AD1184618

NAVAL
POSTGRADUATE
SCHOOL
MONTEREY, CALIFORNIA
THESIS
MODIFICATION OF COMMERCIAL ROCKET

MOTORS FOR TACTICAL APPLICATIONS
by
Nathan D. Stuffle
June 2022
Thesis Advisor: Christopher M. Brophy

Second Reader: Alexis Thoeny
Approved for public release. Distribution is unlimited.
THIS PAGE INTENTIONALLY LEFT BLANK
Form Approved OMB
REPORT DOCUMENTATION PAGE No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing
instruction, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of
information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions
for reducing this burden, to Washington headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project
(0704-0188) Washington, DC, 20503.
1. AGENCY USE ONLY 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
(Leave blank) June 2022 Master’s thesis
4. TITLE AND SUBTITLE 5. FUNDING NUMBERS
MODIFICATION OF COMMERCIAL ROCKET MOTORS FOR TACTICAL
APPLICATIONS R4Q04
6. AUTHOR(S) Nathan D. Stuffle
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING

Naval Postgraduate School ORGANIZATION REPORT
Monterey, CA 93943-5000 NUMBER
9. SPONSORING / MONITORING AGENCY NAME(S) AND 10. SPONSORING /
ADDRESS(ES) MONITORING AGENCY
ONR/NPS REPORT NUMBER
11. SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the
official policy or position of the Department of Defense or the U.S. Government.
12a. DISTRIBUTION / AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE
Approved for public release. Distribution is unlimited. A
13. ABSTRACT (maximum 200 words)
Several design modifications to commercial off-the-shelf solid rocket motors have been evaluated in support
of the Naval Postgraduate School tactical rapid-response payload delivery vehicle. The modifications include a
novel head-end ignition system and a tailorable nozzle end cap designed to provide reliable at-altitude ignition by
improving the transient behavior of the initial combustion chamber pressure rise. The nozzle cap also provides the
additional benefit of extending the shelf life of the propellant by creating an environmental seal to prevent ambient
humidity from affecting the propellant. A preliminary design of a blast tube was proposed to explore how the
motor exhaust could be channeled through a smaller-diameter tube before reaching the nozzle throat, thereby
accommodating the volume requirements of aft fin control servos without sacrificing the overall rocket diameter
or precluding use of larger-diameter rocket motors. Implementation of a blast tube also resulted in a favorable
shift of the center of gravity of the rocket, which preserved and enhanced control authority during simulated
fly-outs. All of the modifications were designed to be directly interchangeable with the OEM hardware to
minimize the cost of implementing the new capabilities. Both the head-end ignition and nozzle enclosure systems
were successfully demonstrated during flight testing, and a design process for the future implementation of a blast
tube was proposed.
14. SUBJECT TERMS 15. NUMBER OF

commercial, commercial off-the-shelf, COTS, modification, rocket, rocket motor, ignition, PAGES
blast tube, forward ignition, head-end ignitor, nozzle, nozzle cap, nozzle plug, head-end 111
ignition, tactical, tactical applications, rapid response, payload delivery, rocket propulsion, 16. PRICE CODE
rocket propellant, commercial rocket, solid rocket motor, SRM
17. SECURITY 18. SECURITY 19. SECURITY 20. LIMITATION OF

CLASSIFICATION OF CLASSIFICATION OF THIS CLASSIFICATION OF ABSTRACT
REPORT PAGE ABSTRACT
Unclassified Unclassified Unclassified UU
NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)

Prescribed by ANSI Std. 239-18
i
ii
Approved for public release. Distribution is unlimited.
MODIFICATION OF COMMERCIAL ROCKET MOTORS

FOR TACTICAL APPLICATIONS
Nathan D. Stuffle
Lieutenant, United States Navy
BS, Embry-Riddle Aeronautical University, Prescott, 2016
Submitted in partial fulfillment of the

requirements for the degrees of
MASTER OF SCIENCE IN SPACE SYSTEMS OPERATIONS
and
MASTER OF SCIENCE IN ASTRONAUTICAL ENGINEERING
from the
NAVAL POSTGRADUATE SCHOOL

June 2022
Approved by: Christopher M. Brophy

Advisor
Alexis Thoeny
Second Reader
James H. Newman
Chair, Space Systems Academic Group
Garth V. Hobson
Chair, Department of Mechanical and Aerospace Engineering
iii
iv
ABSTRACT
Several design modifications to commercial off-the-shelf solid rocket motors have

been evaluated in support of the Naval Postgraduate School tactical rapid-response
payload delivery vehicle. The modifications include a novel head-end ignition system and
a tailorable nozzle end cap designed to provide reliable at-altitude ignition by improving
the transient behavior of the initial combustion chamber pressure rise. The nozzle cap
also provides the additional benefit of extending the shelf life of the propellant by
creating an environmental seal to prevent ambient humidity from affecting the propellant.
A preliminary design of a blast tube was proposed to explore how the motor exhaust
could be channeled through a smaller-diameter tube before reaching the nozzle throat,
thereby accommodating the volume requirements of aft fin control servos without
sacrificing the overall rocket diameter or precluding use of larger-diameter rocket motors.
Implementation of a blast tube also resulted in a favorable shift of the center of gravity of
the rocket, which preserved and enhanced control authority during simulated fly-outs. All
of the modifications were designed to be directly interchangeable with the OEM
hardware to minimize the cost of implementing the new capabilities. Both the head-end
ignition and nozzle enclosure systems were successfully demonstrated during flight
testing, and a design process for the future implementation of a blast tube was proposed.
v
vi
TABLE OF CONTENTS
I. INTRODUCTION..................................................................................................1
A. BACKGROUND ........................................................................................1
B. MOTIVATION ..........................................................................................7
C. OBJECTIVES ..........................................................................................11
II. APPROACH .........................................................................................................13

A. HEAD-END IGNITION SYSTEM ........................................................13
B. NOZZLE ENCLOSURE SYSTEM .......................................................13
C. BLAST TUBE...........................................................................................14
III. DESIGN PROCESSES ........................................................................................17

1. Ignition Squib Holder Version 1.................................................19
1. Nozzle Enclosure Plug .................................................................35
2. Nozzle Enclosure Cap ..................................................................37
3. Enclosure Retainer.......................................................................39
C. BLAST TUBE...........................................................................................41
IV. EXPERIMENTAL RESULTS............................................................................47

A. STATIC MOTOR LIVE-FIRE TESTING............................................47
B. NOZZLE CAP LABORATORY TESTING .........................................51
V. ROCKET FLIGHT TESTING ...........................................................................57

A. OCTOBER 2021 TWO-STAGE ROCKET ..........................................57
1. Rocket Vehicle Design .................................................................57
2. Flight Electronics .........................................................................60
3. Flight Data and Results ...............................................................65
B. APRIL 2022 TWO-STAGE ROCKET ..................................................73
1. Rocket Vehicle Design .................................................................73
2. Flight Electronics .........................................................................74
3. Flight Data and Results ...............................................................76
vii
VI. SUMMARY ..........................................................................................................79
VII. FUTURE WORK .................................................................................................81

C. BLAST TUBE...........................................................................................81
D. ROCKET VEHICLE ...............................................................................82
APPENDIX. DATA LOGGER ARDUINO CODE ......................................................83
LIST OF REFERENCES ................................................................................................87
INITIAL DISTRIBUTION LIST ...................................................................................91
viii
LIST OF FIGURES
Figure 1. Example of COTS drones swarm. Source: [1].............................................2
Figure 2. 2017 model of NPS RPV. Source: [4]. ........................................................3
Figure 3. 3D bomblet model. Source: [5]. ...................................................................4
Figure 4. NPS anti-drone bomblet concept of employment. Source: [6]. ...................4
Figure 5. 2019 RPV multiple bomblets design. Source: [7]. ......................................5
Figure 6. Multi-bomblet deploying mechanism. Source: [7]. .....................................6
Figure 7. NPS RRPD-V multi-mission concept. Source: [8]. .....................................6
Figure 8. Fall 2020 two-stage mid-flight failure. Source: [6]. ....................................7
Figure 9. NASA rocket stability demonstration. Source: [9]. .....................................8
Figure 10. Cesaroni M1540 representative thrust curve. Adapted from [11]. ............10
Figure 11. Hypothetical rocket model with a blast tube (stable). Adapted from
[12]. ............................................................................................................11
Figure 12. Hypothetical rocket model without a blast tube (unstable). Adapted
from [12]. ...................................................................................................11
Figure 13. Ignition squib with 2 g of solid propellant attached ..................................17
Figure 14. Smoke charge replacement hardware ........................................................18
Figure 15. Smoke charge replacement hardware within 4-grain SRM 3D model
cutaway ......................................................................................................18
Figure 16. Head-end ignition leads post-fire ...............................................................19
Figure 17. Squib basket 3D model hole configuration C1 ..........................................20
Figure 18. a) Hole configuration C2. b) Hole configuration C3 .................................21
Figure 19. 7.62 cm (3 in) basket in configurations a) C1, b) C2, and c) C3 ...............22
Figure 20. Test stand setup with 7.62 cm (3 in) C1 basket .........................................22
Figure 21. High-speed camera and test stand setup ....................................................23
ix
Figure 22. 7.62 cm (3 in) basket with 0.159 cm (1/16) in holes high-speed
camera still frame .......................................................................................24
Figure 23. Post-fire 7.62 cm (3 in) basket with clogged holes ....................................24
Figure 24. 5.08 cm (2 in) basket with combination hole sizes a) before and b)
during ignition............................................................................................25
Figure 25. Ignition basket version 2 with 8 holes .......................................................27
Figure 26. Ignition basket version 2 with 0.476 cm (3/16 in) holes test high-
speed camera still frame.............................................................................28
speed camera still frame.............................................................................28
Figure 28. Static SRM live-fire test stand setup ..........................................................29
Figure 29. Test stand head-end after live-fire test .......................................................30
Figure 30. Exterior of motor head-end retaining hardware after live-fire test ............31
Figure 31. Ignition basket version 2 after live-fire test ...............................................32
Figure 32. Semi-flexible squib holder supports ..........................................................32
Figure 33. 3D model of squib holder within smoke charge replacement puck ...........33
Figure 34. Vacuum chamber test article ......................................................................34
Figure 35. Ignition squib holder final design ..............................................................34
Figure 36. Geometry of a converging-diverging (CD) nozzle. Source: [17]. .............36
Figure 37. Plug cap 3D model .....................................................................................36
Figure 38. Example of a) recessed and b) protruding nozzles ....................................38
Figure 39. Nozzle enclosure cap 3D model ................................................................38
Figure 40. Enclosure retainer arm and plug/cap..........................................................39
Figure 41. Shear pin retainer collar and cap ................................................................41
Figure 42. 3D model cutaway of blast tube based on the O8000 motor .....................42
Figure 43. 3D model of blast tube assembled into O8000 motor................................43
x
Figure 44. M3400 chamber pressure vs. time .............................................................48
Figure 45. M3400 thrust vs. time ................................................................................49
Figure 46. M3400 pressure and thrust profiles............................................................49
Figure 47. M3400 representative thrust curve. Adapted from [21].............................50
Figure 48. Zoomed in M3400 pressure vs. time..........................................................51
Figure 49. Enclosed squib test apparatus ....................................................................52
Figure 50. Enclosed squib test pressure vs. time.........................................................52
Figure 51. Nozzle cap test apparatus ...........................................................................53
Figure 52. Four pin nozzle cap/test run 1 pressure vs. time ........................................54
Figure 53. Six pin nozzle cap/test run 1 pressure vs. time ..........................................54
Figure 54. Six pin nozzle cap/test run 2 pressure vs. time ..........................................55
Figure 55. OpenRocket model of rocket design ..........................................................57
Figure 56. First stage booster (right) and parachute bay (left) ....................................58
Figure 57. Booster motor adapter sleeve .....................................................................58
Figure 58. OpenRocket sustainer section model .........................................................59
Figure 59. Sustainer motor section ..............................................................................59
Figure 60. OpenRocket simulation sustainer for October 2021 rocket launch ...........60
Figure 61. Altus Metrum TeleMega. Source: [22]. .....................................................61
Figure 62. PerfectFlite StratoLoggerCF. Source: [23]. ...............................................61
Figure 63. Internal view of booster parachute bay and aft electronics bulkhead ........62
Figure 64. Altus Metrum EasyMega. Source: [24]. ....................................................62
Figure 65. Booster electronics bay forward bulkhead .................................................63
Figure 66. Internal view of drogue parachute bay and sustainer electronics bay
aft bulkhead ................................................................................................64
Figure 67. Multitronix Kate-1 TelemetryPro transmitter. Source: [25]. .....................64
xi
Figure 68. Sustainer electronics bay and nose cone exploded view ...........................65
Figure 69. Booster stage flight data.............................................................................66
Figure 70. Booster stage burn......................................................................................67
Figure 71. Moment of separation ................................................................................67
Figure 72. Sustainer after stage separation ..................................................................68
Figure 73. Flight 1 sustainer atmospheric pressure and height data ...........................70
Figure 74. Moment of head-end ignition squib fire ....................................................70
Figure 75. Flight 1 sustainer stage data .......................................................................72
Figure 76. OpenRocket model of April 2022 rocket ...................................................73
Figure 77. OpenRocket simulated flight for April 2022 rocket launch.......................74
Figure 78. Adafruit Data Logger Shield with Arduino UNO. Source: [27]. ...............75
Figure 79. Sustainer avionics bay................................................................................75
Figure 80. Sustainer avionics bay side view ...............................................................76
Figure 81. First stage boost phase T+2 seconds ..........................................................77
Figure 82. a) Booster burnout, b) stage separation, c) sustainer ignition ....................77
xii
LIST OF TABLES
Table 1. n values at various pressure ranges. Adapted from [15]. ...........................26
Table 2. Shear limits of #2 nylon pins. Adapted from [18]. ....................................40
Table 3. Cured material physical and mechanical properties of carbon

phenolic. Adapted from [20]. .....................................................................45
Table 4. Booster flight statistics...............................................................................65
Table 5. Standard atmosphere properties. Adapted from [26]. ................................69
Table 6. Sustainer flight statistics ............................................................................72
xiii
xiv
LIST OF ACRONYMS AND ABBREVIATIONS
CD Converging-Diverging
CG Center of Gravity
COTS Commercial Off-The-Shelf
CP Center of Pressure
CTI Cesaroni Technology Inc.
FS Factor of Safety
GNC Guidance, Navigation, and Control
MSFC Marshall Space Flight Center
NPS Naval Postgraduate School
RCAF Royal Canadian Air Force
RPL Rocket Propulsion Lab
RPV Rocket-Powered Vehicle
RRPD-V Rapid-Response Payload Delivery Vehicle
SRM Solid Rocket Motor
xv
xvi
ACKNOWLEDGMENTS
I would like to thank the NPS Rocket Lab staff and faculty. My deepest gratitude
goes to my advisor, Dr. Brophy. Thank you for taking me on when I wandered into your
lab without knowing anything about rockets. It has been a life goal of mine to complete the
degree for which this thesis was written, and I owe the achievement of that goal largely to
your willingness to work with me. To Alexis Thoeny, thank you for all of your assistance
throughout this effort and your willingness to explain to me in copious detail why my
idea(s) would not work and point me in the right direction.
Lastly, and most importantly, I want to thank my wife, Lauren. Thank you for your
loving support while I subjected our family to the requirements of earning not one, but two,
master’s degrees at the same time. I am forever grateful for your compassion and
understanding in times of stress and hardship. I love you, and I could not have done this
without you.
xvii
xviii
I. INTRODUCTION
A. BACKGROUND
Over the past two decades, technological advancements have increased the
affordability of and access to microtechnology to the point where components can be
commercially acquired by rogue actors and state governments alike for use in cheap
instruments that can threaten American and friendly forces. The affordability and
availability of such components has created a disparity between the low-cost of these
technologies and the cost of countering them with expensive conventional military
capabilities. A method of contending with the asymmetric cost disparity is to develop a
low-cost solution to tactical problems using similar commercially available components.
One example of how microtechnology poses a threat is in an adversary’s ability to

cheaply constitute a drone fleet. The popularity of airborne drones has significantly
increased among hobbyists and private entities thanks in part to rapid advancements in
battery power and the power-to-size ratio of commercial drone motors [1]. Such
advancements gave rise to the miniature form-factor of modern commercial drones and
have significantly reduced the cost to consumers. With the advent of cheap, easy-to-use
drone technology, commercial drones have become an alluring, easily accessible option
with which extremist organizations and adversaries can threaten U.S. or friendly military
and government targets.
Examples of extremist use of modern commercial drone technology include a 2013

plot by Al-Qaeda to use multiple drones against Pakistan, the incorporation of homemade
and commercial-off-the-shelf (COTS) drones by the Islamic State in Iraq and Syria
beginning in 2014, the 2018 assassination attempt against Venezuelan President Maduro
using explosive-laden drones, and the 2018 drone swarm attacks on two Russian military
bases [1]. The drones used in these examples can be easily built using COTS components
or purchased fully assembled at low cost, allowing for the implementation of swarm tactics
aimed at overwhelming defensive countermeasures. Figure 1 shows an example of a drone
swarm.
1
Figure 1. Example of COTS drones swarm. Source: [1].
Countermeasures to modern drone swarms are frequently limited by cost due to the
number of countermeasures required and their ability to autonomously engage multiple
targets simultaneously. Electronic attack can be a cost-effective solution to disabling small
numbers of unprotected COTS drones, but purely electronic means of defense become
more complicated as swarm numbers increase or if the drone operators implement
electronic counter-countermeasures through electronic hardening and enhanced
cybersecurity software [2]. Therefore, an optimal strategy for engaging adversary drones
should include some form of autonomous kinetic kill capability. Modern self-guided
missiles possess enough capability to engage swarms at long distances and accurately
target incoming drones [2], however, the costs of using such systems is prohibitively high.
Even small semi-active seeker missiles, such as the Hellfire, exceed $100,000 per unit [3].
When compared to the low cost of reconstituting a COTS drone fleet, it becomes obvious
that the asymmetric cost of deploying current military missile systems to achieve
autonomous kinetic kills against drone swarms is unsustainable.
In 2016, the Naval Postgraduate School (NPS) Rocket Propulsion Laboratory

(RPL) initiated development of a low-cost autonomous kinetic kill response to the small
drone threat. This project features a rocket-powered vehicle (RPV) built from the ground
up by RPL students and staff using COTS solid rocket motors (SRM) and commercially
available guidance systems and avionics. An early model of the RPV is depicted in Figure
2
2. Early iterations of the project focused on the aerodynamics, structure design, booster
separation, and guidance, navigation, and control (GNC) needed to maneuver the vehicle
to the target location. Such work was initiated by ENS Fletcher Rydalch, USN in 2016 and
refined by CAPT Kai Grohe, Royal Canadian Air Force (RCAF) the following year.
Follow-on work began to focus on the design of bomblet submunitions that are
intended to be deployed from the RPV sled following apogee booster separation. CAPT
Keith Lobo, RCAF designed the bomblet depicted in Figure 3. The submunitions are
designed to fall under the control of maneuverable air brake fins and terminally self-guide
into the target using onboard sensors.
Figure 2. 2017 model of NPS RPV. Source: [4].

3
Figure 3. 3D bomblet model. Source: [5].
Figure 4 visually depicts the concept of employment for the RPV. The intent is to
provide a fire-and-forget quick-launch response capability on a single-stage solid rocket
booster given a general location. The RPV will maneuver to a given latitude, longitude,
and altitude. Following booster separation at apogee, the vehicle sled will fall towards
Earth under a drogue parachute or other aerobraking mechanism. Once sensors in the nose
cone detect drones, the bomblet will be released to intercept the targets.
Figure 4. NPS anti-drone bomblet concept of employment. Source: [6].
4
More recent iterations of the RPV focused on including multiple deployable
bomblets as well as a possible second stage rocket motor. Figures 5 and 6 show the overall
multi-bomblet RPV and 3D model for a deploying mechanism, respectively, as designed
by LT Matthew Busta, USN, and LT Robert Thyberg, USN, in 2019 [7].
Figure 5. 2019 RPV multiple bomblets design. Source: [7].
5
Figure 6. Multi-bomblet deploying mechanism. Source: [7].
The current direction of the project seeks to expand the application of the RPV to
include acting as a rapid-response payload delivery vehicle (RRPD-V) that can be adapted
for various mission sets, as outlined in Figure 7. Mission “A” of Figure 7 depicts the
deployment of a high-altitude communications relay that provides a rapidly deployable
over-the-horizon communications capability. Mission “B” depicts a use concept similar to
the bomblet concept of deployment shown in Figure 4. In this case, however, the bomblet
bay can be changed out for any payload of appropriate size, allowing for low-cost rapid
delivery, including in denied areas.
Figure 7. NPS RRPD-V multi-mission concept. Source: [8].

6
The requirements for the eventual use of these system design options are to provide
a system that is steerable, that can be fired on-demand, and has a minimal manufacturing
cost to keep the threat response cost symmetric.
B. MOTIVATION
The NPS RPL has made several attempts at developing a two-stage solid propellant
rocket. Two attempts in 2018 resulted in mid-air breakups of the RPV at the point where
staging was expected to occur. Although the suspected cause for these breakups was failure
of the interstage coupler, the exact reasons were indeterminate [4]. In both cases, the second
stage SRM failed to ignite.
More recently, a 2020 attempt of a two-stage launch failed less than three seconds
into flight as the bending torque created by the addition of weight in the nosecone to
manipulate the rocket’s center of gravity (CG) proved too high for the booster-to-interstage
mating coupler [6]. Figure 8 shows the moment of failure as captured by video from the
ground. Although improvements to the structural integrity of the system are being
addressed, the desire for reliable ignition of the second stage remains.
Figure 8. Fall 2020 two-stage mid-flight failure. Source: [6].

7
The inclusion of a second stage will add versatility to the RRPD-V in terms of range
and altitude capabilities. Single-stage systems are limited in motor size, and therefore
range, by the diameter of the rocket body. Larger diameter booster SRMs can preclude the
use of aft fin control servos due to space limitations and can complicate GNC by shifting
the vehicle CG aft of the center of pressure (CP), possibly creating an unstable control
condition during motor burn and the subsequent fly-out. Additionally, propellant mass
depletion during flight will cause the CG to shift, introducing potential challenges for the
system guidance and control due to the shift of the CG. Figure 9 illustrates how having a
CG forward of the CP creates a stable condition.
Figure 9. NASA rocket stability demonstration. Source: [9].
One of the complications of introducing a second-stage motor, however, is

inclusion of a reliable mechanism with which to ignite the second motor. In the world of
amateur rocketry, it is common practice to ignite commercial SRMs via a pyrotechnic squib
attached to a semi-flexible lead that is fed through the nozzle, up the aft end of the rocket.
8
The vibration and acceleration loads during boost make a similar aft-end ignition approach
unreliable for the second stage, as the squib may become dislodged and misaligned.
Furthermore, any excess material, such as a semi-flexible lead, can act as a liability,
potentially clogging the nozzle throat as the material is expelled. Even a partial throat clog
for a short duration can cause a jump in motor chamber pressure that could rupture the
motor case. For the purpose of the RRPD-V, the second stage ignition system must also be
reliable, easily implemented, and directly controlled by the onboard avionics electronics.
Implementation of a head-end ignition system will improve the structural integrity

and effectiveness of the ignition squib and also allow for the inclusion of an aft nozzle
enclosure. The nozzle enclosure will hermetically seal the propellant within the motor
chamber during storage. Solid rocket propellants deteriorate over time due to
environmental storage factors, especially from the absorption of water due to humidity.
This deterioration occurs at a faster rate than that of other rocket systems, causing the shelf
life of the rocket to be determined by the longevity of the propellant [10]. A hermetically
sealed chamber will help reduce inconsistency in propellant behavior born from differences
in storage environments and increase the reliability and repeatability among motors that
have been stored for various amounts of time.
A second improvement gained from implementing a nozzle seal is to promote a

rapid ignition sequence by designing the sealing enclosure such that it allows for a rapid
buildup of pressure and yields once a predetermined chamber pressure is reached. COTS
SRMs typically exhibit longer ignition transient times, between when the ignition is
triggered and when maximum thrust is achieved, than do tactical military SRMs, which
always include such enclosures. Figure 10 shows the representative thrust curve, as
provided by the manufacturer, of the Cesaroni M1540 75mm five-grain SRM used for this
work.
The curve exhibits an underdamped time response to ignition before following a

regressive burn pattern until the propellant is exhausted. For this specific propellant
mixture, the peak thrust is not reached until approximately 0.33 seconds post-ignition
without including any delay between the ignition squib firing and the initial buildup of
pressure. This is the point at which the maximum available propellant surface area is
9
burning for a regressive burn propellant geometry. Prior to this peak, the motor will eject
propellant mass at off-design chamber conditions. A nozzle seal enclosure helps minimize
the escape of mass from the chamber during the ignition transient with the goal of
producing a rapid and reliable ignition event with a thrust curve that more closely
resembles a Heaviside step function.
Thrust (N)
Time (s)
Figure 10. Cesaroni M1540 representative thrust curve. Adapted from [11].
Finally, the inclusion of a blast tube will allow for the use of larger diameter motors
that can maximize volumetric efficiency while maintaining a minimum-diameter rocket
body. In an aft fin-control guided booster stage, the SRM must contend for radial space
with the fin control servos. A blast tube would function as an extension of the motor
chamber prior to the nozzle throat with a reduced diameter that will allow the flow of a
large-diameter SRM to be directed to the nozzle through a diameter that allows for the
volume requirements of fin control servos around the blast tube.
10
A secondary effect of a blast tube is that it can favorably shift the CG of the motor,
and therefore the RPV, forward. Because of the significant mass of the SRM with respect
to the rest of the rocket components, its placement has a large impact on the location of the
overall vehicle CG. If the CG is too far aft, the rocket will be unstable. Figures 11 and 12
depict the location of the CG and CP of a hypothetical single-stage RPV with and without
a blast tube.
Blast tube
CG
CP
Figure 11. Hypothetical rocket model with a blast tube (stable).

Adapted from [12].
CG
CP
Figure 12. Hypothetical rocket model without a blast tube (unstable). Adapted
from [12].
C. OBJECTIVES
This study aims to support the NPS RPL rapid-response payload delivery vehicle
by providing low-cost modifications to COTS rocket components to enhance their
performance and applicability for tactical missions. To that end, the objectives of this effort
include:
11
• Design and implement a head-end ignition system for use in a multi-stage rocket
that utilizes COTS solid rocket motors.
• Design and implement an aft nozzle enclosure along with an in-flight rocket motor
chamber pressure monitoring system.
• Propose a design geometry and material for a blast tube for future manufacture and
integration between the COTS solid rocket motor and nozzle.
12
II. APPROACH
A. HEAD-END IGNITION SYSTEM
The design and integration of an autonomous head-end ignition system for the RPV
second stage motor leveraged as much of the vendor-provided hardware as possible.
Experimental ground tests guided the design such that the system could reliably ignite a
solid propellant motor, and follow-on flight tests demonstrated the system’s ability to
operate effectively under flight conditions. 3D printing was identified as the preferred
method of manufacture of consumable components, as it allowed low-cost, quick
prototyping with resources that were available within NPS and specifically at the Rocket
Propulsion Lab. Criteria for material selection included strength, melting temperature,
ductility, and ease of printing. The squib holder geometry was selected based on ease of
integration into existing COTS hardware and performance characteristics of two primary
holder categories. These categories included a directed-flow squib housing and an open,
unrestricted holder. The ignition squib holders were subjected to laboratory testing and
evaluation, in which high-speed video recording was used to analyze a series of ignition
sequence characteristics. These characteristics included size, duration, and direction of the
flow of ejected hot material. The laboratory tests also ensured that the system was capable
of being triggered with electronic hardware that was onboard the rocket. A squib holder
design was then chosen for use in a static ground live-fire test of a commercial SRM and
was later integrated into the RPV second stage for in-flight testing of the autonomous
sustainer ignition capability.
B. NOZZLE ENCLOSURE SYSTEM
All tactical missile systems have some way of isolating the propellant in an SRM
from ambient conditions. The enclosure designs often involve a plug or a cap placed in or
around the nozzle to create a hermetic seal. This enclosure device was designed to be
ejected once a predetermined chamber pressure was reached.
The designs that were considered were developed using 3D modeling software
based on the geometry of the COTS rocket nozzles and then 3D printed. These prototypes
13
underwent a series of laboratory pressure tests, static motor live-fire tests, and rocket flight
tests. The selection of the final design that was integrated into the RPV was based on
repeatability of the blowout conditions, as determined by the laboratory pressure tests, and
performance measurements taken during the static motor live-fire tests.
The blowout condition was determined based on the pressure created by the squib
ignition process. The data was obtained by initiating an ignition squib within an enclosed
space representative of the open volume within the motor chamber. This volume was
monitored by a pressure transducer that measured the increase in pressure that could be
expected within a sealed motor. The blowout condition was then set at some margin above
this pressure to avoid premature enclosure ejection.
The seal was made via O-ring contact with the nozzle. This study investigated
methods of contact to include a plug that contacted the internal surface of the nozzle and
an exterior cap that sealed around the outer surface of the nozzle. High-speed video
recording and a chamber pressure monitoring system were used during static live-fire tests
to analyze and compare the ability of each design to maintain a seal against gas leaks during
the ignition process.
Several methods of securing the nozzle enclosure to the rocket were also
investigated. This portion of the enclosure system was designed to yield at a desired force
due to the applied pressure. The application of nylon shear pins as well as 3D printed
leverage arms were tested. Determination of the preferred securing method was made based
on the ability of each method to prevent leakage and the repeatability of achieving the
desired yield point. Repeatability testing was conducted in a laboratory setting by
regulating the pressure in a controlled space. High-speed video recording and pressure
monitoring observed the securing mechanism’s ability to prevent leaks during static live-
fire tests.
C. BLAST TUBE
A blast tube design was investigated in order to demonstrate the proof of concept
for incorporating a custom-designed blast tube into a commercial rocket motor and identify
a material with which it may be manufactured in the future. Industry standards were
14
researched and leveraged in order to design a geometry and provide a recommended
manufacturing method.
15
16
III. DESIGN PROCESSES
The design for the head-end ignition system was born of the need for a reliable,
autonomous second-stage ignitor and the simultaneous requirement to seal the aft end of
the SRM. Two main requirements were identified at the outset of the design process—the
ignitor squib needed to remain secure during the booster phase of flight, and the system
needed to reliably light the solid propellant. A Cesaroni Technology Inc. (CTI) commercial
squib was secured to 2 g of ammonium perchlorate-based solid propellant, as shown in
Figure 13. The head-end ignition system and accompanying electronic hardware needed to
be able to deliver the required voltage and current to the squib and also be compatible with
the existing forward retaining hardware, imaged in Figures 14 and 15. Figure 16 shows
how the wire leads feed through the forward hardware.
2 g of solid propellant
Commercial squib
Figure 13. Ignition squib with 2 g of solid propellant attached
17
Mount screw hole
Hole for ignition

leads
Figure 14. Smoke charge replacement hardware
Squib support
mount hole
Figure 15. Smoke charge replacement hardware within 4-grain SRM 3D

model cutaway
18
Figure 16. Head-end ignition leads post-fire
Commercially available JBKwik epoxy was used to bond the holder and wires to
create a pressure seal between the wire leads and the forward retaining hardware. The
epoxy was selected for its 15-minute set time, which was ideal for quick launch site
assembly.
1. Ignition Squib Holder Version 1
Early iterations of the system focused on a 3D printed perforated basket that was
designed to hold the squib. The perforations consisted of holes meant to create hot jets
from the squib material that would be directed at and ignite the propellant. Polycarbonate
was selected as the printing material for its high melting temperature (300°C [13]) vs. that
of PLA (151°C [14]). The intent behind this selection was to ensure the integrity of the
basket throughout the ignition and combustion process. Several variations on basket size
and hole placement were investigated. An example of this design is given in Figure 17.
19
12x 0.159 cm
(1/16 in) holes
Squib
5.08 cm
(2 in)
Figure 17. Squib basket 3D model hole configuration C1
After the design was printed, the holes were drilled out with a 0.159 cm (1/16 in)
drill bit. The basket variation depicted in Figure 17 featured eight radial holes—four sets
of two vertically-aligned holes separated by 90°. Four additional holes were positioned on
the tip of the basket in azimuthal alignment with the pairs. The basket tip was angled at 45°
from vertical in order to give the hot jets an axial vector component. Placement of the holes
intentionally began 1.91 cm (3/4 in) above the base of the basket to ensure the jets made
contact with the propellant without striking the motor forward retaining hardware directly.
The two other hole configurations that were tested are shown in Figure 18.
20
a) b)
12x 0.159 cm
(1/16 in) holes
0.86 cm 0.86 cm
(0.34 in) (0.34 in)
Figure 18. a) Hole configuration C2. b) Hole configuration C3
Hole configurations C2 and C3 incorporated axial and azimuthal offsets,

respectively, between the holes in an attempt to achieve greater propellant surface area
coverage by the jets. Configurations C2 and C3 each had 12 holes.
A 7.62 cm (3 in) long version of the basket was also tested with the same hole
configurations. The extra length of this version allowed both for a larger squib and for a
third row of holes as shown in Figure 19. Each configuration was tested with a squib-
propellant combination as imaged in Figure 13. With the squib inserted into the basket, the
ignition lead wires were fed through the forward retaining hardware assembly and secured
with JBKwik epoxy cold weld. Figure 20 shows a profile view of the test article assembly
secured in the test stand.
21
a) b) c)
0.159 cm
(1/16 in) holes
7.62 cm
1.75 cm 1.75 cm 0.86 cm (3 in)
(0.69 in) (0.69 in) (0.34 in)
Figure 19. 7.62 cm (3 in) basket in configurations a) C1, b) C2, and c) C3
Figure 20. Test stand setup with 7.62 cm (3 in) C1 basket
22
Figure 21. High-speed camera and test stand setup
High speed video of the ignition process was recorded at a rate of 10,000 frames
per second. These videos were used to evaluate the performance of each basket based on
jet development reponse time, jet fullness, and jet duration. The video brightness was
reduced to better visualize the hot gas jets. Figure 22 shows a still image of the jets
produced by a 7.62 cm (3 in) basket with 0.159 cm (1/16 in) holes. The gas jets extruding
from the basket tip were clearly visible, while the radial holes remained mostly dark. Visual
inspection of the basket post-firing revealed that the ignition of the squib created slag that
clogged many of the holes, preventing the gas from escaping, as can be seen in Figure 23.
23
Jets
Basket
Figure 22. 7.62 cm (3 in) basket with 0.159 cm (1/16) in holes high-speed
camera still frame
Figure 23. Post-fire 7.62 cm (3 in) basket with clogged holes

24
To remedy the clogging issue, the hole diameters were enlarged for subsequent
tests. A combination of 0.238 cm (3/32 in) and 0.318 cm (1/8 in) holes were used for the
test of the 5.08 cm (2 in) basket shown in Figure 24a. The bottom row of holes was drilled
out to a 0.318 cm (1/8 in) diameter, while the rest were drilled out to 0.238 cm (3/32 in).
Figure 24b highlights the contrast between the hole sizes. While neither size resulted in
clogging, the larger holes allowed for a significant increase in mass flow and a
corresponding increase in the size of the jets.
a) b)
Figure 24. 5.08 cm (2 in) basket with combination hole sizes a) before and b)
during ignition
Because of the length of the first version of the basket, concerns existed over a
potential nozzle throat clog in the event the basket became dislodged during the motor
burn. The relationship between chamber pressure P c and the nozzle throat area A th for
steady-state operation of a solid rocket motor is given by Equation 1.
1
 Ab a ρ p C *  (1− n )
Pc =   (1)
 Ath g c 
The burn rate constant a, the propellant density ρ p , the characteristic exhaust
velocity C*, and the gravitational constant g c , can be considered constant for a given
25
propellant. The propellant burn area A b is also close to constant for a neutral burn profile.
Therefore, it follows that
1
 1  (1− n )
Pc ∝   (2)
 Ath 
where n is the burn rate exponent specific to the propellant chemical makeup and is a value
less than 1. Typical n values for a high burn rate solid propellant grain are given in
Table 1.
Table 1. n values at various pressure ranges. Adapted from [15].
Pressure Range (MPa) n-Variation

6.865-11.77 0.35-0.36
11.77-14.71 0.44-0.45
14.71-19.61 0.47-0.48
19.61-22.56 0.58-0.59
The relationship in Equation 2 reveals that as A th is reduced, the pressure within the
chamber increases substantially. According to Equation 2, an SRM with a propellant burn
rate exponent of 0.4 that has its nozzle throat area reduced by half will experience a 317%
increase in chamber pressure. Assuming a standard factor of safety (FS) for expendable
launch vehicles of 1.4 [16], the motor case would certainly rupture. Even a small blockage
that results in a pressure build below the FS margin could still result in undesirable nozzle
flow which could lead to flow separation from the nozzle walls. This could cause
unpredictable thrust vectoring and loss of control.
It was therefore imperative that the head-end ignition system be redesigned such
that the dislocation of part or all of the squib holder during motor burn would not result in
nozzle throat blockage. This limited the size of the ignition basket to less than the diameter
of the nozzle throat, measured as 2.45 cm (0.966 in). This axial length constraint eliminated
26
the possibility of including radial jet holes. Two different methods for compensating for
the reduction in the number of holes were tested. The first approach was to increase the
size of the hole diameters to the maximum that space would allow, which was 0.476 cm
(3/16 in). Four holes of this size created a total hole surface area of 0.710 cm2 (0.110 in2)
or 150% of the hole surface area of the ignition basket shown in Figure 19. The second
approach incorporated four additional holes, making a total of eight, each drilled out to a
0.318 cm (1/8 in) diameter. The resulting total surface area was 0.632 cm2 (0.098 in2).
Figure 25 depicts a 3D model of the new ignition basket variant with the eight-hole
arrangement.
8x 0.318 cm
(1/8 in) holes
2.45 cm
(0.966 in)
Figure 25. Ignition basket version 2 with 8 holes
Each of the new ignition basket variants were tested in the same manner as the
version 1 variants. High-speed video capture was used to compare the performance of the
hot gas jets. Figures 26 and 27 show the comparable size and thickness of the jets as well
as the hot particulate that the squib expelled.
27
2 cm
speed camera still frame
2 cm
speed camera still frame
28
While the images showed that the ignition basket with 0.476 cm (3/16 in) holes
created wider jets, the length and duration of the jets did not change appreciably between
the two tests. The most notable difference between the two variants was the spread of the
expelled material. The 0.318 cm (1/8 in) hole variant created a much more evenly dispersed
cone of hot gas and particulate, offering the advantage of striking a larger portion of the
propellant grain surface area. The 0.318 cm (1/8 in) hole variant was chosen for static live-
fire testing for this reason. Because of the reduced ignition basket size, however, a smaller
squib had to be used. In static SRM tests, the smaller squib failed to expel enough energetic
material to light the propellant. A second squib needed to be inserted through the nozzle in
order to ignite the motor. Figure 28 shows the SRM in the test stand.
Figure 28. Static SRM live-fire test stand setup
Post-fire visual inspection of the test stand and motor revealed another critical
failure in the current design; the ignition system failed to prevent a leak through the head-
end hardware. Figures 29 and 30 show the clearly visible scorch marks left by the leak.
Such a leak during flight would likely destroy the flight electronics that reside forward of
29
the motor and cause a catastrophic failure. The cause of the leak was determined to be an
imperfect seal in the port of the forward retaining hardware, shown in Figure 30, through
which the wire leads were fed. These test results led to the design of version 3 of the
ignition system.
Evidence of
vented products
Figure 29. Test stand head-end after live-fire test
30
Evidence of
vented products
Figure 30. Exterior of motor head-end retaining hardware after live-fire test
The design for version 3 of the ignition squib housing focused on allowing the use
of a larger amount of energetic material without introducing the risk of throat blockage.
Critically, the design also needed to assure against leakage through the forward retaining
hardware. To meet these requirements, the design approach was changed to an open holder
that secured the squib in place with semi-flexible supports, shown in Figure 32, rather than
a perforated basket. This approach prevented the limitations that the basket housing
imposed on the size of the squib. To accommodate the size of squib pictured in Figure 13,
however, the holder length needed to be longer than the 2.45 cm (0.9660 in) diameter of
the nozzle throat. To reduce the risk of throat blockage, the supports needed to be designed
to melt in the extreme temperature of the motor chamber. Figure 31 shows that most of the
polycarbonate basket remained intact but had started to melt away from the base,
demonstrating the threat of nozzle clogging from large debris posed by this design. It was
therefore necessary to change the material with which the housing was printed. PLA, which
has a melting temperature of about half that of polycarbonate [13, 14], was selected for this
purpose.
31
Figure 31. Ignition basket version 2 after live-fire test
Squib
Support
Ignition leads
Figure 32. Semi-flexible squib holder supports
The final design consideration focused on how the squib holder seated into the
COTS forward hardware and prevent the escape of gas through the head end. Rather than
rely on the epoxy seal between the wire leads and the forward hardware alone, the new
design fit into the smoke charge replacement puck and the wire leads were split before
32
being threaded through the holder and individually sealed with epoxy. Figure 33 shows a
3D model cutaway of the squib holder seated in the smoke chare replacement puck.
JBKwik was again selected as the bonding agent for the wire leads. A modified
version of the holder, shown in Figure 34, was printed for laboratory pressure testing. The
wire leads were coated in JBKwik and individually threaded through the holes in the test
article. Additional JBKwik was added to where the wires exited the article to ensure that
no gaps existed. Laboratory tests at room temperature demonstrated that JBKwik was able
to withstand at least 5.17 MPa (750 psi) without any leaks.
Squib holder
Smoke charge replacement

E-match leads hardware
O-ring (into page)
Figure 33. 3D model of squib holder within smoke charge replacement puck
33
Ignition lead
feed-through
Figure 34. Vacuum chamber test article
The final design of the squib holder is shown in Figure 35. The holder expanded
into a larger diameter after emerging from the smoke charge replacement puck via a filleted
section that was designed to wedge into the hole in the puck as pressure rose in the motor
chamber, improving the sealing mechanism. The long supports are meant to pinch the squib
in place and provide support during the booster-phase vibration. The thin supports were
designed to quickly melt and be expelled from the chamber.
3.81 cm
(1.5 in)
Ignition lead
feed-through
Figure 35. Ignition squib holder final design

34
The primary driving requirements for the design of the nozzle enclosure system
were the needs for reliable blowout at the desired pressure and to create a hermetic seal
under storage conditions for propellant longevity. Maintaining a complete seal once the
propellant began to burn, however, was not as important so long as the system maintained
the ability to rapidly pressurize.
After observing the performance of the polycarbonate squib basket under

combustion conditions, polycarbonate was again chosen as the material with which the
nozzle enclosure would be 3D printed. As the entirety of the enclosure system existed aft
of the nozzle throat, there was no concern about throat blockage. The ability for the material
to withstand high temperature and force for a short duration with minimal distortion was
desirable.
The system consisted of two parts—a retainer and the nozzle enclosure that had an
O-ring contact with the nozzle surface and was ejected at the desired pressure. Because
there was intended to be no or minimal escape of gas through the nozzle, the pressure
initially pressing on the nozzle enclosure was treated as equal to the stagnation pressure
within the motor chamber.
1. Nozzle Enclosure Plug
The initial design for the nozzle cap considered a plug that contacted the internal
surface in the supersonic (diverging) portion of the nozzle versus an external cap that fits
around the nozzle. Figure 36 depicts a generic converging-diverging (CD) nozzle.
35
Figure 36. Geometry of a converging-diverging (CD) nozzle. Source: [17].
The major benefit to the plug design, an example of which is pictured in Figure 37,
was that it allowed for the precise control of the force required to eject the plug at the
desired blowout chamber pressure. The pressure surface, in this case, was the projected 2D
surface created by the O-ring seal. This is due to the fact that all radial components of the
force acting normal to the plug surface are cancelled out as a result of the plug’s
axisymmetric design.
Figure 37. Plug cap 3D model
36
The design of the plug, however, required meticulous measurement of the geometry
of the COTS nozzle to be used. This limited motor selection for the sustainer phase to the
motors for which the RPL had the hardware on-hand. The design shown in Figure 37 was
based on the nozzle for the Cesaroni M3400 98mm SRM. Once the plug shape was
modeled, the location and size of the O-ring was determined using the following equation,
where D O-ring is the O-ring seal diameter, F is the force at which the failure mechanism
releases the plug, and P is the motor chamber gauge pressure at blowout.
F
DO − ring = 2 (3)
πP
For testing purposes, F was set at 222.4 N (50 lbf) and P was set to 103.4 kPa (15
psi). The O-ring groove was, therefore, placed at the location on the plug where the
diameter was 5.232 cm (2.06 in).
2. Nozzle Enclosure Cap
The benefit of an enclosure cap that fits around the exterior of the COTS nozzle
was that it offered a much more modular design that was agnostic to the internal geometry
of the specific nozzle. The cap enclosure system could theoretically be used with any
nozzle of the same diameter with only minimal adjustments made to the enclosure retainer
based on the extent to which the nozzle protruded from the motor case. Figures 38a and
38b show examples of a recessed and a protruding nozzle, respectively. This modularity
greatly improved motor selection possibility. The cap design also required far fewer nozzle
measurements. Only the outer diameter and protrusion of the nozzle was needed. This
reduced risk of a compromised seal due to imperfect enclosure fit and allowed for the
design of caps for nozzles that were not on-hand.
37
Motor retainer collar
a) b)
Aft motor hardware
Figure 38. Example of a) recessed and b) protruding nozzles
Because the O-ring in the cap design made contact with the outer surface of the
nozzle, the pressure face was determined by the nozzle exit area plus the thickness of the
nozzle walls. In the case of the Cesaroni M1540 75mm SRM, for which the cap shown in
Figure 39 was designed, the pressure face had an area of 15.9 cm2 (2.46 in2).
O-ring groove
Shear pin hole
Figure 39. Nozzle enclosure cap 3D model
38
3. Enclosure Retainer
An enclosure retainer that included a 3D-printed arm, which was intended to bolt
on to the aft bulkhead of the rocket and hold the enclosure in place, was briefly considered.
Figure 40 depicts the positioning of the arm in relation to the enclosure plug/cap.
Fastener
Holder arm
Stress concentrator
notch
Enclosure plug/cap
Figure 40. Enclosure retainer arm and plug/cap
The intent was for the arm to contact the center of the plug/cap. The force of the
pressure acting on the cap was transferred along the arm to a stress concentrator notch that
acted as the failure point. Polycarbonate was chosen as the printing material due to its high
brittleness. Despite this, laboratory strength testing caused a significant amount of bending
in the arm prior to failure. The strength testing was conducted by hanging increasingly
heavier weights from the end of the arm. The holder was fastened to a sturdy table in a
manner representative of the way in which it would be fastened to the bulkhead of a rocket.
It was determined that the flex in the arm would allow the enclosure plug/cap to tilt
to one direction, which would have allowed a significant escape of gas prior to blowout.
Additionally, designing to a specified failure force proved to be a tedious and imprecise
process of trial and error, with each print taking approximately eight hours to complete.
Because of these issues, the arm design was ultimately abandoned in favor of a more
39
reliable securing method involving commercial nylon shear pins which failed in a more
repeatable manner.
Experimentally obtained shear limits of #2 nylon pins in various configurations are

provided in Table 2. Average peak load for a configuration with two shear pins was 109.6
N (24.64 lbf) with a standard deviation of 6.51 N (1.46 lbf). Configurations of three and
four shear pins averaged 95.24 N (21.41 lbf) and 95.01 N (21.36 lbf), respectively. Of note,
the standard deviation of the final failure loads decreased as the number of pins is increased.
Three and four-pin configurations had standard deviations of 4.48 N (1.01 lbf) and 4.28 N
(0.96 lbf), respectively, revealing that as the number of shear pins increased, the system
became less sensitive to the defects and differences of individual pins.
Table 2. Shear limits of #2 nylon pins. Adapted from [18].
#2 Nylon Screws
# of Pins Peak Load in N (lbf) Peak Load (Each Pin)

2 236.3 (53.12) 118.1 (26.56)
2 204.4 (45.95) 102.2 (22.98)
2 226.2 (50.85) 113.0 (25.42)
2 230.4 (51.80) 115.2 (25.90)
2 213.2 (47.92) 106.6 (23.96)
Avg 222.1 (49.93) 109.6 (24.64)
3 278.6 (62.64) 92.88 (20.88)

3 269.4 (60.57) 89.81 (20.19)
3 286.4 (64.40) 95.50 (21.47)
3 277.6 (62.41) 92.52 (20.80)
3 305.9 (68.76) 102.0 (22.92)
3 296.4 (66.64) 98.79 (22.21)
Avg 285.8 (64.24) 95.24 (21.41)
4 385.7 (86.70) 96.39 (21.67)

4 386.4 (86.86) 96.57 (21.71)
4 350.4 (78.77) 87.59 (19.69)
4 374.9 (84.27) 93.72 (21.07)
4 380.8 (85.62) 95.19 (21.40)
4 402.1 (90.40) 100.5 (22.60)
Avg 375.6 (84.44) 95.01 (21.36)
40
Figure 41 shows an exploded 3D cutaway view of the concept for the shear pin
retainer and cap. Each component was drilled with symmetrically-spaced holes for the
shear pins. The retainer took the shape of a collar that fit around the outer surface of the
COTS nozzle. The cap fit between the nozzle and the collar such that the shear pin holes
in the cap and the collar aligned. The collar was designed to fit within the existing motor
retaining hardware without the need for any modification to the aft bulkhead.
Figure 41. Shear pin retainer collar and cap
C. BLAST TUBE
The proposed blast tube design was based on the estimated geometry of the
Cesaroni O8000 150 mm rocket motor. The O8000 was selected because it offered the
highest thrust from a COTS motor that fit within the existing RRPD-V body. It was the
largest available COTS SRM, providing more than 8,007 N (1800 lbf) of initial thrust and
was “originally designed to boost a [227 kg (500 lb)] experimental cruise missile to flight
speed” [19]. Because of its diameter, the O8000 precluded the use of aft fin control servos
in the RRPD-V without incorporation of a blast tube.
The final blast tube geometry would be dependent on the constraint imposed by the
size of the servos needed to control the aft fins of the rocket. Other size and weight
41
characteristics of the RRPD-V at the time of design should also be considered to ensure
the shift in CG that results from incorporating the blast tube could be favorable. What
should remain constant within the design, however, is that the inner port diameter should
remain larger than the nozzle throat diameter. The blast tube was, thus, designed foremost
around this parameter. Because the COTS O8000 motor nozzle was not available to be
measured, the geometry had to be estimated. Based on information provided by CTI, the
nozzle throat was estimated to have a diameter of 3.0 cm (1.2 in). Measurements taken
from the nozzle of an M3400, which used the same White Thunder propellant mixture as
the O8000, were used to estimate the O8000 nozzle expansion ratio. The O8000 propellant
grains were estimated to have an initial inner diameter of 5.08 cm (2 in). This diameter was
also used as the inner blast tube diameter forward of the nozzle. Figure 42 shows a 3D
model of a blast tube designed to fit into the existing O8000 motor case. The shape was
intended to be able to replace the existing COTS nozzle without any modifications to the
motor case. Figure 43 depicts how the blast tube would fit into a fully-assembled O8000
motor.
Figure 42. 3D model cutaway of blast tube based on the O8000 motor
42
Available volume
for servos/
batteries
Propellant grains
Blast tube
Outer pressure
vessel case
Figure 43. 3D model of blast tube assembled into O8000 motor
A common practice in amateur solid propellant rocketry is to machine nozzles out

of solid micro fine-particle graphite rod. These rods, however, become exceedingly
expensive as diameter increases. The size of rod needed to machine a blast tube for a 150
mm diameter motor would need to be special ordered and would cost thousands of dollars.
An alternative method of manufacture is to wrap carbon-phenolic tape around a mandrel
that is pre-machined to the desired geometry of the blast tube. The NASA Marshal Space
Flight Center (MSFC) described this manufacturing process:
The carbon cloth used to fabricate composite solid rocket motor nozzles is
impregnated with the binder or matrix prior to wrap and cure. This
preimpregnated material is commonly called “prepreg” in the composite
industry. The diversity of the manufacturing process requires six different
vendors before final material is produced. These vendors: 1) produce rayon
thread; 2) weave cloth; 3) carbonize cloth; 4) produce resin; 5) produce
43
carbon fillers; and 6) impregnate carbon cloth with resin and filler
(production of prepreg). Constant monitoring of all phases of the
manufacturing process is required to ensure satisfactory quality. The rayon
thread is manufactured, then woven into cloth 60 in. wide. The rayon cloth
is carbonized by slowly heating to 1000 deg. to 1500 deg. C in an inert
atmosphere. Critical factors to be controlled in this process are the rate of
temperature increase, time, and maintenance of an inert atmosphere in the
oven.
After carbonization, the carbon-cloth is impregnated by drawing it through

a heated container of phenolic resin and carbon filler to form prepreg.
Critical factors that must be carefully controlled in the preimpregnation
process are temperature of the resin/filler mixture, tension and speed of the
cloth through the resin/filler mixture, pressure on the roller to ensure
penetration of resin into the cloth, oven temperature after impregnation to
remove volatiles, and control staging of the resin. [20]
MSFC also described the tape wrapping process. In the case of the blast tube, the mandrel
would likely need to consist of at least two sections; one for the converging portion of the
tube/nozzle and one for the diverging section.
The tape wrapping process places the tape at the proper angle and debulks
the tape material to minimize movement of the tape during cure. Debulking
of the tape should be achieved by applying heat and pressure at the point of
contact with the mandrel or the previous ply. Heat should be applied prior
to wrapping to make the tape tacky. The pressure roller forces the fibers to
nest and compact (debulk) within the resin/fabric matrix. CO2 is used to
cool the resin to stop further cure, and dimensionally and thermally
stabilizes the billet for further processing…After wrapping, the billet and
mandrel are vacuum bagged for waterproofing, then installed in a
hydroclave for final cure. Curing of the carbon-cloth liner requires a
pressure of 1000 psi at 310 deg. F for a minimum of five hours. After the
cure cycle, a test ring should be removed and tested to verify the properties
of the cured carbon cloth phenolic. The carbon cloth is then machined to
configuration. [20]
Additional layers of phenolic tape could then be wrapped by first applying a coat of
phenolic resin to the previous layer. At each consecutive stage, the component should be
vacuum-bagged and cured at a pressure of 1.72 MPa (250 psi) and a temperature of 154 °C
[20]. Table 3 lists the properties of the cured carbon phenolic wrap at room temperature
unless otherwise stated.
44
Table 3. Cured material physical and mechanical properties of carbon
phenolic. Adapted from [20].
Limits
Property
Minimum Maximum
Density (g/c3) 1.4 1.52
Resin content (%) 30.0 38.5
Compressive strength (MPa), Wrap direction 172 448
edgewise Fill direction 138 379
Interlaminar double sheer strength (MPa) 24 55
Across ply 0.17 1.9

Thermal conductivity (W/m-K)
@ 121 °C
With ply 0.17 1.9
Coefficient of thermal Across ply 9.0 36.0

expansion (1/°K x10-6) @
204 °C With ply 3.6 16.2
Wrap direction 172 379

Flexural strength (MPa)
Fill direction 138 379
Tensile strength (MPa), Wrap direction 103 276
edgewise Fill direction 68.9 241
45
46
IV. EXPERIMENTAL RESULTS
A. STATIC MOTOR LIVE-FIRE TESTING
One of the goals of the static live-fire test was to record the thrust and chamber
pressure of a motor with both the head-end ignition system and aft nozzle enclosure in
place. To accomplish this, a pressure transducer was connected through the forward
retaining hardware of an M3400 motor using Swagelok tubing. The chamber pressure of
the motor was not provided by the manufacturer, so a conservative approach was taken in
selecting the pressure transducer. In order to avoid transducer saturation, a 20.7 MPa (3000
psi) transducer was selected. The same thrust stand that is pictured in Figure 28 was used
for this test.
The data from the live fire are graphically depicted in Figures 44, 45, and 46. A fair
amount of 60 Hz electronic noise was present in the testing equipment and needed to be
filtered out. To do this, the raw data .tdms files were converted to Microsoft Excel
spreadsheets and imported into Matlab. The data were then run through a band stop IIR
filter and adjusted for the offset imposed by the data acquisition equipment. Figures 44 and
45 depict the data before and after filtering for pressure and thrust, respectively.
The phase shift between the measured pressure and thrust data was due to the extra
processing time incurred as a result of the conditioning electronics through which the load
cell voltages were processed and amplified. The rise in thrust was otherwise expected to
correlate temporally to the rise in pressure. This could also explain the more gradual rise
and drop of the experimental pressure curve compared to that of the manufacturer-provided
curve shown in Figure 47. The measured average thrust was 3505 N. This was on the order
of what was expected, given the manufacturer-provided average thrust of 3421 N [21].
Interestingly, the pressure curve more closely resembled the manufacturer-

provided thrust profile than did the experimental thrust curve. Again, this was likely an
artifact of the conditioning electronics. At any rate, the correlation between the chamber
pressure and the thrust was expected, given the relationship in Equation 4, where T is thrust,
𝑚𝑚̇ is the mass flow out of the motor, v e is the flow exit velocity, P e is the pressure at the
47
exit of the nozzle and is directly tied to chamber pressure, P o is ambient environmental
pressure, and A e is the nozzle exit area. 𝑚𝑚̇, v e , and A e can be assumed nearly constant for
a neutral burn profile.
 e + ( Pe − P0 ) Ae
T = mv (4)
Offset
correction
Figure 44. M3400 chamber pressure vs. time
48
Offset
correction
Figure 45. M3400 thrust vs. time
Figure 46. M3400 pressure and thrust profiles
49
Thrust (N)
Figure 47. M3400 representative thrust curve. Adapted from [21].
The expected pressure response was a profile close to what is circled in green in
Figure 48. The pressure was expected to rise sharply to the blowout pressure, drop slightly
with the expulsion of the nozzle enclosure, then rapidly climb to maximum pressure.
However, the response highlighted in Figure 48 fell within the 1% uncertainty associated
with the pressure transducer. It was, therefore, impossible to make a conclusion about the
performance of the nozzle enclosure from these data. What was able to be gleaned,
however, was the range of chamber pressure that could be expected in subsequent firings.
This informed the selection of pressure transducers for follow-on tests to ensure the
expected nozzle enclosure blowout pressure fell outside of the uncertainty of the transducer
while also avoiding saturating the transducer.
50
Figure 48. Zoomed in M3400 pressure vs. time
B. NOZZLE CAP LABORATORY TESTING
Laboratory testing of the nozzle cap consisted of two parts—the firing of a squib
inside a known closed volume and shear pin blowout tests. The closed volume squib test
was conducted first to determine the pressure increase created by just the ignition of the
squib. Swagelok tubing and fittings were used to create a volume representative of an
M1540/M1400 75 mm motor chamber. A squib was sealed into one end, and a 6895 kPa
(1000 psi) pressure transducer was fixed to the other. Figure 49 shows the test article.
Figure 50 shows a graphical representation of the data obtained from the pressure
transducer. A peak gauge pressure of 120.9 kPa was achieved. This value was used to
inform the baseline pressure that the nozzle cap needed to withstand in order to avoid
premature blowout, assuming no initial mass addition from burning propellant.
51
Squib
Pressure
transducer
Volume: 383 cm3
Figure 49. Enclosed squib test apparatus
120 kPa
Volume: 383 cm3

with P75 CTI squib
Ignition
Figure 50. Enclosed squib test pressure vs. time
A subsequent test involved securing an enclosure cap to a motor assembly as it

would be installed during ground or flight testing, excluding propellant or a squib.
52
Compressed air was then rapidly fed through the forward end of the motor case through
tubing that was also connected to a pressure transducer. This was done for two and four
pin configurations. The test configurations were restricted to even pin counts to ensure load
symmetry. Figure 51 depicts the setup of the forward end of the test apparatus.
Pressure
transducer
Motor case
Pressure feed
Figure 51. Nozzle cap test apparatus
The results of these tests are graphically depicted in Figures 52, 53, and 54. In each
of these figures, the blowout pressure is marked by a black horizontal line and labeled with
the pressure value. Although the pressure transducer shown in Figure 51 did locally
measure transient pressures slightly above the blowout pressure, the ultimate point of cap
blowout was clearly represented by a sharp drop in pressure to 0 kPa. The four pin
configuration withstood up to 354.5 kPa, while the six pin configuration withstood on
average 821.5 kPa. The pressures represented FS margins over the base squib ignition
pressure of 2.93 and 6.79, respectively. Both configurations were, therefore, acceptable in
terms of achieving a margin outside of the transducer’s 1% uncertainty. The four pin
configuration was selected for follow-on flight testing in an effort to mitigate the risk of
inadvertently extinguishing the propellent. The slope of the pressure drop off in Figure 52
was noted to help identify the expected pressure response, referenced in Figure 48, for
follow-on tests.
53
354 kPa
Slope = -1,600
kPa/s
Figure 52. Four pin nozzle cap/test run 1 pressure vs. time
814 kPa
Figure 53. Six pin nozzle cap/test run 1 pressure vs. time
54
828 kPa
Figure 54. Six pin nozzle cap/test run 2 pressure vs. time
55
56
V. ROCKET FLIGHT TESTING
A. OCTOBER 2021 TWO-STAGE ROCKET
1. Rocket Vehicle Design
The rocket utilized in this study was separate from the RRPD-V. It was a two-stage,
unguided solid propellant rocket that acted as a test platform for the SRM modifications
developed in this study. The purpose of the October 2021 launch was to demonstrate the
ability of the head-end ignition system to autonomously ignite the second stage SRM with
an installed nozzle cap under flight conditions. OpenRocket was used to model the rocket
prior to construction to evaluate the stability and flight characteristics of the proposed
design. The OpenRocket software was also used to develop the shape, size, and placement
of the booster and sustainer fins. The final OpenRocket model is depicted in Figure 55.
Figure 55. OpenRocket model of rocket design
A minimum stability threshold of 0.95 cal was set for each stage of the rocket. By
manipulating the fin sweep angle, sweep length, tip chord, and position, suitable stability
margins of 1.7 cal for the booster phase and 0.96 cal for the sustainer phase were achieved
for crosswinds up to 5 m/s. The rocket fins were cut from 0.300 cm (0.118 in) thick
fiberglass sheet and were limited to a span of 17.8 cm (7 in). This limitation was due to the
size of the fiberglass sheets available. Additionally, larger fin sizes complicated storage
considerations. Figure 56 shows the booster body components.
57
Figure 56. First stage booster (right) and parachute bay (left)
The booster section of the rocket consisted of a 96.52 cm (38 in) tube that was
connected to a 30.48 cm (12 in) interstage section that housed the parachute bay and the
booster flight electronics. The interstage section slid over an overlapping coupler that
extended from the booster tube and was secured with four #2 nylon shear pins. The
electronics bay was situated forward of the booster engine block and segregated by 1.27
cm (1/2 in) wooden bulkheads on either side. Between the electronics bay and the booster
engine block was the booster parachute bay. The booster was designed to carry most 98
mm diameter motors and included an adapter sleeve, pictured in Figure 57, to
accommodate smaller 75 mm motors, such as the M2075 used in this launch. Figure 58
depicts the OpenRocket model of the sustainer section of the RPV.
98 mm retainer collar
98 mm retainer ring
75 mm adapter sleeve
Figure 57. Booster motor adapter sleeve
58
Figure 58. OpenRocket sustainer section model
The sustainer section consisted of an 85.09 cm (33.5 in) tube that held the second
stage motor, a 75 mm M1400 in this case. A 45.40 cm (17.875 in) section of tube forward
of the sustainer motor housed the drogue parachute, the sustainer electronics bay, and the
sustainer main parachute. Forward of the parachute bay tube was the nose cone.
The main rocket body was built out of 19.05 cm (7.5 in) diameter phenolic
cardboard tube wrapped in a fiberglass shell. The coupler tubes were made of 18.65 cm
(7.343 in) diameter phenolic cardboard tubes. The nose cone was stock fiberglass with a
gel-coat surface. Figure 59 shows the sustainer motor section.
Figure 59. Sustainer motor section
Simulations ran with OpenRocket for the design pictured in Figure 55 generated an
expected apogee for the sustainer of 4,027 m (13,213 ft) above ground level and a
59
maximum speed of 335 m/s (1100 ft/s), or Mach 0.99. Figure 60 shows the OpenRocket
graphical representation of the simulation for the sustainer section of the rocket.
Figure 60. OpenRocket simulation sustainer for October 2021 rocket launch
2. Flight Electronics
The rocket booster used an Altus Metrum TeleMega altimeter for telemetry
recording and parachute deployment. A PerfectFlite StratoLoggerCF altimeter was used as
a backup. Each altimeter was powered individually by separate 9V batteries. The TeleMega
recorded height, speed, acceleration, and vertical acceleration for the boost phase of the
flight, whereas the StratoLoggerCF recorded just altitude and temperature. These
components are pictured in Figures 61 and 62.
60
Figure 61. Altus Metrum TeleMega. Source: [22].
Figure 62. PerfectFlite StratoLoggerCF. Source: [23].
The TeleMega and the StratoLoggerCF were each connected to E-matches that
initiated black powder charges within the booster parachute bay as pictured in Figure 63.
The ignition of the black powder created enough pressure within the compartment to shear
the pins securing the interstage section to the booster, releasing the booster parachute in
the process.
61
Parachute deployment black powder
charges (2 g each)
Figure 63. Internal view of booster parachute bay and aft electronics bulkhead
The sustainer flight electronics included two Altus Metrum EasyMega altimeters,
pictured in Figure 64, for redundancy. Again, each altimeter was powered by a 9V battery,
and each recorded height, speed, acceleration, and vertical acceleration. These EasyMegas
were used to control stage separation, sustainer ignition, sustainer drogue parachute
deployment, and sustainer main parachute deployment.
Figure 64. Altus Metrum EasyMega. Source: [24].
62
Because the sustainer section of the rocket was expected to reach a much higher
apogee than the booster, a drogue parachute was included to reduce the time aloft and the
subsequent recovery distance. The drogue allowed the rocket body to fall at a reduced
velocity until a specified altitude was reached, at which point the main parachute was
deployed. Two separation charges were utilized. E-match leads were fed through a tube
that ran the length of the sustainer motor section to a black powder charge on the outside
of the forward booster electronics bay bulkhead, shown in Figure 65. The other E-match
lead from the EasyMega was fed through the sustainer engine block bulkhead to the head-
end ignition system. These leads were connected to the sustainer electronics bay via barrel
jacks, as shown in Figure 66, to allow for quick disconnection during drogue parachute
deployment. A CO2 separation charge was used for the drogue parachute deployment since
the altitude at which the drogue deploys precludes the use of black powder due to reduced
reactivity.
Stage separation camera
Stage separation black

powder charge
Figure 65. Booster electronics bay forward bulkhead
63
Barrel jack connectors for
stage separation and
sustainer ignition
CO2 separation charge
Figure 66. Internal view of drogue parachute bay and sustainer electronics
bay aft bulkhead
The final electronic component in the sustainer was the Multitronix Kate-1
TelemetryPro Transmitter, shown in Figure 67, that was housed in the nose cone. Kate-1
was a self-contained flight data recorder and transmitter that operated on a one-watt 900
MHz signal [25]. Figure 68 shows an exploded view of all the components of the sustainer
electronics bay and nose cone.
Figure 67. Multitronix Kate-1 TelemetryPro transmitter. Source: [25].
64
Figure 68. Sustainer electronics bay and nose cone exploded view
3. Flight Data and Results
This flight incorporated the head-end ignition and the nozzle enclosure systems into
the sustainer portion of the rocket. Although there was no pressure monitoring system
aboard this flight, there were two cameras onboard that captured the stage separation and
sustainer motor ignition. Video from the onboard cameras as well as video captured from
the ground confirmed successful second stage ignition. Both stages were successfully
recovered with minimal damage and no loss of flight data.
Table 4. Booster flight statistics
Maximum height 936.8 m / 3073 ft
Maximum speed 164.9 m/s / 541 ft/s / Mach 0.5
Max boost accel 106.9 m/s2 / 351 ft/s2 / 10.90 G
Avg boost accel 62.0 m/s2 / 204 ft/s2 / 6.33 G
Ascent time 2.6 s boost 7.4 s coast
Main descent rate 8.5 m/s / 28 ft/s
Descent time 83.2 s
Flight time 93.2 s
The booster stage functioned as expected and had a burn time of about three
seconds, followed by a two-second coast prior to stage separation. After an apogee of 936.7
65
m (3073 ft), the booster section fell under free fall to a height of 600.0 m (1969 ft), at which
point the main parachute was released. Figure 69 shows a graphical representation of
booster acceleration, height, and speed over time.
Boost
Coast
Drogue
Main
Figure 69. Booster stage flight data
Figures 70 and 71 show stills before and after stage separation, indicating
successful staging. Figure 72 shows a still taken from the video captured by the camera
mounted on the outside of the forward booster electronics bay bulkhead shortly after stage
separation occurred. It shows the white nozzle enclosure cap in place on the sustainer
motor.
66
Figure 70. Booster stage burn
Booster section
Figure 71. Moment of separation
67
Nozzle enclosure cap
Figure 72. Sustainer after stage separation
The external aft-facing camera captured the moment that the head-end ignition
system fired. The next frame of the video revealed that the nozzle cap was ejected at that
time as a result of the pressure build from only the ignition of the squib. Figure 74 shows
still frames from these two points. It took nearly four more seconds for the propellant to
fully light and for the sustainer to enter its boost phase.
The nozzle cap was secured to the collar with two shear pins for an expected shear
force of 222 N (50 lbf) based on Table 2. This equated to a total pressure of 131 kPa (19
psi). Because of the elevation change between the launch site, where the motor chamber
was sealed, and the height at sustainer ignition, the change in pressure from elevation
ΔP elevation needed to be accounted for when calculating the chamber pressure for cap
blowout P c (blowout) .
Fshear n pins
P
=c ( blowout ) − ∆Pelevation (5)
Acap
In Equation 5, F shear is the average shear force per pin, n pins is the number of pins,
and A cap is the area of the pressure surface. Atmospheric pressure could be estimated prior
to launch by linearly interpolating data from Table 5. Flight data recorded on the
68
EasyMega, however, provided more accurate pressure readings. Figure 73 depicts the
atmospheric pressure data recorded by the sustainer. At launch, the atmospheric pressure
was 94.0 kPa (13.6 psi). At squib ignition, indicated by the red line labeled “Squib ignition”
in Figure 73, the atmospheric pressure was 85.0 kPa (12.3 psi). This equated to a ΔP elevation
of 9.0 kPa (1.3 psi) and a P c (blowout) of approximately 122 kPa (17.7 psi). The delay between
the ejection of the nozzle cap and the motor entering the boost phase, as corroborated
between the video capture and flight data recording, indicated that two shear pins were not
sufficient to withstand the pressure spike created by the ignition of the squib itself. Further
testing was needed to determine the minimum number of shear pins needed to withstand
the initial increase in pressure.
Table 5. Standard atmosphere properties. Adapted from [26].
Elevation Temperature Pressure Speed of Sound

(m) (°C) (kPa) (m/s)
0 15.1 101 340.3
500 11.9 95.5 338.4
1000 8.7 89.9 336.4
1500 5.4 84.6 334.5
69
Coast
Fast
Boost
Boost
Coast
Stage separation
Squib ignition
Figure 73. Flight 1 sustainer atmospheric pressure and height data
Nozzle enclosure cap
Figure 74. Moment of head-end ignition squib fire
The sustainer portion of the rocket functioned mostly as expected and climbed to a
height of 3,670.7 m (12,043 ft). This was on the order of what was expected based on the
OpenRocket simulation. It should be noted that OpenRocket was only used as a rough
70
estimation and not as a rigid performance metric due to the software’s underlying analytical
model assumptions that can lead to performance variations. One reason why the maximum
height may have been less than what was simulated is that the second stage motor did not
achieve maximum burn instantaneously. As shown in Figure 75, the motor took about four
seconds to reach peak thrust following squib ignition. During this time, the sustainer was
losing vertical speed as it decelerated following booster-stage burnout. One of the goals of
including the nozzle cap on the sustainer motor was to recapture some of the effective
impulse lost during these four seconds while providing a more predictable staging time
sequence.
Of particular note in reference to Figure 75 is the very shallow decline in height

following where drogue deployment is indicated. Visual observation from the ground
confirmed that this was due to the premature deployment of the main parachute at the same
time. The separation charges fired appropriately, as was indicated by the recorded data and
inspection of the charges after recovery. The most likely explanation for the premature
deployment was early failure of the shear pins that secured the nose cone to the parachute
bay. In this case, four shear pins were used to produce an estimated shear force of 445 N
(100 lbf). When the drogue parachute deployed at apogee, the sustainer body likely still
had enough momentum that the sudden deceleration caused the shear pins to yield and
release the main parachute as well. Although this did not cause a catastrophic failure, it
drastically increased the time aloft of the sustainer. This resulted in a drift of several miles
during the decent and exacerbated recovery efforts. To avoid this problem, more shear pins
were needed to secure the parachute bay to the nose cone on follow-on launches.
71
Squib ignition
Drogue
Full second
stage ignition
Figure 75. Flight 1 sustainer stage data
Table 6. Sustainer flight statistics
Maximum height 3670.7 m / 12043 ft
Maximum speed 274.8 m/s / 902 ft/s / Mach 0.8
Max boost accel 105.6 m/s2 / 346 ft/s2 / 10.77 G
Avg boost accel 55.6 m/s2 / 182 ft/s2 / 5.67 G
Ascent time 6.8 s boost 0.7 s fast 24.6 s coast
Descent rate 7.6 m/s / 25 ft/s
Descent time 462.9 s
Flight time 495.0 s
72
B. APRIL 2022 TWO-STAGE ROCKET
1. Rocket Vehicle Design
The April 2022 launch was conducted using the same rocket body as the October
2021 launch. Slight repairs were made to the epoxy fillets that secured the fins in place on
both the booster and sustainer sections of the rocket. Otherwise, only slight modifications
were made to the internal structure to allow a pressure line to feed from the sustainer motor
to a new health monitoring system in the sustainer electronics bay.
The M2075 75 mm motor was exchanged for a more powerful M3400 98 mm motor
for use as the booster motor. This exchange allowed for the testing of the rocket at faster
speeds and greater altitudes as well as for the demonstration of the versatility of the booster
body design. The sustainer motor selected for this launch was the M1540 75 mm SRM.
The M1540 provided slightly higher impulse than the M1400 used in the October 2021
launch but maintained the same nozzle form factor, preventing the need for a redesign of
the nozzle enclosure system. Figure 76 shows the OpenRocket model of the RPV for the
second flight.
Figure 76. OpenRocket model of April 2022 rocket
OpenRocket simulations generated an expected flight pattern very similar to the

October 2021 rocket as shown in Figure 77. The rocket was expected to climb to a new
maximum height of 5,000 m (16,400 ft) and achieve a supersonic maximum velocity of
366 m/s (1200 ft/s, Mach 1.1).
73
Figure 77. OpenRocket simulated flight for April 2022 rocket launch
2. Flight Electronics
This launch maintained many of the same flight electronics from the previous
launch, including the TeleMega and the two EasyMegas. The Kate-1 unit was also kept
aboard this flight. The PerfectFlite altimeter in the booster was exchanged for a second
TeleMega due to an unknown error that prevented communication with the PerfectFlite
when tested during rocket assembly. The major addition to this flight was a modular health
monitoring system that consisted of an Arduino UNO microcontroller board and an
Adafruit Data Logger Shield, an example of which is pictured in Figure 78.
74
Figure 78. Adafruit Data Logger Shield with Arduino UNO. Source: [27].
The health monitoring system for this flight was set up to read and record inputs
from two pressure transducers—one measuring sustainer motor chamber pressure and the
other measuring pressure within the drogue parachute bay during separation. The final
sustainer electronics bay is shown in Figures 79 and 80.
Figure 79. Sustainer avionics bay
75
Figure 80. Sustainer avionics bay side view
3. Flight Data and Results
Although the stage separation and ignition appeared to function appropriately from
the ground, the ground receiver lost contact with the Kate-1 telemetry system at a distance
of 2.57 km (1.6 mi). Ground observers lost visual contact with the sustainer following
sustainer motor burnout, and no parachute was able to be discerned, preventing recovery
of the sustainer section. The loss in connection was thought to be due to the Kate-1
telemetry system not being properly configured to “launch mode.”
Because of this, no quantitative data was collected on the performance of the aft
nozzle enclosure cap. The only information available was anecdotal evidence from ground
observations that suggested the four shear pin variant of the nozzle cap may have
functioned as intended. The ignition of the rocket second stage was observed to occur very
close to the two-second delay from booster burnout that was programmed into the
EasyMega that controlled stage separation and sustainer ignition. The extra post-ignition
delay that was observed in the onboard video captured during the previous launch was not
present in this case. Furthermore, the ignition sound heard by observers on the ground was
a loud, crisp “bang” that suggested there may have been an improvement in the motor thrust
76
time response over the previous launch. These observations, of course, require
confirmation through further testing. Figures 81 and 82 show stills from video captured
from the ground.
Figure 81. First stage boost phase T+2 seconds
a) b) Stage separation c)
Second
stage
plume
T+3s T+4s T+5s
Figure 82. a) Booster burnout, b) stage separation, c) sustainer ignition
77
78
VI. SUMMARY
Several modifications to COTS SRMs, including a head-end ignition system, a

nozzle enclosure system, and a custom-designed blast tube, were investigated in support of
the NPS RRPD-V. These modifications were designed to be manufactured at low cost and
easily integrated into existing commercial motor hardware with minimal modifications to
the hardware itself. To that end, 3D printing was heavily used as the preferred
manufacturing technique. Alternative manufacturing methods were investigated and
suggested for components that could not be 3D printed because of the pressure and thermal
loads expected on the component.
The design of the head-end ignition system focused on three primary competing
versions of a mechanism that held the ignition squib in place. For all versions, the squib
was inserted through the forward motor hardware. The first two versions incorporated a
perforated basket concept that directed hot jets that would contact and ignite the propellant.
The third version was an open holder that did not restrict the size of the squib used or
influence the direction of the expelled material. Ultimately, the open holder version was
selected for flight tests because it allowed for a squib sizeable enough to reliably light the
propellant, as demonstrated during live-fire ground testing, without presenting a significant
threat of nozzle throat clogging. Two two-stage rocket launches demonstrated the ability
of the head-end ignition system to reliably ignite the second-stage motor and be controlled
by the avionics equipment onboard the rocket. The basket approach may still be viable but
would require custom loading squib material into the basket to maximize the space
available.
The nozzle enclosure system was designed to environmentally seal the motor
chamber and propellant for long-term storage. During launch, the enclosure caused
chamber pressure to build to a specified point before yielding and being expelled from the
aft end of the rocket. This was intended to improve the transient ignition process of the
motor by preventing the escape of exhaust prior to the ignition of the full burn area of the
propellant, resulting in a more rapid increase in chamber pressure. Two sealing methods
were tested—a plug that created a seal on the internal surface of the diverging section of
79
the nozzle and a cap that sealed around the external diameter of the nozzle. The main
difference between these methods was the ability of the former to manipulate the pressure
face area and, therefore, more precisely control the blowout chamber pressure. The benefit
of the external seal method was reduced complexity of the enclosure geometry that led to
a more reliable seal and a more modular “one-size-fits-many” system. Control of the
blowout chamber pressure beyond the precision of adjusting the number of shear pins used
to secure the cap to the retainer was deemed unnecessary to attain the desired results. The
sustainer section of the rocket, which housed the pressure monitoring system, was lost
during the second flight test and was not recovered. Anecdotal evidence observed by
personnel at the launch suggested this enclosure method had the intended effect. More
testing, however, is needed to draw firmer conclusions.
The final design effort proposed a design for a blast tube, which will favorably shift
the SRM CG and allow for the simultaneous use of a larger diameter SRM and aft fin
control servos. The design was intended to be seamlessly integrated into COTS hardware
by replacing the commercial nozzle. The internal geometry of the tube was based on
industry standards for COTS rocket nozzles, and the length could be extended or shortened
based on the requirements of the GNC system of the RRPD-V. Finally, a potentially cost
effective method of manufacture was proposed based on NASA practices, which involved
wrapping layers of carbonized phenolic tape around a mandrel machined to the desired
geometry.
80
VII. FUTURE WORK
Although the open version of the squib holder was selected for live-fire and flight
tests for this study, the perforated basket version may still hold promise. Using a basket
squib holder that is small enough to negate the risk of a nozzle clog would require the
Rocket Lab to custom pour its own ignition squibs rather than use commercial pre-made
squibs. This can be done with commercially available material but will require further
testing. Benefits of using the custom-poured basket include higher repeatability in the
energy of the squib ignition due to the more precise control of the amount of squib material
used in the custom pour and quicker and easier assembly of pre-poured baskets into the
head-end hardware.
Further live-fire tests are needed to determine the enhancement, if any, to the
efficiency of the SRM from the inclusion of the nozzle enclosure design. Whether the
follow-on tests are conducted on the ground or in flight, the health monitoring system
developed as part of this study could be used to collect the desired data. Future experiments
may also wish to study how aging affects the propellant by comparing characteristics, such
as burn rate, of samples of propellant that have been stored sealed versus open to
atmospheric conditions.
C. BLAST TUBE
The blast tube will require the most extensive amount of future work, to include
testing of the viability of the recommended, as well as competing, methods of
manufacturing. Structural tests should be conducted to ensure the selected manufacturing
material can withstand expected chamber pressure and thermal loads. Computational fluid
dynamics should also be leveraged to ensure adequate heat absorption such that the
components surrounding the blast tube are insulated from the exhaust flow. The final
81
design will need to be closely tailored to the requirements of the GNC system in order to
ensure optimal SRM CG shift and adequate space for aft fin control servos.
D. ROCKET VEHICLE
Improvements can be made to the avionics equipment to increase redundancy and

minimize risk of loss of data or equipment. In addition to the Kate-1 telemetry system, the
TeleMega altimeters can be fitted with an antenna that allows telemetry data to be
transmitted to a ground station in real time. Correspondence with the TeleMega vendor
also revealed that TeleMegas can be special ordered with additional signal input pins that
could allow data from the health monitoring system to be transmitted to the ground station
in real time.
82
APPENDIX. DATA LOGGER ARDUINO CODE
/*
SD card datalogger
This example shows how to log data from three analog sensors
to an SD card using the SD library.
The circuit:
analog sensors on analog ins 0, 1, and 2
SD card attached to SPI bus as follows:
** MOSI - pin 11
** MISO - pin 12
** CLK - pin 13
** CS - pin 4 (for MKRZero SD: SDCARD_SS_PIN)
created 24 Nov 2010

modified 9 Apr 2012
by Tom Igoe
This example code is in the public domain.
*/
#include <SPI.h>
#include <SD.h>
const int chipSelect = 10;
char fileName[] = "LOGGER00.txt";

File dataFile;
void setup() {
// Open serial communications and wait for port to open:
Serial.begin(9600);
while (!Serial) {
; // wait for serial port to connect. Needed for native USB port only
}
Serial.print("Initializing SD card...");
// see if the card is present and can be initialized:

if (!SD.begin(chipSelect)) {
Serial.println("Card failed, or not present");
83
// don't do anything more:
while (1);
}
Serial.println("card initialized.");
// create a new file

for (uint8_t i = 0; i<100; i++){
fileName[6] = i/10 + '0';
fileName[7] = i%10 + '0';
if (! SD.exists(fileName)){
// only open a new file if it doesn't exist
Serial.println("Creating file: ");
Serial.println(fileName);
break;
}
}
}
void loop() {
// make a string for assembling the data to log:
String dataString = "";
// read three sensors and append to the string:

for (int analogPin = 0; analogPin < 3; analogPin++) {
int sensor = analogRead(analogPin);
dataString += String(sensor);
if (analogPin < 2) {
dataString += ",";
}
}
dataString += ",";
dataString += String(millis());
// open the file. note that only one file can be open at a time,
// so you have to close this one before opening another.
dataFile = SD.open(fileName, FILE_WRITE);
// if the file is available, write to it:

if (dataFile) {
dataFile.println(dataString);
dataFile.close();
// print to the serial port too:
Serial.println(dataString);
}
// if the file isn't open, pop up an error:
84
else {
Serial.println("error opening datalog.txt");
}
}
85
86
LIST OF REFERENCES
[1] Pledger, T., 2021, “The Role of Drones in Future Terrorist Attacks,” AUSA
[Online]. Available: https://www.ausa.org/publications/role-drones-future-
terrorist-attacks
[2] Guitton, M. J., “Fighting the Locusts: Implementing Military Countermeasures

Against Drones and Drone Swarms,” Scandinavian Journal of Military Studies,
Vol. 4, No. 1, p. 26-36, 2021, doi: 10.31374/sjms.53.
[3] “Rockets Galore; Cheap Smart Weapons,” The Economist (London), vol. 404, no.
8804, p. 85–, 2012.
[4] Busta, M., “Design and Development of an Adaptable Flight Demonstration

Vehicle and Modular Guidance and Control,” Master’s Thesis, Mechanical and
Aerospace Engineering Department, Naval Postgraduate School, Monterey, CA,
2019.
[5] Lobo, K., “Submunition Design for a Low-Cost Small UAS Counter-Swarm
Missile,” Master’s Thesis, Mechanical and Aerospace Engineering Department,
Naval Postgraduate School, Monterey, CA, 2018.
[6] Decker, K., “System Design and Integration of a Rapid Response Payload
Delivery Vehicle Using Commercial Off-The-Shelf Components,” Master’s
Thesis, Mechanical and Aerospace Engineering Department, Naval Postgraduate
School, Monterey, CA, 2021.
[7] Thyberg, R., “Design and Testing of a Multi-Unit Payload Delivery and Tracking
System for Guided Munitions to Combat UAV Swarm Threats,” Master’s Thesis,
Mechanical and Aerospace Engineering Department, Naval Postgraduate School,
Monterey, CA, 2019.
[8] Brophy, C., “ME4704 Missile Design Problem/Motivation,” Mechanical and

Aerospace Engineering Department, Naval Postgraduate School, Monterey, CA,
2021.
[9] Benson, T., “Rocket Stability,” NASA Glenn Research Center, May 2021.
https://www.grc.nasa.gov/www/k-12/rocket/rktstab.html
[10] H. Shekhar, “Prediction and Comparison of Shelf Life of Solid Rocket

Propellants Using Arrhenius and Berthelot Equations,” Propellants, explosives,
pyrotechnics, vol. 36, no. 4, pp. 356–359, 2011, doi: 10.1002/prep.200900104.
87
[11] Cesaroni Technology, Pro75 6819M1540-P Motor Data, May 2009.
http://www.pro38.com/products/pro75/motor/
MotorData.php?prodid=6819M1540-P
[12] Arnold, C., Ong, W. X., Patrick, G., Stuffle, N., and Fick, A. “ME4704: Tactical
missile Design High Altitude Communications Relay,” Naval Postgraduate
School, Monterey, CA, 2021.
[13] Ultimaker, Ultimaker PC Material Properties, Utrecht, Netherlands, retrieved 17

Mar. 2022. https://ultimaker.com/materials/pc
[14] Ultimaker, Ultimaker Tough PLA Material Properties, Utrecht, Netherlands,

retrieved 17 Mar. 2022. https://ultimaker.com/materials/tough-pla
[15] Saraswat, V. K., Woodley Clive, and Reddy G. Satheesh, “Effect of Burn Rate
Exponent Variation on Ballistic Performance of Gas Generator with High Burn
Rate Propellant,” in Ballistics 2019 - 31st International Symposium on Ballistics,
Volume 1 & 2, DEStech Publications, 2019, pp. 1–2.
[16] Zipay, J., Clarence T. Modlin, and Curtis E. Larsen, “The Ultimate Factor of
Safety for Aircraft and Spacecraft – Its History, Applications and
Misconceptions,” in 57th AIAA/ASCE/AHS/ASC Structures, Structural Dynamics,
and Materials Conference, doi: 10.2514/6.2016-1715.
[17] Hall, N., “Converging-Diverging Nozzle,” Nozzle Design, NASA Glenn Research
Center, May 2021. https://www.grc.nasa.gov/www/K-12/airplane/nozzled.html
[18] RocketMaterials.org, Sheer Pins, Rocket Materials, Scranton, PA, Aug. 2004.
https://web.archive.org/web/20131026023457/http://www.rocketmaterials.org/
datastore/cord/Shear_Pins/index.php
[19] Cesaroni Technology, Pro150, Ontario, Canada, retrieved 02 May 2022.

http://www.pro38.com/products/pro150/pro150.php
[20] Marshal Space Flight Center, Application of Ablative Composites to Nozzles for
Reusable Solid Rocket Motors, NASA Lesson Number 672, Feb. 1999.
[21] Cesaroni Technology, Pro98 9994M3400-P Motor Data, Ontario, Canada, Nov
2009. http://www.pro38.com/products/pro98/motor/
MotorData.php?prodid=9994M3400-P
[22] Altus Metrum, TeleMega, Altus Metrum, Colorado Springs, CO, Sep. 2021.
[23] PerfectFlite, StratoLoggerCF Altimeter, PerfectFlite, Andover, NH, 2017.
[24] Altus Metrum, EasyMega, Altus Metrum, Colorado Springs, CO, Dec. 2021.
88
[25] Multitronix, Kate-1 TelemetryPro Transmitter, Multitronix, Boise, ID, retrieved
14 Apr. 2022.
[26] The Engineering ToolBox, International Standard Atmosphere,

EngineeringToolBox.com, retrieved 16 Apr. 2022.
[27] Adafruit.com, “Overview,” Adafruit Data Logger Shield, Adafruit, New York,
NY, Apr. 2013.
89
90
INITIAL DISTRIBUTION LIST
1. Defense Technical Information Center

Ft. Belvoir, Virginia
2. Dudley Knox Library

Naval Postgraduate School
Monterey, California
91

AD1184618

Uploaded by

Copyright:

Available Formats

AD1184618

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AD1184618

Uploaded by

Copyright:

Available Formats

NAVAL

MODIFICATION OF COMMERCIAL ROCKET

Thesis Advisor: Christopher M. Brophy

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING

14. SUBJECT TERMS 15. NUMBER OF

17. SECURITY 18. SECURITY 19. SECURITY 20. LIMITATION OF

NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)

MODIFICATION OF COMMERCIAL ROCKET MOTORS

Submitted in partial fulfillment of the

MASTER OF SCIENCE IN SPACE SYSTEMS OPERATIONS

MASTER OF SCIENCE IN ASTRONAUTICAL ENGINEERING

NAVAL POSTGRADUATE SCHOOL

Approved by: Christopher M. Brophy

Several design modifications to commercial off-the-shelf solid rocket motors have

II. APPROACH .........................................................................................................13

III. DESIGN PROCESSES ........................................................................................17

IV. EXPERIMENTAL RESULTS............................................................................47

V. ROCKET FLIGHT TESTING ...........................................................................57

VII. FUTURE WORK .................................................................................................81

APPENDIX. DATA LOGGER ARDUINO CODE ......................................................83

LIST OF REFERENCES ................................................................................................87

INITIAL DISTRIBUTION LIST ...................................................................................91

Figure 1. Example of COTS drones swarm. Source: [1].............................................2

Figure 2. 2017 model of NPS RPV. Source: [4]. ........................................................3

Figure 3. 3D bomblet model. Source: [5]. ...................................................................4

Figure 4. NPS anti-drone bomblet concept of employment. Source: [6]. ...................4

Figure 5. 2019 RPV multiple bomblets design. Source: [7]. ......................................5

Figure 6. Multi-bomblet deploying mechanism. Source: [7]. .....................................6

Figure 7. NPS RRPD-V multi-mission concept. Source: [8]. .....................................6

Figure 8. Fall 2020 two-stage mid-flight failure. Source: [6]. ....................................7

Figure 9. NASA rocket stability demonstration. Source: [9]. .....................................8

Figure 13. Ignition squib with 2 g of solid propellant attached ..................................17

Figure 14. Smoke charge replacement hardware ........................................................18

Figure 16. Head-end ignition leads post-fire ...............................................................19

Figure 17. Squib basket 3D model hole configuration C1 ..........................................20

Figure 18. a) Hole configuration C2. b) Hole configuration C3 .................................21

Figure 21. High-speed camera and test stand setup ....................................................23

Figure 25. Ignition basket version 2 with 8 holes .......................................................27

Figure 28. Static SRM live-fire test stand setup ..........................................................29

Figure 29. Test stand head-end after live-fire test .......................................................30

Figure 31. Ignition basket version 2 after live-fire test ...............................................32

Figure 32. Semi-flexible squib holder supports ..........................................................32

Figure 34. Vacuum chamber test article ......................................................................34

Figure 35. Ignition squib holder final design ..............................................................34

Figure 36. Geometry of a converging-diverging (CD) nozzle. Source: [17]. .............36

Figure 37. Plug cap 3D model .....................................................................................36

Figure 38. Example of a) recessed and b) protruding nozzles ....................................38

Figure 39. Nozzle enclosure cap 3D model ................................................................38

Figure 40. Enclosure retainer arm and plug/cap..........................................................39

Figure 41. Shear pin retainer collar and cap ................................................................41

Figure 43. 3D model of blast tube assembled into O8000 motor................................43

Figure 45. M3400 thrust vs. time ................................................................................49

Figure 46. M3400 pressure and thrust profiles............................................................49

Figure 47. M3400 representative thrust curve. Adapted from [21].............................50

Figure 48. Zoomed in M3400 pressure vs. time..........................................................51

Figure 49. Enclosed squib test apparatus ....................................................................52

Figure 50. Enclosed squib test pressure vs. time.........................................................52

Figure 51. Nozzle cap test apparatus ...........................................................................53

Figure 55. OpenRocket model of rocket design ..........................................................57

Figure 57. Booster motor adapter sleeve .....................................................................58