KSCE Journal of Civil Engineering (2017) 21(6):2211-2225 Structural Engineering

Copyright ⓒ2017 Korean Society of Civil Engineers

DOI 10.1007/s12205-016-1386-4 pISSN 1226-7988, eISSN 1976-3808
www.springer.com/12205
TECHNICAL NOTE

Review of Analytical and Empirical Estimations for Incident Blast Pressure

Aleem Ullah*, Furqan Ahmad**, Heung-Woon Jang***, Sung-Wook Kim****,
and Jung-Wuk Hong*****
Received May 14, 2015/Revised August 9, 2016/Accepted October 30, 2016/Published Online December 12, 2016

··································································································································································································································

Abstract

Recently, the blast load has become more recognized in the structural engineering field because the blast load can result in not only
disproportionate structural failure but also tremendous casualties of lives and injuries. As an effort to overcome this problem, blast
resistant analyses and designs have been developed, and the methodology would be incorporated into the conventional construction
design. Analysis of structures exposed to blast load is the first step for the design, and it requires good understanding of blast
phenomena and the following dynamic response of structures. This paper provides an up-to-date comprehensive review of the
incident blast wave and its parameters for air and ground blasts. Considering the Unified Facilities Criteria (UFC 3-340-02) as a
benchmark, a quantitative comparison between the empirical results presented by researchers and the result given in UFC charts is
conducted for a span of scaled distances. We discuss the appropriate use of empirical or analytical equations for precise prediction of
blast pressures, and recommend equations that match well the results presented in UFC.
Keywords: blast wave, overpressure, arrival time, impulse, duration, scaled distance
··································································································································································································································

1. Introduction investigate dominant factors affecting on the incident waves.

Secondly, the aim had been to examine the response of a
The effect of explosion on a structure had not been studied structure to blast loads (Beshara, 1994). Brode (1955) performed
comprehensively until World War II. Only limited amount of a numerical integration of the differential equations of gas
research work existed, and even it was kept classified considering its motion in Lagrangian using von Neumann-Richtmyer artificial
sensitivity to the general public (Hopkinson, 1915; Cranz, 1926). viscosity to avoid discontinuities and described dynamic and
For the last several decades, the need of implementation of blast static pressures, positive and negative durations of pressure and
resistant design in major construction projects, especially for velocity, and shock values (Brode, 1955). Henrych and Major
government buildings, had been recognized by observing numerous (1979), based on the analysis of several experimental data,
terrorists’ attacks on civilian structures such as bombing in US presented the formulae to compute peak positive overpressure,
embassy compound, Ankara (1958); US embassy attack, Beirut positive phase duration and positive phase impulse (Henrych and
(1984); Alfred P. Murrah Federal Building attack, Oklahama Major, 1979). Kinney and Graham (1985) utilized both experimental
(1995); World Trade Center attack, New York (2001); Mumbai and theoretical means to get the parameters of the blast wave
attacks, Mumbai (2008); and Mariott Hotel attack, Islamabad such as overpressure, positive phase duration, blast wave arrival
(2008). time and positive phase impulse (Kinney and Graham, 1985). In
A large number of experimental and analytical studies were 1984, Kingery and Bulmash presented the parameters for air
conducted during the second half of the 20th century to understand burst in terms of high order polynomials (Kingery and Bulmash,
blast effects and structural response (Kinney and Graham, 1985; 1984). Swisdak (1994) presented the same results, as were
Henrych and Major, 1979; Brode, 1955; Mills, 1987; Kingery produced by Kingery, in terms of simplified polynomials with
and Bulmash, 1984). The objective had been, firstly, to study the the results accurate to within 1% of the original Kingery values
nature of the blast wave and the characteristics, and had evolved to (Swisdak Jr., 1994). Wu and Hao (2005) developed empirical

*Jr. Structural Engineer, Redco International W.L.L, Precast Factory, Al Mesaieed, Qatar (E-mail: [email protected])
**Assistant Professor, Dept. of Mechanical and Mechatronics, College of Engineering, Dhofar University, Salalah 211, P.O. Box 2509, Sultanate of
Oman (E-mail: [email protected])
***Ph.D. Candidate, Dept. of Civil and Environmental Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141, Korea (E-mail:
[email protected])
****Senior Research Fellow, Structural Engineering Research Institute, Korea Institute of Civil Engineering and Building Technology, Goyang, Korea (E-
mail: [email protected])
*****Member, Associate Professor, Dept. of Civil and Environmental Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141,
Korea (Corresponding Author, E-mail: [email protected], [email protected])

− 2211 −
 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

expressions of air blast pressure in the time domain as a function effects on structures. Computational methods used in these
of surface explosion charge weight, distance to surface, and programs are generally categorized into two types: (a) those used
structure height. They also suggested functions for the ground for the prediction of blast loads on structures and (b) those for the
shock time history spectral density, an envelop function, and a calculation of structural response. Many of those can take into
duration (Wu and Hao, 2005). There are several other researchers account the time varying load, nonlinear material properties,
whose works related to blast wave parameters have been large displacement and the fluid-structure interaction (De Silva,
reported, and the representative achievements among those are 2010).
explained in this review paper in the following sections. There are four types of explosions such as physical, chemical,
The National Fire Protection Association’s Guide to Fire and electrical, and nuclear explosions. The physical explosion means
Explosion Investigation, NFPA 921 (2008), simply defines an the physical gas dynamic and thermodynamic effects including
explosion as “the sudden conversion of potential energy (chemical the bursting of a pressure vessel and/or a rapid phase transition.
or mechanical) into kinetic energy with the production and Chemical explosion involves exothermic reactions and combustion
release of gas under pressure. These high-pressure gases then do explosion from fuel (e.g., natural gas and propane). Electrical
mechanical work such as moving, changing, or shattering nearby explosion occurs from an instantaneous release of electrical
materials.” An ideal blast wave profile can be completely energy such as an arc event or other electrical failure (fault).
defined using blast wave parameters such as blast wave arrival Nuclear explosion is the result of military weapons based on
time (ta), blast overpressure (Ps), positive phase duration (td), nuclear fusion or fission of atomic nuclei (Pape et al., 2009).
blast underpressure (Ps−), negative phase duration (td−), wave Explosive materials also might be classified considering their
decay parameter (b), positive phase impulse (is), and negative physical states (solid, liquid, or gas). On the other hand, military
phase impulse (is−). explosives are divided into two general classes, high and low
The steps for protecting the occupants of a building from the explosives, according to their rate of decomposition. Solid
blast effect include defining of (a) the maximum charge weight explosives usually mean high explosives, and other materials
used, (b) detonation locations, and (c) blast wave parameters. For such as flammable chemicals and propellants might be classified
the prediction of blast wave parameters from a high explosive as low explosive materials. A variety of substances used in the
detonation, unfortunately, the general engineering community manufactures of chemicals, fuels, and propellants are available in
has limited knowledge due to the nature of the explosives and the liquid or gaseous forms.
public security. The primary and widely used Unified Facilities Blast loads on structures can be categorized into two major
Criteria (UFC 3-340-02) (2008) is open to public, and it contains cases in terms of the confinement of the explosive charge
blast wave parameter curves developed by Kingery and Bulmash
for TNT bursts at standard atmospheric pressure and temperature. Table 1. Categories of Explosions Based on Confinement
UFC presents two sets of conditions, of which one is for Charge confinement Category
spherical bursts in free air and the other for hemispherical bursts 1. Free air blast (Spherical explosions)
at the ground surface. Unconfined explosions 2. Air burst
For last few decades, many computer programs such as 3. Surface burst (Hemispherical explosions)
BLASTX, CTH, SHAMRC, FEFLO, FOIL, DYNA3D, ALE3D, 4. Fully vented
LS-DYNA, Air3D, CONWEP, AUTODYN, ABAQUS, FEFLO, Confined explosions 5. Partially confined
FOIL, and 3D BLAST have been developed to simulate the blast 6. Fully confined

Fig. 1. Blast Pressure Effects on a Structure by: (a) Spherical, (b) Hemispherical Explosions

− 2212 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

(unconfined and confined explosions), and the classifications items, explosive confinement (casing, containers, etc.), and the
might be further specified if the blast loading is generated within pressure range (near, intermediate or far ranges) might affect the
the structures or is imposed on the outer surface of structures. equivalency of the material compared to TNT (Unified Facitilities
The detailed classification is summarized in Table 1 (Unified Criteria 3-340-02, 2008). TNT equivalents of some commonly
Facitilities Criteria 3-340-02, 2008). Spherical explosions (Fig. used explosives are given in Table 2 (Mays and Smith, 1995;
1(a)) are defined as “an explosion, which occurs in free air, Bangash and Bangash, 2006). Mathematically, TNT equivalent
produces an initial output, whose shock wave propagates away charge is written as:
from the center of the detonation, striking the protective structure Qx
without intermediate amplification of its wave.” However, W= Wx (1)
QTNT
hemispherical explosions (Fig. 1(b)) are defined as: “a surface
burst explosion will occur when the detonation is located close to where W is the equivalent charge in TNT, Wx is the mass of a
or on the ground so that the initial shock is amplified at the point particular explosive, Qx is the mass specific energy of the
of detonation due to the ground reflections” (Unified Facitilities particular explosive, and QTNT is the mass specific energy of
Criteria 3-340-02, 2008). TNT.
In this paper, a comprehensive review of analytical and TNT equivalence is often estimated by the ratio of the two
empirical estimations for incident blast pressure is presented so heats of explosives and TNT. However, it is also possible to
that the interested practitioners and designers can understand, compute TNT equivalence using the maximum overpressure or
compute, and compare the blast wave profiles. Among several impulse (Esparza, 1986; Krauthammer, 2008; TM-855-1, 1986).
dozens of available equations, we selected most representative For the conservative estimation, it is recommended to increase
expressions by comparing with UFC curves, and the values by TNT equivalent weight by 20% in order to compensate for
those equations are investigated and compared in this paper. The unknown factors such as the variation of unexpected shock wave
limitations of the representative empirical equations are explored, reflections, construction methods and quality of construction
and the equations that yield closest value to that of UFC curves materials. This total charge weight is known as “effective charge
are suggested. Blast load is quite a broad topic. Therefore, to weight” (Unified Facitilities Criteria 3-340-02, 2008).
narrow down the scope, we investigate the incident pressure in
unconfined spherical and hemispherical explosions generated by 3. Blast Scaling Law
solid chemical explosives. Additionally, because much literature
in this field is classified as confidential for security purpose, only Scaled parameters are determined from tests conducted in
a representative discussion is written down. much smaller scales, and are used to predict the properties of
large-scale explosions for various distances and energies (Conrath,
2. TNT Equivalence 1999; Sachs, 1944). Hopkinson-Cranz blast scaling law, in the
form of the cube root scaling, is the most commonly used blast
Explosives can vary in both composition and detonation pressure. scaling law. It was independently formulated by Hopkinson
Each material is usually studied by the shattering effect of its (1915) (Hopkinson, 1915) and Cranz (1926) (Cranz, 1926). In
sudden release of energy or blast pressure. Thus, in this review, the law, self-similar blast waves are produced at identical scaled
we consider TNT as a reference explosive (Chock, 1999). The distances by detonating two explosive charges of similar
mass of an explosive other than TNT calculated by multiplying a geometry and of the same explosive but with different sizes in
conversion factor based on its specific energy and that of TNT. In the same atmosphere. The scaled distance or the proximity factor
addition to the amount of energy, other factors such as explosive (Z) in m/kg1/3 is defined as (Mays and Smith, 1995; Uddin, 2010;
material shape (flat, square, round, etc.), the number of explosive UNODA, 2011):

R
Z = (2)
Table 2. Conversion Factors for Explosives W
1/ 3

Mass specific TNT where R is the distance from the center of a spherical charge in
Explosive energy equivalent
Qx(kJ/kg) Qx/QTNT meters (m), and W is the charge mass expressed in kilograms
Compound B (60% RDX 40% TNT) 5190 1.148 (kg) of TNT (Kim et al., 2009). Similarly, blast wave parameters
RDX (Cyclonite) 5360 1.185 such as the arrival time and positive and negative phase
HMX 5680 1.256 duration and impulse can be scaled using the Hopkinson-
Nitroglycerin (liquid) 6700 1.481 Cranz scaling law. Fig. 2 shows the Hopkinson-Cranz scaling
TNT 4520 1.000 law schematically.
Blasting Gelatin (91% nitroglycerin, 7.9%
4520 1.000
A structure located at a distance (R) from a charge of diameter
nitrocellulose, 0.9% antacid, 0.2% water) (d) will be subjected to a blast wave of the peak overpressure (P),
60% Nitroglycerin dynamite 2710 0.600 positive phase duration (td), and impulse (is). Hopkinson-Cranz
Semtex 5660 1.250 scaling law then states that a structure located at a distance (λR)

Vol. 21, No. 6 / September 2017 − 2213 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

Table 3. Expected Shock Wave Effects on Objects

Overpressure
Expected damage
(MPa × 10−3)
1.0-1.5 Window glass cracks
3.5-7.6 Minor damage in some buildings
7.6-12.4 Metal panels deformed
12.4-20 Concrete walls damaged
Over 35 Wooden buildings demolished
27.5-48 Major damage to steel objects
40-60 Heavy damage to reinforced concrete buildings
70-80 Probable demolition of most buildings

expected effects of a shock waves on objects (Kinney and

Graham, 1985)), the positive phase is typically considered
more important than the negative phase because of the large
Fig. 2. Hopkinson-Cranz blast scaling law amplitude of the overpressure (Ps) and the concentrated
impulse (is) (Uddin, 2010). On the other hand, the negative
from the center of a charge of diameter (λd) will experience a phase has often been ignored because the magnitude is
blast wave of a typical form with amplitude (P), duration (λtd) relatively small, and is difficult to be measured (Nassr, 2012).
and impulse (λis). All characteristic times are modified by the However, some studies have shown that, for scaled distances
same factor (λ) as that of length (Conrath, 1999). (Z) larger than 20 m/kg1/3 (especially for scaled distances (Z)
larger than 50 m/kg1/3), the influence of the negative phase
4. Profiles of Blast Waves cannot be neglected (Larcher, 2008). In addition, for relatively
flexible structures, the negative phase pressure might be
After detonation of a high explosive in free air, the expansion of included (Uddin, 2010).
the hot gases at extremely high pressure causes a shock wave that
moves outward at a high velocity. The pressure-time history of an 4.1 Positive Phase/Shock Wave
idealized explosion at a fixed distance (R) from the center of the In the positive phase, the shock wave travels along a point of
explosion can be described as shown in Fig. 3, which is called the consideration after detonation of an explosive in a time interval,
Friedlander waveform. The pressure-time history can be divided the so-called arrival time (ta). An instantaneous increase in the
into positive and negative phases. The ideal Friedlander waveform ambient pressure occurs due to the highly compressed air of the
has blast wave parameters such as the arrival time (ta), peak positive shock front, and the pressure reaches its peak value, which is
overpressure (Ps), positive duration (td), under or negative pressure known as the incident peak pressure or peak overpressure (Ps).
(Ps−), negative duration (td−), wave decay parameter (b), positive After reaching the peak, pressure decays back to normal
impulse (is), and negative impulse (is−), all of which are discussed in atmospheric pressure (Po) over a period known as the positive
detail in the following sections. phase duration (td). The area under the pressure-time pulse over the
For the study of blast wave effects on structures and also for positive phase is referred to as the positive incident impulse (is)
the design of structures or components (Table 3 shows (Larcher, 2008). Blast pressure variation is the difference of the
peak overpressure (Ps) and negative pressure (Ps−). For the
calculation and presentation of positive blast pressure profiles,
numerous expressions have been proposed. Flynn proposed a linear
decaying pressure profile, which is given by (Chock, 1999):

⎛ t ⎞
P (t ) = Po + Ps ⎜1 − ⎟, 0 < t ≤ td (3)
⎝ td ⎠

Considering the decaying nature of blast profile, a better form

was presented by Ethridge in his 1965 work (Chock, 1999):
−ct
P (t ) = Po + Ps e (4)
where t is measured from the time of arrival (ta). The curve
fitting of Eq. (4) can be performed using the peak overpressure
Fig. 3. Ideal Temporal Pressure Profile Resulting from Explosion in and the wave decay rate as parameters. The next extension of this
Air process is to fit experimental results with three different parameters.

− 2214 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

experimental data at different scaled distances and charge sizes.

The formulae for a spherical TNT explosion in air can be used
for hemispherical explosions with increasing the explosive
weight by multiplying a reflection factor, 2η. The coefficient (η)
accounts for the energy used for deformation of the base
material, and the values for different base materials are listed in
Table 4 (Jeremić and Bajić, 2006).
On the other hand, Unified Facilities Criteria (Unified Facitilities
Criteria 3-340-02, 2008) is widely used as a reference for the
design of protective structures. Figs. 5(a) and (b) show positive
phase wave parameters for a spherical explosion in free air and
for a hemispherical TNT explosion on the ground, respectively
(Unified Facitilities Criteria 3-340-02, 2008).
Fig. 4. Triangular Pressure-time Profile
The empirical and analytical equations for spherical airbursts
presented by several researchers are given below. These presented
This form is usually called as the modified Friedlander’s equation, equations have been modified to have the unit of megapascals
which is expressed as (Chock, 1999; Kangarlou, 2013; Kadid et (MPa), while W is expressed in kg and Z in m/kg1/3.
al., 2012): Sadovskyi (1952) (Kangarlou, 2013; Jeremić and Bajić, 2006;
Chang and Young, 2010, Sadovskiy, 2004)
⎛ t ⎞ −bt td
P (t ) = Po + Ps ⎜1 − ⎟e (5)
⎝ td ⎠ 0.085 0.3 0.82
Ps = + 2+ 3 (6)
Z Z Z
where b is the decay coefficient. For the simplicity, the blast
Brode (1955) (Brode, 1955; Mays and Smith, 1995; Kadid et
profile can by approximated by the linearly decaying triangular
al., 2012; Chang and Young, 2010; Yin et al., 2009; Abdollahzadeh
pulses shown in Fig. 4. This linearly decaying triangular profiles
and Nemati, 2013; Low and Hao, 2001; Smith and Hetherington,
have the same initial peak overpressure, but have different
1994)
durations depending on the expected time of maximum structural
response (Beshara, 1994; Chock, 1999). ⎧ 0.67
⎪⎪ + 0.1 for (1 < Ps )
Z3 (7)
Ps = ⎨
4.1.1 Overpressure ⎪ 0.0975 + 0.1455 + 0.585 − 0.0019 for (0.01 ≤ P ≤ 1)
⎪⎩ Z Z2 Z3
s

Several empirical relations have been presented in literature to

predict blast overpressure based on the analysis of large sets of Naumyenko and Petrovskyi (1956) (Henrych and Major, 1979;

Table 4. Values of the Coefficient η for Various Base Materials

Reinforced Compact sandy clay; Medium
Type of material Steel plate Concrete, rock Water
concrete plate Clay compact soil
η 1 0.95-1 0.85-0.9 0.7-0.8 0.6-0.65 0.55-0.6

Fig. 5. Positive Phase Parameters for: (a) Spherical, (b) Hemispherical TNT Explosions

Vol. 21, No. 6 / September 2017 − 2215 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

Low and Hao, 2001) Gelfand and Silnikov (2004) (Gelfand and Silnikov, 2004)
⎧ 1.050 ⎪⎧1.7 × 10 exp( −7.5 × Z ) + 0.0156 for ( 0.1 ≤ Z < 8 )
for ( Z ≤ 1)
3 0.28

⎪⎪ − 0.0981 Ps = ⎨
Z3 (15)
Ps = ⎨ (8) ⎪⎩ 8 × 10 exp( −10.7 × Z )
3
for (8 ≤ Z ) 0.1

⎪ 0.0745 + 0.250 + 0.637 for (1< Z ≤ 15 )

⎪⎩ Z Z2 Z3 Bajić (2007) (Bajić, 2007)
Adushkin and Korotkov (1961) (Adushkin and Korotkov, W 1/ 3
W 2/ 3
W
1961; Pierre-Emmanuel Sauvan, 2012) Ps = 0.102 + 0.436 + 1.4 (16)
R R 2
R 3

0.08 0.28 0.322 National Defense Engineering Design Specifications (NDEDS),

Ps = + 2 − for ( 0.8 ≤ Z < 18 ) (9)
Z Z Z3 China (Li and Ma, 1992)
Henrych and Major (1979) (Henrych and Major, 1979; Kangarlou, 2 3
2013; Kadid, Nezzar and Yahiaoui, 2012; Yin et al., 2009; ⎛ 3W ⎞ ⎛ 3W ⎞ ⎛ 3W ⎞
Ps = 0.084 ⎜⎜ ⎟⎟ + 0.27 ⎜⎜ R ⎟⎟ + 0.7 ⎜⎜ R ⎟⎟ (17)
Abdollahzadeh and Nemati, 2013; Low and Hao, 2001; Saska et ⎝ R ⎠ ⎝ ⎠ ⎝ ⎠
al., 2011)
For hemispherical surface bursts, the empirical formulae for
⎧1.380 0.543 0.035 0.000613 the peak overpressure proposed by different researchers are
⎪ Z + Z2 − Z3 + Z4
for (0.05 ≤ Z ≤ 0.3)
summarized below. All the equations have been modified to
⎪
⎪ 0.607 0.032 0.209 (10) have the unit of megapascals (MPa).
Ps = ⎨ − + for (0.3 < Z ≤ 1)
⎪ Z Z2 Z3 Newmark and Hansen (1961) (Kadid et al., 2012; Vijayaraghavan
⎪ 0.0649 0.397 0.322 et al., 2012; Newmark and Hansen, 1961)
⎪ + + for (1 < Z ≤ 10)
⎩ Z Z2 Z3
1
W ⎛W ⎞ 2
Held (1983) (Chang and Young, 2010; Held, 1983; Goel et al., Ps = 0.6784 + 0.294 ⎜ 3 ⎟
R3 ⎝R ⎠
2012) (18)
Wu and Hao (2005) (Wu and Hao, 2005)
W2/ 3

Ps = 2 (11) −
R 2
⎧ ⎛ R ⎞
2.56

⎛ R ⎞
⎪1.059 ⎜ ⎟ − 0.051 for ⎜ 0.1 ≤ ≤ 1⎟
⎪ ⎝W ⎠ ⎝ W ⎠
1/ 3 1/ 3

Kinney and Graham (1985) (Kinney and Graham, 1985; Ps = ⎨ − (19)

⎛ R ⎞ R
2.01
Nassr, 2012; Larcher, 2008; Kangarlou, 2013) ⎪ ⎛ ⎞
Kinney and Graham’s equation does not use any polynomial base; ⎪ 1.008 ⎜⎝ W ⎟⎠ for ⎜1 <
⎝ W
≤ 10 ⎟
⎠
⎩
1/ 3 1/ 3

therefore, this equation does not have limits on the valid range.
Ahmad et al. (2012) (Ahmad et al., 2014)
⎡ ⎛ Z ⎞2 ⎤ −2.67
808 ⎢1 + ⎜ ⎟ ⎥ ⎛ R ⎞
Ps = 2.46 ⎜ ⎟ (20)
⎢⎣ ⎝ 4.5 ⎠ ⎥⎦ ⎝W ⎠
1/ 3

Ps = Po
⎡ ⎛ Z ⎞2 ⎤ ⎡ ⎛ Z ⎞2 ⎤ ⎡ ⎛ Z ⎞2 ⎤ (12)
Siddiqui and Ahmad (2007) (Ahmad et al., 2014)
⎢1 + ⎜ ⎟ ⎥ × ⎢1 + ⎜ ⎟ ⎥× ⎢1 + ⎜ ⎟ ⎥
⎢⎣ ⎝ 0.048 ⎠ ⎥⎦ ⎢⎣ ⎝ 0.32 ⎠ ⎥⎦ ⎢⎣ ⎝ 1.35 ⎠ ⎥⎦
−1.91
⎛ R ⎞ ⎛ R ⎞
Mills (1987) (Kadid, Nezzar and Yahiaoui, 2012; Chang and Ps = 1.017 ⎜ ⎟ for ⎜ 1 ≤ ≤ 12 ⎟ (21)
⎝W ⎠ ⎝ W ⎠
1/3 1/ 3

Young, 2010; Low and Hao, 2001; Vijayaraghavan et al., 2012)

Iqbal and Ahmad (2009) (Iqbal and Ahmad, 2009)
1.772 0.114 0.108
Ps = − + (13)
Z3 Z2 Z −1.96
⎛ R ⎞ ⎛ R ⎞
Hopkins-Brown and Bailey (1998) (Hopkins-Brown and Bailey, Ps = 1.026 ⎜ 1/3 ⎟ for ⎜1 ≤ 1/3 ≤ 12 ⎟ (22)
⎝W ⎠ ⎝ W ⎠
1998)
Ahmad et al. (2012a) (Ahmad et al., 2012)
⎧ 1.935 0.2353 0.01065
⎪⎪−1.245 + Z + Z 2 − Z 3 for ( 0.05 ≤ Z ≤ 1.15)
Ps = ⎨ (14) −2.868
⎛ R ⎞
⎪ 0.0707 + 0.3602 + 0.4891 for (1.15 < Z ≤ 40 ) Ps = 1.5956 ⎜ ⎟ for ( R < 5 ) (23)
⎪⎩ Z2 Z3 ⎝W ⎠
1/ 3
Z

Table 5. Simplified Kingery Air Blast Coefficients for Blast Overpressure

1/3
Z (m/kg ) A 1 B 1 C 1 D1 E 1 F
1 G
1

0.2-2.9 7.2106 -2.1069 -0.3229 0.1117 0.0685 0 0

2.9-23.8 7.5938 -3.0523 0.40977 0.0261 -0.01267 0 0
23.8-198.5 6.0536 -1.4066 0 0 0 0 0

− 2216 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

UFC.

4.1.2 Arrival Time of Blast Wave

The empirical relations for the arrival time (ta) of a shock wave
front from the center of an explosion to a structure are not
commonly available and are not usually included in most
studies. In this review, the arrival time formulas given by
different researchers are summarized. All equations presented
have been modified to have the unit of milliseconds (ms).
Kinney and Graham (1985) (Kinney and Graham, 1985)
1/ 2

⎡ ⎤
1 R⎢ 1 ⎥ 1 R⎛
1 ⎞
ta =
∫
⎢
ao r ⎢1 + 6 Ps
c
⎥ dR =
⎥ ao ∫
rc
⎜
M
⎝ x
⎟dR
⎠
(25)
⎢⎣ 7 Po ⎥⎦

where ao is the speed of sound in the undisturbed atmosphere, Mx

Fig. 6. Comparison of Positive Overpressure Computed by Empir- is the Mach number and rc is the charge radius. The given
ical Formulae and UFC 3-340-02 (Fig. 5(a)) integration equation can be calculated using the graphical method,
but this method is less precise than numerical calculation. For
Table 6. Values of Parameters at Sea Level example, in order to calculate the arrival time using the graphical
Temperature Air density Ambient Height above Sound speed
method (Kinney and Graham, 1985), the overpressures (Ps)
(°C) 3
(kg/m ) pressure (MPa) sea level (m/s) (using Eq. (12)) at different scaled distances are converted into
15 1.225 0.101325 0 340 Mach numbers (Mx) and substituted into Eq. (25). Reciprocals of
the Mach number against the distance are plotted in Fig. 7. The
curve is extrapolated to the charge radius (0.227 m) and the area
Swisdak (1994) (Swisdak Jr., 1994)
under the curve to the distance of 20 m is calculated as 18.866 m.
2 3
Ps = exp ( ( A1 + B1 × ( ln ( Z ) ) + C1 × ( ln ( Z ) ) + D1 × ( ln( Z ) ) Dividing by the speed of sound, 340 m/s, the arrival time is
6
(24)
4 5
+ E1 × ( ln ( Z ) ) + F 1 ( ln ( Z ) ) × G1 ( ln ( Z ) ) ) × 10
–3 obtained as 58.8 ms. The expression for the arrival times are
listed:
where Z is the scaled range and A1, B1, C1, D1, F1, and G1 are the Swisdak (1994) (Swisdak Jr., 1994)
simplified Kingery air blast coefficients given in Table 5. All of 2
ta = exp ( A2 + B2 × ( ln ( Z ) ) + C2 × ( ln ( Z ) ) + D2 × ( ln ( Z ) )
3

the Swisdak’s equations uses Kingery air blast coefficients. In 4 5 6 1 3

(26)
+ E 2 × ( ln( Z ) ) + F2 × ( ln ( Z ) ) + G2 × ( ln ( Z ) ) ) × W
⁄

addition to these empirical relations, there exist other experimental

diagrams presented by Kingery and Bulmash, and Baker. Kingery
and Bulmash suggested diagrams valid up to Z = 40 m/kg1/3,
while Baker’s diagram covers up to Z = 1000 m/kg1/3. Fig. 6
shows the log-log plot of overpressure versus scaled distance (Z)
in the conditions listed in Table 6 for spherical air bursts. To
compare the pressures quantitatively, the empirical relations by
hemispherical burst are converted to that of the spherical burst
for a steel plate which has the reflection factor of 2. The pressures
calculated from the above empirical relations are compared with
the results by UFC 3-340-02 (2008). The difference between the
empirical relations and UFC charts is significant, especially in
the range of (0.2 < Z (m/kg1/3) < 1.0) while the variation is
negligible for (Z (m/kg1/3) > 1.0.) It means that overpressure at
large distances from an explosive charge can be predicted with
reasonable confidence by means of empirical equations.
However, when the charge is close to a target, the result might
not be accurate. Bajic’s equation (Eq. (16)) gives the largest
values, and the equation by Gelfand and Silnikov (Eq. (15))
gives the smallest values. The empirical equation by Kinney and
Graham (Eq. (12)) give the values very close to the result by Fig. 7. Estimation of Arrival Time

Vol. 21, No. 6 / September 2017 − 2217 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

Table 7. Simplified Kingery Air Blast Coefficients for Blast Arrival Time
Z (m/kg )1/3
A 2 B 2 C2 D 2 E 2 F 2 G
2

0.06-1.50 -0.7604 1.8058 0.1257 -0.0437 -0.0310 -0.00669 0

1.50-40 -0.7137 1.5732 0.5561 -0.4213 0.1054 -0.00929 0

where Z is the scaled range and A2, B2, C2, D2, F2, and G2 are the Kinney and Graham’s equation (Eq. (25)) produces the largest
simplified Kingery air blast coefficients as given in Table 7. values among all the presented empirical equations.
Wu and Hao (2005) (Wu and Hao, 2005)
4.1.3 Positive Phase Duration
R W− 1.4 0.2

The duration of a blast wave might be defined as the time

ta = 0.34 (27)
ao difference between the passing of the shock front and the passing
of the end of the positive pressure phase (Goel, Matsagar, Gupta
where ao is the speed of sound in air, given in Table 6.
and Marburg, 2012). The positive phase duration might be
Iqbal and Ahmad (2009) (Iqbal and Ahmad, 2009)
calculated using the scaled distance (Z), or the mass of charge
R W− 1.4 0.2
(W) and the standoff distance (R). Several empirical relations of
ta = 560 (28)
ao positive phase duration (td) are listed, and equations are modified
to have the unit of milliseconds (ms) as follows:
Ahmad et al. (2012a) (Ahmad et al., 2012)
Sadovskyi (1952) (Kangarlou, 2013; Jeremić and Bajić, 2006;
−0.138 Sadovskiy, 2004)
4637 ⎛ R ⎞
ta = ⎜ ⎟ (29)
ao ⎝ W ⎠ 1/ 3
td = 1.2 6 W R (31)
Ahmad et al. (2014) (Ahmad et al., 2014) Newmark (1972) (Beshara, 1994; Newmark and Hansen,
1961)
−0.996
8534 ⎛ R ⎞ For a surface burst, the total positive phase duration of blast
ta = ⎜ ⎟ (30)
ao ⎝ W 1/3 ⎠ overpressure is expressed in milliseconds (ms) as:
Figure 8 shows the standard doubly logarithmic plot of the td = 10 ⋅ W 1/ 3
(32)
blast arrival time (ta) with respect to the scaled distance (Z). The
Henrych (1979) (Henrych and Major, 1979)
variation among the results produced by the equations given by
researchers is larger in the region (0.2 < Z (m/kg1/3) < 1) while it 1 3
td = W ( 0.107 + 0.444Z + 0.264Z – 0.129Z + 0.0335Z )
⁄ 2 3 4
(33)
is smaller for the scaled distance range (Z (m/kg1/3) > 1). The for ( 0.05 ≤ Z ≤ 3 )
results obtained from Wu and Hao’s equation (Eq. (27)) show
Kinney and Graham (1985) (Beshara, 1994; Nassr, 2012;
closer agreement with the UFC chart values. On the other hand,
Kangarlou, 2013)
⎡ ⎛ Z ⎞ ⎤ 10

980 ⎢1 + ⎜ ⎟ ⎥
⎣⎢ ⎝ 0.54 ⎠ ⎥⎦
(34)
td = W 1/ 3

⎡ ⎛ Z ⎞ ⎤⎡ ⎛ Z ⎞ ⎤ 3

⎛ Z ⎞
6 2

⎢1 + ⎜ ⎟ ⎥ ⎢1 + ⎜ ⎟ ⎥ 1+ ⎜ ⎟
⎣⎢ ⎝ 0.02 ⎠ ⎦⎥ ⎣⎢ ⎝ 0.74 ⎠ ⎦⎥ ⎝ 6.9 ⎠

Swisdak (1994) (Swisdak Jr., 1994)

2 3
td = exp ( A3 + B3 ( ln ( Z ) ) + C3 × ( ln ( Z ) ) + D3 × ( ln ( Z ) )
(35)
4 5 6 1 3
+ E3 × ( ln ( Z ) ) + F3 × ( ln ( Z ) ) + G3 × ( ln( Z ) ) ) × W
⁄

where Z is the scaled range and A3, B3, C3, D3, F3, and G3 are the
simplified Kingery air blast coefficients, given in Table 8.
Smith (Lam et al., 2004)
⎛ R ⎞
0.25 + 0.27 log10
⎜ 1/3 ⎟
td ≈ W 1/ 3
× 10 ⎝W ⎠ (36)
Wu and Hao (2005) (Wu and Hao, 2005)
1.30 0.72
⎛ R ⎞ ⎛ R ⎞
Fig. 8. Comparison of Scaled Arrival Time Computed by Empirical td = 1.9 ⎜ 1/3 ⎟ + 0.5 ⎜ 1/3 ⎟ W 0.4 (37)
Formulae and UFC 3-340-02 ⎝W ⎠ ⎝W ⎠

− 2218 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

Table 8. Simplified Kingery Air Blast Coefficients for Positive Phase Duration
Z (m/kg ) 1/3
A 3 B3 C 3 D 3 E
3 F
3 G
3

0.2-1.02 0.5426 3.2299 -1.5931 -5.9667 -4.0815 -0.9149 0

1.02-2.80 0.5440 2.7082 -9.7354 14.3425 -9.7791 2.8535 0
2.80-40 -2.4608 7.1639 -5.6215 2.2711 -0.44994 0.03486 0

Iqbal and Ahmad (2009) (Iqbal and Ahmad, 2009) variation in values of the duration is larger for the scaled
0.92 0.89
distance range (0.2< Z (m/kg1/3)<1) and becomes negligible
⎛ R ⎞ ⎛ R ⎞ for the range (Z (m/kg1/3)>1). Iqbal and Ahmad’s equation
td = 4.6 ⎜ 1/3 ⎟ + 1.3 ⎜ 1/3 ⎟ W 0.52 (38)
⎝W ⎠ ⎝W ⎠ (Eq. (38)) gives the largest value, while Kinney and Graham’s
equation (Eq. (33)) yields the smallest over most of the range.
Ahmad et al. (2012a) (Ahmad et al., 2012)
Sadovskyi’s equation (Eq. (30)) calculates values relatively
−0.1 −0.472 closer to the UFC values.
⎛ R ⎞ ⎛ R ⎞
td = 0.3108 ⎜ 1/3 ⎟ + 2.462 ⎜ 1/3 ⎟ (39)
⎝W ⎠ ⎝W ⎠ 4.1.4 Positive Phase Impulse
Izadifard and Maheri (Izadifard and Foroutan, 2010) The incident overpressure impulse can be calculated by
integrating pressure-time curve over the positive phase duration
⎧W 1/ 3
× (−64.86Z +52.32Z -15.68Z +1.794Z+0.1034) for ( Z ≤ 0.37 )
4 3 2

⎪ (td). The positive impulse is the parameter that has central

⎪W
1/ 3
× (4.64Z − 3.86Z + 0.854)
2
for ( 0.37 < Z ≤ 0.82)
td = ⎨ importance for the loading of a structure and can be mathematically
⎪W
1/ 3
× (−2.97 X + 6.27 X + 0.358 X + 0.763)
3 2
for ( 0.82 < Z ≤ 2.5)
⎪
expressed as (Kinney and Graham, 1985):
⎩W
1/ 3
× (0.608 X − 2.38 + 5.62 X − 0.22)
3 2
for ( 2.5 < Z)
ta + td
(40)
where X = log10 ( Z ) (41)
is =
∫ [ P(t ) − P ] dt
ta
o (43)

Ahmad et al. (2014) (Ahmad et al., 2014) Positive impulse also can be calculated using scaled distance
(Z) or mass of charge (W) and standoff distance (R). All equations
−0.759
⎛ R ⎞ ⎛ R ⎞
2.159
presented have been modified to have units of pascal-second
td = 1.4 ⎜ ⎟ + 0.5 ⎜ ⎟ (42)
⎝W ⎠ ⎝W ⎠ (Pa-s). Several empirical relations of positive phase impulse (is)
1/ 3 1/ 3

are listed below:

Positive phase duration computed by above empirical
Sadovskyi (1952) (Henrych and Major, 1979; Kangarlou, 2013;
relations show dissimilar behaviors (Fig. 9). Therefore it is
Jeremić and Bajić, 2006; Chang and Young, 2010; Sadovskiy,
hard to predict the trend of positive phase duration, plotted
2004; Saska et al., 2011)
against scaled distance. However, it should be noted that the
3
W2 (44)
is = 200
R
and

⎧ 3
W2
⎪⎪( 34 ~ 36 ) × 9.81× for ( Z > 0.5 )
is = ⎨ R (45)
⎪147.15 × W for ( Z < 0.25 )
⎪⎩ R2
Henrych (1979) (Henrych and Major, 1979; Saska et al., 2011)

⎧ ⎛ 1115 629 100.4 ⎞

⎪9.81×W ⎜ 663− Z + Z − Z ⎟ for ( 0.4 ≤ Z ≤ 0.75)
1/3

⎪ ⎝ ⎠
2 3

is = ⎨ (46)
211
⎪ 9.81×W ⎛ −32.2 + − 216 80.1 ⎞
+ ⎟ for ( 0.75 < Z ≤ 3)
1/3

⎪⎩ ⎜
⎝ Z Z Z ⎠ 2 3

Held (1983) (Chang and Young, 2010; Held, 1983)

W 2/ 3

(47)
is = 300
R
Fig. 9. Comparison of Scaled Positive Phase Duration Computed
by Empirical Formulae and UFC 3-340-02 Kinney and Graham (1985) (Kinney and Graham, 1985;

Vol. 21, No. 6 / September 2017 − 2219 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

Table 9. Simplified Kingery Air Blast Coefficients for Positive Phase Impulse
Z (m/kg ) 1/3
A 4 B
4 C
4 D 4 E
4 F
4 G 4

0.2-0.96 5.522 1.117 0.6 -0.292 -0.087 0 0

0.96-2.38 5.465 -0.308 -1.464 1.362 -0.432 0 0
2.38-33.7 5.2749 -0.4677 -0.2499 0.0588 -0.00554 0 0
33.7-158.7 5.9825 -1.062 0 0 0 0 0

Nassr, 2012; Larcher, 2008; Kangarlou, 2013) 02. Henrych’s equation (Eq. (46)) values are the largest, while
the Kinney and Graham’s equation (Eq. (48)) yields the smallest
6.7 1 + ( Z / 0.23)
4

is = W 1/ 3 values among all the presented equations.

(48)
Z 1 + ( Z /1.55 )
2 3 3

4.1.5 Wave Decay Parameter

Swisdak (1994) (Swisdak Jr., 1994)
The wave decay parameter (b) of the modified Friedlander
2 3
i = exp ( A4 + B4 × ( ln ( Z ) ) + C4 × ( ln ( Z ) ) + D4 × ( ln( Z ) ) equation describes the shape of the temporal pressure profile.
(49)
4 5
E4 × ( ln ( Z ) ) + F4 × ( ln( Z ) ) + G4 × ( ln ( Z ) ) ) × W
6 1 3
⁄
This parameter is dimensionless unlike other blast wave
parameters, and influences on the length of the negative phase
where Z is the scaled range and A4, B4, C4, D4, F4, and G4 are the although the modified Friedlander equation describes only
simplified Kingery air blast coefficients, given in Table 9. positive phase. The negative phase is important when the value
Hopkins-Brown and Bailey (1998) (Hopkins-Brown and Bailey, of the wave decay parameter is less than one but becomes less
1998) significant for wave decay parameters greater than one (Kadid,
⎧ ⎛ 106.65 18.89 0.401 ⎞ Nezzar and Yahiaoui, 2012). Several empirical relations are
⎪W ⎜ 308.2 − + − ⎟ for ( 0.05 ≤ Z ≤ 1.15 )
1/ 3

⎪ ⎝ Z Z Z ⎠ 2 3 summarized below:
is = ⎨ (50)
⎪ W ⎛ 196.0 8.62 30.5 ⎞ Kinney and Graham (1985) (Kinney and Graham, 1985)
1/ 3

⎜ 1.79 + + + ⎟ for (1.15 < Z ≤ 40 )

⎪⎩ ⎝ Z Z Z ⎠ 2 3
The relationship between blast impulse and the decay parameter
can be expressed as:
Figure 10 shows the log-log plot of the positive phase impulse
td
versus the scaled distance (Z). The variation among positive ⎡1 1 −b ⎤
s d ⎢ − 2 (1− e ) ⎥
∫
is
phase impulse values is larger for scaled distances (0.2 < Z (m/ = Impulse = pdt = Pt (51)
A ⎣b b ⎦
kg1/3) < 1) and smaller for scaled distances (1 < Z (m/kg1/3)). The 0

positive phase impulse values computed by different researchers Dharaneepathy (1993) (Dharaneepathy, 1993)
show different tendencies in different regions. The positive phase
b = 3.18Z − 0.58
(52)
impulse values computed from the equation of Kinney and
Graham (Eq. (48)) is in close agreement with that of UFC 3-340- Lam et al. (2004) (Lam et al., 2004)
b = Z 2 − 3.7Z + 4.2 (53)

Fig. 10. Comparison of Scaled Positive Phase Impulse Computed

by Empirical Formulae and UFC 3-340-02 Fig. 11. Wave Decay Parameter Versus Scaled Distances

− 2220 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

Larcher (2008) (Larcher, 2008; Kangarlou, 2013) Figure 12(a) and (b) show the negative phase shock wave
−1.1975 parameters for spherical and hemispherical explosions in free air
b = 5.2777Z (54)
and on the surface (Unified Facitilities Criteria 3-340-02, 2008).
Teich and Gebekken (Kangarlou, 2013)
b = 1.5Z −0.38
for ( 0.1 < Z < 30 )
4.2.1 Underpressure
(55)
The shock wave is followed by a rarefaction wave. It follows
Figure 11 shows the relationship of wave decay parameter from its nature that:
versus scaled distance (Z). The curves for wave decay parameter
0 < Ps− < Po (58)
given by the researchers show different trends.
Following are different empirical relations for underpressure
4.2 Negative Phase/Rarefaction Wave (Ps−) in the units of megapascals (MPa). The size of the negative
For the negative phase or the partial vacuum phase, after the pressure as well as the duration of the negative part can be taken
positive phase duration (td), the pressure decreases below the from a diagram by Drake (also shown in Krauthammer, 2008
reference pressure to the maximum negative pressure (Ps−), and (Krauthammer, 2008)). Drake used experimental data to obtain
then returns back to the normal atmospheric pressure (Po) in time the mentioned diagram. The diagram can be approximated by
(td−), giving a negative phase impulse (is−). The maximum negative (Larcher, 2008; Cabello, 2011):
pressure or underpressure (Ps−) has much smaller amplitude than
the positive overpressure, while the negative phase duration (td−) ⎧ 0.035
⎪ for ( Z > 3.5 )
Ps− = ⎨ Z
is much longer than the positive phase duration (td). The negative ⎪0.01 for ( Z ≤ 3.5 )
(59)
phase pressure variation versus time can be expressed as follows ⎩
(Brode, 1955; Nassr, 2012): From the work of Brode and the experimental investigation of
−4t Henrych, the underpressure can be written as (Henrych and
−⎛ t ⎞⎛ t ⎞ td− Major, 1979; Mays and Smith, 1995):
P (t ) = Po − Ps ⎜ − ⎟ ⎜1 − − ⎟ e (56)
⎝ td ⎠ ⎝ td ⎠ 0.034335
−
Ps− = − for( Z > 1.6) (60)
where Po is the ambient pressure, P is the peak negative phase
s Z
overpressure, td− is the duration of the negative phase, and t is the
Figure 13 compares the results of available empirical relations
time measured from the end of the positive phase duration (ta+
for underpressure with the value obtained from UFC 3-340-02
td). Larcher (2008) presented a piecewise equation to approximate
(2008). It should be noted that the results of empirical equations
the form of the negative phase (Larcher, 2008):
do not show good agreement with the values of UFC charts.
⎧ 2P− ⎛ −
t ⎞ However, both empirical relations give close values to each other
⎪ Po − −s ( t − td ) for ⎜ td < t < td + d ⎟ for the scaled distance range (Z (m/kg1/3) > 3.5).
⎪ td ⎝ 2⎠
⎪ 2P −
⎛ −
⎞
⎪ t
P ( t ) = ⎨ Po − −s ( td + td− − t ) for ⎜ td + d < t < td + td− ⎟ 4.2.2 Negative Phase Duration
t ⎝ 2 ⎠ (57)
⎪ d The time duration in which air pressure falls below the
⎪
⎪ P o for ( t > t d + t
−
d ) atmospheric pressure and then returns back to the atmospheric
⎪⎩ pressure slowly is known as negative phase duration. The negative

Fig. 12. Negative Phase Parameters for: (a) Spherical, (b) Hemispherical TNT Explosions

Vol. 21, No. 6 / September 2017 − 2221 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

Fig. 13. Comparison of Underpressure by Empirical Formulae and Fig. 15. Comparison of Scaled Negative Phase Impulse by Empiri-
UFC 3-340-02 (Fig. 12(a)) cal Formula and UFC 3-340-02 (Fig. 12(a))

phase may last up to three times longer than the positive phase. Teich and Gebbeken presented the following equation to
The duration of the negative phase (td−) expressed in milliseconds compute the time at which maximum negative pressure occurs
(ms) by Krauthammer (2008) can be described with the following (Goel et al., 2012):
equation (Larcher, 2008; Cabello, 2011):
b +1
td−− peak = td (63)
⎧ b
for ( Z < 0.3)
1/ 3

10.4 × W
⎪⎪
t d = ⎨( 3.125 × log ( Z ) + 12.01) × W for ( 0.3 ≤ Z ≤ 1.9 ) (61) Figure 14 shows the comparison of negative phase duration
− 1/ 3

⎪ computed from the empirical equation of Krauthammer (Eq.

⎪⎩ 13.9 × W 1/ 3
for (1.9 < Z )
(61)) with that of UFC 3-340-02. There is a very small difference
From the theoretical work of Brode and the experimental in the estimation of the negative phase duration.
investigation of Henrych, time duration of the negative pressure
can be calculated by (Cabello, 2011): 4.2.3 Negative Phase Impulse
The area under the negative phase of the pressure time profile
−
td = 1.25 ×W 1/3
(milliseconds) (62) is called as the negative phase impulse or underpressure impulse.
Teich and Gebbeken presented the following equation to
compute the underpressure impulse (Goel et al., 2012):
Ps td − b
is− = e (64)
b2
Brode’s solution for specific impulse in this phase (is-) is given
by (Mays and Smith, 1995):
⎡ 1 ⎤
is− = is ⎢1 − (65)
⎣ 2 Z ⎥⎦
Figure 15 illustrates the comparison of results by empirical
relations and UFC 3-340-02. Large difference of the impulse is
observed for the range of (2 < Z (m/kg1/3)), and the results get
closer to each other in the scaled distance range of (2< Z (m/kg1/3)<6).

5. Examples of Positive and Negative Blast Pro-

files

Fig. 14. Comparison of Scaled Negative Phase Duration by Empir- In this section, an example of a blast wave profile is presented
ical Equation and UFC 3-340-02 (Fig. 12(a)) using the blast wave parameters recommended in the above

− 2222 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

relations for the standoff distances R = 5, 10, 15, 20 m and the

TNT charge weight W = 80 kg are compared with the profile by
UFC. The pressure values are very similar, but large variations
are observed in the arrival time and in the positive phase
duration.
There are few empirical relations available in literature for
negative phase of blast wave, but they do not show good agreement
with the result obtained by UFC 3-340-02. Fig. 17 shows the plot
of Eq. (56). for negative phase profile. In this equation, the
empirical equations of Drake (Eq. (59)) and Krauthammer (Eq.
(61)) are used to calculate the underpressure (Ps−) and the
negative phase duration (ts−), respectively, and the shapes are
triangular as UFC suggests to use. The negative pressure-time
curve of UFC has a time of rise equal to 0.25 times of negative
phase duration (ts−). It should be noted that UFC shows larger
values for underpressure as compared to the results produced
from the given empirical relations.
Fig. 16. Comparison of Positive Pressure Profiles by Empirical
Equations and UFC 3-340-02 6. Conclusions

sections. Among many available equations, several empirical This review provides a comprehensive overview of the
equations that show good agreement with UFC 3-340-02 are empirical equations and blast wave parameters given by various
selected. Fig. 16 shows the plot of modified Friedlander equation researchers. Positive and negative incident blast wave parameters
(Eq. (5)). In this equation, the empirical equations proposed by such as temporal pressure profiles, positive peak overpressures,
Kinney and Graham (Eq. (12)), Wu and Hao (Eq. (27)), Sadovskyi shock arrival times, positive phase durations, positive phase
(Eq. (31)), and Teich and Gebekken (Eq. (55)) are used for impulses, and wave decay parameters are compared with those
overpressure (Ps), arrival time (ta), positive phase duration (td) by Unified Facility Criteria (Unified Facitilities Criteria 3-340-
and wave decay parameter (b), respectively. Empirical equations 02, 2008).
of overpressure (Ps), arrival time (ta), and positive phase duration Typically, it is observed that blast parameters at large distances
(td), which most reflect the UFC chart, are selected among the from an explosive charge can be predicted accurately using the
equations presented. However, wave decay parameter (b) is not presented equations. However, if the explosive is very close to
presented in the UFC chart. Therefore the most recent empirical the target, the accuracy might decrease since there are limited
equation is used in the example. test data available that might be used to validate the parameters
The positive pressure profiles by using given empirical for near regions (Bogosian and Heidenreich, 2012). Especially,
in the negative phase of the blast wave, the results show large
difference compared with the result by UFC. One possible
reason of this variation is that the experimental data contain
weather effects of the explosive performance in the real world
conditions, but those are not included in the equations up till now
(Swisdak Jr., 1994). Therefore, appropriate care and judgment
are required in applying any of these empirical relations to a real
situation. Various types of comparison of the presented equations
against test data and improvement of the empirical formulae are
important to enhance the reliability of the given equations.
Based on the comparative study of several dozens of the
presented empirical equations, the most appropriate equations
are selected to compute the incident blast wave parameters.
These suggested equations give values closer to that of UFC
curves, or they are the only available choices. For example, in
case of negative phase, we do not have many choices available to
select. The peak overpressure (Ps) and positive phase impulse (is)
can be calculated using the equations by Kinney and Graham
Fig. 17. Comparison of Negative Pressure Profiles by Empirical (Eq. (12)) and Kinney and Graham (Eq. (48)), respectively.
Equations and UFC 3-340-02 Additionally, the shock arrival time (ta), the positive phase

Vol. 21, No. 6 / September 2017 − 2223 −

 Aleem Ullah, Furqan Ahmad, Heung-Woon Jang, Sung-Wook Kim, and Jung-Wuk Hong

duration (td) and the wave decay parameter (b) can be obtained Cranz, C. (1926). Lehrbuch der Ballistik, II Band. Berlin.
using the equations by Wu and Hao (Eq. (27)), Sadovskyi (Eq. De Silva, C. W. (2010). Vibration and shock handbook, CRC Press,
(31)), and Teich and Gebekken (Eq. (55)), respectively. Similarly, USA.
Dharaneepathy (1993). Air-blast effects on shell structures, Phd thesis,
the negative phase parameters can be calculated using the equations
Anna University, Madras.
by Drake (Eq. (59)), Krauthammer (Eq. (61)), and Brode (Eq. Esparza, E. D. (1986). “Blast measurements and equivalency for spherical
(65)), respectively. charges at small scaled distances.” Int. J. Impact Eng., Vol. 4, No. 1,
pp. 23-40, DOI: 10.1016/0734-743X(86)90025-4.
Acknowledgements Gelfand, B. and Silnikov, M. (2004). Translation from Russian to
English the Book “Blast Effects Caused by Explosions”, DTIC
This research was supported by a grant from a Construction Document, London, England.
Technology Research Project (Development of impact/blast Goel, M. D., Matsagar, V. A., Gupta, A. K., and Marburg, S. (2012).
“An abridged review of blast wave parameters.” Def. Sci. J., Vol. 62,
resistant HPFRCC and evaluation technique thereof, 13SCIPS02)
No. 5, pp. 300-306, DOI: 10.14429/dsj.62.1149.
funded by the Ministry of Land, Infrastructure, and Transport, Held, M. (1983). “Blast waves in free air.” Propellants, Explos.,
and was also supported by U-city Master and Doctor Course Pyrotech., Vol. 8, No. 1, pp. 1-7, DOI: 10.1002/prep.19830080102.
Grant Program of Korea Ministry of Land, Infrastructure and Henrych, J. and Major, R. (1979). The dynamics of explosion and its
Transport (MOLIT). use, Elsevier, Amsterdam.
Hopkins-Brown, M. A. and Bailey, A. (1998). Chapter 2 (Explosion
Effects) Part 1., AASTP-4 Royal Military College of Science, Cranfield
References University.
Hopkinson, B. (1915). British ordnance board minutes, Rep, 13565.
Abdollahzadeh, G. and Nemati, M. (2013). “Risk assessment of Iqbal, J. and Ahmad, S. (2011). “Improving safety provisions of structural
structures subjected to blast.” Int. J. Damage Mech., Vol. 23, No. 1, design of containment against external explosion.” Proc. International
pp. 3-24, DOI: 10.1177/1056789513482479. conference on opportunities and challenges for water cooled reactors in
Adushkin, V. V. and Korotkov, А. I. (1961). “Parameters of a shock the 21st century, Intenational Atomic Energy Agency (IAEA).
wave near to HE charge at explosion in air.” PMTF, Vol. 5, pp. 119- Izadifard, R. A. and Foroutan, M. (2010). “Blastwave parameters assessment
123. at different altitude using numerical simulation.” Turk. J. Eng. and
Ahmad, S., Elahi, A., Pervaiz, H., Rahman, A. G. A., and Barbhuiya, S. Environ. Sci., Vol. 34, No. 1, pp. 25-42, DOI: 10.3906/muh-0911-39.
(2014). “Experimental study of masonry wall exposed to blast Jeremić, R. and Bajić, Z. (2006). “An approach to determining the TNT
loading.” Materiales de Construccion, Vol. 64, No. 313, pp. 1-11, equivalent of high explosives.” Sci. Tech. Rev., Vol. 56, No. 1, pp.
DOI: 10.3989/mc.2014.01513. 58-62.
Ahmad, S., Taseer, M., and Pervaiz, H. (2012). “Effects of impulsive Kadid, A., Nezzar, B., and Yahiaoui, D. (2012). “Nonlinear dynamic
loading on reinforced concrete structures.” Tech. J., Univ. Eng and analysis of reinforced concrete slabs subjected to blast loading.”
Technol. Taxila, Pakistan (Vibration analysis issue). Asian J. Civ. Eng. (Build. and Hous.), Vol. 13, No. 5, pp. 617-634.
Bajić, Z. (2007). “Determination of TNT equivalent for various explosives.” Kangarlou, K. (2013). “Mechanics of blast loading on the head models
Master’s, University of Belgrade, Belgrade, Serbia. in the study of traumatic brain injury.” Nationalpark-Forschung In
Bangash, M. Y. H. and Bangash, T. (2006). Explosion-resistant buildings, Der Schweiz (Switz. Res. Park J.), Vol. 102, No. 11, pp. 1571-1581.
Springer, London, UK. Kim, J. H. J., Yi, N. H., Kim, S. B., Choi, J. K., and Park, J. C (2009).
Beshara, F. B. A. (1994). “Modelling of blast loading on aboveground “Experiment study on blast loading response of FRP-retrofitted RC
structures—I. General phenomenology and external blast.” Comput. slab structures.” Proc., Asia-Pacific Conference on FRP in Structures,
and Struct., Vol. 51, No. 5, pp. 585-596, DOI: 10.1016/0045-7949 pp. 533-538.
(94)90066-3. Kingery, C. N. and Bulmash, G. (1984). Air blast parameters from TNT
Bogosian, D. D. and Heidenreich, A. N. (2012). “An evaluation of spherical air burst and hemispherical surface burst, Ballistic Research
engineering methods for predicting close-in air blast.” Proc., Structures Laboratories.
Congress 2012, pp. 90-101, DOI: 10.1061/9780784412367.009. Kinney, G. F. and Graham, K. J. (1985). Explosive shocks in air,
Brode, H. L. (1955). “Numerical solutions of spherical blast waves.” J. Springer-Verlag, Berlin and New York.
Appl. phys., Vol. 26, No. 6, pp. 766-775, DOI: 10.1063/1.1722085. Krauthammer, T. (2008). Modern protective structures, CRC Press,
Cabello, B. (2011). Dynamic stress analysis of the effect of an air blast USA.
wave on a stainless steel plate, Master’s thesis, Rensselaer Polytechnic Lam, N., Mendis, P., and Ngo, T. (2004). “Response spectrum solutions
Institute Hartford, Connecticut. for blast loading.” Electron. J. Struct. Eng., Vol. 4, pp. 28-44.
Chang, D. B. and Young, C. S. (2010). “Probabilistic Estimates of Larcher, M. (2008). Pressure-time functions for the description of air
vulnerability to explosive overpressures and impulses.” J. Phys. blast waves, Technical note, JRC.
Secur., Vol. 4, No. 2, pp. 1-29. Li, J. and Ma, S. (1992). Explosion mechanics, Science Press, Beijing.
Chock, J. M. K. (1999). Review of Methods for Calculating Pressure Low, H. Y. and Hao, H. (2001). “Reliability analysis of reinforced concrete
Profiles of Explosive Air Blast and its Sample Application. Mater’s slabs under explosive loading.” Struct. Saf., Vol. 23, No. 2, pp. 157-
thesis, Virginia Polytechnic Institute and State University, Blacksburg, 178, DOI: 10.1016/S0167-4730(01)00011-X.
Virginia. Mays, G. C. and Smith, P. D. (1995). Blast effects on buildings: design of
Conrath, E. J. (1999). Structural design for physical security: State of buildings to optimize resistance to blast loading, Thomas Telford,
the practice, ASCE Publications. London.

− 2224 − KSCE Journal of Civil Engineering

 Review of Analytical and Empirical Estimations for Incident Blast Pressure

Mills, C. A. (1987). “The design of concrete structure to resist explosions explosive shock wave impact onto military vehicles of contemporary
and weapon effects.” Proceedings of the 1st Int. Conference on warfare.” J. KONES, Vol. 18, No. 1, pp. 515-524.
Concrete for Hazard Protections, pp. 61-73. Smith, P. D. and Hetherington, J. G. (1994). Blast and ballistic loading
Nassr, A. A. (2012). Experimental and analytical study of the dynamic of structures, Butterworth-Heinemann Oxford, UK.
response of steel beams and columns to blast loading, Open Access Swisdak Jr., M. M. (1994). Simplified Kingery airblast calculations,
Dissertations and Theses, McMaster University, Hamilton, Canada. Minutes of the Twenty Sixth DOD Explosives Safety Seminar,
Newmark, N. M. and Hansen, R. J. (1961). Design of blast resistant DTIC Document, Maryland.
structures, Shock and vibration handbook, Harris, and Crede, eds., Uddin, N. (2010). Blast protection of civil infrastructures and vehicles
McGraw-Hill, New York, USA. using composites, Elsevier, New York, USA.
NFPA (2008). Guide to fire and explosion investigations NFPA 921, UFC (2008). Unified Facilities Criteria 3-340-02: Structures to resist
National Fire Protection Association, Quincy, Massachussetts. the effects of accidental explosions, Dept. of the Army, the NAVY
Pape, R., Mniszewski, K. R., and Longinow, A. (2009). “Explosion and the Air Force, Washington DC, USA.
phenomena and effects of explosions on structures. I: Phenomena UNODA (2011). Formulae for ammunition management IATG 01.80,
and effects.: Pract. Period. Struct. Des. and Constr., Vol. 15, No. 2, United Nations Office for Disarmament Affairs (UNODA), New
pp. 135-140, DOI: 10.1061/(ASCE)SC.1943-5576.0000038 . York, USA.
Pierre-Emmanuel S. (2012). Etude des phénomènes physiques associés Vijayaraghavan, C., Thirumalaivasan, D., and Venkatesan, R. (2012).
à la propagation d'ondes consécutives à une explosion et leur “A study on nuclear blast overpressure on buildings and other
interaction avec des structures, dans un environnement complexe. infrastructures using geospatial technology.” J. Comput. Sci., Vol. 8,
Autre. Université d’Orléans, Français. No. 9, pp. 1520-1530, DOI: 10.3844/jcssp.2012.1520.1530.
Sachs, R. G. (1944). Dependence of blast on ambient pressure and Wu, C. and Hao, H. (2005). “Modeling of simultaneous ground shock
temperature, BRL-466 Ballistic Research Laboratory, Aberdeen, and airblast pressure on nearby structures from surface explosions.”
Maryland. Int. J. Impact Eng., Vol. 31, No. 6, pp. 699-717, DOI: 10.1016/
Sadovskiy, M. A. (2004). Mechanical effects of air shock waves from j.ijimpeng.2004.03.002.
explosions according to experiments, Selected works: Geophysics Yin, X., Gu, X., Lin, F., and Kuang, X. (2009). Numerical analysis of
and physics of explosion, Nauka Press, Moscow. blast loads inside buildings, Computational Structural Engineering,
Saska, P., Krzystała, E., and Mężyk, A. (2011). “An analysis of an Springer, Netherlands, pp. 681-690.

Vol. 21, No. 6 / September 2017 − 2225 −