The Effects Nuclear Weapons

The Effects Nuclear Weapons
The
Effects
of
Nuclear
Weapons
Compiled and edited by
Samuel Glasstone and Philip J. Dolan
Third Edition
Prepared and published by the
UNITED STATES DEPARTMENT OF DEFENSE
and the
ENERGY RESEARCH AND DEVELOPMENT ADMINISTRATION
~
.~~
~
-'!l
-:8
~"
,,-"
.,,~
..0
1977
,-..,."
.
J!'or sale by the Superintendent
of Documents, U.S. Government Printing
OlBce
Washln~ton. D.C. 20402
'"
ib',c
;
I
I
PREFACE
When "The Effects of Atomic Weapons" was published in 1950, the explosive
energy yields of the fission bombs available at that time were equivalent to some
thousands of tons (i.e., kilotons) of TNT. With the development of thermonuclear
(fusion) weapons, having energy yields in the range of millions of tons (i.e.,
megatons) of TNT, a new presentation, entitled "The Effects of Nuclear Weapons," was issued in 1957. A completely revised edition was published in 1962 and
this was reprinted with a few changes early in 1964.
Since the last version of "The Effects of Nuclear Weapons" was prepared, much
new information has become available concerning nuclear weapons effects. This
has come in part from the series of atmospheric tests, including several at very high
altitudes, conducted in the Pacific Ocean area in 1962. In addition, laboratory
studies, theoretical calculations, and computer simulations have provided a better
understanding of the various effects. Within the limits imposed by security requirements, the new information has been incorporated in the present edition. In
particular, attention may be called to a new chapter on the electromagnetic pulse.
We should emphasize, as has been done in the earlier editions, that numerical
values given in this book are not-and cannot be-exact. They must inevitably
include a substantial margin of error. Apart from the difficulties in making
measurements of weapons effects, the results are often dependent upon circumstances which could not be predicted in the event of a nuclear attack. Furthermore,
two weapons of different design may have the same explosive energy yield, but the
effects could be markedly different. Where such possibilities exist, attention is
called in the text to the limitations of the data presented; these limitations should not
be overlooked.
The material is arranged in a manner that should permit the general reader to
obtain a good understanding of the various topics without having to cope with the
more technical details. Most chapters are thus in two parts: the first part is written at
a fairly low technical level whereas the second treats some of the more technical and
mathematical aspects. The presentation allows the reader to omit any or all of the
latter sections without loss of continuity.
The choice of units for expressing numerical data presented us with a dilemma.
The exclusive use of international (SI) or metric units would have placed a burden
on many readers not familiar with these units, whereas the inclusion of both SI and
common units would have complicated many figures, especially those with
logarithmic scales. As a compromise, we have retained the older units and added an
explanation of the SI system and a table of appropriate conversion factors.
i
,
i
Preface
Many organizations and individuals contributed in one way or another to this
revision of "The Effects of Nuclear Weapons," and their cooperation is gratefully
acknowledged. In particular, we wish to express our appreciation of the help given
us by L. J. Deal and W. W. Schroebel of the Energy Research and Development
Administration and by Cmdr. H. L. Hoppe of the Department of Defense.
Samuel Glasstone
Philip J. Dolan
~
:
I
I
I
I II
,II
ACKNOWLEDGEMENTS
Preparation of this revision of "The Effects of Nuclear Weapons" was made
possible by the assistance and cooperation of members of the organizations listed
below.
Department of Defense
Headquarters, Defense Nuclear Agency
Defense Civil Preparedness Agency
Armed Forces Radiobiology Research Institute
V. S. Army Aberdeen Research and Development Center, Ballistic Research Laboratories
V.S. Army Engineer Waterways Experiment Station
Naval Surface Weapons Center
Department of Defense Contractors
Stanford Research Institute
General Electric, TEMPO
Mission Research Corporation
Department of Commerce
National Oceanic and Atmospheric Administration
Atomic Energy Commission!
Energy Research and Development Administration
Headquarters Divisions and the laboratories:
Brookhaven National Laboratory
Health and Safety Laboratory
Lawrence Livermore Laboratory
Los Alamos Scientific Laboratory
Lovelace Biomedical and Environmental Research Laboratories
Oak Ridge National Laboratory
Sandia Laboratories
i
!
CONTENTS
CHAPTER I-General Principles of Nuclear Explosions.
Characteristics of Nuclear Explosions.
Scientific Basis of Nuclear Explosions.
Page
I
I
12
CHAPTER II-Descriptions
of Nuclear Explosions.
Introduction.
Description of Air and Surface Bursts.
Description of High-Altitude Bursts.
Description of Underwater Bursts.
Description of Underground Bursts.
Scientific Aspects of Nuclear Explosion Phenomena.
26
26
27
45
48
58
63
CHAPTER III-Air
Blast Phenomena in Air and Surface Bursts.
Characteristics of the Blast Wave in Air.
Reflection of Blast Wave at a Surface.
Modification of Air Blast Phenomena.
Technical Aspects of Blast Wave Phenomena.
80
80
86
92
96
CHAPTER IV-Air
Blast Loading
Interaction of Blast Wave with Structures.
Interaction of Objects with Air Blast.
127
127
132
CHAPTER V -Structural Damage from Air Blast.
Introduction
FactorsAffectingResponse
,
Commercial and Administrative Structures.
IndustriaIStructures
ResidentiaIStructures
Transportation
Utilities.
MiscellaneousTargets
Analysis of Damage from Air Blast
154
154
156
158
165
175
189
195
206
212
CHAPTER VI-Shock Effects of Surface and Subsurface Bursts.
Characteristics of Surface and Shallow Underground Bursts.
Deep Underground Bursts
Damage to Structures.
Characteristics of Underwater Bursts.
231
231
238
241
244
I
Technical Aspects of Surface and Underground Bursts.
Technical Aspects of Deep Underground Bursts.
IJoading on Buried Structures.
253
260
263
DamagefromGroundShock
Technical Aspects of Underwater Bursts.
265
268
CHAPTER VII-Thermal
Radiation and Its Effects.
RadiationfromtheFireball
Thermal Radiation Effects
Incendiary Effects.
IncendiaryEffectsinJapan
Technical Aspects of Thermal Radiation.
Radiant Exposure-Distance Relationships.
276
276
282
296
300
305
316
CHAPTER VIII-Initial
Nuclear Radiation.
Nature of Nuclear Radiations
Gamma Rays.
Neutrons.
Transient-Radiation Effects on Electronics (TREE)
Technical Aspects of Initial Nuclear Radiation.
324
324
326
340
349
353
CHAPTER IX-Residual Nuclear Radiation and Fallout.
Sources of Residual Radiation.
Radioactive Contamination from Nuclear Explosions.
Fallout Distribution in Land Surface Bursts.
Fallout Predictions for Land Surface Bursts.
Attenuation of Residual Nuclear Radiation.
Delayed Fallout. .'.'
Technical Aspects of Residual Nuclear Radiation.
387
387
409
414
422
439
442
450
CHAPTER X-Radio and Radar Effects.
Introduction
Atmospheric Ionization Phenomena.
Ionization Produced by Nuclear Explosions
Effects on Radio and Radar Signals.
Technical Aspects of Radio and Radar Effects.
461
461
462
466
479
489
CHAPTER XI-The
514
Electromagnetic Pulse and its Effects.
OriginandNatureoftheEMP
EMP Damage and Protection
Theory of the EMP
CHAPTER XII-Biological
Introduction
Blast Injuries.
BurnInjuries
Nuclear Radiation Injury.
,'-
Effects.
514
523
532
541
541
548
560
575
Characteristics of Acute Whole-Body Radiation Injury.
583
CombinedInjuries.
588
Late Effects of Ionizing Radiation.
Effects of Early Fallout
Long-Term Hazard from Delayed Fallout.
Genetic Effects of Nuclear Radiation.
Pathology of Acute Radiation Injury.
Blast-RelatedEffects
Effects on Farm Animals and Plants.
589
594
604
609
614
618
618
Glossary.
629
GuidetoSIUnits
642
Index.
644
~
i
i
!
!
rr
~
CHAPTERIV
AIR BLAST LOADING
INTERACTION OF BLAST WAVE WITH STRUCTURES
INTRODUCTION
4.01 The phenomena associated
with the blast wave in air from a nuclear
explosion have been treated in the preceding chapter. The behavior of an object or structure exposed to such a wave
may be considered under two main
headings. The first, called the "loading," i.e., the forces which result from
the action of the blast pressure, is the
subject of this chapter. The second, the
"response"or distortion of the structure
due to the particular loading, is treated
in the next chapter.
4.02 For an air burst, the direction
of propagation of the incident blast
wave will be toward the ground at
ground zero. In the regular reflection
This tends to cause crushing toward the
ground, e.g., dished-in roofs, in addition to distortion due to translational
motion.
4.03 The discussion of air blast
loading for aboveground structures in
the Mach region in the sections that
follow emphasizes the situation where
the reflecting surface is nearly ideal
(§ 3.47) and the blast wave behaves
normally, in accordance with theoretical
considerations. A brief description of
blast wave loading in the precursor region (§ 3.79 et seq.) is also given. For
convenience, the treatment will be
somewhat arbitrarily divided into two
parts: one deals with "diffraction loading," which is determined mainly by
region, where the direction of propagation of the blast wave is not parallel to
the horizontal axis of the structure, the
forces exerted upon structures will also
have a considerable downward component (prior to passage of the reflected
wave) due to the reflected pressure
buildup on the horizontal surfaces.
Consequently, in addition to the horizontal loading, as in the Mach region
(§ 3.24 et seq.), there will also be initially an appreciable downward force.
the peak overpressure in the blast wave,
and the other with "drag loading," in
which the dynamic pressure is the significant property. It is important to remember, however, that all structures are
subjected simultaneously to both types
of loading, since the overpressure and
dynamic pressure cannot be separated,
although for certain structures one may
be more important than the other.
4.04 Details of the interaction of a
blast wave with any structure are quite
127
1
l
!
128
complicated, particularly if the geometry of the structure is complex. However, it is frequently possible to consider
equivalent simplified geometries, and
blast loadings of several such geometries are discussed later in this chapter.
AIR BLAST LOADING
4.05 When the front of an air blast
wave strikes the face of a structure,
reflection occurs. As a result the overpressure builds up rapidly to at least
twice (and generally several times) that
in the incident wave front. The actual
pressure attained is determined by
various factors, such as the peak overpressure of the incident blast wave and
the angle between the direction of motion of the wave and the face of the
structure (§ 3.78). The pressure increase is due to the conversion of the
kinetic energy of the air behind the
shock front into internal energy as the
rapidly moving air behind t~e shock
front is decelerated at the face of the
structure. The reflected shock front
propagates back into the air in all directions. The high pressure region expands
outward towards the surrounding regions of lower pressure.
4.06 As the wave front moves forward, the reflected overpressure on the
face of the structure drops rapidly to that
produced by the blast wave without reflection, I plus an added drag force due
same pressure is exerted on the sides
and the roof. The front face, however, is
still subjected to wind pressure, although the back face is shielded from it.
4.07 The developments described
above are illustrated in a simplified form
in Figs. 4.07a, b, c, d, e;2 this shows, in
plan, successive stages of a structure
without openings which is being struck
by an air blast wave moving in a horizontal direction. In Fig. 4.07a the wave
front is seen approaching the structure
with the direction of motion perpendicular to the face of the structure exposed
to the blast. In Fig. 4.07b the wave has
just reached the front face, producing a
high reflected overpressure. In Fig.
4.07c the blast wave has proceeded
about halfway along the structure and in
Fig. 4.07d the wave front has just
passed the rear of the structure. The
pressure on the front face has dropped to
some extent while the pressure is building up on the back face as the blast wave
diffracts around the structure. Finally,
when the wave front has passed completely, as in Fig. 4.07e, approximately
equal air pressures are exerted on the
sides and top of the structure. A pressure difference between front and back
faces, due to the wind forces, will persist, however, during the whole positive
phase of the blast wave. If the structure
is oriented at an angle to the blast wave,
the pressure would immediately be exerted on two faces, instead of one, but
to the wind (dynamic) pressure. At the
same time, the air pressure wave bends
or "diffracts" around the structure, so
that the structure is eventually engulfed
by the blast, and approximately the
the general characteristics of the blast
loading would be similar to that just
described (Figs. 4.07f, g, h, and i).
4.08 The pressure differential between the front and back faces will have
DIFFRAcrION LOADING
'This is often referred to as the "side-on overpressure," since it is the same as that experienced by the
side of the structure, where there is no appreciable reflection.
'A more detailed treatment is given later in this chapter.
~;;~~
-~;
INTERACTION
OF BLAST W AVE WITH STRUCTURES
129
D
a
b
f
Figure 4.07.
c
g
d
e
h
Stages in the diffraction of a blast wave by a structure without openings (plan
view).
its maximum value when the blast wave
has not yet completely surrounded the
structure, as in Figs. 4.07b, c, and d or
g and h. Such a pressure differential will
produce a lateral (or translational) force
tending to cause the structure to deflect
and thus move bodily, usually in the
same direction as the blast wave. This
force is known as the "diffraction loading" because it operates while the blast
wave is being diffracted around the
structure. The extent and nature of the
response will depend upon the size,
shape, and weight of the structure and
how firmly it is attached to the ground.
Other characteristics of the structure are
also important in determining the response, as will be seen later.
4.09 When the blast wave has en-
gulfed the structure (Fig. 4.07e or
4.07i), the pressure differential is small,
and the loading is due almost entirely to
the drag pressure3 exerted on the front
face. The actual pressures on all faces of
the structure are in excess of the ambient
atmospheric pressure and will remain
so, although decreasing steadily, until
the positive phase of the blast wave has
ended. Hence, the diffraction loading on
a structure without openings is eventually replaced by an inwardly directed
pressure, i.e., a compression or squeezing action, combined with the dynamic
pressure of the blast wave. In a structure
with no openings, the loading will cease
only when the overpressure drops to
zero.
' 4.10 The damage caused during the
'The drag pressure is the product of the dynamic pressure and the drag coefficient (§ 4.29).
130
AIR BLAST LOADING
diffraction stage will be determined by
the magnitude .of the loading and by its
duration. The loading is related to the
peak overpressure in the blast wave and
this is consequently an important factor.
If the structure under consideration has
no openings, as has been assumed so
far, the duration of the diffraction loading will be very roughly the time required for the wave front to move from
the front to the back of the building,
although wind loading will continue for
a longer period. The size of the structure
will thus affect the diffraction loading.
For a structure 75 feet long, the diffraction loading will operate for a period of
about one-tenth of a second, but the
squeezing and the wind loading will
persist for a longer time (§ 4.13). For
thin structures, e.g., telegraph or utility
poles and smokestacks, the diffraction
period is so short that the corresponding
loading is negligible.
4.11 If the building exposed to the
blast wave has openings, or if it has
windows, panels, light siding, or doors
which fail in a very short space of time,
there will be a rapid equalization of
pressure between the inside and outside
of the structure. This will tend to reduce
the pressure differential while diffraction is occurring. The diffraction loading on the structure as a whole will thus
be decreased, although the loading on
interior walls and partitions will be
greater than for an essentially closed
structure, i.e., one with few openings.
Furthermore, if the building has many
openings, the squeezing (crushing) action, due to the pressure being higher
outside than inside after the diffraction
stage, will not occur.
:"';;~
DRAG (DYNAMIC PRESSURE)
LOADING
4.12 During the whole of the overpressure positive phase (and for a short
time thereafter) a structure will be subjected to the dynamic pressure (or drag)
loading caused by the transient winds
behind the blast wave front. Under
nonideal (precursor) conditions, a dynamic pressure loading of varying
strength may exist prior to the maximum
overpressure (diffraction) loading. Like
the diffraction loading, the drag loading,
especially in the Mach region, is equivalent to a lateral (or translational) force
acting upon the structure or object exposed to the blast.
4.13 Except at high blast overpressures, the dynamic pressures at the face
of a structure are much less than the
peak overpressures due to the blast wave
and its reflection (Table 3.07). However, the drag loading on a structure
persists for a longer period of time,
compared to the diffraction loading. For
example, the duration of the positive
phase of the dynamic pressure on the
ground at a slant range of I mile from a
I-megaton nuclear explosion in the air is
almost 3 seconds. On the other hand,
the diffraction loading is effective only
for a small fraction of a second, even for
a large structure, as seen above.
4.14 It is the effect of the duration
of the drag loading on structures which
constitutes an important difference between nuclear and high-explosive detonations. For the same peak overpressure
in the blast wave, a nuclear weapon will
prove to be more destructive than a
conventional one, especially for buildings which respond to drag loading.
INTERACTION
OF BLAST WAVE WITH STRUCTURES
This is because the blast wave is of
much shorter duration for a high-explosive weapon, e.g., a few hundredths of
a second. As a consequence of the
longer duration of the positive phase of
the blast wave from weapons of high
energy yield, such devices cause more
damage to drag-sensitive structures
(§ 4.18) than might be expected from
the peak overpressures alone.
131
4.15 In analyzing the response to
blast loading, as will be done more fully
in Chapter V, it is convenient to consider structures in two categories, i.e.,
diffraction-type structures and drag-type
structures. As these names imply, in a
nuclear explosion the former would be
affected mainly by diffraction loading
front and rear exists during the whole of
this period. Examples of structures
which respond mainly to diffraction
loading are multistory, reinforced-concrete buildings with small window area,
large wall-bearing structures such a~
apartment houses, and wood-frame
buildings such as dwelling houses.
4.17
Because, even with large
structures, the diffraction loading will
generally be operative for a fraction of a
second only, the duration of the blast
wave positive phase, which is usually
much longer, will not be significant. In
other words, the length of the blast wave
positive phase will not materially affect
the net translational loading (or the resuIting damage) during the diffraction
stage. A diffraction-type structure is,
therefore, primarily sensitive to the peak
overpressure in the blast wave to which
it is exposed. Actually it is the asso-
and the latter by drag loading. It should
be emphasized, however, that the distinction is made in order to simplify the
treatment of real situations which are, in
fact, very complex. Although it is true
that some structures will respond mainly
to diffraction forces and others mainly to
drag forces, actually all buildings will
respond to both types of loading. The
relative importance of each type of
loading in causing damage will depend
upon the type of structure as well as on
the characteristics of the blast wave.
These facts should be borne in mind in
connection with the ensuing discussion.
4.16
Large buildings having a
moderately small window and door area
and fairly strong exterior walls respond
mainly to diffraction loading. This is
because it takes an appreciable time for
the blast wave to engulf the building,
and the pressure differential between
ciated reflected overpressure on the
structure that largely determines the
diffraction loading, and this may be
several times the incident blast overpressure (§ 3.78).
4.18 When the pressures on different areas of a structure (or structural
element) are quickly equalized, either
because of its small size, the characteristics of the structure (or element), or
the rapid formation of numerous openings by action of the blast, the diffraction forces operate for a very short time.
The response of the structure is then
mainly due to the dynamic pressure (or
drag force) of the blast wind. Typical
drag-type structures are smokestacks,
telephone poles, radio and television
transmitter towers, electric transmission
towers, and truss bridges. In all these
cases the diffraction of the blast wave
around the structure or its component
STRUcrURAL CHARACTERISTICS
AND AIR BLAST LOADING
)
132
AIRBLASTLOADING
elements requires such a very short time
that the diffraction processes are negligible, but the drag loading may be considerable.
4.19 The drag loading on a structure is determined not only by the dynamic pressure, but also by the shape of
the structure (or structural element). The
shape factor (or drag coefficient) is less
for rounded or streamlined objects than
for irregular or sharp-edged structures or
elements. For example, for a unit of
projected area, the loading on a telephone pole or a smokestack will be less
than on an I-beam. Furthermore, the
drag coefficient can be either positive or
negative, according to circumstances
(§ 4.29).
4.20 Steel (or reinforced-concrete)
frame buildings with light walls made of
asbestos cement, aluminum, or corrugated steel, quickly become drag-sensitive because of the failure of the walls at
low overpressures. This fail.ure, accompanied by pressure equalization, occurs
INTERACTION
r
very soon after the blast wave strikes the
structure, so that the frame is subject to
a relatively small diffraction loading.
The distortion, or other damage, subsequently experienced by the frame, as
well as by narrow elements of the
structure, e.g., columns, beams, and
trusses, is then caused by the drag
forces.
4.21 For structures which are fundamentally of the drag type, or which
rapidly become so because of loss of
siding, the response of the structure or
of its components is determined by both
the drag loading and its duration. Thus,
the damage is dependent on the duration
of the positive phase of the blast wave as
well as on the peak dynamic pressure.
Consequently, for a given peak dynamic
pressure, an explosion of high energy
yield will cause more damage to a
drag-type structure than will one of
lower yield because of the longer duration of the positive phase in the former
case (see § 5.48 et seq.).
OF OBJECTS WITH AIR BLAST4
DEVELOPMENT OF BLAST LOADING
4.22 The usual procedure for predicting blast damage is by an analysis,
supported by such laboratory and fullscale observations as may be available.
The analysis is done in two stages: first
the air blast loading on the particular
structure is determined; and second, an
evaluation is made of the response of the
structure to this loading. The first stage
of the analysis for a number of idealized
targets of simple shape is discussed in
the following sections. The second stage
is treated in Chapter V.
4.23 The blast loading on an object
is a function of both the incident blast
wave characteristics, i.e., the peak
overpressure, dynamic pressure, decay,
and duration, as described in Chapter
III, and the size, shape, orientation, and
response of the object. The interaction
of the incident blast wave with an object
is a complicated process, for which a
theory, supported primarily by experimental data from shock tubes and wind
'The remaining (more technical) sections of this chapter may be omitted without loss of continuity.
~~j
,~,--,
INTERACTION
OF OBJECTS WITH AIR BLAST
133
tunnels, has been developed. To reduce
the complex problem of blast loading to
reasonable terms, it will be assumed, for
the present purpose, that (I) the overpressures of interest are less than 50
pounds per square inch (dynamic pressures less than about 40 pounds per
square inch), and (2) the object being
loaded is in the region of Mach reflection.
4.24 To obtain a general idea of the
blast loading process, a simple object,
namely, a cube with one side facing
toward the explosion, will be selected as
an example. It will be postulated, further, that the cube is rigidly attached to
the ground surface and remains motionless when subjected to the loading. The
blast wave (or shock) front is taken to be
of such size compared to the cube that it
can be considered to be a plane wave
striking the cube. The pressures referred
to below are the average pressures on a
particular face. Since the object is in the
region of Mach reflection, the blast front
is perpendicular to the surface of the
ground. The front of the cube, i.e., the
side facing toward the explosion, is
normal to the direction of propagation of
the blast wave (Fig. 4.24).
4.25 When the blast wave strikes
the front of the cube, reflection occurs
producing reflected pressures which
I
may be from two to eight times as great
as the incident overpressure (§ 3.56).
The blast wave then bends (or diffracts)
around the cube exerting pressures on
the sides and top of the object, and
finally on its back face. The object is
thus engulfed in the high pressureof the
blast wave and this decays with time,
eventually returning to ambient conditions. Because the reflected pressure on
the front face is greater than the pressure
in the blast wave above and to the sides,
the reflected pressure cannot be maintained and it soon decays to a "stagnation pressure," which is the sum of the
incident overpressure and the dynamic
(drag) pressure. The decay time is
roughly that required for a rarefaction
wave to sweep from the edges of the
front face to the center of this face and
back to the edges.
4.26 The pressures on the sides and
top of the cube build up to the incident
overpressure when the blast front arrives
at the points in question. This is followed by a short period of low pressure
caused by a vortex formed at the front
edge during the diffraction process and
which travels along or near the surface
behind the wave front (Fig. 4.26). After
the vortex has passed, the pressure returns essentially to that in the incident
blast wave which decays with time. The
air flow causes some reduction in the
loading to the sides and top, because, as
BLAST WAVE FRONT will be seen in § 4.43, the drag pressure
here has a negatiye value.
4.27 When the blast wave reaches
the rear of the cube, it diffracts around
the edges, and travels down the back
surface (Fig. 4.27). The pressure takes a
certain time ("rise time") to reach a
Figure 4.24. Blast wave approachingcube more-or-less steady state value equal to
rigidly attachedto ground.
the algebraic sum of the overpressure
134
AIR BLAST LOADING
BLAST WAVE FRONT and the drag pressure. The latter is related to the dynamic pressure, q(t), by
the expression
VORTEX0)
Drag pressure = Cdq(t),
where Cd is the drag coefficient. The
value of Cddepends on the orientation of
the particular face to the blast wave
front and may be positive or negative.
The drag pressures (or loading) may
thus be correspondingly positive or
negative. The quantities p(t) and q(t)
BLASTWAVEFRONT represent the overpressure and dynamic
pressure, respectively, at any time, t,
Figure 4.26. Blast wavemoving over
sides and top of cube.
VORTEX
~
after the arrival of the wave front
(§ 3.57 et seq.).
4.30 The foregoing discussionhas
referred to the loading on the various
surfaces in a general manner. For a
particular
Figure
rear
4.27.
Blast
wave
moving
down
d
of cube.
pomt
and the drag pressure, the latter having a
negative value in this casealso (§ 4.44).
The finite rise time results from a weakening of the blast wave front as it diffracts around the back edges, accompanied by a temporary vortex action,
and the time of transit of the blast wave
from the edges to the center of the back
face.
4.28 When the overpressure at the
rear of the cube attains the value of the
overpressure in the blast wave, the diffraction process may be considered to
have terminated. Subsequently, essentially steady state conditions may be
assumed to exist until the pressureshave
returned to the ambient value prevailing
prior to the arrival of the blast wave.
4.29 The total loading on any given
face of the cube is equal to the algebraic
sum of the respective overpressure, p(t),
'_c_":'.:
epen
.
d
point
h
to
on a surface,
I
s a so
tee
on
the
h d.
t e
d
IS t ance
d
ges
an
loading
f rom
d
a
more
th e
.
etal
1
e
d
treatment is necessary. It should be
noted that only the gross characteristics
of the development of the loading have
been described here. There are, in actual
fact, several cycles of reflected and
rarefaction waves traveling across the
surfaces before damping out, but these
fluctuations are considered to be of
minor significance as far as damage to
the structure is concerned.
EFFECT OF SIZE ON LOADING
DEVELOPMENT
4.31 The loading on each surface
may not be as important as the net
horizontal loading on the entire object.
Hence, it is necessary to study the net
loading, i.e., the loading on the front
face minus that on the back face of the
cube. The net horizontal loading during
the diffraction process is high because
INTERACTION
OF OBJECTS WITH AIR BLAST
the pressure on the front face is initially
the reflected pressure and no loading has
reached the rear face.
4.32 When the diffraction process is
completed, the overpressure loadings on
the front and back faces are essentially
equal. The net horizontal loading is then
relatively small. At this time the net
loading consists primarily of the difference between front and back loadings
resulting from the dynamic pressure
loading. Because the time required for
the completion of the diffraction process
depends on the size of the object, rather
than on the positive phase duration of
the incident blast wave, the diffraction
loading impulse per unit area (§ 3.59) is
greater for long objects than for short
ones.
4.33 The magnitude of the dynamic
pressure (or drag) loading, on the other
hand, is affected by the shape of the
object and the duration of the dynamic
pressure. It is the latter, and not the size
of the object, which determines the application time (and impulse per unit
area) of the drag loading.
4.34 It may be concluded, therefore, that, for large objects struck by
blast waves of short duration, the net
horizontal loading during the diffraction
process is more important than the dynamic pressure loading. As the object
becomes smaller, or as the dynamic
pressure duration becomes longer, e.g.,
with weapons of larger yield, the drag
loading becomes increasingly importanto For classification purposes, objects
a.re often described as "diffraction targets" or "drag targets," as mentioned
earlier, to indicate the loading mainly
responsible for damage. Actually, all
objects are damaged by the total loading, which is a combination of over-
135
pressure and dynamic pressure loadings,
rather than by anyone component of the
blast loading.
EFFECf OF SHAPE ON LOADING
DEVELOPMENT
4.35 The description given above
for the interaction of a blast wave with a
cube may be generalized to apply to the
loading on a structure of any other
shape. The reflection coefficient, i.e.,
the ratio of the (instantaneous) reflected
overpressure to the incident overpressure at the blast front, depends on the
angle at which the blast wave strikes the
structure. For a curved structure, e.g., a
sphere or a cylinder (or part of a sphere
or cylinder), the reflection varies from
point to point on the front surface. The
time of decay from reflected to stagnation pressure then depends on the size of
the structure and the location of the
point in question on the front surface.
4.36 The drag coefficient, i.e., the
ratio of the drag pressure to the dynamic
pressure (§ 4.29), varies with the shape
of the structure. In many cases an
overall (or average) drag coefficient is
given, so that the net force on the surface can be determined. In other instances, local coefficients are necessary
to evaluate the 'pressures at various
points on the surfaces. The time of
buildup (or rise time) of the average
pressure on the back surface depends on
the size and also, to some extent, on the
shape of the structure.
4.37 Some structures have frangible
portions that are easily blown out by the
initial impact of the blast wave, thus
altering the shape of the object and the
subsequentloading. When windows are
blown out of an ordinary building, the
136
AIR BLAST LOADING
blast wave enters and tends to equalize
the interior and exterior pressures. In
fact, a structure may be designed to
have certain parts frangible to lessen
damage to all other portions of the
structure. Thus, the response of certain
elements in such cases influences the
blast loading on the structure as a
whole. In general, the movement of a
structural element is not considered to
influence the blast loading on that element itself. However, an exception to
this rule arises in the case of an aircraft
in flight when struck by a blast wave.
BLAST LOADING-TIME CURVES
4.38 The procedures
whereby
curves showing the air blast loading as a
function of time may be derived are
given below. The methods presented are
for the following five relatively simple
shapes: (I) closed box-like structure; (2)
partially open box-like structure; (3)
open frame structure; (4) cylindrical
structure; and (5) semicircular arched
structure. These methods can be altered
somewhat for objects having similar
characteristics. For very irregularly
shaped structures, however, the proce-
dures described may provide no more
than a rough estimate of the blast loading to be expected.
4.39 As a general rule, the loading
analysis of a diffraction-type structure is
extended only until the positive phase
overpressure falls to zero at the surface
under consideration. Although the dynamic pressure persists after this time,
the value is so small that the drag force
can be neglected. However, for dragtype structures, the analysis is continued
until the dynamic pressure is zero. During the negative overpressure phase,
both overpressure and dynamic pressure
are too small to have any significant
effect on structures (§ 3.11 et seq.).
4.40 The blast wave characteristics
which need to be known for the loading
analysis and their symbols are summarized in Table 4.40. The locations in
Chapter III where the data may be obtained, at a specified distance from
ground zero for an explosion of given
energy yield and height of burst, are
also indicated.
4.41 A closed box-like structure
may be represented simply by a parallelepiped, as in Fig. 4.41, having a length
L, height H, and breadth B. Structures
Table 4.40
BLAST WAVE CHARACTERISTICS FOR DETERMINATION OF LOADING
Property
Peak overpressure
Time variation of overpressure
Peak dynamic pressure
Time variation of dynamic pressure
Reflected overpressure
Duration of positive phase of
Symbol
p
p(l)
q
q(l)
P,
Source
Figs. 3.73a, b, and c
Fig. 3.57
Fig. 3.75
Fig. 3.58
Fig. 3.78b
overpressure
Duration of positive phase of
I;
Fig. 3.76
dynamic pressure
Blast front (shock) velocity
,;
V
Fig. 3.76
Fig. 3.55
INTERACTION
OF OBJECTS WITH AIR BLAST
137
..-
'"'"
"-""-
H
~
""
'"'" , ,
Figure 4.41.
Representation of closed box-like structure.
with a flat roof and walls of approximately the same blast resistance as the
frame will fall into this category. The
Front Face.-The first step is to determine the reflected pressure, Pr; this
gives the pressure at the time t = 0,
walls have either no openings (doors
and windows), or a small number of
such openings up to about 5 percent of
the total area. The pressures on the interior of the structure then remain near
the ambient value existing before the
arrival of the blast wave, while the outside is subjected to blast loading. To
simplify the treatment, it will be supposed that one side of the structure faces
toward the explosion and is perpendicular to the direction of propagation of the
when the blast wave front strikes the
front face (Fig. 4.42). Next, the time, t,,
is calculated at which the stagnation
pressure, P" is first attained. It has been
found from laboratory studies that, for
peak overpressures being considered (50
pounds per square inch or less), t , can be
represented, to a good approximation,
by
blast wave. This side is called the front
face. The loading diagrams are computed below for (a) the front face, (b)
the side and top, and (c) the back face.
By combining the data for (a) and (c),
the net horizontal loading is obtained in
where S is equal to H or 8/2, whichever
is less, and U is the blast front (shock)
velocity. The drag coefficient for the
front face is unity, so that the drag
pressure is here equal to the dynamic
pressure. The stagnation pressure is thus
(d).
4.42
(a)
A verage
Loading
on
t = ~
.f
U
,
p, = p( t,) + q( t,),
138
AIR BLAST LOADING
where p(t,) and q(ts) are the overpressure and dynamic pressure at the time Is.
The average pressure subsequently
decays with time, so that,
ing at the distance U2 from the front of
the structure, so that
P
a
( )+
( )
= P ~2U
Cdq ~2U
Pressure at time t = p(t) + q(t),
where t is any time between ts and t;.
The pressure-time curve for the front
face can thus be determined, as in Fig.
4.42.
4.43 (b) A verage Loading on Sides
and Top.-Although
loading commences immediately after the blast wave
strikes the front face, i.e., at t = 0, the
sideswave
and top
not fully
until
the
has are
traveled
the loaded
distance
L,
The drag coefficient on the sides and top
of the structure is approximately -0.4
for the blast pressure range under consideration (§ 4.23). The loading increases from zero at t = 0 to the value Pa
at the time UU, as shown in Fig. 4.43.
Subsequently, the average pressure at
any time t is given by
.- t -P
Pressure at time
t -L 2U
i.e., at times t = UU. The average
pressure,Pa' at this time is considered to
be the overpressure plus the drag load-
+
C
dq
(
(
t--
Pr
w
0:
::>
(/)
(/)
W
0:
a. Ps ---
0
s
t;
TIME
Figure 4.42.
Average front face loading of closed box-like structure.
)
L
2U
)
'
INTERACTION
OF OBJECTS WITH AIR BLAST
139
Pa
UJ
Q:;
::>
(f)
P( I --f--
~
R::
2U
0
) + Cdq ( I
f:.2U
L
)
,+ + L
U
p
W
TIME
Average side and top loading of closed box-like structure.
for the postulated blast pressure range.
The average pressure at any time tafter
the attainment of Pb is represented by
L
)
-U
(
t
Cdq
-U
L
+
(
)
Pressure at time t =
P
U2U.
where t lies between U U and t+ +
p
U2U, as seen in Fig. 4.43. The overpressure and dynamic pressure, respectively, are the values at the time t U2 U. Hence, the overpressure on the
sides and top becomes zero at time t;; +
t
Figure 4.43.
4.44 (c) Average Loading on Back
Face.- The shock front arrives at the
back face at time UU, but it requires an
additional time, 4S/U, for the average
pressure to build up to the value Pb(Fig.
4.44), where Pb is given approximately
by
-This
) + C q ( ~
)
=
Pb
P
U
d
U
4.45
(d) Net Horizontal
Loadmg.-The net loading is equal to the
front loading minus the back loading.
subtraction is best performed
graphically, as shown in Fig. 4.45. The
left-hand diagram gives the individual
Here, as before, S is equal to H or B/2
whichever is the smaller. The drag coefficient on the back face is about -0.3
front and back loading curves, as
derived from Figs. 4.42 and 4.44, respectively. The difference indicated by
the shaded region is then transferred to
(~
where t lies between (L + 45)/ U and
t;; + UU, as seen in Fig. 4.44.
~"- ",-
'T'C~;~{
140
~~~
AIR BLAST LOADING
Pb
P(f--fj-)
+ Cdq(/--b)
w
a::
::>
cn
cn
w
a::
a-
D
1::..
L+4S
U
Figure 4.44.
U
fp +1::-
TIME
U
Average back face loading of closed box-like structure.
w
0:
w
::>
0:
::>
(/)
(/)
(/)
W
(/)
W
0:
Q.
0:
Q.
IW
I
I
I
I
Z
I
I
I
I
I
I
I
I
0
L
U
$-
I;
u
TIME
Figure 4.45.
1++1:..
p u
0
1:..
u
t$~
I;
u
1;+10-
u
TIME
Net horizontal loading of closed box-like structure.
~
INTERACTION
OF OBJECTS WITH AIR BLAST
141
the right-hand diagram to give the net
pressure. The net loading is necessary
for determining the frame response,
whereas the wall actions are governed
primarily by the loadings on the individual faces.
4.47 (a) A verage Loading on Front
Face.-The outside loading is computed in the same manner as that used
for a closed structure, except that S is
replaced by S'. The quantity S' is the
average distance (for the entire front
face) from the center of a wall section to
PARTIALLY OPEN BOX-LIKE
an open edge of the wall. It represents
STRUcrURES
the average distance which rarefaction
4.46
A partially open box-like
waves must travel on the front face to
structure is one in which the front and reduce the reflected pressures to the
back walls have about 30 percent of stagnation pressure.
openings or window area and no interior
4.48 The pressure on the inside of
partitions to influence the passage of the the front face starts rising at zero time,
blast wave. As in the previous case, the because the blast wave immediately
loading is derived for (a) the front face, enters through the openings, but it takes
(b) the sides and roof, (c) the back face, a time 2L/U to reach the blast wave
and (d) the net horizontal loading. Be- overpressure value. Subsequently, the
cause the blast wave can now enter the inside pressure at any time t is given by
inside of the structure, the loading-time p(t). The dynamic pressures are ascurves must be considered for both the sumed to be negligible on the interior of
exterior and interior of the structure.
the structure. The variations of the in-
Pr
w
a:
=> Ps
<n
<n
w
g:
OUTSIDE p(t)+q(t)
0
~
U
Figure 4.48.
..?.f.
U
TIME
t;
Average front face loading of partially openbox-like structure.
142
AIR BLAST LOADING
OUTSIDE pv--lu-)
+ Cdq(f --ifJ)
W
a:
=>
(/)
(/)
w
a:
C-
O
Figure 4.49
--f;
U
U
TIME
+ L
"2V
Average side and top loading of partially open box-like structure.
side and the outside pressures with time
are as represented in Fig. 4.48.
4.49 (b) Average Loading on Sides
and Top.- The outside pressures are
obtained as for a closed structure (§
4.43), but the inside pressures, as for
the front face, require a time 2L/U to
attain the overpressure in the blast
wave. Here also, the dynamic pressures
on the interior are neglected, and side
wall openings are ignored because their
effect on the loading is uncertain. The
loading curves are depicted in Fig. 4.49.
4.50 (c) Average Loading on Back
Face.-The outside pressures are the
same as for a closed structure, with the
exception that S is replaced by S', as
described above. The inside pressure,
reflected from the inside of the back
face, reaches the same value as the blast
overpressure at a time L/U and then
decays as p(t -UV};
as before, the
dynamic pressure is regarded as being
negligible (Fig. 4.50).
4.51
(d) Net Horizontal
Loading.-The
net horizontal loading is
equal to the net front loading, i.e., outside minus inside, minus the net back
face loading.
OPENFRAME STRUcruRE
4.52 A structure in which small
separateelements are exposed to a blast
wave, e.g., a truss bridge, may be regarded as an open frame structure.
Steel-frame office buildings with a majority of the wall area of glass, and
industrial buildings with asbestos, light
steel, or aluminum panels quickly become open frame structures after the
initial impact of the blast wave.
4.53 It is difficult to determine the
magnitude of the loading that the frang-
INTERACTION
lJJ
OF OBJECTS WITH AIR BLAST
143
P
a:
::>
U>
U>
lJJ
~
OUTSIDE
p(t-t-)+
0
--fp+-
U
Cdq(t--b-)
..L
U
U
TIME
Figure 4.50.
Average back face loading of partially open box-like structure.
ible wall material transmits to the frame
before failing. For glass, the load transmitted is assumed to be negligible if the
loading is sufficient to fracture the glass.
For asbestos, transite, corrugated steel,
or aluminum paneling, an approximate
value of the load transmitted to the
frame is an impulse of 0.04 pound-second per square inch. Depending on the
span lengths and panel strength, the
panels are not likely to fail when the
peak overpressure is less than about 2
pounds per square inch. In this event,
the full blast load is transmitted to the
frame.
4.54 Another difficulty in the treatment of open frame structures arises in
the computations of the overpressure
loading on each individual member
during the diffraction process. Because
this process occurs at different times for
various members and is affected by
shielding of one member by adjacent
members, the problem must be simplified. A recommended simplification is to
treat the loading as an impulse, the
value of which is obtained in the following manner. The overpressure loading impulse is determined for an average
member treated as a closed structure and
this is multiplied by the number of
members. The resulting impulse is considered as being delivered at the time the
shock front first strikes the structure, or
it can be separated into two impulses for
front and back faces where the majority
of the elements are located, as shown
below in Fig. 4.56.
4.55 The major portion of the loading on an open frame structure consists
of the drag loading. For an individual
member in the open, the drag coefficient
for I-beams, channels, angles, and for
members with rectangular cross section
I
144
AIR BLAST LOADING
REAR WALL
IMPULSE
O.O4Abw + Ibm
W
U
Q:
0
LA.
~
t-
FRONT WALL IMPULSE
c
O.O4Afw + Ifm
0
r.
t-
+
L
-fq+U
2U
TIME
Figure 4.56.
Net horizontal loading of an open frame structure.
is approximately 1.5. However ,because
in a frame the various members shield
one another to some extent from the full
blast loading, the average drag coefficient when the whole frame is considered is reduced to 1.0. The force F, i.e.,
pressure multiplied by area, on an individual member is thus given by
The result may thus be written in the
form
F (frame) = q(t)A,
where A = IA.
I
4.56 The loading (force) versus
time for a frame of length L, having
major areas in the planes of the front and
rear faces, is shown in Fig. 4.56. The
symbols AI'" and Ab'" represent the areas
F (member) = CA(t)Aj,
.of
the front and back faces, respecI y, w h.ICh transml.t Ioadsore
bef
f al.1
where Cd is 1.5 and A I IS the member.
.tlve
area projected perpendicular to the dl.ure,
rection
of
blast
propagatIon.
For
the
I
oa
loading
force
on
.mem
IS
the
frame,
however,
the
F (frame) = Cd'l<t)IAj,
where Cd is 1.0 and IAj is the sum of
the projected areas of all the members.
I
an
d ..'"
d
mg
1.fi
an
Id
lmpu
be
rs,
respec
'"
on
t.
1.bare
ses
lve
th e
overpressure
f ron
Iy.
Alth
t
an d
oug
b ac k
h
d rag
d..
d.
oa mg commences lmme lateIyaterf
the blast wave strikes the front face,
i.e., at t = 0, the back face is not fully
loaded until the wave has traveled the
distance L, i.e., at time t = UU. The
INTERACTION
OF OBJECTS WITH AIR BLAST
average drag loading, qa' on the entire
structure at this time is considered to be
that which would occur at the distance
LI2 from the front of the structure, so
that
q a = C dq
( -.!::-),
2U
and the average force on the frame, Fa
(frame), is
F (frame) = q ( -.!:.-) A,
a
2U
145
ters are small compared to the lengths.
The discussion presented here provides
methods for determining average pressures on projected areas of cylindrical
structures with the direction of propagation of the blast perpendicular to the
axis of the cylinder. A more detailed
method for determining the pressuretime curves for points on cylinders is
provided in the discussion of the loading
on arched structures in § 4.62 et seq.
The general situation for a blast wave
approaching a cylindrical structure is
represented in section in Fig. 4.57.
where t lies between L/ U and t+ +.
2U , as seen in Fig. 4.56.
4.58 (a) A verage Loading on Front
Surface.-When
an ideal blast wave
impinges on a flat surface of a structure,
the pressure rises instantaneously to the
reflected value and then it soon drops to
the stagnation pressure (§ 4.25). On the
curved surface of a cylinder the interaction of the blast wave with the front face
h
I
.
.
ever, In terms of the average pressure,
CYLINDRICAL STRUCTURE
4.57 The following treatment is applicable to structures with a circular
cross section, such as telephone poles
and smokestacks, for which the diame-
the load appears as a force that increases
with time from zero when the blast front
arrives to a maximum when the blast
wave has propagated one radius. This
occurs at a time D/2U, where D is the
diameter of the cylinder. For the blast
where Cd is 1.0, as above. After this
time, the average drag force on the
frame at any time t is given by
.
(
Fa (frame) at tIme t =q
L
)
t -2U
U
A,
q
d
IS
muc
more
comp
ex
In
BLAST
WAVE
FRONT
Figure4.57.
Represenlationof a cylindrical structure.
t
e
.
al.
1
H
ow-
146
AIR BLAST LOADING
lJ.J
~
2p
U>
U>
lJ.J
(1:
C-
p(t) + Cdq(t)
/
..
--tp
2U
U
TIME
Figure 4.58.
Average pressure variation on the front face of a cylinder.
pressure range being considered, the
maximum average pressure reaches a
value of about 2p as depicted in Fig.
4.58. The load on the front.surface then
decays in an approximately linear manner to the value it would have at about
time
t = 2D/U.
Subsequently,
the average pressure
decreases
as shown.
The
Complex vortex formation then causes
the average pressure to drop to a minimum, Ps2' at the time t = 3D/2U; the
value of P'2 is about half the maximum
overpressure at this time, i.e.,
Ps2 -I
-T
P
( W3D )
drag coefficient for the front surface of
the cylinder is 0.8.
4.59 (b) A verage Loading on the
Sides.-Loading
of the sides commences immediately after the blast wave
strikes the front surface but, as with the
closed box discussed in § 4.41 et seq.,
the sides are not fully loaded until the
wave has traveled the distance D, i.e.,
at time t = DIU. The average pressure
on the sides at this time is indicated by
The average pressure on the side then
rises until time 9 DI2 U and subsequently
decays as shown in Fig. 4.59. The drag
coefficient for the side face is 0.9.
4.60 (c) Average Loading on Back
Surface.-The blast wave begins to affect the back surface of the cylinder at
time DIlU and the average pressure
gradually builds up to Phi (Fig. 4.60) at
a time of about 4D/U. The value of Phiis
P,I' given approximately by
given by
INTERACTION
OF OBJECTS WITH AIR BLAST
147
w
a:
:)
cn
cn
PS1
W
g:
P (t-
~)
+ Cdq (t -&)
PS2
030
90
UW
T
W
0
tp+ W
TIME
Figure 4.59.
Average pressure variation on the side face of a cylinder.
The average pressure continues to rise
until it reaches a maximum, Pb2' at a
time of about 20D/U, where
.back
+ Cg
Pb2 = P .:?:QQ
U
U
Consequently, the net horizontal loading cannot be determined accurately by
the simple process of subtracting the
loading from the front loading. A
rough appro~imation
of the net. I~ad
may be obtained by procedures sImIlar
The average pressure at any time t after
to those described for a closed box-like
structure (§ 4.45), but a better approx-
(
the maximum
(~ )
)
is represented by
Pressure at time t = P (
+ Cfl (
imation is given by the method referred
to in § 4.65 et seq.
t --ill-
)
ARCHED STRUCTURES
)
4.62
The following treatment is applicable to arched structures, such as
ground huts, and, as a rough approx-
t -~
2U
where t lies between
20D/U and r; +
imation, to dome shaped or spherical
structures.
The discussion presented
011. U. The drag coefficient for the back
here is for a semicylindrical
surface is -0.2.
4.61
The preceding discussion has
been concerned with average values of
the loads on the various surfaces of a
cylinder, whereas the actual pressures
with the direction of propagation of the
blast perpendicular to the axis of the
cylinder. The results can be applied to a
cylindrical structure, such as discussed
above, since it consists of two suchI.
::~::::::sIY
semicylinders
from point to point.
structure
with identical loadings on
148
AIR BLAST LOADING
...
a:
:)
~
00po.
Po,
P(f-!ii)+Cdq(,-~)
--2U
,
+
U
U
0
P+'ilJ
TIME
Figure 4.60.
Average pressure variation on the back face of a cylinder.
each half. Whereas the preceding treatment referred to the average loads on the
various faces of the cylinder (§ 4.57 et
seq.), the present discussion describes
the loads at each point. The general
situation is depicted in Fig. 4.62; His
the height of the arch (or the radius of
the cylinder) and z repres.ents any point
on the surface. The angle between the
horizontal (or springing line) and the
line joining z to the center of curvature
of the semicircle is indicated by a; and
X, equal to H(l -cos a), is the hori-
zontal distance, in the direction of
propagation of the blast wave, between
the bottom of the arch and the arbitrary
point z.
4.63 When an ideal blast wave impinges on a curved surface, vortex formation occurs just after reflection, so
that there may be a temporary sharp
pressure drop before the stagnation
pressure is reached. A generalized representation of the variation of the
pressure with time at any point, z, is
shown in Fig. 4.63. The blast wave
BLAST
WAVE
FRONT
H
Figure 4.62.
Representation of a typical semicircular arched structure.
INTERACTION
PI
OF OBJECTS WITH AIR BLAST
149
--
w
a:: P3
:)
C/)
--
C/) P z
w
a::
a.
P(tJ+Cdq(tJ
0
U
Figure 4.63.
TIME
Typical pressure variation at a point on an arched structure subjected to a blast
wave.
front strikes the base of the arch at time t
= 0 and the time of arrival at the point z,
regardless of whether it is on the front or
back half, is X/U. The overpressure
then rises sharply, in the time interval tl,
to the reflected value, PI, so that tl is the
rise time. Vortex formation causes the
pressure to drop to P2, and this is followed by an increase to P3,the stagnation pressure; subsequently, the pressure, which is equal to p(t) + Cdt) ,
where Cd is the appropriate drag coefficient, decays in the normal manner.
4.64 The dependence of the pressures PI and P2 and the drag coefficient
Cdon the angle a is represented in Fig.
4.64; the pressure values are expressed
as the ratios to Pr' where Pr is the ideal
reflected pressure for a flat surface.
When a is zero, i.e., at the base of the
arch, PI is identical with Pr' but for
larger angles it is less. The rise time tl
and the time intervals t2 and t3, corresponding to vortex formation and attainment of the stagnation pressure, respectively, after the blast wave reaches
the base of the arch, are also given in
Fig. 4.64, in terms of the time unit H/U.
The rise time is seen to be zero for the
front half of the arch, i.e., for a between 0° and 90°, but it is finite and
increases with a on the back half, i.e.,
for a a between 90° and 180°. The times
t2and t1 are independent of the angle a.
4.65 Since the procedures described
above give the loads normal to the surface at any arbitrary point z, the net
horizontal loading is not determined by
the simple process of subtracting the
back loading from that on the front. To
obtain the net horizontal loading, it is
necessary to sum the horizontal compo-
150
AIR BLAST LOADING
L-
15
'3
1.0
0.8
10 ~
:::>
~ 0.6
,
~
~
L!.
0.4
"If;! 0.2
0
'2
~
,
"~
Z
0
:)
~
ri: -0.2
w
~
-0.4
f=
-0.6
0
20
40
60
80
ANGLE
Figure 4.64.
100
a
120
140
160
180
(DEGREES)
Variation of pressure ratios, drag coefficient, and time intervals for an arched
structure.
p, A
w
u
a:
P' A..?:...I!u
0
lJ..
-.J
<{
IZ
0
N
a:
0
:I:
O.4q(f)A
I-
w
Z
+
TIME
Figure 4.66.
Approximate
fq
equivalent net horizontal force loading on semicylindrical
structure.
INTERACTION
OF OBJECTS WITH AIR BLAST
nents of the loads over the two areas and
then subtract them. In practice, an approximation may be used to obtain the
required result in such cases where the
net horizontal loading is considered to
be important. It may be pointed out that,
in certain instances, especially for large
structures, it is the local loading, rather
than the net loading, which is the significant criterion of damage.
4.66 In the approximate procedure
for determining the net loading, the
overpressure loading during the diffraction stage is considered to be equivalent
to an initial impulse equal to PrA(2HIU),
where A is the projected area normal to
the direction of the blast propagation. It
will be noted that 2HIU is the time taken
for the blast front to traverse the structure. The net drag coefficient for a single
cylinder is about 0.4 in the blast pres-
151
sure range of interest (§ 4.23). Hence,
in addition to the initial impulse, the
remainder of the net horizontal loading
may be represented by the force 0.4
q(t)A, as seen in Fig. 4.66, which applies to a single structure. When a frame
is made up of a number of circular
elements, the methods used are similar
to those for an open frame structure (§
4.55) with Cd equal to 0.2.
NONIDEAL BLAST WAVE LOADING
4.67 The preceding discussions
have dealt with loading caused by blast
waves reflected from nearly ideal
ground surfaces (§ 3.47). In practice,
however, the wave form will not always
be ideal. In particular, if a precursor
wave is formed (§ 3.79 et seq.), the
loadings may depart radically from
100
80
;;;
Go
60
'"a:
'"
on
on
'"
~
40
20
0
0
100
200
300
TIME
400
500
(MSEC)
Figure 4.67a. Nonideal incident air blast (shock)wave.
600
700
152
AIR BLAST LOADING
400
=
300
VI
0IAI
~
200
VI
VI
W
u:
0-
100
0
0
200
400
TIME
600
(MSEC)
b.
100
80
VI
0-
;;;
60
u:
=>
VI
VI
w
40
u:
0-
20
.
0
0
200
400
TIME
600
(MSEC)
C.
50
40
VI
0W
u:
30
20
=>
VI
VI
W
g::
10
0
0
200
400
TIME
600
(MSEC)
d.
Figure 4.67b, c, d.
Loading pattern on the front, top, and back, respectively, on a
rectangular block from nonideal blast wave.
INTERACTION
OF OBJECTS WITH AIR BLAST
those described above. Although it is
beyond the scope of the present treatment to provide a detailed discussion of
non ideal loading, one qualitative exampIe is given here. Figure 4.67a shows a
nonideal incident air blast (shock) wave
and Figs. 4.67b, c, and d give the loading patterns on the front, top, and back,
respectively, of a rectangular block as
observed at a nuclear weapon test.
153
Comparison of Figs. 4.67b, c, and d
with the corresponding Figs. 4.42,
4.43, and 4.44 indicates the departures
from ideal loadings that may be encountered in certain circumstances. The
net loading on this structure was significantly less than it would have been
under ideal conditions, but this would
not necessarily always be the case.
BIBLIOGRAPHY
*AMERICAN SOCIETY OF CIVIL ENGINEERS,
"Design of Structures to Resist Nuclear Weapons Effects," ASCE Manual of Engineering
Practice No. 42, 1%1.
*ARMOUR RESEARCH FOUNDATION, "A Simpie Method of Evaluating Blast Effects on
Buildings,"
Armour Research Foundation,
Chicago, Illinois, 1954.
*BANISTER, J. R., and L. J. VORTMAN, "Effect
of a Precursor Shock Wave on Blast Loading of
a Structure," Sandia Corporation, Albuquerque, New Mexico, October 1960, WT-1472.
JACOBSEN, L. S. and R. S. AYRE, "Engineering
Vibrations,"
McGraw-Hili
Book Co., Inc.,
New York, 1958.
KAPLAN, K. and C. WIEHLE, "Air Blast Loading in the High Shock Strength Region," URS
Corporation, Burlingame, California,
1%5,
URS 633-3 (DASA 146(}-1), Part II.
*MITCHELL, J. H., "Nuclear Explosion Effects
on Structures and Protective Construction-A
Selected Bibliography,"
U.S. Atomic Energy
Commission, April 1961, TID-3092
PICKERING, E. E., and J. L. BOCK HOLT,
"Probabilistic Air Blast Failure Criteria for
Urban Slructures," Stanford Research Institute,
Menlo Park, California, November 1971.
WILLOUGHBY, A. B., etal., "A Study of Loading, Structural Response, and Debris Characteristics of Wall Panels," URS Research Co.,
Burlingame, California, July 1969
WILTON, C., et al., "Final Report Summary,
Structural Response and Loading of Wall
Panels," URS Research Co., Burlingame, California, July 1971.
*These documents may be purchased from the National Technical
Departmenl of Commerce, Springfield, Virginia 22161.
Information
Service, U.S.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising