Experimental and Numerical Studies of
Blast, Fragmentation and Thermal
Effects Mitigation of Energetic
Materials Detonation
By
Khurshid Ahmed
School of Chemical and Materials Engineering
National University of Sciences and Technology
2021
Experimental and Numerical Studies of
Blast, Fragmentation and Thermal
Effects Mitigation of Energetic
Materials Detonation
Name: Khurshid Ahmed
Reg. No: 00000202572
This thesis is submitted as a partial fulfillment of the requirements
for the degree of
Doctor of Philosophy in Energetic Materials Engineering
Supervisor: Dr. Abdul Qadeer Malik
School of Chemical and Materials Engineering (SCME)
National University of Sciences and Technology (NUST)
H-12 Islamabad, Pakistan
June, 2021
List of Publications
1. Ahmed K, Malik AQ, Ahmad IR. Heterogeneous lightweight configuration
for protection against 7.62 × 39 mm bullet impact. International Journal of
Protective Structures. 2019;10(3):289-305.
https://doi.org/10.1177/2041419619839216
2. Ahmed K, Malik AQ. Experimental studies on blast mitigation capabilities of
conventional dry aqueous foam. AIP Advances. 2020;10(6):065130.
https://doi.org/10.1063/5.0010283
3. Ahmed K, Malik AQ, Hussain A, Ahmad IR, Ahmad I. Lightweight
protective configurations against blast and fragments impact: Experimental
and numerical studies. AIP Advances.2020;10(9):095221.
https://doi.org/10.1063/5.0022982
4. Ahmed, K., Malik, A. Q., Hussain, A., Ahmad, I. R. and Ahmad, I. (2021)
Blast and Fragmentation Studies of a Scaled Down Artillery Shell-Simulation
and Experimental Approaches, The International Journal of Multiphysics,
15(1), pp. 49-71. https://doi.org/10.21152/1750-9548.15.1.49
Under Review
1. Ahmed K, Malik AQ. Experimental investigations of the response of a
heterogeneous container to blast, fragmentation and thermal effects of
energetic materials detonation, International Journal of Protective Structures.
2. Ahmed K, Malik AQ. Protective Container for Combined Blast,
Fragmentation and Thermal Effects of Energetic Materials Detonation,
Central European Journal of Energetic Materials (CEJEM).
International Conference
Ahmed K, Malik AQ, Ahmad IR. Heterogeneous lightweight configuration
for protection against 7.62 × 39 mm bullet impact, 5th
International
Conference on Protective Structures (ICPS5), Poznan, Poland, 19-23 August
2018
i
DEDICATED TO MY
Dearest Family
ii
Acknowledgments
All Commendations to Almighty Allah, The Most Beneficent, The Most Merciful,
Who blessed me with intellect and determination to accomplish this task. I would
like to express my deepest and sincere gratitude to my supervisor Dr. Abdul Qadeer
Malik for his valuable guidance and assistance throughout this research work. I can
never forget his keen interest, encouragements and unprecedented support during this
period. My special thanks to my co-supervisor Dr. Iram Raza Ahmad for his
continued technical and moral support to accomplish this work. My thanks are due to
GEC members, Prof. Dr. Arshad Hussain and Dr. Iftikhar Ahmad, for their guidance
and support. The support and cooperation of Dr Ahsan, Dr Sarah, Dr Tayyaba, Ms
Javeria and Ms Ayesha during this period is acknowledged and highly appreciated.
My humble gratitude to Prof. Muhamed Suceska for sharing EXPLO5 calculations
and his unconditional support during these years.
I would like to express my special thanks to Member Technical, PAEC, Mr. Aslam
Hayat and Dr. A. Ghani Akram for providing me this opportunity and their
everlasting support. My thanks to PAEC for meeting the financial requirements and
generous support during experimental work. I would also like to express my thanks
to my friends and colleagues Mr Muhammad Mursaleen, Ahmad Hashmi,
Muhammad Tariq, Mohsan Hassan, Dr. Sultan, Zahoor Sultan, M. Arshad, Khalid
Naeem, Kamran Arif, Fazle Hadi, Sameen Murtaza, Dr. Naveeda, Dr. Javaid, Dr.
Kamran, Kazim Hussain, M. Ashraf, Ajmal Hussain, Nusrat, Kamran and fellow
team members for helping me throughout these years and especially their support
during the experimental work. I also express my thanks to Atif Khurshid for his
assistance in review and computation work. My thanks are also due to Dr Shakeel
Abbas, Mr. Abdul Waheed, Mr Sammiullah and the experimental group for their
generous support in experimental work. I want to thank all the people who helped
and supported me during my PhD.
I would like to express my obedience to my parents and my family whose prayers
have led me to this stage and will accompany me for all future endeavors. It would
not have been easy to complete this work without their sacrifices, cooperation and
prayers.
iii
Abstract
The detonation of an energetic material (EM) is mainly manifested in the form of
blast, fragmentation and thermal effects. These effects are very destructive and cause
injuries-being fatal-and structural damage as well. The suppression or attenuation of
these effects is a prime focus. The present research is related to the concerted
investigations employing lightweight materials capable of mitigating the blast,
fragmentation and thermal effects of explosive devices including lighter improvised
explosive devices (IEDs). Commercially available shaving foam was characterized
and investigated as a potential mitigating material in combination with Kevlar woven
fabric, laminated glass fiber reinforced polymer (GFRP), Bakelite, Polyurethane
(PU)/expanded Polystyrene (EPS) foams and PU-silica to withstand the impact of
blast wave and explosively driven high velocity fragments.
Various amounts of C4 explosive (82, 104, 250 and 800 grams) were tested in air and
immersed in shaving foam. The shaving foam confinement suppressed the fireball
radius by 80% and quenched the afterburning reactions resulting from an EM
detonation. About 70% reduction in blast overpressure and 62% reduction in positive
impulse were observed for shaving foam confinements weighing 1.0 - 2.05 kg
against C4 explosives of 82 - 250 grams.
Lightweight protective configurations comprising different combinations of Kevlar
woven fabric, laminated GFRP, PU/EPS foams and alumina (Al2O3) tile were tested
against blast, fragments and bullet impact. Multi-layer composition of PU-silica and
a mixture of PU-silica and alumina powder were also studied. The protective
configurations were tested under static detonation of geometrically scaled down 155
mm artillery shell. Fragments weighing up to 4.3 grams with velocities in the range
of 961–1555 m/s were produced and impacted the configurations. The Kevlar woven
fabric, laminated GFRP and PU foam compositions provided significant absorption
and attenuation to impacting fragments. Configurations employing alumina tile were
able to resist perforation of 7.62 mm mild steel core (MSC) bullet and also withstood
the blast and multiple fragments impact without significant backface signatures
(blunt force trauma).
Numerical simulations were performed using ANSYS AUTODYN. SPH (Smoothed
Particle Hydrodynamics) solver was used for characterization of shell fragmentation.
iv
Coupled SPH -ALE (Arbitrary Lagrangian-Eulerian) approach was used to simulate
the interaction of fragments with protective configurations. A coupled Euler-ALE
approach was employed to simulate blast wave propagation in air and loading on
protective configurations. The fragments mass, initial velocity and spatial
distributions were in good agreement with the experimental findings. The blast wave
parameters showed good match of the arrival time and peak pressure values with
measured data, however, a discrepancy in incident impulse was observed.
On the basis of experimental and simulation studies a model heterogeneous
containment system was developed to counter combined blast, fragmentation and
thermal effects of energetic material detonation of 1.0 kg bare and 0.6 kg of steel
cased TNT equivalent charge. The two layers container provided 97% overpressure
reduction as well as contained the high velocity fragments. The novel combination of
EPS foam, Bakelite and PU-silica layers provided protection against in contact C4
detonation at the base of the container. The upshot of this research work is that,
besides being of academic significance, it provides ample data to design a device to
combat terrorism against lighter time bomb/IEDs placed at public places, high profile
meeting venues and transportation systems (land, air etc.).
v
Table of Contents
Acknowledgments ........................................................................................................ ii
Abstract ....................................................................................................................... iii
Table of Contents .......................................................................................................... v
List of Figures .............................................................................................................. ix
List of Tables ............................................................................................................ xvii
List of Acronyms and Symbols .............................................................................. xviii
Introduction........................................................................................... 1 Chapter 1
1.1 Motivation ...................................................................................................... 1
1.2 Energetic Materials ........................................................................................ 2
1.2.1 Low Explosives ...................................................................................... 2
1.2.2 High Explosives ...................................................................................... 2
1.3 The Rankine-Hugoniot Relations................................................................... 4
1.4 Equation of State ............................................................................................ 5
1.5 Formation of Blast Wave ............................................................................... 6
1.6 Scaling Laws .................................................................................................. 8
1.7 Effects of Cased Explosive Detonation ......................................................... 9
1.8 Literature Review ......................................................................................... 11
1.9 Gaps in Literature......................................................................................... 15
1.10 Objectives ................................................................................................. 16
1.11 Thesis Organization .................................................................................. 16
References ................................................................................................................... 17
Materials and Methodology for Blast Mitigation ............................ 21 Chapter 2
2.1 Introduction .................................................................................................. 21
2.2 Aqueous Foam ............................................................................................. 21
vi
2.2.1 DENIM Shaving Foam ......................................................................... 22
2.2.2 Microscopic and Viscosity Study ......................................................... 22
2.2.3 Sound Speed Measurement................................................................... 24
2.3 C4 Explosive ................................................................................................ 26
2.4 Diagnostic Tools .......................................................................................... 27
2.4.1 High-Speed Photography ...................................................................... 27
2.4.2 Pressure Transducers ............................................................................ 27
2.4.3 Arrival Time Sensors ............................................................................ 29
2.5 Experimental Work ...................................................................................... 30
2.5.1 Blast Mitigation with Shaving Foam .................................................... 30
2.5.2 Effects of Foam Volume and Liquid Content on Mitigation ................ 35
2.5.3 Blast Parameters at Z < 1.0 m/kg1/3
...................................................... 36
2.6 Numerical Modeling and Simulation ........................................................... 39
2.6.1 Material Modeling ................................................................................ 39
2.6.2 Blast Wave Parameters ......................................................................... 40
2.7 Results and Discussion................................................................................. 42
2.7.1 Fireball and Afterburning Suppression ................................................. 44
2.7.2 Peak Pressure and Impulse Reduction .................................................. 46
References ................................................................................................................... 50
Characterization of Shell Fragmentation ......................................... 52 Chapter 3
3.1 Introduction .................................................................................................. 52
3.1.1 Effects of Fragmentation ...................................................................... 53
3.2 Characterization ........................................................................................... 53
3.2.1 Experimental Work ............................................................................... 54
3.3 Numerical Simulation .................................................................................. 58
3.3.1 Material Modeling ................................................................................ 59
3.3.2 Fragmentation of Shell ......................................................................... 59
3.3.3 Blast Parameters for Shell Detonation .................................................. 62
vii
3.4 Experimental Results ................................................................................... 64
References ................................................................................................................... 71
Protective Configurations against Fragments .................................. 73 Chapter 4
4.1 Protective Mechanisms ................................................................................ 73
4.1.1 Momentum Disruption.......................................................................... 73
4.1.2 Shock Wave Spreading ......................................................................... 74
4.1.3 Shock Energy Absorption ..................................................................... 75
4.2 Protection against Fragments Impact ........................................................... 75
4.2.1 Protective Configurations ..................................................................... 76
4.2.2 Experimental Work ............................................................................... 77
4.3 Protection against Bullet Impact .................................................................. 84
4.3.1 Blunt Force Trauma Test ...................................................................... 85
4.4 Numerical Simulation .................................................................................. 87
4.4.1 Material Modeling ................................................................................ 87
4.4.2 Shell Fragmentation and Impact on Protective Configurations ............ 89
4.4.3 Coupled SPH-ALE Simulation ............................................................. 90
4.4.4 Blast Loading on Protective Configurations ......................................... 98
4.5 Experimental Results ................................................................................. 101
References ................................................................................................................. 107
Containment for Blast, Fragmentation and Thermal Effects .... 109 Chapter 5
5.1 Introduction ................................................................................................ 109
5.2 Materials and Experimental Work ............................................................. 109
5.2.1 Scaled down Container ....................................................................... 109
5.3 Numerical Simulation ................................................................................ 111
5.3.1 C4 Surface Burst Parameters .............................................................. 111
5.3.2 Fragments Impact on Scaled down Container .................................... 114
5.4 Experimental Results ................................................................................. 117
5.4.1 Scaled down Container Test ............................................................... 117
viii
5.4.2 Container Test with 1.0 kg TNT Equivalent Charge .......................... 122
5.4.3 Container Test with Steel Cased Charge (Pipe-bomb) ....................... 125
References ................................................................................................................. 132
Conclusions and Recommendations ................................................ 133 Chapter 6
6.1 Conclusions ................................................................................................ 133
6.2 Recommendations ...................................................................................... 135
ix
List of Figures
Figure 1-1 : Fatalities in terrorist violence in Pakistan, (2000-2018) [3] ..................... 1
Figure 1-2 : The detonation wave structure and pressure profile [10] ......................... 3
Figure 1-3 :Shock wave generation in a compressible fluid [12] ................................. 4
Figure 1-4 : Characteristics of a blast wave (a) ambient pressure (b) positive phase
(c) negative phase [13] .............................................................................. 6
Figure 1-5 : Devastating effects of energetic material detonation .............................. 11
Figure 1-6 : Blast inhibitor: 1-Elastic envelope, 2-liquid gas medium, 3-working
space and 4-HE [37] ................................................................................ 13
Figure 1-7: (a) Episafe container and (b) Resnyansky and Delany setup ................... 13
Figure 1-8 : Blastgard trash receptacle [63] ............................................................... 14
Figure 1-9 : TM International blast containment vessel [64] ..................................... 14
Figure 1-10 : Blast Containment Receptacle (BCRs) [65] ......................................... 15
Figure 2-1 : Denim foam-average bubble size (a) 15 µm at 0 hrs (b) 85 µm at 1 hrs 22
Figure 2-2: Denim foam-average bubble size (c) 120 µm at 2 hrs (d) at 24 hrs ........ 23
Figure 2-3: Gillette foam-average bubble size (a) 20 µm at 0 hrs (b) 105 µm at 1 hrs
(c) 155 µm at 2 hrs (d) at 24 hrs .............................................................. 23
Figure 2-4: The graphical representation of bubbles coarsening of conventional
aqueous foams (a) Denim (b) Gillette ..................................................... 24
Figure 2-5: Shear rate vs viscosity plot for shaving foam .......................................... 24
Figure 2-6: Setup for measuring sound speed in air ................................................... 25
Figure 2-7: Setup for measuring sound speed in shaving foam.................................. 25
Figure 2-8: Waveforms (a) in air (b) in shaving foam ................................................ 26
Figure 2-9: (a) Transducers orientation for Pr and Ps measurement (b) Components of
data acquisition (DAQ) system ............................................................... 28
Figure 2-10: (a) Arrival time sensor (b) section view (c) mounted in the fixture ...... 29
Figure 2-11: Comparison of Time of arrival with pressure transducer and
arrival time probe .................................................................................... 29
Figure 2-12: 2D view of Experimental Layout for blast parameters measurement ... 30
Figure 2-13: Test setup for measuring blast parameters for (a) bare charge (b)
immersed in shaving foam(c) Transducer orientation for measuring
reflected pressure .................................................................................... 31
Figure 2-14: High-speed images, 82g bare C4 detonation in air and propagation of
fireball ..................................................................................................... 31
x
Figure 2-15: A sequence of events after detonation of charge immersed in shaving
foam ....................................................................................................... 32
Figure 2-16: (a) Reflected overpressure (b) Impulse plots for 82g C4 blast at 0.8 m
for bare charge and covered in shaving foam ........................................ 32
Figure 2-17: Test setup for measuring 82g C4 blast parameters for (a) bare charge (b)
covered in shaving foam at 0.85 m ........................................................ 33
Figure 2-18: (a) Incident overpressure and (b) Impulse profiles for 82g C4 bare and
covered in shaving foam at 0.85 m ........................................................ 33
Figure 2-19: (a) Pressure and (b) Impulse plots for bare 250g C4 blast at 1.0 m and
submerged in shaving foam ................................................................... 34
Figure 2-20: Test setup for measuring 250g C4 blast parameters for bare charge and
covered in 0.05 m3 shaving foam .......................................................... 35
Figure 2-21: Peak incident pressure for 250g Bare C4 and covered in shaving foam at
0.8m ....................................................................................................... 36
Figure 2-22: (a) Perspex channel filled with foam (b) Testing setup within shaving
foam ....................................................................................................... 36
Figure 2-23: Experimental setup for measuring blast parameters for Z < 1 (m/kg1/3
)
(a) sensors fixed inside empty Perspex channel (b & c) shaving foam
filled inside channel ............................................................................... 37
Figure 2-24: High-speed images of 200g C4 detonation inside Perspex channel (a)
just after detonation (b) fireball expansion (c) product gasses expansion
............................................................................................................... 37
Figure 2-25 : Pressure and scaled distance plots for HE detonation in shaving foam
for Z<1 ................................................................................................... 38
Figure 2-26: (a) Pressure and (b) Impulse plots for 200g C4 charge detonated inside
shaving foam at 0.25 and 0.30m from charge face ................................ 38
Figure 2-27: (a) ANSYS AUTODYN model (b) expansion of detonation product
gases after detonation of 250g C4 ......................................................... 41
Figure 2-28: (a) Formation of blast wave and (b) propagation in air ......................... 41
Figure 2-29: Experimental and simulation results for 82g C4 bare blast, (a) Incident
overpressure (b) Impulse at 0.5m and 0.7m from charge center ......... 42
Figure 2-30: Experimental and simulation results for 82g C4 bare blast, incident
overpressure at 0.65 m and 0.85 m ........................................................ 42
xi
Figure 2-31: Fireball for bare 82 g C4 (a, b) and formation of secondary fireball (c, d,
e, f) ......................................................................................................... 45
Figure 2-32: (a) Fireball formation and (b, c) quenching for 82 g C4 covered in
shaving foam .......................................................................................... 45
Figure 2-33: Fireball for 250 g C4 (a, b, c) bare C4 and (d, e, f) covered in shaving
foam ....................................................................................................... 46
Figure 2-34: Incident pressure and Impulse plots for 82g C4 blast at 0.5 m-for bare
charge and submerged in shaving foam ................................................. 47
Figure 2-35: (a) Peak incident pressure and (b) impulse plots for 250 g Bare C4 and
covered in shaving foam at 0.9 m .......................................................... 47
Figure 2-36: Incident and reflected pressure plot (left) and impulse plot (right) for
250g C4 at 0.8m ..................................................................................... 48
Figure 2-37: Peak pressure and distance plots with and without shaving foam for (a)
82 g C4 (b) 250 g C4 ............................................................................. 49
Figure 3-1: (a) scaled down shells (b) its cut-view (c) standard 155mm shell ........... 54
Figure 3-2: Layout for blast and fragmentation tests of scaled down shell ................ 55
Figure 3-3: (a) Flat Brass probe (b) 3-D view(c) section view (d) Two probes setup 56
Figure 3-4: Flat probes arrangement for fragment velocity measurement (a) test-1 (b)
test-2 ...................................................................................................... 56
Figure 3-5: Testing setup with (a) fiberglass (F/G) sheets and timing probes (b)
plywood sheet ........................................................................................ 57
Figure 3-6: Setup for fragments impact and spatial distribution (a) Test-2 and (b)
Test-3 ..................................................................................................... 57
Figure 3-7: Fiberglass witness sheets (a, b) before and (c) after fragments impacts .. 57
Figure 3-8: Plywood witness sheets (a) before and (b) after the fragments impact ... 58
Figure 3-9: Fragments recovered in the tests .............................................................. 58
Figure 3-10: SPH Model of scaled down shell with gauge points ............................. 60
Figure 3-11: Fragmentation process (a) at 27 s (b) at 48 s (c) venting of product
gases ....................................................................................................... 60
Figure 3-12: Fragmentation process and radial expansion with time (a) at 57 s (b) at
80 s and (c) 150s (d) at 200 s (e) at 250 s ..................................... 61
Figure 3-13: Number of fragments and mass distribution .......................................... 61
Figure 3-14: Fragment velocities of gauge points defined on shell casing (a) with
ALE solver (b) with SPH ....................................................................... 62
xii
Figure 3-15: Number of fragments and their velocity distribution............................. 62
Figure 3-16: (a) AUTODYN model of shell (b) detonation wave propagation inside
shell (c) Expansion of shell at 20 s ...................................................... 63
Figure 3-17: (a) Venting of pressurized gases in air at t= 40 s and expansion (b) at
t= 61 s (c) at 90 s ............................................................................... 63
Figure 3-18: (a) Blast wave propagation in air at t=0.150ms, (b) at t= 0.604ms and (c)
at t= 0.88ms ........................................................................................... 63
Figure 3-19: Simulated and experimental (a) peak Pressure and (b) Impulse plots for
scaled down shell ................................................................................... 64
Figure 3-20: (a, b) Fragment's velocity measurement from different parts of the shell
(c) timing probe after fragment impact .................................................. 65
Figure 3-21: : (a) Two probes setup (b) Fragment’s impact on timing probe and (c)
arrival time for velocity calculations ..................................................... 65
Figure 3-22: Different positions of shell for fragment velocity calculations ............. 66
Figure 3-23: Fragment impacts and perforation through Fiberglass witness sheets . 67
Figure 3-24: (a, b) Plywood witness sheets (c) Timing probe placed below the shell
base ........................................................................................................ 67
Figure 3-25: Number of fragments and their mass distribution for scaled shell ........ 68
Figure 3-26: : Experimental and simulated peak incident pressure plots at (a) 0.55m
& 0.65m (b) 0.59m & 0.675m ............................................................... 69
Figure 4-1: (a) 7.62 mm bullet impact. (b) Pulverization of alumina disk. (c)
Fragmentation and scattering of alumina .............................................. 74
Figure 4-2 : Alumina tiles used in present work ......................................................... 74
Figure 4-3 : Stress-strain relation of porous material [6] ........................................... 75
Figure 4-4: Fragments penetration and perforation (a) Front side (b) back sides of
the 6 mm thick steel plate ...................................................................... 76
Figure 4-5: Experimental setup for testing protective configurations against
fragments impact ................................................................................... 77
Figure 4-6: (a) Front side (b) back side of C-1 ........................................................... 78
Figure 4-7: (a) Setup for testing C-1 and C-3. (b) Front (c) back sides of C-1 after
fragment’s impact .................................................................................. 78
Figure 4-8: (a, b) Test-2 setup and C-2 view after (c) impacted and (d) perforated
fragments ............................................................................................... 79
Figure 4-9: Assembled view of C-2, (a) front and (b) back sides .............................. 79
xiii
Figure 4-10: (a, b) C-2 testing setup (c) C-2 along with 6 mm thick MS plate .......... 79
Figure 4-11: C-2 (a) front (b) back sides (c) laminated GFRP (d) PU foam after
fragments impact ................................................................................... 80
Figure 4-12: (a) Front and (b) back sides of C-3 ........................................................ 80
Figure 4-13: C-3 front and back sides after fragments impact- first test .................... 80
Figure 4-14: C-3 Second test (a) Front (b) back sides, (c, d) laminated GFRP.......... 81
Figure 4-15: (a, b) Assembled view of C-4 and (c) testing setup for C-4 and C-5 .... 81
Figure 4-16: (a, b) C-4 front and (c, d) back sides after the test ................................. 81
Figure 4-17: C-5 (a) Front side (b) back side ............................................................. 82
Figure 4-18: (a) Front and (b) back sides of C-5 after fragment's impact .................. 82
Figure 4-19: (a, b) A view of C-6 and (c, d) C-7 configurations ................................ 83
Figure 4-20: Testing setup for C-6 and C-7 ................................................................ 83
Figure 4-21: (a, b) front and back sides of C-6 after test, (c, d) front and back sides
of C-7 after test ...................................................................................... 83
Figure 4-22 : Different parts of 7.62x39 mm bullet, (left) steel core, lead filler, and
copper coated steel jacket. (Right) cut sections of bullet [8] ................. 84
Figure 4-23: High speed images of bullet impacting the target configuration ........... 85
Figure 4-24: 7.62 x 39 mm MSC original and plastically deformed recovered bullet
............................................................................................................... 85
Figure 4-25: The testing configuration, including Plastilina clay as backing medium
............................................................................................................... 86
Figure 4-26: High speed photographic images of blunt force trauma test at three
different times ........................................................................................ 86
Figure 4-27: (a) Testing setup (b) 7.62mm bullet impact (c), Rear side of C-5 (d) and
impact on Plastilina clay ........................................................................ 86
Figure 4-28: (a, b) FE model with gauge points defined and Fragment velocities of
gauge points defined on shell casing (c) with ALE solver (d) with SPH
............................................................................................................... 89
Figure 4-29: FE models of (a, b) C-1 and (c, d) C-5 .................................................. 90
Figure 4-30: Isometric view of shell fragmentation with C-1 and C-5, fragments
impacting C-1 & C-5 ............................................................................. 91
Figure 4-31: (a) Fragments penetration through C-1 (b) deformation of Kevlar woven
fabric and (c) Fragments impact on C-5 and bulge on back side of C-5
............................................................................................................... 92
xiv
Figure 4-32: (a) Velocity profile of gauge points before and after impacting C-5 (b)
before and after impacting C-1 .............................................................. 92
Figure 4-33: (a) Fragments impacting on MS plate (b) isometric view ..................... 93
Figure 4-34: Penetration and perforation through (a) MS plate. (b) Close view of MS
plate ........................................................................................................ 93
Figure 4-35: (a) Shell model in SPH (b) radial expansion of fragments and (c)
moving gauges are visible ..................................................................... 94
Figure 4-36: (a) ALE model of C-4 (b, c) Fragments approaching the configurations
............................................................................................................... 95
Figure 4-37: Fragments impact on C-4, (a) front side view, (b, c) back side view .... 95
Figure 4-38: Velocity plot of gauge points (a) before and (b) after impacting C-4 .. 96
Figure 4-39: ALE model of C-2. (a) Grid plot and (b) material plot.......................... 96
Figure 4-40: (a) Isometric view of C-2 and C-4, (b) Fragments perforation through C-
2. (c) Gauge points defined on C-2 ....................................................... 97
Figure 4-41: (a) Fragments penetration through C-2 (b, c) combined view of C-2 and
C-4 ......................................................................................................... 97
Figure 4-42: Velocity profiles of gauge points (a) before and (b) after impacting C-2
............................................................................................................... 98
Figure 4-43: (a) AUTODYN model of shell (b) detonation wave propagation inside
shell (c) Expansion of shell at 20 s ...................................................... 99
Figure 4-44: (a) Expansion of shell at 40 s (b) at 61 s and (c) at t= 90 s ............. 99
Figure 4-45: (a) Blast wave approaching C-2 (b) impacted on C-2 and (c) C-2 before
and (d) after blast impact ....................................................................... 99
Figure 4-46: For gauge point 2 (a) Incident and reflected overpressure (b) Impulse
plots ...................................................................................................... 100
Figure 4-47: (a) Pressure at gauge point 3, (b) Change in Specific internal energy in
PU foam for gauge points 7 and 8 ....................................................... 101
Figure 4-48: Fragments perforation and captured in (a, b) Kevlar and (c, d) GFRP 102
Figure 4-49: (a) C-2, (b) laminated GFRP and (c) PU foam after fragments impact
............................................................................................................. 102
Figure 4-50: (a) Front and (b) back sides of C-3 (c) laminated GFRP after second
test. ....................................................................................................... 103
Figure 4-51: (a) Back side of C-4 (b, c) front and back sides of laminated GFRP
placed behind the ceramic tile ............................................................. 103
xv
Figure 4-52: (a) Front (b) back sides of C-5 after fragment's impact, (c) recovered
fragment & bullet ................................................................................. 104
Figure 4-53: (a, b) Front and back sides of C-6 (c, d) C-7, after fragments impact . 105
Figure 5-1: (a) Empty container (b) C4 placed at the bottom (c) filled with shaving
foam ..................................................................................................... 110
Figure 5-2: A sketch of testing setup for (a) surface burst (b) detonation inside
cylindrical container ............................................................................ 110
Figure 5-3: Experimental setup for bare C4 charge test (a) close view (b) complete
setup ..................................................................................................... 111
Figure 5-4 (a) Experimental setup for C4 detonation inside container (b) C4 and foam
filled in container ................................................................................. 111
Figure 5-5: (a) AUTODYN model (b) product gases expansion ............................. 112
Figure 5-6: (c) Blast wave formation and (d) propagation in air towards gauge points
............................................................................................................. 112
Figure 5-7: Simulated P(t) history of gauge points for 104g bare C4 surface blast 113
Figure 5-8: (a) AUTODYN Model (b) blast wave reflection from base and walls (c)
blast wave loading on container walls ................................................. 114
Figure 5-9: FE model for shell and container ........................................................... 114
Figure 5-10: (a) Shell fragmentation at 63.93 s and (b) velocity plot of the gauge
points.................................................................................................... 115
Figure 5-11: Isometric view of the fragments, radial dispersion and container ....... 115
Figure 5-12: Fragments penetration into steel liner and its deformation .................. 116
Figure 5-13: Fragments velocity after impacting the container walls ...................... 116
Figure 5-14: High speed images of bare C4 (a) detonator fired (b) t= 0.22 ms after
detonation ............................................................................................ 117
Figure 5-15: High speed images (c) t= 0.320 ms and (d) t= 0.725 ms ..................... 117
Figure 5-16: High speed images of bare C4 detonation ........................................... 118
Figure 5-17: high speed images at (a) t=0.075ms (b) t = 0.800 ms .......................... 118
Figure 5-18: High speed images at (c) t=1.075ms and (d) t=1.475ms ..................... 119
Figure 5-19: C4 charge detonation inside container at (e) 1.80ms and (f) t= 2.675ms
............................................................................................................. 119
Figure 5-20: Contact detonation of 30 g C4 (a) test setup (b) post-test view ........... 120
Figure 5-21: Pressure plots for 104g C4 detonated (a) Surface burst (b) inside
container .............................................................................................. 120
xvi
Figure 5-22: Post-test sights of (a & b) PU-sand composite top and side (c) container
............................................................................................................. 121
Figure 5-23: (a) MS cylinder (b) Inner container ..................................................... 122
Figure 5-24: (a, b) PU-silica disc (c) EPS foam and Bakelite sheet ......................... 122
Figure 5-25: (a) GFRP cylinder (b) inner view of composite container (c) outer
composite container ............................................................................. 123
Figure 5-26: (a) C4 placed inside container (b) shaving foam filled around C4 ...... 123
Figure 5-27: Experimental setup for 800 g C4 detonation inside container ............. 124
Figure 5-28: High-speed images of 800g C4 detonation inside container ............... 124
Figure 5-29: Reflected overpressure-time history for 800g C4 detonation inside
container .............................................................................................. 125
Figure 5-30: Steel cased charge (Pipe-bomb) ........................................................... 125
Figure 5-31: (a) SPH- ALE model for steel cased charge and inner container (b) at 15
s of detonation (c) fragmentation at 40 s ......................................... 126
Figure 5-32: (a) Gauge points location at 58s, (b) radial flight at 105 s .............. 126
Figure 5-33: (a) No. of fragments and mass distribution (b) fragments velocity plot
............................................................................................................. 127
Figure 5-34: (a) Shaving foam filled around steel cased charge (b) experimental setup
............................................................................................................. 127
Figure 5-35: High-speed images of steel cased 565g Comp-B detonation inside
container .............................................................................................. 128
Figure 5-36: Fragments impact and perforation (a) inner layer (b) outer layer of
GFRP ................................................................................................... 128
Figure 5-37: (a, b) Fragments perforation through outer layer of Kevlar fabric (c)
bottom MS disc after second test ......................................................... 129
Figure 5-38: Reflected overpressure -time history for steel cased 565 g Comp-B
charge inside container ........................................................................ 130
xvii
List of Tables
Table 2-1 : Blast parameters for bare charges and covered in shaving foam ............. 35
Table 2-2: JWL parameters of C4, EOS data for Balsa wood and air ........................ 40
Table 2-3: Effect of grid size and discretization error in AUTODYN Simulation .... 40
Table 2-4: Comparison of bare charge tests with Kingrey-Bulmash (KB) results ..... 43
Table 2-5: Fireball radius for C4 detonation in air and with shaving foam ............... 46
Table 3-1: Material and dimensions of scaled down and standard 155mm shell ....... 54
Table 3-2: Johnson Cook damage parameters for Steel-1006 [17] ............................ 59
Table 3-3: Comparison of experimental, numerical and analytical fragments
velocities ................................................................................................... 66
Table 3-4: Measured mass and size of the recovered fragments- comparison with
simulation results ...................................................................................... 68
Table 3-5: Summary of the experimental & simulation Pressure-time values ........... 70
Table 4-1: Constituent materials and dimensions of protective configurations ......... 77
Table 4-2: Protective configuration for bullet impact tests [8]................................... 84
Table 4-3: Measured values of 7.62x39 mm MSC bullet ........................................... 84
Table 4-4: Material properties used in simulation ...................................................... 88
Table 4-5: Summary of the experimental and simulated results .............................. 106
Table 5-1: Simulation results for 104g bare C4 surface blast .................................. 113
Table 5-2: Material and dimensions of pipe-bomb for blast and fragmentation study
................................................................................................................ 126
Table 5-3: Comparison of experimental results with Conwep calculations ............. 130
xviii
List of Acronyms and Symbols
EM Energetic Materials
LE Low Explosives
HE High Explosives
IEDs Improvised Explosive Devices
EPS Expanded Polystyrene
PU Polyurethane
GFRP Glass fiber reinforced polymer
psi pounds per square inch
kPa kilopascal
MSC Mild Steel Cover
SPH Smoothed Particle Hydrodynamics
ALE Arbitrary Lagrangian-Eulerian
TNT Trinitrotoluene
VoD Velocity of Detonation
Ps Incident pressure
Pr Reflected pressure
q Dynamic pressure
ts Positive phase duration
Is positive impulse
P0 ambient pressure
Specific heat capacity ratio
Z Scaled distance
W Charge mass
R distance
UFC Unified Facilities Criteria
m/s meter/second
xix
kg kilogram
Liquid volume fraction
w/w weight/weight
m micrometer
hrs hours
cP centipoise
W Volume fractions of water
a Volume fractions of air
DAQ Data Acquisition system
ms millisecond
mm millimeter
m meter
s microsecond
FE finite element
Diameter
GSM grams per square meter
Fragment size variable
C/M ratio of explosive to metal mass/unit length
mph miles per hour
1
Introduction Chapter 1
1.1 Motivation
Numerous terrorist attacks have caused thousands of deaths and left tens of
thousands injured besides damage to valuable infrastructure and property. In March
2004, several IEDs placed in four commuter trains in Madrid exploded causing 193
deaths and injuries to around 2000 people [1]. In July 2005, 52 died in London
underground train explosions [2]. More than 7500 people have died and over 17000
injured by suicidal attacks in Pakistan between 2007 and 2014. The yearly
breakdown is shown in Figure 1-1. About 5000 civilians lost their lives only in IEDs
attacks in Afghanistan between 2009 and 2014 [3]. In October 2015, a Russian
Metrojet was brought down over Egypt's Sinai peninsula by a bomb, killing all 224
people on board [4]. In April 2017, 11 people were killed and 45 injured in blast on
the St. Petersburg metro [5]. In September 2017, an IED was detonated on a Tube
train in south-west London injured 29 people [6]. These are the direct counts of
deaths in such extreme events. Another aspect of these attacks, which has usually
been ignored, is the indirect fatalities due to health, displacement and malnutrition of
the victimized families. According to Geneva Declaration Secretariat, "a reasonable
average estimate would be a ratio of four indirect deaths to one direct death in
contemporary conflicts” [3]. Therefore, safety against EM detonation is of vital
importance to avoid direct and indirect loss of lives.
Figure 1-1 : Fatalities in terrorist violence in Pakistan, (2000-2018) [3]
2
1.2 Energetic Materials
Energetic materials are a class of material which, when suitably ignited by an
external stimulus, release chemical energy stored in their molecular structure at
extremely fast rate. The history of energetic materials goes back to China with the
combustion of the powder mixture of sulfur, charcoal and nitrate salts [7]. These
materials release large amounts of energy and expand rapidly in volume to generate
force in the time scale of microseconds. High explosives, propellants and
pyrotechnics belong to the class of energetic materials. These materials are
commonly used in various military and commercial applications. Based on their
applications, explosives can be classified as low explosives(LE) and high explosives
(HE) [8].
1.2.1 Low Explosives
The decomposition or propagation of reaction rate in these explosives is fast but
subsonic. These explosives are ignited by a spark, flame or impact and this
combustion reaction is termed as deflagration [9]. The deflagration under
confinement turns into explosion. The reaction products move in a direction opposite
to that of reaction propagation. Propellants and pyrotechnics fall in this category of
energetic materials [7]. Propellants are not expected to detonate but combust in a
controlled manner. Propellants are used for the propulsion of different objects
including missiles, rockets, gun bullets etc. [9]. Pyrotechnics produce some special
effects like heat, light, smoke and sound [9].
1.2.2 High Explosives
Once ignited these explosives decompose spontaneously. The propagation of reaction
rate in these materials is supersonic. This type of combustion reaction is termed as
detonation. The unique aspect of detonation is the release of energy at extremely high
rate and production of highly compressed gaseous products [10]. The detonation
creates a shock wave propagating at supersonic speed due to the continuous
exothermic chemical reaction taking place inside the high explosive just behind
detonation front [11]. The point at which the exothermic reaction completes is called
Chapman-Jouguet (C-J) point. Just ahead of C-J point is a narrow pressure spike
known as Von Neumann spike. The thin region between this spike and the C-J point
3
is called reaction zone. A release wave propagating downstream into the detonation
products behind the detonation wave is called Taylor wave. [10]. The Von Neumann
spike is attenuated immediately after encountering a metal whereas the Taylor wave
produced the practical impact. The reaction products follow the direction of
propagation. The rate at which the reaction proceeds to the unreacted part of the high
explosive is called detonation velocity (D). It is a function of the density of the
explosive. For commonly used explosives detonation velocity ranges from 3000 to
9000 m/s. The structure of detonation wave and pressure profile is shown in Figure
1-2.
Figure 1-2 : The detonation wave structure and pressure profile [10]
High explosives (HE) are classified as primary and secondary based on their
sensitivity to initiation. Primary explosives are more sensitive and require only a
small amount of energy for initiation but are less powerful. These are mainly used in
detonators. Lead Azides, Mercury fulminate and Lead Styphnate are few examples
[8].
Secondary explosives are less sensitive but more powerful and can be initiated by a
shock wave. Common examples of secondary HE are TNT, C4, Comp-B, RDX
HMX etc. [8]. These explosives create shock waves resulting in shattering,
penetration, lift etc. [9].
4
1.3 The Rankine-Hugoniot Relations
The concept of shock wave propagation and the conservation equations can easily be
understood by considering a piston-cylinder configuration as shown in Figure 1-3
[10, 12]. The cross-sectional area of cylinder is taken as unity. The piston is initially
at rest and then pushed into the compressible material at velocity UP. After a time t1,
the compressed region ahead of piston traveled a distance USt1, where US is the
velocity of compressed /disturbed medium ahead of piston. In the meantime, piston
has traveled a distance of UPt1. This process is analogous to snow plow behavior with
US>UP. A very thin layer separates the compressed moving region from stationary
material in the cylinder. The plane separating “moving” from “stationary” fluid in the
cylinder is known as shock front. The initial and final parameters like, pressure,
density, specific internal energy and velocity are shown in the Figure 1-3.
Figure 1-3 :Shock wave generation in a compressible fluid [12]
Now, applying the conservation laws of mass, momentum and energy in the
compressed region, the conservation of mass is expressed as;
𝑈𝑆𝑡1𝜌0 = 𝜌1(𝑈𝑆 − 𝑈𝑃)𝑡1 (1.1)
For any time t, this equation can be written as;
𝜌0𝑈𝑆 = 𝜌1(𝑈𝑆 − 𝑈𝑃) (1.2)
The momentum conservation requires that the difference in the momentum ahead and
behind the shock front is equal to the impulse per unit of cross-sectional area:
𝜌1(𝑈𝑆 − 𝑈𝑃)𝑈𝑃𝑡 − 0 = (𝑃1 − 𝑃0)𝑡
𝜌1(𝑈𝑆 − 𝑈𝑃)𝑈𝑃 = (𝑃1 − 𝑃0)
(1.3)
The equation for conservation of energy requires that the work done by the external
forces is equal to the sum of the increase in both internal and kinetic energy (K.E):
𝐸1[𝜌1(𝑈𝑆 − 𝑈𝑃)𝑡 − 𝐸0[𝜌0𝑈𝑆𝑡] = (𝐸1 − 𝐸0)𝜌0𝑈𝑆𝑡 (1.4)
5
The kinetic energy is given by 1
2𝑚𝑣2 , this gives;
1
2𝜌1
(𝑈𝑆 − 𝑈𝑃)𝑈𝑃2𝑡 − 0 =
1
2𝜌1𝑈𝑆𝑈𝑃
2𝑡
(1.5)
For a stationary shock wave, the change in K.E is equal to the change in internal
energy.
𝐸1 − 𝐸0 =
1
2𝑈𝑃
2
(1.6)
The equations (1.2), (1.3) and (1.6) are known as Rankine-Hugoniot relationships for
a material where a pressure discontinuity propagates. The equations applied for a
piston-cylinder configuration with a compressible medium (gas) can be extended to
shock wave propagation into gas, liquid, or solid or to a detonation wave [10]. The
conservation equations for detonation are identical to these three equations with a
change that shock velocity US is replaced by D (detonation velocity). The above
stated three conservation equations have five unknowns US, UP, P1, 1 and E1. By
eliminating US and UP from these equations one can obtain a single equation which is
a function of state variables P1, 1 and E1 [12].
𝐸1 = 𝐸0 +
(𝑃1 + 𝑃0)(𝜌1 − 𝜌0)
2𝜌1𝜌0
(1.7)
This equation (1.7) is known as Hugoniot equation which represents the locus of all
the states that can be reached from an initial density 0 to final compressed density of
1.
1.4 Equation of State
In order to completely define the final state of a system invested by a shock front and
to plot the Hugoniot curve another equation is required. This equation is known as
the equation of state (EOS) of the material [12]. The equation of state of the material
defines a relationship of pressure in terms of internal energy and volume given by;
𝑃 = 𝑓(𝐸, 𝜌). The EOS of a material describes the behavior of a material under shock
wave loading [10]. The above stated three conservation equations together with C-J
condition (D = CCJ + (UP)CJ) and the equation of state of detonation products
completely describe the detonation process. Here, CCJ represents speed of sound at
6
the C-J point [10]. A linear relationship of the following form holds for various
materials under shock wave loading without phase change.
𝑈𝑆 = 𝐶0 + 𝑆 𝑈𝑃 (1.8)
Where, C0 is the sound speed of the uncompressed material and S is the slope of the
straight line. The other well-known EOS are ideal gas, Tillotson, Puff, JWL etc.
1.5 Formation of Blast Wave
The detonation of a HE generates hot gases at pressures from 100 to 300 kilobars and
at temperatures about 3000 – 4000oC. The violent expansion of these product gases
forces out the surrounding air. Consequently, a thin layer of highly compressed air- a
blast wave- forms and starts propagating outwards in front of the hot gases. The
pressure falls to ambient as the blast wave moves away from the source. As the
product gases continue to expand, they cool down and their pressure falls below the
ambient pressure. The expansion of these gases results in the reversal of flow
towards the source due to the pressure difference between the atmosphere and these
gases. This causes an under pressure -negative phase- in the pressure-time profile of
the blast wave. Finally, the conditions return to equilibrium as the air and gases
pushed away from the source come back [9]. A typical waveform of the blast wave
is shown in Figure 1-4 [13]. It is pertinent to define blast parameters associated with
a blast wave and loading estimation on an object. These parameters are also
illustrated in Figure 1-4.
Figure 1-4 : Characteristics of a blast wave (a) ambient pressure (b) positive phase (c)
negative phase [13]
Incident pressure (Ps) - Pressure acting on a surface parallel to the direction of blast
wave propagation [14].
7
Reflected pressure (Pr) - The pressure as measured front–on to the flow. Peak
reflected pressure is a function of incident pressure and angle of incidence [15].
Dynamic pressure (q) - The pressure due to moving air behind the shock front.
Positive phase duration (ts) - The time for which incident or reflected pressure stays
above the atmospheric pressure.
Incident Impulse (Is) - The area under the incident pressure time plot [16].
Stand-off (R) – The physical distance between the surface of a building or
component, and the center of charge.
The blast wave as shown in Figure 1-4 is a kind of shock wave that represents a
discontinuity in the medium. It is a compressive wave that moves at a speed higher
than the speed of sound in the medium. The blast wave decays immediately after the
peak pressure is reached [17]. The decay rate of the blast wave is given by Modified
Friedlander equation and developed by Thornhill is shown in eq. (1.9) [13, 18].
𝑃(𝑡) = 𝑃𝑠 (1 −
𝑡
𝑡𝑠)𝑒
−𝑏𝑡𝑡𝑠
(1.9)
Where, Ps is peak incident overpressure, ts is positive phase duration and b is a
constant. The blast wave undergoes reflection when the forward moving air
molecules are brought to rest and further compressed upon meeting an obstacle, thus
forming a reflected pressure (Pr) [15]. The reflected pressure depends on the angle
of incidence of the blast wave. For a blast wave impacting perpendicular to the
surface, the reflected pressure is determined by relations shown in eq. (1.10) and
(1.11) [9].
𝑃𝑟 = 2𝑃𝑠 + (𝛾 + 1)𝑞 (1.10)
𝑃𝑟 = 2𝑃𝑠[
7𝑃0 + 4𝑃𝑠
7𝑃0 + 𝑃𝑠]
(1.11)
Where, P0 is ambient pressure, specific heat capacity ratio of air ( = 1.4) and q is
dynamic pressure. The wind associated with the blast wave constitutes the dynamic
pressure (q) and is defined as half the density () times the square of the fluid
velocity (v). The dynamic pressure (q) is given in eq. (1.12). By using the Rankine-
Hugoniot relations q can also be computed by eq. (1.13) [19].
8
𝑞 = 1
2𝜌𝑣2 (1.12)
𝑞 =
𝑃𝑠2
2𝛾𝑃0 + (𝛾 − 1)𝑃𝑠
(1.13)
The dynamic pressure is responsible for drag loading on objects [20]. The area under
the pressure time plot for positive phase duration of equation (1.6) determines the
impulse. This impulse is responsible for the transfer of kinetic energy to the structure
which can cause damage to the structure or its components [21]. Impulse is not only
a function of peak pressure and positive phase duration but also a function of decay
rate [22]. The positive or incident impulse (Is) can be computed from the relation
given in eq. (1.14) and (1.15).
𝐼𝑆 = ∫ 𝑃𝑑𝑡
(1.14)
𝐼𝑠 =
𝑃𝑠𝑡𝑠
𝑒 = 0.368𝑃𝑠𝑡𝑠
(1.15)
Kingary-Bulmash [23] empirical equations provide a common approach to determine
the blast loading. The equations describe a range of blast wave parameters for a given
mass of explosive and stand-off distance. The parameters are normalized in terms of
TNT as the explosive and equations are expressed in the form of scaled distances (Z)
[15].
1.6 Scaling Laws
The blast wave parameters are functions of charge mass (W) and standoff distance
(R) [14]. Hopkinson [24] and Cranz [25] independently formulated blast wave
scaling laws which are also known as cube-root scaling. The approach leads to the
specification of scaled distance (Z) defined in eq. (1.16) [9]. The scaled distance (Z)
is widely used to estimate blast related parameters in empirical methods [14].
𝑍 =
𝑅
√𝑊3
(1.16)
For two charge masses W1 and W2 and distances R1 and R2, Z also defines the
constant of proportionality as given in eq. (1.17) and (1.18) [9].
𝑅1 = 𝑍 √𝑊13 (1.17)
𝑅2 = 𝑍 √𝑊23 (1.18)
9
Sachs [26] presented scaling laws that incorporate the effect of altitudes where
ambient conditions are significantly different to that of sea level. The overpressure at
an altitude can be computed as given in eq. (1.19).
𝑃𝑆 = 𝑃1
𝑃𝑎𝑡𝑚
𝑃0
(1.19)
Where P1 is the overpressure at sea level [19].
1.7 Effects of Cased Explosive Detonation
The detonation of a cased explosive device or an IED causes damage by following
three main mechanisms [27]:
Blast overpressure
Fragmentation
Thermal effects
The impact of the blast wave induces stresses in the target materials [9]. Compressive
and tensile waves traverse through the material. This wave propagation can result in
several irreversible phenomena including (but not limited to) plastic deformation,
fracture and phase transformation. The damage produced to a structure by a blast
wave depends on its orientation and usually related to a complex combination of the
hydrostatic and dynamic pressure forces. The response of the structure is a function
of the time-history of the blast wave loading [28].
The blast wind resulting from the blast overpressure leads to injuries and fatalities. A
wind speed of 163 mph is associated with a peak overpressure of 5 psi (34.5 kPa)
[20, 29]. The surrounding objects are thrown violently and crushed [29]. Human
beings are sensitive to overpressure, peak pressure rise rate, positive and negative
phase duration of blast wave and the specific impulse. The injuries caused by a blast
wave are categorized as:
Primary injuries
Secondary injuries
Tertiary injuries
Quaternary or Miscellaneous injuries
10
Primary blast injuries are caused solely by the direct effect of overpressure on tissue.
Organs containing air like ears, lungs and gastrointestinal track are the most
susceptible to this kind of injury [13]. The minimum pressure for fatal primary blast
injury is 29 psi (200 kPa) [27]. Secondary blast injuries are caused by the flying
objects that strike people. In recent years, nails, screws, ball bearings and other
metallic parts have been used in IEDs to enhance this injury mechanism. Tertiary
blast injury is caused by whole-body displacement at high blast overpressures and
impulses. Any part of the body can be affected by secondary and tertiary injuries. All
other injuries caused by the explosion are included in the quaternary (miscellaneous)
type [27, 30].
The ear is the most sensitive to blast but is not the most critical organ. Threshold for
rupture of human eardrum for a fast rising pulse is 5.0 psi (34.5 kPa). Blast
overpressure at which half of tympanic membranes rupture and the minimum
pressure for lungs damage is 15 psi (103.4 kPa) [27, 31]. Unified Facilities Criteria-
UFC 3-340-02 provides various damage criteria for specific organs of the human
body, equipment and explosives [32].
Fragmentation is one of the dominant threats to personnel from cased explosives or
IEDs. Injuries to personnel due to fragment impact can be classified into primary
fragment and secondary fragment injuries. Primary fragments, which are usually
small and have high-velocities, cause injuries by penetration and perforation of vital
organs of the body [27]. Secondary fragments are usually larger and have much less
velocity upon impact and can cause non-penetrating blunt trauma [33]. Therefore, it
is very difficult to effectively deal with this mechanism. It necessitates higher mass
solution or an expensive lightweight ballistic protection technique.
Thermal effect, burning due to heat and fireball, is another potential injury
mechanism associated with explosives detonation. [33]. The primary thermal hazard
posed by an explosive detonation is less significant than the blast and fragmentation
threats.
Besides these, an energetic material detonation causes damage to the surrounding
objects and structures. A few examples of structural damage caused by energetic
material detonation are shown in Figure 1-5.
11
1.8 Literature Review
The highly dynamic nature of the blast wave hampers protective measures. No single
material has been found to provide safety within affordable cost and weight limits.
Sandia National laboratories [34, 35] introduced aqueous foam for mitigation of blast
overpressure. Silnikov et al [36, 37] and Takayama,et al [38] investigated
effectiveness of multiphase material relaxation for blast mitigation in combined
inhibitors and found significant reduction in blast wave parameters. Gelfand et al
[39] studied blast wave mitigation properties of aqueous foam as a function of liquid
fraction. Del Pret et al [40] conducted experimental and numerical studies of HE
detonation inside aqueous foam.
Grujici et al [41] investigated ballistic performance of alumina/S-2 glass combination
against armor piercing (AP) and non- armor piercing projectiles. Poh and Wai [42,
43] investigated multi-layered construction against steel rod at impact velocity of
500 m/s. Zecevic et al [44] compared the fragmentation pattern and effectiveness of
M54 projectile with M107 and concluded that body material and energetic filling
plays important role in fragmentation mass and velocity distribution. Prytz, and
Odegardstuen [45] investigated the fragmentation of a 155 mm artillery shell by
statically detonating the shell and recovering its fragments to analyze the mass
distribution. Fragments’ velocities were estimated with numerical simulation only.
Arnold et al [46] studied the fragmentation behavior of very light and heavier casings
Figure 1-5 : Devastating effects of energetic material detonation
12
and found that the circumferential fragments size depends on material strength.
Nystrom and Gylltoft [47] investigated the combined effects of blast and
fragmentation loading on reinforced concrete and concluded that the combined
effects of blast and fragmentation loading are more severe than their separate effects.
Mohammad A. Abdallah [48] conducted fragmentation analysis of OG-7 warhead
using AUTODYN and compared the results with pit test results for OG-7 warhead.
Discrepancy in mass distribution between experimental and numerical data was
observed.
Ramadhan et al [49] studied impact response of sandwich structures based on
Kevlar/epoxy resin and Aluminum alloy laminated panels at impact velocities up to
400 m/s. The sandwich structures exhibited good energy absorbing characteristics
against impact loading. Liverts et al [50] worked specifically on conventional
aqueous foam Gillette using an exploding wire facility for blast wave mitigation.
Carton [51] experimentally investigated the effectiveness of water foam for the
mitigation of blast from HE detonation. Bornstein [15] studied water-filled
containers, numerically and experimentally, against near-field blast loading and
found that the containers provide 65% reduction to steel plate deformation. Alogla et
al [52] studied metallic protective panels against nearby blast loading. The fragments
effect was not studied. Nayak et al [53] studied 7.62 mm AP bullet impact at
velocity of 550 m/s on aramid fabric-epoxy composite laminates. Li et al [54]
introduced a lab scale experimental technique for studying combined loading effects
of blast and fragments, considering only single fragment impact with blast loading.
Lee et al [55] studied ballistic impact behavior of silica impregnated Kevlar against
fragment impact velocity up to 244 m/s. Soydan et al [56] experimentally and
numerically investigated the impact of a 9 mm FMJ bullet, of velocity 400 m/s, on a
three-layer armor configuration consisting of fiber-cement, Kevlar fabric and steel.
Ansari et al [57] investigated laminated GFRP composite plate under blunt projectile
impact at velocities below 50 m/s. Rasico et al [58] simulated the blast and
fragmentation of HE (high explosive) filled, M795 artillery shell IED (improvised
explosive device) buried in soil. The fragments impacted on vehicle hull were
investigated numerically.
Unified Facilities Criteria-UFC 3-340-02 provides blast resistant protective design
criteria for components, glass windows and structures. [32]. The latest version of
13
UFC 4-010-01 published in 2013 supersedes its previous versions and defined
different levels of protection [14]. Qi et al investigated mitigation of shock loads
from near field and contact detonations of high explosives using auxetic honeycomb-
cored sandwich panels. The combined shield was effective to protect reinforced
concrete structures against impact and near field blast loadings [59]. Langdon et al
investigated numerically and experimentally the mitigation of damage in aircraft
luggage containers subjected to internal blast loading. It was found that allowing a
venting in containers lengthwise along the aircraft body is more beneficial [60].
A destructible container [37] as shown in Figure 1-6 and liquid blast inhibitors
composed of several elastic envelopes can withstand only the blast effects of 0.5 kg
TNT.
Figure 1-6 : Blast inhibitor: 1-Elastic envelope, 2-liquid gas medium, 3-working space and 4-
HE [37]
Episafe [61] containers comprising a hard inner cylinder and an energy absorbing
outer cylinder made of high performance plastic fiber can contain blasts of cased
charges up to 0.47 kg TNT and equivalent. Episafe container as shown in Figure
1-7(a) is relatively bulky and direction dependent.
Figure 1-7: (a) Episafe container and (b) Resnyansky and Delany setup
a b
14
Resnyansky and Delany [62] recorded a large reduction of peak incident pressure by
surrounding 0.5 kg Pentolite with 110 liter of water inside a spherical container. The
setup is shown in Figure 1-7(b). Blastguard technologies Inc. [63] developed trash
receptacles to reduce lethal threats posed by the detonation of an improvised
explosive device (IED). The trash receptacle, shown in Figure 1-8, has a diameter of
30 inches and a height of 36 inches. It can contain a blast of charge up to 0.50 kg of
TNT or equivalent.
Figure 1-8 : Blastgard trash receptacle [63]
TM International [64] developed a steel wired rope blast containment tank for
explosive ordnance disposal (EOD). The vessel shown in Figure 1-9 is employed to
counter emergency scenarios involving an energetic material detonation. The product
is primarily intended to contain fragments but it can also reduce the blast
overpressure by a minimum of 50%.
Figure 1-9 : TM International blast containment vessel [64]
Mistral Security [65] has developed a Blast Containment Receptacle (BCRs) which
provides protection against fragmentation and blast pressure. BCR0.5 weighing 160
kg shown in Figure 1-10 can withstand a blast of 0.5 kg TNT or equivalent and it
costs about USD1800 –USD2000 [66].
15
Figure 1-10 : Blast Containment Receptacle (BCRs) [65]
Greenfield et al [67] developed a container for explosive devices comprising inner
and outer containment vessels to contain explosively driven fragments and blast
overpressure. The inner and outer vessels were made of 12.5 mm thick HY80 Steel.
Edberg et al [68] presented a shield formed of one or more sprays of attenuation
material comprising a gas, such as air, disposed as bubbles in a liquid medium for
blast mitigation. Fly-Bag [69] made of multi-layered textiles is fitted inside the
standard aluminum containers that are used for loading luggage can stop a small-to-
medium-sized (200-500g) bomb hidden in a suitcase or a cargo-hold from damaging
the structure. It can only be fitted in narrow body aircrafts. FLY BAG2 [70] has
developed a technology that could allow planes in the future to survive a Lockerbie-
sized explosion. Fly Bag2 is a joint project of 14 European countries.
1.9 Gaps in Literature
Much of the material in this field is restricted or proprietary which makes the
learning inherently difficult. Although aqueous foam barrier is very effective for
blast and shock wave mitigation, it cannot protect against the fragmentation effects
associated with the detonation of cased charges or IEDs. Sembian and Liverts et al
[50, 71] worked specifically on conventional aqueous foam using an exploding wire
facility. The facility, however, lacks generation of detonation products produced in
case of energetic material detonation. Fountain blast inhibitors [36] and destructible
container [37] can only provide safety against blast effects up to 0.5 kg of TNT
equivalent charge. The details of the working medium are not available. Episafe [61]
is relatively bulky and direction dependent. Grujicic et al [41] used encapsulated
ceramic plates and alumina/S-2 glass for protective armor but the shattering of brittle
ceramic plate and resulting secondary fragmentation was not investigated. Alogla et
16
al [52] studied metallic protective panels against nearby blast loading. The fragments
effect was not studied. Blastgard trash receptacles, TM International containment
vessels and Mistral Security BCRs [63-65] are good solutions but fairly bulky and
expensive. BCR0.5 that can withstand a blast of 0.5 kg TNT or equivalent costs
about USD1800 –USD2000 [66].
1.10 Objectives
The main objectives of the present work are;
1. To measure the blast wave parameters produced by C4 detonation and its
mitigation.
2. To search, characterize and investigate commercially available materials for
fireball suppression, blast mitigation and fragments containment.
3. To study the response of multi-layered combination of Kevlar fabric, GFRP
and PU/PS foams against high velocity fragments and blast loading.
4. To develop a model container against combined blast, fragmentation and
thermal effects of 1.0 kg TNT equivalent charge detonation.
1.11 Thesis Organization
A brief introduction and literature review of the effects of EM detonation and
mitigation strategies is laid down in Chapter 1. Materials, diagnostic tools and
experimental work applied for blast mitigation are presented in Chapter 2.
Experimental and simulation approaches for characterization of shell fragmentation
and testing of protective configurations are discussed in Chapter 3 and Chapter 4
respectively. Design and testing of a protective container based on the simulated and
experimental data obtained is laid down in Chapter 5. Finally, Chapter 6 presents
conclusions from the current research as well as provides some recommendations
17
References
[1] 2004 Madrid train bombings. Available from:
https://en.wikipedia.org/wiki/2004_Madrid_train_bombings.
[2] 7 July 2005 London bombings. Available from:
https://en.wikipedia.org/wiki/7_July_2005_London_bombings.
[3] N.C. Crawford, War-related death, injury, and displacement in Afghanistan
and Pakistan 2001-2014. (2015.
[4] Russian plane crash. Available from: https://www.bbc.com/news/world-
middle-east-34687990.
[5] St Petersburg metro explosion. Available from:
https://www.bbc.com/news/world-europe-39481770.
[6] Parsons Green: Underground blast. Available from:
https://www.bbc.com/news/uk-41278545.
[7] D.I. Millar, Energetic materials at extreme conditions. (2011.
[8] J.A. Zukas, W. Walters, W.P. Walters, Explosive effects and applications.
(2002.
[9] P.D. Smith, Blast and ballistic loading of structures / P.D. Smith, J.G.
Hetherington. (1994.
[10] M.A. Meyers, Dynamic behavior of materials. (1994.
[11] G.L. Hempen, (1993).
[12] M. Scapin, Torino, Italia, (2013).
[13] B. Rutter, (2019).
[14] H.D. Hidallana Gamage, Response of laminated glass panels to near field
blast events(2015).
[15] H. Bornstein, Physical mechanisms for near-field blast mitigation with fluid-
filled containers(2018).
[16] H. Bornstein, S. Ryan, A.P. Mouritz, Physical Mechanisms for Near-Field
Blast Mitigation With Fluid Containers, in Explosion Blast Response of
Composites. 2017, Elsevier.345.
[17] C.E. Needham, Blast waves. Vol. 402. (2010.
[18] C. Thornhill, The shape of a spherical blast wave, Armament Research and
Development Establishment (ARDE) Memo.(B) 41/59(1959).
[19] E. Lozano, Design and analysis of a personnel blast shield for different
explosives applications(2016).
18
[20] S. Glasstone, D.J. Philips, The effects of nuclear weapons, 3rd
ed. (1977.
[21] T. Rahimzadeh, Design of Protective Structures for Optimal Blast and Impact
Mitigation(2016).
[22] G.F. Kinney, K.J. Graham, Explosive shocks in air. (2013.
[23] C.N. Kingery, Bulmash, G., Airblast parameters from TNT spherical air burst
and hemispherical surface burst(1984).
[24] B. Hopkinson, The National Archives, Kew, UK. 11(1915).
[25] C. Cranz, Lehrbuch der Ballistik II(1926).
[26] R.G. Sachs, The dependence of blast on ambient pressure and
temperature(1944).
[27] P. Peters, Military Medicine. 176(2011) 110.
[28] I. Sochet, Blast effects: physical properties of shock waves. (2017.
[29] R.K. Zipf, K.L. Cashdollar, (2006).
[30] K. Ahmed, A.Q. Malik, A. Hussain, I.R. Ahmad, I. Ahmad, AIP Advances.
10(2020) 095221.
[31] A.I.S. M.V. Silnikov, A.I. Mikhaylin, A.V. Orlov, Materials Physics and
Mechanics. 20(2014) 11.
[32] D.o. Defense, Unified Facilities Criteria (UFC): Structures to Resist the
Effects of Accidental Explosions (UFC 3–340–02)(2008).
[33] K.J. Sharpe, J. Waddell, J.F. Gordon, Explosive effect mitigated containers
and enclosing devices(2008).
[34] W.F. Hartman, M.E. Larsen, B.A. Boughton, Blast mitigation capabilities of
aqueous foam(2006).
[35] T.D. Panczak, H. Krier, P.B. Butler, Journal of hazardous materials. 14(1987)
321.
[36] M. Silnikov, A. Mikhaylin, N. Vasiliev, V. Ermolaev, V. Shishkin, NATO
Security through Science Series B: Physics and Biophysics, (2006).
[37] M. Silnikov, A. Sadyrin, A. Mikhaylin, A. Orlov, Materials Physics &
Mechanics. 20(2014).
[38] K. Takayama, М. Silnikov, M. Chernyshov, Acta Astronautica. 126(2016)
541.
[39] B. Gelfand, M. Silnikov, M. Chernyshov, Shock Waves. 20(2010) 317.
[40] E. Del Prete, A. Chinnayya, L. Domergue, A. Hadjadj, J.-F. Haas, Shock
waves. 23(2013) 39.
19
[41] M. Grujicic, B. Pandurangan, K.L. Koudela, B.A. Cheeseman, Applied
Surface Science. 253(2006) 730.
[42] C.W. Poh, Investigation of new materials and methods of construction of
personnel armor, M.Sc(2008).
[43] B.C. Wai, Investigation of Shock Wave Attenuation in Porous Media,
M.Sc(2009).
[44] B. Zecevic, J. Terzic, A. Catovic, S. Serdarevic-Kadic. Influencing
parameters on HE projectiles with natural fragmentation. in International
Conference on New Trends in Research of Energetic Materials. 2006.
[45] A.K. Prytz, G. Odegardstuen, 25th Int. J. Ballistics, September. 12(2011) e22.
[46] W. Arnold, E. Rottenkolber, International journal of impact engineering.
35(2008) 1393.
[47] U. Nyström, K. Gylltoft, International Journal of Impact Engineering.
36(2009) 995.
[48] M.A. Abdalla. Fragmentation Analysis of OG-7 Warhead Using AUTODYN
SPH Solver. in Advanced Materials Research. 2012. Trans Tech Publ.
[49] A. Ramadhan, A.A. Talib, A.M. Rafie, R. Zahari, Materials & Design.
43(2013) 307.
[50] M. Liverts, O. Ram, O. Sadot, N. Apazidis, G. Ben-Dor, Physics of Fluids.
27(2015) 076103.
[51] E. Carton, AIR BLAST MITIGATION USING WATER FOAM
COVERAGE(2018).
[52] A. Alogla, M. Helal, M.M. ElShafey, E. Fathallah, Applied Sciences.
10(2020) 2121.
[53] N. Nayak, A. Banerjee, T. Panda, Procedia engineering. 173(2017) 230.
[54] L. Li, Q.-C. Zhang, R. Zhang, X. Wang, Z.-Y. Zhao, S.-Y. He, B. Han, T.J.
Lu, International Journal of Impact Engineering. 134(2019) 103382.
[55] Y.S. Lee, E.D. Wetzel, N.J. Wagner, Journal of materials science. 38(2003)
2825.
[56] A.M. Soydan, B. Tunaboylu, A.G. Elsabagh, A.K. Sarı, R. Akdeniz,
Advances in Materials Science and Engineering. 2018(2018).
[57] M.M. Ansari, A. Chakrabarti, M.A. Iqbal, Procedia engineering. 173(2017)
153.
20
[58] J. Rasico, C. Newman, M.R. Jensen, MODELING FRAGMENTATION OF
A 155MM ARTILLERY SHELL IED IN A BURIED MINE BLAST
EVENT. (2016.
[59] C. Qi, A. Remennikov, L.-Z. Pei, S. Yang, Z.-H. Yu, T.D. Ngo, Composite
structures. 180(2017) 161.
[60] G. Langdon, S. Kriek, G. Nurick, International Journal of Impact
Engineering, (2020) 103567.
[61] J. Polak, R. Romek, A. Wiśniewski, Problemy Techniki Uzbrojenia.
37(2008).
[62] A. Resnyansky, T. Delaney, Experimental study of blast mitigation in a water
mist(2006).
[63] Blastguard Trash Receptacles. Available from:
http://blastgardtech.com/blastwrap#blastgard-mtr.
[64] Blast Containment Tank. Available from:
https://www.tmi2001.com/products/blast-containment-tank/.
[65] Blast Containment Receptacles. Available from:
http://www.mistralsecurityinc.com/Our-Products/Blast-Management-
Systems/Blast-Containment-Receptacles-BCRs.
[66] Y. Banai, Blast Containment Receptacles-Pricing).
[67] G. Greenfield, P.R. Gefken, J.D. Colton, Container for explosive
device(2006).
[68] D.L. Edberg, S. Schneider, Blast attenuation device and method(2005).
[69] D. Zangani. Blastworthy textile-based luggage containers for aviation safety.
Available from: https://cordis.europa.eu/project/id/213577.
[70] D.Z. Alessandro Bozzolo, Advanced technologies for bomb-proof cargo
containers and blast containment units for the retrofitting of passenger
airplanes(2015).
[71] S. Sembian, M. Liverts, N. Apazidis, Physics of Fluids. 28(2016) 096105.
21
Materials and Methodology for Blast Chapter 2
Mitigation
2.1 Introduction
The chapter covers an introduction of materials used for generation of blast wave and
its mitigation. Characterization of commercially available shaving foam and its blast
mitigation capabilities are presented. The diagnostic tools employed for visual study,
fireball measurement and determination of blast wave parameters are briefly
discussed. Blast wave parameters for C4 explosive detonation in air and covered with
shaving foam are presented in this chapter.
2.2 Aqueous Foam
Aqueous foam is a bi-phase system where air cells are enclosed by thin liquid films.
It consists of a foam concentrate of de-ionized water with a surfactant and a polymer
stabilizer essentially introduced with gas/air component. The aqueous foam is
categorized as dry, wet and bubbly based on the liquid volume fraction () [1].
dry foam: < 0.05
wet foam: 0.05 < < 0.36
bubbly liquid: > 0.36
Aqueous foam in particular has been a subject of research since its first development
at Sandia laboratories in 1983. It was developed to mitigate blast effects and to
capture the hazardous particulate material from unwanted dispersion in the
atmosphere. The blast wave strength can also be significantly attenuated by
surrounding the energetic material with aqueous foam [2, 3].
The aqueous foam developed specifically for blast mitigation may not be available to
the general public which necessitated the search for a commercially available
alternate. Present work addresses the investigation of commonly available health care
product DENIM shaving foam against an energetic material detonation including the
suppression of fireball as well as the overpressure attenuation.
22
2.2.1 DENIM Shaving Foam
The composition of the contents under pressure in the aerosol can is [4];
Bulk (Liquid surfactant system): 96.00 % w/w
Hydrocarbon propellant: 4.00 % w/w
The aerosol system is a sealed system, so both the bulk and the propellant are liquid
and mix under pressure to form a homogeneous liquid system. Only a very small
amount of propellant, residing in the head space of the can, is in vapor form and
helps to expel the product out when the actuator is pressed [4].
2.2.2 Microscopic and Viscosity Study
The shaving foam used in this research work has a density of 60 kg/m3
as measured
using a pycnometer. Microscopic study of the shaving foam was performed using
Olympus microscope. The foam sample was dropped on the slide and studied at
magnification 100X. The foam produces finely dispersed bubbles with an average
initial size of 15 m. The microscopic study of conventional aqueous foams (Gillette
regular and DENIM original) was carried out at different time intervals to observe
the foam stability, bubble size and coalescence. Foam stability is related to its
drainage rate (i.e. the volume of liquid that drains out of expanded foam over time).
Foams were examined in 30 minute intervals for a total time of two hours Initially, a
rapid increase in average bubble size is observed as depicted in Figure 2-1, Figure
2-2 and Figure 2-3. The coarsening due to coalescence of bubbles slowed down with
time. The bubble size increased to 120 m in 2 hours. The aqueous foams stability
deteriorated with time but the DENIM foam was found to be in better condition even
after 24 hours.
a b
Figure 2-1 : Denim foam-average bubble size (a) 15 µm at 0 hrs (b) 85 µm at 1 hrs
23
The graphical representation of bubbles coarsening of both commercially available
shaving foams is shown in Figure 2-4.
c d
Figure 2-2: Denim foam-average bubble size (c) 120 µm at 2 hrs (d) at 24 hrs
c d
a b
Figure 2-3: Gillette foam-average bubble size (a) 20 µm at 0 hrs (b) 105 µm
at 1 hrs (c) 155 µm at 2 hrs (d) at 24 hrs
24
The viscosity of DENIM shaving foam was measured with DV2T viscometer at 24
0C using RV06 spindle for torque ranging from 17.4% to 45%. A rapid drop in
viscosity was observed at low shear rates. The decreasing viscosity trend at higher
shear rates is shown in Figure 2-5 and indicates a shear thinning behavior of the two-
phase medium.
2.2.3 Sound Speed Measurement
Sound speed was measured both in air and in shaving foam at a temperature of 240C.
A motor bike horn was used as the sound source. Two terminal condenser mics,
connected with an RC circuit and a 5V DC power supply, were used to record sound
signals on an oscilloscope. Perspex channel measuring 225 x 75 x 75 mm3 was used
to house the two mics and foam. Mics were fixed at a separation of 84 mm. The
testing setups are shown in Figure 2-6 and Figure 2-7 for sound speed measurement
in air and shaving foam respectively.
Figure 2-4: The graphical representation of bubbles coarsening of conventional
aqueous foams (a) Denim (b) Gillette
Figure 2-5: Shear rate vs viscosity plot for shaving foam
25
The waveforms of sound signals were recorded on mic-1 and mic-2 and displayed on
two channels of oscilloscope. The time interval (T) between the two peaks was
recorded. Sound speed was calculated using the simple expression given in eq. (2.1).
𝐶0 =
𝑺
∆𝑻
(2.1)
Where, S is the distance between two mics and T is time interval obtained from
oscilloscope. The average sound speed measured in air at 240C was 345.8 m/s. The
sound speed is proportional to square root of the absolute temperature. From
literature, the sound speed in air at 240C is 343.0 m/s [5]. The relative error in
measured value with this setup was less than 1.0 %.
The Perspex channel and the complete path including the sound source, horn, was
then filled with shaving foam as shown in Figure 2-7 and T are measured to
compute the sound speed in foam. The recorded waveforms are shown in Figure 2-8.
The sound speed measured in shaving foam was 65.98±3 m/s. A significant reduction
in sound speed was observed in this two-phase medium compared to sound speed in
air and water.
Mic-1 Mic-2 Horn
Perspex channel
Figure 2-6: Setup for measuring sound speed in air
Horn
Foam in Perspex channel
Figure 2-7: Setup for measuring sound speed in shaving foam
26
The two-phase mixture density and sound speed can be computed by Wood’s law
(also known as mixture law) given in eqs. (2.2) and (2.3) [6, 7].
𝜌𝑓 = 𝛼𝑊𝜌𝑊 + (1 − 𝛼𝑊)𝜌𝑎 (2.2)
1
𝜌𝑓 𝐶2=
𝛼𝑊
𝜌𝑊𝐶𝑊2 +
𝛼𝑎
𝜌𝑎𝐶𝑎2
(2.3)
Where W and a are volume fractions of water and air. The subscripts f, w and a
represents foam, water and air respectively. For a liquid volume fraction of 6.0% in
foam the density and sound speed computed from this relation are 61.15 kg/m3 and
52.69 m/s respectively. The measured values for density and sound speed are in close
agreement to the calculated values.
2.3 C4 Explosive
C4 or Composition C4 is RDX based explosive. The heat of detonation of C4 is 5.3
MJ/kg [8, 9]. The prevalent formulation is;
RDX 91.0%,
Di(2-ethylhexyl) Sebacate 5.3%,
Polyisobutylene (PIB) 2.1 %
20-weight motor oil 1.6%
A similar British formulation is known as PE-4 that comprises 88 % RDX and 12 %
plasticizer. An important characteristic of C4 is that it can easily be molded into any
desired shape. It can only be initiated by a shock wave [10]. A cylindrical geometry
was used in the present work. The velocity of detonation and C-J pressure of C4 used
a b
Figure 2-8: Waveforms (a) in air (b) in shaving foam
27
in present work (at C4 density=1.46 g/cm3) were 7300 m/s and 19.98 GPa
respectively [8, 11].
2.4 Diagnostic Tools
2.4.1 High-Speed Photography
High-speed photography is used for recording transient phenomenon. The advances
in high-speed camera technology have enabled researchers to see dynamic events
taking place in microsecond time domain. The events may include gaseous discharge,
detonation, explosion of wires under high current and blast wave propagation. High-
speed camera was employed in present work for visual study of the blast wave
propagation and in particular for fireball measurement. The high-speed Phantom
camera was operated at frame rates of 8000 -54000 fps during these tests.
2.4.2 Pressure Transducers
Transducer is a device that converts one form of energy into another. Piezo-electric
(PE) gauges have been used to measure pressure time, P(t), history in a variety of
applications including- blast waves in air, propellant pressure in gun chamber and
combustion chamber pressure in rocket motors. Many different crystals generate
piezo-electric charge when strained. For pressure gauge application quartz has been
used as active material in most transducers because of its desirable electrical and
mechanical properties [12] .
The transducer comprises a stainless steel (SS) case with a coaxial electrical
connector at one end. A SS diaphragm closes the other end while covering the cavity
that contains the sensing element. The sensing elements are stacked quartz plates.
The applied pressure on the diaphragm compresses the quartz stack and produces
proportional charge output. The piezoelectric crystal generates a very small charge
which is amplified by a charge amplifier. If the charge amplifier is an external device
it is referred as charge output or PE sensor. If the electronic circuit is integrated in to
the sensor housing, it is referred as voltage output or IEPE sensor [12]. The
transducer and its orientation for measuring incident pressure (Ps) and reflected
pressure (Pr) are shown in Figure 2-9(a). Sensitivity of the transducer is expressed in
28
milli-volts per psi (mV/psi). The coaxial cable simultaneously carries operating
power to the transducer and transmits the signal from it.
Kistler-PIEZOTRON 211B series pressure transducers (211B2, 211B3 and 211B5)
were used with the multi-channel coupler as a power source and amplifier for
transducer output. The Data Acquisition (DAQ) system comprises National
Instruments 16-bit, 1.25 MS/s hardware, which was coupled with eight-channel
noise, minimized differential signaling acquisition software to receive transducer
data. The data stored in the PC was plotted for P(t) history and impulse calculations.
The DAQ system may be operated in the frequency range of 100 kHz–1000 kHz. The
components of DAQ are shown in Figure 2-9(b).
Figure 2-9: (a) Transducers orientation for Pr and Ps measurement (b) Components of data
acquisition (DAQ) system
b
Transducer Multi Channel
Coupler
8-Channel Data Acquisistion
Systerm-(16-bit, 1.25MS/sec),
100-1000 kHz
Storage/Display Device
a
29
2.4.3 Arrival Time Sensors
To augment the diagnostic equipment, cylindrical sensor/probe as shown in Figure
2-10 was developed and employed to measure the time of arrival (Ta) of blast waves.
The sensor comprised an aluminum cylinder, an axial pin electrically insulated from
the cylinder and a Brass foil of diameter 20 mm fixed at one end. The Brass foil and
pin were separated by a very small distance. The cylindrical casing and the pin were
connected through a cable to a recording device like an oscilloscope or a transient
recorder. The blast wave impact at the thin foil pushed the foil towards the axial pin.
The electrical path is completed with the contact of foil and the pin and a signal is
registered on the oscilloscope or the transient recorder. The sensors were placed in
parallel to the pressure transducers. The arrival time measured, for 82 grams C4
detonation, with the sensor was in close agreement with that of the pressure
transducer measurements, as shown in Figure 2-11. These sensors are quite
inexpensive.
a c b
Pin
Figure 2-10: (a) Arrival time sensor (b) section view (c) mounted in the fixture
Figure 2-11: Comparison of Time of arrival with pressure
transducer and arrival time probe
30
2.5 Experimental Work
2.5.1 Blast Mitigation with Shaving Foam
Four Experiments were performed to evaluate the fireball suppression and blast
attenuation capabilities of shaving foam. All experiments were performed using C4
explosive. TNT equivalent of C4 is taken as 1.27 [13]. Cylindrical geometry of
explosive was used in the present work. Experiments were conducted with bare
charges as well as covered in shaving foam. The blast wave measurement and
recording equipment (DAQ system) are schematized in Figure 2-9(b). High-speed
photography was employed for visual study in particular, fireball measurement. A 2D
view of the experimental layout is shown in Figure 2-12.
C4 was placed on polystyrene (PS) foam of radius 21 mm and height 25 mm above a
4 mm thick plywood sheet as shown in Figure 2-13. The charge was initiated from
the top side with a Tetryl booster pellet and a detonator. The pressure sensors fitted
inside steel pipe measure incident pressure (Ps) whereas the sensor flush mounted
and facing the charge as shown in Figure 2-13(c) measures reflected pressure
(Pr).The blast wave parameters (Ps, Pr and Ta) were measured for bare 82 g C4
charge and covered in shaving foam of radius 0.2 m having volume 0.0168 m3
(weighing 1.0 kg). This shaving/aqueous foam has relatively lesser liquid volume
fraction (0.6). An increased liquid volume fraction enhances the mitigation as
reported by Kan et al [14, 15]. Therefore, 100 ml water was sprayed uniformly with
Figure 2-12: 2D view of Experimental Layout for blast
parameters measurement
31
the foam. The transducers were placed 230 mm above the C4 stand to avoid the
recording of a reflected blast from the plywood stand.
The fireball formation and expansion with and without shaving foam are shown in
Figure 2-14 and Figure 2-15. The blast wave undergoes reflection when the forward
moving air molecules are brought to rest and further compressed upon meeting an
obstacle, thus forming a reflected overpressure. The reflected overpressure time
history Pr(t) and impulse at 0.8 m is presented in Figure 2-16. Testing setup and blast
wave parameters for the bare and foam-covered 82 g C4 at 0.85 m distance is shown
in Figure 2-17 and Figure 2-18 respectively.
Figure 2-14: High-speed images, 82g bare C4 detonation in air and propagation of
fireball
Figure 2-13: Test setup for measuring blast parameters for (a) bare charge (b) immersed in
shaving foam(c) Transducer orientation for measuring reflected pressure
a b c
32
Figure 2-15: A sequence of events after detonation of charge immersed
in shaving foam
Figure 2-16: (a) Reflected overpressure (b) Impulse plots for 82g C4 blast
at 0.8 m for bare charge and covered in shaving foam
a
b
33
Figure 2-17: Test setup for measuring 82g C4 blast parameters for (a) bare
charge (b) covered in shaving foam at 0.85 m
b a
Figure 2-18: (a) Incident overpressure and (b) Impulse profiles for 82g C4 bare
and covered in shaving foam at 0.85 m
a
b
34
Another experiment was performed using 250 grams C4 charge. The charge was
immersed in 0.038 m3 volume (weighing 2.28 kg) of shaving foam without water
spray. Incident overpressure time history Ps(t) and positive impulse at 1.0 m distance
was measured and shown in Figure 2-19. Peak incident overpressure of 81 psi (558.4
kPa) and 32 psi (220.6 kPa) was recorded at 1.0 m distance for C4 detonation in air
and covered in shaving foam respectively. The corresponding blast wave arrival time
was 0.63 ms and 1.17 ms respectively.
Results for bare and foam-covered charges at different distances are presented in
Table 2-1. Arrival time (Ta) delay of 61%, peak pressure reduction of 72 and impulse
reduction of 61% were observed with shaving foam confinement. Since data was
clipped at 1.0 ms for sensor at 1.1 m distance, the impulse value for the shaving foam
should be higher than recorded. An extrapolated value of positive impulse is
tabulated at this distance.
Figure 2-19: (a) Pressure and (b) Impulse plots for bare 250g C4 blast at 1.0 m and
submerged in shaving foam
b
a
35
Table 2-1 : Blast parameters for bare charges and covered in shaving foam
C4 Dist. Blast Parameters
Bare C4 charge With shaving foam
(g) (m) Ta(ms) Ps(psi) Pr(psi) Impulse
(psi-ms)
Foam
(grams)
Ta(ms) Ps(psi) Pr(psi) Impulse
(psi-ms)
82
0.50 0.23 143 7.93 1000 0.39 43 2.72
0.70 0.45 62 4.91 0.67 18 2.20
0.80 0.57 185 18.87 0.90 51 6.11
0.85 0.72 47.6 4.1 1.20 17 2.9
1.10 0.99 55
250
0.8 0.42 133 602 6.5 2280 0.74 42 141 3.51
1.0 0.62 81 6.2 1.17 32 4.86
250 0.8 0.42 133 6.5 2700 0.72 36 3.44
2.5.2 Effects of Foam Volume and Liquid Content on Mitigation
The test was repeated with an increased volume of shaving foam having a radius 0.28 m
and volume of 0.045 m3 (weighing 2.7 kg). Moreover, 300 ml water was uniformly
sprayed with the foam. The setup is shown in Figure 2-20. The pressure plot with
shaving foam is shown in Figure 2-21. About 76% overpressure reduction is observed in
this case. The increased volume of shaving foam around the charge and the liquid
volume fraction increased the attenuation level. Further experimental results are
presented in section 2.7.2.
--0.56m-
--0.76m-
250 g C4
Figure 2-20: Test setup for measuring 250g C4 blast parameters for bare
charge and covered in 0.05 m3 shaving foam
36
2.5.3 Blast Parameters at Z < 1.0 m/kg1/3
Overpressure measurements were made within the shaving foam at scaled distances
(Z) less than 1.0 m/kg1/3
. The testing setup is schematized in Figure 2-22and Figure
2-23. The pressure transducers were fixed within the shaving foam at close distances
from the charge. The transducers and data cables were sealed and fitted in safety
jackets to avoid damage from fireball at such close distances. The charges weighing
150, 200 and 250 grams were detonated within the Perspex channels measuring
(1000 x 200 x 200) mm3 filled with shaving foam. The foam thickness behind C4
cylinder was 150 mm. The high-speed images for the blast mitigation within shaving
foam, shown in Figure 2-24, depict the rapid blast propagation in air (left side in each
image) and its attenuation inside the foam (right side in each image).
Figure 2-21: Peak incident pressure for 250g Bare C4 and covered in
shaving foam at 0.8m
a b
Figure 2-22: (a) Perspex channel filled with foam (b) Testing setup
within shaving foam
37
Since the blast wave parameters for an open air HE detonation could not be measured
at such proximity due to fireball effects, therefore, numerical simulations were
performed to acquire blast parameters for bare charges at 0.20 – 0.50 m standoff
distances. Experimental and simulated results are shown in Figure 2-25 and Figure
2-26. Peak pressure of 731 psi (5040 kPa) measured at 0.25 m was reduced to 244 psi
(1682 kPa) at 0.3 m for 200 g C4 within shaving foam. The corresponding impulse
values were reduced from 74 psi-ms to 40 psi-ms.
b a c
Figure 2-24: High-speed images of 200g C4 detonation inside Perspex channel (a)
just after detonation (b) fireball expansion (c) product gasses expansion
a b
c
Figure 2-23: Experimental setup for measuring blast parameters for Z < 1 (m/kg1/3
) (a)
sensors fixed inside empty Perspex channel (b & c) shaving foam filled inside channel
38
Figure 2-25 : Pressure and scaled distance plots for HE detonation in shaving foam for Z<1
The pressure-time history measured at 0.25 m and 0.30 m are shown in Figure
2-26(a). The impulse at these distances is also shown in Figure 2-26(b). Average
a
b
Figure 2-26: (a) Pressure and (b) Impulse plots for 200g C4 charge
detonated inside shaving foam at 0.25 and 0.30m from charge face
39
peak pressure attenuation of 73% within the shaving foam is observed for Z<1. The
dry aqueous foam significantly reduced the peak pressure and positive impulse.
2.6 Numerical Modeling and Simulation
ANSYS AUTODYN [16] was used to develop models and simulate the blast wave
parameters for bare C4 detonation. Multi-material (MM) Euler solver was used to
simulate C4 detonation and expansion in air in axial-symmetric form.
2.6.1 Material Modeling
JWL (Jones-Wilkins-Lee) Equation of state (EOS) [17] was used for C4 detonation
and product expansion. The JWL EOS is given in equation (2.4).
𝑃 = 𝐴 (1 −𝜔
𝑅1𝑉) 𝑒−𝑅1𝑉 + 𝐵 (1 −
𝜔
𝑅2𝑉) 𝑒−𝑅2𝑉 +
𝜔𝐸
𝑉 (2.4)
Where E is the detonation energy per unit volume, V is the ratio of the detonation
product volume with the original volume of the explosive, and A, B, R1, R2 and ω are
empirical fitting parameters. Ideal gas EOS shown in equation (2.5) was used for air
surrounding the shell.
𝑃 = 𝜌𝑎(𝛾 − 1)𝑒 (2.5)
Where P is the pressure, is specific heats ratio (=1.4 for air), e is specific internal
energy and a is air density. The JWL parameters for C4, as shown in Table 2-2,
were calculated by EXPLO5 code [8]. The difference from the standard published
data was due to relatively lesser density of C4 used in the present study. The
parameters for Ideal gas EOS used for air were obtained from ANSYS AUTODYN
material library. The space surrounding the charge was filled with air for an initial
pressure of 101 kPa (14.7 psi) by assigning specific internal energy of 2.068e+5
kJ/kg. Flow out boundary condition was used at the boundaries. Material data for
polystyrene (PS) foam was taken from AUTODYN library. Material model for balsa
wood [18], as given in Table 2-2, was used for simulating the plywood sheet as data
for plywood was not available.
40
Table 2-2: JWL parameters of C4, EOS data for Balsa wood and air
2.6.2 Blast Wave Parameters
The numerical simulation results of detonation events greatly depend on the mesh
size. An optimal mesh size that speeds up the calculations and give adequate results
is important [19]. The errors due to mesh discretization can be fixed by evaluating
the quality and adequacy of the mesh [20]. The discretization error in present
simulation work was minimized by optimizing the grid size. Simulation was run as
per experimental setup for bare 82 g C4 with different grid sizes and results are
presented in Table 2-3. Simulation with grid size of 1 x 1 mm2 reasonably well
reproduced the experimental results.
Table 2-3: Effect of grid size and discretization error in AUTODYN Simulation
The expansion of detonation product gases after the detonation of C4 in air and
subsequent formation of blast wave is schematized in Figure 2-27 and Figure 2-28
Simulations were run for bare 82 and 250 g C4.
JWL parameters for C4 Balsa wood
Equation of State JWL EOS Shock
Density (g/cm3) 1.46 Density (g/cm
3) 0.123
A 3.13640E+08 (kPa ) 1.415
B 6.39000E+06 (kPa ) C (m/s) 61
R1 4.0600 S 1.31
R2 9.80000E-01 Air
W 3.50000E-01 EOS Ideal gas
Detonation velocity 7.301E+03 (m/s ) Density (g/cm3) 0.00123
C-J-Energy/unit
volume
7.40E+06 (kJ/m3 ) 1.4
C-J Pressure 1.990E+07 (kPa ) e 2.068e5 kJ/kg
Stand
–off
Dist.
(m)
AUTODYN Simulation at different grid size Experimental
Results Grid
(mm2)
No. of
elements
Grid
(mm2)
No. of
elements
Grid
(mm2)
No. of
elements
4 x 4 41,250 3 x 3 80,000 1 x 1 353,000
Ta(ms) P(psi) Ta(ms) P(psi) Ta(ms) P(psi) Ta(ms) P(psi)
0.5 0.28 179.34 0.27 134.505 0.25 138.18 0.23 143
0.7 0.45 101.43 0.47 57.036 0.46 60.858 0.45 62
0.8 0.56 72.03 0.63 44.688 0.6 42.63 0.57 47.3
1.1 1.00 32.046 1.14 23.226 1.11 22.785 0.99 18.8
41
A comparison of experimental and simulated results for overpressure and impulse
values is depicted in Figure 2-29 and Figure 2-30 respectively. The simulated results
for peak overpressures and arrival time were in close agreement to experimental
results; however, a discrepancy in impulse is observed.
a b
Figure 2-27: (a) ANSYS AUTODYN model (b) expansion of
detonation product gases after detonation of 250g C4
Figure 2-28: (a) Formation of blast wave and (b) propagation in air
a b
42
2.7 Results and Discussion
The microscopic study of shaving foam revealed evenly distributed bubbles with an
average initial size of 15 m. The coarsening due to coalescence of bubbles slowed
down with time. The bubble size increased to 120 m in 2 hours showing good
stability. The viscosity of shaving foam measured at shear rate of 24.4 S-1
was 11650
Figure 2-30: Experimental and simulation results for 82g C4
bare blast, incident overpressure at 0.65 m and 0.85 m
Figure 2-29: Experimental and simulation results for 82g C4 bare blast, (a)
Incident overpressure (b) Impulse at 0.5m and 0.7m from charge center
a
b
43
cP. The two-phase medium exhibited a decreasing viscosity trend at higher shear
rates. The density and average sound speed measured in shaving foam at temperature
of 240C was 60.0 kg/m
3 and 65.98±3 m/s respectively. The measured values were in
good agreement to the values calculated using mixture law or Wood’s formula [6].
Several experiments were performed with bare C4 charges and covered in shaving
foam. A computer program written in Python was used to plot the pressure-time
history of the transducer’s data. Impulse was computed by integrating the pressure-
time plot for positive phase duration. The blast wave parameters obtained for bare
charges were compared with Kingery-Bulmash (KB) blast parameter calculator [21,
22]. Results are presented in Table 2-4.
Table 2-4: Comparison of bare charge tests with Kingrey-Bulmash (KB) results
A discrepancy in measured peak pressure and impulse values with the KB
calculations was observed. The reason for this deviation is the geometry of the
charges and the relatively lower density (1.46 g/cm3 in contrary to 1.6 g/cm
3) of C4
explosive used. Cylindrical shape (L/D1) of C4 was used in the present work
contrary to the hemi-spherical shape assumed for KB calculations. Moreover, tests
were conducted by placing the explosive on a 4.0 mm thick plywood sheet that was
770 mm above the ground level. This setup has not provided the same magnitude of
reflection as obtained for hemi-spherical surface burst charges. The higher pressures
generated in surface burst charges lead to higher values of positive phase duration
and consequently the higher impulse values.
C4
mass
Dist. Blast Parameters for bare charges
Present Experimental work Kingrey-Bulmash (KB)
(g) (m) Ta(ms) Ps(psi) Pr(psi) Impulse
(psi-ms)
Ta(ms) Ps(psi) Impulse
(psi-ms)
82 0.50 0.23±0.01 143±4 -- 7.93±0.2 0.25 173.4 15.4
0.70 0.47±0.02 61±1.5 -- 4.71±0.12 0.46 81.73 11.8
0.80 0.57±0.03 -- 185±9 18.87±0.45 0.60 58.8 10.4
0.85 0.73±0.03 46.6±1.3 -- 4.1±0.11 0.67 51.45 9.9
1.1 0.98±0.04 -- 56±2.8 -- 1.08 28.95 7.7
250
0.8 0.41±0.02 133±3.6 -- 6.5±0.16 0.43 141 21.5
0.9 0.56±0.02 95±2.7 -- 4.76±0.12 0.53 108 19.7
1.0 0.62±0.03 81±2.1 -- 6.2±0.15 0.65 85 17.8
44
2.7.1 Fireball and Afterburning Suppression
The suppression of fireball and the attenuation of highly pressurized product gases is
important for mitigation purposes. The dry aqueous foam, when placed on charge,
appears to serve this purpose [23]. Generally, a fireball is related to hot products
which emit light when sufficiently hot [8]. C4 (C3.86H7.57N5.22O5.32) has an oxygen
balance of −46.6%. The detonation products contain a lot of C(solid), CO, H2 etc.
During expansion some products are further oxidized (e.g. C and CO to CO2, H2 to
H2O, etc.) producing additional heat. This process is called afterburning reactions.
Total heat of combustion is, thus, the sum of detonation heat and afterburning heat.
For C4 the heat of detonation is only 44% of the total combustion heat [9]. EXPLO5
calculations for C4, assuming all combustible products (Cs, CO, H2…) are fully
oxidized, yielded the following results [8];
Heat of detonation 5.137 MJ/kg
Afterburning heat 6.145 MJ/k
Total heat of combustion 11.282 MJ/kg
The experimentally measured value of heat of detonation for C4 is 5.1±0.2 MJ/kg
[9]. It is worth mentioning here that the latent heat of vaporization for water is 2.25
MJ/kg and its specific heat capacity at constant pressure is 4.187 kJ/kg-K. That is
why water is able to absorb the energy of detonation (and combustion) of high-
energy explosives, provided it is aerosolized [24]. The even distribution of thin liquid
films covering micron sized gas bubbles in shaving foam readily evaporate under C4
detonation conditions and reduced the temperature to a level that suppressed fireball
and the afterburning reaction.
Fireball was measured with the help of high-speed images and is shown in Figure
2-31 - Figure 2-33. The detonation of bare 82 g C4 produced an initial fireball radius
of 0.8 m as shown in Figure 2-31(a, b). The afterburning in this fireball resulted in
formation of the secondary fireball with radius 1.1 m. The sequence of events with
increasing time is shown in Figure 2-31(c - f).
45
For C4 covered in shaving foam, the fireball appeared below the stand as shown in
Figure 2-32(a), while mainly suppressed by the shaving foam above the plywood
stand. The expansion of carbonaceous soot is more pronounced in Figure 2-32(b, c).
Fireball for bare 250 g C4 could not be measured accurately as it exceeded the
window diameter of 2.0 m as evident from Figure 2-33(b, c). The actual radius
was greater than 1.0 m. The fireball measurement and suppression for 250 g C4
covered in shaving foam is shown in Figure 2-33(d, e, f). The emergence of fireball
and partial suppression indicates that the volume of shaving foam for this charge was
insufficient.
a b c
f d e
Figure 2-31: Fireball for bare 82 g C4 (a, b) and formation of secondary fireball (c, d, e, f)
b
c
a
-0.25m
-0.76m
Figure 2-32: (a) Fireball formation and (b, c) quenching for 82 g C4 covered
in shaving foam
46
The presence of much higher concentration of carbonaceous soot (C) in high-speed
images shown in Figure 2-32(a, b, c) is an evidence of the suppression of
afterburning reaction and, hence, the reduction of the fireball radius. Almost 80%
reduction in fireball was observed with this shaving foam. The measured data is also
presented in Table 2-5.
Table 2-5: Fireball radius for C4 detonation in air and with shaving foam
C4 (g) Fireball Radius (m)
In air With shaving foam
82 1.1±0.01m 0.125±0.01m
250 >1m 0.265±0.01m
2.7.2 Peak Pressure and Impulse Reduction
The shaving foam confinement slows down the shock propagation due to high
compressibility of gas bubbles in aqueous foam. The bubbles readily accommodate the
change in volume caused by the pressure [23]. This deformation of gas bubbles and the
final bursting of the thin liquid films dissipate energy and cause blast wave mitigation.
The reduction in pressure is also attributed to lower sound speed of this two-phase
medium. The presence of air between the transducers and shaving foam, as shown in the
experimental setups, also plays a role in mitigation. An impedance mismatch at air-
shaving foam interface attenuates the overpressure and impulse [25]. The impedance is
the resistance offered by a material to transmission of shock and is the product of initial
density (0) and shock velocity (Us) [26]. Stated simply, the shock impedance is
-= 2.0m-
a
-=2.0m-
b
-= 2.0m-
c
d
--0.53m-
f
--0.53m-
e
Figure 2-33: Fireball for 250 g C4 (a, b, c) bare C4 and (d, e, f) covered in shaving foam
47
analogous to acoustic impedance. The numerical values of acoustic impedances (C00)
of shaving foam and air are 3904 and 419 respectively, showing a difference of an order
of magnitude at the interface.
The overpressure and impulse reduction for 82 g C4 charge at a distance of 0.5 m is
depicted in Figure 2-34. Peak pressure reduction of 68 – 74 % is recorded for 82g C4
tests at 0.5, 0.7 and 0.8 m distances. An average impulse reduction of 63% is recorded,
corresponding to these overpressure values. The 250 g C4 produces stronger effects and
requires more shaving foam confinement (foam radius> 0.28 m). A shaving foam
confinement of radius 0.28 m with uniform water spray of 300 ml was used for 250 g
C4. The incident overpressure-time history for this charge is shown in Figure 2-35.
Figure 2-34: Incident pressure and Impulse plots for 82g C4 blast at 0.5 m-
for bare charge and submerged in shaving foam
Figure 2-35: (a) Peak incident pressure and (b) impulse plots for 250 g Bare C4
and covered in shaving foam at 0.9 m
a
b
48
The Figure 2-35 indicates a pressure reduction of 80% for the above stated setup. The
test was then repeated without external water spray with shaving foam and blast
parameters were measured at 0.8 m distance. The incident and reflected peak
pressure plot and the corresponding impulse plot are shown in Figure 2-36. The
reflected pressure plots are overlapping the incident overpressure plots. Peak
reflected pressure reduced by 74% whereas the reduction in peak incident pressure
was 68%. Impulse reduction of 60% was observed. The additional water or higher
liquid content available in confinement slightly enhanced the blast mitigation. That is
why wet aqueous foams provide better attenuation against blast waves. Figure 2-37
depicts the variation of peak pressure with distance for bare and shaving foam-
covered C4.
Figure 2-36: Incident and reflected pressure plot (left) and impulse
plot (right) for 250g C4 at 0.8m
a
b
49
Figure 2-37: Peak pressure and distance plots with and without shaving
foam for (a) 82 g C4 (b) 250 g C4
a
b
50
References
[1] P.E. Del, C. A., H. A., D. L., EPJ Web of Conferences, (2011).
[2] W.F. Hartman, M.E. Larsen, B.A. Boughton, Blast mitigation capabilities of
aqueous foam(2006).
[3] T.D. Panczak, H. Krier, P.B. Butler, Journal of hazardous materials. 14(1987)
321.
[4] S. Daniel, DENIM Shaving Foam-Orginal(2020).
[5] Speed of sound. [cited 2020 12-25-2020]; Available from:
https://en.wikipedia.org/wiki/Speed_of_sound.
[6] A. Wood, Bell and Sons, Ltd, (1949) 361.
[7] A. Kapila, R. Menikoff, J. Bdzil, S. Son, D.S. Stewart, Physics of fluids.
13(2001) 3002.
[8] M. Suceska, Personal communication(March, 2020).
[9] L.S. Lebel, P. Brousseau, L. Erhardt, W.S. Andrews, Combustion and flame.
161(2014) 1038.
[10] P.W. Cooper, Explosives engineering. (2018.
[11] M. Sućeska. Calculation of detonation parameters by EXPLO5 computer
program. in Materials Science Forum. 2004. Trans Tech Publ.
[12] Kistler- Pressure Transducers. [cited 2020 17-12-2020]; Available from:
https://www.kistler.com/en/products/components/pressure-
sensors/?pfv_metrics=metric.
[13] E. Del Prete, A. Chinnayya, L. Domergue, A. Hadjadj, J.-F. Haas, Shock
waves. 23(2013) 39.
[14] K. Kann, A. Kislitsyn, Colloid Journal. 65(2003) 26.
[15] K. Kann, A. Kislitsyn, Colloid Journal. 65(2003) 31.
[16] U. ANSYS Inc., Century Dynamics. Release 14.0 documentation for ANSYS
AUTODYN.(2011).
[17] E. Lee, H. Hornig, J. Kury, Adiabatic expansion of high explosive detonation
products(1968).
[18] A. Bushman, I. Lomonosov, K. Khishchenko, V. Kogan, P. Levashov, I.
Lomov, Available on line: http://www. ficp. ac. ru/rusbank, (2004).
[19] H. Draganić, D. Varevac, Shock and Vibration. 2018(2018).
[20] C. Shah. Mesh discretization error and criteria for accuracy of finite element
solutions. in Ansys users conference. 2002.
51
[21] D. Hyde, US Army Engineer Waterways Experiment Station, USA. 2(1991).
[22] C.N. Kingery, Bulmash, G., Airblast parameters from TNT spherical air burst
and hemispherical surface burst(1984).
[23] M. Liverts, Shock Wave Interaction with Aqueous Particulate Foams(2012).
[24] M. Grujicic, B. Pandurangan, C. Zhao, B. Cheeseman, A computational
investigation of various water-induced explosion mitigation
mechanisms(2007).
[25] A. Britan, H. Shapiro, M. Liverts, G. Ben-Dor, A. Chinnayya, A. Hadjadj,
Shock Waves. 23(2013) 5.
[26] X. Kong, W. Wu, J. Li, F. Liu, P. Chen, Y. Li, Materials & Design. 51(2013)
729.
52
Characterization of Shell Chapter 3
Fragmentation
3.1 Introduction
The detonation of an explosive device results in the formation of a blast wave in air.
When an explosive is encased, the detonation energy and momentum are partitioned
into formation of fragments and the blast wave [1]. Fragmentation is the breaking of
shell body into a number of pieces. It is a complex phenomenon where the
fragmenting material fractures under intense shock wave loading produced by the
detonation of high explosive. The fragmentation process starts with the radial
expansion of the casing material. The outer surface fractures develop into cracks and
grow to the inner casing surface. The highly pressurized explosive product gases
begin to flow through the cracks, causing massive venting. The casing expands to 50-
60 % of the initial diameter [1]. Once the metal casing starts to break, the highly
pressurized product gases escape, resulting in the formation of a blast wave [2]. The
blast wave propagates outwards and leads the fragments. The neighboring objects are
first hit by this blast wave followed by the high velocity fragments, causing severe
damage. The highly accelerated fragments have sufficient energy to penetrate hard
targets and damage vulnerable components. The understanding of the fragmentation
phenomenon is important to devise protective measures against the damaging effects
[2, 3]. The fragment velocity, shape and mass distribution are important parameters
in the characterization of the fragmentation process [2]. Mott and Linfoot [4],
presented the relationship for fragment mass distribution given in eq. (3.1);
𝑁(> 𝑚) = 𝑁0exp [−(𝑚 𝛽)⁄ ] (3.1)
Where m is the fragment mass, N(>m) is the number of fragments with mass higher
than m, No is total number of fragments and is fragment size variable [1]. The
fragment mass distribution and velocity vectors are important to the assessment of
the lethal radius of a munition. The lethal radius is evaluated based on the velocity of
fragments and average fragment mass [5]. Gurney [6] proposed a relation for
estimating initial fragment velocity for cylindrical casing exploded under energetic
53
filling. The relation, given in eq. (3.2), is used for the computation of the fragment
velocity of a cylindrical shell.
𝑉 = √2𝐸 √
𝐶𝑀
1 + 0.5 𝐶𝑀
(3.2)
Where C and M are explosive and metal mass per unit length and √2𝐸 is the Gurney
constant for the explosive in km/s. This constant can also be approximated as;
√2𝐸 = 0.338𝐷, where D is the detonation velocity [7]. Huang et al [8] proposed a
relationship for initial fragment velocity calculations along the axis of cylindrical
casing by incorporating the influence of rarefaction waves at the ends. The velocity
relation shown in eq. (3.3) is given by;
𝑉 = (1 − 0.361𝑒1.111𝑥 𝑑⁄ )(1 − 0.192𝑒3.03(𝐿−𝑥) 𝑑⁄ ). √2𝐸 √𝐶
𝑀
1+0.5 𝐶
𝑀
(3.3)
Where x is the distance to the detonation end along the axis of cylindrical casing, d is
the explosive diameter and L is length of the casing.
3.1.1 Effects of Fragmentation
Fragmentation is the most lethal among cased explosive detonation effects because
fragments can fly large distances and cause serious injuries to humans [9]. Injuries to
personnel due to fragment impact can be classified into primary fragment and
secondary fragment injuries. Primary fragments are usually small, having high-
velocities that cause injury by penetration and perforation of vital organs of the body.
Secondary fragments are usually larger and have less velocity. Upon impact these
can cause non-penetrating blunt force trauma. Therefore, it is very difficult to
effectively deal with this mechanism. It necessitates a higher mass solution or an
expensive lightweight ballistic protection technique [3, 10].
3.2 Characterization
The fragment initial velocity, shape and mass distribution are important parameters
in the characterization of the fragmentation process. For this purpose, a scaled down
artillery shell (155mm) was considered. The field testing and required data
acquisition for a standard shell is a difficult task and requires significant resources.
54
Therefore, a geometrically scaled down (1:4) model of this shell was selected to
investigate its blast and fragmentation effects on the surrounding. The shell, as
shown in Figure 3-1, comprised a hollow steel casing filled with 104 grams of Comp-
B explosive. The Comp-B filling is illustrated in the cut-view of Figure 3-1(b). The
other parameters of shell along with parameters of a standard 155 mm shell are
presented in Table 3-1.
Table 3-1: Material and dimensions of scaled down and standard 155mm shell
Mass
(g)
Length
(mm)
Cyl.
OD(mm)
ID (mm)
Scaled down
shell
Steel casing 450 140 37 28
Comp-B 104 127 28 --
Standard
155mm Shell
Steel casing 35700 630 154 132
Comp-B 6800 535 133 --
3.2.1 Experimental Work
Three experiments were conducted with this scaled down (1:4) model to study the
blast and fragmentation effects. Pressure transducers were used with Data
Acquisition (DAQ) system to measure peak overpressure (Ps) and arrival time (Ta) of
the blast wave at different distances from the shell center. The experimental layout is
shown in Figure 3-2.
Figure 3-1: (a) scaled down shells (b) its cut-view (c) standard 155mm shell
b c a
55
The destructive potential of fragments is a function of their kinetic energy distribution.
Therefore, the initial velocity and mass distribution of the fragments need to be
determined [9]. Fiberglass and plywood sheets were employed to witness the fragment
impact. Flat Brass probes were used to measure the fragment velocities. The following
sections describe the characterization of the shell fragmentation.
Flat Brass timing probes, shown in Figure 3-3, were used to measure the fragment’s
velocity and arrival time at predefined distances. The probe comprised two Brass
foils, measuring 125 x 125 and thickness 0.1 mm, separated by a 0.1 mm thick Mylar
sheet. The Mylar sheet has enough strength to provide sufficient insulation. A cable
was used to connect the two Brass foils with a recording device like an oscilloscope
or a transient recorder. A magnified section view of probes is shown in Figure 3-3(c).
Upon impact and penetration by a metallic fragment, the electrical path was
completed momentarily and the event was registered by a recording device. Two
such probes, when placed in parallel and separated by fixed distance (S), as shown
in Figure 3-3(d), can record the fragment’s impact time. The fragment’s velocity was
computed with the help of the measured time interval and known separation distance
of the probes. The probes were fixed rigidly to avoid any disturbance by the
approaching blast wave.
Timing Probes
Fiberglass
Plywood sheets
Tran
sdu
cers
Shell
Figure 3-2: Layout for blast and fragmentation tests of scaled
down shell
56
Fragment velocities from different locations of the shell were measured using these
timing probes. The brass probes were also placed just below the shell base to find the
fragments velocity and number of fragments from this part of the shell. The setup is
shown in Figure 3-4.
The spatial distribution of the fragments was determined with the help of holes in
fiberglass and plywood witness plates penetrated and perforated by fragments. Two
fiberglass sheets measuring 457 x 457 x 10 mm3 and spanning 25.8
0 both in azimuth
and in elevation were placed one meter apart from the center of the scaled shell for
test 1. The testing setup is shown in Figure 3-5(a). For the other two experiments,
plywood sheets measuring 600 x 600 x 8 mm3 as shown in Figure 3-5(b) were placed
at 700 mm from the shell. The plywood sheets spanned 46o in azimuth and 44
o in
elevation. A layout of the experimental setup is shown in Figure 3-6.
a b
Brass
Insulator
Brass
c
S
d
Figure 3-3: (a) Flat Brass probe (b) 3-D view(c) section view (d) Two probes setup
a b
Figure 3-4: Flat probes arrangement for fragment velocity
measurement (a) test-1 (b) test-2
57
F/G
Sheets
Timing
Probe
s
Shell
a
Plywood
Sheet
b
Figure 3-5: Testing setup with (a) fiberglass (F/G) sheets and timing
probes (b) plywood sheet
Figure 3-7 and Figure 3-8 show the impact and perforation of fragments through
fiberglass and plywood witness sheets which were used to estimate the total number
of fragments and their spatial distribution.
Plywood
sheets
Shell
Brass sheet
a b
Figure 3-6: Setup for fragments impact and spatial
distribution (a) Test-2 and (b) Test-3
a b c
Figure 3-7: Fiberglass witness sheets (a, b) before and (c) after fragments impacts
58
The majority of the fragments, especially small ones, could not be recovered. A few
of the recovered fragments are shown in Figure 3-9. The fragments’ mass, size and
velocity distribution is presented in the results and discussion section.
3.3 Numerical Simulation
ANSYS AUTODYN [11] was used to simulate the fragmentation phenomenon of the
scaled down shell. Smoothed Particle Hydrodynamics (SPH) solver was used to
simulate explosively driven fragmentation. Being a mesh-free method, SPH can
handle nonlinear problems with large deformation without mesh degeneration or
tangling. Unlike Lagrange solver non-physical numerical erosion model is not
required [12]. Euler Multi-Material (MM) solver coupled with ALE was used to
simulate the blast phenomenon of the exploded shell in air.
a b
Figure 3-8: Plywood witness sheets (a) before and (b) after the fragments impact
Figure 3-9: Fragments recovered in the tests
59
3.3.1 Material Modeling
JWL (Jones-Wilkins-Lee) [13] equation of state (EOS) was used for expansion of
Comp-B product gases. The JWL parameters for Comp-B were used from ANSYS
AUTODYN material library. Ideal gas EOS was used for air surrounding the shell.
Steel AISI-1006 was used as the casing material. The Shock EOS (Mie-Grüneisen
form) [14] was used as the equation of state model for Steel-1006. This EOS, shown
in equation (3.4), is widely used for materials under shock loading.
𝑃 = 𝑃𝐻 + 𝜌(𝐸 − 𝐸𝐻) (3.4)
The Johnson–Cook strength model [15] was used to simulate the behavior of the steel
(AISI-1006) shell under high strain rate loading of explosive detonation. The model
presented in equation (3.5) reproduce very well the strain hardening, strain rate and
thermal softening effects of steel casing subjected to such high strain rate loading.
𝜎𝑦 = [𝐴 + 𝐵𝜀𝑝𝑛][1 + 𝐶 𝑙𝑛 𝜀𝑝
∗][1 − 𝑇𝐻𝑚] (3.5)
Where A, B, C, n and m are constants for each material and TH is homologous
temperature. The material model parameters for steel, Comp-B and air were used
from ANSYS AUTODYN library. The Johnson-Cook failure model [16] shown in
equation (3.6) was used along with the strength model for casing material (AISI-
1006).
𝜀𝑓 = [𝐷1 + 𝐷2 𝑒𝐷3𝜎∗ ][1 + 𝐷4 𝑙𝑛|𝜀∗|][1 + 𝐷5𝑇𝐻] (3.6)
The values of constants D1 to D5 for steel 1006 are presented in Table 3-2.
Table 3-2: Johnson Cook damage parameters for Steel-1006 [17]
D1 D2 D3 D4 D5
-0.8 2.1 0.5 0.0002 0.61
3.3.2 Fragmentation of Shell
The computational grid influences the accuracy and reliability of the numerical
prediction results [18]. A particle size of 1 was used for packing after optimizing the
size for SPH solver [3, 19]. Quarter geometry of the shell was modeled with 28820
nodal points to reduce the computational time. The shell model and gauge points
defined on steel casing are shown in Figure 3-10.
60
ANSYS AUTODYN generates fragment analysis when material status plot option is
activated or checked in during the simulation. When fragmentation option is checked,
it can output fragment analysis in HTML format. The analysis comprised number of
fragments alongside mass, kinetic energy, momentum, length, origin, coordinates and
velocity of each fragment. The fragmentation of shell casing at different time and the
venting of pressurized detonation product gasses are shown in Figure 3-11and Figure
3-12. The escape of product gases leads to the formation of the blast wave which
surpasses the fragments.
Figure 3-10: SPH Model of scaled down shell with gauge points
b a c
Product Gases
Steel Fragmenting
Figure 3-11: Fragmentation process (a) at 27 s (b) at 48 s (c) venting of
product gases
61
The number of fragments and their mass distribution is schematized in Figure 3-13.
Fragments with mass ranging from tens of milligrams to a few grams were produced.
The fragments’ velocity profile is shown in Figure 3-14. The fragment mass
distribution and velocity are important to evaluate the hit density and lethal radius
calculations [5].
0
25
50
75
100
125
150
175
200
No
. of
Frag
men
ts
Mass (g)
No. and mass distribution of fragments scaled down shell)
Figure 3-13: Number of fragments and mass distribution
Figure 3-12: Fragmentation process and radial expansion with time (a) at 57 s (b) at
80 s and (c) 150s (d) at 200 s (e) at 250 s
a b c
d e
62
The velocity distribution with the number of fragments is shown in Figure 3-15. The
majority of the fragments exhibited velocities ranging from 1000 - 1400 m/s. Only a
few fragments have velocities below 800 m/s.
3.3.3 Blast Parameters for Shell Detonation
Simulation for blast parameters was performed using coupled ALE-Euler Multi-
Material (MM) approach in ANSYS AUTODYN. The hollow steel shell was
modeled in ALE whereas Comp-B filling and surrounding air were modelled in Euler
MM solver. The ALE mesh was embedded inside the Euler MM fixed mesh. The
interaction between the two solvers was controlled by an automatic coupling option
available in ANSYS AUTODYN. An optimized grid size of 1x1 mm2 was used for
both solvers [3, 19].
a b
Figure 3-14: Fragment velocities of gauge points defined on shell
casing (a) with ALE solver (b) with SPH
Figure 3-15: Number of fragments and their velocity distribution
63
The air surrounding the shell was filled for ambient pressure of 101 kPa (14.7 psi) by
assigning specific internal energy of 2.068e+5 kJ/kg. Flow out boundary condition
was used at the Euler subgrid boundaries. The numerical model of the shell,
detonation propagation and expansion were shown in Figure 3-16. The Comp-B
filling was completely detonated in 15.8 s. The expansion and fracturing of the
casing material lead to initial escape of product gases and the subsequent formation
of a blast wave in air. The venting of product gasses at 40 s, subsequent blast wave
formation and its propagation in surrounding air is depicted in Figure 3-17. The
radial propagation of blast wave with time is shown in Figure 3-18.
a b c
Figure 3-16: (a) AUTODYN model of shell (b) detonation
wave propagation inside shell (c) Expansion of shell at 20 s
Gas
es v
enti
ng
a b c
Figure 3-17: (a) Venting of pressurized gases in air at t= 40 s and
expansion (b) at t= 61 s (c) at 90 s
a b c
Figure 3-18: (a) Blast wave propagation in air at t=0.150ms, (b) at t= 0.604ms
and (c) at t= 0.88ms
64
The pressure-time history at gauge points was integrated for positive phase duration
to obtain positive impulse. The peak incident pressure and positive impulse at
distances of 0.60 m and 0.76 m for simulated and experimental results are shown in
Figure 3-19. The blast arrival time and peak pressure values were in close agreement
with the experimental findings. However, a discrepancy in impulse was observed at
0.60 m distance.
3.4 Experimental Results
Three tests were conducted with a geometrically scaled down (1:4) model of the 155
mm artillery shell to study the blast and fragmentation characteristics. A simple
experimental approach was employed to measure the fragments’ velocity. The testing
setup with the timing probe for fragments’ velocity measurement is shown in Figure
3-20(a, b).
a
b
Figure 3-19: Simulated and experimental (a) peak Pressure and (b) Impulse plots
for scaled down shell
65
The fragments’ impacts on the timing probe can also be witnessed in Figure 3-20(c).
The impact time (t1) of the fragment on front probe was registered by an oscilloscope
and transient recorder. When this fragment perforated the second probe (placed
behind the front one), time (t2) was recorded. Fragment velocity was computed by
dividing the distance between the two probes (S) as shown in Figure 3-21(a) by the
time interval t. The penetration time through Brass probe is much smaller compared
to the total flight time of fragment and is therefore ignored for velocity calculations.
The fragment impact and arrival time at timing probes placed at 430 mm and 710
mm from shell center is shown in Figure 3-21(b) and Figure 3-21(c) respectively.
The impact time of the fragments produced from different locations of the shell
(base, cylindrical, conical etc.), as shown in Figure 3-22, was recorded with the help
of oscilloscope and transient recorder. The velocities were calculated from the known
distances (S) and recorded time intervals (t).
Timing
probes
a b c
Figure 3-20: (a, b) Fragment's velocity measurement from different parts of
the shell (c) timing probe after fragment impact
b c
t1
t2
a
s
Figure 3-21: : (a) Two probes setup (b) Fragment’s impact on
timing probe and (c) arrival time for velocity calculations
66
Fragment velocities ranging from 960 to 1555 m/s were measured for different parts
of the shell. Fragments from the cylindrical portion of the shell were found flying
with maximum velocities of 1346 to 1555 m/s due to the highest C/M ratio.
Fragments with relatively lesser velocities were produced from conical region on
account of a smaller C/M ratio. However, the minimum velocity of 960 m/s was
recorded for fragments produced from the base region of the shell as this was the
thickest metal portion. The fragments from this part of the shell were most likely to
be relatively massive. The fragment velocities were also computed from Gurney
relation [6] given in eq. (3.2 and Huang modification [8] shown in eq. (3.3). A
comparison is presented in Table 3-3.
Table 3-3: Comparison of experimental, numerical and analytical fragments velocities
The measured fragments velocities were found closer to Huang modified
formulation; however, Gurney relation predictions were also in close agreement with
other methods for the cylindrical part of the shell. The fragment velocities simulated
by ANSYS AUTODYN were in close agreement with the experimentally measured
Position Fragment velocity (m/s)
Frag. Velocities measured Experimentally Autodyn Gurney Huang
S1 mm
t1 ms
S2 mm
t2 ms
S t V=S/t m/s m/s m/s
mm ms m/s
Cylindrical 430 0.276 710 0.456 280 0.180 1555 1510 1340 1349
Cylindrical 500 0.379 675 0.509 175 0.130 1346 1350 1340 1349
Base 370 0.307 550 0.484 170 0.177 960 910 -- --
Conical 500 0.520 -- 500 0.520 961 1025 1123 948
Base
Cone Cylinder
1 2 3 5 6 7
11
Figure 3-22: Different positions of shell for fragment velocity calculations
67
values. The proposed method provides a fairly economical and simple approach to
measure fragment velocities accurately.
The total number of fragments produced could not be collected as the fragments
dispersed over a large area. Two fiberglass sheets each measuring 457 x 457 mm2
and spanning 25.80 both in azimuth and in elevation placed at a distance of 1.0 m
from the shell witnessed the fragment impact for test-1 as shown in Figure 3-23.
Plywood sheets measuring 600 x 600 x 8 mm3 and spanning 46
0 in azimuth and
elevation were used to witness the impact and penetration/perforation for the second
and third experiments. Fragments weighing tens of milligrams to a maximum of 6.3 g
were observed to impact the fiberglass/plywood sheets, as can be seen from Figure
3-24. Most of the small-sized fragments could not be recovered. A few of the
fragments recovered are shown in Figure 3-9 of the experimental work section. The
mass and size of recovered fragments are presented in Table 3-4. ANSYS
AUTODYN results are also tabulated for the corresponding recovered fragments.
a b c
Figure 3-24: (a, b) Plywood witness sheets (c) Timing probe
placed below the shell base
Figure 3-23: Fragment impacts and perforation through Fiberglass witness sheets
68
Table 3-4: Measured mass and size of the recovered fragments- comparison with simulation
results
Measured
Mass
(g)
Measured ANSYS AUTODYN
Length
(mm)
Width (mm) Length
(mm)
Mass (g)
3.9±0.1 29.8±0.7 9.4±0.2 46 4.07
3.69±0.1 48±1.2 6±0.1 40.52 4.10
3.0±0.1 37.8±0.9 5.4±0.1 31 2.63
1.5±0.05 21.9±0.5 6.7±0.1 15.35 1.46
0.9±0.04 10±0.2 8±0.2 11.94 0.35
0.9±0.04 12.6±0.3 4.5±0.1 12 0.52
0.6±0.03 10.5±0.2 4.8±0.1 14.7 0.56
0.3±0.02 13.8±0.3 5.4±0.1 9.96 0.30
0.2±0.01 7.3±0.2 4.9±0.1 8.8 0.31
0.4±0.02 7.8±0.2 4.2±0.1 10.33 0.33
Considering the shell geometry and assuming a symmetrical fragment distribution in
radial direction, as witnessed from the fiberglass and plywood sheets shown in Figure
3-23 and Figure 3-24, the total number of fragments was grouped into the following
three categories.
Small (< 0.04g)
Medium (0.04 - 0.4g)
Large (0.41 – 6.5g)
0
200
400
600
Small Medium Large
No
. of
Frag
me
nts
No. of Fragments and Mass Distribution
Experimental
Simulation
Figure 3-25: Number of fragments and their mass distribution for
scaled shell
69
A comparison of the number of fragments and their mass distribution is presented in
Figure 3-25. Most of the fragments have mass less than 0.5 g. The SPH simulation
results shown in Figure 3-25 provided a good agreement with the experimental
findings. A few of the recovered fragments of masses ranging between 0.2 – 3.9 g
were in fair agreement with the simulation results.
The peak overpressures (Ps) at 0.55, 0.59, 0.60, 0.65, 0.675 and 0.76 meter distances
were recorded. The pressure time history for experimental and simulated data at four
of the above stated distances are shown in Figure 3-26. Peak overpressure values of
44.2 psi (304.7 kPa) at 0.55 m and 23.38 psi (161.2 kPa) at 0.675 m were measured.
The corresponding time of arrival (Ta) was 0.505 ms and 0.733 ms respectively.
Incident impulse of 3.3 and 1.8 psi-ms were computed corresponding to these peak
overpressure values.
A summary of the experimental and simulated results is presented in Table 3-5. A
good agreement between simulated and experimental results for the blast parameters
was obtained for the scaled down model of 155 mm artillery shell. Therefore, one
a
b
Figure 3-26: : Experimental and simulated peak incident pressure plots at
(a) 0.55m & 0.65m (b) 0.59m & 0.675m
70
can utilize the simulation techniques to predict the blast and fragmentation effects of
such munitions and hence minimize the cost and time spent on full scale testing.
Table 3-5: Summary of the experimental & simulation Pressure-time values
Distance Experimental Results Simulation Results
R(mm) Ta(ms) Ps(psi) Ta(ms) P
s(psi)
550 0.505 44.2±1.1 0.507 43.08
590 0.573 36.54±0.9 0.57 36.22
600 0.603 35.5±0.9 0.59 34.87
650 0.668 26.5±0.6 0.675 28.85
675 0.733 23.38±0.5 0.720 26.43
760 0.882 16.79±0.4 0.887 18.94
71
References
[1] R.M. Lloyd, Reston, VA: American Institute of Aeronautics and
Astronautics, Inc.(Progress in Astronautics and Aeronautics. 179(1998).
[2] X. Kong, W. Wu, J. Li, F. Liu, P. Chen, Y. Li, Materials & Design. 51(2013)
729.
[3] K. Ahmed, A.Q. Malik, A. Hussain, I.R. Ahmad, I. Ahmad, AIP Advances.
10(2020) 095221.
[4] N. Mott, E. Linfoot, HA theory of fragmentation(1943).
[5] G. Tanapornraweekit, W. Kulsirikasem, International Journal of Mechanical
and Mechatronics Engineering. 6(2012) 1070.
[6] R.W. Gurney, The initial velocities of fragments from bombs, shell and
grenades(1943).
[7] B. Zecevic, J. Terzic, A. Catovic, S. Serdarevic-Kadic. Influencing
parameters on HE projectiles with natural fragmentation. in International
Conference on New Trends in Research of Energetic Materials. 2006.
[8] G.-y. Huang, W. Li, S.-s. Feng, International Journal of Impact Engineering.
76(2015) 20.
[9] E. Lozano, Design and analysis of a personnel blast shield for different
explosives applications(2016).
[10] K.J. Sharpe, J. Waddell, J.F. Gordon, Explosive effect mitigated containers
and enclosing devices(2008).
[11] U. ANSYS Inc., Century Dynamics. Release 14.0 documentation for ANSYS
AUTODYN.(2011).
[12] M.A. Abdalla. Fragmentation Analysis of OG-7 Warhead Using AUTODYN
SPH Solver. in Advanced Materials Research. 2012. Trans Tech Publ.
[13] E. Lee, H. Hornig, J. Kury, California: University of California, (1968).
[14] M.A. Meyers, Dynamic behavior of materials. (1994.
[15] G.R.a.C. Johnson, W.H. A Constitutive Model and Data for Metals Subjected
to Large Strains, High Strain Rates, and High Temperatures. in Proceedings
7th International Symposium on Ballistics. 1983. Hague.
[16] G.R. Johnson, T.J. Holmquist. An improved computational constitutive
model for brittle materials. in AIP Conference Proceedings. 1994. American
Institute of Physics.
72
[17] B. Adams, M. Geers, v. Dommelen, A. Huizinga, Simulation of ballistic
impacts on armored civil vehicles. (2006.
[18] Q. Li, L. Lu, T. Cai, AIP Advances. 10(2020) 095107.
[19] K. Ahmed, A.Q. Malik, AIP Advances. 10(2020) 065130.
73
Protective Configurations against Chapter 4
Fragments
4.1 Protective Mechanisms
Protective systems are usually comprised of high strength steel, ceramics, and
composite materials etc. to defeat impacting fragments/projectiles [1]. But armor
steel becomes too heavy while Ceramics are inherently brittle and may cause
secondary fragmentation unless treated properly. The development of a protective
system depends on the type of threats to be defeated. However, several lightweight
materials are available for use in a protective configuration. Ceramics, fibre-
reinforced polymers (FRP) and porous materials are a few that were selected for this
study. These materials offer high strength at much lower densities. In present study,
the protective mechanism was configured to undermine the momentum of incoming
threat (projectile/fragment) and subsequent dispersion and absorption of the energy
produced in the form of shock waves.
4.1.1 Momentum Disruption
The explosively driven fragments/projectiles (bullets etc.) can move at very high
velocities. However, unlike bullets, fragments do not have regular shapes required
for deep penetration/perforation of hard targets. The momentum of these fragments
can be disrupted using a high strength material capable of resisting penetration from
compressive forces and spreading the momentum before it damages the target [2].
Ceramics like silicon carbide (SiC), alumina (Al2O3), boron carbide (B4C), etc. have
been used for the containment of blast and fragment penetration due to their high
hardness and erosion properties. These are highly brittle and are usually backed by
metal plates. The backup metal delays the initiation of tensile failure in the ceramic at
the ceramic-backing plate interface and thus allowing more projectile erosion. The
blunting of nose and projectile erosion are the dominant factors in the reduction of
projectile penetration. The fracturing and pulverization of ceramics are also effective
ways to dissipate a part of the kinetic energy generated by the impacting projectile.
The response of a brittle ceramic tile (Al2O3) to bullet impact is shown in Figure 4-1.
The plastic flow of the hard ceramic fragments around the projectile erodes the tip or
74
even the entire length of the projectile, which further dissipating energy and
increasing the impact area [3]. Alumina tiles measuring 170 x 170 mm2, shown in
Figure 4-2, having hardness greater than 75 HRC were used in present study. The
tiles were fabricated by iso-statically pressing the alumina powder and sintering at
1600oC.
4.1.2 Shock Wave Spreading
Orthotropic materials offer anisotropy in materials along different directions.
Therefore, sound speed along the length of fibers is found to be several times higher
than its value in other directions. This property was used to spread the impacting
shock laterally, thereby reducing the shock strength along the impact direction [4].
After the early-stage disruption by the momentum disruption mechanism of the
ceramic tile, the shock wave transmitted into the protective material was further
attenuated by dispersion in the lateral direction. This layer of fabric can also capture
the secondary fragmentation of the brittle ceramic. In present work, laminated GFRP
and the high-strength aramid fabric Kevlar were used for this purpose.
Figure 4-1: (a) 7.62 mm bullet impact. (b) Pulverization of alumina disk. (c) Fragmentation
and scattering of alumina
Figure 4-2 : Alumina tiles used in present work
17
0 m
m
75
4.1.3 Shock Energy Absorption
After disrupting the fragment momentum and dispering the major portion of the
transmitted shock in the lateral direction, a third layer of a porous material is
introduced in some of the configurations. The remaining shock wave energy can be
absorbed by porous materials like polystyrene (PS) and polyurethane PU) foams. The
porous materials (PE, PU, Al foams etc.) convert shock energy into heat or work
done, thus attenuating the shock energy. These materials absorb the energy through
the pore collapse mechanism [5]. A stress-strain curve of a porous material, as shown
in Figure 4-3, illustrates the energy absorption mechanism [6]. PU/PS foams were
used in present work.
Figure 4-3 : Stress-strain relation of porous material [6]
4.2 Protection against Fragments Impact
Fragmentation is the most lethal effect of EM detonation because fragments can
travel to large distances and cause serious injuries to humans. It is also one of the
dominant threats to vulnerable components, light structures and even hard targets.
IEDs and cased charges are often used to generate this damaging effect. It is
imperative to describe the threat level for which a protective configuration is
developed and tested.
A combination of lightweight materials can be employed for protection against both
blast and fragment impact. Present work deals with the design, development and
testing of various lightweight protective configurations against blast and
fragmentation. To evaluate fragmentation loading and protection against
fragmentation, scaled down artillery shell was used in present work. Details of the
76
blast parameters and fragmentation characterization of this shell has been presented
in Chapter 3. The protective configurations were tested against reflected overpressure
of 235 psi (1620 kPa) and fragments weighing up to 4.3 g impacting with velocities
ranging from 961 to 1555 m/s. A steel plate measuring 250 x 250 x 6 mm3 was
placed at a distance of 0.60 m from the shell. The fragments’ impact, penetration and
perforation through the steel plate are evident from Figure 4-4. The protective
configurations were exposed to such extreme loading conditions.
4.2.1 Protective Configurations
Numerous lightweight materials have been used in a protective configuration. Kevlar
fabric has shown excellent ballistics impact resistance [7]. Kevlar woven fabric, 110
GSM (grams per square meter), was used with epoxy resin to maximize its
mechanical properties. Combinations of Kevlar fabrics, laminated GFRP, PU foam
and alumina tile were selected on the basis of their inherent properties of shock
dispersion, absorption and momentum disruption. In the present work the material’s
combination, order and dimensions were optimized using AUTODYN simulations
and their performance was then validated with experimental work. Two novel
compositions containing PU-alumina and PU-silica were also investigated. The seven
configurations were designated as C-1, C-2, C-3, C-4, C-5, C-6 and C-7. The
dimensions and constituent materials for each configuration are summarized in Table
4-1. Configuration and its protective capability against the blast and fragments
impact are briefly described in the following section of experimental work.
a b
Figure 4-4: Fragments penetration and perforation (a) Front side
(b) back sides of the 6 mm thick steel plate
77
Table 4-1: Constituent materials and dimensions of protective configurations
Sr. # Alumina tile GFRP PU Foam Kev-epoxy Size (mm) Mass (g)
C-1 --- 5mm thick, 6mm thick, 12 layers 470 GSM
285x210x24 1123
C-2 --- 5mm thick, 6mm thick, 12 layers 110 GSM
240x150x17 435
C-3 --- 5mm thick, 6mm thick, 12 layers 110 GSM
220x220x18 626
C-4 9mm thick, Al back-1mm
5mm thick,
---
6 layers 470 GSM
165x165x21 1500
C-5 10mm thick, Al back-1mm
5mm thick,
5mm thick,
Size- 7˝ x 7˝ 12 layers 110 GSM
165x165x26 --
C-6 Multi-layers- 85% PU (4-layers) with 15% sand (3-layers) -212, 22 762
C-7 Mixture of 85% PU and 15% sand - Mixture of 85% PU and 15% Alumina powder
-212, 27 698
MS Plate 225x225x6 2396
4.2.2 Experimental Work
Two experiments were conducted with the scaled down shell (155 mm) to study the
blast and fragmentation effects on the protective configurations listed in Table 4-1.
The experimental setup for testing these configurations against the fragmentation of
this shell is shown in Figure 4-5.
C-1
Shell Flat probesss
Figure 4-5: Experimental setup for testing protective configurations against
fragments impact
78
C-1 comprised Kevlar woven fabric (470 GSM) wrapped on laminated GFRP and
PU foam sheets. The configuration weighed 1123 grams with areal density of 1.87
g/cm2. The assembled view of the specimen is shown in Figure 4-6.
A closer view of the testing setup for C-1 and C-3 is also shown in Figure 4-7. Both
configurations were placed at 0.65 m away from the shell. Out of 15 fragments that
impacted C-1, only 5 perforated the configuration. The majority of the high velocity
fragments were captured within the configuration.
The configuration was tested again with a second shell detonation under the same
conditions. The impact spots of the first test were marked with circles to distinguish
them from the fresh impacts. The test setup and post-test impact on the configuration
is shown in Figure 4-8. Eleven fragments impacted the configuration and only three
could perforate C-1.
a b
Figure 4-6: (a) Front side (b) back side of C-1
b c C-1 C-3
a
Figure 4-7: (a) Setup for testing C-1 and C-3. (b) Front (c) back sides of C-1
after fragment’s impact
79
C-2 comprised Kevlar woven fabric (110 GSM) wrapped on 5 mm thick laminated
GFRP and PU foam sheets. The configuration weighed 435 grams with an areal
density of 1.2 g/cm2. The assembled view of C-2 is shown in Figure 4-9.
C-2 and its testing setup are shown in Figure 4-10. Flat Brass probes are also visible
in Figure 4-10(a) for fragment velocity measurement. The Figure 4-11depicts the
post-test view of both sides of the configuration. A view of the laminated GFRP and
PU foam is also shown in this figure. Nine fragments impacted the configuration and
six were able to perforate the configuration. Being the lightest configurations, it
could not offer sufficient strength to capture most of the impacted fragments. As
expected the Kevlar-470 GSM has offered more resistance to fragments penetration
than equal number of layers of Kevlar-110 GSM.
a b c d
Figure 4-8: (a, b) Test-2 setup and C-2 view after (c) impacted and (d)
perforated fragments
a b
Figure 4-9: Assembled view of C-2, (a) front and (b) back sides
a b c
Figure 4-10: (a, b) C-2 testing setup (c) C-2 along with 6 mm thick MS plate
80
C-3 comprised a combination of Kevlar fabric (110 GSM) wrapped on 5 mm thick
laminated GFRP and PU foam. It weighed 626 g with an areal density of 1.29 g/cm2.
A view of C-3 before test is shown in Figure 4-12.
The assembled view of C-3 after testing is shown in Figure 4-13. The relatively
wider configuration placed at the nearest location of 0.50 m from the shell received
16 fragments out of which 8 perforated C-3. The second test of C-3 was conducted
by placing it at a distance of 0.80 m from the shell. The net fragments that perforated
C-3 in first and second test were 14. Figure 4-14 schematizes the tests results of C-3.
The configuration was able to capture almost 60% of the impacted fragments.
b a c d
Figure 4-11: C-2 (a) front (b) back sides (c) laminated GFRP (d) PU foam
after fragments impact
a b
Figure 4-12: (a) Front and (b) back sides of C-3
Figure 4-13: C-3 front and back sides after fragments impact- first test
81
C-4 comprised a combination of Kevlar fabric (470 GSM) wrapped on an alumina
tile and laminated GFRP sheet. The configuration weighed 1500 grams. The
assembled view and testing setup are shown in Figure 4-15. The Kevlar-470 GSM
was wrapped on alumina tile and laminated GFRP sheets. The configuration was
placed at 0.80 m from the shell. Being relatively farther and having relatively smaller
frontal area, only 4 fragments impacted and none of these could perforate the
configuration as evident from the back-side view of C-4 shown in Figure 4-16.
C-5 comprised a combination of alumina tile (10 mm) backed by laminated GFRP (5
mm) and PU foam (6 mm), all wrapped in 12 layers of Kevlar-epoxy. The front and
back side views of C-5 are shown in Figure 4-17. This configuration was also tested
a b c d
Figure 4-14: C-3 Second test (a) Front (b) back sides, (c, d) laminated GFRP
a b c
Figure 4-15: (a, b) Assembled view of C-4 and (c) testing setup for C-4 and C-5
a b c d
Figure 4-16: (a, b) C-4 front and (c, d) back sides after the test
82
against 7.62 mm bullet impact. The bullet impact is evident on front side of
configuration without a significant backface signature [8].
C-5 was located 0.725 m from the shell. Seven (7) fragments impacted the
configuration, as shown in Figure 4-18(a), and none of these could perforate this
configuration as seen from the back side view of the configuration shown in Figure
4-18(b). The Kev-epoxy covering was observed to contain the secondary
fragmentation of brittle alumina and retained the integrity of the configuration
against severe impacts of fragments.
C-6 comprised a multi-layered combination of 85% Polyurethane and 15% silica.
Three layers of silica were sandwiched in four layers of polyurethane. The 22 mm
thick sample weighed 762 grams and is shown in Figure 4-19(a, b).
C-7 comprised a combination of two composites. The first one was prepared by
mixing 15% (by weight) alumina powder in 85% Polyurethane. The 10 mm thick
sample disc weighed 321 grams. The second sample was prepared by mixing 15%
(by weight) silica with 85% polyurethane. The 17 mm thick sample weighed 367
grams. Both these samples are shown in Figure 4-19(c, d) and joined together with an
adhesive and cotton tap to test against fragments impact.
a b
Figure 4-17: C-5 (a) Front side (b) back side
b a
Figure 4-18: (a) Front and (b) back sides of C-5 after fragment's impact
83
The testing setup for C-6, C-7 and other samples is shown in Figure 4-20. C-1, C-4
and MS witness plate are also visible in this figure. C-6 was located at a stand-off
distance of 0.60 m. The post-test scenario is shown in Figure 4-21. Eleven (11) out of
eighteen (18) impacted fragments perforated C-7 as shown in Figure 4-21(c, d). The
composition did not offer significant resistance to impacting fragments. However,
these two configurations have shown good blast and thermal effects resistance
against the nearby detonation of the scaled shell. Scabbing from rear side is also
witnessed from this Figure 4-21(b, d). Flat timing probes were placed behind C-7 to
measure the exit velocity of the fragments. Fragments with velocity ranging between
1349–1555 m/s were supposed to impact the configuration due to its position. The
fragment’s exit velocity measured with flat Brass probes was 1038 m/s.
a b c d
Figure 4-19: (a, b) A view of C-6 and (c, d) C-7 configurations
Figure 4-20: Testing setup for C-6 and C-7
Figure 4-21: (a, b) front and back sides of C-6 after test, (c, d) front
and back sides of C-7 after test
a b c d
84
4.3 Protection against Bullet Impact
The protective configuration (C-5) consisting of alumina (Al2O3) tile, followed by
laminated GFRP and PU foam, and covered with 9 layers of Kev-epoxy was tested
against 7.62 × 39 mm (MSC) mild steel-core bullet impact. This bullet is fired from
popular AK-47 Kalashnikov assault rifle [9]. The details of the configurations are
also presented in Table 4-2. The bullet shape and its material properties play a vital
role in penetration. The bullet, weighing 7.77 grams, was cut into two sections and its
constituent materials (mild steel core, copper-coated steel jacket and lead filler), as
shown in
Figure 4-22, were inspected, weighed and measured. The measured values are
summarized in Table 4-3[8].
Table 4-2: Protective configuration for bullet impact tests [8]
Sr. # Alumina GFRP PU Foam Cladding Thickness
1 10 mm thick,
Size- 7˝ x 7˝
6mm thick,
Size- 7˝ x 7˝
6mm thick,
Size- 7˝ x 7˝
Kev-
epoxy 26 mm
2 Blunt Force Trauma test - depth of depression = 10 mm
Figure 4-22 : Different parts of 7.62x39 mm bullet, (left) steel core, lead filler, and copper
coated steel jacket. (Right) cut sections of bullet [8]
Table 4-3: Measured values of 7.62x39 mm MSC bullet
Material Mass (g) Hardness (HRC)
Mild steel-core (MSC) 3.87 17
Copper-coated steel jacket 1.82 16
Lead filler 2.08 --
85
The impact velocity measured using high-speed photography was 715 m/s. The bullet
was stopped and captured by the configuration. The target configuration retained its
integrity without any escape of secondary fragmentation from brittle alumina. The
Kevlar-epoxy resin provided enough strength and support to the target constituent
materials. Figure 4-23 illustrates the bullet impacting the target configuration C-5.
The ceramic tile stopped the bullet penetration through erosion and blunting of the
nose. No evidence of bullet penetration behind the ceramic tile was observed. The
recovered mushroom-shape bullet is shown in Figure 4-24.
4.3.1 Blunt Force Trauma Test
Soft body armor comprising higher molecular weight polyethylene (HMWPE) or
high strength fabrics like Kevlar are prone to back face signature (blunt force
trauma). Despite the soft body armor vest’s protection, the possibility exists of blunt
injury resulting from the impact [10]. Body armor is supposed to stop the projectile
and provide protection from internal injury resulting from the impact energy. Blunt
force trauma or backface signature is the result of the energy transferred from the
impacting projectile to the human body (shielded by a bullet resistant vest) without
penetrating the skin. The severity of such an injury may result in ruptured organs,
Figure 4-24: 7.62 x 39 mm MSC original and plastically deformed recovered bullet
Figure 4-23: High speed images of bullet impacting the target configuration
86
internal bleeding and ultimate death. This effect can be minimized by spreading the
impact of the projectile over a wide area of the torso.
The backface signatures (blunt force trauma) test of this configuration was performed
using Plastilina clay [11] as backing material in a wooden frame. Any external cover
or fabric was not used during this test. The configuration was placed inside this
frame, as shown in Figure 4-25, and tested against 7.62 x 39 mm MSC bullet impact
at a distance of four meters. High-speed photographic images of this test are shown
in Figure 4-26.
The bullet penetrated the configuration and was captured inside as seen from Figure
4-27(b). Any bulging effect was not observed on the backside of the Kevlar-epoxy
covered configuration, as shown in Figure 4-27(c). The impact of the bullet was
a b c d
Figure 4-27: (a) Testing setup (b) 7.62mm bullet impact (c), Rear side
of C-5 (d) and impact on Plastilina clay
Figure 4-25: The testing configuration, including Plastilina clay as backing medium
Figure 4-26: High speed photographic images of blunt force trauma test at
three different times
b c a
87
observed on the clay, as shown in Figure 4-27(d). The impact was spread out,
indicating the dispersion of the shock wave over a wide area, thus decreasing the
intensity as desired. The maximum depth of depression measured was 10 mm. This
was well within the European, German and British standards for backface signatures
which allow 20 to 25 mm ‘Backface Signature’ [8, 11].
It was observed that the front ceramic tile was able to completely defeat the
penetration of the bullet owing to its high hardness and eroding effects. The shock
energy transferred from the ceramic tile was well distributed by the laminated GFRP
in lateral directions. Finally, the porous layer of PU foam was able to absorb the rest
of the shock energy, as evident from the blunt force trauma test. The initial
momentum disruption with transmitted shock dispersion and absorption methodology
was found effective to counter a bullet/fragment impact threat. The fiber reinforced
cladding of Kevlar-epoxy was able to prevent the disintegration of the brittle alumina
tile, maintaining the integrity of the configuration.
4.4 Numerical Simulation
ANSYS AUTODYN [12] was used to simulate the fragmentation phenomenon of
scaled down artillery shell (155 mm). In a Lagrangian solver the grid moves with the
material and undergoes distortion causing inaccuracies in solution. This solver is
ideal for simulating materials undergoing low distortion. An erosion model is used
with this solver to avoid mesh tangling. Smoothed Particle Hydrodynamics (SPH)
solver was used to simulate explosively driven fragmentation of shell and the
response of protective configurations to impacting fragments. SPH being a meshfree
method can handle nonlinear problems with large deformation without mesh
degeneration or tangling unlike the Lagrange solver. The non-physical, numerical
erosion model is not required for SPH [13]. A coupled Euler-ALE approach was used
to simulate the blast effects. A coupled SPH-ALE approach was also used to simulate
the fragments impact on protective configurations.
4.4.1 Material Modeling
JWL (Jones-Wilkins-Lee) equation of state [14] was used for expansion of Comp-B
product gases. Shell body was modelled with (AISI-1006) steel. Ideal gas EOS was
used for air. Material models for these materials were used from ANSYS
88
AUTODYN material library which has already been discussed under the section on
material modeling in Chapter 3. Material properties for Alumina, GRFP laminate and
Kev-epoxy used for simulating protective configurations are presented in Table 4-4.
Alumina and Kev-epoxy were retrieved from AUTODYN library. Material
properties for GFRP, PU foam and Kevlar were used from [15], [2] and [16]
respectively.
Table 4-4: Material properties used in simulation
Al2O3-99.5 GFRP laminate [15] Kev-Epoxy
Density – 3.89 g/cm3 Density – 1.80 g/cm
3 Density – 1.29 g/cm
3
EOS - Polynomial EOS - Orthotropic EOS - Puff
A1 = 231 GPa E 11 (kPa) = 6.0e+6 Stiffness Matrix (kpa)
A2 = -160.0 GPa E 22 (kPa) = 1.97e+7 C11=3.42e+6,C22= 1.35e+7,
C33= 1.35e+7, C12=1.14e+6
C23=1.20e+6, C31=1.14e+6
A3 = 2774 GPa
B0=B1=0
E 33 (kPa) = 1.97e+7
T1= 231 Gpa Poisson ratio 12 = 0.15 Shear Modulus(Gpa)
G12=G23=G31=1.0
T2 = 0 Poisson ratio 23 = 0.13 Volum. response-Polynomial
A1=4.15Gpa, A2=40Gpa
,T1=4.15Gpa
T (Ref) = 0 Poisson ratio 13 =0.15
Strength – JH2 Strength – Elastic Strength – Elastic
G(Gpa) = 152 G(kpa) = 1.79e+6 G(kpa) = 1.0e+6
HEL=6.57GPa, A = 0.88, N
= 0.64, C= 0.007, B = 0.28,
M = 0.6
Max frac. Ratio =1.0
Failure-JH2 Failure-Stress/strain Failure-Mat Stress/strain
D1= 0.01, D2=0.7, β = 1
Hyd. Tensile limit =-2.620e-
1 Gpa
Ten fail stress 22(kpa)
4.318e+5
Max. shear stress 23 (kpa)
8.0e+4
Ten fail strain 11 = 0.009
Ten fail strain 22 = 0.02
Ten fail strain 33 = 0.02
Ten fail strain 11 = 0.01
Ten fail strain 22 = 0.08
Ten fail strain 33 = 0.08
89
4.4.2 Shell Fragmentation and Impact on Protective Configurations
SPH solver in ANSYS AUTODYN was used to model the scaled down shell for
fragmentation studies. Particle size of 1 was used for packing. The quarter symmetry
of the shell was modeled with 28,820 nodal points. The shell model, number of
fragments, fragments mass and velocity distribution have already been discussed in
the section on simulation in Chapter 3. The gauge points defined in the model and the
fragments velocity distribution will be discussed frequently in this section, therefore,
the FE model of the shell with gauge points and their velocity plot is shown in Figure
4-28. Present work is focused on the simulation of the behavior of various protective
configurations against fragment impact and the study of blast wave loading on these
configurations. The protective capability of the configurations is presented in this
section.
Figure 4-28: (a, b) FE model with gauge points defined and Fragment velocities of
gauge points defined on shell casing (c) with ALE solver (d) with SPH
c d
a b
90
4.4.3 Coupled SPH-ALE Simulation
The fragmentation phenomenon was modeled and simulated in SPH solver. A
coupled SPH-ALE approach was used to simulate the impact of explosively driven
fragments on protective configurations. ALE solver combines the best features of
both Lagrange and Euler methods. Due to symmetry on two axes, quarter symmetry
of the shell was modeled in SPH and the simulation was run until the fragments
traveled a radial distance of 560 mm. Protective configurations C-1 and C-5 were
then modeled using ALE solver. Kevlar 470 GSM layers were modeled with a
macro-homogeneous model that considers the whole layers as homogenous in
geometry with orthotropic mechanical properties [16].
A grid size of 50 x 25 x 3 was used for Kevlar 470 GSM layer measuring 200 x 100
x 4 mm3. A ratio of 4:1 between SPH particle and ALE cell size best defines the
interaction between the two solvers [17].
Figure 4-29 depicts the FE models of C-1(a, b)) and C-5 (c, d) using ALE solver.
Both grid and material plots of the configurations are shown in this figure.
Figure 4-29: FE models of (a, b) C-1 and (c, d) C-5
a b
c d
91
The fragments approaching both configurations located perpendicular to Y and Z-
axes are shown in Figure 4-30. Gauge points defined on steel casing approaching the
protective configurations are also visible in this figure. The fragments traveling in
radial direction and their interaction with C-1 and C-5 are also shown in Figure 4-31.
Both C-1 and C-5 were placed at a position where fragments from the cylindrical part
of the shell body were most likely to impact, as can be seen from Figure 4-30. The
alumina tile in C-5 offers high resistance to high velocity fragments. The laminated
GFRP backing provides strength to alumina by absorbing the shock energy and
further causing a delay in fracturing of the brittle alumina tile. The combination of
Kevlar-470 GSM and GFRP in C-1 resists the fragments penetration. The Kevlar
fabric undergoes deformation and absorbs the impacting energy. The fabric
deformation at the rear end of the configuration can be seen in Figure 4-31(a, b). The
results are consistent with the experimental testing of C-1 and C-3 discussed in
experimental section. C-5 captures all the impacting fragments; however, a bulge on
the rear side of the configuration is seen in the simulated results of Figure 4-31.
Figure 4-30: Isometric view of shell fragmentation with C-1 and C-5,
fragments impacting C-1 & C-5
C-1
C-5C-1
C-5
a b
c
92
The velocity profile of fragments defined on the cylindrical portion of the shell is
shown in Figure 4-32. The fragments impacting C-5 are captured within the
configuration. An abrupt drop in velocity profile can be seen as the gauge points hit
the protective configurations. Velocity plot of these gauge points is shown in Figure
4-32(a) for configuration C-5. The velocity profile of gauge points impacted on C-1
is shown in Figure 4-32(b). The velocity of gauge points 11 and 12 is approaching
200 m/s while gauge point 10 still has a velocity 400 m/s showing a potential to
perforate the remaining part of the configuration.
Figure 4-31: (a) Fragments penetration through C-1 (b) deformation of Kevlar woven
fabric and (c) Fragments impact on C-5 and bulge on back side of C-5
a b
c
Figure 4-32: (a) Velocity profile of gauge points before and after impacting
C-5 (b) before and after impacting C-1
a b
93
The fragment impact on a 6 mm thick MS (mild steel) plate was also simulated. The
simulation setup is shown in Figure 4-33. Most of the fragments were stopped by the
MS plate as also observed in experimental work shown in Figure 4-4. Fragments
penetration and perforation through MS plate is shown in Figure 4-34.
The simulation setup for C-4 and C-2 is shown in Figure 4-35. SPH model of the
shell with gauge points defined on shell casing are visible in this figure. The
fragment expansion in the radial direction along with the moving gauge point is also
shown in Figure 4-35.
Figure 4-34: Penetration and perforation through (a) MS plate.
(b) Close view of MS plate
a b
Figure 4-33: (a) Fragments impacting on MS plate (b) isometric view
a b
94
The protective model for C-4 in ALE is shown in Figure 4-36. An enlarged view of
the configuration and the approaching fragments can be seen from this figure. As
most of the high velocity fragments hit the central region of the configuration, a
graded zoning was employed in the central part of the layered protective
construction. The graded zoning is visible in the grid plots of Figure 4-36 for C-4.
Figure 4-35: (a) Shell model in SPH (b) radial expansion of fragments
and (c) moving gauges are visible
a
b c
95
Figure 4-36: (a) ALE model of C-4 (b, c) Fragments approaching the configurations
a b
c
Figure 4-37: Fragments impact on C-4, (a) front side view, (b, c) back side view
a b
c
96
The impact of fragments on C-4 is shown in Figure 4-37. Soon after impacting the
configuration, the fragments lost most of their kinetic energy on meeting the hard
alumina tile. The backing laminated GFRP and Kevlar fabric also offered penetration
resistance alongside providing structural support to the brittle alumina against its
inherent shattering. The abrupt drop in velocity to below 200 m/s is an indication of
the protective capability of the configuration. Contrary to the experimental results, a
slight bulge on rear side of C-4 is visible in Figure 4-37. The velocity profile of the
gauge points just before and after impacting C-4 is shown in Figure 4-38.
The ALE model of C-2 is shown in Figure 4-39 to simulate the behavior against
fragment impact. The grid and material plots are shown in Figure 4-39(a) and Figure
4-39(b) respectively.
An isometric view of fragment impact on C-2 and C-4 is shown in Figure 4-40. The
fragments penetrating configuration C-2 and the gauge points are also shown in this
figure. An enlarged view of C-2 shows that fragments have almost penetrated the
configuration.
Figure 4-38: Velocity plot of gauge points (a) before and (b) after impacting C-4
a b
Figure 4-39: ALE model of C-2. (a) Grid plot and (b) material plot
a b
97
A 3D view of the fragmentation and protective configurations C-2 and C-4 is shown
in Figure 4-41. The Kevlar fabric (110 GSM) and GFRP laminate exert a resisting
force to fragments, reducing their velocity. At the same time, the fabric is deformed
Figure 4-40: (a) Isometric view of C-2 and C-4, (b) Fragments perforation
through C-2. (c) Gauge points defined on C-2
a b
c
Figure 4-41: (a) Fragments penetration through C-2 (b, c) combined view of
C-2 and C-4
a
b c
98
and accelerated, thereby dissipating the energy. Several fragments were captured
within the configuration. A few of them were able to perforate. The velocity profile
of the gauge points before impact and after penetrating C-2 is shown in Figure 4-42.
A sudden drop in velocity of the gauge points is observed. However, two of the
gauge points still show a residual velocity of 500 m/s, indicating their capability of
perforation through C-2 as shown in Figure 4-42(b).
Both C-1 and C-2 were having equal number of layers of two different Kevlar
fabrics. From the simulated and experimental results, it was evident that Kevlar
fabric-110 GSM has presented a relatively lesser resistive force to the impacting
fragments as compared to the Kevlar-450 GSM fabric used in configuration C-1.
4.4.4 Blast Loading on Protective Configurations
Simulation for blast parameters was performed using coupled ALE-Euler multi-
material approach in ANSYS AUTODYN. The hollow steel shell was modeled in
ALE, whereas Comp-B filling and surrounding air were modeled in the Euler solver.
An optimized grid size of 1x1 mm was used for both solvers [18]. Configuration C-2
was selected for simulating the blast loading as this was the lightest among C-1 to C-
5 configurations. The numerical model and shell detonation are shown in Figure
4-43. Upon detonation, the shell casing starts expanding under pressurized detonation
product gases as shown in Figure 4-43. The escape of detonation products in air and
subsequent formation of a blast wave and its propagation is already shown in Figure
3-11 of Chapter 3. The formation of blast wave and its outward propagation is shown
in Figure 4-44.
Figure 4-42: Velocity profiles of gauge points (a) before and (b) after impacting C-2
a
C2
b
99
The blast wave arrival, impact and loading on C-2 are depicted in Figure 4-45. The
figure also illustrates the configuration before and after the blast impact and confirms
the integrity of this most vulnerable configuration in present study.
Figure 4-43: (a) AUTODYN model of shell (b) detonation wave
propagation inside shell (c) Expansion of shell at 20 s
a b c
Figure 4-44: (a) Expansion of shell at 40 s (b) at 61 s and (c) at t= 90 s
a b
c
Figure 4-45: (a) Blast wave approaching C-2 (b) impacted on C-2 and (c) C-2 before
and (d) after blast impact
a b
c d
100
The blast wave undergoes reflection when the forward moving air molecules are
brought to rest and further compressed upon meeting the front surface of C-2,
forming a reflected wave with pressure designated by Pr. The reflected blast wave on
the face of C-2 delivers a compressive shock wave from Kev-epoxy layers to PU
foam and laminated GFRP. The shock wave energy is absorbed and dispersed by PU
foam and GFRP layers respectively.
The gauge point 2 was defined 10 mm ahead of C-2 and gauge point 3 was defined
on the back side of C-2 for peak pressure measurement, as shown in Figure 4-45(a).
Peak incident (Ps) and reflected (Pr) pressures of 57 psi (393 kPa)and 261 psi (1799
kPa) respectively were observed at the front of configuration. The pressure-time
history P(t) at gauge point 2 was integrated for the positive phase duration to obtain
reflected impulse. The pressure and impulse plot are shown in Figure 4-46(a) and
Figure 4-46(b) respectively. IS and IR represents incident and reflected impulse
respectively.
The 10 mm gap differentiates the incident and reflected pressure rise time as well as
the impulse shown in Figure 4-46. A peak pressure value of 2.3 psi (15.85 kPa) for
gauge point 3 was obtained in simulation, as shown in Figure 4-47(a). The gauge
Figure 4-46: For gauge point 2 (a) Incident and
reflected overpressure (b) Impulse plots
a
b
IS
IR
101
point 7 was defined to face the blast impact on PU foam whereas the gauge point 8
was located 4.0 mm behind point 7. The shock wave transmitted into the PU foam
and its absorption and shock mitigation in terms of specific internal energy is shown
in Figure 4-47(b). The figure demonstrates that the pore collapse mechanism in PU
foam absorbs substantial energy.
4.5 Experimental Results
Two tests were conducted with a scaled down model of 155 mm artillery shell to study
the fragmentation and blast phenomenon and their loading effects on seven protective
configurations. The fragmentation phenomenon involves the detonation of the explosive
filling in the shell producing a very high pressure which is imparted to the metal casing
within a time scale of microseconds. The detonation wave exerts high pressure on
successive cross-sections of the shell body until it fractures and disintegrates into high
velocity fragments.
Figure 4-47: (a) Pressure at gauge point 3, (b) Change in
Specific internal energy in PU foam for gauge points 7 and 8
b
a
102
The protective configuration C-1 located at 0.65 m from the shell was impacted by
15 fragments and only 5 were able to perforate. In the second test, 3 fragments could
perforate out of 11 those impacted the configuration. The Kevlar fabric exerted a
resisting force to impacting fragments and reduced their velocity. The high strength
Kevlar fabric (470 GSM) and laminated GFRP combination was able to contain the
majority (70%) of the high velocity fragments. The perforation of few of the
fragments and a closer view of the fragments captured in Kevlar and laminated
GFRP are shown in Figure 4-48.
Unlike C-1, both C-2 and C-3 employed Kevlar fabric-110 GSM with the same
number of layers as that of C-1. Configuration C-2 captured only 3 of the 9
fragments impacted on it, whereas C-3 captured 8 out of a total 16. The fragment
impact and perforation through C-2 and C-3 is shown in Figure 4-49 and Figure 4-50
respectively. While resisting the fragment impact, the Kevlar fabric was elongated.
The post impact view on laminated GFRP and PU form (of C-2) is also shown in
Figure 4-50(b, c).
Figure 4-48: Fragments perforation and captured in (a, b) Kevlar and (c, d) GFRP
a b c d
Figure 4-49: (a) C-2, (b) laminated GFRP and (c) PU foam after
fragments impact
a b c
103
The total number of fragments that perforated C-3 in both tests was 14. The
configuration was able to capture almost 60% of the impacted fragments. The Kevlar
fabric-470 GSM and laminated GFRP have offered significant resistance to high
velocity fragments owing to their higher modulus values compared to Kevlar fabric-
110 GSM. From the simulated and experimental results, it was evident that Kevlar
fabric-110 GSM offered less resistive force to impacting fragments compared to 450
GSM fabric used in configurations C-1 and C-4.
C-4 was relatively distant, located 0.80 m away, it was impacted by only 4 fragments
and none of these could perforate the configuration. This was also evident from the
back-side view of C-4 shown in Figure 4-51(a). The front and back sides of
laminated GFRP, shown in Figure 4-51(b & c) did not show any evidence of
fragment impact. Moreover, no bulging effect was noticed on backside of the Kevlar
covered configuration, as evident from Figure 4-51(a). However, the simulation
results show a slight bulge on back side of the configuration.
Configuration C-5, located 0.725 m from shell, was impacted by 7 fragments but no
perforation or backface signature was observed, as shown in Figure 4-52(b).
Figure 4-50: (a) Front and (b) back sides of C-3 (c) laminated GFRP after second test.
a b c
Figure 4-51: (a) Back side of C-4 (b, c) front and back sides of laminated GFRP
placed behind the ceramic tile
a b c
104
The alumina tiles in C-4 and C-5, with hardness greater than 75 HRC, offered very
high resistance to impacting fragments. The backing laminated GFRP sheet further
dispersed the shock energy transmitted through the ceramic tile in lateral directions.
After the test, a few outer layers of Kev-epoxy, as shown in Figure 4-52(a, b) were
removed. A fragment weighing 3.69 g and measuring 48 x 5 mm2 was recovered. A
previously captured bullet was also recovered. The configuration was previously
tested against 7.62 mm bullet at an impact velocity of 715 m/s. The configuration
stopped the bullet without a significant backface signature [8]. The recovered bullet
and fragment are also shown in the Figure 4-52(c). The figure did not show any
evidence of fragment perforation through this configuration. The front alumina tile
completely disrupted the impacting fragment as well as the impacted bullet because
of its high hardness. The shock transmitted from alumina was dispersed in lateral
directions by the backing GFRP, and finally absorbed by the pore collapse
mechanism in the PU foam. The Kev-epoxy covering also served to contain the
secondary fragmentation of brittle alumina and retained the integrity of the
configuration against multiple fragment’s impact.
The configurations C-4 and C-5 employing a ceramic (alumina tile) were able to
completely capture the high velocity fragments, thus providing 100% protection
against fragments weighing up to 4.5 g and moving with velocities up to 1555 m/s.
The backface signatures (also known as blunt force trauma) test of C-5 was
performed and maximum depth of depression of 10 mm was measured [8] , which
was well within the US, European, German and British standards for backface
signatures [11, 19]. The test and the results are evident from the Figure 4-27.
Figure 4-52: (a) Front (b) back sides of C-5 after fragment's
impact, (c) recovered fragment & bullet
a b c
105
C-6 comprised a multi-layered composition of PU-silica of density 1.1 g/cm3. C-6
was located at a stand-off distance of 0.60 m. A total of 14 fragments impacted this
configuration, of those 9 were able to perforate. Another low-density configuration
comprising mixtures of PU, silica and alumina powder was tested at a distance of
0.57 m. Eleven (11) fragments perforated from a counted eighteen (18) impacted on
C-7. A fragment velocity of 1038 m/s at the free surface was calculated by means of
flat timing probes placed behind this configuration. Both C-6 and C-7, being lightest
and most thermally resistant of all configurations, did not offer significant resistance
to the impacting fragments. The post- test view of C-6 and C-7 is shown in Figure
4-53.
A summary of the experimental and simulated results is presented in Table 4-5.
Although all of the configurations withstood the blast wave loading when fixed
rigidly, only C-4 and C-5 provided full protection against both threats of blast wave
and high velocity fragment loading.
a b
c d
Figure 4-53: (a, b) Front and back sides of C-6 (c, d) C-7, after
fragments impact
106
Table 4-5: Summary of the experimental and simulated results
Confi
gura
tions
Thickness
(mm)
Results Protection
Blast/Fragments
Experimental Simulation
Fragments
Impacted
Fragments
Perforated
C-1 24 26 8 perforation Full/Partial
C-2 17 9 6 perforation Full/Partial
C-3 18 18 8 perforation Full/Partial
C-4 21 8 0 Slight bulge Full /Full
C-5 26 7 0 Slight bulge Full/Full
C-6 22 14 9 perforation Full/Partial
C-7 27 18 11 perforation Full/Partial
107
References
[1] P. Qiao, M. Yang, F. Bobaru, Journal of Aerospace Engineering. 21(2008)
235.
[2] C.W. Poh, Investigation of new materials and methods of construction of
personnel armor, M.Sc(2008).
[3] W.W. Chen, ", Journal of the American Ceramic Society. 4(2007).
[4] J.S. I. M. Ward, An Introduction to the Mechanical Properties of Solid
Polymers, 2nd Edition. (2004), 394.
[5] B.C. Wai, Investigation of Shock Wave Attenuation in Porous Media,
M.Sc(2009).
[6] L.J. Gibson, M.F. Ashby, Cellular solids: structure and properties. (1999.
[7] Y. Ma, Ballistic strength of multi-layer fabrics against fragment simulating
projectiles(2017).
[8] K. Ahmed, A.Q. Malik, I.R. Ahmad, International Journal of Protective
Structures. 10(2019) 289.
[9] L. Carbajal, J. Jovicic, H. Kuhlmann, Assault riffle bullet-experimental
characterization and computer (FE) modeling, in Experimental and Applied
Mechanics, Volume 6. 2011, Springer.651.
[10] A.C. Merkle, E.E. Ward, J.V. O'Connor, J.C. Roberts, Journal of Trauma and
Acute Care Surgery. 64(2008) 1555.
[11] R. Kaiser. Understanding Blunt Force Trauma/ Backface Signature. Available
from: https://www.ppss-group.com/blog/understanding-blunt-force-trauma-
backface-signature/.
[12] U. ANSYS Inc., Century Dynamics. Release 14.0 documentation for ANSYS
AUTODYN.(2011).
[13] M.A. Abdalla. Fragmentation Analysis of OG-7 Warhead Using AUTODYN
SPH Solver. in Advanced Materials Research. 2012. Trans Tech Publ.
[14] E. Lee, H. Hornig, J. Kury, California: University of California, (1968).
[15] M.M. Ansari, A. Chakrabarti, M.A. Iqbal, Procedia engineering. 173(2017)
153.
[16] A.M. Soydan, B. Tunaboylu, A.G. Elsabagh, A.K. Sarı, R. Akdeniz,
Advances in Materials Science and Engineering. 2018(2018).
[17] R. Messahel, M. Souli, CMES: Computer Modeling in Engineering &
Sciences. 96(2013) 435.
108
[18] K. Ahmed, A.Q. Malik, AIP Advances. 10(2020) 065130.
[19] N.L. Enforcement, C.T. Center, U.S.o. America, O.o.L.E. Standards, (2001).
109
Containment for Blast, Fragmentation Chapter 5
and Thermal Effects
5.1 Introduction
The detonation of a cased energetic material or an IED is manifested by blast
overpressure, fragmentation and thermal effects. Human beings are sensitive to
overpressure and their air filled organs like, ears, lungs and the gastrointestinal track
are the most susceptible to damage [1]. Thermal effect is a potential injury
mechanism and is associated with the burning due to heat and fireball [2, 3]. A
rapidly propagating fire hazard is also probable if any hydrocarbon fuel is present in
the close vicinity. Fragmentation is the most lethal effect because fragments can
travel to large distances and cause serious injuries to humans. In recent years, nails,
screws, ball bearings and other metallic parts have been used in IEDs to enhance this
injury mechanism. Blast wave mitigation and on-spot quenching of the fireball are
imperative to avoid loss of invaluable lives and infrastructure. Efforts have been
made in the past by individuals and organizations to reduce such life threatening and
infrastructure damaging effects associated with EM detonation. The present work
deals with the experimental and numerical investigation of the on-spot quenching of
fireball, blast wave attenuation and containment of primary and secondary fragments
using lightweight composite materials discussed in previous chapters. The open area
energetic material detonation effects were compared with the detonation inside this
containment system.
5.2 Materials and Experimental Work
5.2.1 Scaled down Container
To limit the devastating effects of a cased EM or an IED detonation, a heterogeneous
cylindrical container of diameter 250 mm and height 525 mm, comprising 2 mm
thick steel liner (height-250 mm) wrapped in laminated GFRP and Kevlar fabric was
developed and studied. Shaving foam was used as filling material in this container as
shown in Figure 5-1. Kevlar woven fabric 470 GSM (grams per square meter) and
110
laminated GFRP were wrapped on 2 mm thick steel liner to provide sufficient
strength against highly repulsive detonation product gases. A combination of
composites, including Bakelite, PU-silica and EPS foam was employed at the base of
the container. The PU-silica composite has density of 1.1 g/cm3 and compressive
strength of 232 MPa. The material has shown good thermal resistance and shock
absorbance at extreme conditions [4]. C4 charge weighing 104 grams was tested in
open area as a surface burst to measure the blast wave parameters.
C4 weighing 104 grams was placed in contact with the base of the container, as
shown in Figure 5-1(b). The container was then filled with shaving foam shown in
Figure 5-1(c). Pressure transducers (S1 – S4) were employed at radial distances of
280 - 620 mm and different heights to measure the pressure-time history of the blast
wave. Transducers were sealed in steel pipes for protection against fragment impact.
The blast wave data was obtained through the transducers with a data acquisition
(DAQ) system. High-speed photography was used for visual studies of the fast events
at 54000 fps. A sketch of testing setup for C4 surface burst and inside composite
container is shown in Figure 5-2.
a b
Figure 5-2: A sketch of testing setup for (a) surface burst (b) detonation
inside cylindrical container
Figure 5-1: (a) Empty container (b) C4 placed at the bottom (c) filled with shaving foam
a b c
111
The experimental setups for surface burst and inside composite container are shown
in Figure 5-3 and Figure 5-4 respectively. For bare C4 test, the charge was placed on
25 mm thick expanded polystryne (EPS) foam which was placed on 60 mm thick
steel disc. The EPS foam was employed to avoid a fragment or debris flight and their
impact on sensors and recording instruments. The EPS foam has absorbed a part of
explosive energy before reflection of blast wave from the steel disc. A part of the
energry may also be wasted in crater formation if placed directly on ground.
5.3 Numerical Simulation
5.3.1 C4 Surface Burst Parameters
Simulation for blast parameters was performed using coupled ALE-Euler approach in
ANSYS AUTODYN [5]. C4 explosive filling and surrounding air were modelled in
Figure 5-4 (a) Experimental setup for C4 detonation inside container (b) C4 and
foam filled in container
b
S1
S2
S3
S4
a
Figure 5-3: Experimental setup for bare C4 charge test (a) close view
(b) complete setup
C4
a
S1 S2
S3
S4
C4
b
112
Euler solver whereas the steel reflecting surfaces with the transducers were modeled
in ALE. JWL (Jones-Wilkins-Lee) equation of state [6] was used for expansion of C4
product gases and ideal gas EOS was used for modeling air. An optimized grid size
of 1x1 mm2 was used for both solvers [2, 4, 7]. The numerical model, product gases
expansion and blast wave propagation are shown in Figure 5-5 and Figure 5-6. C4
charge was placed on 25 mm thick EPS foam and a 60 mm thick steel disc as shown
in Figure 5-3(a). A Steel reflecting medium measuring 10 x 5 mm2 was modeled in
ALE close to gauge points 2, 3 and 4. These points give reflected overpressure (Pr)
whereas the gauge point 1 represents incident overpressure (Ps).
The simulation results for the surface burst are presented in Table 5-1. Here, X
represents the radial distance from the center of cylindrical charge and Y is the height
from ground level. The blast wave arrival time is indicated by Ta.
c d
Figure 5-6: (c) Blast wave formation and (d) propagation in air towards gauge points
a b
Figure 5-5: (a) AUTODYN model (b) product gases expansion
113
Table 5-1: Simulation results for 104g bare C4 surface blast
Gauge Nos. Position Overpressure (psi) Ta (ms)
X Y Ps/Pr
1 280 990 26.6 (Ps) 0.905
2 620 810 58 (Pr) 0.83
3 390 790 75.3 (Pr) 0.671
4 370 550 203 (Pr) 0.381
The pressure-time history at gauge points 1 to 4, defined for positions of pressure
transducers, is shown in Figure 5-7. Only radial distances (X), rather than slant
heights, are shown in pressure-time plot. The blast arrival time and peak pressure
values are in close agreement with the experimental findings.
The exact material model for shaving foam was not available. Two-phase EOS for
water was used for the foam filling inside container. This EOS incorporates the effect
of evaporation. The simulation inside container filled with shaving foam was only
performed to have an idea of the wave propagation pattern and reflection from the
bottom and walls of the container. The AUTODYN model and C4 detonation inside
the container is shown in Figure 5-8. The blast wave reflection from the base of the
container and walls are shown in Figure 5-8(b). The nearest part of the container with
the C4 charge has undergone deformation as seen from Figure 5-8(c). However, any
escape of the blast pressure in lateral directions is not observed in the simulation.
Figure 5-7: Simulated P(t) history of gauge points for 104g bare C4
surface blast
114
5.3.2 Fragments Impact on Scaled down Container
The response of the container to fragment impact was investigated using coupled
SPH-ALE approach. The scaled down shell and its fragmentation was modeled and
simulated using SPH solver. The container was modeled in ALE. Quarter symmetric
FE model of shell and container is shown in Figure 5-9. The height of the container
was adjusted to simulate only the impact of the highest velocity fragments from the
cylindrical part of the shell. That is why the fragments from the base of the shell were
deleted at 76.0 s after detonation just to reduce the computational time and efforts.
Figure 5-9: FE model for shell and container
a b c
Figure 5-8: (a) AUTODYN Model (b) blast wave reflection from base and walls (c) blast
wave loading on container walls
115
Fragments weighing tens of milligrams to 6.5 grams were produced. The fragments’
mass, spatial distribution and velocity distribution has already been presented in
Figure 3-13. The shell fragmentation at 63.93 s after detonation and the velocity
plot are shown in Figure 5-11. The fastest fragments traversed a radial distance of 86
mm as shown in this figure. The detonation product gases were deleted to study the
impact of only fragments with the container. An isometric view of the fragments with
the container wall at 78.8 s is shown in Figure 5-12.
Figure 5-11: Isometric view of the fragments, radial dispersion and container
Figure 5-10: (a) Shell fragmentation at 63.93 s and (b) velocity plot of the gauge points
a b
116
The fragments impact and penetration is shown in Figure 5-12(a). The steel liner
deformation and perforation by few of the fragments through this lining material at
131 s is visible in Figure 5-12(b). The composite layers of laminated GFRP and
Kevlar fabric has offered a good resistance to impacting fragments leading to a drop
in fragments velocity. Fragments perforated the steel liner were captured within the
composite layers. The fragments lost most of their kinetic energy due to high
resistance offered by the composite materials combination. The velocity plot of the
gauge points after impacting the container walls is shown in Figure 5-13.
Although the container was not tested experimentally against the fragments, the
tested configurations against the shell fragmentation presented in Chapter 4 validate
the simulated results.
Figure 5-13: Fragments velocity after impacting the container walls
Figure 5-12: Fragments penetration into steel liner and its deformation
a b
117
5.4 Experimental Results
5.4.1 Scaled down Container Test
The small-sized bubbles (15 – 100 m) in shaving foam were found to be stable over
time. The measured sound speed in the two-phase medium, 65.98±3 m/s, was far
below the sound speed in air and water. This significant difference in the acoustic
property played an important role in peak pressure attenuation. Two experiments
were performed with 104 grams C4 charge as surface burst and inside the cylindrical
container filled with shaving foam. The appearance of fireball is the most immediate
event upon detonation of an energetic material. For bare C4 detonation, the entrapped
oxygen within the fireball assists in afterburning reactions and enhances the total
energy. The total heat of combustion is the sum of heat of detonation and the
afterburning heat [2]. The measured fireball radius from high-speed images was 1.1
m. The resulting high pressure and temperature trigger the formation of a blast wave
in air. Consequently, the blast wave begins to propagate outwards in air. The
sequence of events is shown in Figure 5-14, Figure 5-15 and Figure 5-16. The high-
speed Phantom camera was operated at frame rate of 54000 fps for these tests.
Figure 5-15: High speed images (c) t= 0.320 ms and (d) t= 0.725 ms
d c
Figure 5-14: High speed images of bare C4 (a) detonator fired (b) t= 0.22 ms
after detonation
a b
118
The high-speed images of 104 grams C4 detonation inside the cylindrical container
filled with shaving foam are shown in Figure 5-17, Figure 5-18 and Figure 5-19. The
shaving foam has completely suppressed the fireball owing to a rapid heat and
momentum transfer mechanisms. This transfer mechanism is enhanced by the
presence of finely distributed small-sized bubbles. The suppression of fireball also
seized the afterburning reactions due to a reduction in temperature [2, 8]. The
quenching of fireball also diminishes the thermal affects. The reduction in pressure is
also attributed to the much lower sound speed in foam compared to sound speed in
air and water.
The emergence of carbonaceous soot is also an indication of the fireball suppression
and quenching of afterburning reactions. The carbonaceous soot with the product
gases emerged from the open top of the container, along with the product gases, at
about 0.80 ms after the detonation, as seen from Figure 5-17(b). The high-speed
images shown in Figure 5-18 and Figure 5-19 further illustrate the relatively slower
movement compared to an open area detonation. The product gases did not reach the
transducer located above the container, 0.99 m above ground, until 1.8 ms, as shown
in Figure 5-19(e).
Figure 5-16: High speed images of bare C4 detonation
f e
a b
Figure 5-17: high speed images at (a) t=0.075ms (b) t = 0.800 ms
119
The protective container also restricted the escape of product gases in lateral
directions. The combination of high strength Kevlar fabric 470 GSM and laminated
GFRP provided sufficient resistance to the expansion of product gases. The
laminated GFRP retained its integrity against the loading whereas a slight
deformation (elongation) was observed in Kevlar fabric. Adhesive debonding under
shock effect was observed in the outer layer of Kevlar fabric.
The container base was the most vulnerable place due to in contact detonation of C4.
The combination of EPS foam, Bakelite and PU-silica composite layers provided
protection against the extreme loading conditions of C4 detonation. Therefore, the
platform or object holding the container at the base (vehicle, ship and other
vulnerable objects or structures) will remain protected against the damaging effects
of energetic material detonation. The severity of the contact detonation can be seen
from Figure 5-20 where 30 grams C4 was detonated on top of a 5 mm thick steel
plate stand. A complete perforation of the steel plate was observed [2]. The novel
combination for the base proposed in present work has provided good protection
against any perforation, even with a higher charge.
c d
Figure 5-18: High speed images at (c) t=1.075ms and (d) t=1.475ms
Figure 5-19: C4 charge detonation inside container at (e) 1.80ms
and (f) t= 2.675ms
e f
120
The pressure-time histories at the four transducers placed around the charge, in open
-air and inside container, are shown in Figure 5-21(a & b) respectively. Transducer
S1 has recorded side-on (incident) overpressure whereas transducers S2, S3 & S4
have experienced some levels of pressure reflection due to their orientation and
location with respect to the charge as shown in the experimental setups of Figure 5-3.
Figure 5-20: Contact detonation of 30 g C4 (a) test setup (b) post-test view
a b
a
b
Figure 5-21: Pressure plots for 104g C4 detonated (a) Surface
burst (b) inside container
121
The P(t) plot for 104g bare C4 surface burst is shown in Figure 5-21(a). A reflected
overpressure of 203 psi (1399.6 kPa) was measured at transducer S4 located 0.55 m
above ground at a radial distance of 0.370 m from cylindrical axis (X). The
corresponding arrival time (Ta) was 0.38 ms. A reflected overpressure of 70.5 psi
(486 kPa) was measured at S2 located 0.81 m above ground at the radial distance(X)
of 0.61 m.
In case of the detonation inside container, the blast wave undergoes several
reflections from the bottom and the walls of the container, as evident from the shape
of the P(t) plot in Figure 5-21(b). The pressure transducer S4 placed at the height of
0.550 m and radial distance of 0.470 m recorded a reflected overpressure of 8.4 psi
(57.9 kPa) with an arrival time of 1.75 ms. The maximum pressure measured was
18.3 psi (126 kPa) at transducer S1 placed on top and 80 mm away from the outer
walls of the container. The arrival time for the peak value was 1.83 ms.
A pressure reduction of more than 80% is obtained with the arrangement. The
shaving foam also caused a significant delay in the upward flow of the product gases,
as evident from high-speed images in Figure 5-17(a-b) and pressure plot in Figure
5-21(b). The layered combination of laminated GFRP and Kevlar fabric provided
sufficient strength to contain the rest of the detonation product gases and the blast
wave. The post-test sights of the container and PU-silica composite are shown in
Figure 5-22
Although the effect of fragmentation was not studied experimentally, a simulation
was performed to study the fragment impact on the container. The fragment’s
velocities were immediately reduced to below 200 m/s on meeting the walls of the
container, showing evidence that the container would also provide protection against
c a b
Figure 5-22: Post-test sights of (a & b) PU-sand composite top and side (c) container
122
fragments. The investigation has produced a reliable approach towards on-spot
fireball suppression leading to thermal effects diminution, blast wave mitigation and
ultimate containment against fragments. A good agreement between simulated and
experimental results for bare blast parameters was obtained.
5.4.2 Container Test with 1.0 kg TNT Equivalent Charge
The scaled down container for 104 grams C4 charge was scaled up to contain the
detonation effects of 1.0 kg bare or 0.6 kg steel cased TNT equivalent charge (pipe
bomb). The full scale container comprised an inner and an outer cylindrical
container. The inner layer was made of 3 mm thick mild steel (MS) cylinder with an
inner diameter (ID) 500 mm and height of 350 mm. A 5 mm thick MS disc was
welded at the bottom of the cylinder. Three steel fixtures of tubular configuration 40
mm high were welded 120o apart to the bottom plate. Four layers of Kevlar woven
fabric 470 GSM were wrapped on the MS cylinder. The weight of the inner cylinder
shown in Figure 5-23 was 24.4 kg.
EPS Foam 50 mm thick and of diameter 490 mm was positioned on MS bottom
plate. Bakelite sheet measuring 380 x 295 x 20 mm3 and weighing 3.2 kg was then
placed on top of EPS foam. PU-Silica disc prepared with 6 alternate layers of PU and
silica was placed on the Bakelite sheet. Another sheet of EPS foam with thickness of
50 mm and diameter of 200 mm was then placed on the PU-silica composite. The
PU-silica, EPS foam and Bakelite sheets are shown in Figure 5-24.
a b
Figure 5-23: (a) MS cylinder (b) Inner container
Figure 5-24: (a, b) PU-silica disc (c) EPS foam and Bakelite sheet c a b
123
The outer cylinder was made of 8 mm thick laminated GFRP sheets with inner
diameter 550 mm and height 800 mm. A single layer of Kevlar fabric 110 GSM was
employed at the inner surface of the GFRP cylinder with silicon sealant. Eight layers
of Kevlar fabric 470 GSM were then wrapped on the outer surface of the GFRP
cylinder shown in Figure 5-25. The top and bottom of the outer cylinder were left
open. The outer container shown in Figure 5-25 weighed 34 kg.
The inner cylinder was then placed inside the outer composite cylinder. The net
weight of the combined container was 67 kg. C4 charge weighing 800 grams (1.0
kg TNT equivalent) was placed at the center of the container as shown in Figure
5-26(a). Shaving foam was filled around the charge and inside the inner container as
shown in Figure 5-26(b). The charge was initiated at the top end.
b c a
Figure 5-25: (a) GFRP cylinder (b) inner view of composite container (c) outer
composite container
a b
Figure 5-26: (a) C4 placed inside container (b) shaving foam filled
around C4
124
The testing setup is shown in Figure 5-27. Three pressure transducers were placed
0.9 m above ground level at radial distances of 0.8, 0.9 and 1.0 meter from center of
charge. The transducers were positioned to measure the reflected overpressures (Pr).
The shaving foam completely suppressed the fireball and the container restricted the
movement of product gases in lateral directions. The high-speed images of the event
at different time steps are shown in Figure 5-28(a – e). The product gases appeared
from the open top after 0.482 ms of C4 detonation. The expansion of carbonaceous
soot with product gases is shown in the high-speed images. The outer layer of the
Kevlar fabric was detached due to debonding of adhesive. Figure 5-28(f) shows the
post experiment view of the container.
Figure 5-27: Experimental setup for 800 g C4 detonation inside container
(a) at t= 0 (b) at t= 0.482ms (c) at t= 1.50ms
(d) at t= 2.0 ms (e) at t= 4.2ms (f)
Figure 5-28: High-speed images of 800g C4 detonation inside container
125
There was a 40 mm gap between the base of the inner cylinder and ground level to
allow for the expansion of MS base plate. However, the intense pressure detached the
welded base plate from the MS cylinder. Some leakage of the gases at the bottom
side was observed at a later stage. These gases were less pressurized and reached the
1.5 m radial distance after 35.0 ms of detonation.
The measured reflected overpressure and arrival time data is shown in Figure 5-29. A
reflected overpressure peak of 25.5 psi (175.8 kPa) was recorded at 0.9 m from the
charge center. A maximum reflected overpressure of 12.4 psi (85.5 kPa) was
recorded at 1.0 m distance from the center of the container.
5.4.3 Container Test with Steel Cased Charge (Pipe-bomb)
The second experiment of the container was performed to study its protective
capability against combined blast, fragmentation and thermal effects. A steel cased
charge (pipe-bomb), shown in Figure 5-30, was considered. This cased explosive
also simulates the effects of a lighter IED. The details of the steel cased charge are
given in Table 5-2.
Figure 5-30: Steel cased charge (Pipe-bomb)
Figure 5-29: Reflected overpressure-time history for
800g C4 detonation inside container
126
Table 5-2: Material and dimensions of pipe-bomb for blast and fragmentation study
Material Mass (g) Length(mm) OD(mm) ID (mm)
MS Casing 1275 173 61 50
Comp-B filling 565 170 50 --
The blast wave parameters and fragmentation characteristics of this steel cased
charge were not studied experimentally. However, numerical simulations were
performed to have an estimate of these parameters. The SPH simulations of the
fragmentation within the MS cylinder are shown in Figure 5-31 and Figure 5-32.
A total of 732 fragments were produced. These fragments were grouped in to four
categories based on their mass in grams (g);
a b c
Figure 5-31: (a) SPH- ALE model for steel cased charge and inner container (b) at 15
s of detonation (c) fragmentation at 40 s
a b
Figure 5-32: (a) Gauge points location at 58s, (b) radial flight at 105 s
127
Very small < 0.1 g
Small 0.1 – 1.0 g
Medium 1.0 – 5.5 g
Large > 5.5 g
The number of fragments and their mass distribution is shown in Figure 5-33(a)
while the velocity of the gauge points defined on the casing is shown in Figure
5-33(b). Most of the fragments have velocities around 1550 m/s. The fragment
velocities were comparable to that of the scaled down shell described in Chapter 3.
However, the mass distribution is on the higher side.
The overall strength of the container was already weakened by the explosive loading
during the first experiment of 1.0 kg TNT equivalent charge. An additional layer of
GFRP was laid inside the MS cylinder for this test. The container filled with shaving
foam and the testing setup is shown in Figure 5-34(a, b).
a b
Figure 5-33: (a) No. of fragments and mass distribution (b) fragments velocity plot
a b
Figure 5-34: (a) Shaving foam filled around steel cased charge (b) experimental setup
128
The amount of shaving foam filled for this test was relatively less as shown in Figure
5-34(a). That is why the fire ball was partially suppressed as can be seen from the
high-speed images in Figure 5-35(b – c). However, the container still provided
significant protection against overpressure. The appearance of a partially quenched
fireball and the product gases containing carbonaceous soot from the open top are
evident in Figure 5-35(b – f).
Although, the container collapsed during this test, there was no evidence of any
debris flight beyond 1.5 m from the blast. The additional layer of GFRP inside the
MS cylinder was able to offer considerable resistance against high velocity
fragments. Majority of the fragments lost their kinetic energy while perforating this
layer. A view of the fragments impact and perforation through inner and outer layers
of GFRP is shown in Figure 5-36(a) and Figure 5-36 (b) respectively.
(a) t=0 (b) t=0.407ms (c) t= 0.648ms
(d) t=1.222ms (e) t=2.500ms (f) t=6.055ms
Figure 5-35: High-speed images of steel cased 565g Comp-B detonation inside container
a b
Figure 5-36: Fragments impact and perforation (a) inner layer (b) outer layer of GFRP
129
The outer composite container was able to contain the majority of the fragments.
Only 13 fragments were able to perforate the outer layer of Kevlar fabric as shown in
Figure 5-37(a, b). The bottom MS disc is shown in Figure 5-37(c). A combination of
lightweight materials (EPS foam, Bakelite and PU-silica) at the bottom of the
container and in contact with the charge was able to mitigate the extreme loading.
Although the MS disc at the bottom of the container was deformed, no signs of a
fragment penetration were observed, as visible in Figure 5-37(c).
The maximum reflected overpressure measured at 0.85 m was 6.8 psi (46.88 kPa)
with an arrival time of 1.21 ms. The first peak value at 0.9 m distance was 5.8 psi (40
kPa) with arrival time of 1.41 ms. The multiple reflections from the ground and the
walls of the container yielded a reflected overpressure of 11.79 psi (81 kPa) at the
distance of 0.9 m. The reflected overpressure-time history is shown in Figure 5-38.
Figure 5-37: (a, b) Fragments perforation through outer layer of Kevlar fabric
(c) bottom MS disc after second test
b
c
a
130
Table 5-3 presents comparison of the present studies with Conwep results. A hemi-
spherical surface burst was considered for Conwep calculations. As the charge was
positioned 166 mm above the ground level, the slant heights for Conwep calculations
[9] corresponding to given radial distances of 1.0 and 0.9 meter were 1.2 and 1.1
meter respectively. The transducers were 0.9 m above the ground level.
Table 5-3: Comparison of experimental results with Conwep calculations
W (TNT
eq.)
Conwep Results- Surface Burst With Container Reduction
R(m) Ta(ms) Ps(psi) Pr(psi) Ta(ms) Pr(psi) Ps(psi) %
1.0 kg
1.0 0.65 133 705 1.73 12.8 3.3 97
0.9 0.55 163 940 1.48 25.5 5.5 98
0.6 kg
Steel
Cased
0.9 0.64 109 570 1.41 11.79 3.1 97
As seen from the observed results the container was able to limit the devastating
blast, fragmentation and thermal effects of 1.0 kg TNT equivalent EM detonation.
Besides, for bare charge, the container also provides protection against cased charges
up to 0.6 kg TNT equivalent, pipe-bomb and lighter IEDs. The two layers container
provides safety against blast, fragmentation and thermal effects produced by higher
Figure 5-38: Reflected overpressure -time history for steel cased
565 g Comp-B charge inside container
131
explosive mass compared to existing trash receptacles, BCRs etc. cited in references
[10-15] at reduced cost, size and weight.
132
References
[1] B. Rutter, (2019).
[2] K. Ahmed, A.Q. Malik, AIP Advances. 10(2020) 065130.
[3] P. Peters, Military Medicine. 176(2011) 110.
[4] K. Ahmed, A.Q. Malik, A. Hussain, I.R. Ahmad, I. Ahmad, AIP Advances.
10(2020) 095221.
[5] U. ANSYS Inc., Century Dynamics. Release 14.0 documentation for ANSYS
AUTODYN.(2011).
[6] E. Lee, H. Hornig, J. Kury, Adiabatic expansion of high explosive detonation
products(1968).
[7] H. Draganić, D. Varevac, Shock and Vibration. 2018(2018).
[8] L.S. Lebel, P. Brousseau, L. Erhardt, W.S. Andrews, Combustion and flame.
161(2014) 1038.
[9] D. Hyde, US Army Engineer Waterways Experiment Station, USA. 2(1991).
[10] M. Silnikov, A. Sadyrin, A. Mikhaylin, A. Orlov, Materials Physics &
Mechanics. 20(2014).
[11] Blastguard Trash Receptacles. Available from:
http://blastgardtech.com/blastwrap#blastgard-mtr.
[12] Blast Containment Tank. Available from:
https://www.tmi2001.com/products/blast-containment-tank/.
[13] A. Resnyansky, T. Delaney, Experimental study of blast mitigation in a water
mist(2006).
[14] J. Polak, R. Romek, A. Wiśniewski, Problemy Techniki Uzbrojenia.
37(2008).
[15] G. Greenfield, P.R. Gefken, J.D. Colton, Container for explosive
device(2006).
133
Conclusions and Recommendations Chapter 6
6.1 Conclusions
This PhD thesis focused on mitigating the damaging effects of bare and cased
energetic materials detonation, including lighter IEDs. A number of experimental and
numerical simulation studies were conducted to optimize the performance of
lightweight material combinations against combined blast, fragmentation and thermal
effects of cased EM detonation.
Commercially available Denim shaving foam was characterized. It was found that
the coarsening due to coalescence of bubbles slowed down with time. The bubble
size increased from an initial value of 15 m to 120 m in 2 hours showing good
stability. The viscosity of shaving foam measured at shear rate of 24.4 S-1
was 11650
cP. The two-phase medium exhibited a decreasing viscosity trend at higher shear
rates. The density and average sound speed measured in shaving foam at temperature
of 240C was 60.0 kg/m
3 and 65.98±3 m/s respectively. A significant reduction in
sound velocity is observed in shaving foam compared to its value in air. The
measured values were in good agreement to the values calculated using mixture law
or Wood’s formula.
The mitigation properties of foam were analyzed. Blast wave parameters, generated
by the C4 explosive detonation in air and covered in shaving foam, were measured
for scaled distances ranging from 0.39 – 1.80 m/kg1/3
. The shaving foam confinement
reduced the explosive fireball radius up to 80% and eliminated the afterburning
reactions. The suppression of fireball and quenching of afterburning reactions
diminished the thermal effects. An average blast overpressure reduction of 70% with
a corresponding 62% in impulse reduction was observed. This attenuation to blast
wave parameters is due to a number of factors including fast heat and momentum
transfer. The deformation of foam bubbles and the ultimate bursting of thin liquid
films dissipate energy. The evaporation of the liquid content contributes to decrease
the high pressure and temperature.
The fragmentation effect was studied by considering a scaled down (1:4) model of an
artillery shell. The shell fragmentation was characterized by measuring the fragments
134
initial velocity, mass and spatial distributions. Fragments weighing from tens of
milligram(s) to 6.4 grams were produced with velocities ranging from 960 to 1555
m/s. The majority of the fragments produced were weighing below 1.0 gram. The
cylindrical part of the shell has larger contribution among high velocity fragments
1369-1555m/s than the conical and rear parts due to higher charge to mass (C/M)
ratio. The shell fragmentation was numerically characterized by SPH solver in
ANSYS UTODYN. The SPH simulations reproduced fragment mass, size and initial
velocity distribution reasonably well.
Lightweight protective configurations were tested against the combined blast and
fragmentation effects of a cased EM detonation. The protective capabilities were
tested against peak reflected overpressure of 235 psi (1620 kPa) and fragments
weighing up to 4.3 grams with velocities ranging from 961 - 1555 m/s. The
multilayer combinations of Kevlar woven fabrics, laminated GFRP and PU foam
demonstrated significant absorption and attenuation of impacting fragments. The
number of layers of Kevlar woven fabric has a profound effect on fragments
absorption. The configuration C-1 was able to capture 70% of the impacted
fragments. However, the Configurations employing Ceramic (alumina) tile were able
to withstand 7.62 x 39 mm MSC bullet impact, as well as blast and high velocity
fragments impact, without any significant backface signatures. The maximum depth
of depression measured for this configuration was 10 mm. This was well within the
European, German and British standards for backface signatures. The fiber
reinforced cladding of Kevlar-epoxy successfully prevented the disintegration of the
brittle alumina tile, maintaining the integrity of the configuration against multiple
fragments’ impact. The PU-silica composite having density of 1.1 g/cm3 and
compressive strength of 232 MPa has shown ductile behavior and good toughness.
The composite was not able to resist majority of the fragments penetration, however,
it was effective against severe blast loading.
Coupled SPH (Smoothed Particle Hydrodynamics)-ALE (Arbitrary Lagrangian-
Eulerian) approach was used to simulate the interaction of fragments with protective
configurations. A coupled Euler-ALE approach was employed for blast wave loading
on protective configurations. The blast wave parameters determined using coupled
ALE-Euler approaches were in good agreement with experimental results. These
approaches can be used to predict the blast and fragmentation effects produced by a
135
cased EM detonation and hence minimize the cost and time consumption on full
scale testing.
A scaled down container weighing 13.5 kg and comprising 2 mm thick steel liner
layered by GFRP strips and nine layers of Kevlar woven fabric (470 GSM) was
tested with 104 grams of C4 after filling with shaving foam. The arrangement has
completely suppressed the fireball and hence diminished the thermal effects. It also
provided more than 80 % peak overpressure reduction.
A model container weighing 67 kg was developed and tested against detonation of
1.0 kg bare and 0.6 kg steel cased (pipe-bomb) TNT equivalent charges. This two
layers container completely quenched the thermal effects, provided 97%
overpressure reduction as well as contained the high velocity fragments. The novel
combination of lightweight materials (EPS foam, Bakelite and PU-silica) provided
protection against in contact explosive detonation at the base of the container. The
upshot of this research work, besides being of academic significance, is that it
provides ample data for the development of a blast mitigation system to combat
terrorism against lighter time bomb/IEDs placed at public places, high profile
meeting venues and transportation systems (land, air etc.).
6.2 Recommendations
The work presented in this thesis pertained to the studies for the development of
protective configurations using commercially available lightweight materials to
counter the effects of energetic material detonation.
The work can be extended to devise container that can withstand higher amounts of
charge. The development of a material model for shaving foam will be helpful in
simulating the response of this two-phase medium under extreme loading conditions.
Top Related