INVESTIGATION OF A GENERIC WARHEAD CONTAINING PLASTIC BONDED EXPLOSIVE UNDER LIQUID FUEL FIRE

BY NUMERICAL AND TEST METHODS

A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES

OF MIDDLE EAST TECHNICAL UNIVERSITY

BY

HAKAN ŞAHİN

IN PARTIAL FULLFILLMENT OF THE REQUIREMENTS FOR

THE DEGREE OF MASTER OF SCIENCE IN

AEROSPACE ENGINEERING

DECEMBER 2015

iii

Approval of thesis:

INVESTIGATION OF A GENERIC WARHEAD CONTAINING PLASTIC BONDED EXPLOSIVE UNDER LIQUID FUEL FIRE

BY NUMERICAL AND TEST METHODS

submitted by HAKAN ŞAHİN in partial fulfillment of the requirements for the degree of Master of Science in Aerospace Engineering Department, Middle East Technical University by, Prof. Dr. Gülbin Dural Ünver Dean, Graduate School of Natural and Applied Sciences Prof. Dr. Ozan Tekinalp Head of Department, Aerospace Engineering Assoc. Prof. Dr. Dilek Funda Kurtuluş Supervisor, Aerospace Engineering Dep., METU Dr. Bekir Narin Co-Supervisor, TÜBİTAK SAGE Examining Committee Members: Prof. Dr. İsmail H. Tuncer Aerospace Engineering Dep., METU Assoc. Prof. Dr. Dilek Funda Kurtuluş Aerospace Engineering Dep., METU Prof. Dr. Yusuf Özyörük Aerospace Engineering Dep., METU Assoc. Prof. Dr. Sinan Eyi Aerospace Engineering Dep., METU Assist. Prof. Dr. Sıtkı USLU Mechanical Engineering Dep., TOBB University of Economics and Technology

Date: 29 December, 2015

iv

I hereby declare that all information in this document has been obtained and presented in accordance with academic rules and ethical conduct. I also declare that, as required by these rules and conduct, I have fully cited and referenced all material and results that are not original to this work.

Name, Last Name : Hakan ŞAHİN

Signature :

v

ABSTRACT

INVESTIGATION OF A GENERIC WARHEAD CONTAINING PLASTIC BONDED EXPLOSIVE UNDER LIQUID FUEL FIRE

BY NUMERICAL AND TEST METHODS

Şahin, Hakan M.S., Department of Aerospace Engineering Supervisor : Assoc. Prof. Dr. Dilek Funda Kurtuluş Co-Supervisor : Dr. Bekir Narin

December 2015, 78 pages

The present work provides a design methodology for a munition against fast cook-off

threat for type V insensitivity requirement. The experimental and numerical results

for a generic test item (warhead) filled with a PBXN-109 explosive are presented.

Two key points for the design of an insensitive munition against fast cook-off threat,

which are time to reaction and critical vent area is studied. Dividing the problem as

pre-ignition and post- ignition will allow one to manage this complex problem to

handle easier than its original form.

The first part of the study covers 2-D and 3-D simulations of the problem by

modeling the generic test item geometry by commercial CFD software (ANSYS -

Fluent). The second part of the study reveals low pressure (2-10 MPa) burn

characteristics of a PBXN-109 by strand burner tests. After obtaining pressure

dependent burning rate, conservation of mass equation is used to determine the

chamber pressure using MATLAB Simulink software.

Calculations are compared with the tests performed. Results are seen to be in

reasonable agreement with some discrepancies at 8.9% for time to reaction

prediction and at 10.9% for ventilation characteristics analyses. Possible reasons of

these differences are discussed in this study.

vi

Keywords: Insensitive Munition, Fast Cook-Off, Time-To-Reaction, Strand Burner

Test, PBXN-109

vii

ÖZ

SIVI YAKIT YANGININA MARUZ KALAN PLASTİK PATLAYICI

İÇERİKLİ BİR HARP BAŞLIĞININ NÜMERİK VE TEST METODUYLA İNCELENMESİ

Şahin, Hakan Yüksek Lisans, Havacılık ve Uzay Mühendisliği Tez Yöneticisi : Doç. Dr. Dilek Funda Kurtuluş Ortak Tez Yöneticisi : Dr. Bekir Narin

Aralık 2015, 78 sayfa

Yapılan bu çalışmada sıvı yakıt yangını tehdidine karşı tip V duyarsızlık isterine

sahip bir mühimmatın tasarımına yönelik bir metodolojisi sunulmuştur. PBXN-109

patlayıcısı içeren test kalemi (harp başlığı) ile yapılan deneysel ve nümerik sonuçlar

paylaşılmıştır.

Reaksiyon başlangıç zamanı ve kritik tahliye deliği çapı, sıvı yakıt yangınına karşı

tasarımı yapılan bir mühimmat için iki önemli anahtar parametredir. Problemin

reaksiyon öncesi ve sonrası olarak ikiye bölünmesi bu kompleks problemin çözümü

için tasarımcıya kolaylık sağlayabilir.

Çalışmanın ilk aşaması test kaleminin 2 ve 3 boyutlu olarak ticari hesaplamalı

akışkanlar dinamiği programı ANSYS-Fluent ile modellenmesini içermektedir.

İkinci aşama ise PBXN-109 patlayıcısına ait düşük basınçlı (2-10MPa) yanma

karakteristiğinin strand burn testleri ile çıkarıldığı çalışmayı içermektedir. Basınca

bağlı yanma hızının elde edilmesi ile birlikte test kalemi içerisinde oluşacak basıncın

tespiti için MATLAB Simulink ticari yazılımında kütlenin korunumu denklemleri

kullanılmıştır.

Hesaplamalar yapılan testler ile karşılaştırılmıştır. Buna göre, sonuçlar makul kabul

edilebilecek seviye birbiri ile örtüşmekte olup; reaksiyonun süresinin tespiti için

yapılan öngörüler için %8,9 olup, kritik tahliye deliğinin tespiti için yapılan

viii

çözümlemeler için %10,9’dur. Bu farklara olması muhtemel sebepler çalışma

içerisinde paylaşılmıştır.

Anahtar Kelimeler: Duyarsız Mühimmat, Sıvı Yakıt Yangını, Reaksiyon Süresi,

Yanma Hızı Testleri, PBXN-109

ix

To My Family

x

ACKNOWLEDGEMENTS

I would like to express my deepest thanks and appreciation to my supervisor, Assoc.

Prof. Dr. Dilek Funda KURTULUŞ, for her supervision and continuous support

during all stages of this thesis.

I would like to give my special appreciations to the Chief of the Terminal Ballistics

and Fuze Technologies Division at TÜBİTAK SAGE, Dr. Bekir NARİN for

initiating this thesis and encouragement on my studies.

I would like to express my sincere appreciation to Mr. Başar ESİRGEN for his

crucial advices and invaluable efforts during the preparation of this thesis.

My special thanks also go to my colleagues at TÜBİTAK SAGE, Mr. Oğuzhan

AYISIT, Mr. Tahir TURGUT, Mr. Salih SARAN and Mr. Fatih CENGİZ for their

discussion on the subject.

Finally, the greatest thanks go to my parents and lovely wife who supported and

encouraged me throughout my whole life. It would be impossible without their

patience and understanding.

xi

TABLE OF CONTENTS ABSTRACT ................................................................................................................. v

ÖZ .............................................................................................................................. vii

ACKNOWLEDGEMENTS ......................................................................................... x

TABLE OF CONTENTS ............................................................................................ xi

LIST OF TABLES .................................................................................................... xiii

LIST OF FIGURES .................................................................................................. xiv

LIST OF SYMBOLS ............................................................................................... xvii

LIST OF ABBREVIATIONS ................................................................................. xviii

1. INTRODUCTION................................................................................... 1

1.1. Definition of Insensitive Munitions, Threats and Reaction Levels ......... 2

1.2. Historical Background ............................................................................ 4

1.3. Literature Survey ..................................................................................... 8

1.4. Motivation and Objectives .................................................................... 11

2. METHOLODOGY ................................................................................ 13

2.1. Pre-Ignition Phase: Thermal Initiation .................................................. 13

2.2. Post-Ignition Phase: Ventilation of Explosive Burn Products .............. 16

3. CHARACTERIZATION OF THE PBXN-109 EXPLOSIVE .............. 21

3.1. STRAND BURNER TESTS ................................................................. 21

4. FAST COOK-OFF TESTS ................................................................... 27

4.1. Description of the Test Item .................................................................. 27

4.2. Experimental Setup ............................................................................... 29

5. RESULTS AND DISCUSSION ........................................................... 31

5.1. Fast Cook-Off Test Results ................................................................... 31

xii

5.1.1. Test 1 ........................................................................................ 31

5.1.2. Test 2 ........................................................................................ 34

5.1.3. Test 3 ........................................................................................ 37

5.1.4. Test 4 ........................................................................................ 42

5.1.5. Test 5 ........................................................................................ 45

5.1.6. Test 6 ........................................................................................ 48

5.1.7. Test 7 ........................................................................................ 51

5.1.8. Test 8 ........................................................................................ 55

5.2. Comparison of Tests and Analyses ....................................................... 58

5.2.1. Determination of “True” Boundary Condition ......................... 58

5.2.2. Computation Results for Time to Reaction Analyses and Comparison of the Results with Test Results ........................................ 62

5.2.3. Computation Results for Time to Reaction Analyses and Comparison with Test Results ............................................................... 66

6. CONCLUSION AND FUTURE WORK .............................................. 69

6.1. Conclusion ............................................................................................. 69

6.2. Future Work .......................................................................................... 70

REFERENCES ........................................................................................................... 71

APPENDICIES .......................................................................................................... 77

A.MATLAB FUNCTIONS .................................................................................. 77

xiii

LIST OF TABLES TABLES Table 1. Heat Generation Constants........................................................................... 14

Table 2. PBXN-109 Gaseous Product Thermodynamic Properties in Chamber from NASA CEA Software ....................................................................................... 18

Table 3. Mesh Refinement Study ............................................................................... 63

Table 4. Time Refinement Study ............................................................................... 63

Table 5. Comparison of Calculated and Test Observed TtR ..................................... 65

xiv

LIST OF FIGURES FIGURES Figure 1. Threats against Munitions, Related Insensitive Munitions Test and

Phenomena .......................................................................................................... 3

Figure 2. Reaction Types of Munitions to Accidental Stimuli .................................... 4

Figure 3. USS Forrestal before the Accident ............................................................... 5

Figure 4. Forrestal Aircraft Carrier Accident, 1967 ..................................................... 6

Figure 5. IM and non-IM Munition Accidental Causality Approximation .................. 6

Figure 6. Camp Doha Accident, 1991 .......................................................................... 7

Figure 7. Temperature Profile Predictions at Ignition for an Energetic Material for Different Heating Rates ...................................................................................... 9

Figure 8 (a) Schematic of the Problem Description of Critical Ventilation Problem, (b) Detailed Description of the Mass Inlet and Mass Outlet ............................ 16

Figure 9. MATLAB Simulink –Main Body ............................................................... 19

Figure 10. MATLAB Simulink – Solver ................................................................... 20

Figure 11. Strand Burner Cylindrical Vessel ............................................................. 21

Figure 12. Schematic of the Strand Burner Experimental Setup ............................... 22

Figure 13. Sample Holder and Wire Connections ...................................................... 23

Figure 14. Sample Make Wire Readings ................................................................... 24

Figure 15. PBXN-109 Burn Rate Change with Pressure (0-10MPa) ......................... 25

Figure 16. PBXN-109 Burn Rate Change with Pressure (1-1000MPa) ..................... 25

Figure 17. Generic Test Item (Short Version) ............................................................ 27

Figure 18. Polycarbonate Rod for Holding Steel Sections Together Within the Explosive .......................................................................................................... 28

Figure 19. Thermocouple Placement on 1st Test Item ............................................... 30

Figure 20. Test Item 1 – Before the Test .................................................................... 31

Figure 21. Test Item 1 – After the Test ...................................................................... 32

xv

Figure 22. Flame, Surface and Internal Temperature Readings of Test Item 1 ......... 33

Figure 23. Internal Pressure Readings of Test Item 1 ................................................ 34

Figure 24. Test Item 2 – Before the Test ................................................................... 35

Figure 25. Test Item 2 – After the Test ...................................................................... 35

Figure 26. Flame and Internal Temperature Readings of Test Item 2 ....................... 36

Figure 27. Internal Pressure Readings of Test Item 2 ................................................ 37

Figure 28. Test Item 3 – Before the Test ................................................................... 38

Figure 29. Test Item 3 – After the Test ...................................................................... 38

Figure 30. Surface Type Thermocouples That Are Located on the Steel Sections ... 39

Figure 31. Surface Type Thermocouple That Is Located in the Middle of Test Item Sidewall ............................................................................................................ 39

Figure 32. Flame and Internal Temperature Readings of Test Item 3 ....................... 40

Figure 33. Section View of Test Item 3 and Dimensions from the Measuring Points to Explosive Surface ......................................................................................... 41

Figure 34. Internal Pressure Readings of Test Item 3 ................................................ 42

Figure 35. Test Item 4 – Before the Test ................................................................... 43

Figure 36. Test Item 4 – After the Test ...................................................................... 43

Figure 37. Flame and Internal Temperature Readings of Test Item 4 ....................... 44

Figure 38. Internal Pressure Readings of Test Item 4 ................................................ 45

Figure 39. Test Item 5 – Before the Test ................................................................... 46

Figure 40. Test Item 5 – After the Test ...................................................................... 46

Figure 41. Flame and Internal Temperature Readings of Test Item 5 ....................... 47

Figure 42. Internal Pressure Readings of Test Item 5 ................................................ 48

Figure 43. Test Item 6 – Before the Test ................................................................... 49

Figure 44. Test Item 6 – After the Test ...................................................................... 49

Figure 45. Flame and Internal Temperature Readings of Test Item 6 ....................... 50

xvi

Figure 46. Internal Pressure Readings of Test Item 6 ................................................ 51

Figure 47. Test Item 7 – Before the Test .................................................................... 52

Figure 48. . Test Item 7 – After the Test .................................................................... 52

Figure 49. Flame and Internal Temperature Readings of Test Item 7 ....................... 53

Figure 50. Internal Pressure Readings of Test Item 7 ................................................ 54

Figure 51. Deflagration of PBXN-109 Explosive, Test 7 .......................................... 54

Figure 52. Test Item 8 – Before the Test .................................................................... 55

Figure 53. Test Item 8 – After the Test ...................................................................... 56

Figure 54. Flame and Internal Temperature Readings of Test Item 8 ....................... 57

Figure 55. Internal Pressure Readings of Test Item 8 ................................................ 57

Figure 56. Flame Temperature Readings from the 1st Test ....................................... 58

Figure 57. (a) Front Side, (b) Rear Side Surface Temperature Readings from the 1st Test Item ........................................................................................................... 59

Figure 58 Initial Numerical Results of Test Flame Temperature Compared with Test Surface Temperatures ....................................................................................... 60

Figure 59 Tuned and Simplified Flame Temperature Distribution ............................ 61

Figure 60. Surface Temperature Comparison of Test and Numerical Results - Test Case 1 ................................................................................................................ 62

Figure 61. Mesh Distribution (top) and Temperature Gradient (bottom) of Test Items 7-8 ..................................................................................................................... 63

Figure 62. Maximum Temperature Plot of the Explosive and Ignition Time ............ 64

Figure 63. Sample MATLAB Simulink Output, Test Item with Ø9mm Vent ........... 66

Figure 64. Comparison of Experimental and Numerical Results of Peak Pressure ... 67

xvii

LIST OF SYMBOLS

a burn rate coefficient

A molar fraction

Avent ventilation area, m2

Aburn total burn surface area, m2

cp specific heat capacity, J/kg/K, Cal/g/K

CD discharge coefficient

E activation energy, J/kg/K

specific heat ratio

k1, k2, k3 chemical reaction kinetic model step

L/D length over diameter

M Mach number

mass flux

burning rate pressure exponent

∗ particle velocity at throat, m/s

pressure, MPa

∗ pressure at throat, MPa

R universal gas constant, J/mol/K

explosive burn rate, m/s, mm/sn

S rate of chemical heat production, J/s

heat of reaction, J/kg

t time, s

T temperature, K

∗ temperature at throat, K

V volume, m3

Z frequency factor, 1/s

Greek Symbols

λ thermal conductivity, J/m/s/K

density, kg/m3

∗ density at throat, kg/m3

xviii

LIST OF ABBREVIATIONS

IM Insensitive Munition

PBXN Plastic Bonded Explosive

TtR Time to Reaction

TÜBİTAK Türkiye Bilimsel ve Teknolojik Araştırma Kurumu

SAGE Savunma Sanayi Araştırma Geliştirme Enstitüsü

TNT Trinitrotoluene

STANAG Standardization Agreement of NATO

HMX Octahydrotetranitrotetrazine

RDX Cyclotrimethylenetrinitramine

DDT Deflagration to Detonation Transition

1

CHAPTER 1

1. INTRODUCTION

Energetic materials such as explosives, pyrotechnics and propellants are widely used

in many military applications. Explosives are defined as materials which react to

produce a violent expansion of hot gas, an explosion, which rapidly delivers energy

to its surroundings. The rate of transformation from solid to hot gas takes place on

timescales of microseconds. Propellants are less violent in reaction and are used to

accelerate objects such as missiles and bullets. For propellants transformation takes

place in a slower timescale which is milliseconds. Pyrotechnics are systems which

react to produce an effect such as smoke, light or noise with reaction times from

milliseconds to many minutes [1].

Comparing energetics materials with other materials such as petrol, petrol contains

six times more available energy as the same mass of TNT but unlike petrol, TNT

releases its energy a hundred million times faster [1]. These reaction (transformation)

time form the basis of the explosion.

Today, safety design of the munitions against unplanned stimuli like terrorist attack,

accidental fire environment etc. becomes as important as the design for performance

and operational requirements. Many nations have been undertaken significant work

to reduce the response of munition systems under attack. These efforts include

understanding explosive response under thermal load and mechanical shocks.

Knowing the material behavior of the explosive, it becomes possible to prevent

unwanted reaction of the munition while it is in non-operational condition by taking

precautions at the design phase.

Several studies have been performed in TÜBİTAK-SAGE for the design of munition

components like warheads and rocket motors against threats especially after

insensitivity became a requirement by the Turkish Ministry of Defense. Liquid fuel

fire is one of the threads that munition can be exposed to. Past studies that have been

2

made includes test of munitions having a case material that can melt and allow

explosive to have a contact with atmosphere directly. But not all munitions can take

advantage of the same solution such as munitions that have a penetrator type of

warhead.

In this thesis study, investigation of behavior of the warhead under liquid fuel fire is

performed by using analytical formulas, commercial Fluent program and performing

tests. It is aimed to develop a general method for designing insensitive munitions

against liquid fuel fire threat and minimize the test number required in the test phase.

1.1. Definition of Insensitive Munitions, Threats and Reaction Levels

Insensitive munition by the definition of STANAG 4439 is: “…munitions which

reliably fulfil their performance, readiness and operational requirements on demand

but which minimize the probability of inadvertent initiation and severity of

subsequent collateral damage to weapon platforms, logistic systems and personnel

when subjected to unplanned stimuli…” [2]. These unplanned stimuli are stated as

follow:

Fast heating (fast cook-off)

Slow heating (slow cook-off)

Bullet impact

Fragment impact

Shaped charge jet impact

Sympathetic reaction

In Figure 1, possible threats against munitions, related insensitive munitions tests and

phenomena are shown schematically [3].

Figure

There are

stimuli an

De

ene

Pa

rea

dec

Ex

sub

De

wi

les

1. Threats

e different r

d classified

etonation (T

ergetic mate

artial Deton

action whe

composition

xplosion (T

b-sonic dec

eflagration

th ignition

ss violent pr

against Mu

reaction typ

d under 6 typ

Type I): Th

erial is cons

nation (Ty

ere some e

n.

Type III): T

omposition

(Type IV)

and burnin

ressure relea

3

unitions, R

Phenom

ypes that m

pes which a

he most vio

sumed in a

ype II): Th

energetic m

The third mo

n of energeti

): The fourt

ng of confin

ase.

Related Inse

ena [3]

munitions re

are [4]:

olent type o

supersonic

he second m

material is

ost violent t

ic material a

th most vio

ned energet

ensitive Mu

eact under

of munition

decomposit

most violen

consumed

type of mun

and extensiv

olent type o

tic material

unitions Te

the effect

n reaction w

tion.

nt type of m

d in a su

nition react

ve fragment

of munition

ls which le

st and

of these

where the

munition

upersonic

tion with

tation.

reaction

eads to a

B

e

a

s

In Figu

shown s

1.2.

History

unintend

consequ

and Cam

amount

In July

was con

Burn (Typ

energetic m

No-reactio

any reactio

stimulus.

ure 2 the re

schematical

Figure 2.

Historical

is replete w

ded functio

uences to th

mp Doha a

of personn

1967, USS

nducting com

pe V): The

material ignit

on (Type V

n is self-ex

eaction type

lly [5].

Reaction T

Backgroun

with accide

oning or r

he owning o

accident in

el and equip

Forrestal, a

mbat operat

fifth most v

tes and burn

VI): The lea

xtinguished

es to accid

Types of M

nd

ents and inc

reaction of

or using for

1991 are t

pment losse

a USS Nava

tions in the

4

violent type

ns non-prop

ast violent t

immediate

dental stimu

Munitions to

cidents that

f munitions

rces and N

two of the

es.

al aircraft ca

waters near

e of munitio

pulsive.

type of mun

ely upon rem

uli that are

o Accidenta

involved o

s, which re

ation. Forre

well-known

arrier which

r Vietnam [

on reaction

nition respo

moval of th

explained

al Stimuli [5

or were cau

esulted in

estal accide

n disasters

h is shown i

6].

n where the

onse where

he external

above are

5]

used by the

hazardous

ent in 1967

with large

in Figure 3

During pre

the deck o

missile str

igniting th

additional

tanks to s

These act

destroyed,

Figure 4, a

Fig

eparations f

of the carri

ruck anothe

he fuel wh

planes and

pill and ign

tions and

, 39 aircraf

a photo of s

ure 3. USS

for a comba

er which w

er plane cau

hich is kno

d caused an

nite and mo

reactions r

ft damaged

sailors attem

5

S Forrestal

at sortie, a f

was loaded w

using exter

own as JP-

M-117 bom

ore bombs

resulted in

and the sh

mpting to fig

before the

forward firin

with planes

rnal fuel tan

-5. Flowing

mb to explo

and several

n 134 deat

hip removed

ght fires dur

Accident [6

ng unguided

s fully fuell

nk of the ai

g burning

ode. This ha

l missile wa

hs, 161 in

d from com

ring the inci

6]

d rocket flew

led and arm

airplane to s

fuel then e

as caused m

arheads to

njuries, 21

mbat action

ident is sho

w across

med. The

spill and

engulfed

more fuel

explode.

aircraft

s [6]. In

wn [7].

Figure 5

It is ob

munitio

dramati

Figur

Figur

5 shows the

bvious that

ons charact

cally [8].

re 5. IM an

re 4. Forre

e operationa

if the mu

teristics, th

nd non-IM

stal Aircra

al impact of

unitions exi

he amount

Munition A

6

aft Carrier

f IM and non

isted on th

of total l

Accidental

Accident, 1

n-IM bomb

he deck had

loss could

Causality A

1967 [7]

bs to an aircr

d had the

have been

Approxima

raft career.

insensitive

n reduced

ation [8]

Camp Doh

and armor

initiated b

engine fire

Developin

reduces th

the resear

Researche

insensitive

The IM t

platform a

costs, faci

which faci

In this stu

threat. In

simulate l

to record

ha accident

red personal

by internally

e. It is resul

ng and intro

he life losse

chers about

ers have m

e munitions

technology

and crew su

ilitate reduc

ilitates redu

udy among

fast cook-

iquid fuel f

the test item

which occu

l carriers is

y stored non

lted in 3 dea

Figure 6. C

oducing ins

s and opera

t the outcom

mentioned a

s concept [1

objectives

urvivability,

ction in ha

uced quantit

all other th

off test mu

fire accident

m response

7

urred in Kuw

shown in F

n-IM muniti

aths, 52 inju

Camp Doha

sensitive m

ational costs

mes of imp

about the

0], [11], [12

are to inc

increase op

azard classif

ty-distance r

hreats, main

unitions are

tal conditio

e due to a r

wait 1991 a

Figure 6. Th

ions that are

uries and 15

a Accident,

munitions sy

s. Many ana

plementing

benefits a

2].

crease tota

peration saf

fication thr

requiremen

n concern is

e subjected

on. The purp

rapid increa

and its effec

is incident i

e subjected t

50 vehicle d

, 1991 [9]

ystems in th

alyses have

insensitive

nd the fut

l life cycle

fety, reduce

ough increa

ts.

s focused o

to a heat

pose of the

ase in tempe

ct to multipl

is known to

to single ve

destroyed [9

he operatio

been perfo

munitions

ture trends

e safety, in

ed weapon l

ased storag

on the fast

source in

fast cook-o

erature. An

e tanks

o be

ehicle

].

onal area

ormed by

policies.

s in the

ncreased

ife cycle

ge safety

cook-off

order to

off test is

n aircraft

8

fuel fire with an average steady-state temperature of at least 800 °C is used as the test

stimulus [4].

1.3. Literature Survey

Thermal Initiation Theory describes the initiation of deflagration due to thermal

effects from surrounding conditions and the heat generated inside the energetic

material. Understanding the ignition characteristics of energetic materials is the basis

of the studies in order to predict the cook-off behavior of munitions and development

of design techniques for insensitive munitions.

Energetic materials are unstable and decompose rapidly as their temperature is

raised. Energetic materials can be characterized by an ignition temperature; however,

the temperature at which ignition occurs more accurately depends upon the geometry

of the energetic material and the rate at which it is heated [13].

Energetic materials exothermically decompose when exposed to external heat source.

A part of the heat release accompanied with the decomposition is dissipated out

while some is accumulated inside the energetic material. Any material that gives

exothermic decomposition reactions during heating like explosive in a warhead,

propellant in a rocket motor or even a bale of wool may end up with ignition. Self-

heating can be defined as the increase of temperature of a material due to the heat

generated inside with exothermic decomposition reactions [14]. Decomposition rate

increases as the temperature increases, thus self-heating rate increases.

To determine whether the energetic material will undergo a thermal initiation under

certain surrounding conditions, it is required to know the critical temperature of that

material which is a function of the chemical, physical, and the geometrical properties

of the material. Critical temperature is the limit surrounding temperature under which

no ignition occurs regardless of the exposition time to that temperature. On the other

hand, under a surrounding temperature conditions above the critical temperature,

ignition eventually occurs and the level of temperature only affects time or location

of ignition.

In Figure

material is

boundary

temperatu

the surrou

materials,

increase m

where exp

Figure 7.

Since fast

600°C/h),

geometry

Studies of

made the

zero order

However,

However,

7, the effe

s visualized

but with d

ure limit, sel

unding. Be

the tempe

most. This c

plosive mate

. Temperat

t cook-off

critical tem

and ignition

f thermal ini

uniform tem

r reaction ki

this theor

as the cond

ect of heati

d [14]. In all

different he

lf-heating ra

ecause of t

rature of m

causes igniti

erial density

ture Profile

for Dif

is a case

mperature

n occurs on

itiation of e

mperature d

inetics [15]

y was only

duction coef

9

ing rate on

l cases, ener

eating rates.

ate become

the low co

more isolate

ion point to

y over the v

e Prediction

fferent Hea

where the

of the ene

n the edges c

energetic ma

distribution

. This was t

y applicabl

fficient of s

the locatio

rgetic mater

. For the lo

es more dom

onduction c

ed part of

o be moved

volume is lar

ns at Ignitio

ating Rates

heating ra

ergetic mate

close to bou

aterials start

assumption

the origin o

le for syste

solid energe

on of igniti

rial is heate

ow heating

minant over

coefficient

the explosi

from the ou

rge.

on for an E

s [14]

te is very

erial is ind

undaries whe

t with Seme

n in the ener

f the therma

ems like w

etic material

ion of an e

ed from surr

rate cases

r the heating

of solid e

ive filling

uter edges t

Energetic M

large (bigg

dependent f

here heating

enov [15]. S

rgetic mate

al initiation

well stirred

ls are relativ

energetic

rounding

, after a

g rate by

xplosive

tends to

to places

Material

ger than

from the

occurs.

Semenov

rial with

n studies.

liquids.

vely low

10

and uniform temperature distribution assumption is not applicable, Frank-

Kamenetskii [16] proposed a theory based on the conductive heat transfer in the

energetic material which allows time dependent temperature distribution.

Numerical studies which is based on Frank-Kamenetskii’s work on predicting the

ignition times and temperatures for energetic materials have been performed starting

from 1960’s. Zinn and Mader [17] applied Fourier series spatial representation for

the solution to the reactive heat conduction equation to obtain ignition times for

explosive material. Merzhanov and Abramov [18] used finite difference method for

one-dimensional reactive heat conduction with the zero-order kinetic model. Suceska

[19] used finite difference method and developed code THERMEX for the solution

of the reactive heat conduction equation problem with the zero-order kinetic model.

Anderson [20] developed the code TEPLO using finite difference method, which

took temperature dependent material properties into account. Isler and Kayser [21]

and McGuire and Tarver [22] used different kinetic models instead of zero-order

kinetic model such as power law kinetic model to predict the ignition behavior of

different energetic materials. There are also some methods presented by Pakulak [23]

and Victor [13] for predicting the time to reaction by using analytical equations to

solve 1-D heat transfer. Victor has implemented these methods to excel spread sheets

for determining internal temperatures by taking into account the self-heating of the

explosive, however none of those studies were involved into a chemical reaction

modelling code. Instead, all used simple kinetic models as stated above to estimate

and implement heat generated by the explosive as it gets heated by the stimuli.

Other than those studies, Lawrence Livermore National Laboratory, a high explosive

cook-off study has been conducted by a group of engineers and physicist to model

the response of energetic materials to thermal stimuli and the processes involved in

the energetic response [24], [25]. This study had coupled the two parts of the

problem that was stated above which are pre-ignition and post-ignition. It is stated

that several new algorithms have been developed to increase the accuracy and

fidelity of the modeling process including a level set driven multi-material

deflagration model, a multi-temperature mixed material treatment, self-consistent

11

thermal-hydro coupling, full implicit quasi-static hydrodynamics, ALE slide surfaces

and ALE slide deletion.

On the other hand, ventilation of burn products might be vital for a system containing

energetic material for unplanned thermal or mechanical stimuli. The question that is

generally raised is “what is the required ventilation size for preventing system to

react violently”.

Graham [26] and Victor [13] derived a methodology for the determination of critical

ventilation requirements of rocket motor and warheads using the internal ballistics.

They both assumed that the flow over the ventilation area becomes sonic, and

simplified and derived equations of internal ballistics accordingly. Graham [26] also

used ballistic analysis methodology for the determination of critical ventilation

requirements of rocket motor.

1.4. Motivation and Objectives

Design phase of an insensitive warhead can be complex considering the number of

tests which should be carried on. Considering the cost and environmental concerns,

among all other insensitive munition sign tests, fast cook-off will probably lead in a

"number of tests to be decreased" list for most IM design authorities.

This thesis study is related to the insensitive munition studies that are going on at

TÜBİTAK SAGE Terminal Ballistics Division. The aim of the study is to have an

improved knowledge on the theory behind the fast cook-off stimuli and have a

methodology for being able to design a munition that has reduced impact on its

surroundings. By doing so, this will allow us to decrease the number of tests on the

design phase leading into the decrease of cost and negative environmental effects.

12

13

CHAPTER 2

2. METHOLODOGY

In IM design of a typical warhead against fast cook-off test, there are two important

parameters to be predicted which are time to reaction (TtR) and critical ventilation

area. TtR is important for the precautions that are taken to work before reaction

occurs, whereas critical ventilation area is important for safe disposal of burn

products of energetic materials. Dividing the problem as pre-ignition and post-

ignition will allow one to manage this complex problem to handle easier than its

original form.

Although the present methodology is valid for all energetic materials, PBXN-109

(64% RDX, 20% Al, 16% DOA/HTPB) is considered as the energetic material for

the purpose of illustration and as it is widely used as the main charge for most

warheads designed/manufactured worldwide. Although plastic-bonded explosives

show relatively low responses to auxiliary thermal/mechanical effects, as long as

they are exposed to heat within a closed chamber/casing any explosive, it will lead to

deflagration to detonation transition unless necessary precautions are taken.

2.1. Pre-Ignition Phase: Thermal Initiation

The kinetic model to predict the thermal decomposition of a RDX based explosive

has been developed by McGuire and Tarver [22]:

1

where, S is the chemical heat source during the chemical reaction, λ is the thermal

conductivity, ρ is the density and cp is the specific heat. The scheme proposed by

McGuire and Tarver’s model reaction for RDX based explosives is a 3 step kinetic

model where mass is converted from one species to another:

14

→ →2 → 2

In which A=RDX, B=H2C, C=CH2O+N2O and D=final gaseous products. Each of

the reduced chemical reactions follows the Arrhenius equation:

⁄ 3

Where k is the reaction rate coefficient, Z is the frequency factor (collision number),

E is the activation energy and A is the molar fraction. In our model the full reaction

is modeled as an instantaneous one step exothermic chemical reaction thus, A is

taken to be 1 and heat generation is calculated with the equation (4).

S ρQ AZe ⁄ 4

Temperature dependent heat generation term (1) using constants for PBXN-109

given in Table 1 is implemented into Fluent with a user defined function (UDF),

[27].

Table 1. Heat Generation Constants, [27]

ρ [kg/m3]

Qact

[J/kg]

Z [1/s]

E [J/mol]

R [J/mol.K]

1680 2198070 1.023x1014 152716 8.314

After defining the kinetic model for the thermal decomposition of the explosive, it is

necessary to define the boundary condition to be able to solve thermal conduction

heat transfer to explosive. There is a boundary condition definition proposed by

Victor [13] where temperature is assumed to be constant throughout the test and is

1073 K with a convection coefficient of 6 W/m2K and emissivity of 1. Applying heat

flux as a boundary condition is another option. There are some studies going on by

Fuel Fire Experts (FFE) Working Group [28] to measure heat flux rates and compare

the data with the alternative fast cook-off test measurements. This is being done to

check if there is any alternative method eligible to mimic standard hydrocarbon fuel

15

fire thermal load. Measurements within the hearth of fire shows that, heat flux values

vary between 100-150 kW/m2 [29].

The most time consuming but accurate way of predicting the time to reaction is

probably modeling the fire itself instead of applying the thermal load as a boundary

condition. There are some examples of fire modelling codes such as C-SAFE [30]

Sierra/Fuego [31] and LES solvers [32]. But as these codes are not available for

commercial use and requires huge amount of computational work, it has been

decided to go on with the effort of finding the least time consuming and “accurate

enough way” of predicting the time to reaction.

Although 1-D (infinitely long cylinder approach, axisymmetric) thermal analysis

approach might satisfy the current geometry of our generic test item, most of the time

because of the complex geometrical shapes of the energetic systems this is

considered not to be applicable. Thus, only 2-D and 3-D thermal analysis are

performed, using commercial ANSYS Fluent software as it is commonly used in

similar applications [33], [34]. The aim is to conclude to a point of an improved

boundary condition definition and to see how accurate the predictions are from the

analysis.

Both 2-D axisymmetric and 3-D transient analyses are performed with pressure

based solver as it is known to be better performing for incompressible low velocity

flows [35]. 3-D analyses are made in order to see the effect of buoyancy (trapped

heated air within the test chamber).

Gravitation force is enabled in order to let heated air move within test chamber. PISO

(Pressure-implicit with splitting of operators) pressure-velocity scheme is used

because of the limitations of the SIMPLE and SIMPLEC is that new velocities and

corresponding fluxes do not satisfy the momentum balance after the pressure

correction equation is solved. Although the PISO algorithm takes more CPU time per

solver iteration, it can dramatically decrease the number of iterations required for

convergence, especially for transient problems [35].

For spa

force w

For 2-D

convect

equation

2.2.

To be a

system

material

pressuri

discharg

solve th

having o

total bur

Figu

P

atial discret

eighted disc

D analysis, c

tion and rad

n with the h

Post-Igniti

able to dete

shall be d

l that cont

ize the free

ge burn pro

his problem

one inlet an

rn area whe

ure 8 (a) Sc

Problem, (b

tization, sec

cretization i

constant de

diation term

heat generat

ion Phase: V

rmine the c

derived firs

tains its o

e volume o

oducts that

free volum

nd one outle

ereas outlet

chematic o

b) Detailed

cond order

is chosen.

nsity is def

s are applie

tion and con

Ventilation

critical vent

st. Reaction

wn oxidize

of the cont

will produ

me within th

et as shown

represents t

f the Probl

Description

16

for energy

fined for the

ed as bound

nduction ter

n of Explos

tilation area

n of PBXN

er produce

tainer. Ven

uced by the

he chamber

n in Figure 8

the total are

(a)

(b)

lem Descrip

n of the Ma

y and mom

e air inside

dary conditio

rm only.

ive Burn P

a, the mathe

N-109 like

es burn pro

ntilation on

e burn of e

is defined

8. Inlet in th

ea of ventila

ption of Cr

ass Inlet an

mentum whe

the test ch

on; leaving

Products

ematical mo

any other

oducts that

the other

energetic m

as the contr

his case rep

ation precau

ritical Venti

nd Mass Ou

ereas body

hamber, the

the energy

odel of the

r energetic

t fills and

hand will

material. To

rol volume

presents the

utions.

tilation

utlet

17

As the purpose of this work is to predict critical ventilation area, it is possible to

assume that the flow is choked and Mach is equal to one at the orifice ventilation

hole. The aim here is to assume that the ventilation at its highest rate and see if the

pressure within the chamber is still rising or not. As long as the initial pressure that is

described in the problem tends to decrease this will point out that the ventilation

diameter will not cause a dangerous situation where the burn speed of the explosive

diverge, cause it to deflagrate and even explode with the pressure increase. So, one

can basically write conservation of mass equations to solve this problem assuming

that the burn area of the explosive is constant (no burn back).

As shown in Figure 8, problem can be defined with a single mass inlet (explosive

burn products) and a single mass outlet (total ventilation area). As long as explosive

self-ignites due to heat generation, burn products will start to fill chamber increasing

the pressure within and exiting the chamber due to pressure difference with respect

free atmosphere.

m md m

dt 5

The rate of mass transfer to the control volume can be defined as the product of total

explosive area Aburn, burn speed of the explosive rburn and the density of burn

products within the chamber ρburn.

m A . . 6

As the chamber pressure exceeds the outside pressure (atmospheric pressure) by a

ratio more than 0.8 MPa, the flow becomes sonic for air [26]. Thus, mass transfer

from the control volume can be solved at the section where Mach number is assumed

to be 1 and very simple expression for the rate of mass discharged results in equation

7. Although this ratio will change for burn products the same approach can be

m A . ∗. ∗. C 7

where CD is the discharge coefficient. It is generally used as 1 in ideal flow but in

actuality, because of the ventilation hole shapes (square-edged orifice) that results in

a coefficient of 0.82 is used as vena contracta is formed by the exiting gases [36].

18

Unknown terms at the orifice can be solved using the isentropic equations given

below (8, 9, 10, and 11) [36].

∗ 8

∗ k 9

∗

1 12

10

∗

1 12

11

Burned explosive gaseous product properties in the chamber of the test item are

obtained from the NASA Chemical Equilibrium and Applications (CEA) software

[37] by modelling the PBXN-109 components as given in Table 2.

Table 2. PBXN-109 Gaseous Product Thermodynamic Properties in Chamber

from NASA CEA Software

Input

Reactant Weight Energy

Fraction [J/Mol]

1 Al 0.21 0.00

2 HTPB 0.08 -260.17

3 IPDI 0.01 -

372249.00

4 DOA 0.07 0.29

5 RDX 0.63 66984.78

6 Antioxidant 0.00 -

165587.00

Output

Burn Temperature Cp λ

Product [K] [J/g.K]

2552.94 1.92 1.22

Although

ventilation

burn rate (

a and n

(octahydro

burning c

[38]. Ther

both pure

pressure v

can withst

time warh

mentioned

calculate c

samples o

Simulink

of it’s easy

Main bod

rate coeff

Appendix

functions

Appendix

definition

n area, it is

(∀) of the en

should be

otetranitrote

haracteristic

re are some

e RDX [39

values of 10

tand to very

heads are n

d above. To

critical vent

of PBXN-10

body as sho

y to use inte

F

dy basically

ficients and

A. Solver

of the solv

A.

of constan

also possib

nergetic ma

∀ ∀

known in

etrazine) an

cs that is a

e studies in

] and PBX

0 MPa. Alt

y high inte

not designe

o be able to

tilation area

09. Above

own in Figu

erface and g

Figure 9. M

uses the “

d explosive

that is used

ver which a

19

nt burn ar

ble to defin

aterial as sho

.

n order to

nd RDX e

affected by

n the literatu

XN-109 [40

though ther

ernal pressu

ed to withs

o model low

a, strand bur

given equa

ure 9. MAT

graphical ou

MATLAB Si

input.m” fi

e thermody

d in main b

are Explosi

9

rea is enou

ne burn area

own in equa

.

solve this

nergetic m

the change

ure for char

0]; however

re are some

ures (hard ta

stand intern

w pressure

rner tests ar

ations are im

TLAB Simu

utput capabi

imulink –M

ile for initia

ynamic con

body is giv

ive Geomet

ugh for de

a as a varia

ations (13) a

s problem.

materials, PB

e of pressu

racterizing

r they are

e application

arget penetr

nal pressure

burning of

re conducte

mplemented

ulink tool is

ilities.

Main Body

als and vari

nstants. Thi

en in Figur

try and Bur

efining the

able with re

and (14).

Like othe

BXN-109 a

ure and tem

the burning

mostly ma

ns where w

rators), mos

res of those

f the explo

ed with the p

d into a M

s preferred

iables such

is file is g

re 10 where

rn Rate is

critical

espect to

13

14

er HMX

also has

mperature

g rate of

ade over

warheads

st of the

e values

sive and

prepared

MATLAB

because

as burn

given in

e as two

given in

Figuree 10. MATL

20

LAB Simullink – Solveer

3.

3.1. ST

Similar to

shown in

material.

Explosive

strand bur

well-contr

experimen

CHARAC

TRAND BU

its previou

Figure 11 i

Fi

samples ar

rner vessel i

rolled oper

ntal setup.

CTERIZAT

URNER TE

s applicatio

s used to m

igure 11. St

re ignited b

in a continu

rating cond

21

CHAPT

TION OF T

ESTS

ons [41], stra

measure pres

trand Burn

by igniter w

uous pressur

ditions. Fig

1

TER 3

THE PBXN

and burner

ssure depen

ner Cylindr

wires and bu

rized nitroge

gure 12 sh

N-109 EXPL

test setup o

ndent burn r

rical Vessel

urned insid

en supplied

hows schem

LOSIVE

of TÜBİTAK

rate of the e

l

de the high-

d environme

matic view

K SAGE

energetic

-pressure

ent under

w of the

Single p

is appli

occurred

ohms w

The stra

with a m

selected

environ

KISTLE

pressure

explosiv

stability

which i

discharg

Figure 12

piece of nich

ied through

d as the w

while the ext

and burner i

manual ope

d for the t

nmental cond

ER pressure

e within the

ve burning

y in the ves

includes a

ge of burn p

2. Schemati

hrome wire

h the nichro

wire heated.

ternal voltag

is pressurize

erated need

tests, by t

ditions.

e transducer

e chamber.

as the gas

ssel during

computer c

products aft

ic of the Str

e is used to i

ome wire b

Approxima

ge that is us

ed to the de

dle valve. N

this explosi

r (model RA

Pressure in

generation

the test is

controlled s

er the explo

22

rand Burn

ignite explo

by an exte

ate resistan

sed to start i

esired pressu

Nitrogen wh

ive combu

AG25A200

nside the ve

n takes plac

s maintaine

solenoid va

osive ignitio

er Experim

osive sample

ernal power

nce of the n

ignition is 2

ure by high

hich is an i

ustion is no

0BV1K) is u

essel tends t

ce from the

d by a pre

alve. This v

on.

mental Setu

es. An elect

r supply an

nichrome w

24 Volts.

pressure ni

nert gas is

ot affected

used to mea

to increase

e burn front

ssure contr

valve is us

up

tric current

nd ignition

wires are 3

itrogen gas

especially

d with the

asure static

during the

nt. Pressure

rol system,

sed for the

In our test

are drilled

igniter wi

holes are

for the pla

Generally

TUBITAK

wires can

a delay a

connected

During the

that the m

wires do n

two isolat

let electric

between t

with break

ts, 3 make w

d on the ex

re where it

equally spa

acement of m

Figu

, break wire

K SAGE. Bu

cause probl

after burn f

d to a circuit

e pass of bu

monitored vo

not used for

ed cables th

c to pass o

wo wires in

k wires proc

wires are us

xplosive sam

t is located

aced betwee

make wires

ure 13. Sam

es are prefe

ut in our ap

lems as som

front passes

t and are mo

urn front fr

oltage is no

r monitorin

hat are wrap

one to anoth

ncreases or

cess has bee

23

sed to meas

mples. The

on the nea

en each oth

(Figure 13)

mple Holder

erred for the

pplications,

metimes thes

s the locati

onitored if t

rom the loc

o more 5V b

ng continuou

pping on ea

her. But as

r are just sh

en reversed

3

sure burn ra

e first hole

ar top end o

her with 30

).

r and Wire

e burn rate m

we have ex

se wires are

ion of the

the supply v

ation of bre

but instead

us supply v

ach other. A

s the isolati

hort cut. Du

d and instea

ate of the ex

is used fo

of the samp

mm of disp

e Connectio

measuremen

xperienced th

e not breakin

wire etc. B

voltage is v

eak wire the

0V. Unlike

oltage. Mak

As they are i

on starts to

ue to the m

d of monito

xplosive. Fo

or the place

ple. The oth

placement a

ons

nts of prope

that the use

ng or break

Break switc

valid during

hey tend to b

e break wire

ke wires are

isolated the

o burn conn

monitoring p

oring the cu

our holes

ement of

her three

and used

ellants at

of break

king with

ches are

the test.

break so

es, make

e simply

ey do not

nectivity

problems

ut signal,

short sig

is show

1V cont

Five dif

Unfortu

the flam

rates of

Althoug

and 10-

the data

gnal is mon

wn in Figure

tact signal c

fferent pres

unately, belo

me front an

f PBXN-109

gh there see

-500 MPa, t

a obtained w

nitored. A sa

e 14. Althou

can be obser

Figure 1

ssure range

ow 2 MPa

d tend to b

9 are given

ems to be a

test data ma

within this st

ample, mon

ugh all thre

rved clearly

14. Sample

es is tested

range, it is

burn out jus

in Figure

a discontinu

ade by the L

tudy.

24

nitored data

ee signals h

y.

Make Wir

between 2

s observed t

st after expl

15 and Figu

uity between

Livermore f

a from a test

have noisy

re Readings

2-10 MPa w

that the exp

losive is ig

ure 16 with

n the curve

for 10-20 M

t (PBXN-10

oscillation

s

with three r

plosive cann

nited. Meas

h some liter

fit data of

MPa is cons

09, 2 MPa)

between ±

repetitions.

not sustain

sured burn

rature data.

f 1-10 MPa

sistent with

Fig

Figu

gure 15. PB

ure 16. PBX

BXN-109 B

XN-109 Bu

25

Burn Rate C

urn Rate Ch

5

Change wit

hange with

h Pressure

h Pressure (

e (0-10MPa

(1-1000MP

a)

Pa)

26

4.1. De

There are

contacting

designed t

test item i

liner and a

tested. Ve

SAGE tha

and chose

enter the t

test as thre

Test items

temperatu

escription of

2 differen

g cylinder c

to investiga

is shown in

a thermopla

entilation ho

at can melt a

en to be thi

test chambe

ead or tight

s are design

ure and burn

Fig

4. FAS

f the Test Ite

nt types of

casing. The

ate if there

n Figure 17

astic liner) a

oles are clos

at temperatu

ck enough

er before th

fit accordin

ned so that o

ning rate fro

gure 17. Ge

27

CHAPT

ST COOK-

em

test items

e length rat

is an effect

7. Two diffe

and two diff

sed by an e

ures around

so that the

he ignition

ng to the ve

one can be a

om the intern

eneric Test

7

TER 4

-OFF TEST

which diff

tio over tes

t of length

ferent liner

ferent thickn

eutectic mat

d 400 K. Th

y will not m

starts. Plug

ntilation dia

able to take

nal parts as

t Item (Sho

TS

fer on the l

st specimen

on time to

types (HTP

nesses (1 m

terial develo

hickness of

melt compl

gs are manu

ameter choi

internal me

shown in F

ort Version)

length of e

ns are abou

reaction. T

PB based th

mm and 3.5

oped by TÜ

the plugs a

letely and l

ufactured af

ice.

easurements

Figure 17.

n)

xplosive

ut ¾ and

The short

hermoset

mm) are

ÜBİTAK

re tested

let flame

fter each

s such as

There a

there ar

other. M

front is

hole tha

the sect

rod that

casting.

applicat

Figure

Some p

added t

change

applied

are 4 steel s

re 4 inner a

Make switch

moving wit

at let make s

tions are as

t cables of

Silicone is

tion is show

e 18. Polyc

pre-analysis

to test item

time to rea

in order to

sections loc

and 4 outer

hes are con

thin the exp

switches co

sembled to

f make swit

also used f

wn in Figure

arbonate R

is made to

m do not aff

action. How

o protect the

cated on a p

r holes in e

nnected to t

plosive duri

ome through

the test ite

tches are p

for preventin

e 18.

Rod for Hol

Exp

o ensure th

fect the hea

wever, insula

e cables out

28

polycarbona

every sectio

these holes

ing the test.

h the back p

em before th

passing is f

ng short cir

lding Steel

plosive

hat the appl

ating rate o

ation at the

t of the test

ate rode wi

on located a

in order to

The rod is

plate of the

he explosiv

filled with

rcuit betwee

Sections T

lication of

over explosi

e back of th

t item. Alth

thin the tes

at 90 degre

o locate how

designed a

test item. T

e is cast. Th

a silicone

en make swi

Together W

internal rea

ive and hen

he test item

ough insula

st item and

ees to each

w the burn

as there is a

The rod and

The hole on

before the

itches. The

Within the

ading parts

nce do not

m should be

ation at the

29

back side does affect the heating regime over the test item, ignition point of the

explosive does not change.

Other than internal readings, one pressure sensor (KISTLER RAG25A80BV1K) and

one temperature sensor (ORDEL KTTE3x0.50 5K) are used on the test item. As

flame temperatures are expected to be around 2500 K according to the NASA CEA

solutions, thermocouple has been placed close to wall to avoid bead getting into

ventilation flow path of burn products of explosive and becoming too hot and

destroyed. To protect the pressure sensor from the high temperature environment it is

placed (buried) outside of the test pool. Connection of the sensor to the test item is

made via Ø3 mm stainless steel pipe. The volume increase of test chamber by adding

a pipe for the connection of the pressure will delay the instantaneous pressure

readings due to additional volume added.

4.2. Experimental Setup

Tests have been performed with the mini fuel fire test setup as the test item

dimensions are within the range of declaration of NATO standard 4240 [41].

Total of eight surface thermocouples (STC) is used to measure test item surface

temperature. Measurements are made both in the front and rear side. Thermocouples

are placed circumferentially with an angle of 90º (Figure 19). To avoid contact

dislocations and increase the heat transfer to STCs a thermal paste (Thermigrease TG

20033) is used which can withstand temperatures up to 1200°C and has relatively

high thermal conductivity.

Total nu

area. Al

some te

conditio

Temper

switch m

Data Tr

Figu

umber of 8

lthough the

ests, back

ons with pre

ratures mea

measuremen

rap 2 – S/N:

ure 19. The

tests has be

ere are no in

side of tho

evious tests.

asurements

nts are take

: 9077)

ermocouple

een made fo

nternal mea

ose test item

.

are taken

en with 100

30

e Placemen

or the determ

asurements

ms are stil

with 100 H

00 Hz by ou

nt on 1st Tes

mination of

taken from

ll isolated t

Hz wherea

ur data acqu

st Item

f TtR and cr

inside of t

to simulate

s pressure

uisition syst

ritical vent

test case in

e the same

and make

tem. (Mrel

5.1. Fa

5.1.1.

Test item

mm. It ha

mentioned

temperatu

conditions

According

melting on

5

ast Cook-O

Test 1

1 is a long

as a 3.5 m

d above the

ures on the

s accordingl

g to the firs

n the test ite

5. RESU

Off Test Res

version of

mm of HTP

difference

casing for

ly. Test item

Figure 20.

st inspection

em is observ

31

CHAPT

ULTS AND

sults

f two types.

PB liner be

of test 1 an

determinati

m is shown

. Test Item

n made afte

ved as show

1

TER 5

DISCUSS

It has a ven

tween the

nd other test

ion of surfa

in Figure 20

1 – Before

er tests no d

wn in Figure

ION

ntilation ho

explosive a

t items is th

ace tempera

0.

the Test

deformation

e 21.

ole diameter

and the cas

hat there are

ature and b

n due to pre

r of 32.9

sing. As

e surface

boundary

essure or

After th

after 16

and othe

Temper

to be ap

[42]. Su

conditio

Internal

(361.4s)

explosiv

after wh

he ignition o

61 seconds o

er events th

rature measu

ppropriate a

urface temp

ons and disc

l thermocou

) and t0+287

ve whereas

hole explosi

Figure

of the fuel w

of first data

hat are detec

urements ar

according t

eratures are

cussed withi

uple reading

7s (448s). F

the second

ive burn out

21. Test Ite

within the p

is recorded

cted during t

re given in F

to the requi

e taken to e

in the comp

gs show tha

First one is c

d one is due

t.

32

em 1 – Afte

pool first in

d. This is ac

the tests wi

Figure 22. F

irements m

evaluation fo

parison of te

at there are

considered

e to the hot

er the Test

ncrease in te

ccepted to b

ll be referen

Flame temp

mentioned by

or the deter

ests with an

two steep i

to be as a re

t flame entr

emperature

e the referen

nced to this

eratures are

y the STAN

rmination of

alyses secti

increases at

esult of the

rance to tes

is detected

ence time t0

point.

e examined

NAG 4240

f boundary

ion.

t t0+200.4s

ignition of

st chamber

Figure 2

Chamber

chamber s

t0+213s (3

whereas s

softens (a

longer elig

22. Flame,

pressure re

starts to in

374s). Incre

sudden drop

and melts a

gible to sust

Surface an

eadings giv

crease at t0

ease of press

p indicates

at outer ope

tain high int

33

nd Internal

ven in Figu

0+200.3s (3

sure is cons

the remova

en surfaces

ternal press

3

Temperatu

ure 23 indi

361.3s) and

sidered to b

al of ventila

to flame)

sures after c

ure Readin

icate that p

d suddenly

be the igniti

ation plug d

and is estim

ertain point

ngs of Test

pressure wi

drops to “

ion of the e

debris. As

mated that

t.

Item 1

ithin the

“zero” at

xplosive

the plug

it is no

Althoug

useful d

item.

Accordi

recordin

ignition

5.1.2.

Test ite

mm. It

Test item

Figu

gh cables th

data could b

ing to the

ngs, flame

n of fuel.

Test 2

m 2 is a sh

has a 3.5 m

m is shown

ure 23. Inte

hat are com

be acquired

visualizati

starts to co

hort version

mm of ther

in Figure 2

ernal Press

ing out of t

from the m

on of test

ome out of

of two typ

rmoplastic l

24.

34

sure Readin

the test item

make switch

video reco

ventilation

pes. It has a

liner betwee

ngs of Test

m seems to

hes that are

ording and

hole 200 s

a ventilation

en the expl

Item 1

be well pro

placed with

inspection

seconds afte

n hole diam

losive and t

otected, no

hin the test

n of audio

er the first

meter of 5.3

the casing.

According

the test ite

After the i

after 140.2

other even

g to the firs

em due to pr

ignition of

2s of first d

nts detected

Figure 24.

t inspection

ressure or m

Figure 25

the fuel wit

data is recor

during the

35

. Test Item

n made afte

melting is ob

5. Test Item

thin the poo

rded. This i

tests will be

5

2 – Before

er the test, d

bserved as s

m 2 – After

ol first incre

is accepted

e referenced

the Test

deformation

shown in Fi

the Test

ease in temp

to be the re

d to this poi

n on the sid

igure 25.

mperature is

eference tim

int.

dewall of

detected

me t0 and

Temper

observe

readings

be as a r

F

Chambe

chambe

t0+178s

explosiv

being to

rature measu

ed to be app

s show that

result of the

Figure 26. F

er pressure

er starts to

(318.2s).

ve whereas

orn.

urements ar

propriate wit

t there is a s

e ignition of

Flame and I

readings g

increase a

Increase o

sudden dro

re given in

th STANAG

steep increa

f explosive.

Internal Te

given in Fi

at t0+172s

of pressure

op indicates

36

Figure 32.

G 4240 requ

ase at t0+17

emperature

igure 27 in

(312.2s) an

is consid

s the possib

Flame tem

uirements. I

1.8s (312s).

e Readings

ndicate that

nd suddenly

ered to be

ble reason

mperature re

Internal the

. This is con

s of Test Ite

t pressure

y drops to

e the igniti

of test item

eadings are

ermocouple

nsidered to

em 2

within the

“zero” at

ion of the

m side wall

Although

useful dat

item.

According

recordings

ignition of

5.1.3.

Test item

mm. It ha

Test item

Figure

cables that

a could be

g to the vi

s, flame sta

f fuel.

Test 3

3 is a short

as a 3.5 mm

is shown in

e 27. Intern

are coming

acquired fro

isualization

arts to come

t version of

m of thermo

n Figure 28.

37

nal Pressur

g out of the

om the mak

n of test vi

e out of ve

f two types.

oplastic line

7

re Readings

e test item s

ke switches

ideo record

entilation ho

It has a ve

er between

s of Test Ite

seems to be

that are pla

ding and in

ole 172 sec

ntilation ho

the explos

em 2

e well prote

aced within

nspection o

conds after

ole diameter

sive and the

ected, no

n the test

of audio

the first

r of 12.3

e casing.

Accordi

melting

Instead

thermoc

steel se

(Figure

ing to the f

on the test

of using

couples are

ction (Figu

31) of the t

Figure 2

first inspecti

item is obs

Figure

make swi

used within

ure 30) whe

test item to

28. Test Ite

ion made a

erved as sho

29. Test Ite

itches in t

n the test ite

ereas the 4t

measure lin

38

em 3 – Befo

after tests no

own in Figu

em 3 – Afte

this applica

em. 3 of theth one is lo

ner temperat

ore the Test

o deformati

ure 29.

er the Test

ation, total

e thermocou

cated on th

ture and det

t

ion due to p

l of 4 sur

uples are pla

he middle o

tect side ign

pressure or

rface type

aced on the

of sidewall

nition.

Figure 30

Figure 3

After the i

after 158 s

and other

Temperatu

observed t

readings s

0. Surface T

31. Surface

ignition of

seconds of f

events dete

ure measure

to be approp

show that

ITC-1

Type Therm

e Type Ther

the fuel wit

first data is

cted during

ements are

priate with

there is a

ITC

Liner S

Therm

39

mocouples

rmocouple

Item Sid

thin the poo

recorded. T

g the tests w

given in Fi

STANAG 4

slight incr

C-2

Side Wall

mocouple

9

That Are L

That Is Lo

dewall

ol first incre

This is accep

will be refere

igure 32. F

4240 requir

rease at t0+

ITC-3

Located on

ocated in th

ease in temp

pted to be th

enced to this

lame tempe

rements. Int

+178s (336

n the Steel S

he Middle o

mperature is

the referenc

s point.

erature read

ternal therm

6s) conside

Sections

of Test

detected

ce time t0

dings are

mocouple

ering the

tempera

of the ig

energeti

bead is

number

F

Side wa

t0+173s

tempera

increase

Other th

explosiv

explosiv

failed t

protecte

thermoc

burn fro

whereas

ature increa

gnition of e

ic materials

left within

r of instrume

Figure 32. F

all temperat

(331s) m

ature for P

e in tempera

hermocoupl

ve shows s

ve is relativ

o measure

ed and bre

couples are

ont surface

s this time i

ase over the

explosive. T

s is less com

n the body o

entation fail

Flame and I

ture starts t

measured te

PBXN-109.

ature can be

les that are

lower incre

vely low. It

flame tem

eak instead

tried to be

has interac

is estimated

whole test

The total in

mpared to

of front clo

lure.

Internal Te

to increase

emperature

As self-ign

e observed.

placed on

ease in temp

is observed

mperature as

d. However

estimated a

cted with IT

d to be 162 s

40

t period. Th

ncrease in te

first two te

osure. This

emperature

as the flam

exceeds 2

gnitions occ

steel sectio

mperature as

d that these

s the bead

r, the time

according to

TC-1 56 sec

seconds for

his is consid

emperature

ests as the i

is done be

e Readings

me within th

200 °C and

cur at t0+1

ons which a

the therma

surface typ

of these th

e that flam

o increase ra

conds after

the TC-2 a

dered to be

during the

internal the

cause to de

s of Test Ite

he pool is i

d reaches

78s (336s)

are located

al conductiv

e thermocou

hermocoupl

me interacts

ate. It is esti

the ignition

and TC-3. R

as a result

ignition of

ermocouple

ecrease the

em 3

ignited. At

to critical

a steeper

within the

vity of the

uples were

les are not

s with the

imated that

n occurred

Relaying on

the assum

explosive

thermocou

are measu

TC-2 and

whereas 0

Figure

Chamber

chamber s

Increase o

t0+266s is

“zero”. T

pressures

mption that

burn speed

uple to exp

ured and giv

d TC-3 the

0.125 mm/s

33. Section

pressure re

starts to inc

of pressure

s sudden bu

his is assu

falling belo

flame cov

d can be ca

losive surfa

ven in Figur

burn speed

for TC-1.

n View of T

Poin

eadings giv

crease at t0+

is consider

ut the press

umed to be

ow “zero” af

41

vers all ex

alculated by

ace. Using

re 33. If the

d of the ex

Test Item 3

nts to Explo

ven in Figu

+177s (335

red to be th

sure that is

e a pressur

fter 500 sec

1

xplosive su

y knowing t

the 3-D mo

e calculation

xplosive can

and Dimen

osive Surfa

ure 34 indi

s) and sudd

he ignition

measured

re transduce

conds of firs

urface just

the distance

odel of the

n is made fr

n be found

nsions from

ace

icate that p

denly drops

of the expl

at that mom

er failure e

st data recor

after the

e (shortest)

test item d

rom the side

to be 0.15

m the Meas

pressure wi

s at t0+266s

losive. The

ment is not

especially w

rding.

ignition,

of each

distances

e wall to

51 mm/s

uring

ithin the

s (424s).

e drop at

t exactly

with the

Accordi

recordin

ignition

5.1.4.

Test ite

mm. It h

item is

further t

get succ

Figu

ing to the

ngs, flame

n of fuel.

Test 4

m 4 is a sh

has a 1 mm

shown in F

test as no im

cessful data

ure 34. Inte

visualizati

starts to co

hort version

m of thermop

Figure 35. T

mprovemen

.

ernal Press

on of test

ome out of

of two typ

plastic liner

There are n

nt could be

42

sure Readin

video reco

ventilation

pes. It has a

r between th

no make sw

made on th

ngs of Test

ording and

hole 179 s

a ventilation

he explosiv

witches are

he applicatio

Item 3

inspection

seconds afte

n hole diam

e and the ca

used for th

on and it ha

n of audio

er the first

meter of 8.7

asing. Test

his test and

as failed to

According

melting on

After the i

after 216 s

and other

Temperatu

observed t

measurem

g to the firs

n the test ite

ignition of

seconds of f

events dete

ure measure

to be appro

ments are no

Figure 35.

st inspection

em is observ

Figure 36

the fuel wit

first data is

cted during

ements are

opriate with

ot given as

43

. Test Item

n made afte

ved as show

6. Test Item

thin the poo

recorded. T

g the tests w

given in Fi

h STANAG

the thermo

3

4 – Before

er tests no d

wn in Figure

m 4 – After

ol first incre

This is accep

will be refere

igure 37. F

G 4240 requ

ocouple was

the Test

deformation

e 36.

the Test

ease in temp

pted to be th

enced to this

lame tempe

uirements. In

s failed dur

n due to pre

mperature is

the referenc

s point.

erature read

nternal tem

ring the test

essure or

detected

ce time t0

dings are

mperature

t and no

useful d

tests be

pressure

F

Chambe

chambe

Increase

drop ind

drops to

data is obta

ecause lack

e readings.

Figure 37. F

er pressure

er starts to in

e of pressur

dicates the

o zero, it sta

ained. Intern

of thermoc

Flame and I

readings g

ncrease at t

re is conside

removal of

arts to rise to

nal measure

couples and

Internal Te

given in Fi

0+176s (392

ered to be th

f ventilation

o 0.015 MP

44

ement therm

d it is possi

emperature

igure 38 in

2s) and sud

the ignition

n plug debr

Pa and drops

mocouple i

ble to indic

e Readings

ndicate that

ddenly drops

of the expl

ris. 20 secon

s to zero aga

s replaced

cate ignition

s of Test Ite

t pressure

s to “zero”

osive where

nds after th

ain slowly.

for further

n from the

em 4

within the

at t0+204s.

eas sudden

he pressure

According

recordings

ignition of

5.1.5.

Test item

mm. It ha

item is sho

Figure

g to the vi

s, flame sta

f fuel.

Test 5

5 is a shor

s a 1 mm o

own in Figu

e 38. Intern

isualization

arts to come

rt version o

f thermopla

ure 39.

45

nal Pressur

n of test vi

e out of ve

of two type

astic liner b

5

re Readings

ideo record

entilation ho

s. It has a

between the

s of Test Ite

ding and in

ole 177 sec

ventilation

explosive a

em 4

nspection o

conds after

hole diame

and the casi

of audio

the first

eter of 8

ing. Test

Accordi

melting

After th

after 24

and othe

ing to the f

on the test

he ignition o

44 seconds o

er events de

Figure 3

first inspecti

item is obs

Figure

of the fuel w

of first data

etected durin

39. Test Ite

ion made a

erved as sho

40. Test Ite

within the p

is recorded

ng the tests

46

em 5 – Befo

after tests no

own in Figu

em 5 – Afte

pool first in

d. This is ac

will be refe

ore the Test

o deformati

ure 40

er the Test

ncrease in te

ccepted to b

erenced to t

t

ion due to p

emperature

e the referen

this point.

pressure or

is detected

ence time t0

Temperatu

observed t

the lack o

test.

Figu

Chamber

chamber s

0.015 MP

pressure lo

of ventilat

ure measure

to be appro

of thermoco

ure 41. Fla

pressure re

starts to inc

Pa and reach

oss during

tion plug de

ements are

opriate with

ouples no in

ame and Int

eadings giv

rease at t0+

hes to a pea

the explosiv

ebris remova

47

given in Fi

h STANAG

nternal tem

ternal Tem

ven in Figu

+106s. Press

ak value of

ve burn at t

al from the

7

igure 41. F

4240 requi

mperature m

mperature R

ure 42 indi

sure than in

f 0.032 MPa

t0+138s. Th

ventilation

lame tempe

irements. A

measurement

Readings of

icate that p

ncreases to a

a. It is obse

his is consid

hole.

erature read

As mentione

ts are taken

f Test Item

pressure wi

an average

erved that th

dered to be

dings are

ed due to

n in this

m 5

ithin the

value of

here is a

because

Accordi

recordin

ignition

5.1.6.

Test item

mm. It h

item is s

Figu

ing to the

ngs, flame

n of fuel.

Test 6

m 6 is a lon

has a 1 mm

shown in Fi

ure 42. Inte

visualizati

starts to co

ng version

m of thermop

igure 43.

ernal Press

on of test

ome out of

of two type

plastic liner

48

sure Readin

video reco

ventilation

es. It has a v

r between th

ngs of Test

ording and

hole 106 s

ventilation

he explosiv

Item 5

inspection

seconds afte

hole diame

e and the ca

n of audio

er the first

eter of 6.85

asing. Test

According

the test ite

After the i

after 314 s

and other

g to the firs

em due to pr

ignition of

seconds of f

events dete

Figure 43.

t inspection

ressure or m

Figure 44

the fuel wit

first data is

cted during

49

. Test Item

n made afte

melting is ob

4. Test Item

thin the poo

recorded. T

g the tests w

9

6 – Before

er the test, d

bserved as s

m 6 – After

ol first incre

This is accep

will be refere

the Test

deformation

shown in Fi

the Test

ease in temp

pted to be th

eced to this

n on the sid

igure 44.

mperature is

the referenc

point.

dewall of

detected

ce time t0

Temper

observe

F

Chambe

chambe

estimate

After th

pressure

sidewal

explosiv

rature measu

ed to be app

Figure 45. F

er pressure

er starts to

ed due to t

hat instance

e drop after

l of the tes

ve is consid

urements ar

ropriate wit

Flame and I

readings g

increase at

the removal

e pressure

r the peak v

st item. Wi

dered to dec

re given in

th STANAG

Internal Te

given in Fi

t t0+95s. P

l of ventila

starts to in

values is re

th the incre

rease.

50

Figure 45.

G 4240 requ

emperature

igure 46 in

Pressure sta

ation plug d

ncrease to a

ached is pr

eased venti

Flame tem

uirements.

e Readings

ndicate that

arts to drop

debris from

a peak valu

robably bec

lation area,

mperature re

s of Test Ite

t pressure

ps at t0+97s

m the ventila

ue of 0.35

ause of the

, burn veloc

eadings are

em 6

within the

s which is

ation hole.

MPa. The

e torn apart

city of the

According

recordings

ignition of

5.1.7.

Test item

mm. It has

shown in F

Figure

g to the vi

s, flame sta

f fuel.

Test 7

1 is a long

s a 1 mm o

Figure 47.

e 46. Intern

isualization

arts to com

version of

f HTPB lin

51

nal Pressur

n of test vi

me out of v

f two types.

ner between

1

re Readings

ideo record

entilation h

It has a ven

the explosi

s of Test Ite

ding and in

hole 95 sec

ntilation ho

ive and the

em 6

nspection o

conds after

ole diameter

casing. Tes

of audio

the first

r of 6.85

st item is

Accordi

the test

After th

after 24

and othe

ing to the f

item due to

he ignition o

40 seconds o

er events de

Figure 4

first inspecti

o pressure or

Figure 4

of the fuel w

of first data

etected durin

47. Test Ite

ion made a

r melting is

48. . Test It

within the p

is recorded

ng the tests

52

em 7 – Befo

after the test

observed a

tem 7 – Aft

pool first in

d. This is ac

will be refe

ore the Test

t, deformati

as shown in

ter the Test

ncrease in te

ccepted to b

erenced to t

t

ion on the s

Figure 48.

t

emperature

e the referen

this point.

sidewall of

is detected

ence time t0

Temperatu

from the 4

readings a

Figu

Chamber

chamber s

first pressu

the ventil

increases

decrease

Unfortuna

evidence

chamber.

and smalle

ure measure

4th thermoc

are observed

ure 49. Fla

pressure re

starts to inc

ure drop is

lation hole.

to a value

is consider

ately, pressu

of second

Thus, press

er ventilatio

ements are g

couple. But

d to be appr

ame and Int

eadings giv

crease at t0+

considered

. After tha

e of 1.7 M

red to be d

ure starts to

hole is sti

sure starts t

on hole is fo

53

given in Fig

t still it is p

ropriate with

ternal Tem

ven in Figu

+107s and s

to be the re

at pressure

MPa and rap

due to the

o increase ag

ill not eno

to increase t

ormed.

3

gure 49. No

possible to

h STANAG

mperature R

ure 50 indi

suddenly dr

emoval of th

starts to i

pidly decre

e torn apart

gain at t0+1

ough to sus

till the seco

o useful data

accept that

G 4240 requi

Readings of

icate that p

rops to “zer

he ventilatio

increase at

eases afterw

t side wall

129s. This i

stain the pr

ond vent ho

a could be

t flame tem

irements.

f Test Item

pressure wi

ro” at t0+10

on plug deb

t t0+116s.

wards. This

l of the te

is accepted

ressure low

ole enlarges

obtained

mperature

m 7

ithin the

08s. The

bris from

Pressure

s second

est item.

to be an

w in the

or third

Accordi

recordin

Figu

ing to the

ngs, flame s

Figur

ure 50. Inte

visualizati

starts to com

re 51. Defla

ernal Press

on of test

me out of ve

agration of

54

sure Readin

video reco

entilation ho

f PBXN-109

ngs of Test

ording and

ole at t0+107

9 Explosive

Item 7

inspection

7s.

e, Test 7

n of audio

According

recordings

ignition of

5.1.8.

Test item

mm. It has

shown in F

According

melting on

g to the vi

s, flame sta

f fuel.

Test 8

1 is a long

s a 1 mm o

Figure 52.

g to the firs

n the test ite

isualization

arts to come

g version of

f HTPB lin

Figure 52.

st inspection

em is observ

55

n of test vi

e out of ve

f two types.

ner between

. Test Item

n made afte

ved as show

5

ideo record

entilation ho

. It has a ve

the explosi

8 – Before

er tests no d

wn in Figure

ding and in

ole 107 sec

entilation h

ive and the

the Test

deformation

e 53.

nspection o

conds after

hole diamete

casing. Tes

n due to pre

of audio

the first

er of 7.1

st item is

essure or

After th

after 76

and othe

Temper

given in

STANA

he ignition o

6 seconds of

er events de

rature meas

n Figure 32

AG 4240 req

Figure

of the fuel w

f first data i

etected durin

urements a

2. Flame tem

quirements.

53. Test Ite

within the p

is recorded.

ng the tests

re given in

mperature re

56

em 8 – Afte

pool first in

. This is acc

will be refe

n Figure 54.

eadings are

er the Test

ncrease in te

cepted to be

erenced to t

. Temperatu

e observed t

emperature

e the referen

this point.

ure measure

to be approp

is detected

nce time t0

ements are

priate with

Figu

Chamber

chamber

Increase o

drop indic

ure 54. Fla

pressure re

starts to in

of pressure i

cates the rem

Figure

ame and Int

eadings giv

ncrease at t

is considere

moval of ven

e 55. Intern

57

ternal Tem

ven in Figu

t0+114s an

ed to be the

ntilation plu

nal Pressur

7

mperature R

ure 55 indi

d suddenly

ignition of

ug debris.

re Readings

Readings of

icate that p

y drops to

f the explosi

s of Test Ite

f Test Item

pressure wi

“zero” at

ive whereas

em 8

m 8

ithin the

t0+116s.

s sudden

Accordi

recordin

ignition

5.2.

5.2.1.

Flame t

used to

done by

piecewi

time ra

curves

radiatio

Temper

surface

the ana

tempera

ing to the

ngs, flame

n of fuel.

Compariso

Determi

temperature

simulate bo

y averagin

ise high ord

anges (t[0

are used a

n temperatu

Figure

rature measu

are also av

alysis repre

atures are gi

visualizati

starts to co

on of Tests

ination of “

e measurem

oundary con

g the flam

der polynom

0,33.1s), t

as an input

ure.

56. Flame

urements th

veraged to c

esents thes

iven in Figu

on of test

ome out of

and Analy

“True” Bou

ents that ha

ndition of th

me temperat

mial curves

[33.1s,71.1

parameter

Temperatu

hat are taken

check if the

e measurem

ure 57.

58

video reco

ventilation

yses

undary Co

ave been m

he test item

ture over t

s are fitted

s), t[71.1

r for free s

ure Readin

n both from

e boundary

ments. Bot

ording and

hole 114 s

ndition

ade around

model in A

the entire

(Figure 56)

1s,130.2s) a

stream temp

ngs from th

m front side

condition th

th measure

inspection

seconds afte

the first te

ANSYS Flue

event. Afte

) along fou

and t130.2

perature an

he 1st Test

surface an

hat is imple

ements and

n of audio

er the first

est item are

ent. This is

erwards, 3

ur different

2s). These

nd external

nd rear side

emented to

d averaged

Figure 5

After the

compared

results bet

57. (a) Fron

analyses of

with the t

tween test a

nt Side, (b)

f the first te

test measur

and analysis

59

(a)

(b)

) Rear Side

the 1st Te

est item, su

rements. A

s are observ

9

)

)

e Surface T

est Item

urface temp

ccording to

ed as it can

emperatur

peratures fro

o the comp

be seen in F

re Readings

om the ana

parison inco

Figure 58.

s from

lysis are

onsistent

Figure

From th

represen

common

measuri

type of

by Shan

are due

the env

tempera

interest

and from

reaction

Beside

between

is also c

time req

heated.

between

58 Initial N

he results, it

nt real heat

n sense is t

ing flame te

the thermo

nnon and B

to the temp

vironments

atures signi

[44]. These

m the TC

ns between t

instantaneo

n temperatu

commented

quired to tr

The greate

n the thermo

Numerical

Te

t is obvious

t transfer a

to use dura

emperature,

ocouple cho

Butler [43].

perature bet

with high

ficantly dif

e errors are

bead, temp

the metals c

ous tempera

ure increases

in Shannon

ansfer ener

er the mass

ocouple and

Results of

est Surface

s that define

applied to t

able, long la

, the error o

ice. Reason

Most error

tween senso

intensity li

fferent than

attributed t

perature var

comprising

ature measu

s in the very

n and Butle

gy to the ce

s of the th

d actual gas

60

Test Flame

e Temperat

ed thermal b

the test item

asting therm

of the meas

ns of measu

rs associated

or and surro

ike fires, th

n the actual

to variation

riations alo

the bead at

urement err

y beginning

er’s work [4

enter of the

hermocouple

temperatur

e Temperat

tures

boundary co

m during th

mocouples

urement is

urement erro

d with the

ounding med

hermocoupl

l temperatu

ns in the rate

ng the lead

the surroun

ror of the th

g of the test

43] as this d

e thermocou

e leads the

re.

ture Comp

ondition is n

he test. Alt

for the app

directly rel

or are well

use of therm

dium. When

les are tend

ure of the m

e of energy

d wires, an

nding gases.

hermocoupl

is also obse

delay is beca

uple bead w

greater the

pared with

not able to

though the

plication of

lated to the

concluded

rmocouples

n placed in

d to sense

medium of

transfer to

nd catalytic

.

le, a delay

erved. This

ause of the

when being

e lag time

In our app

for both g

necessary

temperatu

As it is n

condition,

readings t

flame tem

Fi

After anal

curve show

the test tem

plications, 3

gas and surf

as indicat

ure sensor m

not possible

flame tem

that are tak

mperature da

igure 59 Tu

lyzing diffe

wn in Figur

mperature m

mm thick th

face temper

ted in NAT

might lead to

e to use tem

mperature h

ken from th

ata is given i

uned and Si

erent simpli

re 60 is obt

measuremen

61

hermocoupl

rature measu

TO publica

o repetition

mperature d

has been tu

he first test

in Figure 59

implified F

ified flame

tained from

nts.

1

les are used

urements as

ation [42],

of a test.

data directly

uned to sa

t and obtain

9.

Flame Tem

temperature

m Fluent wh

d (ORDEL K

s survivabil

[5]. Other

y in the ana

atisfy the s

ned from th

perature D

e distributio

hich success

KTTE3x0.5

lity along th

rwise, failu

nalyses as b

surface tem

he analyses

Distribution

ons as a mo

sfully match

50 Ö 5K)

he test is

ure of a

boundary

mperature

s. Tuned

n

odel, the

hes with

Figure

Correcte

analyses

5.2.2.

After th

TtR va

axisymm

Time st

Table 3

not affe

e 60. Surfa

ed and sim

s.

Comput

of the R

he first analy

lues are co

metric appr

tep and elem

and Table

ect the result

ace Temper

mplified tem

tation Resu

Results with

yses, it is co

ompared. T

roach. Intern

ment size de

4 and tune

t by more th

rature Com

Test

mperature bo

ults for Tim

h Test Resu

oncluded th

Thus, furthe

nal parts are

ependency o

ed as the eff

han 0.1%.

62

mparison of

t Case 1

oundary con

me to React

ults

hat the effec

er analyses

e not mode

of the analy

ffect of both

f Test and N

ndition is a

tion Analys

ct of buoyan

s are carrie

eled in axisy

yze model is

h element si

Numerical

applied for t

ses and Com

ncy is below

ed out with

ymmetric si

s checked a

ize and tim

Results -

the further

mparison

w 1% when

h the 2-D

imulations.

as shown in

me step will

Total of 4

version of

item conta

Figure 61

Run Typ

Fine Mesh

MediumMesh

Coarse Mesh

RunNumb

1

2

3

4

5

6

49334 eleme

f the analys

aining 1 mm

1. Mesh Dis

Table 3

e Max

Element

h 0.000

0.001

0.002

Table 4

n ber

TimeSi

0

0

0

1

2

1

ents in 2-D

ses. Both m

m of HTPB l

stribution (

63

3. Mesh Re

x t Size

NumElem

05 49

1 12

2 34

4. Time Re

e Step ize

0.1

0.2

0.3

1

2

10

are used w

mesh constr

liner (for te

(top) and T

Items

3

efinement S

ber of ments

TtR[s]

334 189

724 190

407 196

efinement S

TtR D

189.7

189.6

188.9

188

186

180

whereas 1654

ruction and

est cases 7-8

Temperatur

s 7-8

Study

R

DifferenFine Me

.7 0.0

.6 0.4

.4 3.5

Study

ifference withRun (%)

0.00

0.05

0.42

0.90

1.95

5.11

4200 eleme

d temperatur

8) are shown

re Gradien

nce with esh (%)

00

47

53

th 1st

ents are use

ure gradient

n in Figure

nt (bottom)

d in 3-D

s of test

61.

of Test

Ignition

explosiv

(Eq. 4)

above t

generate

Figu

Results

CPU X5

Summa

to reacti

n can be d

ve volume.

that is impl

the critical

ed is conver

ure 62. Max

are given fo

5650 @ 2.6

ary of the te

ion are give

detected by

As specifie

lemented in

temperatur

rged to infin

ximum Tem

for fine mesh

67GHz (2 pr

est configura

en in Table

y tracking

ed above, h

to the Fluen

e, which is

nite so does

mperature P

h since com

rocessors) c

ations and c

5.

64

the maxim

heat generat

nt by UDF t

s in this ca

s the temper

Plot of the

mputation tim

computer.

comparison

mum tempe

tion is a fu

tool. If temp

ase ~220°C

rature (Figu

Explosive a

me is low in

n of calculat

erature plot

nction of te

perature of

for PBXN

ure 62).

and Ignitio

n an 8 core

ted and obs

t over the

emperature

a cell rises

N-109, heat

on Time

Intel Xeon

served time

65

Table 5. Comparison of Calculated and Observed TtR

Test Item No

Explosive L/D

Liner Thickness

[mm] Liner Type

Predicted TtR [s]

ObservedTtR [s]

Difference [s]

Error (Experiment - Numerical)

%

1 2.67 3.50 HTPB Based Thermoset

189.7 200 -10.3 -5.2

2 1.71 3.50 Thermoplastic

Liner 179.7 172 7.7 4.5

3 1.71 3.50 Thermoplastic

Liner 179.7 178 1.7 1.0

4 1.71 3.50 Thermoplastic

Liner 179.7 177 2.7 1.5

5 1.60 1.00 Thermoplastic

Liner 98.8 106 -7.2 -6.8

6 1.60 1.00 Thermoplastic

Liner 98.8 95 3.8 4.0

7 1.60 1.00 HTPB Based Thermoset

103.8 107 -3.2 -3.0

8 1.60 1.00 HTPB Based Thermoset

103.8 114 -10.2 -8.9

The largest difference between the predicted and observed ignition times occurred

with 8.9 percent difference for 8 tests. Addressing the 8.9% difference, the prediction

does not take into account the decomposition/melting of the liner and decomposition

of the explosive that occurs before ignition in the analyses. It is assumed that the

liner thickness is constant along the solution. These phase shifting and/or chemical

reactions are thought to effect heat transfer to explosive through the test item hence

increasing the error.

Another source of the error is the unavailability of simulating the variation of

temperature in medium hence on the test item. Flame temperature may directionally

vary within the hearth either caused by wind or the chaotic environment of its nature.

as shown in Figure 56 different thermocouples have different temperature readings.

The temperature distribution between thermocouples is shown in Figure 57. As

mentioned above, the average of all 4 thermocouples is used as a temperature

boundary condition. On the other hand, because of directionality of the flame, heat

flux on one surface might be higher than the other three causing a hot spot along the

circular direction. Thus, even if by small amounts we are already expecting to have

different ignition times for the test item having the same geometry and liner

type/thi

to be clo

5.2.3.

Determi

each te

fragmen

the cha

Therefo

measure

from di

having

to sprea

applicat

taken as

Figu

ckness. Eve

ose to the ex

Comput

with Te

ination of

est is obtai

ntation imp

amber, som

ore, pressur

ed. Experim

imensionles

a Ø9 mm o

ad all over t

tions in roc

s total burn

re 63. Sam

en with the

xperimental

tation Resu

st Results

ventilation

ined. Unfor

act upon ex

me of the

res above t

mental resu

ss MATLAB

of ventilatio

the explosiv

cket motors

area of the

mple MATL

ese limitatio

l data.

ults for Tim

characteris

rtunately, a

xplosion du

test items

this are int

lts are com

B Simulink

on hole is gi

ve surface a

s. Because

explosive.

LAB Simuli

66

ons, predicte

me to React

stic of the

as the test

ue to instan

above 0.7

terpreted to

mpared with

k code. Sam

iven in Figu

as there are

of that, ini

ink Output

ed ignition

tion Analys

system, pr

item is de

ntaneous pre

MPa are o

o be much

h the nume

mple output

ure 63. Bur

no inhibito

itial burn a

t, Test Item

time can b

ses and Com

essure read

esigned to

essure built

observed to

higher tha

erical result

t data of an

rn reaction i

or line is us

area in calc

m with Ø9m

be accepted

mparison

dings from

have low

t up within

o be torn.

an what is

ts obtained

n test item

is assumed

sed like the

culations is

mm Vent

As there

given as

(Avent/Abur

Figur

Looking a

deflagratio

having exp

accepted t

smallest A

between th

unstable in

Numerical

than 0.00

numerical

is predicta

this is acc

especially

are differen

the ratio

rn).

re 64. Com

at the graph

on to deton

perienced th

that this is

Avent/Aburn a

he safe and

n an engine

l calculatio

115 will ca

results pred

able. Burn a

ceptable for

y for low bu

nt test item

of ventilat

mparison of

of pressure

ation transi

hat the test

the limit fo

are assume

DDT limit

ering point

ons of diffe

ause chamb

dict using a

area is take

r a critical

urn rates. T

67

ms having d

tion area o

Experimen

Press

e change wi

ition (DDT)

items are t

or DDT. On

d to be the

ts where the

of view.

erent scenar

ber pressure

a slightly sm

en to be the

design met

The reason

7

different am

over total

ntal and Nu

sure

ith ventilati

) will occur

torn above A

n the other

e limit of

ere is no tes

rios predict

e to diverge

maller ventil

e whole exp

thodology,

that numer

mount of ex

explosive a

umerical R

on area, it i

r or where t

Avent/Aburn o

hand, test

safe region

sts made can

having an

e and lead

lation area,

plosive area

the real ca

rical calcula

xplosive, res

area in Fi

Results of P

is hard to sa

the safe sid

of 0.00104

items that h

n. Unknown

an be expect

n Avent/Aburn

to DDT. A

reasons beh

a initially. A

ase might n

ations show

sults are

gure 64

eak

ay where

e is. But

we have

have the

n region

ted to be

n smaller

Although

hind this

Although

ot be so

w higher

68

peak pressures at higher Avent/Aburn (> 0.00125) is probably because of extrapolated

burn data between 0-2 MPa and burn surface area approximation.

69

CHAPTER 6

6. CONCLUSION AND FUTURE WORK

6.1. Conclusion

Total of 8 tests are examined in this study utilizing thermal modeling of liquid fuel

fire. They are performed in order to observe ignition time of generic test items. From

the first test, an improved temperature boundary condition is obtained. With the more

realistic boundary condition, 4 different analyses are conducted for different test

items with 2-D, axisymmetric approach. This is done since the 3-D model where the

buoyancy terms are included does not change the predicted time by more than 1%.

Predicted ignition times and test results shows a maximum difference of 8.9%.

Possible reasons of this error are discussed. An improvement on the methodology

can be made upon implementing a melting and/or decomposition model into the

ANSYS Fluent. By considering the current results methodology, using a 2-D

axisymmetric thermal analysis for predicting the time to reaction seems to be

applicable.

Ventilation characteristic of a system that contains an explosive of type PBXN-109 is

also studied. For this purpose, burn characteristics under relatively low pressures (0-

10 MPa) are obtained using a strand burner test setup. Next, this data is used within

the self-developed dimensionless code to calculate the critical ventilation area and

compared with the experimental data. According to the comparison made, results

show some discrepancies at 10.9%. Possible reasons for the prediction of critical

ventilation area are also discussed in the previous section.

By this work, it will be possible for one to predict the ignition time of a munition

containing an energetic material by knowing the heat generation characteristic of that

energetic material by using the ANSYS Fluent or any other software that is eligible

to model the heat transfer. By doing so, it will be also possible to study critical

70

thermal path of the energetic system and take necessary precautions to avoid any

violent reactions.

Another benefit of this work is that, it will guide a designer to design ventilation

mechanism that will allow venting the system safely where the explosive will not

lead to deflagration to detonation transition.

6.2. Future Work

A methodology for a design phase of an insensitive munition is discussed in this

study. Although the methodology is valid for all solid type energetic materials, the

current study only covers an explosive named PBXN-109. Further studies can be

made for other type of explosives.

To improve the accuracy of the time to reaction analyses, melting and decomposition

of the liner and explosive can be implemented by complex user defined functions and

other tools. However, before proceeding with modeling studies detailed physics

behind this should be understood and several tests should be made.

Despite a methodology to understand and improve the way energetic material acts

under thermal stimuli, a coupled thermal, chemical and mechanical solver can be

developed. However, it is obvious that this will require a huge amount of effort.

71

REFERENCES

[1] Proud G. W., Ignition and Detonation in Energetic Materials: An

Introduction, STO-EN-AVT-214 Lecture Series September 2013.

[2] STANAG 4439 JAIS (Edition 3) – Policy for Introduction and Assessment of

Insensitive Munitions, 17 March 2010.

[3] Banton, R., Model Development for Insensitive Munitions (IM) Simulations,

U.S. Army Research Laboratory, 2006.

[4] NATO Allied Ordnance Publication (AOP) 39, Edition 3, Guidance on the

Assessment and Development of Insensitive Munition, March 2010.

[5] Zukas, A.Z., Walters, P.W., Ed., Explosive Effects and Applications,

Springer-Verlag New York Inc., New York, USA, 1998.

[6] Lee, P.R., Course Notes on Initiation and Mechanical Properties of

Energetic Materials and Their Influence on Insensitive Munitions (IM)

Design, Computational Mechanics Associates, Baltimore, USA, 2005.

[7] Desailly, D., Guengant, Y., Numerical Simulation of Reaction Violence to

Cook-off Experiments, NDIA Insensitive Munitions and Energetic Materials

Symposium, Orlando, USA, 2003.

[8] Roger L. S., Lecture Notes on AVT-214, Approach to IM Policy – Defining

the Need, STO-EN-AVT-214 Lecture Series September 2013.

[9] Hayles, D.R., How Safe Should Our Weapons Be? Maxwell Air Force Base,

Alabama, USA, 2005.

72

[10] White, A., Parker, R.P., Cost-Benefit Analysis Concept for Insensitive

Munitions Policy Implementation, DSTO Aeronautical and Maritime

Research Laboratory, Australia, 1999.

[11] Doherty, R., Watt, D., Insensitive Munitions-Coming of Age, International

Annual Conference – ICT, Fraunhofer Institut Für Chemische Technologie,

Germany, 1999.

[12] Isler, J., The Transition to Insensitive Munitions, Propellants, Explosives,

Pyrotechnics, Vol.23, p.283-291, 1998.

[13] Victor A.C., Simple Calculation Methods for Munitions Cook-Off Times and

Temperatures, Propellants, Explosives, Pyrotechnics, Vol. 20, pp. 252-259,

1995.

[14] Aydemir, E., Ulas, A., Serin, N., Reacon-1D: Thermal Analysis Code for

Energetic Materials Using Finite Element Method, New Trends in Research

of Energetic Materials, Pardubice, Czech Republic, 2007.

[15] Semenov, N.N., Z. Phys., Vol. 48, p. 571, 1928.

[16] Frank-Kamenetskii, D.A., Acta Physiochim, URSS, Vol.10, p. 365, 1939.

[17] Zinn, J., Mader, C.L. Thermal Initiation of Explosives, J. Appl. Phys., Vol.

31(2), p. 323, 1960.

[18] Merzhanov, A.G., Abramov, V.G., Propellants, Explosives and

Pyrotechnics, Vol. 6, p. 130, 1981.

[19] Sućeska, M., A Computer Program Based on Finite Difference Method for

Studying Thermal Initiation of Explosives, J. Thermal Analysis and

Calorimetry, Vol. 68, p 865-875, 2002.

73

[20] Anderson, C.A., TEPLO – A Heat Conduction Code for Studying Thermal

Explosion in Laminar Composites, Los Alamos Scientific Laboratory, LA-

4511, Los Alamos, USA, 1970.

[21] Isler, J., Kayser, D., Correlation between Kinetic Properties and Self-

Ignition of Nitrocellulose, 6th Symp. Chem. Probl. Connected Stab. Explos,

Kungalav, Sweden, 1982.

[22] McGuire, R.R., Tarver, C.M., Chemical Decomposition Models for Thermal

Explosion of Confined HMX, RDX, and TNT Explosives, Lawrence

Livermore Laboratory, UCRL-84986, Livermore, USA, 1981.

[23] Pakulak J. M., Simple Techniques for Predicting Sympathetic Detonation

and Fast and Slow Cookoff Reaction of Munitions, Technical Report, Naval

Weapons Center, June 1988.

[24] Nichols, A.L., Anderson, A., Neely, R., Wallin, B, A Model For High

Explosive Cookoff, 12th International Detonation Symposium, 2002.

[25] McClelland, M.A., Maienschein, J. L., Nichols, A.L., Wardell, J.F., Atwood,

A.I., Curran, P.O., ALE3D Model Predications and Materials

Characterization for The Cookoff Response of PBXN-109, 20th Propulsion

Systems Hazards Subcommittee, 2002.

[26] Kenneth J., Graham, Aerojet, Culpeper VA, Mitigation of Fuel Fire Threat

to Large Rocket Motors by Venting, 2010 Insensitive Munitions & Energetic

Materials Technology Symposium, Munich, Germany, 11-14 October 2010.

[27] Victor, A.C., Cookoff Protocol for Munition Assessment, Victor

Technology, Ridgecrest, California, USA, 1994.

[28] Swierk T., Fuel Fire Experts Working Group IV Meeting Summary,

September 2014.

74

[29] Yagla J., D. Griffiths, D. Hubble, K. Ford, E. Washburn, Experimental

Development of Propane Burners for Fast Cook Off Testing, Insensitive

Munition and Energetic Material Technology Symposium, October 2013.

[30] Pershing D. W., Center for the Simulation of Accidental Fire and Explosion

Annual Report, September 1999.

[31] Luketa A, Assessment of Simulation Predictions of Hydrocarbon Pool Fire

Tests, Fire and Aerosol Sciences Department Sandia National Laboratories,

2010.

[32] Rawat R., Pitsch H., and J. F. Ripoll, Large-Eddy Simulation of Pool Fires

with Detailed Chemistry Using an Unsteady Flamelet Model, Center of

Turbulence Research Proceeding of the Summer Program, 2002.

[33] Echeverria F., Bernardez P., Gonzalo R., Use of CFD Modelling in The

Design of a IM Warhead to Cook-Off Stimuli, Insensitive Munition and

Energetic Material Technology Symposium, May 2015.

[34] Ford K.P., Davis N.C., Farmer A.D., Washburn E.B, Atwood A.I., Wilson

K.J., Abshire J.P., M.L. Shewmaker, Z.P. Goedert, C.J. Wheeler, P.O.

Curran, J. Covino, Subscale Fast Cookoff Test Result, Insensitive Munition

and Energetic Material Technology Symposium, 2010.

[35] http://www.aerojet.engr.ucdavis.edu/fluenthelp/html/ug/node998.htm,

FLUENT 6.3 User’s Guide, last access date: June 2015

[36] Anderson B. W. The Analysis and Design of Pneumatic Systems, Wiley,

NY, 1967, p. 19.

[37] http://www.grc.nasa.gov/WWW/CEAWeb, last access date: June 2015.

75

[38] Maienschein J.L., Wardell J.F., Weese R.K., Cunningham B.J., Tran T.D,

Understanding and Predicting the Thermal Explosion Violence of HMX-

Based and RDX-Based Explosives – Experimental Measurements of Material

Properties and Reaction Violence, 12th International Congerence

Symposium, San Diego, California, August, 2002.

[39] Ulas A., Y.-C. Lu & Kenneth K. Kuo (2003) Ignition and Combustion

Characteristics of RDX-Based Pseudopropellants, Combustion Science and

Technology, 175:4, 695-720, DOI: 10.1080/00102200302391.

[40] Maienschein J.L., Wardell J.F., Deflagration Behavior of PBXN-109 and

Composition B at High Pressures and Temperatures, Lawrance Livermore

National Laboratory, March 11, 2002.

[41] Curdaneli S., Experimental Analysis on the Measurement of Ballistic

Properties of Solid Propellants, M.Sc. thesis in Mechanical Engineering

submitted to Middle East Technical University, Turkey, May 2007.

[42] NATO Standard 4240 (Edition 2), NATO Liquid Fuel / External Fire,

Munition Test Procedures, April 2003.

[43] Shannon K.S., Butler B.W., A Review of Error Associated with

Thermocouple Temperature Measurement in Fire Environments, Second

International Wildland Fire Ecology and Fire Management Congress and

Fifth Symposium On Fire And Forest Meteorology, November 2003

[44] Jones J.C., On the Use of Metal Sheathed Thermocouples in a Hot Gas

Layer Originating From a Room Fire. Journal of Fire Science, 13, 257-260.,

1995

76

77

APPENDIX A

MATLAB FUCNTIONS Input.m

clear, clc, close all

Pi=0; % Pa, gauge pressure

ro=1670; % kg/m3

ri=0.039; % m, initial radius of the explosive

Li=0.125; % m, initial length of the explosive

Vi=9.3669E-005; % m3

T_chamber=2552.84; % K

R_chamber=339.28; % Pa.m3/kg/.K

k_chamber=1.2153;

Cd=0.82;

Explosive Geometry Calculation Function

function Ab = Expo_Geo(ri,Li,d)

r=ri-d; % Remaining radius of the explosive

L=Li-2*d; % Remaining length of the explosive

if r<=0 || L<=0

r=0;

L=0;

end

Ab=(2*pi*(r^2)+2*pi*r*L); % Explosive burn area, based on the assumption that

burning takes places on all explosive surfaces

Burn Rate Calculation Function

function rb = Burn_rate(p)

n1=0.93; % m/s - pascal

a1=1.223E-9; % m/s - pascal - Ref: SAGE curve fit, 0-10 MPa

n2=1.3196; % m/s - pascal

78

a2=2.90626E-12; % m/s - pascal - Ref: Livermore Lab., 0-10 MPa

if p<=10^7

rb=a1*p^n1;

else

rb=a2*p^n2;

end