogram mass category.
Image credit: Axelspace [11]

power the rocket engine. Extensive research and development are currently being
performed on both these technologies throughout the rocket industry. Leading
the development of LCH4/LOX engines are SpaceX with the Raptor, Blue Origin
with the BE-4, by Relativity Space with the Aeon, among others. Likewise, Rocket
Lab and Astra have actively developed electropump systems for their Rutherford
and Delphin rocket engines respectively. Yet no engine for any launcher has thus
far been announced that employs both these technologies together.

Why is that the case? How good is a methane-oxygen-powered rocket engine?

What edge does an electric-pump-fed engine cycle provide? Is a propulsion system
employing these two techniques objectively better for a small-lift orbital-class
rocket such as ours? These are the questions which we intend to address in our
study.

With the success of Electron within the light-payload launch market as an exemplar,
we strive to emulate—or rather to better—the precedent Electron has set [7]. Con-
sequently, we target the commercial small satellite industry and plan for a payload
mass between 150 kg and 250 kg for our hypothetical rocket. And accordingly, we
aim for a vehicle diameter of about 1.2 m and a rated thrust of around 25 kN per
engine system in our study.
 9

Now, even though the reusability of rocket boosters and additive manufacturing
of rocket components are active areas of investigation within the rocket industry at
present, we do not delve into those topics in this project. A typical flight profile for
an expendable orbital-class twin-stage launch vehicle is demonstrated in figure
1.4, wherein: after a few minutes of flight post liftoff, the first stage shuts down
its engines, separates from the vehicle and drops back to Earth; soon thereafter,
the second stage commences engine ignition, discards the fairing, deploys the
payload at its targeted trajectory and velocity, and ultimately detaches away from
the payload to eventually burn up in the atmosphere. This is the profile we plan
our rocket to follow. In figure 1.5, Electron stands ready to launch on its 33rd
(and the latest) mission called “Virginia Is For Launch Lovers” for a commercial
customer. Figure 1.6 shows the example of a typical commercial small satellite
in the 200 kilogram mass category, the type our payload will likely comprise of.
Figure 1.7 presents an overview along with a few key specifications of the Electron.

1.3 THESIS OUTLINE

This chapter puts forth a mission definition for us, the objectives of this project,
and the motivation behind our study.

Chapter two defines the fundamental principles of liquid rocket propulsion sys-
tems: from performance parameters and quasi-one-dimensional flow processes for
an ideal rocket engine analysis to limitations and correction factors for a real propul-
sion system operation. All crucial equations used in the theoretical calculation of
engine performance are presented as well.

The third chapter touches on a few fundamental aspects of propellant selection,

and examines the myriad implications of bipropellant mixture characteristics on
the design criteria of a rocket engine. A suite of performance parameters (such
as specific impulse, mixture ratio, chamber pressure and temperature) obtained
from the combustion of LOX with LH2, RP-1 and LCH4 are analysed in detail. This
serves to choose the best bipropellant combination for our mission.

Chapter four delves into the functional details of pressure and pump feed systems,
and investigates the practicality of each type of feed system for the operational
regime of our engine. As a result, propellant pressures inside the combustion
chamber as well as within the feed line are settled on. The selection criteria between
turbopump and electropump systems are established on a multitude of component
characteristics exclusive to the feed system, such as gas generator, battery pack, and
10

Figure 1.7.: An overview along with some key specifications for the different stages
of the Electron launch vehicle.
Image credit: Rocket Lab [7]
 11

the like. Eventually, mass fractions of the feed systems are estimated as functions
of engine operation duration along with technological maturity of today.
The fifth chapter incorporates design methodologies to preliminarily design two
thrust chamber versions for our rocket—a sea-level variant for the booster stage and
a vacuum-optimized variant for the upper stage. After determining the expansion
ratios and correcting the performance parameters, the dimensions of the thrust
chamber are ascertained. An in-depth analysis of the convergent-divergent nozzle
contour is performed based on TIC, TOC, TOP, and the method of characteristics
techniques. Ultimately, the design configuration of the combustion chamber is
presented.
Chapter six initially addresses the topic of thermal management together with the
selection of compatible wall materials for our engine. As a consequence, compre-
hensive cooling procedures are employed for both the thrust chamber variants. The
process of propellant metering, injection, atomization, vaporization and mixing
for effective combustion are finally looked into, which culminates in the design of
an injector assembly.
The last chapter lays out the conclusions of our study, discusses the circumstances
wherein our design of methane-oxygen-powered electric-pump-fed liquid rocket
propulsion system has an edge, and sets forth the opportunities for improvement
of this project as future work prospects.
2
Fundamentals of Rocket Propulsion

In liquid rocket propulsion systems, the mathematical tools necessary to analyse

performance and to ascertain the myriad parameters for thrust chamber operations
stem from the principles of gas dynamics and thermodynamics processes. These
relations allow the determination of combustion chamber and nozzle shape and
size for a specific performance requirement, and apply to any propulsion system
that utilizes gas expansion as the mechanism for ejecting matter at high velocities
to generate thrust. This chapter is an introduction to the theory of gas properties
in chemical rocket propulsion systems, and identifies the relevant thermochemical
fundamentals, basic analytical approaches and key equations.

2.1 IDEAL ROCKET PROPULSION SYSTEM

The concept of an ideal rocket propulsion system is useful because the mathe-
matical relationships defining the primary underlying thermodynamic principles
describe “quasi-one-dimensional nozzle flows, which represent an idealization
and simplification of the full two-dimensional or three-dimensional equations
of real aerothermochemical behaviour” [12, p. 46]. These descriptions are ade-
quate to obtain useful solutions for preliminary rocket propulsion designs, and
compute results that are usually only within 1 to 6% of their measured actual
performance figures [12]. The ideal rocket parameters used in the design of new
rocket propulsion systems are often modified by appropriate correction factors so
as to get theoretical results as close to the empirical values.

The following assumptions define the operation of an ideal rocket propulsion unit
[12, p. 46]:
14

▪ The working fluid (which usually consists of chemical reaction products) is

homogeneous in composition.

▪ All the species of the working fluid are treated as gaseous. Any condensed
phases (liquid or solid) add a negligible amount to the total mass.

▪ The working fluid obeys the perfect gas law.

▪ There is no heat transfer across any and all gas-enclosure walls; therefore,
the flow is adiabatic.

▪ There is no appreciable wall friction and all boundary layer effects may be
neglected.

▪ There are no shock waves or other discontinuities within the nozzle flow.

▪ The propellant flow rate is steady and constant. The expansion of the working
fluid is uniform and steady, without gas pulsations or significant turbulence.

▪ Transient effects (i.e., startup and shutdown) are of such short duration that
may they be neglected.

▪ All exhaust gases leaving the rocket nozzles travel with a velocity parallel to
the nozzle axis.

▪ The gas velocity, pressure, temperature, and density are all uniform across
any section normal to the nozzle axis.

▪ Chemical equilibrium is established within the preceding combustion cham-

ber and gas composition does not change in the nozzle (i.e., frozen composi-
tion flow).

▪ Ordinary propellants are stored at ambient temperatures. Cryogenic propel-

lants are at their boiling points.

These assumptions allow for the derivation of compact, quasi-one-dimensional set

of equations for the ideal operation of a rocket propulsion system.

2.2 PERFORMANCE PARAMETERS

2.2.1 Thrust

The quintessential parameter defining the performance of a rocket engine is the

propulsive force or thrust it generates.
 15

The total thrust (in N) produced by a propulsion system is given as

𝐹 = 𝑚̇ 𝑣𝑒 + 𝐴𝑒 (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ) (2.1)

The term 𝑚̇ 𝑣𝑒 represents momentum thrust (given by the product of the propellant
mass flow rate 𝑚̇ and the average exhaust velocity of the reaction mass at nozzle
exit 𝑣𝑒 ), and the term 𝐴𝑒 (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ) denotes pressure thrust (given by the product
of the cross-sectional area of the nozzle exit 𝐴𝑒 and the difference between the exit
pressure of the combustion fluid 𝑃𝑒 and the external pressure of the surrounding
fluid 𝑃𝑒𝑥𝑡 ). While the momentum thrust is constant during steady operation of a
thrust chamber, the pressure thrust does vary with any variation in the ambient
pressure around the thrust chamber, and can amount from 10 to 30% of the overall
thrust during the ascent of a rocket [13].

At altitudes where the ambient pressure is higher than the exhaust pressure, the
pressure thrust term is negative and contributes to a lower-than-rated overall
thrust; the nozzle in this condition is said to be over-expanded (and creates shock
diamonds in the exhaust plume, thus losing efficiency). In contrasting conditions
where the ambient pressure is lower than the exhaust pressure, the pressure thrust
term, although positive, yet again contributes to a lower overall thrust; the nozzle is
said to be under-expanded (and ejects the exhaust jet without complete expansion,
thus again losing efficiency).

When the exhaust pressure equals the ambient pressure, the pressure thrust term
becomes zero, and the nozzle is said to operate at its optimum expansion ratio. This
condition of perfect expansion maximizes thrust (and thereby efficiency) from the
nozzle for a given ambient pressure, and is only achieved at a particular altitude
during the upward flight of a launch vehicle.

The total thrust from an optimally-expanded nozzle, therefore, can be expressed as

𝐹 = 𝑚̇ 𝑣𝑒 (2.2)

2.2.2 Specific Impulse

The fundamental metric determining the efficiency of a rocket engine is the specific
impulse of the propulsion system. It directly contributes to the change in velocity
(𝛥𝑣) of a rocket stage at propellant burnout and has a pronounced effect on mass
of the payload: the higher the specific impulse, the less is the propellant mass
required to provide the desired impulse.
16

Expressed in units of time (in s), specific impulse is defined as the ratio of the
thrust generated to the weight flow rate 𝑤̇ of the ejected propellant.

𝐹 𝐹
𝐼𝑠 = = (2.3)
𝑤̇ 𝑚̇ 𝑔0

The expression 2.3 can be formulated in terms of exhaust velocity 𝑣𝑒 , nozzle exit
area 𝐴𝑒 , pressure difference at the nozzle exit (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ), and mass flow rate 𝑚̇ as
follows
𝑣 𝐴
𝐼𝑠 = 𝑒 + 𝑒 (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ) (2.4)
𝑔0 𝑚̇ 𝑔0

Specific impulse, when defined as the total impulse delivered per unit mass of the
propellant consumed (or as thrust per unit propellant mass flow rate), does in
essence represent the effective exhaust velocity of the expelled propellant relative
to the rocket, and can thus have units of speed (in m/s).

𝐼𝑡 𝐹
𝐼𝑠 = = (2.5)
𝑚𝑝 𝑚̇

The specific impulse of an engine operating in a vacuum environment is called its

vacuum specific impulse, and is denoted as 𝐼𝑠𝑣𝑎𝑐 .

2.2.3 Effective Exhaust Velocity

The efflux of propellants from a thrust chamber creates a non-uniform velocity

profile across the exit cross sectional area of the nozzle. Because an accurate
measurement of the actual exhaust velocity is difficult, “a uniform axial velocity
is assumed for all calculations which employ one-dimensional problem descrip-
tions” [12, p. 27]. Effective exhaust velocity, thus, represents “an average or
mass-equivalent velocity” [12, p. 28] at which the exhaust jet is ejected from the
thrust chamber, and can be expressed as

𝑐 = 𝐼 𝑠 𝑔0
𝐴𝑒
= 𝑣𝑒 + (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ) (2.6)
𝑚̇

Presence of the external pressure dependent term in the equation 2.6 implies that
the effective exhaust velocity equals the average actual nozzle exhaust velocity 𝑣𝑒
only for an optimum expansion ratio condition.
 17

Effective exhaust velocity may further be defined as the product of characteristic

velocity 𝑐∗ and thrust coefficient 𝐶𝑓 .

𝑐 = 𝑐∗ 𝐶𝑓 (2.7)

2.2.4 Characteristic Velocity

Characteristic velocity reflects the design characteristics of a thrust chamber in-
dependent of the nozzle performance, and represents the effective energy of the
propellant combustion within the combustion chamber. It is not a physical ve-
locity but a measure of the combustion performance—propellant properties and
combustion characteristics—of a thrust chamber, and thus can be used as a conve-
nient parameter when comparing different propellant combinations for propulsion
systems [13].
Characteristic velocity has units of speed (in m/s) and may be defined as

𝑃𝑐𝑐 𝐴𝑡
𝑐∗ = (2.8)
𝑚̇

The equation 2.8 indicates that the lower the rate of propellant consumption in
order to maintain the required nozzle stagnation pressure, the higher is the energy
(and efficiency) of the combustion process and thereby the higher is the value of
characteristic velocity.
Characteristic velocity may again be determined from the measurements of specific
heat ratio 𝑘, gas constant 𝑅, and combustion chamber temperature 𝑇𝑐𝑐 as follows

√𝑘 𝑅 𝑇𝑐𝑐
𝑐∗ = (2.9)
( 𝑘+1 )
𝑘√( 𝑘+1
2
) 𝑘−1

In the equation 2.9, specific heat ratio 𝑘 is the ratio of heat capacity at constant
pressure 𝑐𝑃 to heat capacity at constant volume 𝑐𝑉 , and gas constant 𝑅 is the
universal gas constant 𝑅∗ (= 8.314462618 J/K/mol [14]) divided by the effective
molecular mass 𝔐 of the fluid (in g/mol).

2.2.5 Thrust Coefficient

Thrust coefficient reflects the design characteristics of a de Laval nozzle, and rep-
resents the expansion properties of the combustion fluid within the supersonic
18

nozzle of a thrust chamber. It measures the amplification of thrust generated due

to the reaction mass expansion through the nozzle of a thrust chamber regardless
of its combustion chamber characteristics, and is a convenient parameter for esti-
mating the effects of chamber pressure and/or altitude variation on a given nozzle
configuration [12].

Thrust coefficient is a dimensionless quantity with values ranging from just under
1.0 to roughly about 2.0 [12], and may be defined as

𝐹
𝐶𝑓 = (2.10)
𝑃𝑐𝑐 𝐴𝑡

Thrust coefficient may again be formulated in terms of specific heat ratio 𝑘, com-
bustion chamber pressure 𝑃𝑐𝑐 , exit pressure 𝑃𝑒 , external pressure 𝑃𝑒𝑥𝑡 , and nozzle
expansion area ratio 𝜖 as follows
√
√ 2 𝑘2 ( 𝑘+1 ) ( 𝑘−1 )
2 𝑘−1
⎡ 𝑃𝑒 𝑘
⎤ 𝑃𝑒 − 𝑃𝑒𝑥𝑡
√
𝐶𝑓 = ( )( ) ⎢1 − ( ) ⎥+( )𝜖 (2.11)
𝑘−1 𝑘+1 𝑃𝑐𝑐 𝑃𝑐𝑐
⎷ ⎣ ⎦

In the equation 2.11, expansion ratio 𝜖 is the ratio of exit area 𝐴𝑒 to throat area 𝐴𝑡
of a nozzle.

Combining the equations 2.10 and 2.11, thrust can now be expressed as
√
√ 2 𝑘2 ( 𝑘+1 ) ( 𝑘−1 )
2 𝑘−1
⎡ 𝑃𝑒 𝑘
⎤
√
𝐹 = 𝑃𝑐𝑐 𝐴𝑡 ( )( ) ⎢1 − ( ) ⎥ + 𝐴𝑒 (𝑃𝑒 − 𝑃𝑒𝑥𝑡 ) (2.12)
𝑘−1 𝑘+1 𝑃𝑐𝑐
⎷ ⎣ ⎦

The equation 2.12, known as the ideal thrust equation, shows that thrust is propor-
tional to throat area and chamber pressure, and is a function of the pressure ratio
across the nozzle, the pressure thrust component and specific heat ratio.

2.2.6 Average Exhaust Velocity

The theoretical exhaust velocity of the reaction mass ejected from an ideal thrust
chamber can be expressed as
√
√ 𝑇𝑐𝑐 ( 𝑘−1 )
2 𝑘 ⎡ 𝑃 𝑒
𝑘
⎤
√
𝑣𝑒 = 𝑅∗ ( ) ⎢1 − ( ) ⎥ (2.13)
𝔐 𝑘−1 𝑃𝑐𝑐
⎷ ⎣ ⎦
 19

The equation 2.13 indicates that exhaust velocity is a function of pressure ratio
𝑃𝑐𝑐 /𝑃𝑒 across the nozzle and specific heat ratio 𝑘, and is directly proportional to
the absolute temperature at the nozzle inlet 𝑇𝑐𝑐 and inversely proportional to the
molecular mass of the combustion fluid 𝔐. With other parameters remaining
constant, an increase in the ratio 𝑇𝑐𝑐 /𝔐 will increase the exhaust velocity thereby
increasing the specific impulse of the propulsion system.

2.2.7 Propellant Mass Flow Rate

Now, the theoretical mass flow rate of propellants (in kg/s) exhausted through a
thrust chamber operating under ideal conditions can be expressed as
√ 𝑘+1
√ 𝑘 ( 2 )( 𝑘−1 )
√ 𝑘+1
𝑚̇ = 𝑃𝑐𝑐 𝐴𝑡 (2.14)
⎷ 𝑅 𝑇𝑐𝑐

And for a propellant mass mixture ratio 𝑂/𝐹, the respective oxidizer and fuel mass
flow rates equal

(𝑂/𝐹)
𝑚̇ 𝑜 = 𝑚̇ (2.15a)
(𝑂/𝐹) + 1
1
𝑚̇ 𝑓 = 𝑚̇ (2.15b)
(𝑂/𝐹) + 1

The respective mass of the oxidizer and the fuel can be determined from the product
of their mass flow rates and the nominal burn time 𝑡𝑏 of the rocket engine

𝑚𝑜 = 𝑚̇ 𝑜 𝑡𝑏 (2.16a)
𝑚𝑓 = 𝑚̇ 𝑓 𝑡𝑏 (2.16b)

2.2.8 Density Impulse

The density impulse “expresses the total impulse delivered per unit volume of
the propellant” [13, p. 19], and is defined as the product of specific weight 𝑑 of a
propellant combination and the specific impulse. Denoted as 𝐼𝑠𝑑 , it is an important
measure in launch vehicle design since a lower density impulse implies the need
for a larger tank volume to store the propellant, which in turn has a detrimental
effect on the mass ratio of a rocket.

𝐼𝑠𝑑 = 𝐼𝑠 𝑑 (2.17)
20

2.3 QUASI 1-D FLOW PARAMETERS

2.3.1 Isentropic Flow Condition

For any quasi-one-dimensional flow process undergoing isentropic expansion, the

following relations hold true between two nozzle sections x and y

( 𝑘−1 ) 𝑉𝑦 (𝑘−1)
𝑇𝑥 ⎛ 𝑃𝑥 ⎞ 𝑘
=⎜ ⎟ =( ) (2.18)
𝑇𝑦 ⎝ 𝑃𝑦 ⎠ 𝑉𝑥

Stagnation conditions arise when the flow is stopped isentropically inside a com-
bustion chamber, and the local temperatures and pressures approach the stagnation
temperatures and pressures.

The Mach number 𝑀 is a dimensionless flow parameter used to locally define the
ratio of the flow velocity 𝑣 to the local acoustic velocity 𝑎, and is given as
𝑣 𝑣
𝑀= = (2.19)
𝑎 √𝑘 𝑅 𝑇

Now, the relations between Mach number with stagnation temperature and stag-
nation pressure are written as

1
𝑇0 = 𝑇 (1 + (𝑘 − 1) 𝑀2 ) (2.20)
2
𝑘
( 𝑘−1 )
1
𝑃0 = 𝑃 (1 + (𝑘 − 1) 𝑀2 ) (2.21)
2

The nozzle area ratio for isentropic flow in terms of Mach number at two locations
x and y within the nozzle can be written as
√
√ 𝑘−1 ( 𝑘+1
𝑘−1
)
𝑀 1 + ( ) 𝑀 2
𝐴𝑥 𝑦 √⎛ 2 𝑥⎞
= √⎜
⎜ ⎟
⎟ (2.22)
𝐴𝑦 𝑀𝑥 1 + ( 𝑘−1
) 𝑀 2
⎷⎝ 2 𝑦 ⎠

2.3.2 Nozzle Throat Conditions

From the above relations, the temperature 𝑇𝑡 , pressure 𝑃𝑡 and specific volume
𝑉𝑡 at the throat can be expressed in terms of the respective parameters from the
 21

combustion chamber as follows

2
𝑇𝑡 = 𝑇𝑐𝑐 ( ) (2.23)
𝑘+1
𝑘
( 𝑘−1 )
2
𝑃𝑡 = 𝑃𝑐𝑐 ( ) (2.24)
𝑘+1
1
( 𝑘−1 )
𝑘+1
𝑉𝑡 = 𝑉𝑐𝑐 ( ) (2.25)
2

Likewise, the throat velocity 𝑣𝑡 can be written as

2𝑘
𝑣𝑡 = √( ) 𝑅 𝑇𝑐𝑐 = √𝑘 𝑅 𝑇𝑡 = 𝑎𝑡 (2.26)
𝑘+1

By definition, the sonic velocity 𝑎𝑡 , or the Mach number, at the nozzle throat equals
1.0.

Now, for any region y downstream of the throat, the following relations, expressed
in terms of pressure ratio, hold true

( 𝑘−1 )
𝑃𝑦 𝑘
𝑇𝑦 = 𝑇𝑐𝑐 ( ) (2.27)
𝑃𝑐𝑐
(1)
𝑃 𝑘
𝑉𝑦 ⎜ 𝑐𝑐 ⎞
= 𝑉𝑐𝑐 ⎛ ⎟ (2.28)
⎝ 𝑃𝑦 ⎠
√ 𝑘−1
𝑣𝑦 √ 𝑘+1 ⎛ 𝑃𝑦 ( 𝑘 ) ⎞
= √( )⎜⎜1 − ( ) ⎟
⎟ (2.29)
𝑣𝑡 𝑘−1 ⎜ 𝑃𝑐𝑐 ⎟
⎷ ⎝ ⎠
1
√ 𝑘−1
( 1 ) 𝑃𝑦 ( 𝑘 ) √ 𝑘 + 1 ⎛ 𝑃𝑦 ( 𝑘 ) ⎞
𝐴𝑡 𝑘 + 1 𝑘−1
=( ) ( ) √( )⎜
⎜1 − ( ) ⎟
⎟ (2.30)
𝐴𝑦 2 𝑃𝑐𝑐 𝑘−1 ⎜ 𝑃𝑐𝑐 ⎟
⎷ ⎝ ⎠
𝑉𝑡 𝑣𝑦
=
𝑉𝑦 𝑣𝑡

2.4 REAL ROCKET PROPULSION SYSTEM

The assumptions defining the operation of an ideal rocket propulsion system

allow for the analysis and design a real rocket propulsion unit using relatively
22

simple mathematical approximations and relations. In order to represent a better

simulation of energy losses or physical and chemical phenomena, and contain more
complex theoretical descriptions, it is possible either to use empirical correction
factors based on experimental data or to implement more accurate algorithms
besides the above assumptions. Compared to ideal thrust chambers, real thrust
chambers suffer from numerous operational losses, some of which are mentioned
below [12, p. 81]:

▪ Divergence of the flow in the nozzle exit sections is a loss that varies as a
function of the cosine of the divergence angle for conical nozzles. These
losses can be reduced with bell-shaped nozzle contours.

▪ Low nozzle contraction ratios cause pressure losses in the chamber and
slightly reduce the thrust and exhaust velocity.

▪ The lower velocities at the wall boundary layers reduce the effective average
exhaust velocity by 0.5 to 1.5%.

▪ Solid particles and/or liquid droplets in the gas may cause losses of perhaps
up to 5% (depending on particle size, shape and percentage).

▪ Unsteady combustion and/or flow oscillations may result in small losses.

▪ Chemical reactions within nozzle flows change gas composition and gas
properties, amounting to typically a 0.5% loss.

▪ Chamber pressures and overall performance are lower during start and stop
transient operations.

▪ Any gradual erosion of the throat region increases its diameter by perhaps
1 to 6% during operation with uncooled nozzle materials. This, in turn,
reduces the chamber pressure and thrust by about 1 to 6%.

▪ Nonuniform gas compositions may reduce performance due to incomplete

mixing or incomplete combustion.

▪ Real gas properties may noticeably modify gas composition, that is, actual
values of 𝑘 and 𝔐 cause a small loss in performance by about 0.2 to 0.7%.

When the expansion of the working fluid in a nozzle is sufficiently rapid, its chem-
ical composition may be assumed as invariant throughout the nozzle, meaning,
there are no chemical reactions or phase changes and the reaction products compo-
sition at the nozzle exit are identical to those of the chamber exit. Such composition
 23

results are known as frozen equilibrium rocket performance. This approach, being
the simplest, tends to underestimate the system’s performance by 1 to 4% [12].

Instantaneous chemical equilibrium among all molecular species may be significant

in certain circumstances under the continuously variable pressure and temperature
conditions of the nozzle expansion process. Product compositions do shift in this
case because the chemical reactions and phase change equilibria occurring between
gaseous and condensed phases in all exhaust gas species are fast compared to their
nozzle transit time; this results in different gas composition mass fractions at the
chamber and nozzle exits. The composition results so calculated are called shifting
equilibrium performance. This analysis, being more complex, usually overstates
real performance values such as 𝑐∗ or 𝐼𝑠 by 1 to 4%.

Correction factors based on experimental data from extant rocket propulsion

systems are implemented during the preliminary analysis and design of new
rocket propulsion systems in order to account for much of the above non-ideal
phenomena [12], such as internal and/or kinetic energy losses, imperfect mixing
and/or combustion, heat transfer, friction, etc.

The thrust correction factor 𝜁𝐶𝑓 (= 𝐹𝑎 /𝐹𝑖 ) is determined from the ratio of thrust
measurements to its corresponding ideal values. The 𝑐∗ correction factor 𝜁𝑐∗ repre-
sents a combined effectiveness of the combustion chamber and the injector design
while the 𝐶𝑓 correction factor 𝜁𝐶𝑓 represents the effectiveness of the nozzle design
at its operating conditions. The discharge correction factor 𝜁𝑚̇ (= 𝑚̇ 𝑎 /𝑚̇ 𝑖 ) can be
determined from the ratio of mass flow rate measurements with the corresponding
theoretical values, and is, because of compressible flow properties, somewhat
greater than 1.0 [12].

Determining the above correction factors paves the way for estimating efficiencies
such as those of the exhaust velocity, specific impulse or of the nozzle itself, and
for calculating parameters such as the actual throat and exit areas or the local
stagnation temperatures and pressures of the working fluid across the thrust
chamber.

2.5 ENGINE NOMENCLATURE

The fundamental components/sections of a thrust chamber are the injector, the

combustion chamber, and the (convergent-divergent) nozzle, which in turn is
segmented into the inlet, the throat, and the exit. The schematic and convention
24

our engine design parameters follow are illustrated in figure 2.1 and presented in
table 2.1 below.
Thrust Chamber

Injector Combustion Chamber Nozzle

OXIDIZER
EXHAUST
FUEL

Nozzle Nozzle Nozzle

Inlet Throat Exit

External Convergent Divergent

Conditions Section Section

Figure 2.1.: Schematic of a typical liquid propellant rocket engine together with its
major components/sections.

Table 2.1.: Convention used for the design parameters of our liquid propellant
rocket engine.
Thrust Chamber 𝑡𝑐
Injector 𝑖𝑛𝑗
Combustion Chamber 𝑐𝑐
Nozzle 𝑛
Inlet 𝑖
Throat 𝑡
Exit 𝑒
Convergent section 𝑐
Divergent section 𝑑
External conditions 𝑒𝑥𝑡
First stage 1st
Second stage 2nd
3
Liquid Propellants

A propellant is largely defined as any substance that undergoes chemical changes

and thermodynamical processes within the thrust chamber of a propulsion system
in order to generate thrust. A liquid rocket is primarily distinguished as a mono-
propellant system or a bipropellant system according to the number of liquid
propellants it employs for achieving propulsion. While a monopropellant sys-
tem provides propulsive reaction from the decomposition of a single propellant,
a bipropellant system comprises an oxidizer and a fuel, and produces resultant
thrust from the combustion of the propellant mixture. Based on the temperature
range of its storage, a propellant can be classified as a cryogenic or a storable liquid.
Depending on the process of its ignition, a propellant mixture can be classified as
a hypergolic or a nonhypergolic combination. And as a function of its application, a
liquid rocket engine may further be categorized as a boost propulsion system (used
predominantly for imparting significant total impulses and velocity changes to
payloads) or an auxiliary propulsion system (used principally for attitude control,
orbit maintenance and minor trajectory corrections).

3.1 PROPELLANT CHARACTERISTICS

The choice of propellants for a liquid bipropellant system has paramount implica-
tions on design criteria of the rocket engine and performance metrics of the overall
rocket, and significantly influences such considerations as production, storage,
handling, cost and even mission compatibility. When it comes to performance,
there is no one propellant combination that has an absolute advantage over another
combination: selection of a propellant is therefore a compromise between some
important aspects and some desirable features.
26

3.1.1 Important Aspects & Desirable Features

Thrust, as expressed in the equation 2.2, has been defined as the product of propel-
lant mass flow rate and average exhaust velocity. And as specified in the equations
2.14 and 2.13 respectively, mass flow rate and exhaust velocity are in turn de-
pendent on a combination of thrust chamber characteristics and thermochemical
reaction properties. With other parameters remaining constant and for a given rate
of consumption of propellants, therefore, an increase in the ratio 𝑇𝑐𝑐 /𝔐 does result
in an increased overall thrust. Because exhaust velocity is directly proportional to
combustion chamber temperature and inversely proportional to effective molecular
mass of the exhaust jet, a propellant combination that generates the highest adia-
batic flame temperature and/or produces the lowest average molecular weight of
the combustion products is usually preferred. Additionally, a bipropellant mixture
delivering a low specific heat ratio enhances the performance parameters of the
rocket engine and is thus favoured [13].
In addition to the above propellant characteristics, there are numerous physical
properties which, although don’t have any direct consequence on flight perfor-
mance parameters, are yet of considerable importance during the design of a rocket
engine. One such property is the propellant’s specific gravity. The higher the spe-
cific gravity of a propellant is, the smaller is the volume of tank required for a
certain mass of the propellant, the lower, therefore, get the vehicle’s structural
mass and aerodynamic drag, and the higher, consequently, become the delta-v
and payload carrying capacity of the rocket [12]. Moreover, increasing the propel-
lant’s density allows for an increase in its mass flow rate, which yet again enables
increases in the rocket engine’s combustion chamber pressure, total impulse, and
thrust. Likewise, heat transfer properties such as high specific heat, high thermal
conductivity, low freezing point and high boiling point are of crucial importance
for a propellant when it serves as the coolant fluid in a regeneratively cooled thrust
chamber. Furthermore, a propellant with low vapour pressure and low viscosity
permits effective pump feed system designs and minimizes pressure drops through
the plumbing system.

3.1.2 Classification of Propellants

A hypergolic propellant combination is one where the oxidizer and the fuel ignite
spontaneously upon contact, and thus do not require any source of ignition after
injection for their combustion. While enabling rocket engine designs with great
simplicity and reliability, these propellants usually are rather difficult to handle
owing to their extreme corrosivity, toxicity and oftentimes carcinogenicity [15]. A
 27

few commonly used hypergols include oxidizers like hydrogen peroxide, dinitrogen
tetroxide (NTO) and inhibited red fuming nitric acid (IRFNA), and fuels like
hydrazine, monomethylhydrazine (MMH) and unsymmetrical dimethylhydrazine
(UDMH). Hypergolic propellants typically yield specific impulses between 260 s
and 290 s under optimum conditions [12] [13].

Cryogenic propellants in essence are liquefied gases with extremely low boiling
points (from −140 °C to −260 °C) under ambient pressure conditions [13]. And
because of such low temperatures, handling and storing these propellants can
have certain inherent difficulties: cryogenics necessitate elaborate procedures in
order to minimize losses due to propellant boiloff. Propellant tanks and plumbing
systems, therefore, employ insulating systems to keep the propellants cold and
venting systems to dispose of boiloff vapours—two critical measures which, besides
adding complexity, increase the inert mass of the rocket. These drawbacks, however
consequential, are unqualifiedly eclipsed by high specific impulse and thereby
high overall vehicle performance that cryogenic propellants offer as compared to
most other storable propellants [15] [16]. Using, for instance, liquid oxygen (LOX)
in conjunction with liquid hydrogen (LH2) can yield specific impulse as high as
465s under the right circumstances [17] [18]. Combining LOX with liquid methane
(LCH4) is another similar mixture with highly desirable performance parameters
[12] [13]. Oftentimes, a non-cryogenic fuel like Rocket Propellant 1 (RP-1) is
paired with LOX to produce a combination that yet again delivers satisfactorily
high engine performance [12] [13].

3.1.3 Propellant Selection Criteria

In a nutshell, the most important aspects and desirable features used in the selection
of liquid propellants have been listed below [13, p. 19], order of importance of
which may vary as a function of the application:

▪ High chemical energy content per unit mass of propellant mixture consumed
for higher overall thrust.

▪ Low molecular mass of the combustion or decomposition products for higher

specific impulse.

▪ High density impulse (total impulse/volume) for minimizing the mass and
size of propellant tanks and feed systems.

▪ Optimum combination of high specific heat, high thermal conductivity and

high critical temperature for propellants used as thrust chamber coolants.
28

▪ Low viscosity for minimizing pressure drops through feed system and injec-
tor.
▪ Low vapor pressure for low tank mass and low net positive pump suction
head requirement.
▪ Low freezing point to facilitate engine operation at low temperature.
▪ Stability of combustion.
▪ Ease of ignition.
▪ Availability.
▪ Cost.

3.2 CANDIDATE PROPELLANTS

Having touched on a few fundamental factors influencing the selection of propel-

lants for a liquid rocket propulsion system, this project aims to take propellant
combinations that are nonhypergolic and preferably cryogenic in nature into con-
sideration. A few candidate propellants, along with a handful of their pros and
cons, have therefore been looked at below.

3.2.1 Liquid Oxygen (LOX)

Among the most widely used oxidizers, liquid oxygen has (at different times
throughout history) been combined with such fuels as liquid hydrogen, petroleum
derivatives like gasoline, JP-4, LNG and RP-1, organic compounds like methane,
dimethoxymethane, methanol, ethanol, ethylene and methylamine, as well as
even ammonia, hydrazine, UDMH and hydyne [19]. Being fairly cryogenic (with
a normal boiling point of 90.20 K or −182.95 °C [20]), LOX cannot be used in
applications necessitating storage of the propellant for extended lengths of time.
Notwithstanding the drawback, LOX has some quite advantageous properties to
offer. It is a stable, non-corrosive and non-toxic fluid that supports combustion
vigorously despite itself being non-combustible [12]. It has low molecular mass (of
31.999 g/mol), high specific gravity (of 1.142), low vapour pressure, low viscosity,
and delivers relatively high performance with most fuels [20] [21] [22].
We, therefore, use liquid oxygen as the oxidizer of choice in the design of our
propulsion system. Doing so, however, essentially limits our fuel options to the
 29

following three: liquid hydrogen (LH2), Rocket Propellant 1 (RP-1) and liquid
methane (LCH4).

3.2.2 Liquid Hydrogen (LH2)

Considered an ideal rocket propellant, liquid hydrogen offers the highest perfor-
mance metrics of all fuels used in chemical rocket propulsion [19] [23]. Hydrogen
has the lowest molecular weight of any known substance (2.016 g/mol [24]) and
the highest specific energy of any non-radioactive fuel (141.766 MJ/kg [24]) [23].
When paired with an oxidizer, LH2, consequently, produces relatively low average
molecular weight of the combustion products and generates comparatively high
adiabatic flame temperature—two factors that help maximize the exhaust velocity
and thus the specific impulse of the rocket engine. Moreover, the combustion
of LH2 with LOX is extremely energetic and efficient (producing typically only
water vapour), and yields the highest theoretical specific impulse of any conven-
tional bipropellant combination [23]. In addition to the above characteristics, such
properties as low viscosity, high specific heat and high thermal conductivity make
LH2 a desirable coolant for thrust chambers [25] [12]. LH2 is manufactured from
gaseous hydrogen by successive compression, cooling and expansion processes
[12], and is a non-corrosive, non-toxic, colourless and odourless fluid [24].
LH2 does come with a few major downsides, however. Firstly, liquid hydrogen has
an awfully low density (of only 70.78 kg/m3 at its boiling point under normal pres-
sure [24]). Such low density implies that LH2 “takes an inordinate tankage volume
to hold any great amount of the fuel” [19, p. 106]; requiring voluminous and thus
massive propellant tanks in turn compounds drag on the rocket during ascent and
detracts mass from that of the payload. Despite hydrogen’s exceptionally high
specific energy (energy per unit mass), LH2 offers a remarkably low volumetric
energy density (energy per unit volume)—an attribute that stems solely from its
low specific gravity [26]. Secondly, LH2 is a highly cryogenic liquid (with a normal
boiling point of just 20.37 K or −252.78 °C [24]). Dealing with the fuel, therefore,
poses formidable technical challenges such as adequacy of thermal insulation
(to minimize propellant evaporation and boiloff losses), allowance for venting
mechanism (to vent any gaseous fluid overboard), and selection of compatible
materials (since many materials become brittle at such temperatures) [19]. And
thirdly, hydrogen is extremely flammable over a wide range of mixture ratios (thus
making it a ferocious fire and explosion hazard) [12].
Despite having its shortcomings, the advantages of liquid hydrogen far outweigh
any disadvantages of the fuel. It offers a theoretical specific impulse of about
30

380 s (under the following conditions: combustion chamber pressure = 1000 psi
= 6894.757 kPa; nozzle exit pressure = 1 atm = 101.325 kPa; optimum expansion)
when used in conjunction with LOX [27] [12]. And therefore, LH2 finds application
in a number of rocket engines worldwide. Some of those engines/rockets include
the USA’s J-2 (used on Saturn V), RL-10 (used on Alas V, Delta IV, Space Launch
System and Vulcan Centaur), RS-25 (used on the Space Shuttle orbiter and Space
Launch System) and RS-68 (used on Delta IV); Europe’s HM-7B, Vulcain and Vinci
(used on the Ariane launch vehicle family); India’s CE-7.5 and CE-20 (used on
GSLV Mark II and GSLV Mark III respectively); Japan’s LE-5 and LE-7 (for H-I and
the H-II series of launch vehicles); and China’s YF-73, YF-75 and YF-77 (used on
the Long March family of rockets) [16] [28] [29]. And it’s for the same reason that
this project takes performance parameters from the LOX/LH2 combination into
account for analysis.

3.2.3 Rocket Propellant 1 (RP-1)

The first experimental flight of a liquid rocket (back in the March of 1926 by Robert
Goddard) was fueled by gasoline [19]. And ever since those early days of rocketry,
a number of petroleum derivative fuels have been used to power liquid propellant
rockets—a handful of which include (variants of) kerosene, gasoline and liquefied
natural gas. But these complex hydrocarbon mixtures exhibited wide variance in
physical properties and chemical compositions among each fuel of the same class
(due to inconsistencies in process of refining and in degree of control exercised
during their manufacture), which meant that no two barrels of the fuel were alike
enough to produce accurate and precise results across engine tests and to meet “the
narrow limits of specified engine performance that is needed for a predictable flight
path of many a vehicle” [12] [19] [15]. This led to the requirement of petroleum
fuels meeting rather rigorous specifications and ultimately to the development of a
highly refined product known as RP-1. As described by Clark, RP-1 is “a kerosene
in the C12 region, with a H/C ratio between 1.95 and 2.00, containing about 41
percent normal and branched paraffins, 56 of naphthenes, 3 of aromatics, and no
olefins at all” [19, p. 105].

Despite offering only a modest specific impulse, RP-1 is used quite often in com-
bination with liquid oxygen for high-total-impulse applications—perhaps more
frequently than even LH2 [28]. This is due in part to the high energy density of
RP-1 which produces a significantly energetic reaction for each volume of the fuel
consumed, which in turn generates a correspondingly high temperature of com-
bustion [30] [12]. But because it’s a hydrocarbon mixture comprising numerous
 31

complex molecular structures, its combustion with LOX yields products with fairly
high effective molecular weights, which eventually lowers the ratio 𝑇𝑐𝑐 /𝔐, and
consequently delivers only a decent exhaust velocity [31]. Besides energy density,
RP-1 has a high specific gravity (within 0.799 and 0.815 at about 15 °C [32]), which
allows for pretty small and light propellant tanks for a given mass of the fuel.
Additionally, it is a storable, non-corrosive, non-toxic propellant that is, despite
being flammable, relatively easy to handle [30].
RP-1, nevertheless, has its own challenges to surmount. Primarily, it has a tendency
to undergo polymerization under high enough temperatures [19]. Petroleum
derivative fuels, due to the thermal instability of their complex organic compounds,
experience thermal cracking (wherein long hydrocarbons chains break down into
simpler molecules) as temperature of the fuel rises [30]. This eventually results
in the deposition of soot, coke, and other assorted carbonized residues at regions
of high heat transfer rates which, while occasionally acting as a thermal insulator,
mostly impedes heat transfer, inhibits propellant flow, and ultimately fouls up
intricately designed components of (typically regeneratively cooled) rocket engines
[19] [15]. Moreover, and much like any other fuel, RP-1 is combusted slightly fuel
rich with LOX in order to marginally raise the specific impulse of the bipropellant
combination; doing so, however, further exacerbates the polymerization problem.
Notwithstanding the downsides, RP-1 together with LOX continues being the semi-
cryogenic combination of choice worldwide. Burning the pair delivers a theoretical
specific impulse of about 290 s (under these conditions: combustion chamber
pressure = 1000 psi = 6894.757 kPa; nozzle exit pressure = 1 atm = 101.325 kPa;
optimum expansion) [27] [12]. The bipropellants have notably been used in the
following engines/rockets: the F-1 engine (used on Saturn V); RD-170 (the Energia
launch vehicle) and its variants like RD-171 (the Zenit family), RD-180 (Atlas
V) and RD-191 (in both Angara and Antares); RD-107 and RD-108 (in different
configurations in the R-7 derivative vehicles including Vostok, Voskhod and Soyuz);
Merlin (Falcon 9 and Falcon Heavy); and Rutherford (Electron) [16] [28] [31].
This project, therefore, looks at performance parameters from the bipropellant
combination for analysis.

3.2.4 Liquid Methane (LCH4)

Methane, chemically called carbon tetrahydride, is the simplest hydrocarbon
molecule, and is composed of four hydrogen atoms attached to one carbon atom.
Also known as marsh gas, it is the primary constituent of natural gas, and finds
extensive usage as a hydrocarbon fuel owing to its relative abundance on Earth. It is
32

a colourless and odourless gas with a boiling point of 111.65 K or −161.50 °C under
normal conditions [33]. Upon stoichiometric combustion, one unit of methane
reacts with two units of oxygen to release one unit of carbon dioxide and two units
of water [34].

Although the usage of liquid methane as a rocket propellant had been proposed
since the 1960s, it has only been recently that the cryogenic fuel accrued any
pragmatic thrust chamber operation and/or test flight experience [35]. Methane is
regarded as the propellant of choice for future long-duration crewed missions to
Mars where it can be synthesized via the Sabatier process using in-situ resources on
the planet [27]. When burned with liquid oxygen, LCH4 generates a high adiabatic
flame temperature (of 3953 K or 3679.85 °C under stoichiometric and normal
temperature and pressure conditions [36]) due to its high heat of combustion
(of −55536 kJ/kg [33]). And being the lightest organic compound (with a molar
mass of just 16.04 g/mol [33]), molecular weights of its combustion products are
comparably low [34]. In addition, methane has a low ratio of specific heat (at 1.306
[33]) which improves both characteristic velocity and thrust coefficient.

Each above characteristic advantages LCH4 in delivering a remarkably high exhaust

velocity and thereby specific impulse. The bipropellant combination yields, as Sut-
ton asserts [15], a theoretical specific impulse of approximately 310 s (under these
conditions: combustion chamber pressure = 1000 psi = 6894.757 kPa; nozzle exit
pressure = 1 atm = 101.325 kPa; optimum expansion and shifting thermochemical
equilibrium in the nozzle).

The remarkably high specific impulse from the LOX/LCH4 combination is, whilst
lower than that of LH2, higher than that of most storable propellants including
RP-1 [27]. Yet unlike LH2, liquid methane has a relatively high density—nearly
six times as much as that of hydrogen—of 422.8 kg/m3 at its normal boiling point
[33]. This implies that LCH4 offers a higher density impulse than LH2 does
and requires much smaller and lighter propellant tanks and associated systems
than LH2 does. Propulsion systems burning liquid methane, therefore, yield
a better overall performance than even liquid hydrogen. Moreover, having the
best hydrogen to carbon ratio of all hydrocarbons, methane burns much cleaner
and more completely than every other hydrocarbon fuel [30]. This forestalls the
negative effects of propellant dissociation and polymerization—which fuels with
complex molecular structures like RP-1 typically foster—within a rocket engine
and facilitates reusability of the propulsion system [30]. Furthermore, temperature
difference between LOX and LCH4 (about 20 K) is much smaller than that between
LOX and either LH2 (roughly 70 K) or RP-1 (nearly 175 K) which permits tank
 33

designs with less stringent requirements for insulation in between the tanks [37].
Additionally, methane has such desirable properties as high specific heat, high
critical temperature and low dynamic viscosity, and is a non-corrosive and non-
toxic albeit flammable propellant that is relatively inexpensive and abundantly
available [38]. The combination of properties above gives LCH4 an edge over the
shortcomings of using either LH2 or RP-1.

Even though liquid methane has a suite of attractive characteristics to offer, devel-
opment of a LOX/LCH4 propulsion system had largely been put off until now [12].
There were a few proposals to build and test methalox rocket engines prior to the
2010s, but the efforts were limited—partly because of bureaucratic quagmire which
deemed any urgency of pursuing qualification of such an engine unwarranted—
and the projects were cancelled [35]. Over the past decade, however, there has
been a renewed interest in the development of LCH4-fueled propulsion systems.
Most notably, the spaceflight company SpaceX has built, tested and flown the
Raptor engine for its Starship launch vehicle. The BE-4 is another similar engine
undergoing development for use on the New Glenn and the Vulcan Centaur. Re-
spective organizations from Europe, Russia, China and India have also announced
their initiatives to develop rocket engines running on methane [16] [28]. Keeping
in line with the current interests, and in a quest to further investigate the perfor-
mance metrics of the bipropellant, this project intends to employ the LOX/LCH4
combination for the development of our experimental engine.

3.3 PROPELLANT PERFORMANCE ANALYSIS

The generation of thrust in a rocket propulsion system ideally involves the following
two processes: combustion and expansion. All species of the propellant mixture
undergo a rapid enough combustion reaction within a long enough residence
time under an isobaric condition inside a combustion chamber, attaining chemical
equilibrium in the process [12]. Then the combustion fluid experiences an adiabatic
and reversible expansion process on entering a supersonic nozzle and undergoes
a drop in temperature and pressure, reflecting the conversion of thermal energy
into kinetic energy [12].

Several methods with varying degrees of complexity have been implemented to

analyse relevant chemical and thermodynamical properties associated with the
above processes. Yet there typically exist two simplistic approaches in analysing
the gas flow characteristics within a supersonic nozzle: frozen composition flow and
shifting composition flow. In a frozen flow condition, the gas expansion is considered
34

instant and its composition is assumed invariant throughout the nozzle, that is, the
combustion fluid experiences no chemical reactions or phase changes during the
expansion process and chemical composition of the reaction products is identical
at both the combustion chamber and the nozzle exit. This approach tends to
underestimate performance values like specific impulse and characteristic velocity
of a propulsion system typically by 1 to 4% [12]. For a shifting flow condition,
however, the gas is presumed to undergo instantaneous chemical reaction and
hence exist in equilibrium among all reaction species during the nozzle flow, i.e.,
the gas composition does change between the combustion chamber and the nozzle
exit. This method usually overestimates performance values by 1 to 4% [12].

Because of the complex chemical and thermodynamical nature of the processes

involved, exact dependence of the performance parameters on the propellant
mixture is difficult to demonstrate through simple analytical formulae; iterative
computing methods have thus been developed to obtain solutions to such problems.
The requisite computations in this project are performed using the tool CEARUN,
an online computer program facilitating use of the NASA code CEA (Chemical
Equilibrium with Applications) [39]. Published in 1994 by Sanford Gordon and
Bonnie J. McBride at the NASA Glenn (formerly Lewis) Research Center, CEA
employs chemical and thermodynamical equilibrium codes in order to analyse
problems involving rocket performance, combustion properties, detonation pa-
rameters, shock parameters, and other similar applications [40]. In addition, the
software Rocket Propulsion Analysis (RPA Standard Edition v.2.x), developed by
Alexander Ponomarenko, is used to perform certain conceptual computations in
the preliminary design of our chemical rocket propulsion system [41].

All calculations in this project have implemented the frozen composition flow
condition in order to obtain slightly conservative values for performance and
account for uncertainties in our engine design procedures. Additional assumptions
in the project include ideal gas behaviour, one-dimensional forms of the continuity,
energy and momentum equations, negligible velocity at the forward end of the
combustion chamber, isentropic expansion in the nozzle, and chemical equilibrium
in the combustion chamber.

3.4 MIXTURE RATIO

The mixture ratio of a bipropellant combination is the ratio of mass flow of the
oxidizer to mass flow of the fuel. For a certain combination of oxidizer and fuel,
the highest flame temperature is obtained under the stoichiometric mixture ratio,
 35

wherein both the species react in the exact stoichiometric proportion resulting in
a complete combustion reaction [42]. However, the maximum exhaust velocity
and thus the highest specific impulse from burning the propellants is achieved
at the optimum mixture ratio, in which the rocket engine typically operates with
an excess proportion of the fuel [13]. A “fuel rich” mixture is oftentimes more
advantageous than otherwise because it not only leaves unburned fuel species in
the combustion fluid (thus lowering effective molecular weight of the exhaust jet)
but also burns cooler inside the combustion chamber (thereby mitigating excessive
thermal loads on the thrust chamber). Yet occasionally, an off-optimum mixture
ratio may be preferred when it allows for such desirable characteristics as better
combustion stability, improved chemical equilibrium during expansion, required
coolant flow rate, or even reduced propellant tank size [13] [42].

3.4.1 Liquid Oxygen & Liquid Methane

The combustion of methane may be summarized as

CH4 + 2 O2 ⟶ CO2 + 2 H2 O

The reaction above is in stoichiometric proportions, comprising 1 mol of CH4 and

2 mols of O2 as reactants and 1 mol of CO2 and 2 mols of H2 O as products. This
implies, with standard atomic weights for the constituent elements being H = 1.008,
C = 12.011 and O = 15.999, complete combustion of methane requires 64 kg of O2
and 16 kg of CH4 for a stoichiometric mixture ratio of 4 ∶ 1 by mass [42].
Yet by convention, a fuel rich mixture instead of the stoichiometric mixture is
employed for the operation of our thrust chamber. In order to determine the best
propellant mixture ratio, we examine the variation of parameters like specific
impulse, combustion temperature, characteristic velocity and molecular mass with
changes in mixture ratio.
Again, the optimum mixture ratio of a propellant mixture is dependent on the
pressures at which a thrust chamber operates. We, therefore, preliminarily perform
our calculations for mass mixture ratios ranging from 1.0 through 7.0 with an
interval of 0.25 and under the following conditions: combustion chamber pressure
𝑃𝑐𝑐 = 1000 psi = 6894.757 kPa; nozzle exit pressure 𝑃𝑒 = 1 atm = 101.325 kPa
(sea level condition); nozzle area ratio 𝜖 = 40 (vacuum condition); frozen flow
characteristic and infinite area combustor option.
The combustion chart, computed in [43], for determining optimum mixture ratio
as a function of combustion chamber pressure at two different nozzle exit pressures
36

for the liquid oxygen and liquid methane combination is presented in figure 3.1.

Mixture Ratio for the highest IS

Combustion Chamber Pressure (atm)

Figure 3.1.: The dependence of optimum mixture ratio as a function of combustion

chamber pressure at two different nozzle exit pressures (of 0.1 atm and
1.0 atm) for liquid oxygen/liquid methane [43].

Results depicting the variation of specific impulse (both at sea level and in vac-
uum), temperature (both inside combustion chamber and at throat), characteristic
velocity, and effective molecular mass as a function of mixture ratio are plotted in
figures 3.2 to 3.4.

On examining the figures, it is evident that an oxidizer-to-fuel mass mixture ratio

of 2.75 yields the highest performance parameters, viz. specific impulse (both
at sea level and in vacuum) and characteristic velocity, for the aforementioned
operating conditions. In addition, the optimum mixture lowers both the effective
molecular mass of combustion fluid by roughly 3 units as well as the temperature
(both inside combustion chamber and at throat) by about 200 K in comparison to
the respective metrics of the stoichiometric mixture. Figure 3.5 portrays the mixture
ratios, optimum and stoichiometric, that respectively produce the highest values
of specific impulse (at sea level) and temperature (inside combustion chamber)
for the bipropellant combination.
 37

4000.00

3500.00
Speci�ic Impulse (m/s)

3000.00

2500.00

2000.00

1500.00 — Vacuum

— Sea Level
1000.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Mixture Ratio

Figure 3.2.: Variation of specific impulse, both at sea level and in vacuum, with
mixture ratio.

4000.00

3500.00

3000.00
Temperature (K)

2500.00

2000.00

1500.00 — Combustion Chamber

— Throat
1000.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Mixture Ratio

Figure 3.3.: Variation of temperature, both inside combustion chamber and at

throat, with mixture ratio.
38

2000.00 35.00

Characteristic Velocity (m/s) 1800.00 30.00

Molecular Mass (g/mol)

1600.00 25.00

1400.00 20.00

1200.00 15.00

1000.00 10.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00
Mixture Ratio

Figure 3.4.: Variation of characteristic velocity and molecular mass with mixture
ratio.

4000.00
Speci�ic Impulse (m/s) & Temperature (K)

3500.00

3000.00

2500.00

2000.00

1500.00

1000.00
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Mixture Ratio

Figure 3.5.: Variation of specific impulse (at sea level) and temperature (inside
combustion chamber) with mixture ratio.
 39

3.4.2 Liquid Oxygen & Liquid Hydrogen

As pledged, we take a look at the performance parameters from the LOX/LH2
combination in this segment. Unlike for other bipropellants, an optimum mixture
ratio for liquid oxygen and liquid hydrogen is not necessarily the best mixture
ratio for the combination. The highest values of specific impulse usually occur at a
mixture ratio of around 3.5 [12]. But because of the awfully low specific gravity of
liquid hydrogen, a slight increase in the oxidizer-to-fuel mass mixture ratio results
in a significant decrease in the volume of the fuel consumed. Braeunig [43] states
that employing a mixture ratio of about 5.5 reduces the tankage volume of the
fuel by almost a quarter. And as mentioned before, the smaller the volume of the
propellant tank is, the lower are the vehicle’s structural mass and aerodynamic drag,
and the higher, consequently, become the delta-v and payload carrying capacity of
the rocket. Employing a higher (and off-optimum) mixture ratio, therefore, offsets
any loss in the specific impulse and results in a better overall vehicle performance.
In practice, most liquid oxygen/liquid hydrogen rocket engines typically operate
at mixture ratios between 5.0 and 6.0 [43].
Mixture Ratio for the highest IS

Combustion Chamber Pressure (atm)

Figure 3.6.: The dependence of optimum mixture ratio as a function of combustion

chamber pressure for LOX/LH2 [43].

The combustion chart, computed in [43], for determining the best mixture ratio as
a function of combustion chamber pressure for liquid oxygen and liquid hydrogen
is presented in figure 3.6. Regardless, we perform our calculations for specific
40

impulse, combustion temperature and molecular mass as a function of mixture

ratio—for the same range of O/F mass mixture ratios and under the same condi-
tions of thrust chamber operation—as in the subsection 3.4.1. And results of the
calculations have been plotted in figure 3.7.

3.4.3 Liquid Oxygen & Rocket Propellant 1

Going ahead, we now look at the performance parameters from the LOX/RP-1
combination here. Since RP-1 does not have any idiosyncrasies that LH2 has,
the procedure to compute the best oxidizer-to-fuel mass mixture ratio for this
bipropellant combination is fairly straightforward.
Mixture Ratio for the highest IS

Combustion Chamber Pressure (atm)

Figure 3.8.: The dependence of optimum mixture ratio as a function of combustion

chamber pressure at two different nozzle exit pressures (of 0.1 atm and
1.0 atm) for LOX/RP-1 [43].

The combustion chart, computed in [43], for determining optimum mixture ratio as
a function of combustion chamber pressure at two different nozzle exit pressures
for LOX and RP-1 is presented in figure 3.8. And as in the previous segment,
we perform our calculations for specific impulse, combustion temperature and
molecular mass as a function of mixture ratio—for the same range of O/F mass
mixture ratios and under the same conditions of thrust chamber operation—as in
the subsection 3.4.1. Results of the calculations have been plotted in figure 3.9.
 41

4000.00 40.00
Speci�ic Impulse (m/s) & Temperature (K)
35.00
3500.00
30.00

Molecular Mass (g/mol)

3000.00
25.00

2500.00 20.00

15.00
2000.00
10.00
1500.00
5.00

1000.00 0.00
1.00 2.00 3.00 4.00 5.00 6.00 7.00
Mixture Ratio

Figure 3.7.: Variation of specific impulse (at sea level), temperature (inside combus-
tion chamber) and molecular mass with mixture ratio for LOX/LH2.

4000.00 40.00
Speci�ic Impulse (m/s) & Temperature (K)

35.00
3500.00
30.00
Molecular Mass (g/mol)

3000.00
25.00

2500.00 20.00

15.00
2000.00
10.00
1500.00
5.00

1000.00 0.00
1.00 2.00 3.00 4.00 5.00 6.00 7.00
Mixture Ratio

Figure 3.9.: Variation of specific impulse (at sea level), temperature (inside combus-
tion chamber) and molecular mass with mixture ratio for LOX/RP-1.
42

4000.00

3500.00
Speci�ic Impulse (m/s)

3000.00

2500.00

2000.00
— LOX/LH2

1500.00 — LOX/LCH4

— LOX/RP -1
1000.00
1.00 2.00 3.00 4.00 5.00 6.00 7.00
Mixture Ratio

Figure 3.10.: Variation of specific impulse (at sea level) with mixture ratio for all
the three bipropellants considered.

3.0

2.5

2.0
Volume Ratio

1.5

1.0

0.5
1.00 2.93 0.98 0.62
(O/F) (5.50) (2.75) (2.25)
0.0
LOX LH2 LCH4 RP-1

Figure 3.11.: The relative volume of respective fuels per unit volume of the oxidizer
for the considered mass mixture ratios.
 43

3.5 SUMMARY

The choice of liquid propellants has phenomenal implications on the design of

a rocket engine. Of notable significance for a bipropellant combination are such
parameters as specific heat ratio, combustion chamber temperature and effective
molecular mass of the exhaust jet. Again, physical properties such as specific grav-
ity, specific heat, thermal conductivity, vapour pressure and viscosity, properties
which do not directly improve performance, are also of remarkable importance
during propellant selection.
We look in depth at one oxidizer, LOX, and three fuels, LH2, RP-1 and LCH4, for
our liquid rocket engine. In addition to the experimental prospect, our motiva-
tion to employ methalox over the other two bipropellants stems from its suite of
properties, both pros and cons, being right between those of hydrolox and kerolox
combinations.
Using CEA, we analyse the performance parameters of each mixture for a frozen
flow composition along with certain additional assumptions. The optimum mixture
ratio for each bipropellant mixture is thus preliminarily determined under the
following conditions: combustion chamber pressure 𝑃𝑐𝑐 = 1000 psi = 6894.757 kPa;
nozzle exit pressure 𝑃𝑒 = 1atm = 101.325kPa (sea level condition); nozzle area ratio
𝜖 = 40 (vacuum condition); frozen flow characteristic and infinite area combustor
option.
From their respective plots, it is observed that the mixture ratios of 2.75, 5.50
and 2.25 yield specific impulses of 296.88 s, 372.75 s and 285.19 s for methalox,
hydrolox and kerolox combinations. Corrected for a 4% over-performance, we
obtain specific impulses of 285.00 s, 357.84 s and 273.78 s for the bipropellant
combinations respectively. Figure 3.10 depicts the variation of specific impulse as a
function of mixture ratio for all three bipropellant mixtures considered. Figure 3.11
helps visualize the relative volume of each fuel with respect to that of liquid oxygen
at our considered mixture ratio: the volume of liquid methane is almost equal
to that of LOX for a mixture ratio of 2.75. Moving ahead, we carry the optimum
mixture ratio of LOX/LCH4 forward to the subsequent chapter(s).
4
Propellant Feed Systems

Transfer of liquid propellants from their tanks to the thrust chamber at the desired
flow rates and injection pressures requires a feed system (also called an engine
cycle). The selection of one feed system over another depends on a multitude of
factors: from propellant combination and performance specification of the engine
to mission requirement and operational envelope of the vehicle. Even the choice
of a sub-type of feed system—regulated vs blowdown pressurization system or
staged combustion vs gas generator turbopump mechanism—is influenced by
aspects such as complexity, controllability, reliability and reusability. Largely, a
propellant feed system can be classified as a pressure feed system or a pump feed
system; and depending on how power is derived to run the pump, the latter can
again be classified as a turbopump feed system or an electropump feed system.

4.1 TYPES OF FEED SYSTEMS

In a nutshell, a pressure feed system needs a difference in pressure between the

propellant tank and the combustion chamber in order to drive the propellant down-
stream. A pump feed system, contrarily, keeps the tank pressurization compara-
tively low and utilizes a pump near the thrust chamber inlet in order to maintain
adequate propellant flow rate and raise the fluid pressure to the desired level.

As a rule, liquid rocket propulsion systems involving fairly high thrust magnitudes
and/or long operation durations typically employ a pump feed system [12] [13].
Because the more powerful propulsion systems require relatively high combustion
chamber pressures to operate, accommodating propellant tanks that are capable
of withstanding such high tank pressurization, as with pressure feed systems,
46

results only in an inordinate inert mass penalty, which is hardly feasible for a flight
vehicle. Notwithstanding the foregoing, designing a rotodynamic pump along
with all associated components can permit little margin for uncertainties, and
employing a pump feed system can present tremendous operational complications
if not downright unforgiving circumstances.

Each propellant feed system comes with a suite of benefits and drawbacks. The
preference of one to another therefore entails a compromise between some critical
design parameters and some crucial performance metrics. The more common
types of feed systems, viz. gas-pressure feed systems, turbopump feed systems
and electrically-driven-pump feed systems, along with some major sub-types, have
been looked at below.

4.1.1 Pressure Feed Systems

Perhaps the simplest among all engine cycles, a pressure feed system uses a pres-
surant fluid to pressurize the propellant tank and expel the propellant into the
combustion chamber. The pressurant can either be a gas stored in a tank, a gas
generated through chemical reaction or even the evaporated propellant itself [13].
Mere actuation of the start/shutoff valves delivers the pressurant into the propel-
lant tanks and/or drives the propellants into the combustion chamber, making the
system extremely reliable [12].

There are two common types of pressure feed systems: regulated systems and blow-
down systems [12]. A regulated pressurization system includes a high-pressure
gas-supply tank and uses a pressure regulator in the feed line to maintain constant
propellant tank pressurization, and thus operates the thrust chamber essentially
at constant pressure and constant thrust. A blowdown pressurization system, on
the other hand, contains the pressurant gas within the propellant tank at an initial
maximum pressure and loses pressurization steadily with consumption of the
propellant, resulting in a gradual decrease in thrust over time.

Pressure feed systems have practical limitations on propellant tank pressurization

which in turn limits engine output and vehicle performance. These systems, how-
ever, offer superior performance to pump feed systems when the total impulse is
relatively low, the propellant mass is relatively low, the combustion chamber pres-
sure is relatively low, the engine thrust-to-weight ratio is relatively low, and when
repeated short-duration thrust pulses are the requirement [12]. A few renowned
engines that have employed the pressure feed system include the Service Propul-
sion System (SPS) engine on the Apollo Command/Service Module; the Descent
 47

Figure 4.1.: Schematic diagram of a generic blowdown-type pressure feed system.

Source: Wikimedia Commons

Propulsion System (DPS) and the Ascent Propulsion System (APS) engines on
the Apollo Lunar Module; and the Orbital Maneuvering System (OMS) engines
on the Space Shuttle orbiter [44] [28]. Figure 4.1 shows the schematic diagram of
a typical pressure feed system.

4.1.2 Turbopump Feed Systems

A turbopump feed system is, by definition, a combination of one or more gas
turbine(s) and rotodynamic pump(s). The turbine derives its energy from the
expansion of a high-enthalpy gaseous working fluid which imparts a rotational
impulse and/or reaction force on the rotor assembly [45]; this mechanical energy
provides the shaft power, with or without a gear transmission, for driving the
propellant pump at the desired speeds and torques. Turbines are classified into
radial-flow turbines and axial-flow turbines; the latter again are sub-classified
into reaction turbines and impulse turbines, which in turn have multiple types
such as single-rotor impulse turbines, velocity-compounded impulse turbines,
pressure-compounded impulse turbines, and the like [13].
The rocket propellant pump imparts kinetic energy from the rotor to the propellant
fluid flow and converts the energy into hydrostatic pressure necessary for the
thrust chamber operation [45]. Pumps principally are categorized as radial (or
centrifugal) pumps and axial pumps. The fundamental elements of a pump are
the rotor assembly (which usually comprises an inducer, an impeller, a shaft and
48

bearings) and the stator assembly (which consists of a casing with diffuser vanes,
a volute with discharge ducts, wear rings and shaft seals). The inducer, an axial-
flow rotor positioned at the pump inlet, provides the propellant fluid entering
the main pump section with an increased total pressure sufficient to permit non-
cavitating operation of the impeller. The impeller of a radial pump is a rotating
wheel with blades that discharges the flow radially outwards, whereas that of an
axial pump is a cylindrical rotor with multiple rows of rotating blades which keeps
the fluid flowing largely axial. While the impeller blades accelerate the fluid flow
by imparting kinetic energy to it, the diffuser vanes decelerate, or diffuse, the flow
to convert the fluid’s velocity head into pressure head. The volute collects and
redirects the pressurized fluid to the pump’s discharge outlet.

The engine cycle of a turbopump feed system defines the source of the energetic
drive fluid to power the turbopump assembly and the destination of the turbine
exhaust fluid after doing the work, and outlines propellant flow paths through ma-
jor engine components. Based on how the working fluid from the turbine exhaust
is handled, an engine cycle may be classified as open or closed [12]. In an open
cycle engine, the turbine working fluid is discharged into the diverging nozzle
section of the thrust chamber at a location downstream of the nozzle throat or
is ejected overboard altogether usually after undergoing expansion through its
own low-area-ratio nozzle. The turbine working fluid in a closed cycle engine,
however, is injected into the combustion chamber through the injector and under-
goes expansion through the entire pressure ratio across the thrust chamber nozzle.
While a closed cycle engine offers somewhat better specific impulse than its open
cycle counterpart, the performance advantage occurs at the expense of engineering
complexity from ducting the turbine exhaust fluid into the combustion chamber at
the flow rates and injection pressures necessary for the engine operation [12].

For any engine cycle, an increased flow rate of the working fluid into the turbine
increases the power output of the pump, therefore increasing flow rate of the
propellants into the combustion chamber, and consequently increasing the thrust
magnitude from the engine. Moreover, the working fluid comprises a less-than-
optimal mixture ratio—either very fuel-rich or very oxidizer-rich—so as to preclude
excessive fluid temperatures and to provide a more benign environment for the
turbine elements [12].

A key component of all high-total-impulse rocket engines, the turbopump is a

highly intricate piece of high-precision high-accuracy high-shaft-speed machinery.
High reliability, along with high performance and high efficiency, is therefore of the
utmost importance for the operation of the turbopump feed system. Selection of a
 49

(a) Gas Generator Cycle (b) Staged Combustion Cycle

(c) Expander Cycle

Figure 4.2.: Schematic diagrams of typical turbopump feed system cycles.

Source: Wikimedia Commons

specific turbopump configuration is consequently dependent on factors such as

desired flow rate, required discharge pressure, propellant combination, operation
duration, throttling range, among others.

The three most common types of turbopump feed systems include the gas generator
cycle, the staged combustion cycle and the expander cycle. Schematic diagrams of
the engine cycles are shown in figure 4.2.
50

Gas Generator Cycle

In a gas generator engine cycle, the working fluid constitutes a tiny percentage
of the total oxidizer and fuel flow quantities, and undergoes combustion inside
the gas-generator in order to generate the high-enthalpy gases for the turbine.
The engine derives little performance advantage from the turbine exhaust and
subsequently dispenses with the fluid, making the system an open cycle type. The
gas generator cycle operates with a high pressure ratio and a low flow rate in order
to minimize propellant loss and maximize engine efficiency [45].

Staged Combustion Cycle

In a staged combustion engine cycle, the working fluid comprises a small quantity
of one propellant type and a large quantity of the other, and, true to its name,
undergoes combustion in stages. The fluid is initially burned inside the preburner,
which produces the energetic fuel-rich or oxidizer-rich gas mixture for the turbop-
ump. The entire turbine exhaust fluid is injected into the combustion chamber,
along with the remainder propellant, and is eventually burned at the designed
bipropellant mixture ratio. A variant of this is the full flow staged combustion
engine cycle, where two working fluids, one fuel-rich and another oxidizer-rich, are
burned in separate preburners to drive separate turbopump assemblies; both the
exhaust fluids are subsequently ducted and injected into the combustion chamber
for combustion [42].

Expander Cycle
In an expander cycle, the working fluid consists usually of a cryogenic fuel, such
as liquid hydrogen or liquid methane, and flows through the cooling jackets of the
thrust chamber. The absorbed heat serves to vaporize and expand the propellant
flow and, in the process, produces the gases to power the turbine. The entire
exhaust fluid is then either ducted into the combustion chamber (allowing for a
high specific impulse) or bled overboard (allowing for a high thrust). Nonetheless,
the limited availability of heat for the turbine drive fluid, due to the square–cube
rule, limits the turbopump power, which in turn limits the thrust of an expander
cycle engine [42].

4.1.3 Electropump Feed Systems

Much like a turbopump feed system, an electropump (a portmanteau for electric-

pump or electrically-driven-pump) feed system uses propellant pumps in order
 51

Figure 4.3.: Schematic diagram of a generic electropump feed system.

Source: Wikimedia Commons

to drive the propellants into the combustion chamber. Yet unlike the former, the
latter swaps out the gas turbine and utilizes electric motors powered by battery
packs in order to run the propellant pumps. The inherent advantage of the system
lies in its operational ease owing to its reduced part count, from doing away with
the gas turbine and all associated components, and to its lack of complexity, from
not having to deal with the high-enthalpy turbine drive fluid.

Nonetheless, the mass penalty from the battery pack had historically been a major
bottleneck in the development of electric-pump-fed rocket engines: the battery
mass is dictated by both the electrical power and the electrical energy available
to drive the propellant pump throughout the burn time of the engine [30] [46].
However, with advancement in battery technologies and refinement in electric mo-
tors, adopting such a system has only recently become feasible. This system offers
a comparable performance to—or even outperforms—a gas generator cycle engine
for applications where the thrust-to-weight ratio is moderate and the operation
duration is long [47] [46].

To date, electric pump drive systems have successfully been tested on experimental
rocket engines [46]. And yet more remarkably, such an engine cycle has reli-
ably been employed on a small-lift orbital-class launch vehicle called Electron,
referred to in figure 1.7 [7]. Developed by Rocket Lab, Electron uses ten Rutherford
engines—nine sea-level versions on the booster stage and one vacuum-optimized
variant on the upper stage—to loft a payload mass of 300 kg to LEO or 200 kg to
52

500 km SSO. And the Rutherford is an electric-pump-fed engine, i.e., its propel-
lant pumps are driven by brushless DC motors which are powered from lithium
polymer batteries. Running on the LOX/RP-1 combination, the first-stage engine
variant produces a thrust of 5600 lbf (roughly 24 kN), delivers a specific impulse
of 311 s, and weighs in at 35 kg.
Under the right conditions, as stated in references [48] and [47], an electropump
feed system is a viable alternative to both the pressure feed system and the tur-
bopump feed system [30] [46]. Owing to its rather limited flight experience, this
project has an impetus to look into the development of such a system for our
experimental rocket engine running on liquid oxygen and liquid methane. The
schematic diagram of such a system is illustrated in figure 4.3.

4.2 FEED SYSTEM SELECTION CRITERIA

The selection of a propellant feed system for any application is a multifaceted

procedure entailing numerous conditions and considerations [12]. Of paramount
importance are the features that influence the mission design: bipropellant combi-
nation, mixture ratio, propellant physical properties, hardware inert mass, engine
reusability, system reliability, engine operating environment, and so on. Of even
more importance are factors dependent on the engine operation: thrust, burn du-
ration, combustion chamber pressure, nominal flow rate, feed line pressure losses,
and the like. But because the fundamental function of a feed system is to ensure
propellant flow into the combustion chamber at the required pressure, basing
the selection criteria of the system on combustion chamber pressure is perhaps a
prudent approach.
There currently exists a considerable overlap of applicability in the upper limit of
pressure feed systems and the lower limit of pump feed systems [13]. Combustion
chamber pressures as high as 400 psi and as low as 350 psi have been achieved
by employing pressure feed systems and pump feed systems respectively [15];
(1 psi = 6894.757293168 Pa). Yet, just because a feed system can operate near its
limits does not make it an ideal choice for a propulsion system. Rather, every feed
system type/sub-type has a range of combustion chamber pressures which it is
the most suited for [30] [45].
Prior to selecting a feed system, we need to ascertain the combustion chamber
pressure for our thrust chamber from its optimum pressure ratio condition. And
subsequently, we compare the performance parameters of the selected system with
other major types and examine its suitability for our application.
 53

4.2.1 Optimum Pressure Ratio

It has been established that an improved pressure ratio—by increasing the com-
bustion chamber pressure and/or decreasing the nozzle exit pressure—results in
an enhanced overall engine performance. Because the exit pressure is dictated
by the ambient pressure external to the thrust chamber and can only be modified
so much, the usual practice therefore is to raise the chamber pressure. With all
other parameters remaining constant, a higher combustion chamber pressure not
only increases the thrust coefficient (by reducing the external-to-chamber pressure
ratio) but also increases the characteristic velocity (by raising the adiabatic flame
temperature) [13]. In addition to directly improving specific impulse and overall
thrust, the higher combustion chamber pressure allows for a reduction in the thrust
chamber dimensions (for the same nozzle expansion area ratio) by permitting a
shorter nozzle divergent section with a smaller nozzle exit diameter, which saves
the vehicle inert mass and further enhances its performance [12].

Nonetheless, an unreasonably high combustion chamber pressure comes with

disadvantages as well [15] [12]. Firstly, the rate of heat transfer is approximately
proportional to the chamber pressure; the increased pressure, thus, raises heat
fluxes and thermal stresses on the thrust chamber, thereby necessitating elaborate
cooling procedures in order to keep the engine from overheating. Secondly, the
power required to drive a propellant pump (in a pump feed system) rises nearly lin-
early with the chamber pressure; increasing the pressure, thus, marginally reduces
the system performance, since the power for the pumps is obtained either from
electric motors running on battery packs or from turbines deriving energy from
propellant preburners/combustion chambers. And thirdly, as does the combustion
chamber pressure increase, so does the inert mass of the propulsion system from,
say, increased wall dimensions of ducts and valves or heavier components such as
regulators and manifolds. The optimum chamber pressure for a given application
is, therefore, a compromise between the higher performance of the propulsion
system and the operational complexity of such a system [15] [12].

In order to determine our combustion chamber pressure, we first scrutinize the

variation of thrust coefficient and specific impulse with changes in pressure ratio
across the nozzle. We preliminarily perform the calculations for a combustion
chamber pressure of 𝑃𝑐𝑐 = 1000 psi = 6894.757 kPa using CEARUN [39] for liq-
uid oxygen/liquid methane at a mass mixture ratio of 2.75 (assuming optimum
expansion, frozen composition flow condition and infinite area combustor option
as before). Additionally, we examine how nozzle expansion area ratio varies with
pressure ratio, since an area ratio greater than 20 is hardly reasonable for a thrust
54

chamber operating near sea level [12].

The results are plotted in figures 4.4 and 4.5 respectively. It is observed that a
pressure ratio between 75 and 150 delivers good enough performance values for
the above combustion chamber pressure [12] [13]: the thrust coefficient is neither
too much nor too little for sea level conditions, the specific impulse is about right
for atmospheric flight regimes, and the expansion area ratio of less than 15 permits
room for a cluster of thrust chambers on the booster stage.

4.2.2 Combustion Chamber Pressure

With pressure ratios across the nozzle narrowed down, we cannot help but realize
from figure 4.5 that the nozzle exit pressure drops below the ambient sea level
pressure even towards the lower pressure ratios. A substantial decrease in exit
pressure, to around 25 to 35% of local ambient pressure, does result in flow sepa-
ration instabilities (along with compression wave formation) within the divergent
nozzle section [13]; and during start and stop transients when the thrust is not
always axially symmetric, such instabilities may momentarily produce side loads
on the nozzle wall from uneven oscillations of the exhaust jet, which potentially
can damage the nozzle due to excessive forces and resonant frequencies [12]. Even
a slight decrease in exit pressure contributes to a noticeable reduction in thrust
and specific impulse from the thrust chamber.
In order to forestall flow separation without having to adopt an unreasonably
high chamber pressure, we need a minimum exit pressure that is nearly 40% of
the ambient pressure at sea level. We examine numerous combinations of exit
pressures (from 0.4 atm to 0.8 atm) and pressure ratios (from 75 to 150) to obtain
our chamber pressure (between 30 bar and 120 bar), and compare the values with
resultant area ratio (between 14 and 16). This helps us corroborate our engine
parameters with those of prior successful engines of similar configurations.
With a nozzle exit pressure of 𝑃𝑒 = 40 kPa, we tentatively opt for a pressure ratio
of 150, and settle on a combustion chamber pressure of 𝑃𝑐𝑐 = 6000 kPa (= 60 bar);
this yields an expansion area ratio of about 15. We finally run CEA again to verify
values of all the metrics calculated/considered thus far for sea level operation, and
summarize the results in table 4.1.
Having selected the chamber pressure helps us decide between the type/sub-type
of feed system best suited for our application. And the selected mixture ratio in
conjunction with propellant flow rates lets us design and/or determine the pumps’
components and/or parameters, if need be.
 55

2.40 4000

2.20
3500

2.00

Speci�ic Impulse (m/s)

3000
Thrust Coef�icient

1.80
2500
1.60

2000
1.40

1500
1.20

1.00 1000
1 10 100 1000 10000 100000
Pressure Ratio

Figure 4.4.: Variation of thrust coefficient and specific impulse with pressure ratio.
𝑃𝑐𝑐 = 1000 psi = 6894.757 kPa.

10000 1000.00

1000 100.00
Expansion Area Ratio

Exit Pressure (kPa)

100 10.00

10 1.00

1 0.10
1 10 100 1000 10000 100000
Pressure Ratio

Figure 4.5.: Variation of nozzle expansion area ratio and nozzle exit pressure with
pressure ratio. 𝑃𝑐𝑐 = 1000 psi = 6894.757 kPa.
56

Table 4.1.: Preliminary parameters for thrust chamber operation at sea level. [39]
Oxidizer Liquid oxygen
Fuel Liquid methane
Mixture ratio 𝑂/𝐹 2.75
Combustion chamber pressure 𝑃𝑐𝑐 6000 kPa

Specific impulse 𝐼𝑠 3068.30 m/s

Thrust coefficient 𝐶𝑓 1.6723
Characteristic velocity𝑐∗ 1834.80 m/s

Nozzle exit pressure 𝑃𝑒 40 kPa

Pressure ratio 𝑃𝑐𝑐 /𝑃𝑒 150
Nozzle expansion area ratio 𝜖 14.966

4.3 SUITABILITY OF FEED SYSTEMS

As specified earlier, the objective of our hypothetical mission is to deliver a light

payload into LEO using a two-stage rocket. The delta-v budget of the launch vehicle
is 9500 m/s, with the assumed delta-v distribution as 4500 m/s for the first stage
and 5000 m/s for the second stage. And delta-v—the most significant parameter
in the performance assessment of a launch vehicle—is immensely influenced by
the mass ratio of a rocket stage. In order to minimize inert mass and/or maximize
payload capacity, evaluation of the feed system mass serves as the foundation for
establishing its suitability for our application. The analysis for doing so follows
the framework laid out in references [48], [47] and [46].

4.3.1 Pressure Feed System

The practicality of a pressure feed system is considered first. This project intends
to operate our experimental engine not only on the upper stage but also on the
booster stage of our launch vehicle. And a pressure feed system powering the first
stage of a rocket is hitherto unheard of.
This is because, in order to preclude undesirable flow separation instabilities, a
thrust chamber operating from the sea level can afford a minimum nozzle exit
 57

pressure of around, or only slightly below, 0.4 atm (40.530 kPa). This means that,
for any decent values of thrust coefficient and specific impulse (for the LOX/LCH4
propellant combination), the propulsion system can have a minimum combustion
chamber pressure of roughly 75 times the nozzle exit pressure. This in turn implies
that, even with a minimum propellant tank pressurization of nearly 1.5 times the
combustion chamber pressure (to ensure adequate flow rate and injection pressure
into the chamber [13]), the tank pressurization can become immoderately high.
So, for a chamber pressure of, say, 3000 kPa, the required tank pressurization can
be upwards of 4500 kPa. Again, the pressurization required for pressurant tanks
can easily be manyfold more [48] [13].

Moreover, mass of the propellant tank (for a given wall material density) is deter-
mined by both its surface area (from its volume) and its wall thickness, which in
turn are dictated by the propellant mass and the Laplace’s law respectively [48].
Again, added to that mass is the mass of the pressurant tank. Now, for high-total-
impulse applications (like at liftoff, where the propellant mass is monumental and
at its maximum), an inordinately high mass from the propellant tank(s) exacer-
bates the propellant mass fraction and does not lend itself to good delta-v values
for the stage.

Merely through qualitative reasoning and without much quantitative inspection,

it can unequivocally be asserted that employing the pressure feed system for our
rocket engine would not result in a sound solution. Therefore, we disregard this
engine cycle and consider the pump feed systems for our application.

4.3.2 Turbopump Feed System

The feasibility of a turbopump feed system is examined here. Turbopump cycle

engines are capable of accomplishing the highest figures in terms of thrust, thrust-
to-weight ratio or combustion chamber pressure, as is evident from engines such as
the RD-170 (highest thrust engine), the F-1 (highest thrust single-chamber engine),
the Merlin (highest thrust-to-weight ratio engine) and the Raptor (highest com-
bustion chamber pressure engine). For more moderate performance requirements,
however, turbopump-fed engines are oftentimes bested by the other engine cycles.
Regardless, we follow the convention here and predicate the gas generator cycle as
the benchmark for good performance.

For the evaluation of overall mass, only masses of the major components are
considered while those of the minor and common components are neglected. It is
to be noted that helium is used as the pressurant fluid here for pressurizing both
58

oxidizer and fuel tanks and preventing cavitation in the propellant pumps.

The mass of the turbopump feed system (TPFS) is defined as the combined masses
of oxidizer (o), fuel (f), pressurant (prs), oxidizer tank (o.t), fuel tank (f.t), pres-
surant tank (prs.t), turbopump assembly (tp) and gas generator (gg).

𝑚𝑇𝑃𝐹𝑆 = 𝑚𝑜 + 𝑚𝑓 + 𝑚𝑝𝑟𝑠 + 𝑚𝑜.𝑡 + 𝑚𝑓 .𝑡 + 𝑚𝑝𝑟𝑠.𝑡 (4.1)

+ 𝑚𝑡𝑝 + 𝑚𝑔𝑔

Turbopump Assembly Mass

Assuming no mechanical losses between the turbine and the pumps, the turbop-
ump power equals the power generated from the turbine or the power consumed
by the pumps.
𝑝𝑡𝑝 = 𝑝𝑡𝑢 = 𝑝𝑝𝑢 = 𝑝𝑜.𝑝 + 𝑝𝑓 .𝑝 (4.2)

The turbopump power is proportional to the propellant mass flow rate times the
head rise in both the pumps [46].

𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝

𝑝𝑡𝑝 = + (4.3)
𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓

where, 𝑚̇ 𝑜.𝑝 , 𝑚̇ 𝑓 .𝑝 , 𝛥𝑃𝑜.𝑝 , 𝛥𝑃𝑓 .𝑝 , 𝜂𝑜.𝑝 and 𝜂𝑓 .𝑝 represent mass flow rates, head rises
and efficiencies for both the propellant pumps respectively.

Because turbopump mass is related to its power via its power density (power per
unit mass), the turbopump mass can be expressed in terms of mass flow rate and
pump head rise as [46]

1 1 1
𝑚𝑡𝑝 = 𝑝𝑡𝑝 = 𝑝𝑡𝑢 = (𝑝 + 𝑝𝑓 .𝑝 )
𝛿𝑡𝑝 𝛿𝑡𝑝 𝛿𝑡𝑝 𝑜.𝑝
1 ⎛ 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝 ⎞
= ⎜ + ⎟ (4.4)
𝛿𝑡𝑝 ⎝ 𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓 ⎠

Gas Generator Mass

Mass of the gas generator is proportional to its volume and its wall thickness.

The gas generator volume can be determined by the stay time 𝑡𝑠 method as

𝑚̇ 𝑔𝑔 𝑡𝑠
𝑉𝑔𝑔 = (4.5)
𝜌𝑔𝑔
 59

where, 𝑚̇ 𝑔𝑔 and 𝜌𝑔𝑔 are the mass flow rate and the density of the gas generator
working fluid respectively.

And assuming a spherical shape, the gas generator mass can be expressed as [46]

3 𝜌𝑔𝑔.𝑚 𝑚̇ 𝑔𝑔 𝑡𝑠
𝑚𝑔𝑔 = 𝑃 𝜅 (4.6)
2 𝜎𝑔𝑔.𝑚 𝜌𝑔𝑔 𝑔𝑔 𝑔𝑔

where, 𝜌𝑔𝑔.𝑚 , 𝜎𝑔𝑔.𝑚 , 𝑃𝑔𝑔 and 𝜅𝑔𝑔 represent material density, material yield strength,
pressure and structural safety factor of the gas generator respectively.

Now, for a single stage supersonic impulse turbine undergoing an adiabatic process,
the turbine power is given as
1−𝑘𝑔𝑔
( )
⎛ 𝑃𝑡𝑢.𝑖𝑛𝑙𝑒𝑡 𝑘𝑔𝑔 ⎞
𝑝𝑡𝑢 = 𝑚̇ 𝑔𝑔 𝑐𝑃.𝑔𝑔 𝜂𝑡𝑢 𝑇𝑡𝑢.𝑖𝑛𝑙𝑒𝑡 ⎜
⎜
⎜
⎜ 1 − ( ) ⎟
⎟
⎟
⎟ (4.7)
𝑃𝑡𝑢.𝑜𝑢𝑡𝑙𝑒𝑡
⎝ ⎠
where, 𝑐𝑃.𝑔𝑔 , 𝑘𝑔𝑔 , 𝑇𝑡𝑢 and 𝑃𝑡𝑢 are specific heat of constant pressure, specific heat
ratio, turbine temperature and turbine pressure respectively. The subscript gg here
represents the gas generator working fluid.

Propellant Masses
The mass of each propellant consumed for propulsion is proportional to the mass
flow rate through its pump and the burn time 𝑡𝑏 of the engine [46].

𝑚𝑜 = 𝑚̇ 𝑜.𝑝 𝑡𝑏 𝜅𝑝 (4.8a)
𝑚𝑓 = 𝑚̇ 𝑓 .𝑝 𝑡𝑏 𝜅𝑝 (4.8b)

where, 𝜅𝑝 represents the margin of residual propellant mass trapped inside pipes,
valves, fittings, filters, pumps, etc. after engine operation.

The total propellant mass therefore is given as

𝑚𝑝 = 𝑚 𝑜 + 𝑚 𝑓
= (𝑚̇ 𝑜.𝑝 + 𝑚̇ 𝑓 .𝑝 ) 𝑡𝑏 𝜅𝑝 (4.9)

Propellant Tank Masses

Mass of a propellant tank can be expressed as the product of its surface area, its
wall thickness and its material density.
60

For the sake of simplicity, it is assumed that each propellant tank is spherical.
Additionally, it is assumed that the pressurant tank is immersed and integrated
inside the oxidizer tank, as is common practice for propellant tank arrangements.

Volumes of the oxidizer tank and the fuel tank are, therefore, defined as

𝑉𝑜.𝑡 = 𝑉𝑜 𝜅𝑢 + 𝑉𝑝𝑟𝑠.𝑡 (4.10a)

𝑉𝑓 .𝑡 = 𝑉𝑓 𝜅𝑢 (4.10b)

where, 𝜅𝑢 is a factor representing the ullage volume within the tank.

Thus, radius and surface area of the oxidizer tank, respectively, can be written in
terms of its volume as
1/3
3
𝑟𝑜.𝑡 = ( 𝑉𝑜.𝑡 ) (4.11a)
4𝜋
2/3
3
𝑆𝑜.𝑡 = 4𝜋 ( 𝑉 ) (4.11b)
4𝜋 𝑜.𝑡

Now, the maximum pressure exerted on the lower and the upper hemispheres
of the propellant tank is different due to differences in fluid head from different
loading conditions, and can be expressed as [46]

𝑃𝑜.𝑡.𝑙𝑜𝑤𝑒𝑟 = 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 + 𝜌𝑜 𝑔𝑚𝑎𝑥 ℎ𝑜.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 (4.12a)

𝑃𝑜.𝑡.𝑢𝑝𝑝𝑒𝑟 = 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 + 𝜌𝑜 𝑔𝑚𝑎𝑥 (ℎ𝑜.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 − 𝑟𝑜.𝑡 ) (4.12b)

where, 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 , 𝑔𝑚𝑎𝑥 and ℎ𝑜.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 represent internal tank pressure, maximum
vehicle acceleration and initial fluid head respectively. Consequently, the thickness
of the lower and the upper tank hemispheres is different, and can be calculated
from the thin wall theory.

Mass of the oxidizer tank can, therefore, be expressed as

3 𝜌𝑜.𝑡.𝑚
𝑚𝑜.𝑡 = 𝜅 𝑉 (𝑃 + 𝑃𝑜.𝑡.𝑢𝑝𝑝𝑒𝑟 ) (4.13)
4 𝜎𝑜.𝑡.𝑚 𝑡 𝑜.𝑡 𝑜.𝑡.𝑙𝑜𝑤𝑒𝑟

Mass of the fuel tank can, likewise, be expressed as

3 𝜌𝑓 .𝑡.𝑚
𝑚𝑓 .𝑡 = 𝜅 𝑉 (𝑃 + 𝑃𝑓 .𝑡.𝑢𝑝𝑝𝑒𝑟 ) (4.14)
4 𝜎𝑓 .𝑡.𝑚 𝑡 𝑓 .𝑡 𝑓 .𝑡.𝑙𝑜𝑤𝑒𝑟
 61

Pressurant Mass
In order to suppress cavitation within the propellant pumps, it is necessary to pres-
surize both the propellant tanks up to the required pump inlet pressure. Mass of
the pressurant fluid can be determined assuming ideal gas and adiabatic expansion
properties.
𝑘𝑝𝑟𝑠 ⎛
⎜ 𝑃𝑉 ⎞
⎟
𝑚𝑝𝑟𝑠 = 𝜅𝑝𝑟𝑠 ⎜
⎜ ⎟
⎟ (4.15)
𝑅𝑝𝑟𝑠 𝑇𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎜
⎜ 𝑃𝑝𝑟𝑠.𝑓 𝑖𝑛𝑎𝑙 ⎟ ⎟
1− 𝑃
⎝ 𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎠

where, 𝜅𝑝𝑟𝑠 , 𝑘𝑝𝑟𝑠 , 𝑅𝑝𝑟𝑠 , 𝑇𝑝𝑟𝑠 and 𝑃𝑝𝑟𝑠 represent mass margin, specific heat ratio, gas
constant, temperature and pressure of the pressurant fluid respectively.

Additionally, 𝑃 and 𝑉 above represent propellant tank internal pressure and pro-
pellant volume, expanding which for both the oxidizer and the fuel yields [46]

𝑘𝑝𝑟𝑠 ⎛
⎜ 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑉𝑜 + 𝑃𝑓 .𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑉𝑓 ⎞
⎟
𝑚𝑝𝑟𝑠 = 𝜅𝑝𝑟𝑠 𝜅𝑢 ⎜
⎜ ⎟
⎟ (4.16)
𝑅𝑝𝑟𝑠 𝑇𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎜
⎜ 𝑃 𝑝𝑟𝑠.𝑓 𝑖𝑛𝑎𝑙
⎟
⎟
1− 𝑃
⎝ 𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎠

Pressurant Tank Mass

Volume of the pressurant tank is given as

𝑇𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙
𝑉𝑝𝑟𝑠.𝑡 = 𝑚𝑝𝑟𝑠 𝑅𝑝𝑟𝑠 (4.17)
𝑃𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙

And assuming a spherical pressurant tank, mass of the tank can thus be expressed
as [46]

3 𝜌𝑝𝑟𝑠.𝑡.𝑚 ⎛
⎜ 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑉𝑜 + 𝑃𝑓 .𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 𝑉𝑓 ⎞
⎟
𝑚𝑝𝑟𝑠.𝑡 = 𝜅𝑝𝑟𝑠.𝑡 𝜅𝑝𝑟𝑠 𝜅𝑢 𝑘𝑝𝑟𝑠 ⎜
⎜
⎜
⎟
⎟
⎟ (4.18)
2 𝜎𝑝𝑟𝑠.𝑡.𝑚 ⎜ 𝑃𝑝𝑟𝑠.𝑓 𝑖𝑛𝑎𝑙 ⎟
1− 𝑃
⎝ 𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎠

where, 𝜅𝑝𝑟𝑠.𝑡 , 𝜌𝑝𝑟𝑠.𝑡.𝑚 and 𝜎𝑝𝑟𝑠.𝑡.𝑚 represent mass margin, material density and
material yield strength of the pressurant tank respectively.

Turbopump Feed System Mass

Total mass of the turbopump feed system, therefore, may be obtained from sub-
stituting the individual mass expressions above in the equation 4.1. The total
turbopump feed system mass, expressed here in terms of pump power and burn-
62

ing time after rearranging and recombining the above expressions, is as follows

1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝

𝑚𝑇𝑃𝐹𝑆 = ⎛
⎜ + 𝑋1 𝑃𝑔𝑔 ⎞
⎟⎛⎜ + ⎞
⎟ (4.19)
𝛿
⎝ 𝑡𝑝 𝜂
⎠ ⎝ 𝑜.𝑝 𝑜 𝜌 𝜂 𝜌
𝑓 .𝑝 𝑓 ⎠
𝑚̇ 𝑜.𝑝 𝑡𝑏
+ (𝑋𝑜 + 𝜅𝑝 𝜌𝑜 + 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 (𝑋2 + 𝑋𝑜 𝑋3 ))
𝜌𝑜
𝑚̇ 𝑓 .𝑝 𝑡𝑏
+ (𝑋𝑓 + 𝜅𝑝 𝜌𝑓 + 𝑃𝑓 .𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 (𝑋2 + 𝑋𝑓 𝑋3 ))
𝜌𝑓

where, the new variables can be expanded as

3 𝜌𝑜.𝑡.𝑚
𝑋𝑜 = 𝜅𝑝 𝜅𝑢 𝜅𝑡 (𝑃 + 𝑃𝑜.𝑡.𝑢𝑝𝑝𝑒𝑟 ) (4.20)
4 𝜎𝑜.𝑡.𝑚 𝑜.𝑡.𝑙𝑜𝑤𝑒𝑟
3 𝜌𝑓 .𝑡.𝑚
𝑋𝑓 = 𝜅𝑝 𝜅𝑢 𝜅𝑡 (𝑃 + 𝑃𝑓 .𝑡.𝑢𝑝𝑝𝑒𝑟 ) (4.21)
4 𝜎𝑓 .𝑡.𝑚 𝑓 .𝑡.𝑙𝑜𝑤𝑒𝑟
1−𝑘𝑔𝑔 −1
( )
3 𝜌𝑔𝑔.𝑚 𝜅𝑔𝑔 𝑡𝑠 ⎡ ⎛
⎜ 𝑃𝑡𝑢.𝑖𝑛𝑙𝑒𝑡 𝑘𝑔𝑔 ⎞
⎟⎤
⎢ ⎜ ⎟
𝑋1 = 𝑐 𝜂 𝑇 ⎜1 − (
2 𝜎𝑔𝑔.𝑚 𝜌𝑔𝑔 ⎢ 𝑃.𝑔𝑔 𝑡𝑢 𝑡𝑢.𝑖𝑛𝑙𝑒𝑡 ⎜ 𝑃𝑡𝑢.𝑜𝑢𝑡𝑙𝑒𝑡
) ⎟⎥
⎟⎥ (4.22)
⎣ ⎝ ⎠⎦
⎛
⎜ 𝑘𝑝𝑟𝑠 ⎞
⎟ 1 3 𝜌𝑝𝑟𝑠.𝑡.𝑚
𝑋2 = 𝜅𝑝 𝜅𝑢 𝜅𝑝𝑟𝑠 ⎜ ⎜
⎜ 𝑃𝑝𝑟𝑠.𝑓 𝑖𝑛𝑎𝑙 ⎟
⎟
⎟ ⎛
⎜ + 𝜅 ⎞
⎟ (4.23)
⎜ ⎟ ⎝ 𝑅𝑝𝑟𝑠 𝑇𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 2 𝜎𝑝𝑟𝑠.𝑡.𝑚 𝑝𝑟𝑠.𝑡 ⎠
1− 𝑃
⎝ 𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 ⎠

𝜅𝑝𝑟𝑠 𝑘𝑝𝑟𝑠
𝑋3 = (4.24)
𝑃𝑝𝑟𝑠.𝑖𝑛𝑖𝑡𝑖𝑎𝑙 − 𝑃𝑝𝑟𝑠.𝑓 𝑖𝑛𝑎𝑙

4.3.3 Electropump Feed System

The viability of an electropump feed system is scrutinized now. As mentioned

before, the lack of a breakthrough in battery technology over the recent decades
has perhaps been the greatest impediment in the widespread adoption of electric
vehicles—from rockets and airplanes to cars and boats [49] [50]. Yet, as is estab-
lished in references [48], [47] and [46], and as is evident from the launch vehicle
Electron [7], electropump feed systems are indeed practicable under the proper
circumstances and/or requirements of engine operation. Our objective, therefore,
is to optimize the performance parameters from the electropump feed system.

Among the most crucial factors dictating the mass of a battery pack is its critical
density in terms of both power and energy. Both power and energy are in turn
influenced by the required discharge time of the battery pack. Again, because
the efficiency of a battery cell is rather sensitive to its environmental temperature,
 63

the use of a coolant (usually the fuel) to keep the battery pack within the desired
temperature range becomes a necessity.

For the mass of the electropump feed system (EPFS), the masses of oxidizer (o),
fuel (f), pressurant (prs), oxidizer tank (o.t), fuel tank (f.t), pressurant tank (prs.t),
oxidizer pump (o.p), fuel pump (f.p), electric motor (mot), inverter (inv) and
battery pack (bat) are considered.

𝑚𝐸𝑃𝐹𝑆 = 𝑚𝑜 + 𝑚𝑓 + 𝑚𝑝𝑟𝑠 + 𝑚𝑜.𝑡 + 𝑚𝑓 .𝑡 + 𝑚𝑝𝑟𝑠.𝑡 (4.25)

+ 𝑚𝑜.𝑝 + 𝑚𝑓 .𝑝 + 𝑚𝑚𝑜𝑡 + 𝑚𝑖𝑛𝑣 + 𝑚𝑏𝑎𝑡

Propellant Pump Masses

As described for the turbopump assembly mass, mass of each propellant pump is
related to its power via its power density, and can be expressed in terms of mass
flow rate and pump head rise [46].

𝑝𝑜.𝑝 1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝

𝑚𝑜.𝑝 = = (4.26a)
𝛿𝑜.𝑝 𝛿𝑜.𝑝 𝜂𝑜.𝑝 𝜌𝑜
𝑝𝑓 .𝑝 1 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝
𝑚𝑓 .𝑝 = = (4.26b)
𝛿𝑓 .𝑝 𝛿𝑓 .𝑝 𝜂𝑓 .𝑝 𝜌𝑓

Electric Motor Mass

Assuming no mechanical losses between the motor and the pumps, the output
power from the motor equals the power consumed by both the propellant pumps.
For a motor mechanical output power density 𝛿𝑚𝑜𝑡 , mass of the motor is given as

1 1
𝑚𝑚𝑜𝑡 = 𝑝𝑚𝑜𝑡.𝑜𝑢𝑡𝑝𝑢𝑡 = (𝑝𝑜.𝑝 + 𝑝𝑓 .𝑝 ) (4.27)
𝛿𝑚𝑜𝑡 𝛿𝑚𝑜𝑡

Inverter Mass
Now, the efficiency of the motor is defined as the ratio of motor mechanical output
power to inverter electrical output power [46]. Likewise, the inverter efficiency is
the ratio of its electrical output power to its electrical input power [46].
𝑝𝑚𝑜𝑡.𝑜𝑢𝑡𝑝𝑢𝑡
𝜂𝑚𝑜𝑡 = (4.28a)
𝑝𝑖𝑛𝑣.𝑜𝑢𝑡𝑝𝑢𝑡
𝑝𝑖𝑛𝑣.𝑜𝑢𝑡𝑝𝑢𝑡
𝜂𝑖𝑛𝑣 = (4.28b)
𝑝𝑖𝑛𝑣.𝑖𝑛𝑝𝑢𝑡
64

Mass of the inverter is therefore given as [46]

1 1
𝑚𝑖𝑛𝑣 = 𝑝𝑖𝑛𝑣.𝑜𝑢𝑡𝑝𝑢𝑡 = 𝑝
𝛿𝑖𝑛𝑣 𝛿𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑚𝑜𝑡.𝑜𝑢𝑡𝑝𝑢𝑡
1
= (𝑝 + 𝑝𝑓 .𝑝 ) (4.29)
𝛿𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑜.𝑝

Battery Pack Mass

A battery is constrained in both its power and its energy capacities. Therefore,
depending on the requirement for either power or energy, either power density or
energy density dictates the mass of a battery pack.

The electrical power of the battery pack required to run the propellant pumps is
given as [46]

1
𝑝𝑏𝑎𝑡 = 𝑝𝑖𝑛𝑣.𝑖𝑛𝑝𝑢𝑡 = 𝑝
𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑚𝑜𝑡.𝑜𝑢𝑡𝑝𝑢𝑡
1
= (𝑝 + 𝑝𝑓 .𝑝 ) (4.30)
𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑜.𝑝

Likewise, the electrical energy required to run the pumps for the entire engine
operation duration 𝑡𝑏 is given as [46]

1
𝑒𝑏𝑎𝑡 = 𝑡𝑏 𝑝𝑏𝑎𝑡
𝜂𝑏𝑎𝑡
1
= 𝑡𝑏 (𝑝𝑜.𝑝 + 𝑝𝑓 .𝑝 ) (4.31)
𝜂𝑏𝑎𝑡 𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡

where, 𝜂𝑏𝑎𝑡 is the energy efficiency of the battery.

Upon introducing power density 𝛿𝑏𝑎𝑡.𝑝 and energy density 𝛿𝑏𝑎𝑡.𝑒 of the battery, the
power-constrained mass and the energy-constrained mass of the battery pack can,
respectively, be expressed as [46]

1 1
𝑚𝑏𝑎𝑡.𝑝 = 𝜅𝑏𝑎𝑡 𝑝𝑏𝑎𝑡 = 𝜅 (𝑝 + 𝑝𝑓 .𝑝 ) (4.32a)
𝛿𝑏𝑎𝑡.𝑝 𝛿𝑏𝑎𝑡.𝑝 𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑏𝑎𝑡 𝑜.𝑝
1 1
𝑚𝑏𝑎𝑡.𝑒 = 𝜅 𝑒 = 𝜅 𝑡 (𝑝 + 𝑝𝑓 .𝑝 ) (4.32b)
𝛿𝑏𝑎𝑡.𝑒 𝑏𝑎𝑡 𝑏𝑎𝑡 𝛿𝑏𝑎𝑡.𝑒 𝜂𝑏𝑎𝑡 𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝑏𝑎𝑡 𝑏 𝑜.𝑝

In the equations 4.32, 𝜅𝑏𝑎𝑡 represents a mass margin adopted to account for the
necessary sub-component masses of the battery pack, such as battery management
 65

systems, wires, casing, etc.

Mass of the battery pack is therefore given as

𝑚𝑏𝑎𝑡 = 𝑀𝐴𝑋 [𝑚𝑏𝑎𝑡.𝑝 , 𝑚𝑏𝑎𝑡.𝑒 ]

1 1 𝑡𝑏
= 𝜅𝑏𝑎𝑡 (𝑝𝑜.𝑝 + 𝑝𝑓 .𝑝 ) 𝑀𝐴𝑋 ⎡
⎢ , ⎤
⎥ (4.33)
𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 ⎣ 𝛿𝑏𝑎𝑡.𝑝 𝛿𝑏𝑎𝑡.𝑒 𝜂𝑏𝑎𝑡 ⎦

Propellant Masses
Now, because no quantity of propellant from the pump outlet is diverted out into
the gas generator before being exhausted overboard, the revised oxidizer and fuel
masses for the electropump cycle can be given as

𝑚𝑜 = (𝑚̇ 𝑜.𝑝 − 𝑚̇ 𝑜.𝑔𝑔 ) 𝑡𝑏 𝜅𝑝 (4.34a)

𝑚𝑓 = (𝑚̇ 𝑓 .𝑝 − 𝑚̇ 𝑓 .𝑔𝑔 ) 𝑡𝑏 𝜅𝑝 (4.34b)

where, 𝑚̇ 𝑜.𝑔𝑔 and 𝑚̇ 𝑓 .𝑔𝑔 represent oxidizer and fuel mass flow rates through the
gas generator, and can be computed using 𝑚̇ 𝑔𝑔 and the mixture ratio (𝑂/𝐹)𝑔𝑔 of
the gas generator working fluid. The gas generator mass flow rate typically varies
between 1 to 7% of the total propellant flow [12] [13].

Electropump Feed System Mass

For masses of the other four terms in the equation 4.25, the same expressions from
the turbopump feed system mass can be applied. Therefore, total mass of the
electropump feed system, expressed in terms of pump power and burning time as
before, is as follows

1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 1 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝

𝑚𝐸𝑃𝐹𝑆 = ⎛
⎜ + 𝑋4 ⎞
⎟⎛⎜ ⎞
⎟+⎛ ⎜ + 𝑋4 ⎞
⎟⎛⎜ ⎞
⎟ (4.35)
𝛿
⎝ 𝑜.𝑝 𝜂 𝜌
⎠ ⎝ 𝑜.𝑝 𝑜 ⎠ ⎝ 𝑓 .𝑝 𝛿 𝜂
⎠ ⎝ 𝑓 .𝑝 𝑓 ⎠ 𝜌
(𝑚̇ 𝑜.𝑝 − 𝑚̇ 𝑜.𝑔𝑔 ) 𝑡𝑏
+ (𝑋𝑜 + 𝜅𝑝 𝜌𝑜 + 𝑃𝑜.𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 (𝑋2 + 𝑋𝑜 𝑋3 ))
𝜌𝑜
(𝑚̇ 𝑓 .𝑝 − 𝑚̇ 𝑓 .𝑔𝑔 ) 𝑡𝑏
+ (𝑋𝑓 + 𝜅𝑝 𝜌𝑓 + 𝑃𝑓 .𝑡.𝑖𝑛𝑡𝑒𝑟𝑛𝑎𝑙 (𝑋2 + 𝑋𝑓 𝑋3 ))
𝜌𝑓

where, the new variable can be expanded as

1 1 1 1 𝑡𝑏
𝑋4 = + + 𝜅𝑏𝑎𝑡 𝑀𝐴𝑋 ⎡
⎢ , ⎤
⎥ (4.36)
𝛿𝑚𝑜𝑡 𝛿𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 ⎣ 𝛿𝑏𝑎𝑡.𝑝 𝛿𝑏𝑎𝑡.𝑒 𝜂𝑏𝑎𝑡 ⎦
66

4.3.4 Analysis of the Feed System Mass

For a comparative evaluation of both the feed system masses, we neglect every
common element between the systems and enumerate specifications for each
parameter above. The two total mass expressions stated in their respective exclusive
terms are presented again below.

𝑚𝑇𝑃𝐹𝑆 = 𝑚𝑜 + 𝑚𝑓 + 𝑚𝑡𝑝 + 𝑚𝑔𝑔 (4.37)

= 𝑚̇ 𝑜.𝑝 𝑡𝑏 𝜅𝑝
+ 𝑚̇ 𝑓 .𝑝 𝑡𝑏 𝜅𝑝
1 ⎛ 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝 ⎞
+ ⎜ + ⎟
𝛿𝑡𝑝 ⎝ 𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓 ⎠
3 𝜌𝑔𝑔.𝑚 𝑚̇ 𝑔𝑔 𝑡𝑠
+ 𝑃 𝜅
2 𝜎𝑔𝑔.𝑚 𝜌𝑔𝑔 𝑔𝑔 𝑔𝑔

𝑚𝐸𝑃𝐹𝑆 = 𝑚𝑜 + 𝑚𝑓 + 𝑚𝑜.𝑝 + 𝑚𝑓 .𝑝 + 𝑚𝑚𝑜𝑡 + 𝑚𝑖𝑛𝑣 + 𝑚𝑏𝑎𝑡 (4.38)

(𝑂/𝐹)𝑔𝑔
=⎛⎜𝑚̇ 𝑜.𝑝 − 𝑚̇ 𝑔𝑔 ⎞
⎟ 𝑡𝑏 𝜅𝑝
⎝ (𝑂/𝐹) 𝑔𝑔 + 1 ⎠
1
+⎛⎜𝑚̇ 𝑓 .𝑝 − 𝑚̇ 𝑔𝑔 ⎞
⎟ 𝑡𝑏 𝜅𝑝
⎝ (𝑂/𝐹)𝑔𝑔 + 1 ⎠
1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝
+
𝛿𝑜.𝑝 𝜂𝑜.𝑝 𝜌𝑜
1 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝
+
𝛿𝑓 .𝑝
𝜂𝑓 .𝑝 𝜌𝑓
1 ⎛ 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝 ⎞
+ ⎜ + ⎟
𝛿𝑚𝑜𝑡 ⎝ 𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓 ⎠
1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝
+ ⎛
⎜ + ⎞
⎟
𝛿𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 ⎝ 𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓 ⎠
1 𝑚̇ 𝑜.𝑝 𝛥𝑃𝑜.𝑝 𝑚̇ 𝑓 .𝑝 𝛥𝑃𝑓 .𝑝 1 𝑡𝑏
+ 𝜅𝑏𝑎𝑡 ⎛
⎜ + ⎞
⎟ 𝑀𝐴𝑋 ⎡
⎢ , ⎤
⎥
𝜂𝑖𝑛𝑣 𝜂𝑚𝑜𝑡 ⎝ 𝜂𝑜.𝑝 𝜌𝑜 𝜂𝑓 .𝑝 𝜌𝑓 ⎠ ⎣ 𝛿𝑏𝑎𝑡.𝑝 𝛿𝑏𝑎𝑡.𝑒 𝜂𝑏𝑎𝑡 ⎦

From the values laid out in references [48], [47] and [46], we obtain power den-
sities for the turbopump, the oxidizer pump and the fuel pump as 12.50 kW/kg,
20 kW/kg and 15 kW/kg respectively. The efficiencies of the oxidizer pump and
the fuel pump are adopted as 0.66 and 0.60 respectively, and both their pressure
increments are assumed to be 1.25 𝑃𝑐𝑐 = 7500 kPa and 1.50 𝑃𝑐𝑐 = 9000 kPa respec-
tively; the additional pressure is necessary to account for any pressure drop within
 67

the injector assembly as well as, in case of fuel, within the thrust chamber heat
exchanger.

Now, power density and efficiency of the electric motor are taken as 5.50 kW/kg
and 0.95 respectively. The electric motor used is a permanent magnet synchronous
motor (PMSM) with a rated power of around 340 kW and a rotational speed of
about 45000 rpm. Likewise, power density and efficiency of the inverter are taken
as 60 kW/kg and 0.85 respectively. The inverter properties are assumed the same
as in the study [47].

For the battery pack, the structural mass margin is assumed to be 25%. Desirable
characteristics of the battery cell include both higher power densities and higher
energy densities for a lighter battery pack. A lithium polymer cell with an efficiency
of 0.90, a power density of 6.50 kW/kg and an energy density of 300 Wh/kg is
therefore adopted.

For the gas generator subassemblies, we use Inconel 625 [51] as the construction
material, which has a yield strength of 500 MPa and a density of 8440 kg/m3 , and
opt for a structural safety factor of 2.50. A mass mixture ratio of 0.333 is selected for
the gas generator working fluid, which corresponds to a density of 12.624 kg/m3 ;
the fuel rich mixture ensures a low combustion temperature of 931.41 K and averts
any erosion of the turbine blades [39]. Referring to [12] and [13], we adopt a
working fluid mass flow rate of 0.0075 times the total propellant flow rate through
the thrust chamber.

The mass margin for residual propellant after the thrust chamber operation is
rationally assumed to be 5%. The pressure within the gas generator is considered
same as that within the combustion chamber, while the stay time of the working
fluid is assumed to be 0.01 s. The operation duration for the rocket engine is varied
from 0 s to 1000 s so as to analyse the mass dependence of both the feed systems
on engine burn time.

All pertinent specifications considered thus far are summarized in table 4.2, with
the attendant plots from our calculations presented further below. From figure
4.6, we can observe that the mass of the feed system (i.e., the entire pump feed
system mass without the thrust chamber propellant mass) varies, for the most part,
linearly with the engine burn duration. This is because, for the turbopump system,
the gas generator working fluid mass is directly proportional to the burn time and
outweighs the combined masses of other feed system components. In case of the
electropump system, however, the battery pack mass dominates the masses of all
other feed system components and is dictated by its critical parameter in terms of
68

Table 4.2.: Specifications used for the estimation of feed system mass.
Oxidizer Pump
Power density 𝛿𝑜.𝑝 20.00 kW/kg
Efficiency 𝜂𝑜.𝑝 0.66
Pressure increment 𝛥𝑃𝑜.𝑝 7500 kPa
Fuel Pump
Power density 𝛿𝑓 .𝑝 15.00 kW/kg
Efficiency 𝜂𝑓 .𝑝 0.60
Pressure increment 𝛥𝑃𝑓 .𝑝 9000 kPa
Electric Motor
Power density 𝛿𝑚𝑜𝑡 5.50 kW/kg
Efficiency 𝜂𝑚𝑜𝑡 0.95
Inverter
Power density 𝛿𝑖𝑛𝑣 60.00 kW/kg
Efficiency 𝜂𝑖𝑛𝑣 0.85
Battery Pack
Power density 𝛿𝑏𝑎𝑡.𝑝 6.50 kW/kg
Energy density 𝛿𝑏𝑎𝑡.𝑒 300.00 Wh/kg
Efficiency 𝜂𝑏𝑎𝑡 0.90
Mass margin 𝜅𝑏𝑎𝑡 25%
Turbopump Assembly
Power density 𝛿𝑡𝑝 12.50 kW/kg
Gas Generator
Material density 𝜌𝑔𝑔.𝑚 8440 kg/m3
Material strength 𝜎𝑔𝑔.𝑚 500 MPa
Safety factor 𝜅𝑔𝑔 2.50
Gas Generator Working Fluid
Mixture ratio (𝑂/𝐹)𝑔𝑔 0.333
Density 𝜌𝑔𝑔 12.624 kg/m3
Mass flow rate 𝑚̇ 𝑔𝑔 0.08 kg/s
Stay time 𝑡𝑠 0.01 s
Mass margin 𝜅𝑝 5%
 69

100

90

80

70

60
Mass (kg)

50

40

30

20 — Turbopump system
10
— Electropump system
0
0 100 200 300 400 500 600 700 800 900 1000
Burn Time (s)

Figure 4.6.: Variation of mass of both the feed systems (without thrust chamber
propellant) with engine burn time.

2.5
— Turbopump system

2.0
— Electropump system

1.5
Mass Ratio

1.0

0.5

0.0
0 100 200 300 400 500 600 700 800 900 1000
Burn Time (s)

Figure 4.7.: Variation of the ratio of feed system mass to thrust chamber propellant
mass with engine burn time.
70

both power and energy densities. When the burn times are short, the minimum
battery cell discharge time imposed from its critical power density warrants the
need for a larger number, and thus mass, of battery cells for the engine operation;
this is evident from the horizontal segment of the electropump system plot in
figure 4.6. For long burn times, the functional energy capacity of the battery pack,
which is proportional to the energy necessary for engine operation, establishes the
total mass of the battery pack. In our case, the switch from critical power density
to critical energy density occurs around 150 s.
Figure 4.7 illustrates the variation of the ratio of feed system mass to thrust chamber
propellant mass as a function of engine operation duration. Noticeably, the power-
imposed and the energy-imposed burn times for the electropump system are
signified by the curved and the straight sections of the plot. Overall, we recognize
that, for our configuration of the feed systems, the mass fraction of the turbopump
system is lighter than that of the electropump system up to a burn time of about
88 s after which the electric pump feed system yields a lighter overall mass.
These results conclusively validate that an electric-pump-fed engine cycle can
indeed outperform a turbopump feed system under the proper circumstances
of engine operation. Since typical rocket engine burn durations vary anywhere
between 180 s and 300 s, our feed system is never constrained by the power density
of the battery pack. We, therefore, unqualifiedly employ the above configuration of
the electropump feed system for our liquid rocket propulsion system requirements.

4.4 SUMMARY

A liquid rocket propulsion system can be categorized according to its engine cycle,
that is, the manner power is derived to feed propellants to the thrust chamber. A
pressure feed system, although is of the simplest type, needs a high tank pressur-
ization to drive the propellant downstream. A pump feed system, at the expense
of simplicity however, utilizes a rotodynamic pump near the thrust chamber inlet
to maintain the desired flow rate and injection pressure. With each feed system
having its suite of benefits and drawbacks, our preference of one to another encom-
passes a compromise between some critical design parameters and some crucial
performance metrics. And we fairly quickly deem the pressure feed system un-
suitable for our application owing to our range of thrust magnitudes and thereby
combustion chamber pressures.
We delve into the operational details of turbopump and electropump systems next,
and work out the fundamental elements of the pump assembly. With regard to
 71

the turbopump feed system, we define the engine cycle based on the source of
the energetic drive fluid to power the turbopump assembly and the destination of
the turbine exhaust fluid after doing the work. The three most common turbop-
ump systems, viz. the gas generator cycle, the staged combustion cycle and the
expander cycle, are outlined in the process. We realize that increasing the turbine
working fluid flow rate increases the pump power output which translates into
an increased combustion chamber propellant flow rate and an increased engine
thrust magnitude. In an electropump feed system, the gas turbine assembly is
swapped out for electric motors powered by battery packs that drive the propellant
pumps instead. While the mass penalty from such a system had historically been
an impediment, rocket engines employing electric-pump-fed cycles are at present
feasible by dint of advancement in battery technologies and refinement in electric
motors. And under the right set of circumstances, an electropump system can even
outperform a turbopump system.
In our assessment of the practicality of each feed system type for our engine
application, we settle on a combustion chamber pressure of 𝑃𝑐𝑐 = 6000kPa and head
rises of 9000kPa and 7500kPa for the oxidizer pump and the fuel pump respectively.
Eventually, the overall mass and mass fractions of turbopump and electropump
systems are evaluated for a multitude of component characteristics exclusive to the
feed system, like those of gas generator or battery pack for instance. We observe
that our electropump feed system does indeed outperform the turbopump feed
system for our requirements of engine operation.
5
Design of the Thrust Chamber

The design of a thrust chamber assembly presents among the most elaborative
tasks in the realization of a liquid rocket propulsion system. Intertwined with
the overall development of a number of rocket subsystems and subassemblies,
such a procedure necessitates numerous iterations and optimizations in order
to accomplish adequate performance parameters. In addition to performance, it
entails a multitude of rationalizations behind the selection of engine components
and component arrangements that help satisfy stringent mission objectives with
minimum life cycle cost.

Yet, any design, unless verified experimentally, is at best considered preliminary—

primarily because the myriad physical, chemical and thermodynamical processes
of combustion within a thrust chamber are impossible to model exactly, making
any analytical approach imperfect, and in turn making any designed specification
approximate [13] [12]. Regardless, the defining criteria in the preliminary design
of a rocket engine stem from the performance metrics of the propulsion system,
which in turn derive from the objectives of the mission.

5.1 DESIGN REQUIREMENTS

Our hypothetical mission, as specified earlier, requires us to deliver a light payload

of nearly 200 kg into LEO using a two-stage rocket. With the delta-v budget from
the launch vehicle being 9500 m/s, we decide to target for a generated thrust of
approximately 30 kN from each thrust chamber. Additionally, the outer diameter
of our launch vehicle is capped at 1.25 m, which but constrains the size and/or
number of thrust chamber(s) per stage. Moreover, the delta-v for the first stage is
74

presumed to be about 4500 m/s, whereas that for the second stage is presumed to
be around 5000 m/s.

We, therefore, are required to design two thrust chamber versions—a sea-level
variant for the booster stage and a vacuum-optimized variant for the upper stage.
Both propulsion systems are obligated to be as similar as pragmatically possible,
allowing only for minimal modifications to accommodate the disparate pressure
conditions of sea level and vacuum.

The second stage, consequently, requires an engine with a high expansion-area-

ratio nozzle. And the first stage, in order to generate the necessary thrust for liftoff
and ascent, requires a cluster of thrust chambers in a parallel burn arrangement.
In addition, to meet the total impulse requirements of the stage, and following a
typical flight profile, the booster stage engines are required to nominally operate
for a duration of up to 180 s.

Furthermore, a high thrust-to-weight ratio and a high specific impulse, being highly
desirable characteristics of a propulsion system, are what we aspire to obtain. With
all the initial conditions laid out, we now can work out the parameters for designing
both our thrust chambers.

5.2 DESIGN PARAMETERS

It has heretofore been settled on that our propulsion system runs on the liquid
oxygen/liquid methane combination at a mass mixture ratio of 𝑂/𝐹 = 2.75 and
operates at a combustion chamber pressure of 𝑃𝑐𝑐 = 6000 kPa (= 60 bar). Now, we
need to determine the thrust chamber operating parameters such as combustion
chamber temperature, effective molecular mass, and specific heat ratio. Prior to
doing so however, we ascertain the expansion area ratios of both the nozzles.

5.2.1 Expansion Area Ratios

The ideal values of thrust chamber performance for a given ambient pressure are
obtained under the conditions of optimum expansion. With ambient pressure
varying substantially during the ascent of launch vehicles, trajectory computations
typically determine the optimum exit pressure for booster-stage engines. And
because ambient pressure is rather low at high altitudes, upper-stage engines
employ nozzles with high expansion area ratios so as to achieve close to optimum
expansion.
 75

While the expansion ratio ought to approach infinity in order to satisfy ideal
expansion at zero or near zero ambient pressure, the maximum theoretical exhaust
velocity remains but finite (since it corresponds to an exhaust fluid with finite
energy content) [12]. Besides, an increase in expansion area ratio in vacuum, even
though increases performance, yields diminishing returns in terms of payload due
to an increased engine size and thus an increased stage mass [13]. An optimization
between engine performance and payload capacity, therefore, dictates the optimum
expansion area ratio of a nozzle.

Standard values of expansion area ratio are between 3 and 30 for nozzles operating
from sea level and between 40 and 400 for those optimized for vacuum operation
[12] [13]—in practice, however, the opted ratios tend to be towards the lower
margin of their respective ranges. Thrust chamber parameters exclusive to the
first/booster stage and the second/upper stage are hereinafter designated with the
subscripts 1st and 2nd respectively.

After some scrutiny into propulsion systems of comparable performance, we settle

on nozzle expansion area ratios of 𝜖1st = 15 and 𝜖2nd = 100 for the sea-level engine
and the vacuum-optimized engine. And from CEA [39], we figure out that the
chosen expansion area ratios result in nozzle exit pressures of 𝑃𝑒.1st = 39.878 kPa
and 𝑃𝑒.2nd = 3.058 kPa for the respective engine variants. We, therefore, obtain
pressure ratios of (𝑃𝑐𝑐 /𝑃𝑒 )1st = 150.459 and (𝑃𝑐𝑐 /𝑃𝑒 )2nd = 1962.067 across the
respective nozzle variants.

Because performance is negatively affected by ambient pressure, we consider the

maximum ambient pressure in all our performance calculations. The maximum
external pressure for the booster-stage nozzle equals 𝑃𝑒𝑥𝑡.1st = 101325.0 Pa (at sea
level). For a typical flight profile, wherein stage separation occurs at about 75 km
and engine ignition commences at around 80 km, the maximum external pressure
for the upper-stage nozzle equals 𝑃𝑒𝑥𝑡.2nd = 0.886280 Pa (at an altitude of 80 km)
[52]. All thrust chamber parameters obtained above are tabulated below.

1st Stage 2nd Stage

Expansion area ratio 𝜖 15 100
Exit pressure 𝑃𝑒 39878.0 3058.0 Pa
Pressure ratio 𝑃𝑐𝑐 /𝑃𝑒 150.458900 1962.066710
External pressure 𝑃𝑒𝑥𝑡 101325.0 0.886280 Pa
76

Furthermore, the dimensions of a thrust chamber—of vacuum-optimized thrust

chambers in particular—are often constrained by the outer diameter as well as
the interstage length of a launch vehicle. With the diameter of our vehicle being
1250 mm, it is but prudent to restrict the maximum diameter of the upper-stage
nozzle to 750 mm so as to accommodate any support structure and/or additional
component within the rocket interstage. We preliminarily select an exit diameter
of 600 mm for the aforesaid nozzle, thereby leaving us a margin of some 300 mm
radially between the thrust chamber and the interstage fairing.

Now, from the exit diameter of 𝐷𝑒.2nd = 600.0 mm and the expansion ratio of
𝜖2nd = 100, we obtain a throat diameter of 𝐷𝑡.2nd = 60.0 mm for the second-stage
engine variant. Because the throats of the upper-stage nozzle and the booster-stage
nozzle are dimensionally identical, the expansion ratio of 𝜖1st = 15 gives us an exit
diameter of 𝐷𝑒.1st = 232.379 mm for the first-stage engine variant.

With major external parameters now preliminarily ascertained, we next proceed

with determining all thrust chamber performance parameters, and later move on
to scrutinizing and optimizing the obtained values.

5.2.2 Performance Parameters

Without delving deep into solving the chemical/thermodynamical quantities of

combustion, we run CEA [39] to figure out the following crucial parameters:
combustion chamber temperature 𝑇𝑐𝑐 , effective molecular mass 𝔐, and specific
heat ratio 𝑘.

Combustion chamber temperature 𝑇𝑐𝑐 3338.670 K

Effective molecular mass 𝔐 19.503 g/mol
Specific heat ratio 𝑘 1.207

Because these metrics form the basis of all subsequent calculations, we run RPA
[41] and corroborate the accuracy of the values; in addition, combustion charts
computed in [43] and presented in figure 5.1 only help validate our values.

From the obtained values, and using equations from chapter 2, we compute the
following performance parameters for the first-stage thrust chamber variant: gas
constant 𝑅, characteristic velocity 𝑐∗ , thrust coefficient 𝐶𝑓 , and effective exhaust
velocity 𝑐.
 77

Adiabatic Flame Temperature (K)

Combustion Chamber Pressure (atm)

(a) Combustion Chamber Temperature

Gas Molecular Mass (g/mol)

Combustion Chamber Pressure (atm)

(b) Effective Molecular Mass

Speci�ic Heat Ratio

Combustion Chamber Pressure (atm)

(c) Specific Heat Ratio

Figure 5.1.: Variation of the three crucial combustion parameters with chamber
pressure. [43]
78

1st Stage
Gas constant 𝑅 426.317111 J/K/kg
Characteristic velocity 𝑐∗ 1835.757507 m/s
Thrust coefficient 𝐶𝑓 1.531862
Effective exhaust velocity 𝑐 2812.128075 m/s

Now, propellant mass flow rate for the first-stage thrust chamber calculated in
terms of combustion chamber pressure, characteristic velocity and throat area
(using equation 2.14) equals 𝑚̇ 1st = 9.241199 kg/s. However, the mass flow rate
when calculated as minimum rated thrust (at sea level) divided by effective ex-
haust velocity equals 𝑚̇ 1st = 10.668077 kg/s. The significant mismatch in flow rate
persists for the second-stage thrust chamber, which but confirms that our initial
assumptions of nozzle dimensions are incorrect. Therefore, we redetermine the
nozzle throat diameter to be 𝐷𝑡 = 65.0 mm, which yields a nozzle throat area of
𝐴𝑡 = 0.003318307240 m2 . The corrected throat dimension results in propellant
mass flow rate variations of less than 2% (and is essential to account for design
uncertainties and correction factors). The thrust chamber dimensions determined
thus far are tabulated here.

1st Stage 2nd Stage

Throat diameter 𝐷𝑡 65.0 65.0 mm
Exit diameter 𝐷𝑒 251.744 650.0 mm
Exit area 𝐴𝑒 0.049775 0.331831 m2

The ideal values of mass flow rate 𝑚,̇ specific impulse 𝐼𝑠 , and thrust 𝐹 for the
first-stage thrust chamber variant are presented below.

1st Stage
Mass flow rate 𝑚̇ 10.845574 kg/s
Specific impulse 𝐼𝑠 286.757259 s
Thrust 𝐹 30499.142443 N
 79

5.2.3 Performance Correction Factors

Prior to proceeding with ascertaining the remaining thrust chamber parameters,

we correct our ideal values of performance using, in the absence of any experi-
mental data, rather reasonable correction factors. Because our assumption for flow
characteristic across the nozzle involves frozen composition at the throat (which
tends to underestimate performance typically by 1 to 4%), we presume relatively
optimistic figures for all correction factors and yet expect to obtain theoretical
performance somewhat close to actual numbers.

Characteristic velocity 𝜁𝑐∗ 0.99

Thrust coefficient 𝜁𝐶𝑓 0.97
Mass flow rate 𝜁𝑚̇ 1.01
Effective exhaust velocity 𝜁𝑐 0.960300
Thrust 𝜁𝐹 0.969903

The following corrected performance parameters are adopted hereinafter in the

calculation of all additional thrust chamber metrics.

1st Stage 2nd Stage

Characteristic velocity 𝑐∗ 1817.399932 1817.399932 m/s
Thrust coefficient 𝐶𝑓 1.485907 1.885631
Effective exhaust velocity 𝑐 2700.486590 3426.945580 m/s
Mass flow rate 𝑚̇ 11.064676 11.064676 kg/s
Specific impulse 𝐼𝑠 275.372996 349.451197 s
Thrust 𝐹 28980.711195 36776.824036 N

5.3 DESIGN OF THE NOZZLE

With the primary performance metrics ascertained, we now look into the shape
of the convergent-divergent nozzle. Because the flow of the combustion fluid
is subsonic within the convergent nozzle section, fine details in the profile of
the convergent section do not appreciably affect nozzle performance. On the
80

contrary, the contour of the divergent section is of profound significance to nozzle

performance owing to the supersonic velocities of the exhaust jet contained within.

There primarily exist two configurations of supersonic nozzles: conical-shaped

nozzles and bell-shaped nozzles. While a conical nozzle entails little design com-
plexity and allows ease of manufacture, it undergoes an appreciable performance
loss due to the nonaxial component of the exhaust gas velocity. A bell nozzle, on
the other hand, provides an efficient expansion of the supersonic exhaust gas flow
with a shorter and lighter divergent section, albeit at the expense of a complex con-
tour design. A few geometries for the divergent nozzle section have been analysed
below.

5.3.1 Ideal Nozzle

The flow within a convergent-divergent nozzle—which is considered compressible,

isentropic, irrotational, inviscid, steady and uniform—may again be assumed
quasi-one-dimensional. Even though the flow intrinsically is three-dimensional,
variations in the flow variables are far more pronounced in the axial (x) direction
than in the radial (y and z) direction due to the varying cross-sectional area of
the nozzle along its axis [53]; the flow variations along y and z are neglected and
the fluid properties are considered as functions of x only. Although merely an
approximation to the actual flow, the quasi-one-dimensional analysis approach can
help determine flow properties within a rocket nozzle to a fair degree of accuracy.
The theory of the method of characteristics is further applied in analysing the flow
processes and designing the divergent section of a supersonic nozzle.

For a supersonic flow in a two-dimensional space, the flow variables are continuous
and the governing partial differential equations are hyperbolic. Now, there exist
unique curves in the x-y space, called characteristic curves, on which derivatives
of the flow variables are indeterminate and may as well be discontinuous [54];
such curves transform the nonlinear partial differential equations into integrable
ordinary differential equations and, as a result, generate the compatibility equations
with exact solutions that remain constant along the characteristics [54]. Besides,
because it relies on developing a network of curves in the 2-D space, application
of the method of characteristics is inherently a graphical technique. Figure 5.2
illustrates the existence of two characteristics through a point A. Since exploring
the entire mathematical theory of hyperbolic equations is beyond the scope of
this study, only the essential elements of the method of characteristics have been
summarized below.
 81

y
C+

θ
A
μ
Streamline

Characteristic Curves

C–
x

Figure 5.2.: Definition sketch of the velocity field and the local characteristic curves
for a point in a supersonic flow.

⃗⃗⃗ ⃗⃗ through
The velocity potential equation of a streamline defined by a velocity field 𝑉
Cartesian coordinates can be expressed as

𝑢2 𝜕2 𝜙 𝑣2 𝜕2 𝜙 2𝑢𝑣 𝜕2 𝜙
(1 − ) + (1 − ) − 2 =0 (5.1)
𝑎2 𝜕𝑥2 𝑎2 𝜕𝑦2 𝑎 𝜕𝑥 𝜕𝑦

where, 𝜙 is the velocity potential, 𝑎 is the local speed of sound, and 𝑢 and 𝑣 are the
velocity components along x and y for the point A on the streamline.

Because the velocity potential and its derivatives are functions of x and y, the
differential relations are given as follows

𝜕𝜙 𝜕2 𝜙 𝜕2 𝜙
𝑑( ) = 𝑑𝑢 = 𝑑𝑥 + 𝑑𝑦 (5.2a)
𝜕𝑥 𝜕𝑥2 𝜕𝑥 𝜕𝑦
𝜕𝜙 𝜕2 𝜙 𝜕2 𝜙
𝑑 ( ) = 𝑑𝑣 = 𝑑𝑥 + 2 𝑑𝑦 (5.2b)
𝜕𝑦 𝜕𝑥 𝜕𝑦 𝜕𝑦

On examining the system of equations 5.1 and 5.2 and solving for the ensuing
second order derivatives as described by Anderson in [53], it becomes apparent
that there are values of 𝑑𝑥 and 𝑑𝑦 for which the velocity derivative 𝜕2 𝜙/(𝜕𝑥 𝜕𝑦)
becomes undefined. This characterizes the presence of specific directions through
the point A which satisfy the condition for indeterminacy of the characteristic lines.
82

The precise direction of the characteristic can, therefore, be formulated as

2 2
𝑑𝑦 − 𝑢𝑣
𝑎2
± √( 𝑢 𝑎+𝑣
2
)−1
( ) = 2
(5.3)
𝑑𝑥 𝑐ℎ𝑎𝑟 1 − 𝑢𝑎2

where, 𝑢 = 𝑉 𝑐𝑜𝑠 𝜃, 𝑣 = 𝑉 𝑠𝑖𝑛 𝜃, and (𝑑𝑦/𝑑𝑥) denotes the slope of the characteristic.

Using the Mach angle 𝜇, the expression for the slope can subsequently [53] be
obtained as
𝑑𝑦
( ) = 𝑡𝑎𝑛 (𝜃 ∓ 𝜇) (5.4)
𝑑𝑥 𝑐ℎ𝑎𝑟

The result signifies that two characteristic curves run through the point A: one with
a slope of 𝑡𝑎𝑛 (𝜃 − 𝜇) and labelled as 𝐶− (inclined below the streamline), and the
other with a slope of 𝑡𝑎𝑛 (𝜃 + 𝜇) and labelled as 𝐶+ (inclined above the streamline).

After substantial mathematical manipulation of the system of above equations as

described in [53], the compatibility equations along the characteristic lines can be
expressed in terms of the Prandtl–Meyer function 𝜈 as follows

𝜃 + 𝜈 = 𝐾− = constant [along C− characteristic] (5.5a)

𝜃 − 𝜈 = 𝐾+ = constant [along C+ characteristic] (5.5b)

In essence, the characteristic lines symbolize the Mach lines through a point in a
supersonic flow [53]. Accordingly, an expansion fan is formed whenever the flow
encounters a convex corner and is turned away from itself. And correspondingly,
an oblique shock is formed whenever the flow encounters a concave corner and is
turned into itself. Thus, the function of these waves is to gradually change the
local flow direction within the divergent section so as to achieve an axial wave-free
flow with an exit Mach number 𝑀𝑒 . Although an infinite number of Mach waves
crisscross the flow field, only a finite number of characteristic curves are deemed
adequate for determining the divergent section contour [54].

Application of the method of characteristics is depicted schematically in figure 5.3.

While the sketch here is rather coarse, the characteristic network used in practice
is developed significantly finer in an effort to generate an optimum profile for a
supersonic nozzle. The profile of the expanding section, as [53] states, is somewhat
arbitrary and is typically a circular arc of a large radius. Of paramount importance,
however, is the contour of the straightening section since it serves to cancel the
expansion waves produced within the expanding section. Taking advantage of
 83

n
ing Sectio
Straighten 13
ction 12
ng Se
Ex pandi 8
5 θw = θmax
1
7 9
Limiting
Characteristic
2 4

3 6 10 11

2’ 4’
Sonic
Line 9’
7’
1’
5’
8’

12’
13’

Figure 5.3.: Design schematic of a supersonic nozzle contour from a network of

crisscrossing characteristics and intersecting grid points.

flow symmetry with respect to the nozzle axis, only the upper half of the divergent
section is commonly analysed, wherein the centerline is treated as a solid wall
and the characteristics emanating through it are treated as reflections of the upper
centered expansion waves.

Mapping of the supersonic flow field, therefore, begins at internal and/or wall
grid points on an initial data line of known properties (usually at the throat) and
propagates downstream along characteristic directions 𝐶+ and 𝐶− . This establishes
local flow angle 𝜃 and local Prandtl–Meyer angle 𝜈 from respective compatibility
constants 𝐾+ and 𝐾− , and in turn obtains the local Mach number 𝑀. Additional
flow parameters (like pressure 𝑃, temperature 𝑇 and specific volume 𝑉) at each
grid point within the flow field can subsequently be computed using the relations
for isentropic condition and frozen composition [12]. The methods outlined above,
along with any steps in between as discussed in [54], can be implemented in a
computer program to design a proper divergent section contour for our thrust
chamber.

We use MATLAB to generate an ideal ‘minimum length nozzle’ contour for a

uniform inlet flow at the throat. Employing 100 characteristic lines for our analysis,
we obtain expansion area ratios of 𝜖1st = 14.016 and 𝜖2nd = 108.229 from exit
Mach numbers of 𝑀𝑒.1st = 3.724 and 𝑀𝑒.2nd = 5.952 for both our thrust chamber
variants respectively. Results of the computation are plotted in figure 5.4, where
84

700
70

600
60

Expansion Area Ratio = 108.229

Expansion Area Ratio = 14.016

500
50

400
40
x/y*

x/y*
•

•
Exit Mach Number = 3.724

Exit Mach Number = 5.952

300
30

200
20

100
10
0

0
150

100
15

10

50
5

y/y* y/y*

(a) Booster-stage nozzle contour (b) Upper-stage nozzle contour

Figure 5.4.: The ‘minimum length nozzle’ contours for our thrust chamber variants
computed using the method of characteristics. The scale of the plot
5.4a is one-tenth that of the plot 5.4b above.
 85

the x-y plane represents the ratio of corresponding nozzle dimension to throat
dimension for a 2-D nozzle. Contour coordinates of the exit wall point, therefore,
are at (64.210, 14.016) and (730.838, 108.229) for nozzles of the booster stage and
the upper stage respectively, with the throat wall point being at (0.0, 1.0) for each
nozzle. Moreover, the obtained area ratios are within 0.166% and 0.297% of their
respective theoretical values for both the nozzle configurations. This signifies that
the degree of accuracy of our graphical solutions, even when computed with only
100 characteristic lines, is remarkable.

Yet, an ideal contour design for our divergent sections is anything but practicable.
Coordinates of the exit wall points above imply that our nozzles are dispropor-
tionately long for rocket applications. Besides, the supersonic flows accomplish
the exit Mach numbers along the centerline only at 21.672% and 13.110% of their
respective nozzle lengths. Furthermore, any adverse effects of chemical kinetics
are mitigated by employing a circular arc instead of a sharp corner at the throat,
which only lengthens the nozzle some more [55]. Such a long divergent contour,
although produces a one-dimensional exhaust profile thereby maximising the
geometric efficiency, also raises viscous drag, inert mass and heat flux of the nozzle
thus reducing its overall efficiency. The factors above validate that an ideal contour
design does little good to our engine configurations and warrant the need for a
more optimum nozzle contour.

5.3.2 Truncated Ideal Contoured (TIC) Nozzle

Although an ideal nozzle is necessary to produce an axial uniform exit flow condi-
tion, any thrust generated towards the end of the straightening section is negligible
owing to the small divergence angles near the exit. It is, therefore, prudent to trun-
cate the nozzle to attain a more feasible contour design; the contour thus obtained
is called a truncated ideal contour (TIC). A graphical technique for selecting the
optimum contour from a family of TIC nozzles is presented in [56]. A complete
set of ideal contours along with lines representing constant length, vacuum thrust
coefficient, surface area and exit diameter is initially synthesised in a plot. An
optimisation process is finally implemented to determine where the contour has to
be truncated so as to maximise performance for a given constraint such as nozzle
length or exit diameter. Figure 5.5 illustrates an enlarged section of such a plot
to help visualize the optimisation procedure. Point A, which is where the thrust
coefficient line is tangent to a line of constant radius, is the optima representing
the nozzle exit point that yields the maximum thrust for a given expansion ratio.
Point B, which is where the thrust coefficient line is tangent to the constant surface
86

Figure 5.5.: Illustration of an optimum TIC nozzle for given constraints. [55]

area line, represents the optima of a nozzle contour for a given surface area. Point
C represents the contour of a nozzle with the maximum performance for a given
length. Point D represents the exit point at which the maximum thrust is attained
from any given nozzle contour. Among these, the TIC nozzle represented by point
A is of the most practical interest.

Gogish in [57] presents a method to further shorten the nozzle length, the basis
of which is to linearly compress a TIC nozzle in the axial direction to produce a
compressed truncated ideal contour (CTIC). Any discontinuity in the CTIC nozzle
wall slope produced as a result is eliminated by a cubic equation which smoothly
connects the linearly compressed curve with the initial circular curve. The above
compression yields a nozzle with a more rapid initial expansion followed by a more
severe turn back, therefore resulting in the propagation of strong compression
waves into the flow field. With a strong enough compression, the characteristic
lines can coalesce to form a right-running oblique shock wave that increases the
static pressure of the supersonic flow across it. When the shock wave lies near the
nozzle wall, the increased local pressure along the wall helps marginally increase
 87

the thrust. And under the right circumstances, a CTIC nozzle may even yield a
higher performance than a Rao nozzle [55].

5.3.3 Thrust Optimised Contoured (TOC) Nozzle

The application of the calculus of variations provides a direct and elegant approach
of designing the optimal nozzle contour. A complicated method formulated to
produce a nozzle with a thrust optimised contour (TOC) for prescribed values of
nozzle length was simplified significantly by Rao in [58] and was later adopted
extensively for rocket propulsion applications; the TOC nozzle is hence commonly
referred to as a Rao nozzle, and the procedure to design one is as follows.
A constant Mach number line TO is first generated at the throat using transonic flow
analysis. With the solid boundary at TN and the flow condition along TO defined,
a kernel flow field TNKO is developed next using the method of characteristics for
specified values of radius of curvature 𝑅𝑡.𝑑 and angle of divergence 𝜃𝑡.𝑑 downstream
of the throat. Location of the exit wall point E is established now from the given
design parameters of nozzle length and expansion area ratio. The characteristic
line NK is determined next by concurrently satisfying the following conditions:
▪ A control surface PE is created between the exit point E and a select point P
along the kernel line NK.
▪ The mass flow across PE equals the mass flow across NP.
▪ The thrust obtained from the nozzle is the maximum.
The contour NE is ultimately constructed by selecting points P’, P’’, etc. along the
line NK that create a series of control surfaces P’E’, P’’E’’, etc. as described above

E
E’
E’’
Rt.d
θt.d
T N

Rt
O

P’’ Control
P’ P Surfaces
T N
Kernel
O K

Figure 5.6.: Schematic of a thrust optimised contoured (TOC) nozzle.

88

which define points E’, E’’, etc. on the contour NE. The design procedure detailed
above is illustrated in figure 5.6.

Compared to a TIC nozzle, a TOC nozzle features a greater initial expansion and,
as a result, induces a more drastic turning of the flow [55]. The corresponding
values of wall angle and Mach number are, consequently, higher immediately
downstream of the throat and lower near the exit. The ensuing compression waves
formed in the region NPE coalesce into a right-running shock downstream of the
control surface PE and, in the process, keep the region shock free. By definition, a
TIC nozzle is created when point P equals point K.

5.3.4 Thrust Optimised Parabolic (TOP) Nozzle

Because the computation necessary to construct a TOC nozzle is rather complex,

Rao in [59] proposed a simple geometric approach to design a nozzle with a thrust
optimised parabolic (TOP) contour. Also known as a parabolic bell nozzle, the
profile of such a nozzle involves two circular arcs at the throat T and a parabolic
curve from the inflection point I to the exit point E. Figure 5.7 portrays the design
configuration of a TOP nozzle. The convergent contour immediately upstream
of the throat is an arc having a radius of 𝑅𝑡.𝑢 = 1.5 𝑅𝑡 . And the divergent contour is
composed of an arc with a radius of 𝑅𝑡.𝑑 = 0.4 𝑅𝑡 immediately downstream of the
throat followed by a parabola that can be defined with the following five variables:
throat radius 𝑅𝑡 , exit radius 𝑅𝑒 , wall slope at the inflection point 𝜃𝑖 , wall slope at
the exit point 𝜃𝑒 , and nozzle length from the throat plane to the exit plane 𝐿𝑛 .

The bell nozzle length is typically specified as a fractional length of an equivalent

standard conical nozzle with a 15° divergence half angle. The two parabolic contour

E
Rt.u
θe
Rt.d
θi

T
Re
I θi

Rt Ln

Figure 5.7.: Design schematic of a thrust optimised parabolic (TOP) nozzle.

 89

angles 𝜃𝑖 and 𝜃𝑒 are functions of nozzle length fraction 𝐿𝑓 and expansion area
ratio 𝜖, and can be determined from figure 5.8. With careful selection of the input
parameters, a parabolic contour can accurately approximate an optimum contour
generated using the method of characteristics. Yet, minor variations in the above
variables incur only a miniscule loss in nozzle performance, making the design of
a TOP nozzle somewhat forgiving.

Figure 5.8.: The upper plot shows variation of geometric efficiency of a TOP nozzle
with fractional length of a 15° conical nozzle. Optimal values for inflec-
tion 𝜃𝑖 and exit 𝜃𝑒 contour angles as functions of nozzle length fraction
and expansion area ratio are shown with the lower set of curves. [55]
90

Now, strong compression waves are formed due to the discontinuity in the contour
curvature at the point I where the circular arc transitions into the parabolic curve.
These waves coalesce into a right-running oblique shock wave upstream of the last
left-running characteristic line (unlike as in a TOC nozzle) and, in consequence,
marginally raise the wall pressure at the nozzle exit. This additional margin in exit
pressure can, to an extent, avert any adverse effects of flow separation in sea-level
thrust chambers and, in turn, may even offset any loss in the performance of a TOP
nozzle.

Having scrutinized the crucial characteristics of each nozzle configuration, we can

avow that a TOP nozzle yields the most favourable contour—not only in terms of
performance but also with regard to convenience—for a range of requirements
and a variety of applications. Even when a single engine is required to operate
throughout the disparate lower and upper atmospheric pressure conditions (such
as the SSME), thrust chambers with parabolic bell nozzles have proven to be the
most useful [12]. Moreover, the entire contour geometry can be ascertained from
the mere determination of the above contour variables. These aspects only serve to
underscore the advantages of the TOP contour over the others even in the absence
of any empirical data.

We, therefore, employ the TOP nozzle for both our thrust chamber variants. And
in doing so, we strive to as meticulously choose values for each contour variable
as theoretically possible. Now, the dimensions of throat and exit for each nozzle
variant have already been established. For the length of the nozzle, decent values
range anywhere from 75% to 85% of an equivalent standard conical nozzle length
[12], with 80% bell offering a good compromise between efficiency and penalty for
both sea-level nozzles and vacuum-optimized nozzles [13]; this can not only be
verified from the upper plot in figure 5.8 but may also be evidenced by the ubiquity
of 80% bell nozzles across the rocket industry. From trigonometric analyses, we
obtain lengths of 𝐿𝑛.1st = 278.775 mm and 𝐿𝑛.2nd = 873.300 mm for the respective
nozzle variants.

The two divergence angles of the parabolic segment can be established from the
lower plot in figure 5.8. For our expansion ratios of interest, the inflection and exit
angles for the sea-level nozzle are 𝜃𝑖.1st = 28.0° and 𝜃𝑒.1st = 10.0°, while those for
the vacuum-optimized nozzle are 𝜃𝑖.2nd = 34.0° and 𝜃𝑒.2nd = 8.0°. As stated above,
the convergent and divergent sections immediately upstream and downstream of
the throat are, for each nozzle variant, circular arcs of 1.5 times and 0.4 times the
respective radius of the throat. And for the convergent section, we rationally choose
a nozzle convergence angle of 𝜃𝑐 = 30.0° for each thrust chamber configuration.
 91

With all nozzle variables now ascertained, we tabulate the pertinent data below
and proceed with designing the combustion chamber.

1st Stage 2nd Stage

Throat radius 𝑅𝑡 32.50 32.50 mm
Exit radius 𝑅𝑒 125.87 325.00 mm
Nozzle length 𝐿𝑛 278.78 873.30 mm
= 0.8 × 348.468894 1091.624861 mm
Throat upstream radius 𝑅𝑡.𝑢 48.75 48.75 mm
Throat downstream radius 𝑅𝑡.𝑑 13.00 13.00 mm
Nozzle inflection angle 𝜃𝑖 28.0° 34.0°
Nozzle exit angle 𝜃𝑒 10.0° 8.0°
Nozzle convergence angle 𝜃𝑐 30.0° 30.0°

5.4 DESIGN OF THE COMBUSTION CHAMBER

The process of dimensioning a combustion chamber can—in the absence of any

experimental data from prior successful engines of similar design configurations
and operating conditions—be somewhat arbitrary. While chamber geometries can
theoretically take any shape, actual design approaches tend to keep the combustor
proportions neither too long and thin nor too short and fat but not of spherical
proportions either [13]. Customarily, chambers feature a moderate longer-length
smaller-diameter proportion so as to favour combustion stability and enhance
combustion efficiency [13]. An optimum combustor volume, therefore, provides
an adequate stay time for complete combustion while trying to minimize both inert
mass and heat flux.

Because our chamber pressure is quite high, we can afford to adopt a marginally
smaller combustor volume for our thrust chamber. With standard values anywhere
between 0.56 m to 0.76 m for our propellant combination, we judiciously select a
characteristic length of 𝐿∗ = 0.7 m. And in order to restrict pressure losses, we
examine contraction area ratios of similar engine designs [13] [12], and tentatively
select a combustion chamber area of 𝐴𝑐𝑐 = 4.0 𝐴𝑡 = 0.013273 m2 . These dimen-
sions correspond to a stay time of 𝑡𝑠 = 0.885 ms, which corroborates well with
92

the experimental evaluations performed in [60]. Having computed all internal

dimensions for both our thrust chamber variants, we tabulate the data and present
configuration layouts and visual renders of both the variants.

Characteristic length 𝐿∗ 0.7 m

Contraction area ratio 𝜖𝑐 4.0
Chamber diameter 𝐷𝑐𝑐 130.00 mm
Chamber area 𝐴𝑐𝑐 0.013273 m2
Chamber volume 𝑉𝑐𝑐 0.002323 m3
Stay time 𝑡𝑠 0.885 ms
Cylindrical section length 𝐿𝑐 142.16 mm
Frustum section length 𝐿𝑓 56.29 mm
Combustion chamber length 𝐿𝑐𝑐 198.45 mm

140.414
.75
48

30°
48.75

30°
65

58.041
32.5

198.455

Figure 5.9.: Internal configuration layout of the combustion chamber. All dimen-
sions are in mm.
 93

(a) Booster-stage thrust chamber (b) Upper-stage thrust chamber

𝜖1st = 15 𝜖2nd = 100
𝐷𝑒.1st = 251.74 mm 𝐷𝑒.2nd = 650.00 mm
𝐿1st = 477.23 mm 𝐿2nd = 1071.75 mm

Figure 5.10.: Internal cross-sectional profiles for our thrust chamber variants. All
dimensions are identical up to the throat, and the scale is equal for
both the schematics.

5.5 SUMMARY

From the initial conditions outlined and the operational metrics ascertained, we
incorporate methodologies to preliminarily design two thrust chamber versions for
our rocket—a sea-level variant for the booster stage and a vacuum-optimized vari-
ant for the upper stage—with both propulsion systems obligated to be as similar
as pragmatically possible. To do so, we determine the thrust chamber performance
parameters such as expansion area ratio, combustion chamber temperature, effec-
tive molecular mass, and specific heat ratio. We settle on nozzle expansion area
ratios of 𝜖1st = 15 and 𝜖2nd = 100 for our thrust chamber variants, which lets us
compute, using CEA, the other parameters listed. Because we have implemented
procedures to underestimate performance typically by 1 to 4%, we account for our
94

theoretical performance values using relatively optimistic correction factors. Doing

so ultimately helps us obtain rather realistic values of characteristic velocity, thrust
coefficient, effective exhaust velocity, mass flow rate, specific impulse, and thrust
for both the thrust chamber variants.
Further, we scrutinize the shape of the convergent-divergent nozzle, of the diver-
gent section in particular. The expansion and shock characteristics of the supersonic
exhaust jet are examined for an ideal nozzle using the method of characteristics,
a truncated ideal contoured (TIC) nozzle, a thrust optimised contoured (TOC)
nozzle, a thrust optimised parabolic (TOP) nozzle, as well as a conical nozzle. We
subsequently adopt the TOP contour for our nozzle variants.
With the nozzle throat diameter of 𝐷𝑡 = 65.0 mm, we next work out the exact
dimensions of our nozzles. For the combustion chamber size, we adopt a marginally
smaller combustor volume—with a characteristic length of 𝐿∗ = 0.7 m and a
contraction ratio of 𝐴𝑐𝑐 = 4.0 𝐴𝑡 —albeit with an adequate propellant stay time for
complete combustion. Detailed dimensions of both our thrust chamber variants
are tabulated/presented in figures 5.9 and 5.10.
6
Thrust Chamber Design Specifics

For any engineering task, there are almost always more than one design approaches
that could produce an excellent design outcome. This notion is especially true
when the applications encompass thermal management and propellant injection
of a liquid rocket propulsion system.

We have, over the previous chapters, designed the propellant feed system and
the thrust chamber configurations for our twin-staged rocket. The topic of heat
transfer together with material selection and the process of metering, injection,
atomization, vaporization and mixing of propellants for their effective combustion
are addressed in this chapter. And, of course, the design results unless validated
through experimental investigations of engine hot-fire testing may only be regarded
preliminary.

6.1 THERMAL MANAGEMENT

The high temperatures generated from the combustion fluid together with the
high heat fluxes convected to the thrust chamber wall in consequence necessitate
management of the high thermal energy within an engine to an adequate degree
so as to obviate any failure during its operation. While there are sophisticated FEA
tools to compute the distribution of thermal loads throughout the thrust chamber
assembly, we analyse one-dimensional steady-state heat transfer rates only at the
most critical wall sections, such as around key regions of the chamber, at and near
the throat, and towards the exit of the nozzle. Further, we attempt to forestall any
overheating through the adoption of appropriate cooling techniques as well as
with the selection of suitable wall materials for the thrust chamber.
96

M=0 M=1 M=Me

(a) Any typical nozzle parameter vs Mach number

1.2 60
& Pressure Ratio

1.0 50

0.8 40

Nozzle Area Ratio

0.6 30
Temperature Ratio

0.4 20

0.2 10

0.0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Mach Number

4000 8000

3500 7000

3000 6000
Temperature (K)

2500 5000
Pressure (kPa)

2000 4000

1500 3000

1000 2000

500 1000

0 0
0 1 2 3 4 5 6
Mach Number

Figure 6.1.: Variation of nozzle area ratio, temperature (with its ratio) and pressure
(with its ratio) with Mach number across the nozzle. The gradient in
the topmost plot showcases variation of any typical nozzle parameter
with Mach number—from stagnation to exit.
 97

4000 8.0

3500 7.0

3000 6.0

Local Velocity (m/s) 2500 5.0

Mach Number
2000 4.0

1500 3.0

1000 2.0

500 1.0

0 0.0
6000 5000 4000 3000 2000 1000 0
Pressure (kPa)

4000 4.0

3500 3.5

3000 3.0

Speci�ic Volume (m 3/kg)

Temperature (K)

2500 2.5

2000 2.0

1500 1.5

1000 1.0

500 0.5

0 0.0
6000 5000 4000 3000 2000 1000 0
Pressure (kPa)

0.15

0.12
Area (m 2) & Radius (m)

0.09

0.06

0.03

0.00
6000 5000 4000 3000 2000 1000 0
Pressure (kPa)

Figure 6.2.: Variation of Mach number, local velocity, temperature, specific volume,
radius and area with pressure across the nozzle.
98

The exhaust gas temperatures at the above locations for both our thrust chamber
configurations, as determined from CEA and using the equations for isentropic
flow, are 𝑇𝑐𝑐 = 3339 K and 𝑇𝑡 = 3021 K while 𝑇𝑒.1st = 1190 K and 𝑇𝑒.2nd = 564 K for
nozzle expansion area ratios of 15 and 100 respectively. Because the temperatures
up to a nozzle area ratio of approximately 10 are substantially high—as is evidenced
from the plots in figures 6.1 and 6.2—for most conventional high-temperature
materials to comfortably withstand, we need to regeneratively cool the thrust
chamber at least down to this expansion area ratio. Beyond this area ratio, the
free-stream gas temperatures and the wall heat fluxes are adequately low for us
to radiatively cool the divergent nozzle section until the exit. For the sake of
convenience, and keeping in line with design similarity, we decide to employ
regenerative cooling throughout the booster-stage engine, and adopt radiative
cooling after the area ratio of 15 for the upper-stage nozzle.

6.1.1 Booster-Stage Thrust Chamber

A suite of attractive physical properties such as high specific heat and high thermal
conductivity make liquid methane a good coolant for rocket propulsion applica-
tions [33] [60]. In our regeneratively cooled thrust chamber, therefore, the entire
fuel flow is circulated as the coolant fluid through the cooling jacket prior to its in-
jection for combustion. Incidentally, whether the fuel is injected in a liquefied state
or a supercritical phase depends on the net heat energy it absorbs from the walls
[60]; this absorbed heat in being returned for combustion replenishes the thermal
energy of the system that can be utilized to supplement engine performance [12].
With heat flux being the highest at the throat of a nozzle, the cooling channels at this
region require the maximum coolant flow velocity and the minimum cumulative
cross-sectional area compared to at any other location [12]. Heat transfer can be
augmented by injecting fresh subcooled coolant into the jacket at or near the throat
from where the flow usually turns both up towards the injector and down within
the divergent section before returning up through adjacent channels near the exit.
Any turbulence and/or surface roughness within the channels further improves
heat transfer rates to the coolant.
In our thrust chamber, the main fuel feed is injected through the inlet manifold
at the exit to flow unidirectionally upward within the longitudinal channels after
which it enters directly into the injector manifold. And instead of a brazed tubular
wall configuration, our engine employs a milled slot jacket design, where the
smooth rectangular machined grooves are enclosed within and bonded to an outer
structural shell using electrolytic plating techniques. Selection of the number of
 99

milled slots, the channel dimensions and the wall thicknesses is dependent on the
coolant pressure, its mass flow rate, its local flow velocity, as well as on the wall
material [12].

Thrust Chamber Axis

Coolant Outlet Hot Exhaust Gases
(into Injector)

Inner Wall
Coolant Channel
Outer Wall
Coolant Inlet
(from Pump)

Figure 6.3.: Schematic diagram of the thrust chamber coolant channel.

We choose Inconel 625—a nickel-based superalloy with a chromium-molybdenum

component—as the wall material for both the inner jacket as well as the outer shell.
With its marvellous mechanical properties and fantastic thermal properties, Inconel
625 [51] finds prevalent usage in extreme environments of aerospace applications,
like turbine blades and afterburner combustors inside jet engines. Besides, Inconel
625 [51] possesses such desirable characteristics as a high melting point (of about
1560 K), high design strengths at high service temperatures, good density, excellent
fabricability, easy availability, and lends itself to additive manufacturing techniques
(which this project intends to explore the possibility of). All pertinent properties
of Inconel 625 are presented in table 6.1.
The thrust chamber coolant, which constitutes the entire fuel flow, has a mass
flow rate of 𝑚̇ 𝑓 = 2.950580 kg/s. Assuming a fuel pressure drop of 0.5 times the
combustion chamber pressure distributed between the cooling jacket and the
injector assembly [13], we adopt a coolant inlet pressure of 𝑃𝑓 = 1.5 𝑃𝑐𝑐 = 9000 kPa.
Besides, the coolant inlet temperature is presumed to be 100 K which corresponds
to a density of 445.340 kg/m3 at the stated inlet pressure. The local flow velocity
of the coolant requisite for adequate convective heat transfer within the channels
at every location along the thrust chamber profile can be determined using the
Sieder–Tate equation [13]. The equation comprises of such parameters as the
Nusselt number, the Reynolds number, the Prandtl number, the coolant constant,
its heat transfer coefficient, thermal conductivity, dynamic viscosity, specific heat,
100

Table 6.1.: Physical, mechanical and thermal properties of Inconel 625. [51]
Melting temperature 1560 K
Upper service temperature 1255 K
Density 8440 kg/m3
Specific heat capacity 620 J/kg/K

Young’s modulus 147.5 GPa

Poisson’s ratio 0.336
Ultimate strength 284.1 MPa
Yield strength 275.8 MPa
Ultimate strength (Room temp.) 965.3 MPa
Yield strength (Room temp.) 479.2 MPa

Thermal conductivity 22.8 W/m/K

Linear expansion coefficient 15.8 𝜇m/m/K

density, and the like, all of which vary with both temperature and pressure of
the coolant. Instead, we run RPA [41] to iteratively ascertain the number and
dimensions of the channels at all crucial thrust chamber locations, and verify via
trial-and-error that the maximum wall temperature at any location stays below the
softening temperature of Inconel 625.

We rationally select a maximum allowable working stress of 𝜎𝑤.𝑖𝑛𝑛 = 200 MPa and
a maximum allowable wall temperature of 𝑇𝑤.𝑚𝑎𝑥 = 1000 K for the entire thrust
chamber inner jacket, both values being well below their respective safe limits for
Inconel 625. The superalloy [51] has a softening temperature of around 1100 K
and a yield stress of about 275 MPa at such elevated temperatures. The working
stress, in conjunction with internal pressure and radius along the thrust chamber
profile, obtains us a thickness of 𝑡𝑤.𝑖𝑛𝑛 = 1.00 mm that can safely be adopted as the
minimum wall thickness throughout the inner jacket. A conservative value for the
inner wall thickness is deemed acceptable not only because it reduces temperature
drops and thermal stresses across the wall but also because the inner jacket, being a
non-structural component, does not endure the induced mechanical loads usually
associated with rocket engine operation.
 101

For the thrust chamber outer shell, we consider a design stress of 𝜎𝑤.𝑜𝑢𝑡 = 400 MPa,
twice as much that of the inner wall owing to the much cooler operating tem-
peratures of the outer wall, and a safety factor of 2.0 so as to accommodate any
uncertainty in loads, vibrations or shocks. This, together with the higher coolant
pressure, results in an outer wall thickness of 𝑡𝑤.𝑜𝑢𝑡 = 3.00 mm. The ribs between
the milled channels have a minimum thickness of 𝑡𝑟𝑖𝑏 = 0.75 mm—25% thinner
than that of the inner wall because of the cooler working temperatures as well as
the balanced loading conditions within the cooling channels.

Coolant-fluid-side Twc
wall surface
Tc

Outer wall thickness

Width
Depth
Rib thickness
Inner wall thickness

Twg
Tg
Combustion-gas-side
wall surface

Twc = Coolant-side Wall Temperature

Twg = Gas-side Wall Temperature
Tc = Coolant Temperature
Tg = Gas Temperature

Figure 6.4.: Configuration layout of the thrust chamber cooling jacket.

In order to investigate the temperature distribution along the inner wall and by
extension the coolant pressure drop, we preliminarily test with selecting the num-
ber of channels between 50 and 150 while varying the channel depth between
0.75 mm and 7.50 mm at each key thrust chamber location. We employ a jacket
design with 100 channels—resulting in the channel widths of 3.40 mm around the
chamber, 1.35 mm at the throat, and 7.22 mm near the exit of the nozzle—and a
uniform depth of 1.00 mm throughout the channel profile. Such a configuration
offers an optimum combination of heat transfer rates and hydraulic losses within
the channels as well as inert mass and fabrication feasibility for the engine.

All crucial inner wall and coolant parameters at key thrust chamber sections are
summarized in table 6.2. Moreover, the convective heat flux distribution along the
102

50

45

Convective Heat Flux (MW /m 2) 40

35

30

25

20

15

10

0
0 100 200 300 400 500
Location (mm)

Figure 6.5.: Axial distribution of convective heat flux on the inner thrust chamber
wall. [41]

900
— Combustion -gas-side
800
— Coolant-�luid-side
Wall Surface Temperature (K)

700

600

500

400

300

200

100
0 100 200 300 400 500
Location (mm)

Figure 6.6.: Axial distribution of temperature on the thrust chamber inner wall
surfaces. [41]
 103

Table 6.2.: Cooling jacket parameters at key thrust chamber sections.

Chamber Throat Exit
Channel width 3.40 1.35 7.22 mm
Channel depth 1.00 1.00 1.00 mm
Cumulative area 339.69 135.49 722.16 mm2

Film coefficient 5.52 18.47 0.56 kW/m2 /K

Heat flux 16904.82 48902.60 1830.23 kW/m2
Flow velocity 68.97 72.35 8.90 m/s

inner wall, the temperature distribution on the wall surfaces, and the variation of
both pressure and temperature for the coolant are plotted in figures 6.5, 6.6, and
6.7 respectively.

We recognize that the combustion-gas-side wall surface reaches its peak tempera-
tures both at the throat, due to the immense rates of heat flux around this region,

400 10.0

350 9.5

300 9.0
Coolant Temperature (K)

Coolant Pressure (kPa)

250 8.5

200 8.0

150 7.5

100 7.0

50 6.5

0 6.0
0 100 200 300 400 500
Location (mm)

Figure 6.7.: Axial variation of both temperature and pressure for the coolant within
the cooling jacket. [41]
104

as well as near the injector, because of the gradual accretion of coolant temperature
that culminates at this section. Likewise, the coolant-fluid-side wall surface tem-
perature reaches its highest towards the injector end after a small spike near the
throat. Regardless, our maximum wall surface temperature, of 842.84 K, lies well
below the softening temperature of Inconel 625. In addition, the coolant obtains an
outlet pressure of 7772 kPa and a bulk temperature of 301.98 K within the jacket,
thereby transitioning the liquid fuel into a supercritical fluid; methane has a critical
pressure and a critical temperature of 4604 kPa and 190.58 K respectively.

6.1.2 Upper-Stage Thrust Chamber

As stated earlier, the upper-stage thrust chamber employs a regeneratively cooled
section down to an area ratio of 15 and a radiatively cooled section towards aft
of the divergent nozzle extension. Depicted schematically in figure 6.8, such an
arrangement is feasible due to the moderate heat transfer criteria near the nozzle
exit. Because the highest gas temperature in the region is about 1200 K, we opt to
use Niobium C-103—a refractory-metal alloy with 10% hafnium and 1% titanium
constituents—as the wall material for the radiative section. Nb C-103 has a melting
point of 2620 K, an upper service temperature of 1755 K, a thermal conductivity
of 44.6 W/m/K, a density of 8850 kg/m3 and an emissivity of 0.92, in addition to
such properties as high tensile strength, good corrosion resistivity and excellent
formability [61].

Thrust Chamber Axis

Hot Exhaust Gases

Fuel Outlet

Fuel Inlet

Radiative Heat
Regeneratively cooled Section
Radiatively cooled Section

Figure 6.8.: Schematic diagram of the upper-stage regenerative/radiative cooling

thrust chamber.

With a uniform wall thickness of 𝑡𝑤.𝑟𝑎𝑑 = 1.00 mm, the nozzle extension obtains
a peak wall surface temperature of 1584.13 K—a value well within the operating
 105

temperature threshold for the material. The wall surface temperature distribution
for the entire thrust chamber wall is presented in figure 6.9.

1700

1500
Wall Surface Temperature (K)

1300

1100

900

700

500
— Inner surface
300
— Outer surface
100
0 100 200 300 400 500 600 700 800 900 1000 1100
Location (mm)

Figure 6.9.: Axial distribution of temperature on the upper-stage thrust chamber

wall surfaces. [41]

6.2 INJECTOR DESIGN

The quintessential function of an injector is to inject propellants into the combustion

chamber in the correct proportions and the right conditions to yield efficient and
stable combustion. Placed at the forward end of the combustion chamber, the
injector also performs the structural task of closing off the top of the thrust chamber
against the combustion fluid it contains and transferring the generated thrust into
the structure of the rocket. High combustion efficiencies can be obtained from
metered distribution, proper atomization and uniform mixing of the propellants at
their required flow rates and mixture ratios. Combustion stability is also a crucial
requirement for a satisfactory injector design, since “high performance can become
secondary if the injector is easily triggered into destructive instabilities” [13, p.
105]. Stable operation depends in good part on the selection of the injector element
type and the provision for damping any oscillatory phenomena.
The design of an injector, much like other application before, entails numerous
conditions and considerations [12]. The selection of injection elements has largely
106

been dominated by the type of propellants used, the collective properties of the
bipropellant combination, and the conditions for propellant injection prior to igni-
tion. And the placement of the elements on the injector plate has predominantly
been determined by the spatial distribution of spray atomization and vaporization,
the provisions for uniform mass distribution of and adequate mixing interaction
between the propellants after ignition, and the possibility for tailoring the mixture
ratio distribution to minimize chamber wall heat flux. Considering the aforemen-
tioned criteria and with regard to the LOX/LCH4 bipropellant combination, we
examine below the applicability of a few injection elements and manifold configu-
rations for our combustion chamber.

6.2.1 Injection Elements

Depending on the pattern of oxidizer and fuel injection into the combustion cham-
ber, injection elements may broadly be classified into the following types [13] [12]
[62].

Showerhead Injectors
A showerhead element is among the simplest injection elements and typically injects
axial non-impinging streams of either oxidizer or fuel from each orifice into the
combustion chamber. These elements exhibit low rates of mass transfer across the
injector, and rely on turbulence and diffusion between the jets and the combus-
tion fluid in order to achieve adequate atomization and mixing. Because of their
non-uniformity in spray distribution across the injector face and their ineffective
bipropellant atomization and mixing along the chamber length, these elements are
seldom used for primary thrust generation. Their poor atomization and mixing
characteristics, however, make them well suited for film cooling applications, with
the fuel jets acting as coolant between the combustion fluid and the chamber wall.

Impinging Injectors
The like impinging elements accomplish atomization by the self-impingement of
respective jets from oxidizer doublets and fuel doublets. Mixing of the propellants
is achieved downstream of the jet impingement point and takes place as a result
of turbulent intermixing between adjacent oxidizer and fuel spray fans. Effective
atomization occurs as a function of injection velocity, impingement angle and
impingement distance. And attainment of efficient mixing is dependent upon
the relative geometric arrangement of and between oxidizer doublets and fuel
doublets, such as orifice size, spacing between respective fans and fan inclination
 107

angle. Again, like quadlet elements are pairs of like doublets that are canted toward
each other in order to induce better bipropellant mixing and deliver higher com-
bustion performance for similar atomization and vaporization efficiencies and
equal number of impinging pairs.

Unlike the above however, the unlike impinging elements such as unlike doublets
impinge an oxidizer jet directly upon a fuel jet, and achieve virtually all atom-
ization and mixing in the immediate vicinity of the impingement point. Ignition
and combustion of the bipropellant mixture occur closer to the injector face and
result in a relatively high injector plate heat flux. Unlike doublets work best for
propellant combinations which have nearly equal oxidizer and fuel injection orifice
areas and nearly equal injection stream momentum ratios. And again, unlike triplet
elements impinge two opposing jets of one propellant upon a single jet of the other
propellant, and are arranged in oxidizer-fuel-oxidizer or fuel-oxidizer-fuel configu-
rations depending upon properties of the bipropellant mixture and characteristics
such as droplet atomization, spray distribution, mixing uniformity and thermal
compatibility. While unlike impinging elements tend to produce finer atomization
and better mixing than like impinging elements of comparable orifice dimensions
and pressure drops, they tend to exhibit poorer spatial controllability of spray
distribution and higher susceptibility toward combustion instabilities.

Concentric Injectors
A concentric—or coaxial—injection element injects two concentric streams of propel-
lants, with the central jet usually being the oxidizer while the annular jet typically
being the fuel, in a self-impinging conical spray pattern into the combustion cham-
ber. The already atomized and mixed spray from each element further impinges
on sprays from adjacent elements to produce an even finer atomized and better
mixed bipropellant spray. Ignition and combustion of the mixture, much like for
unlike impinging elements, occur fairly close to the impingement point. Because
of their indirect nature of propellant impingement, these elements require pro-
portionately lower fuel-to-oxidizer injection velocity ratios in order for sufficient
injection momentum differences to adequately atomize and mix the propellant
jets. Now, it can be difficult to achieve uniform combustion fluid distribution and
tailored bipropellant mixture ratios across an injector face with concentric injection
elements without evoking combustion instability conditions.

Shear coaxial injectors are typically used with liquid oxidizer and gaseous fuel
combinations since these elements require even larger injection momentum dissim-
ilarities between the propellant jets to satisfactorily shear the streams for effective
108

atomization and mixing. And with their relatively narrow spray exit cone angles,
these elements tend to cause high temperatures near the convergent nozzle section
of a thrust chamber because of incomplete combustion of oxidizer droplets by the
end of the chamber length.
Swirl coaxial injectors, on the other hand, can be used with liquid/liquid, liq-
uid/gaseous or gaseous/gaseous bipropellant combinations, for these elements
induce a swirl motion on either of or both the bipropellant jets, either by tangential
flow into the element or by mechanical channels inside the element, prior to their
injection which helps self-atomize the swirled stream without the added assistance
of the other stream. These elements, because of their rather wide spray cones, can
cause overheating of the forward section of a thrust chamber if positioned too close
to the perimeter of an injector plate due to impingement of the oxidizer spray near
the forward chamber wall.

Hybrid Injectors
A hybrid injection element, such as a pintle injector, is one that combines two or
more of the elements above into a single unit for the injection of bipropellants
into the combustion chamber. In the pintle element, one propellant flows down
the inside of the pintle and is ejected radially through a series of holes or slots
near the tip of the pintle while the other propellant leaves the manifold through
an annular sheet around the base of the pintle. Depending on the propellant
flowing inside the pintle tip, pintle flows are classified as either oxidizer-centered
or fuel-centered patterns. Among the crucial advantages of a pintle injector are its
inherent combustion stability, deep throttling capability, small dribble volume and
face shutoff features.

6.2.2 Design of Injection Elements

Considering the benefits and limitations of each injection element above and re-
ferring to a few successful design studies [60] [63] [64], we employ unlike type
impingement elements for the primary injection of oxidizer and fuel in our project.
Figure 6.10 sketches a typical unlike doublet impingement injection pattern. Per-
haps the most crucial parameter in the design of an injection element is the pro-
pellant pressure drop across the injector. A high pressure drop—although strains
resources within the feed system—increases injection velocity of the propellants,
which not only aids in their atomization and mixing, thereby ensuring a more com-
plete and stable combustion, but also mitigates the inefficiencies associated with
propellant dribbling. We opt for a loss in pressure equal to 25% of the combustion
 109

Fuel
Manifolds

Impingement
Points

Oxidizer
Manifolds Injector
Faceplate

Injection
Ori�ices

Figure 6.10.: Typical injection configuration of an unlike doublet impingement

injector. [12]

chamber pressure, resulting in an injector pressure of 𝑃𝑖𝑛𝑗 = 1.25 𝑃𝑐𝑐 = 7500 kPa
and a pressure drop of 𝛥𝑃 = 1500 kPa across the injection orifice system.

For the flow of an incompressible fluid through hydraulic orifices, the mass flow
rate 𝑚̇ and the injection velocity 𝑣 can be expressed as

𝑚̇ = 𝐶𝑑 𝐴√2 𝛥𝑃 𝜌 (6.1)
𝑣 = 𝐶𝑑 √2 𝛥𝑃/𝜌 (6.2)

where, 𝐶𝑑 is the dimensionless discharge coefficient, 𝐴 the cumulative cross-

sectional area of the orifices, and 𝜌 the propellant mass density. The injection
temperature of the oxidizer is assumed to be about 95 K. And from subsection
6.1.1, we determined the fuel temperature following its passage through the thrust
chamber heat exchanger to be around 305 K. These temperatures (and pressures)
correspond to densities of 𝜌𝑖𝑛𝑗.𝑜 = 1134.26 kg/m3 and 𝜌𝑖𝑛𝑗.𝑓 = 53.17 kg/m3 for the
oxidizer and the fuel respectively.

It is now apparent that a discharge coefficient approaching unity maximizes the

injection velocity for a given pressure drop. Therefore, assuming “smooth well-
rounded entrances to injection holes with clean bores” [12, p. 282], we cautiously
adopt a discharge coefficient of 0.80 for our injector assembly. Doing so gives us
propellant injection velocities of 𝑣𝑖𝑛𝑗.𝑜 = 41.14 m/s and 𝑣𝑖𝑛𝑗.𝑓 = 190.03 m/s, and
thus a fuel-to-oxidizer velocity ratio of 4.62. Since the velocity ratio falls below
the minimum range (of 10) typical for a coaxial injection element [13] [60], our
110

selection of the impingement type injector can be deemed appropriate.

From the propellant mass flow rates of 𝑚̇ 𝑜 = 8.114096 kg/s and 𝑚̇ 𝑓 = 2.950580 kg/s,
we obtain cumulative injection areas of 𝐴𝑖𝑛𝑗.𝑜 = 173.87mm2 and 𝐴𝑖𝑛𝑗.𝑓 = 292.03mm2 .
Now, [62] asserts that for unlike impingement elements, a specific oxidizer-to-fuel
orifice diameter ratio is necessary for an optimum mixing efficiency. This ratio,
when the impingement angle is 60°, can be expressed as

2 𝜌𝑖𝑛𝑗.𝑓 𝑚̇ 𝑜 ⎞ 2 0.7
𝑑
⎜ 𝑜⎞
⎛ ⎟ = 𝑀⎛
⎜ ⎛ ⎞
⎜ ⎟ ⎟ (6.3)
⎝ 𝑑𝑓 ⎠ ⎝ 𝜌𝑖𝑛𝑗.𝑜 ⎝ 𝑚̇ 𝑓 ⎠ ⎠

where, 𝑀 represents the mixing factor for a specific injection element. For 1-on-1
and 2-on-1 unlike element types, the mixing factors equal 1.0 and 1.6 respectively.

Again, [62] affirms that, even for an optimum diameter configuration, the smaller
the orifice diameter, the finer is the droplet size, the higher becomes the vapor-
ization rate and level of mixing, and the better becomes the performance. This
implies maximizing the number of oxidizer and fuel orifices within the feasibility
of fabrication. From references [63] and [64], and with consideration to the above
design aspects (without delving deep into any complications relevant to combus-
tion instabilities), we preliminarily adopt 120 oxidizer orifices, which results ideally
in 96 fuel orifices.

However, by employing a marginally smaller fuel orifice at the expense of a slightly

higher pressure drop (There in fact is the pressure budget available within the fuel
feed system to accommodate this!), we opt to match the number of oxidizer and
fuel orifices to 120 each. Doing so gives us orifice diameters of 𝐷𝑜𝑟𝑖.𝑜 = 1.36 mm
and 𝐷𝑜𝑟𝑖.𝑓 = 1.76 mm. The unlike doublet element sets are arranged in two concen-
tric patterns on the injector faceplate with 80 pairs on the outer ring divided evenly
into two rows of 48 and 32 elements and 40 pairs on the inner ring distributed
again into two rows of 24 and 16 elements. The configuration layout of our injector
is presented in figure 6.11.

Because performance is further enhanced “when the resultant momentum of

impinging streams is approximately axial” [12, p. 283], and since the impingement
angle has been accepted as 60°, our oxidizer and fuel orifices are oriented at 38°
and 22° respectively from the thrust chamber axis towards one another. Any net
spray momentum can effectively be cancelled by directing the injection element
pairs from the inner ring and the outer ring at each other.

In order to abate the adverse effects of jet misimpingement from long partially-
 111

Outer Ring Elements

Inner Ring Elements

Unlike Doublet Set

Oxidizer Orifice

Fuel Orifice

Figure 6.11.: Configuration layout of the injector faceplate. The colours blue and
red represent the oxidizer and fuel orifices respectively. Distance
within the unlike doublet set orifices equals 7.738 mm. Propellant
impingement occurs at 6.528 mm from the injector surface.

disintegrated propellant streams, the ratio of free-stream jet length to injection

orifice diameter ought to be, as [62] suggests, between 5 and 7. With our length-
to-diameter ratios approximately 6 and 4 for the oxidizer and fuel jets respectively,
we obtain an impingement distance of 6.528 mm from the injector faceplate, thus
resulting in a distance of 7.738 mm between centres of the orifices within a doublet
set. Now, to adequately manage heat flux to the injector, we choose Inconel 625 [51]
as the material for the injector assembly. Table 6.3 below summarizes all pertinent
parameters for our injector.

6.3 LAUNCH VEHICLE SPECIFICS

Our study, through the sections covered thus far, has revolved around design
details of the engine assembly and/or its associated subsystems. This section,
instead, encompasses the launch vehicle specifics, such as engine configurations,
mass ratios, mission velocities, and the like.
112

Table 6.3.: Design parameters for the injector faceplate assembly.

Combustion chamber pressure 𝑃𝑐𝑐 6000 kPa
Injector pressure 𝑃𝑖𝑛𝑗 7500 kPa
Oxidizer parameters
Mass flow rate 𝑚̇ 𝑜 8.114 kg/s
Density 𝜌𝑖𝑛𝑗.𝑜 1134.26 kg/m3
Injection velocity 𝑣𝑖𝑛𝑗.𝑜 41.14 m/s
Injection area 𝐴𝑖𝑛𝑗.𝑜 173.87 mm2
Orifices 120
Orifice diameter 𝐷𝑜𝑟𝑖.𝑜 1.36 mm
Orifice angle 𝜃𝑜𝑟𝑖.𝑜 38°
Fuel parameters
Mass flow rate 𝑚̇ 𝑓 2.951 kg/s
Density 𝜌𝑖𝑛𝑗.𝑓 53.17 kg/m3
Injection velocity 𝑣𝑖𝑛𝑗.𝑓 190.03 m/s
Injection area 𝐴𝑖𝑛𝑗.𝑓 292.03 mm2
Orifices 120
Orifice diameter 𝐷𝑜𝑟𝑖.𝑓 1.76 mm
Orifice angle 𝜃𝑜𝑟𝑖.𝑓 22°
Impingement distance 6.528 mm
Orifice distance 7.738 mm
Material Inconel 625

Impingement Point
Oxidizer Jet

22°

Fuel Jet
Momentum of
38° Resultant Jets

Figure 6.12.: Propellant impingement angles and resultant jet momentum.

 113

As stated earlier, the objective of our hypothetical mission is to deliver a light

payload into LEO. With a payload capacity of 200 kg, we target for a perfectly
circular orbit at 500 km altitude. This corresponds to an orbital speed of 7616.6 m/s
for our payload. In order to accommodate losses due to gravity and drag during
the flight, we add a factor of 1.25 times the orbital speed and obtain a mission
velocity “delta-v” of about 9500 m/s.

In section 5.2, we have established mass flow rate of 𝑚̇ = 11.064676kg/s and average
exhaust velocities of 𝑣𝑒.1st = 2700.486590 m/s and 𝑣𝑒.2nd = 3335.262440 m/s for our
thrust chamber variants. (Here, the lesser quantity between the average exhaust
velocity 𝑣𝑒 and the effective exhaust velocity 𝑐 is considered the exhaust velocity.)
We initially ascertain delta-v and mass metrics for the upper stage and finally move
on to those of the booster stage with its engine cluster configuration.

Now, launch vehicles typically have structural mass fractions between 10 to 15%
for the upper stage and between 5 to 10% for the booster stage to account for
the inert hardware exclusive to the stage [12]. Structural mass fraction is usually
higher for the upper stage than for the booster stage because the net propellant
mass is significantly higher for the latter than for the former [65, ch. 13]; with
larger tanks having better volume-to-surface ratios, there is less tankage mass per
unit propellant mass for the booster stage. We, therefore, conservatively adopt
structural mass fractions of 𝜀2nd = 0.125 and 𝜀1st = 0.075 which correspond to
propellant mass fractions of 𝜁2nd = 87.5% and 𝜁1st = 92.5% for both our rocket
stages respectively.

The rocket equation for a twin-stage launcher is expanded and expressed as [12]

1 1
𝛥𝑣 = 𝑣𝑒.1st 𝑙𝑛 ( ) + 𝑣𝑒.2nd 𝑙𝑛 ( ) (6.4)
MR1st MR2nd

Equation 6.4 can be expressed in terms of structural mass fraction as [65, ch. 13]

𝑣𝑒.1st 𝜂 − 1 𝑣 𝜂−1
𝛥𝑣 = 𝑣𝑒.1st 𝑙𝑛 ( ) + 𝑣𝑒.2nd 𝑙𝑛 ( 𝑒.2nd ) (6.5)
𝑣𝑒.1st 𝜀1st 𝜂 𝑣𝑒.2nd 𝜀2nd 𝜂

where, 𝜂 represents the Lagrange multiplier and can be solved iteratively.

From the mass flow rate and for a given burn time, we attain the stage propellant
mass as
𝑚𝑝𝑟𝑜 = 𝑚̇ 𝑡𝑏 (6.6)

Using the above propellant and structural mass fractions, we obtain both wet and
114

dry stage masses as

1
𝑚𝑤𝑒𝑡 = 𝑚𝑝𝑟𝑜 (6.7a)
𝜁
𝑚𝑑𝑟𝑦 = 𝜀 𝑚𝑤𝑒𝑡 (6.7b)

For the operation duration of each stage, we follow standard flight profiles for
small launchers (such as that of Electron [7])—wherein the upper stage operates
anywhere upwards of 180 s, whereas the booster stage operates anywhere up to
180 s. And in order to generate the necessary liftoff/ascent thrust, we require a
cluster of thrust chambers in a parallel burn arrangement for the first stage. In
addition, we add in 15 s of propellant margin to each engine to account for start and
stop transients during startup and shutdown operations. By cycling the nominal
engine burn time between 180 s, 240 s and 300 s for the upper stage and between
120 s, 150 s and 180 s for the booster stage while employing a cluster of 5, 7 and 9
thrust chambers therein, we determine both the liftoff mass and the delta-v for our
launch vehicle using the equations above. Results of the calculations are presented
in table 6.4.

It is observed that certain combinations of burn times and cluster configurations

are downright unfeasible in terms of delta-v, while others yield a rather exorbitant
liftoff mass. Further, a few combinations offer superfluous mission velocities at
the expense of unwarranted launch vehicle mass. Any combination with a delta-v
below 9500 m/s or a liftoff mass above 20 t or a 2% excess delta-v is disregarded,
and thus greyed out in our tabulation. (The relative 𝛥𝑣 therein corresponds to the
minimum mission velocity, i.e., 9500 m/s.)

We decide to allow some margin (of nearly +1.5%) for vehicle performance with
regard to delta-v and select a nominal burn time of 𝑡𝑏.2nd = 180 s for the upper
stage and of 𝑡𝑏.1st = 120 s for the booster stage with a cluster of 9 thrust chambers.
Accordingly, the total propellant budget is of 195 s and 135 s for both the stages.
Correspondingly, we obtain a mass of around 𝑚𝐿𝑉 = 17200 kg and a delta-v of
about 𝛥𝑣𝐿𝑉 = 9636.6 m/s for our launch vehicle. The delta-v budget is thereby
distributed as 5527.7 m/s for the second stage and 4108.9 m/s for the first stage.
Configuration of the engine cluster on our booster is portrayed in figure 6.13.

6.4 SUMMARY

Having designed the thrust chamber configurations for our twin-staged rocket, we
 115

Table 6.4.: Preliminary mass and delta-v values of our launch vehicle for the nomi-
nal burn times at the structural mass fractions considered.
𝑡𝑏.2nd (s) 𝑡𝑏.1st (s) 𝑁1st 𝑚𝐿𝑉 (kg) 𝛥𝑣𝐿𝑉 (m/s) 𝛥𝑣 (%)

180 120 5 10740.065 8737.897657 − 8.02

7 13969.755 9254.994660 − 2.58
9 17199.444 9636.616163 + 1.44
150 5 12534.337 9046.236315 − 4.78
7 16481.735 9560.492737 + 0.64
9 20429.133 9932.552578 + 4.55
180 5 14328.609 9302.881990 − 2.07
7 18993.716 9809.524642 + 3.26
9 23658.822 10170.068158 + 7.05

240 120 5 11498.786 8623.640510 − 9.22

7 14728.475 9134.503122 − 3.85
9 17958.165 9520.982790 + 0.22
150 5 13293.058 8926.512933 − 6.04
7 17240.456 9443.232184 − 0.60
9 21187.854 9826.458687 + 3.44
180 5 15087.330 9182.552216 − 3.34
7 19752.436 9698.835948 + 2.09
9 24417.543 10075.459234 + 6.06

300 120 5 12257.507 8510.705225 − 10.41

7 15487.196 9009.029692 − 5.17
9 18716.885 9393.550566 − 1.12
150 5 14051.778 8804.780770 − 7.32
7 17999.177 9315.668708 − 1.94
9 21946.575 9702.146594 + 2.13
180 5 15846.050 9056.480901 − 4.67
7 20511.157 9572.711413 + 0.77
9 25176.264 9956.830522 + 4.81

(+ 15 s) (+ 15 s)
116

325.548 mm

∅251.744 mm

425.349 mm
∅1250 mm

Figure 6.13.: Cluster configuration of the booster stage engines.

delve into analysing the heat transfer and mitigating the excessive thermal energy
by implementing steady-state cooling techniques. Our one-dimensional analyses
are performed only at the most critical wall sections, such as around key regions
of the chamber, at and near the throat, and towards the exit of the nozzle. We
employ regenerative cooling for the combustion chamber and the nozzle down to
an area ratio of 15, i.e., encompassing entirety of the booster-stage engine, beyond
which we employ radiative cooling owing to the relatively moderate temperatures
therein, i.e., towards aft of the upper-stage nozzle.

In our regeneratively cooled thrust chamber, the main fuel feed is circulated as
the coolant fluid within the cooling jacket. We inject the coolant through the inlet
manifold at the nozzle exit from where it flows unidirectionally upward within
the longitudinal channels following which it enters into the injector manifold.
We adopt a milled-slot jacket design and choose Inconel 625 as the wall material
for both the inner jacket and the outer shell. With heat flux being the highest
at the throat, the cooling channels at the region are designed for the maximum
flow velocity and the minimum cross-sectional area. For our radiatively cooled
nozzle segment, we use Niobium C-103 as the wall material. Our mechanisms
 117

and materials are, as a consequence, perfectly compatible to withstand the high

thermal energy generated within our propulsion system.
Next, we explore design principles of the injector and look into the process of
metering, injection, atomization, vaporization and mixing of propellants for their
effective combustion. High combustion efficiency and good combustion stabil-
ity can be obtained from metered distribution, proper atomization and uniform
mixing of the propellants at their required flow rates and mixture ratios. Having
considered quite a few injection elements, we select the unlike type impingement
elements for our thrust chamber design, wherein an oxidizer jet is impinged di-
rectly upon a fuel jet and virtually all atomization and mixing is achieved in the
immediate vicinity of the impingement point; ignition and combustion occur closer
to the injector face, which results in a relatively high injector plate heat flux. The
unlike doublet element sets are arranged in two concentric patterns with 80 pairs
on the outer ring divided evenly into two rows of 48 and 32 elements and 40 pairs
on the inner ring distributed again into two rows of 24 and 16 elements.
In the end, we strive to determine the tentative liftoff mass of our launch vehicle
from mass flow rates and average exhaust velocities of both our engines as well as
with rational assumptions of structural mass fractions for each stage. Following
standard flight profiles for small launchers, we obtain the operation duration for
both our rocket stages. And using the minimum mission velocity necessary, we
conclude to employ a cluster of 9 engines in a parallel burn arrangement on the
booster stage of our rocket.
7
Conclusion

The objective of our project has been to preliminarily develop a “new generation”
propulsion system that could be employed in both the booster stage and the upper
stage of a small-lift orbital-class rocket to deliver a light payload into LEO. The
launch vehicle for our hypothetical mission is to have a delta-v budget of 9500 m/s
and a payload capacity of 200 kg. And above all, both the liquid rocket engines are
obligated to be as similar to each other as pragmatically possible. Such an engine
configuration for a launch vehicle closely mirrors that of Rutherford/Electron from
Rocket Lab or of Merlin/Falcon 9 from SpaceX.

While the conception of engine systems driven by electric pump feed systems or
running on the liquid oxygen and liquid methane mixture is anything but new,
what is novel about our study is the amalgamation of these two conceptualizations
to design one single propulsion system that can outperform current rocket en-
gines in the comparable performance class. In that regard, we have unequivocally
demonstrated the feasibility of our liquid rocket propulsion system to accomplish
a typical real-life mission, such as the one outlined in our objectives.

7.1 CONCLUSION

We, at first, looked into the theoretical notions of ideal propulsion systems along
with the means to incorporate losses to the incurred over-performance through
realistic correction factors. In our quest to select the best bipropellant combination
for our mission, we investigated a multitude of propellant characteristics critical
to engine performance. In particular, we analysed the combustion of LOX with
LH2, RP-1 and LCH4 using CEA for a frozen composition flow, and determined
120

the best mixture ratio to yield the highest specific impulse: for hydrolox, kerolox
and methalox bipropellants, the optimum mixture ratios are 5.50, 2.25 and 2.75
respectively. We also examined the variation of combustion temperature, character-
istic velocity and molecular mass as functions of mixture ratio. For the LOX/LCH4
combination at its optimum mixture ratio, we obtained a chamber temperature of
3350.27 K at a chamber pressure of 6894.757 kPa.

With regard to the propellant feed system, we determined fairly quickly that a pres-
sure feed system would be anything but suitable for our engine specifications. We
concentrated our effort studying the pump feed system instead. Having skimmed
over a few pump-fed-engine cycles, we considered the gas generator cycle as the
yardstick for turbopump system performance. With electropump systems, we in-
vestigated the significance of both power density and energy density in governing
the battery pack mass. Since the mass penalty from battery packs was historically a
major bottleneck in the development of electric-pump-fed rocket engines, adoption
of such a system has only just become feasible owing to recent advancements in
battery technologies. Electropump feed systems at present, therefore, offer compa-
rable performance to—or even can outperform—gas generator cycle engines under
the right conditions. In our study, where the thrust is 30 kN and the combustion
chamber pressure is 6000 kPa, we realized that the electropump system yielded
lesser mass, thereby better mass ratio, than the turbopump feed system for nominal
engine burn times above 88 s; and the switch from critical power density to critical
energy density for our battery pack occurred around 150 s.

Next, we designed two thrust chamber versions for our launch vehicle—a sea-level
variant with a nozzle expansion area ratio of 15 for the booster stage and a vacuum-
optimized variant with an area ratio of 100 for the upper stage. We obtained a few
crucial thrust chamber performance parameters from CEA and calculated the rest
with MATLAB using the theoretical equations from chapter 2. In order to account
for a slight over-performance, we incorporated somewhat conservative correction
factors and worked out all our results again. Consequently, we attained a mass flow
rate of 11.065 kg/s and a throat diameter of 65 mm. For our convergent-divergent
nozzle contours, we examined the expansion and shock characteristics of the
exhaust jet based on TIC, TOC, TOP, and the method of characteristics techniques.
We subsequently adopted the TOP “thrust optimised parabolic” contour for both
the nozzle variants, and ascertained all pertinent internal dimensions therein.

Thereafter, we investigated heat transfer from the combustion fluid to the thrust
chamber walls; we employed steady-state mitigation methods to manage the high
thermal energy within and obviate any failure. We implemented regenerative cool-
 121

ing, using Inconel 625, for the combustion chamber and the nozzle down to an area
ratio of 15, i.e., encompassing entirety of the booster-stage engine, beyond which
we implemented radiative cooling, with Niobium C-103, owing to the relatively
moderate temperatures, i.e., towards aft of the upper-stage nozzle. Further, we
designed the injector assembly to inject propellants in the correct proportions and
the right conditions so as to yield efficient and stable combustion. Our injector
faceplate, also of Inconel 625, comprised two concentric patterns of unlike impinge-
ment doublet sets—with 80 pairs on the outer ring divided evenly into two rows of
48 and 32 elements and 40 pairs on the inner ring distributed again into two rows
of 24 and 16 elements.

We, at last, strived to evaluate the liftoff mass of our launch vehicle from per-
formance parameters of both the thrust chamber configurations as well as with
rational assumptions of structural mass fractions for each stage: 0.075 being for
the first and 0.125 for the second. In doing so, and following standard flight pro-
files, we settled for nominal operation durations of 120 s and 180 s (plus 15 s each
to accommodate transient conditions) for the booster stage and the upper stage
respectively; both our burn times are, as compared to electropump systems, well
within the optimal envelope for electropump feed system performance. With a
cluster of 9 engines on the booster, our hypothetical launch vehicle has a mass
of roughly 17200 kg (200 kg of which is the payload) and a delta-v of approxi-
mately 9600 m/s—quite within the desirable range of specifications for small-lift
orbital-class twin-stage rockets of today. And thus concludes the preliminary
design of our propulsion system.

7.2 LIMITATIONS & FUTURE WORK

Limitations
Having accomplished our objective, we do recognize that our design has, much
like any other preliminary design, certain implicit shortcomings. First, ours is a
theoretical study based on idealistic models of the principles/processes involved
in liquid rocket propulsion systems. We adopt our crucial (and in a way initial)
performance figures, viz. combustion chamber temperature, effective molecular
mass and specific heat ratio, from CEA with the assumption that NASA CEA
does not overestimate its values; that need not necessarily be so. Likewise, RPA
generates results, both operational and analytical, that could be a touch optimistic.
Adopting perfect correction factors, consequently, is perhaps a tricky endeavour.
Without any empirical data to back our design, this study is to be deemed but
122

provisional.

Second, while we have proven the viability of electric-pump-fed engine cycles to

outperform both pressure feed systems and turbopump feed systems, the outper-
formance only exists within a narrow window of engine operating conditions. Put
another way, above or below a certain combustion chamber pressure or a given
burn time or a specific thrust, a pressure-fed system or a turbopump-fed system
instead will reign supreme. Moreover, the batteries at present are still quite limited
by the technologies of today. Nonetheless, with advances in battery technologies
lately, there is room for improvement of both power density and energy density
so as to save mass and/or enhance performance. Regardless, it will perhaps be a
very long time (if not downright impossible) before batteries come anywhere near
matching the energy densities of fossil fuels.

And third, throughout our design process, we have performed numerous iterations
to various parameters whenever we realized the need of our propulsion system to
attain and/or assume values and/or details necessary for the mission objective. Our
reconsiderations might have been, at times, somewhat biased towards achieving
a favourable design outcome. An instance of this would be the structural mass
fraction of each stage to obtain a certain launcher liftoff mass. Another similar
example would likely be the selection of nozzle contours for both our engines. But
because little could be done to verify the validity of our theoretical design, we
regard all our assumptions perfectly sound in this design study.

Future Work
The ultimate realization of the task initiated in this project is, quite unequivocally,
the flight qualification of our “new generation” propulsion system for a small-
lift orbital-class launch vehicle like ours. A more prudent target would perhaps
be to hot-fire test our rocket engine after finalising the preliminary design. To
accomplish that, we ought to perform at least the following tasks:

▪ Conduct computational fluid dynamics analyses for the combustion fluid

characteristics to preclude instabilities and model stable engine operating
conditions.

▪ Design the rotodynamic pump elements, like the stator assembly, the rotor
assembly, inducer, impeller, diffuser vanes and volute for our propellant feed
properties.

▪ Develop the remainder systems and/or components necessary for engine

operation, such as detailed designs for injector manifolds, ignition devices,
 123

propellant feed assembly, engine control system, thrust vector control, to list
a few.
▪ Perform structural and thermal analyses using the more appropriate tools.
▪ Build and test the liquid rocket propulsion system.

With this, we conclude the objective of this project and the scope of my thesis.
A
CEA Computation

This appendix presents the output of the CEA analysis [39] on the basis of which
we perform all our design calculations.

NASA-GLENN CHEMICAL EQUILIBRIUM PROGRAM CEA2, FEBRUARY 5, 2004

BY BONNIE MCBRIDE AND SANFORD GORDON
REFS: NASA RP-1311, PART I, 1994 AND NASA RP-1311, PART II, 1996

****************************************************************

# Problem Type: "Rocket" (Infinite Area Combustor)

prob case=METHALOX_6000_2.750 ro frozen
# Pressure (1 value):
p,bar= 60
# Supersonic Area Ratio (2 values):
supar= 15, 100
# Oxidizer/Fuel Wt. ratio (1 value):
o/f= 2.75
# You selected the following fuels and oxidizers:
reac
fuel CH4(L) wt%=100.0000
oxid O2(L) wt%=100.0000
# You selected these options for output:
output short
output massf
output siunits
# The following line is the end of your CEA input file!
end
126

THEORETICAL ROCKET PERFORMANCE ASSUMING FROZEN COMPOSITION

Pin = 870.2 PSIA

REACTANT WT FRACTION ENERGY TEMP

(SEE NOTE) KJ/KG-MOL K
FUEL CH4(L) 1.0000000 -89233.000 111.643
OXIDANT O2(L) 1.0000000 -12979.000 90.170

O/F= 2.75000 %FUEL= 26.666667

R,EQ.RATIO= 1.450641 PHI,EQ.RATIO= 1.450641

CHAMBER THROAT EXIT EXIT

Pinf/P 1.0000 1.7781 150.46 1962.05
P, BAR 60.000 33.744 0.39878 0.03058
T, K 3338.67 3022.92 1316.75 748.18
RHO, KG/CU M 4.2155 0 2.6185 0 7.1040-2 9.5875-3
H, KJ/KG -1780.73 -2560.60 -6489.75 -7597.78
U, KJ/KG -3204.04 -3849.31 -7051.10 -7916.74
G, KJ/KG -45034.4 -41723.7 -23548.7 -17290.7
S, KJ/(KG)(K) 12.9554 12.9554 12.9554 12.9554
M, (1/n) 19.503 19.503 19.503 19.503
Cp, KJ/(KG)(K) 2.4856 2.4533 2.0726 1.8173
GAMMAs 1.2070 1.2103 1.2590 1.3065
SON VEL,M/SEC 1310.7 1248.9 840.7 645.5
MACH NUMBER 0.000 1.000 3.651 5.284

PERFORMANCE PARAMETERS
Ae/At 1.0000 15.000 100.00
CSTAR, M/SEC 1834.8 1834.8 1834.8
CF 0.6807 1.6726 1.8590
Ivac, M/SEC 2280.8 3251.8 3504.4
Isp, M/SEC 1248.9 3068.9 3410.9

MASS FRACTIONS
*CO 0.34809 *CO2 0.18457 COOH 0.00002
*H 0.00110 HCO 0.00003 HO2 0.00002
*H2 0.01741 HCOOH 0.00001 H2O 0.42142
H2O2 0.00001 *O 0.00148 *OH 0.02273
*O2 0.00312

* THERMODYNAMIC PROPERTIES FITTED TO 20000.K

NOTE. WEIGHT FRACTION OF FUEL IN TOTAL FUELS AND OF OXIDANT IN
TOTAL OXIDANTS
 B
MATLAB Script

This appendix presents the MATLAB code snippets, together with their output,
used to compute our design parameters. The code is divided into three sections:
first, for the thrust chamber performance; second, for the pump feed system, and
third, for the rocket mass and delta-v metrics.

B.1 THRUST CHAMBER PARAMETERS

1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2
3 function [stage] = ThrustChamberParameters()
4
5 G0 = 9.80665; %% Earth Standard Gravity
6 RSTAR = 8.314462618; %% Universal Gas Constant
7 rho_O = 1142.0; %% Density Oxidizer
8 rho_F = 0422.8; %% Density Fuel
9 P_cc = 6000000; %% Combustion Chamber Pressure
10 OF = 2.75; %% O/F Mixture Ratio
11 F_nomi = 0;
12 C_L = 100; %% Characteristics Lines
13 MLNC = 0; %% Function Switch
14 expn = 0; %% Nozzle Expansion Area Ratio
15 P_e = 0;
16 P_ext = 0;
17 T_cc = 3338.67; %% Combustion Chamber Temperature
18 MM = 19.503; %% Effective Molecular Mass
19 k = 1.2070; %% Specific Heat Ratio
20 k_t = 1.2103; %% Specific Heat Ratio _ Throat
 128

21 rho_cc = 4.2155; %% Combustion Chamber Fluid Density

22 D_t = 60.00;
23 D_t = 65.00;
24 % D_t = 64.465918; %% optimized for 30kN uncorrected
25 % D_t = 65.130381; %% optimized for 30kN corrected
26
27 alpha = 15;
28 beta = 30;
29 cntr = 4.0; %% Nozzle Contraction Area Ratio
30 L_STAR = 0.70; %% Characteristic Length
31 corrfact_c_STAR = 0.99;
32 corrfact_C_f = 0.97;
33 corrfact_m_DOT = 1.01;
34
35 DS = 275000000; %% Design Strength = Yield @ Temp
36 FS = 1.3750000; %% Safety Factor = 0.727272727273
37 chnl_num = 100; %% Coolant Channel Number
38 dpth = 1.000; %% Coolant Jacket Depth
39 MAWT = 1000.0; %% Maximum Allowable Wall Temperature
40 DeltaP = 0.25; %% Pressure Drop
41 rho_inj_O = 1134.2600; %% Density @ IM K & 7.50 MPa
42 rho_inj_F = 0053.1685; %% Density @ IM K & 7.50 MPa
43 ori_num_O = 120; %% D = 1.36
44 ori_num_F = 120; %% D = 1.76
45 angle_ori_O = 38;
46 angle_ori_F = 22;
47 dis_coeff = 0.80; %% Discharge Coefficient
48 MixFac = 1.0; %% Unlike 1-1 Mixing Factor // 0.696
49 FACNUM = 4; %% Factor Number
50 LENFAC = 0; %% Length Factor
51
52 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
53
54 clc;
55
56 fprintf('\n\n');
57 fprintf('\tRocket Stages:\n\n');
58 fprintf('\t\t1\t\tFIRST\tBOOSTER\t •\n');
59 fprintf('\t\t2\t\tSECOND\tUPPER\t ••\n');
60 fprintf('\n\t\tPlease enter your stage… ');
61 stage = input('', 's');
62 % stage = '1';
63 % stage = '2';
64 % stage = 'AA'; %% Select! Stage! Here! Now! Go!
 129

65 if (strcmpi(stage, '1') == 1)
66 stage = '1ST';
67 stagename = 'FIRST "BOOSTER" STAGE' •;
68 expn = 15;
69 k_e = 1.2590;
70 P_e = 39878;
71 P_ext_aa = 101325.0;
72 P_ext_zz = 2.067920;
73 div_i = 28;
74 div_e = 10;
75 end
76
77 if (strcmpi(stage, '2') == 1)
78 stage = '2ND';
79 stagename = 'SECOND "UPPER" STAGE' ••;
80 expn = 100;
81 k_e = 1.3065;
82 P_e = 03058;
83 P_ext_aa = 0.886280;
84 P_ext_zz = 000000.0;
85 div_i = 34;
86 div_e = 08;
87 end
88
89 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
90
91 P_ext = P_ext_zz;
92 P_ext = P_ext_aa;
93 % % P_ext = P_e;
94 A_t = pi / 4 * D_t ^ 2 / 1000000;
95 D_e = D_t * sqrt(expn);
96 A_e = pi / 4 * D_e ^ 2 / 1000000;
97 corrfact_c = corrfact_c_STAR * corrfact_C_f;
98 corrfact_F = corrfact_c_STAR * corrfact_C_f * corrfact_m_DOT;
99 corrfact_n = corrfact_c * corrfact_c ;
100 M = MM / 1000;
101 R = RSTAR / M;
102 c_STAR = (sqrt(k * R * T_cc)) / (k * sqrt((2 / (k + 1)) ...
103 ^ ((k + 1)/(k - 1))));
104 c_STAR = c_STAR * corrfact_c_STAR;
105 C_f_opt = sqrt(2 * k * k / (k - 1) * (2 / (k + 1)) ...
106 ^ ((k + 1) / (k - 1)) * (1 - (P_e / P_cc) ^ ((k - 1) / k)));
107 C_f = C_f_opt + (P_e - P_ext) / P_cc * expn;
108 C_f = C_f * corrfact_C_f;
 130

109 m_DOT = P_cc * A_t / c_STAR;

110 m_DOT = m_DOT * corrfact_m_DOT;
111 c = c_STAR * C_f;
112 I_s = c_STAR * C_f / G0;
113 F = P_cc * C_f * A_t * corrfact_m_DOT;
114 F_DASH = F * corrfact_F;%% => F' = F" × corrfact_F ^ 2
115 m_DOT_DASH = 30000 / c;
116 % m_DOT_DASH = F / c;
117 % m_DOT_DASH = F_DASH / c;
118 P_t = P_cc * (2 / (k_t + 1)) ^ (k_t / (k_t - 1));
119 T_t = T_cc / ((P_cc / P_t) ^ ((k_t - 1) / k_t));
120 T_e = T_cc / ((P_cc / P_e) ^ ((k_e - 1) / k_e));
121 SV_cc = R * T_cc / P_cc;
122 SV_t = SV_cc * (P_cc / P_t) ^ (1 / k_t);
123 SV_e = SV_cc * (P_cc / P_e) ^ (1 / k_e);
124 v_cc = sqrt(2 * k / (k - 1) * R * T_cc * (1 - (P_cc / P_cc) ...
125 ^ ((k - 1) / k)));
126 v_t = sqrt(2 * k_t / (k_t - 1) * R * T_cc * (1 - (P_t / P_cc) ...
127 ^ ((k_t - 1) / k_t)));
128 v_e = c - (P_e - P_ext) * A_e / m_DOT;
129 sv_e = sqrt(k_e * R * T_e);
130 M_e = v_e / sv_e;
131
132 Cline = C_L;
133 Gamma = k_e;
134 Mexit = M_e;
135 Stage = stage;
136 if (MLNC > 0)
137 MLNC = MinimumLengthNozzleContour(Cline, Gamma, Mexit, Stage);
138 end
139 velout = min(c, v_e);
140
141 L_N = (D_e - D_t) / tand(alpha) / 2;
142 L_n = 0.8 * L_N;
143 V_cc = L_STAR * A_t;
144 D_cc = D_t * sqrt(cntr);
145 A_cc = pi / 4 * D_cc ^ 2 / 1000000;
146 L_f = (D_cc - D_t) / tand(beta) / 2;
147 V_f = pi / 24 / tand(beta) * (D_cc * D_cc * D_cc - D_t * D_t * D_t);
148 V_c = V_cc * 1000000000 - V_f;
149 L_c = V_c / (pi / 4 * D_cc * D_cc);
150 L_cc = L_c + L_f;
151 L = L_n + L_cc;
152 t_s = V_cc * rho_cc / m_DOT;
 131

153 m_DOT_O = 1 / (OF + 1) * m_DOT * (OF);

154 m_DOT_F = 1 / (OF + 1) * m_DOT * 1.00;
155
156 MAWS = DS / FS; %% Maximum Allowable Working Stress
157 thck_inn = (D_cc * 0.50) * (P_cc * 0.50) / (MAWS * 1.00) * (1.00);
158 thck_inn = round(thck_inn);
159 thck_out = (D_cc * 0.50) * (P_cc * 1.50) / (MAWS * 2.00) * (2.00);
160 thck_out = round(thck_out);
161 thck_rib = thck_inn * 0.75;
162 thck_cc = thck_inn;
163 thck_t = thck_inn;
164 thck_e = thck_inn;
165 crcm_cc = 2 * pi * (D_cc / 2 + thck_cc);
166 crcm_t = 2 * pi * (D_t / 2 + thck_t);
167 crcm_e = 2 * pi * (D_e / 2 + thck_e);
168 chnl_wdth_cc = crcm_cc / chnl_num - thck_rib;
169 chnl_wdth_t = crcm_t / chnl_num - thck_rib;
170 chnl_wdth_e = crcm_e / chnl_num - thck_rib;
171 chnl_dpth_cc = dpth;
172 chnl_dpth_t = dpth;
173 chnl_dpth_e = dpth;
174
175 DeltaP = P_cc * DeltaP;
176 P_inj = P_cc + DeltaP;
177 A_inj_O = m_DOT_O / (dis_coeff * sqrt(2 * DeltaP * rho_inj_O));
178 A_inj_F = m_DOT_F / (dis_coeff * sqrt(2 * DeltaP * rho_inj_F));
179 A_ori_O = A_inj_O / ori_num_O;
180 A_ori_F = A_inj_F / ori_num_F;
181 D_ori_O = sqrt(A_ori_O * 4 / pi) * 1000;
182 D_ori_F = sqrt(A_ori_F * 4 / pi) * 1000;
183 v_inj_O = dis_coeff * sqrt(2 * DeltaP / rho_inj_O);
184 v_inj_F = dis_coeff * sqrt(2 * DeltaP / rho_inj_F);
185 v_inj_O = m_DOT_O / rho_inj_O / A_inj_O;
186 v_inj_F = m_DOT_F / rho_inj_F / A_inj_F;
187 D_ori_OF = sqrt(MixFac * ((rho_inj_F / rho_inj_O) ...
188 * (m_DOT_O / m_DOT_F) ^ 2) ^ (0.7));
189 angle_inj = atand(((m_DOT_O * v_inj_O * sind(angle_ori_O)) ...
190 - (m_DOT_F * v_inj_F * sind(angle_ori_F))) ...
191 / ((m_DOT_O * v_inj_O * cosd(angle_ori_O)) ...
192 + (m_DOT_F * v_inj_F * cosd(angle_ori_F))));
193 LENFAC = (D_ori_F * cosd(angle_ori_F)) / (D_ori_O * cosd(angle_ori_O));
194 LENFAC = LENFAC * FACNUM;
195 len_O = D_ori_O * LENFAC;
196 len_F = D_ori_F * FACNUM;
 132

197 hei_O = cosd(angle_ori_O) * len_O;

198 hei_F = cosd(angle_ori_F) * len_F;
199 dis_O = sind(angle_ori_O) * len_O;
200 dis_F = sind(angle_ori_F) * len_F;
201 lenfac_O = len_O / D_ori_O;
202 lenfac_F = len_F / D_ori_F;
203 dis_oriinj = dis_O + dis_F;
204 dis_imping = max(hei_O, hei_F);
205
206 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
207
208 stage = 'ZZ';
209
210 return;
211
212 end
213
214 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 133

1st Stage Thrust Chamber Output

1 >>
2
3 FIRST "BOOSTER" STAGE •
4
5 --------------------------------------------------
6
7 cofa_c* = 99.000 […]
8 cofa_Cf = 97.000 […]
9 cofa_mDOT = 101.000 […]
10 cofa_c = 96.030000 […]
11 cofa_F = 96.990300 […]
12 cofa_n = 92.217609 […]
13 R = 426.317111 […]
14 c* = 1817.399932 […]
15 Cf = 1.485907 […]
16 c = 2700.486590 […]
17 Is = 275.372996 […]
18 F = 29880.009851 […]
19 F' = 28980.711195 […]
20 mDOT = 11.064676 […]
21 mDOT.O = 8.114096 […]
22 mDOT.F = 2.950580 […]
23 mDOT' = 11.109109 […]
24
25 vinj.O = 41.142844 […]
26 vinj.F = 190.030543 […]
27 Ainj.O = 173.873427 […]
28 Ainj.F = 292.031450 […]
29 Aori.O = 1.448945 […]
30 Aori.F = 2.433595 […]
31 Dori.O = 1.358254 […]
32 Dori.F = 1.760270 […]
33 θinj = -0.330 […]
34 θ.O = 38 […]
35 θ.F = 22 […]
36 DSoriinj = 7.738160 […]
37 DSimping = 6.528375 […]
38 FAC.O = 6.099467 […]
39 FAC.F = 4.000000 […]
40 LEN.O = 8.28 […]
41 LEN.F = 7.04 […]
42 HEI.O = 6.528375114945 […]
43 HEI.F = 6.528375114945 […]
 134

44
45 Dcc = 130.000000 […]
46 Rcc = 65.000000 […]
47 CRCMcc = 414.690230 […]
48 Acc = 0.013273228961 […]
49 Lc = 142.163203 […]
50 Lf = 56.291651 […]
51 Lcc = 198.454855 […]
52 Pcc = 6000000.000000 […]
53 Tcc = 3338.670000 […]
54 SVcc = 0.237222 […]
55 vcc = 0.000000 […]
56 Vcc = 0.002322815068 […]
57 ts = 0.000885 […]
58
59 Dt = 65.000000 […]
60 Rt = 32.500000 […]
61 CRCMt = 210.486708 […]
62 At = 0.003318307240 […]
63 Pt = 3374871.418511 […]
64 Tt = 3021.010723 […]
65 SVt = 0.381617 […]
66 vt = 1248.501396 […]
67 θcon = 30 […]
68 θd.i = 28 […]
69 θd.e = 10 […]
70
71 De = 251.743918 […]
72 Re = 125.871959 […]
73 CRCMe = 797.160027 […]
74 Ae = 0.049774608605 […]
75 Pe = 39878.000 […]
76 Te = 1190.248514 […]
77 SVe = 12.724392 […]
78 ve = 2976.906822 […]
79 Me = 3.724493 […]
80 Pext = 101325.000 […]
81
82 LN = 348.468894 […]
83 Ln = 278.775115 […]
84 Rtu = 48.750 […]
85 Rtd = 13.000 […]
86
87 L = 477.229970 […]
 135

88 THCK.i = 1.000 […]

89 THCK.o = 3.000 […]
90 DPTHcc = 1.000 […]
91 DPTHt = 1.000 […]
92 DPTHe = 1.000 […]
93 WDTHcc = 3.397 […]
94 WDTHt = 1.355 […]
95 WDTHe = 7.222 […]
96
97 VelOut = 2700.486590391 […]
98 L* = 0.700000 […]
99 Expn = 15.000
100 Cntr = 4.000
101 MAWS = 200.000 […]
102 Pcc/Pe = 150.458900
103 Pcc/Pext = 59
104 Pe/Pext = 0.393565
105 Pe-Pext = -61447.000
106 F../mDOT = 0.266667
107 O../mDOT = 0.733333
108 De/Dcc = 1.936492
109 De/Dt = 3.872983
110 Lcc/Dcc = 1.526576
111 Lc/Lf = 2.525476
112 ..F/..O = 4.618799
113 Acc/..O = 76.338456
114 Acc/..F = 45.451368
115
116 --------------------------------------------------
117
118 >>
 136

2nd Stage Thrust Chamber Output

1 >>
2
3 SECOND "UPPER" STAGE ••
4
5 --------------------------------------------------
6
7 cofa_c* = 99.000 […]
8 cofa_Cf = 97.000 […]
9 cofa_mDOT = 101.000 […]
10 cofa_c = 96.030000 […]
11 cofa_F = 96.990300 […]
12 cofa_n = 92.217609 […]
13 R = 426.317111 […]
14 c* = 1817.399932 […]
15 Cf = 1.885631 […]
16 c = 3426.945580 […]
17 Is = 349.451197 […]
18 F = 37918.043388 […]
19 F' = 36776.824036 […]
20 mDOT = 11.064676 […]
21 mDOT.O = 8.114096 […]
22 mDOT.F = 2.950580 […]
23 mDOT' = 8.754151 […]
24
25 vinj.O = 41.142844 […]
26 vinj.F = 190.030543 […]
27 Ainj.O = 173.873427 […]
28 Ainj.F = 292.031450 […]
29 Aori.O = 1.448945 […]
30 Aori.F = 2.433595 […]
31 Dori.O = 1.358254 […]
32 Dori.F = 1.760270 […]
33 θinj = -0.330 […]
34 θ.O = 38 […]
35 θ.F = 22 […]
36 DSoriinj = 7.738160 […]
37 DSimping = 6.528375 […]
38 FAC.O = 6.099467 […]
39 FAC.F = 4.000000 […]
40 LEN.O = 8.28 […]
41 LEN.F = 7.04 […]
42 HEI.O = 6.528375114945 […]
43 HEI.F = 6.528375114945 […]
 137

44
45 Dcc = 130.000000 […]
46 Rcc = 65.000000 […]
47 CRCMcc = 414.690230 […]
48 Acc = 0.013273228961 […]
49 Lc = 142.163203 […]
50 Lf = 56.291651 […]
51 Lcc = 198.454855 […]
52 Pcc = 6000000.000000 […]
53 Tcc = 3338.670000 […]
54 SVcc = 0.237222 […]
55 vcc = 0.000000 […]
56 Vcc = 0.002322815068 […]
57 ts = 0.000885 […]
58
59 Dt = 65.000000 […]
60 Rt = 32.500000 […]
61 CRCMt = 210.486708 […]
62 At = 0.003318307240 […]
63 Pt = 3374871.418511 […]
64 Tt = 3021.010723 […]
65 SVt = 0.381617 […]
66 vt = 1248.501396 […]
67 θcon = 30 […]
68 θd.i = 34 […]
69 θd.e = 8 […]
70
71 De = 650.000000 […]
72 Re = 325.000000 […]
73 CRCMe = 2048.318410 […]
74 Ae = 0.331830724035 […]
75 Pe = 3058.000 […]
76 Te = 563.787177 […]
77 SVe = 78.597816 […]
78 ve = 3335.262441 […]
79 Me = 5.951841 […]
80 Pext = 0.886 […]
81
82 LN = 1091.624861 […]
83 Ln = 873.299889 […]
84 Rtu = 48.750 […]
85 Rtd = 13.000 […]
86
87 L = 1071.754744 […]
 138

88 THCK.i = 1.000 […]

89 THCK.o = 3.000 […]
90 DPTHcc = 1.000 […]
91 DPTHt = 1.000 […]
92 DPTHe = 1.000 […]
93 WDTHcc = 3.397 […]
94 WDTHt = 1.355 […]
95 WDTHe = 19.733 […]
96
97 VelOut = 3335.262440675 […]
98 L* = 0.700000 […]
99 Expn = 100.000
100 Cntr = 4.000
101 MAWS = 200.000 […]
102 Pcc/Pe = 1962.066710
103 Pcc/Pext = 6769870
104 Pe/Pext = 3450.376856
105 Pe-Pext = 3057.114
106 F../mDOT = 0.266667
107 O../mDOT = 0.733333
108 De/Dcc = 5.000000
109 De/Dt = 10.000000
110 Lcc/Dcc = 1.526576
111 Lc/Lf = 2.525476
112 ..F/..O = 4.618799
113 Acc/..O = 76.338456
114 Acc/..F = 45.451368
115
116 --------------------------------------------------
117
118 >>
 139

B.2 ENGINE CYCLE PARAMETERS

1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2
3 function [cycle] = EngineCycleParameters()
4
5 cycle = 'AA';
6
7 G0 = 9.806650; %% Gravity - Earth
8 P_cc = 6000000; %% Pressure - Thrust Chamber
9 eta_op = 0.66; %% Efficiency - Oxidizer Pump
10 eta_fp = 0.60; %% Efficiency - Fuel Pump
11 eta_mot = 0.95; %% Efficiency - Motor
12 eta_inv = 0.85; %% Efficiency - Inverter
13 eta_bat = 0.90; %% Efficiency - Battery
14 delta_op = 20000.00; %% Density Power - Oxidizer Pump
15 delta_fp = 15000.00; %% Density Power - Fuel Pump
16 delta_tp = 12500.00; %% Density Power - Turbopump
17 delta_mot = 5500.00; %% Density Power - Motor
18 delta_inv = 60000.00; %% Density Power - Inverter
19 delta_batP = 6500.00; %% Density Power - Battery
20 delta_batE = 300.00 * 3600;
21 %% Density Energy - Battery
22 delta_batP = delta_batP / 1000;
23 %% Density Power - Battery
24 delta_batE = delta_batE / 1000;
25 %% Density Energy - Battery
26 OF_gg = 0.333333; %% Mass Mixture Ratio - Gas Generator W. F.
27 rho_o = 1142.00; %% Density - Oxidizer
28 rho_f = 0422.80; %% Density - Fuel
29 rho_gg = 12.6240; %% Density - Gas Generator W. F.
30 rho_ggm = 8440.00; %% Density - Gas Generator Material
31 dotm_op = 8.114096; %% Mass Flow Rate - Oxidizer Pump
32 dotm_fp = 2.950580; %% Mass Flow Rate - Fuel Pump
33 dotm_gg = 0.75 * (dotm_op + dotm_fp) / 100;
34 %% Mass Flow Rate - Gas Generator W. F.
35 kappa_p = 1.05; %% Mass Margin - Propellant
36 kappa_gg = 2.50; %% Safety Factor - Gas Generator
37 kappa_bat = 1.25; %% Mass Margin - Battery Pack
38 Deltap_op = 1.25 * P_cc / rho_o / G0;
39 %% Head Rise - Oxidizer Pump
40 Deltap_fp = 1.50 * P_cc / rho_f / G0;
41 %% Head Rise - Fuel Pump
42 p_gg = 1.00 * P_cc; %% Pressure - Gas Generator
43 sigma_ggm = 500000000; %% Strength - Gas Generator Material
 140

44 t_s = 0.0100; %% Stay Time - Gas Generator W. F.

45 t_b = 0; %% Burn Time - Thrust Chamber
46
47 ...
48 t_b = 135.00; %% Burn Time - 1ST STAGE
49 t_b = 195.00; %% Burn Time - 2ND STAGE
50 t_b = 00:01:1000; %% Burn Time - Thrust Chamber
51 t_b = [135 195]; %% Burn Time - Thrust Chamber
52 ...
53
54 P_op = 0; %% Power - Oxidizer Pump
55 P_fp = 0; %% Power - Fuel Pump
56 P_tp = 0; %% Power - Turbopump
57
58 m_o_TPFS = 0; %% Mass - TP Oxidizer
59 m_f_TPFS = 0; %% Mass - TP Fuel
60 m_tp_TPFS = 0; %% Mass - Turbopump
61 m_gg_TPFS = 0; %% Mass - Gas Generator
62 m_xo_TPFS = 0; %% Mass - Gas Generator Oxidizer
63 m_xf_TPFS = 0; %% Mass - Gas Generator Fuel
64 m_xx_TPFS = 0; %% Mass - Gas Generator Propellant
65 M_FS_TPFS = 0; %% Mass - TPFS - Feed System
66 M_P_TPFS = 0; %% Mass - TPFS - Propellant
67 MR_TPFS = 0; %% Mass Ratio - TPFS
68
69 m_o_EPFS = 0; %% Mass - EP Oxidizer
70 m_f_EPFS = 0; %% Mass - EP Fuel
71 m_op_EPFS = 0; %% Mass - Oxidizer Pump
72 m_fp_EPFS = 0; %% Mass - Fuel Pump
73 m_mot_EPFS = 0; %% Mass - Motor
74 m_inv_EPFS = 0; %% Mass - Inverter
75 m_bat_EPFS = 0; %% Mass - Battery
76 M_FS_EPFS = 0; %% Mass - EPFS - Feed System
77 M_P_EPFS = 0; %% Mass - EPFS - Propellant
78 MR_EPFS = 0; %% Mass Ratio - EPFS
79
80 M_TPFS = 0; %% Mass - TPFS
81 M_EPFS = 0; %% Mass - EPFS
82
83 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
84
85 P_op = ((dotm_op * Deltap_op) / (eta_op * rho_o));
86 P_fp = ((dotm_fp * Deltap_fp) / (eta_fp * rho_f));
87 P_tp = P_op + P_fp;
 141

88
89 m_o_TPFS = kappa_p * t_b * dotm_op;
90 m_f_TPFS = kappa_p * t_b * dotm_fp;
91 m_tp_TPFS = (1 / delta_tp) * (P_tp);
92 m_gg_TPFS = (3 / 2) * (rho_ggm / sigma_ggm) ...
93 * (dotm_gg * t_s / rho_gg) * p_gg * kappa_gg;
94 m_xo_TPFS = kappa_p * t_b * (1 / (OF_gg + 1) * (dotm_gg * OF_gg));
95 m_xf_TPFS = kappa_p * t_b * (1 / (OF_gg + 1) * (dotm_gg * 1.000));
96 m_xx_TPFS = m_xo_TPFS + m_xf_TPFS;
97 M_FS_TPFS = m_tp_TPFS + m_gg_TPFS + m_xx_TPFS;
98 M_P_TPFS = m_o_TPFS + m_f_TPFS;
99 MR_TPFS = M_FS_TPFS ./ M_P_TPFS .* 100;
100
101 m_o_EPFS = kappa_p * t_b * dotm_op;
102 m_f_EPFS = kappa_p * t_b * dotm_fp;
103 m_op_EPFS = (1 / delta_op) * P_op;
104 m_fp_EPFS = (1 / delta_fp) * P_fp;
105 m_mot_EPFS = (1 / delta_mot) * (P_tp);
106 m_inv_EPFS = (1 / (delta_inv * eta_mot)) * (P_tp);
107 m_bat_EPFS = (1 / (eta_inv * eta_mot)) * kappa_bat * (P_tp) ...
108 * (max((1 / (delta_batP)), (t_b / (delta_batE * eta_bat))));
109 M_FS_EPFS = m_op_EPFS + m_fp_EPFS ...
110 + m_mot_EPFS + m_inv_EPFS + m_bat_EPFS;
111 M_P_EPFS = m_o_EPFS + m_f_EPFS;
112 MR_EPFS = M_FS_EPFS ./ M_P_EPFS .* 100;
113
114 M_TPFS = m_o_TPFS + m_f_TPFS + m_tp_TPFS + m_gg_TPFS ...
115 + m_xo_TPFS + m_xf_TPFS;
116 M_EPFS = m_o_EPFS + m_f_EPFS + m_op_EPFS + m_fp_EPFS ...
117 + m_mot_EPFS + m_inv_EPFS + m_bat_EPFS;
118
119 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
120
121 cycle = 'ZZ';
122
123 return;
124
125 end
126
127 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 142

Engine Cycle Parameters Output

1 >>
2
3 --------------------------------------------------
4
5 OP H.R. = 669.691 […]
6 FP H.R. = 2170.635 […]
7 OP P. = 7.209 […]
8 FP P. = 25.247 […]
9
10 GG M. = 0.083 […]
11 GG D. = 50.073 […]
12
13 ※
14
15 --------------------------------------------------
16
17
18 --------------------------------------------------
19
20 • T P F S
21
22 MASS TPFS = 1580.208520 […]
23 MASS TPFS = 2282.511167 […]
24
25 → M.o = 1150173.108 […]
26 → M.o = 1661361.156 […]
27 → M.f = 418244.715 […]
28 → M.f = 604131.255 […]
29 → M.tp = 2.597 […]
30 → M.gg = 24.967 […]
31 → M.xgg = 11763.134 […]
32 → M.xgg = 16991.193 […]
33
34 → P = 1568.417823 […]
35 → P = 2265.492411 […]
36 → FS = 11.790697 […]
37 → FS = 17.018756 […]
38
39 → FS / P • = 0.752 […]
40 → FS / P • = 0.751 […]
41
42 --------------------------------------------------
43
 143

44
45 --------------------------------------------------
46
47 • E P F S
48
49 MASS EPFS = 1576.155889 […]
50 MASS EPFS = 2275.580356 […]
51
52 → M.o = 1150173.108 […]
53 → M.o = 1661361.156 […]
54 → M.f = 418244.715 […]
55 → M.f = 604131.255 […]
56 → M.op = 0.360 […]
57 → M.fp = 1.683 […]
58 → M.mot = 5.901 […]
59 → M.inv = 0.569 […]
60 → M.bat = 7729.552 […]
61 → M.bat = 10079.431 […]
62
63 → P = 1568.417823 […]
64 → P = 2265.492411 […]
65 → FS = 7.738066 […]
66 → FS = 10.087945 […]
67
68 → FS / P • = 0.493 […]
69 → FS / P • = 0.445 […]
70
71 --------------------------------------------------
72
73
74 --------------------------------------------------
75
76 M TPFS = 1580.208520 […]
77 M TPFS = 2282.511167 […]
78
79 M EPFS = 1576.155889 […]
80 M EPFS = 2275.580356 […]
81
82 ※
83
84 --------------------------------------------------
85
86 >>
 144

B.3 VEHICLE MASS PARAMETERS

1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2
3 function [delta] = VehicleMassParameters()
4
5 delta = 'AA';
6
7 G0 = 9.80665;
8 R0 = 6371000;
9 mDOT = 11.064676254169;
10 mass_payl = 0200; %% Payload Mass [kg]
11 orbit = 0500; %% Orbit Altitude [km]
12 velo_1ST = 2700.486590390697; %% Exhaust Velocity
13 velo_2ND = 3335.262440674872; %% Exhaust Velocity
14 time_burn_1ST = 120; %% Operation Duration == 135
15 time_burn_2ND = 180; %% Operation Duration == 195
16 SMF_1ST = 0.075; %% Structural Mass Fraction >> ΔT = 135
17 SMF_2ND = 0.125; %% Structural Mass Fraction >> ΔT = 195
18 ETA = 0.000641075875; %% 1# = 0.075 || 2# = 0.125 >> <<--
19 margin = 15;
20 NUM = 09;
21
22 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
23
24 orbit = orbit * 1000;
25 velo_orbit = R0 * sqrt(G0 / (R0 + orbit));
26 velo_orbit = sqrt(0.398600441 * 10 ^ 15 / (R0 + orbit));
27 velo_mission = velo_orbit * 1.250;
28 velo_mission = round(velo_mission, 2, "significant");
29
30 velo_delta_stage_1ST = velo_1ST * log((velo_1ST * ETA - 1) ...
31 / (velo_1ST * ETA * SMF_1ST));
32 velo_delta_stage_2ND = velo_2ND * log((velo_2ND * ETA - 1) ...
33 / (velo_2ND * ETA * SMF_2ND));
34 velo_delta_stage = velo_delta_stage_1ST + velo_delta_stage_2ND;
35
36 OMR_1ST = (velo_1ST * ETA - 1) / (velo_1ST * ETA * SMF_1ST);
37 OMR_2ND = (velo_2ND * ETA - 1) / (velo_2ND * ETA * SMF_2ND);
38 PMF_1ST = 1 - SMF_1ST;
39 PMF_2ND = 1 - SMF_2ND;
40 CHK_1ST = ETA * velo_1ST * (SMF_1ST * OMR_1ST - 1) ^ 2 + 2 ...
41 * SMF_1ST * OMR_1ST - 1;
42 CHK_2ND = ETA * velo_2ND * (SMF_2ND * OMR_2ND - 1) ^ 2 + 2 ...
43 * SMF_2ND * OMR_2ND - 1;
 145

44 mass_wet_1ST = (OMR_1ST - 1) / (1 - OMR_1ST * SMF_1ST) ...

45 * (mass_payl + 000000000000);
46 mass_wet_2ND = (OMR_2ND - 1) / (1 - OMR_2ND * SMF_2ND) ...
47 * (mass_payl);
48 mass_wet_1ST = (OMR_1ST - 1) / (1 - OMR_1ST * SMF_1ST) ...
49 * (mass_payl + mass_wet_2ND);
50 mass_wet_2ND = (OMR_2ND - 1) / (1 - OMR_2ND * SMF_2ND) ...
51 * (mass_payl);
52 mass_dry_1ST = mass_wet_1ST * SMF_1ST;
53 mass_dry_2ND = mass_wet_2ND * SMF_2ND;
54 mass_prop_1ST = mass_wet_1ST - mass_wet_1ST * SMF_1ST;
55 mass_prop_2ND = mass_wet_2ND - mass_wet_2ND * SMF_2ND;
56 % time_burn_1ST = mass_prop_1ST / mDOT / NUM;
57 % time_burn_2ND = mass_prop_2ND / mDOT / 1.0;
58 % time_burn_1ST = round(time_burn_1ST, 2, "significant");
59 % time_burn_2ND = round(time_burn_2ND, 2, "significant");
60 time_burn_1ST = time_burn_1ST + margin;
61 time_burn_2ND = time_burn_2ND + margin;
62 mass_prop_1ST = mDOT * time_burn_1ST * NUM;
63 mass_prop_2ND = mDOT * time_burn_2ND * 1.0;
64 mass_wet_1ST = mass_prop_1ST / PMF_1ST;
65 mass_wet_2ND = mass_prop_2ND / PMF_2ND;
66 mass_dry_1ST = mass_wet_1ST * SMF_1ST;
67 mass_dry_2ND = mass_wet_2ND * SMF_2ND;
68 mass_vehicle = mass_payl + mass_wet_1ST + mass_wet_2ND;
69 velo_delta_1ST = velo_1ST * log((mass_payl + mass_wet_2ND ...
70 + mass_wet_1ST) / (mass_payl + mass_wet_2ND + mass_dry_1ST));
71 velo_delta_2ND = velo_2ND * log((mass_payl + mass_wet_2ND) ...
72 / (mass_payl + mass_dry_2ND));
73 velo_delta_vehicle = velo_delta_1ST + velo_delta_2ND;
74
75 % ×××××××××××××××××××××××××××××××××××××××××××××××××××××××
76
77 delta = 'ZZ';
78
79 return;
80
81 end
82
83 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
 146

Vehicle Mass Parameters Output

1 >>
2
3 --------------------------------------------------
4
5 Orbit = 500.000000 […]
6 V.Orbit = 7616.561 […]
7 V.Mission = 9500.000 […]
8
9 ΔV.1 ϕ = 4667.513924 […]
10 ΔV.2 ϕ = 4832.486076 […]
11 ΔV • ϕ = 9500.000000 […]
12
13 ΔV.1 θ = 4108.950112 […]
14 ΔV.2 θ = 5527.666051 […]
15 ΔV • θ = 9636.616163 […]
16
17 B.T.1 = 135.000 […]
18 B.T.2 = 195.000 […]
19
20 SMF.1 = 0.075 […]
21 SMF.2 = 0.125 […]
22 PMF.1 = 92.500 […]
23 PMF.2 = 87.500 […]
24 OMR.1 = 5.631621 […]
25 OMR.2 = 4.258459 […]
26 ETA = 0.000641075875 […]
27
28 •
29
30 M.Prop.1 = 13443.582 […]
31 M.Dry..1 = 1090.020 […]
32 M.Wet..1 = 14533.602 […]
33 M.Prop.2 = 2157.612 […]
34 M.Dry..2 = 308.230 […]
35 M.Wet..2 = 2465.842 […]
36
37 M.Payl.. = 200.000 […]
38 M.Vehi.. = 17199.444 […]
39
40 --------------------------------------------------
41
42 >>
Bibliography

[1] Carl Sagan. Cosmos. New York, USA: Random House, 1980. isbn: 0-394-50294-
9.
[2] Tim Dodd and Mariia Kiseleva. “Soviet Rocket Engines”. 2021. url: https :
//everydayastronaut.com/soviet-rocket-engines/ (visited on 06/01/2022).
[3] “Falcon Nine - SpaceX”. 2021. url: https://www.spacex.com/vehicles/falcon-9/
(visited on 06/01/2021).
[4] SpaceNews. “SpaceNews Archive”. 2023. url: https://spacenews.com/section/
news-archive/ (visited on 03/08/2023).
[5] Vox. “Private spaceflight, explained”. 2015. url: https://www.vox.com/2015/2/6/
18073658/private-space-flight (visited on 03/01/2023).
[6] Harbard Business Review. “The Commercial Space Age Is Here”. 2021. url: https:
//hbr.org/2021/02/the-commercial-space-age-is-here (visited on 03/01/2023).
[7] “Electron - Rocket Lab”. 2021. url: https : / / www . rocketlabusa . com / rockets /
electron/ (visited on 06/01/2021).
[8] Harry Jones. “The Recent Large Reduction in Space Launch Cost”. Conference
Paper, 48th International Conference on Environmental Systems, ICES-2018-
81. California, USA, July 2018. url: http://hdl.handle.net/2346/74082 (visited
on 03/08/2023).
[9] Space.com. “8 ways that SpaceX has transformed spaceflight”. 2022. url: https:
//www.space.com/ways-spacex-transformed-spaceflight (visited on 03/08/2023).
[10] John Gardi and Jon Ross. “The Future of Space Launch is Here!” 2022. url:
http://justatinker.com/Future/ (visited on 03/01/2023).
[11] Axelspace Corporation. “Announcing the new development of critical components
to prepare for the mass production of next-generation microsatellites”. Aug. 12,
148

2020. url: https://www.axelspace.com/news/press_ 20201012/ (visited on

03/12/2023).
[12] George P. Sutton and Oscar Biblarz. Rocket Propulsion Elements. Ninth Edition.
New Jersey, USA: John Wiley & Sons, Inc., 2017. isbn: 978-1-118-75388-0.
[13] David H. Huang and Dieter K. Huzel. Modern Engineering for Design of Liquid-
Propellant Rocket Engines. Revised and Updated Edition. Washington DC, USA:
American Institute of Aeronautics and Astronautics, 1992. isbn: 1-56347-013-6.
[14] “CODATA Value: Molar Gas Constant”. Fundamental Physical Constants. The
NIST Reference on Constants, Units, and Uncertainty. May 20, 2019. url:
https://physics.nist.gov/cgi-bin/cuu/Value?r (visited on 02/10/2021).
[15] George P. Sutton. History of Liquid Propellant Rocket Engines. Virginia, USA:
American Institute of Aeronautics and Astronautics, 2006. isbn: 978-1-56347-
649-5.
[16] Jonathan C. McDowell. “Engine Data Table”. Engines, Launch Vehicles. Gen-
eral Catalog of Artificial Space Objects. 2020. url: https://planet4589.org/
space/gcat/data/tables/engines.html (visited on 03/05/2021).
[17] Mark Wade. “Vinci”. Encyclopedia Astronautica. 2019. url: http : / / www .
astronautix.com/v/vinci.html (visited on 03/04/2021).
[18] Mark Wade. “RL-10 B-2”. Encyclopedia Astronautica. 2019. url: http://www.
astronautix.com/r/rl-10b-2.html (visited on 03/04/2021).
[19] John D. Clark. Ignition! An Informal History of Liquid Rocket Propellants. New
Jersey, USA: Rutgers University Press, 1972. isbn: 0-8135-0725-1.
[20] PubChem. “PubChem Compound Summary for CID 977, Oxygen”. National
Center for Biotechnology Information. 2021. url: https://pubchem.ncbi.nlm.
nih.gov/compound/Oxygen (visited on 03/12/2021).
[21] Engineering ToolBox. “Oxygen - Thermophysical properties”. 2008. url: https:
//www.engineeringtoolbox.com/oxygen-d_1422.html (visited on 03/12/2021).
[22] Mark Wade. “LOX”. Encyclopedia Astronautica. 2019. url: http://astronautix.
com/l/lox.html (visited on 03/12/2021).
[23] Virginia P. Dawson and Mark D. Bowles. Taming Liquid Hydrogen: The Centaur
Upper Stage Rocket 1958-2002. The NASA History Series. Washington DC, USA:
National Aeronautics and Space Administration, Office of External Relations,
2004. Chap. Introduction. url: https://history.nasa.gov/SP-4230.pdf .
[24] PubChem. “PubChem Compound Summary for CID 783, Hydrogen”. National
Center for Biotechnology Information. 2021. url: https://pubchem.ncbi.nlm.
nih.gov/compound/Hydrogen (visited on 03/20/2021).
[25] Engineering ToolBox. “Hydrogen - Thermophysical Properties”. 2008. url: https:
//www.engineeringtoolbox.com/hydrogen-d_1419.html (visited on 03/20/2021).
 149

[26] College of the Desert. “Module 1: Hydrogen Properties”. Hydrogen Fuel Cell
Engines and Related Technologies: Revision 0, Dec. 2001. url: https://www1.
eere.energy.gov/hydrogenandfuelcells/tech_validation/pdfs/fcm01r0.pdf (visited
on 03/21/2021).
[27] Robert A. Braeunig. “Rocket Propellants”. Basics of Space Flight. Rocket & Space
Technology. 2008. url: http://www.braeunig.us/space/propel.htm (visited on
02/15/2021).
[28] Norbert Brügge. “Space Launch Vehicles”. Spacerockets: Database & Analytics.
B14643.de. 2021. url: http://www.b14643.de/Spacerockets/index.htm (visited
on 03/22/2021).
[29] Mark Wade. “LOX/LH2”. Encyclopedia Astronautica. 2019. url: http : / /
astronautix.com/l/loxlh2.html (visited on 04/02/2021).
[30] Pablo Rachov, Hernán Tacca, and Diego Lentini. “Electric Feed Systems for Liq-
uid Propellant Rocket Engines”. Research rep. 2010. url: https://www.aacademica.
org/hernan.emilio.tacca/9.pdf (visited on 02/01/2021).
[31] Mark Wade. “LOX/Kerosene”. Encyclopedia Astronautica. 2019. url: http :
//astronautix.com/l/loxkerosene.html (visited on 04/23/2021).
[32] “Detail Specification—Propellant: Rocket Grade Kerosene”. MIL-DTL-25576D.
Washington DC, USA: Department of Defense, May 20, 2005.
[33] PubChem. “PubChem Compound Summary for CID 297, Methane”. National
Center for Biotechnology Information. 2021. url: https://pubchem.ncbi.nlm.
nih.gov/compound/Methane (visited on 05/21/2021).
[34] Robert A. Braeunig. “Rocket Thermodynamics”. Rocket & Space Technology.
2012. url: http://www.braeunig.us/space/thermo.htm (visited on 02/15/2021).
[35] Mark Wade. “LOX/LCH4”. Encyclopedia Astronautica. 2019. url: http://
astronautix.com/l/loxlch4.html (visited on 05/21/2021).
[36] Engineering ToolBox. “Adiabatic Flame Temperatures”. 2005. url: https://www.
engineeringtoolbox.com/adiabatic-flame-temperature-d_996.html (visited on
04/10/2021).
[37] Gerald Hagemann. “LOX/Methane - The Future is Green”. Airbus Defence &
Space, Nov. 4, 2015. url: http://www.academie-air-espace.com/upload/doc/
ressources/Launchers/slides/hagemann.pdf (visited on 05/21/2021).
[38] Engineering ToolBox. “Methane - Thermophysical Properties”. 2008. url: https:
//www.engineeringtoolbox.com/methane-d_1420.html (visited on 05/21/2021).
[39] Sanford Gordon and Bonnie J. McBride. “CEARUN”. CEA (Chemical Equi-
librium with Applications). NASA Glenn Research Center, Ohio, USA: NASA,
1994. url: https://cearun.grc.nasa.gov/.
150

[40] Sanford Gordon and Bonnie J. McBride. “Introduction to CEA”. Computer

Program for Calculation of Complex Chemical Equilibrium Compositions and Appli-
cations. NASA. 1994. url: https://cearun.grc.nasa.gov/intro.html (visited on
06/01/2021).
[41] Alexander Ponomarenko. “Rocket Propulsion Analysis”. RPA Standard Edition
v.2.x. Version v.2.3.2. Germany, 2008. url: https://www.rocket-propulsion.com/
index.htm.
[42] Robert A. Braeunig. “Rocket Propulsion”. Basics of Space Flight. Rocket & Space
Technology. 2012. url: http://www.braeunig.us/space/propuls.htm (visited on
02/15/2021).
[43] Robert A. Braeunig. “Propellant Combustion Charts”. Rocket & Space Tech-
nology. 2005. url: http : / / www . braeunig . us / space / comb . htm (visited on
02/15/2021).
[44] Mike Jetzer. “heroicrelics.org”. 2021. url: http://heroicrelics.org/ (visited on
06/20/2021).
[45] NASA. “Turbopump Systems for Liquid Rocket Engines”. Space Vehicle Design
Monograph. NASA SP-8107. Aug. 1974. url: https://ntrs.nasa.gov/citations/
19750012398 (visited on 06/26/2021).
[46] Hyun-Duck Kwak, Sejin Kwon, and Chang-Ho Choi. “Performance assessment
of electrically driven pump-fed LOX/kerosene cycle rocket engine: Comparison with
gas generator cycle”. Aerospace Science and Technology, Vol. 77. Republic of
Korea, 2018. doi: 10.1016/j.ast.2018.02.033. url: https://www.sciencedirect.com/
science/article/pii/S1270963817320953 (visited on 07/20/2021).
[47] Pablo Rachov, Hernán Tacca, and Diego Lentini. “Electric Feed Systems for
Liquid-Propellant Rockets”. Journal of Propulsion and Power, Vol. 29, No. 5.
Argentina & Italy, 2013. doi: 10.2514/1.B34714. url: https://arc.aiaa.org/doi/10.
2514/1.B34714 (visited on 07/20/2021).
[48] N. Soldà and D. Lentini. “Opportunities for a Liquid Rocket Feed System Based on
Electric Pumps”. Journal of Propulsion and Power, Vol. 24, No. 6. Italy, 2008.
doi: 10.2514/1.35074. url: https://arc.aiaa.org/doi/10.2514/1.35074 (visited on
07/20/2021).
[49] Jeffrey Ball. “The Race Is on to Build a Better Battery”. FORTUNE. May 24,
2019. url: https://fortune.com/longform/race-build-better-battery/ (visited on
06/15/2021).
[50] Eric Adams. “The Age of Electric Aviation Is Just 30 Years Away”. WIRED. May 31,
2017. url: https://www.wired.com/2017/05/electric-airplanes-2/ (visited on
06/15/2021).
 151

[51] Special Metals Corporation. “INCONEL Alloy 625”. UNS N06625/W.Nr. 2.4856.
2013. url: https://www.specialmetals.com/documents/technical-bulletins/inconel/
inconel-alloy-625.pdf (visited on 06/14/2022).
[52] “U.S. Standard Atmosphere, 1976”. NOAA-S/T 76-1562. Washington DC, USA:
National Oceanic, Atmospheric Administration, National Aeronautics, Space
Administration, and U.S. Air Force, Oct. 1976.
[53] John David Anderson Jr. Fundamentals of Aerodynamics. Sixth Edition. New
York, USA: McGraw-Hill Education, 2017. isbn: 978-1-259-12991-9.
[54] Roelof Vos and Saeed Farokhi. Introduction to Transonic Aerodynamics. Fluid
Mechanics and Its Applications, Volume 110. The Netherlands: Springer, 2015.
isbn: 978-94-017-9747-4.
[55] Jan Östlund. “Flow Processes in Rocket Engine Nozzles with Focus on Flow Sep-
aration and Side-Loads”. Stockholm, Sweden: Royal Institute of Technology,
2002. url: http://www.diva-portal.org/smash/get/diva2:7328/FULLTEXT01.pdf
(visited on 11/15/2021).
[56] J. H. Ahlberg, S. Hamilton, D. Migdal, and E. N. Nilson. “Truncated Perfect
Nozzles in Optimum Nozzle Design”. ARS Journal, Vol. 31, No. 5. Connecticut,
USA, May 1961. doi: 10.2514/8.5577. url: https://arc.aiaa.org/doi/10.2514/8.
5577 (visited on 01/04/2022).
[57] L. V. Gogish. “Investigation of Short Supersonic Nozzles”. AN SSSR, Izvestiya,
MechanicaZidkosti i Gaza. 1966.
[58] G. V. R. Rao. “Exhaust Nozzle Contour for Optimum Thrust”. Journal of Jet
Propulsion, Vol. 28, No. 6. California, USA, June 1958. doi: 10.2514/8.7324.
url: https://arc.aiaa.org/doi/10.2514/8.7324 (visited on 03/01/2022).
[59] G. V. R. Rao. “Approximation of Optimum Thrust Nozzle Contour”. Tecnical
Notes. ARS Journal, Vol. 30, No. 6. California, USA, June 1960, p. 561. doi:
10.2514/8.5151. url: https://arc.aiaa.org/doi/10.2514/8.5151 (visited on
03/10/2022).
[60] Nobuyuki Yatsuyanagi. “Comprehensive Design Method for LOX/Liquid-Methane
Regenerative Cooling Combustor with Coaxial Injector”. Transactions of the Japan
Society for Aeronautical and Space Sciences, Vol. 52, No. 177. Japan, 2009,
pp. 180–187. doi: 10.2322/tjsass.52.180. url: https://www.jstage.jst.go.jp/article/
tjsass/52/177/52_177_180/_pdf/-char/en (visited on 07/27/2022).
[61] H. C. Starck. “C-103 Nb Alloy: Properties & Applications”. Aug. 10, 2020. url:
https : / / www . hcstarcksolutions . com / c103 - nb - alloy - properties - applications/
(visited on 08/15/2022).
152

[62] G. S. Gill and W. H. Nurick. “Liquid Rocket Engine Injectors”. Space Vehicle
Design Monograph. NASA SP-8089. Mar. 1976. url: https://ntrs.nasa.gov/
citations/19760023196 (visited on 08/14/2022).
[63] John C. Melcher and Robert L. Morehead. “Combustion Stability Characteristics
of the Project Morpheus Liquid Oxygen / Liquid Methane Main Engine”. Con-
ference Paper, 50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
Ohio, USA, 2014. doi: 10.2514/6.2014-3681. url: https://arc.aiaa.org/doi/10.
2514/6.2014-3681 (visited on 07/27/2022).
[64] James R. Hulka and Gregg W. Jones. “Performance and Stability Analyses of
Rocket Thrust Chambers with Oxygen/Methane Propellants”. Conference Paper,
46th AIAA/ASME/SAE/ASEE Joint Propulsion Conference. Tennessee, USA,
July 2010. doi: 10.2514/6.2010-6799. url: https://arc.aiaa.org/doi/abs/10.2514/
6.2010-6799 (visited on 07/27/2022).
[65] Howard D. Curtis. Orbital Mechanics for Engineering Students. Fourth Edition.
Butterworth-Heinemann, 2020. isbn: 978-0-08-102133-0.
& to my Mom
 Division of Space Technology
Department of Computer Science, Electrical and Space Engineering

Luleå University of Technology 2023