the influence of blast and earth pressure loadings …calhoun: the nps institutional archive theses...
TRANSCRIPT
Calhoun: The NPS Institutional Archive
Theses and Dissertations Thesis Collection
1960
The influence of blast and earth pressure loadings
on the dynamic response of flexible underground
two-hinged arches.
Walls, Worthen Allen
University of Illinois
http://hdl.handle.net/10945/12331
DUDLEY KNOX LIBRARY
NAVAL POSTGRADUATE SCHOOI
MONTEREY CA 93943-5101
THE INFLUENCE OF BLAST AND EARTH PRESSURELOADINGS ON THE DYNAMIC RESPONSE OFFLEXIBLE UNDERGROUND TWO-HINGED
ARCHES
BY
WORTHEN ALLEN WALLSB.S., University of Arkansas, 1948M.S., University of Illinois, 1954
THESISSUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN CIVIL ENGINEERING
IN THE GRADUATE COLLEGE OF THE
UNIVERSITY OF ILLINOIS, 1960
URBANA, ILLINOIS
DUDLEY KNOX LIBRARY
NAVAL POSTGRADUATE SCHOOl
MONTEREY CA 93943-5101
ACKNOWLEDGMENTS
This thesis was undertaken under a postgraduate educational
program sponsored by the United States Navy for Civil Engineer Corps
Officers. The author expresses his appreciation to the United States
Navy and to the Bureau of Yards and Docks, United States Navy, for the
opportunity of participating in this program.
Appreciation for timely personal interest and encouragement
is expressed to Dr. N. M. Nevmark, Head, Department of Civil Engineering,
University of Illinois, under whose immediate supervision this thesis
was prepared.
The author is deeply indebted to Dr. J. D. Haltiwanger,
Professor of Civil Engineering, for constructive criticism throughout
the preparation of the thesis, and to Dr. J. L. Merritt, Assistant
Professor of Civil Engineering, for frequent exchanges of ideas.
iv
TABLE OF CONTENTS
Page
CHAPTER I. INTRODUCTION 1
1.1 General Statement of the Problem 1
1.2 Purpose of this Thesis 1
1*3 Method of Approach 4
CHAPTER II. GENERAL DISCUSSION OF THE VARIABLES 7
2.1 Blast Pressure Loading 72.2 Passive Ecjrth Pressure Loading 8
2.3 Structures Considered 9
CHAPTER III. GENERAL RELATIONS COMMON TO EACH LOADING FUNCTION ... 11
3.1 Stress Transformation with Direction 12
3.2 Distributed Load Functions 13
3.3 Basis for the Determination of Analogous Loads 153.3 •! General Equations for the Right Reactions of
Circular-Segment Arches 16
3.3»2 Moment Equations for 180 Degree Arches 18
3.3»3 Moment Equations for 120 Degree Arches 213.3*^ Procedure in Determining Analogous Loads 23
CHAPTER IV. BLAST PRESSURE LOADING 33
4.1 Pressure-Time Relations at a Point 334.2 Harmonic Analysis of Radial Blast Load with Time a Constant 374.3 Blast Loading Approximations 40
4.3.1 180 Degree Arches 434.3.2 120 Degree Arches 48
4.3.3 Summary 52
CHAPTER V. PASSIVE EARTH PRESSURE LOADING 6l
5.1 Unit Passive Earth Pressure 6l5.1.1 180 Degree Arches 625.1.2 120 Degree Arches 63
5.2 Distributed Earth Pressure Load 64
5.3 Analogous Earth Pressure Load 66
CHAPTER VI. ANALYSIS OF THE RESPONSE OF THE ARCH TO THEDYNAMIC LOADING 70
6.1 Method of Analysis 706.2 Sample Calculation 736.3 Discussion of Results 75
CHAPTER VII. QUALITATIVE DESIGN METHOD FOR ARCH RIBS 82
TABLE OF CONTENTS (Continued)
Page
CHAPTER VIII. CONCLUSIONS AND RECOMMENDATIONS 85
8.1 Conclusions 858.2 Recommendations 87
APPENDIX A. BIBLIOGRAPHY 08
APPENDIX B. NUMERICAL CALCULATION RESULTS 89
VITA 125
CHAPTER I
INTRODUCTION
1.1 General Statement of the Problem
When a nuclear weapon explodes, a high pressure wave develops
and moves away^from the point of the explosion, traveling over the ground
surface and through the ground. A shallov-buried arch subjected to this
pressure responds in various ways, depending on the orientation of the
structure relative to the direction of travel of the blast wave, the parame-
ters of the blast loading, and the properties of the soil surrounding the
arch. If the arch is oriented with its longitudinal axis perpendicular to
the direction of travel of the blast wave, the windward side of the arch
is subjected to the pressure a few milliseconds before the leeward side.
This unsymmetrical sequence of loading produces flexural stresses and de-
flection in the arch. The deflection of the arch into the surrounding soil
gives rise to an unsymmetrical passive earth pressure loading.
An understanding of how these loadings occur and their effects on
the arch is necessary for the proper consideration of the design of the
arch rib.
1.2 Purpose of this Thesis
When the arch rib is subjected to the above described loads it
responds in a configuration which is a function of an infinite number of
shapes or modes of vibration, since the arch rib has distributed mass and
elasticity. An infinitely large number of coordinates would be required
to specify the position of the deflected arch. The participation of each
mode depends on the spatial distribution of the loads along the arch rib
2
as veil as the time variation of the dynamic load in relation to the natural
frequencies of vibration of each mode.
The distortion of the arch rib is a function of the stiffness of
the rib, the passive earth pressure loading resulting from the stiffness of
the soil surrounding the arch, the distribution of the soil and dynamic
loads, the time variation of the dynamic load, and the support condition of
the abutments. It is convenient to consider the earth pressure loading due
to the stiffness of the soil as an integral part of the resistance of the
arch. For flexible arches, the stiffness of the soil is greatly in excess
of the stiffness of the arch rib, consequently the properties of the arch
rib may be neglected. Henceforth, unless specifically designated otherwise,
the resistance of the arch rib means the resistance given to the arch by
the soil.
Design of structures for dynamic loading is usually based on the
principle of conservation of energy. When the dynamic load acts on the
structural element, the element deflects. Work is done on the structure by
the load at any point as that point deflects. Due to the speed of the load-
ing, kinetic energy of motion is given to the mass of the element, and, as
the element deflects, strain energy is stored. The strain energy may involve
both elastic and plastic action. A general statement of the principle of
conservation of energy is:
work done * kinetic energy + strain energy
To evaluate this equation the displacement of any point on the
structure under the load must be known. The strain energy is the sum of
all the strain energy in t he element, and may be caused by compressive,
flexural, torsional and shear stresses in the element. The kinetic energy
is the energy of translation or rotation of all the masses in the element.
The evaluation of these quantities would be exceedingly difficult
if no more than the arch rib were involved. Evaluation of the energy-
stored in the soil is a hopeless task, impossible for practical purposes.
Fortunately, the principal modes of response of the arch rib are a sym-
metrical mode corresponding to the compression stresses and an anti symmetrical
mode corresponding to the flexural stresses. This allows a considerable
simplification to be made in the treatment of the problem.
A single-degree-of-freedom system consists of a lumped mass
restrained by a veightless spring. When this system is subjected to a load,
only one coordinate is required to fully define the position of the mass.
By assuming that the antisymmetrical mode of response of the arch rib is
defined by the radial deflection of a point at the quarterpoint of the arch,
which is the point of maximum deflection of the arch, the entire deflected
shape of the arch may be defined with one coordinate, and the arch may be
considered a single-degree-of-freedom system. In essence this neglects any
tangential movements of the arch rib associated with the radial deflection.
After representing the action of the arch rib as that of a single-
degree-of-freedom system, dynamic equivalence relations should be established
between the actual and idealized masses, spring stiffnesses, and loads acting
on the two systems. However, for the purposes of this study, this refinement
is omitted. Instead, the following assumptions are made:
(1) The mass of the single-degree-of-freedom system is equal to
the unit mass of the arch, which is obtained by uniformly distributing along
the arch rib the weight of the soil which creates the passive earth pressure.
(2) The resistance function of the single-degree-of-freedom
system is equal to the unit resistance given the arch by the earth pressure.
(5) The forcing function on the single-degree-of-freedom system
is equal to the unit pressure of the forcing function acting on the arch rib.
Once the mass, resistance function, and forcing function acting
on the single-degree-of-freedom system have been defined, the response of
the system may be determined by the vell-knovn differential equation of
motion of such *a system:
mx + kx = p(t)
vhere k* = the second derivative of the displacement vith respectto time
kx = the resistance of the system as a function of thedisplacement
m = mass of the system
p(t) = the externally applied force as a function of time
The main purpose of this thesis is to investigate the response of
various arch ribs under a limited range of variables -when acted upon by blast
pressure. In order to do this, it is necessary to relate the parameters of
the single-degree-of-freedom system to those of the actual structure under
investigation
.
1.5 Method of Approach
The blast pressure acting on the surface of the ground causes
vertical and horizontal pressures in the soil. These pressures are a
function of the maximum surface pressure, and are time-dependent upon a
pressure wave vhich passes through the soil. Pressure-time relations of
these pressures are formulated.
These pressures are assumed to be principal stresses, the resultant
of vhich acts radially on the arch. The tangential component is omitted.
The total radial load vhich results from these pressures is designated as
a load profile p (<P, t), which is then represented as a distributed load by
the series ) p (t) sin ~-
where pn(t) = iL
f9pr
(cp,t) sin H|-2 Rd 9
lo£ central angle of the arch
R = arch radius
An analogous, uniformly-distributed, antisymmetrical, time-
dependent load, p (t), is then determined. This load has the same generala
r1
effect on the entire haunch of the arch as the load / p (t) sin _ .
The pressure-time diagram for p (t) and the numerical relation between p (t)
and the dynamic load \ p (t) sin are established by equating their
effects at various times as though they both acted statically. The analogous
antisymmetrical load po(t) defines the forcing function acting on the arch
and also on the single-degree-of-freedom system.
The unit, horizontal, passive earth pressure acting on the arch
is determined by Rankine's theory for passive earth pressure. The unit
vertical pressure acting on the arch is found by statics. These two pressures
are assumed to be principal stresses, and their radial resultant is assumed
to resist the movement of the arch. The tangential component is omitted.
The total radial soil load is designated as a load profile p (<p), and is
represented as a distributed load by the series \ p^ sin '
vhere pn= TrJ pr^ 9)
sin ^T2 Rd 9
6
o
An analogous, uniformly-distributed, antisymmetrical load which
has the same general effect on the entire haunch of the arch as the load
Ip sin is determined by statics. The passive earth pressure is
assumed to reach its maximum value when a point located at d/k on the arch
rib deflects radially an amount equal to 0.001B. In addition, the resistance
diagram of this pressure is assumed to be elasto -plastic . The analogous
antisymmetrical load defines the unit resistance of the arch rib and also
the resistance function for the single-degree-of-freedom system.
The weight of the soil that causes the passive earth pressure is
determined by Rankine's theory and by statics. This weight is assumed to
be uniformly distributed along the arch rib and defines the unit mass of the
arch rib and also the mass of the single-degree-of-freedom system.
The response of the equivalent single-degree-of-freedom system is
determined by a method developed by Dr. N. M. Newmark for blast resistant
design, Reference ( 8)
.
CHAPTER II
GENERAL DISCUSSION OF THE VARIABLES
The variables of the problem may be divided into three general
groups corresponding to the blast loading, the earth pressure loading, and
the configuration of the arch rib.
2.1 Blast Pressure Loading
The blast forces acting on the arch rib are conceived to act as a
pressure vave passing through the soil, Figure 2.1. The direction and
velocity of the propagation of this wave in the soil is assumed to be related
to the shock velocity of the surface pressure wave, U, and to the seismic
velocity of the soil surrounding the arch, C . The magnitude of C iss s
assumed to be less than U.
Within the stressed-soil zone, the vertical pressure p (t) is
assumed to be equal to the surface overpressure P (t), but to have a dif-s
ferent pressure-time curve to allow for the retardation effects on the shock
transmitted through the soil. The horizontal pressure P^t) is assumed to
be proportional to the vertical pressure.
The time of arrival t of the pressure wave at any point in the
soil, or on the arch, is determined by the propagation velocity of the
wave. The time of rise t of the vertical pressure to its maximum value Pr m
is assumed to be a function of the vertical transit time t, of the pressure
wave. The rate of decay of the earth pressure wave is assumed to be the
same as that for the surface pressure. These relations are shown
qualitatively in Figure 2.2.
Study of Figure 2.1 indicates that as the shock velocity of the
blast wave increases relative to the seismic velocity of the soil, the
8
unsyrametrical loading decreases, since the pressure wave approaches the
arch more nearly vertically. For the usual range of these variables the
earth pressure wave qualitatively approaches the arch as shown in Figure 2.1.
It should be noted that this pressure wave is considered to have
been induced solely by the surface pressure. The direct ground shock wave
associated with" a nuclear weapon explosion is not considered in this thesis.
Reference (l) states that for pressures of less than 300 psi and for depths
of less than fifty feet the direct ground shock may be ignored.
The arch is assumed to be acted upon by the radial resultant of
the vertical and horizontal pressures. If the ratio of the horizontal
pressure to the vertical pressure is as great as 1:1, the radial pressure
is equal to the vertical pressure. Moreover, the total radial force in-
creases as the ratio of the horizontal to the vertical pressure increases.
The effect of the rise time may be seen in Figure 2.2. The
pressure wave arrives at some point cp, on the arch, Figure 2.1, and the
pressure increases to its maximum value in some time t . At some adjacent
point cp the pressure is at a different level, depending on the relative
arrival and rise times. Reference (2) recommends that the rise time be
taken equal to one-half the vertical transit time.
The following range of variables will be studied in this paper:
P = 100 and 200 psi with U = 3000 and 1*000 fps
C - 1000 and 2000 fpss
t » t . and 1/2 t.r t x>
Ph/py- 0.25, 0.50, 0.75 and 1.00
2.2. Passive Earth Pressure Loading
The unsymmetrical blast load on the arch causes the arch to deflect
into the soil on the leeward side. This deflection brings into action the
9
passive earth pressure of the soil as well as the inertia of the mass of
soil associated with this pressure, since the blast load acts dynamically.
This pressure contributes materially to the resistance of the arch and must
be considered.
While precise evaluation of the passive earth pressure under static
conditions is complex, evaluation under dynamic conditions is equally, or
even more, complex. The passive earth pressure can vary in the length of
the arch due either to the properties of the soil or to the construction
technique used in placing the backfill. Rankine's theory has been chosen as
the simplest method for evaluating this passive earth pressure.
In the application of Rankine's formulas, Reference (3)> the unit
weight of the soil is assumed to be 120 pounds per cubic foot and the angle
of internal friction is assumed to be thirty degrees.
2.3 Structures Considered
The blast loading on arches with central angles of 180 and 120
degrees and radii of 150 and 300 inches is investigated to determine the
effect of the soil variables, arch geometry, and relative rise time on the
arch loading, and to determine a conventionalized dynamic loading function.
Subsequently investigated is the response of arches with central angles
of 180 and 120 degrees and radii of 200 to 600 inches to this conventionalized
dynamic loading function.
10
Air Blast
Wave Front
Air InducedGround Pressure Wave
Ground Surface
Figure 2.1 Underground Arch Subjected to Blast Pressure
For $
For d>
Time
Figure 2.2 Vertical Pressure-Time Diagram for 9
11
CHAPTER III
GENERAL RELATIONS COMMON TO EACH LOADING FUNCTION
To predict the effect of the loads on the axch rib, the loads must
be defined in a convenient manner.
Certain general relations may be derived for defining either the
blast load or the earth pressure load. These general relations concern the
transformation of stress with direction, the method used to determine the
value of the coefficients p (t) and p for the distributed radial load
functions \ p (t) sin _ and \ p sin • which represent the actual
load on the arch, and the basis for assuming analogous loads to represent
these distributed loads.
For purposes of dynamic analysis, the arch is idealized to a
single-degree-of-freedom system. The time-dependent load \ p (t) sinfl
is replaced by a uniformly-distributed, antisymmetrical, time-dependent load
p (t), and the distributed earth pressure load \ p sinn ™
is replaced
by a uniformly-distributed antisymmetrical load p . These antisymmetricalct
loads act outwardly on one half of the arch, and they act inwardly on the
other half.
The direction of the analogous earth load is opposite to that of
the analogous blast load. These anti symmetrical loads define the effects
of the actual loads on the arch haunch.
The values of these analogous loads are found by the application
of the principles of statics. The value of the analogous load for the soil
pressure may be found by equating the moment produced by the distributed load
Ip sin —r— at a point 6/k to those of the analogous antisymmetrical loadn v
p at the same point. The replacement of the dynamic load \ p (t) sinn ™
with an analogous load requires that a basic assumption be made. It is
12
assumed that if two generally similar static loads produce equal static
effects on a structure at a specified point, then the same two loads applied
dynamically with similar pressure-time diagrams of equal duration would
cause equal dynamic effects at the same point. Of course the dynamic
effects of the loads are not defined. The point considered is 0/k.
To determine the analogous loads, the moments caused by the anti-
symmetrical loads p (t) and p are equated with the moments caused by thea a
Z/\ ntf cp V nfl cp
p (t) sin' and \ p sin respectively. In
addition, the effect of a uniform compressive load on the central third of
the arch coupled with a uniform outward load on the outer thirds of the arch
is determined. This load provides an analogous load for study of the crown
of the arch.
5«1 Stress Transformation with Direction
Let it be assumed that a horizontal unit stress p, and a vertical
unit stress p are acting on a segment Rdcp of a semicircular arch rib located
at the angle cp measured from the right abutment. Assuming that p and p.
are principal stresses, the radial pressure p (cp) at cp may be found by the
well-known method for determining the relations between stresses at a point
on different planes passing through the point, Reference (k) .
Rdcp may be assumed to be a plane surface for an infinitesimal
segment. The vertical projection of the plane Rdcp is Rdcp cos cp; the hori-
zontal projection is Rdcp sin cp. Equating the sum of the radial forces to
zero yields the following:
2 2p (cp) Rdcp = u Rdcp cos cp + p Rdcp sin cp
Using the trigonometric identities
pcos cp = 2(1 + cos 2cp)
psin cp m ^(l - cos 2cp)
13
and cancelling Rdcp from both sides of Equation (a),
Pr ((P) - iU + cos 29) 1^ + i(l - cos 2<p) py
- id^ + Pv ) + 2(1^ - Pv ) cos 2<p (1)
The ratios of p./p to be studied are 0.25, 0.50, 0.75 and 1.0.
In addition, these pressures are time-dependent, so p (sP) must be a time-
dependent pressure and is therefore expressed as p (cp>t) when the pressures
considered arise from the blast pressure acting on the surface of the ground.
Let PyXt) = Cp (t). Then Equation (l) becomes:
Pr(<P,t) = C py
(t) + py(t) (1 - C) sin
2<p (2)
This equation is evaluated for various values of C as follows:
For C * 0.25, p„(<P,t) m fc.25 + 0.75 8in2cp) p (t) (3)
For C = 0.50, p (cp,t) =(0.50 + 0.50 sin2cp) p (t) (h)
For C - 0.75, Pr(<P,t) =(0.75 + 0.25 sin
2cp) Py(t) (5)
For C = 1.00, pr(cp,t) = py
(t) (6)
Equations (3) through (6) express the time-dependent unit radial
pressure acting on the arch rib at any point cp for the various assumed
ratios of the horizontal pressure Pj^t) to the vertical pressure p (t).
Evaluation of these equations is set forth in Table I.
3.2 Distributed Load F unctions
Any radial load profile p (cp), such as that shown in Figure J.l,
when acting upon a segment of a circular arch rib with a central angle 6,
may be represented as a distributed load by a trigonometric series in the
following manner:
Pr((P) = Px
sin -|- + p2sin—- + + Pq sin — +
The coefficient p is obtained by multiplying both sides of
Ik
(7)
/ \ nic 9Equation (7) by sin and integrating over the length of the arch rib.
J pr(<p) eln^Rdcp =J px
sin 2 8in £2 Rd9o w
/.6
fP2
sin -2 sin £2 Rd9o
.2 nncp+ + / p sin —-2- Rdcp + --X pn~ a
ou °
(8)
The value of the integral / sin —^nsin —— Rd9 is zero provided
that m and n are integers and m does not equal n. Equation (8) then yields:
2n«9
= i v R0n
Pn=
$R
o0
J pr(cp) sin £2 Rdcp (9)
If p (cp) can be expressed analytically, the value of p may be
found by integration. If p (9) is an irregular load profile for which
there is no convenient analytical expression, the integral may be evaluated
by numerical methods since the integral / p (9) sin £2- Rdq> represents the
nwparea under the load profile multiplied by sin —~ .
The sign convention adopted throughout this thesis is that a
positive load acts inwardly and a negative load acts outwardly. The
15
xiupharmonic components of loading p 6in —~ shown in Figure ).l are in
n u
qualitative agreement with the original load p (<p); therefore the sign of
the coefficient p2 is definitely negative. The sign of the coefficient p_
has been taken as negative, although no positive statement can be made that
this is the case for the original load p (<p) in Figure 3.1.
Included in Figure 3*1 are two cases of uniform radial loading.
p sin —
r
2- on the crown and haunchn 6
of the arch may be represented by such uniformly distributed loads.
If the load profile p (cp) is time-dependent, it is expressed as
P (<P>t). The procedure described above may be used to determine the co-
efficient p (t), with the qualification that p (t) must be evaluated at a
specified time.
Reference (5) contains a full development of the methods of
harmonic analysis as applied to structural engineering problems.
3.3. Basis for the Determination of Analogous Loads
As stated at the beginning of this chapter, the analogous loads
p (t) and p are derived by equating their moments to the sum of the moments& EL
of the distributed loads \ p (t) sin —S- and \ p sin —2J., respectively.
The time dependency of p (t) is not considered in this section since thecL
load \ p (t) sin -— is assumed to be acting statically at the instant
the analogous load p (t) is evaluated.
Moment equations for all load profiles in Figure 3»1 &ce found
in this section after general equations for the right reaction of any two-
hinged semicircular arch are established. Since the blast load and soil
Zn«Pp sin —~ , one set
of moment equations suffices for both analogous loads.
16
5.3«1» General Equations for the Right Reactions of C ircular-Segment Arches
Since the circular-segment two-hinged arch, Figure 3.2, is one
degree indeterminate, one equation based on Castigliano's theorem,
Reference (k), is used for the determination of the reactions. The horizontal
thrust at the right reaction is chosen as the redundant. It is assumed that
the arch rib may* be treated as a thin ring, making it unnecessary to consider
the theory of curved beams. The effects of rib shortening also are not
considered
.
The following geometrical relations are derived in accordance with
Figure 3*2:
Figure 3.2a BC = 2R sinJ9
AB = BC cos^ d - <p) = 2R singCp cog^< e - 9)
AC = BC sin^{0 - 9) = 2R sini9 sin^ 9 - 9)
BD = 2R sin^e
Figure 3.2b DF = 2R sinj(e - 0)
EF = 2R sini(e - p) coe^e - 0)
= moment arm of P about D
Figure 3.2c FG = R sin (9 - P)
a moment arm of P about G
Figure 3. 2d AB = moment arm of V about C
AC = moment arm of H about C
DB = moment arm of V about D
In Figure 3»2d, the total strain energy U in the arch rib due to
the moment M caused by the radial load P acting at a point on the rib
located by p may be expressed as
17
°"/ 2©^=/ 2EIRdCP
Since the abutments are assumed not to move, the contribution of the right
reaction H to the strain energy is zero and may be expressed as
ai-J ^ Rdcp =< a >
Two equations are required to express the moment in the arch rib;
one equation for angle <p less than or equal to £, and one for cp equal to or
greater than p . For < cp 5- 0,
1^ = V(AB) - H(AC)
» 2VR sin ^p cosi(e - cp) - 2HR sin^cp sinj(0 - cp)
(b)
^g- -2R sin^cp sin^S - cp) (c)
For B < cp < 6,
Mg = V(AB) - H(AC) - P(FG) (d)
* 2VR sin^qp cosi( - 9) - 2HR sin^cp eln^ ©-cp) - PR sin(9"-£)
^g- = -2R sin^cp sin^<0 - 9) (e)
Substituting Equations (b) through (e) into Equation (a),
°-j 1S. OH"
RdcP + JM2 5T Rd<P
= / [ 2VR sin^cp cosj( 0-cp) - 2HR sin^cp sini< 0-cp) ]
J o
[ -2R sin^cp sinj(0-cp)] Rdcp +
r e
/ [ 2VR sin^cp cos^( 0-cp) - 2HR sinjcp sini( 0-cp)
- PR sin(cp-B)][-2R sin^cp sin5<0-cp)] Rdcp (f)
Solving Equation (f) for H yields
18
e
o4/ VR5 sin
2 §sin(±2) cos(^)d*
-2 / PR5 sin(cp-3) sin | sin(-^)d<p
H % («)
i*/6E5 Bin
2| si„
2(±2) av
Referring to Figure 3.2, the sum of the moments equated about
the left reaction gives
VcZiM=2PR sin(ffi cos(ffi p^^)
1)52R sin| 2 sin |
Substituting Equation (10) into Equation (g),
P {"SUfl [l . * sine - i cose] + [cosI
-I cos(f - (3)
- £cos(|2-0) +(6^) sin(f -e)J|H =
e - 2/ e% ~ . e e^ + e cos (75) - 3 sin ^ cos £
(11)
Equations (10) and (ll) define the right reaction of any circular-segment
two-hinged arch acted upon by a radial load P located at any angle £.
3.3*2. Moment Equations for l8o Degree Arches
For a semicircular arch, the central angle 9 is * , and
Equation (ll) reduces to
H =I [sin + (p - «) cos p] (12)
Using Equation (12), the horizontal right reaction may now be
DSRPfound for the harmonic load component p sin —~ .
19
For P » p, sin — Rdcp = p sincp Rdcp, since Q = x,
if* P1R
H «= — / p, sin <p[sin<p + (cp-x) coscp] Rdcp = -j—
For P = -p2
sin ^2 Rdcp >
-a f -p2
sin2cp[sincp + (cp-x) coscp] Rdcp = - PgR
For P = -p., sin 39 Rdcp,
if* 5H = - I -p, sin39[sincp + (cp-*) coscp] Rdcp « 4 p R
nxcpThe loading configurations p sin —~ are shovn in Figure 3»1«
n o
It is also necessary to investigate the effect of the uniform
antisymmetrical load p and the uniform symmetrical load p shovn in
Figure 3.1. For the uniform antisymmetrical load,
n
H = - / -P tsincp + (cp-x) coscp] Rdcp + - / p[sincp + (cp-x)coscp]Rdcp = pR
2
For the uniform symmetrical load,
x 2x
H = j I ^-p[sincp + (cp-x) coscp] Rdcp + i / ^ p[sincp + (cp-a)coscp] Rdcp
3
+ j \2% p [sincp + (cp-x) coscp] Rdcp = | f$ pR
TEquation (10) is used to find the right vertical reaction.
With 6 = x, Equation (10) reduces to V = £ P sincp.
20
f n
For P = pxsin -2 Rd9, V = £ / px
sin 9 Rd9 = ^KR
For P = -p2
sin ^2 Rd9, V = - £ / pg
sin2<p sin9 Rd9 =
p, sin39 sincp Rdcp =
o 5
For the uniform antisymmetrical load,
it
V = £ /2
-P sincp Rdcp + £ / p sincp Rdcp =
2
For the uniform symmetrical load
* 2*
V = i I^ -P sin Rd + £ /
5p sincp Rdcp
5
2j2« p sincp Rdcp =
The value of zero for the last four loads was to be expected from
the symmetry of the arch and the loading configurations. It is interesting
to note that the two-hinged semicircular arch is statically determinate for
the last four loadings.
All the above values for the right reactions for a 180 arch are
tabulated in Table II.
The moment at any point on the arch in Figure 3-1 located by 9
due to a distributed load p, sin -3- Rd£ acting at (3 isi &
dM = -p sin£[R sin (9-0)1 Rdp
The total moment at 9 is
21
M - -H(AC) + V(AB) - / P1sinp [R sin(cp-p) ] Rdp
. - !£ [2R six, | sin^] + !|f
[2R sin | cos(i2)]
+ p^2
£ sinq> sin2p - cos(p ( | - -^^ )
p.R x - ( *-2q>) cos9 - 3 sin<p1u
A similar procedure is used to find the moment at 9 due to the
other loading configurations in Figure 3«1« The results are tabulated in
Table III.
3. 3.3« Moment Equations for 120 Degree Arches
When the central angle of the arch is 120 degrees, Equation (11)
reduces to
sin(^ - p)
sin .
3
p-¥+ cos - -
I cos(| - p) - j cos (n-p) +g
? sin(| - p) |»
™« 2« 2 « I ! * *:= + -=- cos =• - 3 sin rr cos =3 3 3 3 3
P Ucosp + 2j- sinp)( | -3J *) + i -(g + -jr) cosp
+(f
- ^) sinp + -£ p cos p - £ p sinp
= P I.83979 - 2.70578 cosp + O.16665 sinp
+ 1.59330 p cos p - O.9199O p sinp (13)
22
Equation (13) may be used to find the right horizontal reaction
of an arch with a 120 degree central angle acted upon by a concentrated
radial load P located at any angle p. The reaction for p, sin -^ Rdcp is
found as follows:
H = / T Pxsin
Icp 1.83979 - 2.70578 coscp + O.16665 sincp
+ 1*59330 9coscp - 0.91990 cpsincp Rdcp = O.53076p R
The horizontal reactions for the remaining harmonic loads and the
uniform loads are found in the same manner, and are set forth in Table II.
Equation (10) is used to find the vertical reaction.
jtcp
For p1sin ~ Rdcp ,
v = /5
I^—5 Pi sin I <P sin <TT " ') Rd(P = °'6 Pi RJ o L 2 sin J J -
1 * > x
3
The remaining vertical reactions have been found, and are stated
in Table II.
For a segment Rd£ at £ loaded with p1 sin -& Rd £ , the moment
at cp in Figure 3.1 is
dM = -p sin | £[R sin(cp-P) ] Rd£
and the total moment at cp is
M = -0.53076 pxRlAC] + 0.6p
1RtAB]
- / Pxsin
I plRsin(<p-e) 1 Rd?
P- cp
= -0.53076 pxR J2R sin I
sin(-2_I)j
2*9
+ 0.6 p^R 2R sin I cos(-2—g)
-/ px sin ^ p [R sin(cp-p)] Rdp
= (0.78^99 - 1.35965 sin9 - 0.781*99 coscp + 0.8 sin ^) ^R2
23
The moment equations for the remaining load configurations shown
in Figure 3*1 for an arch with a 120 degree central angle have been found
and are contained in Table III.
3 .3 • ^ • Procedure in Determining Analogous Loads
The moment equations tabulated in Table III have been evaluated
for every ten degrees. The results are tabulated in Table IV and shown
in Figures 3*3 &a& 3«^« These values are for static loads. They do not
indicate the dynamic moment which may exist at any time in the arch.
Inspection of Figures 3«3 and 3*^ shows that the effect of the
analogous antisymmetrical load is almost identical to that of the load
2»Pp? sin —-r1 • If Pp bas a value of one, then the value of the antisymmetrical
load must be slightly less than one if the effect of each is to be the same.
However, the antisymmetrical load must also include the effect of the loads
p, sin ~- and p, sin ^^- . Therefore, the moment that \ p sin "— causes10 3 / n 6
at a point e/h is first determined, and subsequently a value of the anti-
symmetrical load is found which will cause the same moment at e/h. The
point 6/k is selected since the unsymmetrical blast load produces the maximum
effect at this point. The time must remain constant for each numerical
evaluation of p (t), as described below,a
The actual blast load on the arch is p (<p, t) . At any given time
the load profile p {y,t) can be determined and analyzed to give a distributed
load \ Pn(t) sin —^- . By assuming that \ p (t) sin —~ acts statically,
its effects may be found at 6/k. An analogous load can be found which
causes the same effect. By repeating this operation at various times, the
pressure-time diagram of p (t) can be determined. This load must then be
applied dynamically to determine its actual effect on the arch.
2k
In determining the static load of the earth, the analogous load
is found by equating the moments at Q/k of the distributed static load
\ p sin ^~- with the moment of the anti symmetrical load. Since the earth
pressure is not a dynamic load, only one evaluation is needed.
Further inspection of Figures 3-3 and J.k shovs that the effect of
Znfltpp sin —^ on the central third of the arch may be
represented by the symmetrical load which acts in compression on the central
third of the arch and in tension on the outer thirds. This load is
designated p . If the distributed load is a time-dependent load, the same6
procedure is used in finding the pressure-time diagram for p (t) as that
used for p (t)
.
25
TABLE I
COEFFICIENTS FOR RESOLVING VERTICAL AND HORIZONTAL PRESSURE INTORADIAL PRESSURE
Angle Radial pressure p (9>t) in terms of vertical pressure
10
20
30
ko
50
60
70
80
90
100
110
120
130
140
150
160
170
180
Assumed ration P^/p
0.50 0.75
0.250 0.500 0.750 1.00
0.273 0.515 0.758 1.00
0.338 0.559 0.779 1.00
0A38 0.625 0.813 1.00
0.560 0.707 0.853 1.00
0.690 0.79^ 0.897 1.00
0.815 0.875 0.938 1.00
0.913 0.9U2 0.971 1.00
0.978 0.985 0.993 1.00
1.000 1.000 1.000 1.00
0.978 0.985 0.993 1.00
0.913 0.91*2 0.971 1.00
0.813 0.875 0.938 1.00
0.690 0.794 0.897 1.00
0.560 0.707 0.853 1.00
0A38 0.625 0.813 1.00
0.338 0.559 0.779 1.00
0.273 0.515 0.758 1.00
0.250 0.500 0.750 1.00
26
I
I wa -h
CO
P 5>CM
oi
c5
£1
51
CO
CM
OI
>Q
W
HB
3>
cr>
-HCD
PiI
8t— 3
O
a s
a
COI
CMCD
•Hto
Pi
\f\ of- lf\-* c—CM CON> H• •
o o1
o
c5
I
C
00
Hft
CU
ITSCO
sdo•HPO0)
co
xio
%HB
9•H
w 3n
«
27
CVJ
a
1ePi
58-
Ba
8
§4>
OO
8-PO0)
w8
°8HI
cd
3
K M M K|CVj «|K> CVl|^ 8h 5h ah W|KA cvi|on-4-pv
Vn8-
Vn
3h
Vn Vn Vn Vn Vi Vu Vn Vu Vn Vl Vl9- 8- e- 8- 8- e- 8- &• 9- 8- 8-
Vl Vl Vn Vn Vn Vn Vll Vn Vn Vn Vll
o o o o o «|fO, o o o o O
ICD
K\
I
8-
co
oo
8^CVJ
I
«
I
ITSCVJ
t| CD
9CO
HPi
CVJ
cd
d•Ha
PiI
OJCO
OO
*Va•HCO
I
8-
co
ITS
CD
(J•HCO
K"\Pii
8-
aCO
I
9-
co
oo
5CO
pIfN
oi
8-
co
ooI
H
I
8-
co
ITS
It
CD
9
d•HCO
00
o+8-
CO
8
Io
a•HCO
8-|q,
dCO
8-
•SCO
tr\
oi
CVJ
CD
dCO
PiI
8-
ONlCVJ
d
&ooI
8-co
8
io+
o
CD
CO
8-
O
ON
oI
ICO
oo
8-
dCO H
1 CO
J"l>- 8- t-H IT\
8 a Hvo
• CO KSo •
U"\ O+ p 1
8 HH8•
•
o1
eg
ON
H i
8
•
8-
co
oo
c—-4-
•
o+8-
dHCO
Hsevj
On
P*l
o
w
28
OJOH co
Pi 3O
gH CO> «a
9 CO I
H oo
O 0)H eo
i g
2•HP•HCO
O
a1
§•HPO0)
CO
u
&CO
o
Si5S
CO
a ^
Oil
d•H10
PiI
tlCD
•HCO
t—vO OvOvO t—O H C--H r^Olf\t— -* t— U"N ITN
S3CO KMAO H H
OOOOOOOOOOOI I I I I I
v^J H KNK>01
ITNCO^QHH OJ K\^ j*
6060066660I I I I I I I I I
_ CO u\
-* OJ ITN O C— CO K^ ON O Q' rliOKNOtncvi oncvjO O O H
OOOOOOOOOOO1 1 l l l
HvO t^ t>-fcv C— t--\Q"
jQb* i^GI^O OJCO knco 1
3 CVJ
HHCOOJ^OJCOHH-J.OJOJKM<>vK>>OJOJr-l• ••••••••OOOOOOOOOOOI I I I I I I I I
m5 H O OWO t/N ON QnvO -*OJ tr\K-\0 04 «\-$ OHlA0000000000
• •••••••••O OOOOOOOOQO
OOOQIAOQC-CO ON
I
egHII
CD
OJ K>Qi>- kS£-
OOO
_ KNON^OV0 4-0 WKMT\
rHOMA83S
00000001 1 1
t-vO OvOt-O H C-H OITNC— -* t— lf\CO iOia r^opO H H H O0666060
I I I I I
OvOcp h
00001
O l/NIAOJOJCO
8
3
IfNOQJ IAcb oj
9«61 1 1
6 61 1
OJ 2|t—OCH HO O
• • •
OI
O1
O1
H O ON
OJ
800* • •OOO
> q o o q o £H OJ K"N-* lAv
"8 °<
» i
3
Pr W±_
29
(a) Radial Load Profile p (<p) and Harmonic Load Components
* n * q>
p sin —t-1^n
(b) Uniform Symmetrical and Antisymmetrical Loadings
Figure 3.1 Various Load Configurations
50
L CBE =-£ $
L CBA =-+-(#-<£
4-^-/3)
DE // To FO
E F is a Tangent Line
Figure 3.2 Geometry of a Circular Arch Segment
51
+ 0.5
180 120 90 60
Angle<f> , degrees
Figure 5.3 Moment vs Angle cp for 180 Degree Central AngleTwo-Hinged Arch
52
1
/- Uniform Antisymmetricol
/ • 2t</>
t r p2s,n-
a
g «^^l
/—Uniform S
/"S/--p3
sir
ymmetrical
3tt<£
9
If
it <h
P, s,n-f-
/^-Unifor/
m Symmetrical
LP3
s,n\TTJ>
e
+0.20
+0.15
+ 0.10
+ 0.05
-0.05
v>•ocoQ.
-Coc
c0)
Eo
CVJ
ccQ.
-0.10
-0.15
120 90 60 30-0.20
Angle 4> , degrees
Figure 3.U Moment vs Angle cp for 120 Degree Central Angle Tvo-Hinged Arch
33
CHAPTER IV
BLAST PRESSURE LCADIHG
k.l Pressure-Time Relations at a Soint
As stated in the general discussion of the variables, Section 2.1,
the blast pressure acting on the surface of the ground above the point of
interest causes a vertical pressure at that point. Accompanying this
vertical pressure is a horizontal pressure of varying magnitude, depending
on the soil. The horizontal pressure may be as low as fifteen percent in
loose, dry soil, fifty percent in moist soil, and one hundred percent in
saturated soil, Reference (l). One of the purposes of this thesis is to
study the effects of this variation.i
The vertical pressure is propagated downward with a velocity
equal to the seismic velocity of the soil. At the point of interest, the
time for the pressure to increase to a maximum is a function of the vertical
transit line. The effect of varying the rise time will be studied.
It has been shown in Section J.l that the horizontal pressure
may be expressed in terms of the vertical pressure. Vertical pressure-time
relations at a point on the arch in terms of surface pressure are there-
fore required to define the total radial load profile p (<P>t) acting on
the arch. The equations below are derived from the geometry of Figure 4.1.
The velocity of the wave propagation is:
v = (1*)
where v = velocity of propagation of the earth pressure wave
Cs= seismic velocity of the soil
U = surface blast wave velocity
3*
The time required for the pressure wave to arrive at any point <f>
on the arch rib is:
=R - R sin(cp -
p)a v x '
where <p = angle locating a point on the arch rib
R = radius of the arch
t = time measured after initial contact of the pressurevave with the arch rib
c*
£ = arc tan rr-
The time required for the pressure at 9 to rise to its peak
is:
R - R sin? + Ht = S. for t » t. (16)r C ' r t % '
8
R - R sincp + Htr
«gg
£ , for tr
= i tt
8
where H = depth of cover on the crown
t = rise time of the pressure
t. = vertical transit time of the pressure wave to 9
R - R sin<p + Ht = £t C
8
The vertical pressure-time variation at a point on the arch rib
is shown qualitatively in Figure 2.2. The pressure-time diagram is assumed
to be a straight line in the rise part of the diagram and to have an ex-
ponential decay rate after reaching its peak value P . The total impulse
of the force pulse at any point must be the same as the total impulse of
the surface force pulse. This means that as the rise time increases, the
decay rate must also increase in order to maintain the same impulse in
55
the force pulse. It is assumed in this study that the decay rate at the
shallow points in the soil considered in this thesis is the same as the
decay rate of the surface force pulse. This is a reasonable assumption
since the maximum unsymmetrical loading occurs within the first few milli-
seconds after the blast wave begins to act on the arch.
The decay rate for a side-on overpressure P having a positive
phase duration t, may be obtained from Reference (6) . The decay rates for
the two overpressure levels studied in this thesis are shown in Figure k.2.
For the 100 psi overpressure, t, is assumed to be 1500 milliseconds, while
for 200 psi overpressure, t, is assumed to be l6i*0 milliseconds. The
weapon yield is 1 HT.
The complete vertical pressure-time diagram for any point can be
determined only by numerical evaluation of the pressure on that point at
various times. A sample calculation of a pressure-time diagram follows.
Assume that a 180 degree central angle arch with a radius of
twelve and one-half feet is subjected to a 200 psi overpressure blast wave
originating from a 1 MI weapon. The duration of the positive phase of the
loading is l6kO milliseconds. The arch is buried in soil having a seismic
velocity of 1000 fps, with zero depth of cover on the crown. The time of
rise of the earth pressure at a point in the soil is assumed to equal the
vertical transit time of the earth pressure wave. The ratio of the
horizontal to vertical pressure is assumed to be 0.25
•
Reference (7) give6 the following equation for the shock velocity
U of the surface blast wave:
U= cq(1 + 6p/7P )
2
where c = ambient sound velocity ahead of the shocko
36
p = peak overpressure behind the shock
p = ambient pressure ahead of the shock
.'. U - 1130 (1 + 1200/103) 2 - **023 fps
By Equation (HO, v = S » 1000 x UOZ}= ^ fps
\|C2
+ U2
\l 10002x 1»023
2
By Equation (15), p « arctan (IOOO/U023) = 13°58'
t - 3~2 [1 - sin(<p - 13°58')] , secondsa y [d
- 12.86 [1 - sin(cp - 13 58»] , milliseconds
By Equation (l6) t » iqqq ( ^ " sinCP) > seconds
= 12.5 (l - sin<p)> milliseconds
Let 9 = kO°, Then
t = 12.86 [1 - sin(lK>° - 13 58')3 = 7.2 ms81
t = 12.5 (1 - sin 1+0°) = h.5 ms
Therefore at cp = kQ degrees, the earth pressure wave arrives
7.2 ms after the wave initially contacts the arch rib at 9 = 90° + 13°58'.
The vertical pressure rises at a rate of 200/4.5 = ^5 psi per ms. At 8 ms,
the pressure is (8 - 7.2) V? = 36 psi. At 10 ms, the pressure is
(10 - 7.2)^5 = 126 psi. At 11.7 ms, the pressure reaches a maximum value
equal to the surface overpressure, or 200 psi. Thereafter the pressure
decays at a rate assumed to be equal to the decay rate for the surface
overpressure as shown in Figure k,2. Inspection of Figure k.2 shows that
the decay may be represented accurately by a straight line down to a value
of A p/Ap = O.65 and At/t, = ~ 0.02. The pressure decrement is then
\57
0.55(200) = 70 psi, and the time interval is 0.02(161*0) ~ 55ms. So the
pressure decays at a rate of 70/55 = 2 psi per ms. Therefore at 12 ms, the
pressure is 200 - 2(12 - 11. 7) = 199 psi. At Ik ms, the pressure is
(199 - 4) = 195 psi. This procedure is continued until a time greater than
(55 + 11) ms is reached. Then the time interval is expressed as a fraction
of t, and Ap/Ap *is read from Figure k.2. The complete pressure-timed m
relation for 9 = UO degrees is recorded in Appendix B, Table B5.
To obtain the complete vertical pressure-time relations for the
entire arch, the above procedure is accomplished for every ten degrees on
the arch. Table B5, Appendix B, shovs the vertical pressure-time variation
for the entire arch up to 200 milliseconds. The unsymmetrical loading is
readily seen. For example, at ten milliseconds the windward side of the
arch has pressure acting on its entirety, whereas the leeward side is not
fully loaded until after sixteen milliseconds. The unsymmetrical loading
shifts from the windward side to the leeward side during the time the blast
pressure acts. This effect may be seen at thirty milliseconds. The
pressure on the windward side, cp greater than ninety degrees, averages some
eight pounds less than the pressure on the leeward side. The effect of
the decay rate may be seen by comparing the pressure at the crown with
that at the abutments. For example, at twenty-four milliseconds the pressure
at the crown has reached its maximum and decayed to approximately seventy-
five percent of the maximum surface pressure, while the pressure at the
abutment is approximately equal to the maximum surface pressure.
k. 2. Harmonic Analysis of Radial Blast Load with Time a Constant
Based on the vertical pressures tabulated in Table B5, a radial
load profile p (<?>, t) can be determined for any instant of time given in
38
that table. After p (<P>t) has been defined, the distributed load
I p (t) sin ——• can be evaluated. A sample calculation, for t = 10 milli-
seconds, is given in Tables V and VI.
In Table V, the data in column "p (t)" is obtained from Table B5,
while that in column "p coefficient" is derived from Table I for PyVp
0.25. The data in column "p (<P>t)" of Table V is derived by multiplying
p (t) by the p coefficient. The resulting radial pressure load profile
p (<p>t) is shown graphically in Figure U.3. The coefficients p (t) of
Equation (9) are determined by numerical integration. This is done by
tiR nfltpmultiplying the load ordinate at each ten degree interval by r-g sin —~ ,
2adding the results and multiplying the sum by •= . The resulting dis-
tributed radial loading \ p (t) sin —jp , with t = 10 milliseconds, is
shown as a dotted line in Figure ^.3.
Table VI contains the harmonic analysis of the same blast vave
acting on a 150 inch radius arch with a central angle of 120 degrees. The
radial pressure p (9>t) in Table V from 30 degrees through 150 degrees is
used in Table VI from degrees through 120 degrees. The resulting dis-
tributed radial loading \ p (t) sin ^~ is shown in Figure ^.3»
Inspection of the data in column "p (9,t)", Table VI, reveals that
there is a radial pressure of ten or more pounds acting at all points on
the arch rib. Therefore, a uniform all -around pressure of ten pounds may
be subtracted from the radial pressure before the harmonic analysis is made,
since it is desired to resolve harmonically only the radial pressure causing
flexure in the arch rib. The column "p adjusted" contains the remaining
radial pressure.
The coefficients p (t) of Equation (9) are found by numerical
integration. This is done by multiplying the load ordinate at each ten
degree interval by J| sin £|2 , adding the results, and multiplying by JL .
39
The resulting distributed load, p (t) +\ p (t) sin —* , vith t = 10 milli-
seconds, is shovn as a dotted line in Figure 4.5
•
I
It is assumed that the first three terms of the series
nfltpp sin —~ adequately represent the loading on the arch for the purposesn v
of this thesis.
The radial load on the 180 degree arch at a time t = 10 milli-
seconds is:
p (<p,t) = 166 sin9 - 13 sin29 - 26 sin3<p
Since the time is constant, the load may be considered to be a static load.
An analogous load may be determined which will have the same effect on a
given point of the arch. Let the point be located at cp = 1+5 degrees.
p sin -~ all cause
negative moment at this point. Using the coefficients in Table IV, the
moment of the load is
M= [l66( -0.02262) + 13( -0.33333) + 26( -O.O8857) ] R2
= [-3.75 - 4.33 - 2.30] R2
= -10.38R2
A uniformly distributed antisymmetrical load acting in on the
windward half of the arch and out on the leeward half produces a moment of
0.1al+22p R at cp = k$ degrees. Thereforea
2
P * ^ 25 psi for a time t * 10 msa
o;iute2R
However, the actual load on the arch is time-dependent and consequently p
must be a time-dependent load, p (t). To define the pressure-time diagram
for p„(t), the load p (9>t) on the arch is harmonically analyzed at variousa r
times.
The analogous load for the crown may be found in a similar manner.
The moment caused by the above load at the crown is
1<0
M « [166( 0.035^) + 26(0.125)] R2
2= 9.13 R inch, povinds
A symmetrical load p acting in compression on the central thirds
of the arch and in t ension on the outer thirds of the arch produces a
moment of 0.15^7 p R at the crovn
s
2.'. PR(t) = ? ,:L?R
p = 59 pel, t = 10 ms8
0.15^7R
If .3 Blast Loading Approximations
Appendix B contains the results of numerous harmonic analyses of
the blast load on an arch arising from a 1 MT nuclear weapon. The analyses
were made and analogous loads found as described in the preceding section.
These analyses were made in order to determine the effects of the variables
discussed in Chapter II on the loading the arch receives, and to provide a
basis for ideal! zing the blast load on the structure without the necessity
for actually harmonically analyzing the blast load.
The blast load must be defined in relation to the effect being
studied. In this thesis, the effect studied is the deflection of the arch
due to flexural stresses, the area of primary interest being the haunch of
the arch. The effect of the unsymmetrical blast load on the haunch of the
arch is most conveniently related to an analogous uniformly-distributed
load which acts inwardly on one half of the arch and outwardly on the other
half. The effect of the blast load on the crown is conveniently related
to a uniform load acting in compression on the central third of the arch
and in tension on the outer thirds. Values of such loads for various
assumed conditions are shown in Tables Bll and B12.
kl
After the blast load has been related to the specific effect
being studied, a load-time curve must be established to shov the variation
of the load vith time. It will be noted in Tables Bll and B12 that the
analogous loads are for a specified time only, and that they vary appreciably
with time. Appendix B contains a fev typical analogous load-time curves.
Similar plots were made of all analogous loads.
Study of these load-time curves revealed that the peak values of
the analogous loads are some fairly uniform function of the maximum surface
overpressure P for all assumed values of the variables. In addition, the
shapes of these load-time curves are approximately initially peaked tri-
angular load pulses, which decay rapidly to a relatively small value and
then continue to decay at a much slower rate. The duration of the rapid
decay phase of the loading corresponds approximately to the time required
for the pressure to engulf the arch. The duration of the slow decay phase
of the loading is relatively much longer.
The peak values of the analogous loads found in Appendix B are
shown in Table VTI and plotted in Figures k.k and k.J. The data in
Appendix B and Figures k.k and k.5 may be qualitatively summarized as
follows:
(a) For any given overpressure, the peak value of the
analogous load for the haunch increases as the value of
the seismic velocity of the soil increases. This is ex-
plained by the fact that a higher seismic velocity allows
the pressure to penetrate the soil to a greater extent
and thus a larger segment of the arch on the windward side
is loaded before the leeward side is loaded.
k2
(b) For any given overpressure, the peak value of the
analogous load for the crown decreases as the seismic
velocity of the soil increases. This is explained by
considering the effect of the load profile p (9>t) on
m2P (<fct) sin ^2T Rd<P • ^e11 c „ has
%. r v s
a relatively high value in comparison with U, a greater
portion of the segment from 9-, to 92 is located on the
windward side of the arch and the values of the sine
function of the angle 9 are both positive and negative.
When C has a lower value in comparison with U, thes
load profile p (9>t) is located more nearly in the central
third of the arch, where the value of the sine function
of the angle 9 is negative. This increases the numeri-
cal value of the analogous crown load.
(c) The higher ratios of P^/p increase the peak value
of the analogous load for the haunch. This is explained
by the fact that the area of the total radial load profile
p (9>t) is greater for the higher ratios, and thus the
total unsymmetrical load is greater.
(d) The p./p ratio has little effect on the peak value
of the analogous crown load. The peak value of the crown
load is primarily a function of the location of the load,
i.e., the central third of the arch. In this area, the
value of the coefficient for determining the resultant
of the horizontal and vertical forces is approximately
one for all values of Ph/p • Consequently, changing the
ratio has little effect on the radial load.
*3
(e) The assumed rise time has relatively little effect
on the peak value of the analogous loads. The peak
valuee of both loads are reached very early in the load-
ing of the arch -when the pressure wave has not penetrated
the soil to any extent. For shall ov depths, the rise
time is* practically equal to the rise time of the surface
pressure, -which is assumed to be instantaneous.
(f) Although the higher ratios of PjVp cause a higher
peak value of the analogous haunch load, they actually
cause a less severe dynamic load effect. If the ratio
is high, the pressure, after the pressure vave has en-
gulfed the arch, is more nearly a uniform compression,
causing less flexural action in the arch. On the other
hand, if the ratio is low, the uniformity of the compres-
sion is reduced, increasing the flexural action in the
arch. For the higher ratios of Pl/p it is found that
the analogous loads change direction, thereby creating
a negative phase of loading. The peak value of the
pressure in this negative phase is small, but could pos-
sibly cause an outward movement of the crown of the arch
since it has a relatively long duration.
k.J.l l8o Degree Arches
Since the blast loading may act over all, or over part, of the
arch segment, one approximation to the peak analogous loads may be obtained
by considering the blast loading acting on the arch in such a way that the
maximum value is given to each component of loading in the series
Uk
I p (t) sin ^^ . The ratio of the horizontal pressure to the verticaln 6
pressure which causes the most severe dynamic loading is 0.25, therefore
this value is used in the following derivations. The effect of the rise
time of the pressure and the time of arrival of the pressure wave is not
considered. The vertical pressure at all points on the arch is assumed to
be equal to P . The unit radial pressure at any point is, by Equation (j),m
Pr(<P,t) = (0.25 + 0.75 sin
29) Pm
Using Equation (9),
*nW - h J pr( *' t) 8in H^Rd,
2P n*maximum pn
(t) = -^ / (0.25 + 0.75 sin <p) sincp Rd9 - 0.955 Pi hk j m
2P_ r*
o
2P r *ntfQf 1 "AT™
tnfnrynmm
P2(t) = iffJn (°* 25 + °* 75 8in <p
)sin2cp Rdq> = -°-398 Pm
2P r¥p5(t) = ir|j«
5(0 -25 + °- 75 8in2cp
) 8in3<P^ -o. 1^ Pm
The maximum effect of the blast loading which may occur, under an assumed
ratio of PjVp 0.25, can never exceed the effect of the sum of maximum
values of the harmonic components found in this manner. The value of the
individual maximum values found, and the effect of the sum, are always
reduced when the seismic velocity of the soil, the shock velocity of the
surface blast wave, and the pressure-time relationships are considered.
If these components acted statically and separately, they would
produce the following moments at the leeward haunch and crown:
*5
M = - 0.023(0.95 5Pm)R
2+ 0.353(0.398Pm
)R2
+ 0.088(0.4oaPm)R
2
= -0.190 P R2
m
M = 0.035(0.955P )R2
+ 0.125(0. 40&T )R2
= O.O85 P R2
c m m ' m
The peak value of the analogous haunch load is
0.190 P R2
p ( t ) = 2_ = o.U6 P0.4l4 r
2
The peak value of the analogous crown load is
0.084 P R2
These values of the analogous loads can never occur, and, for
vant of a better name, they are designated as the maximax analogous loads.
They are shown in Figure 4.4 so that they may be compared with the analogous
loads found by actual harmonic analysis. Such comparisons indicates that
these approximations are conservative. The following reductions are
arbitrarily made:
Maximum p&(t) = 0.37 P
ffl
(17)
Maximum p (t) - 0.45 P (18)
Equations (17) and (18) are plotted in Figure k.k.
To obtain an idea of the magnitude of these analogous loads, they
may be related to a uniformly distributed load acting on a simply supported
beam. Inspection of Figure 3*3 shows that the moment caused by the analogous
load p on the haunch of the arch is very similar to the simple beamCI
moment. The moment at ^5 degrees caused by the analogous load is
M = 0.37(0.41422) P R2
k6
Assume that the simple beam has a length equal to the length of half the
arch, or L = -r . The moment is wL /8. Equating this to the moment in
the arch, it is found that:
8 x 9-37 x O.M^33 P R2
v =
= 0.5 Pm
A similar analogy may be drawn for the analogous load acting on
the crown, with the exception that the length of the simple beam is now
one-third the length of the arch, or -5- . The moment in the arch at
90 degrees due to the analogous load is
M = 0.45(0. 15^7) P R2
0^ 8 x 0.^5 x (O.15V7) PmR2
and w -— .
<f»
- °-5 p„
The above maximax loads occur within approximately two to four
milliseconds after the loading begins to act on the arch. This fact is
evident by inspection of the data in Appendix B. These loads decrease
rapidly since p~(t) becomes practically zero when the pressure wave has
engulfed the arch, and p,(t) also decreases. However, p,(t) and p,(t) are
still acting. It is now assumed that the pressure has completely engulfed
the arch. Since the pressure has engulfed the arch, the all-around uniform
pressure may be subtracted from the expression for p (<P*t). Then
Pr(<P,t)= 0.75 sin
2q>Pm
hi
This radial pressure may be resolved into harmonic components. It is
found that
22m I * Pp^t) = -j-S
/ (0.75 Bin 9) sinqp Rd9 = O.657 Pm** o
2P r * pp (t) = ^ / (0.75 sin 9) sin39 Rd9 = -0.127 Pm
The moment in the haunch at ^5 degrees caused by these two
components is
J^ - - 0.023(0.637Pm)R2
+ 0.088(0.127Pm)R2
= -0.026 PmR2
The haunch analogous load is
0.026P R2
».<*> " o.fa"°'06 p
m («»
The components p,(t) and p,(t) act for the total duration of the
surface blast pressure. Reference (2) states that the duration of the
positive phase of the side-on overpressure may be considered as
t± = o.Uo (-^9-) v ^5(20)
*m
•where W veapon size in MT
t. = the effective duration in seconds of a triangular pulsehaving the same impulse as the actual shock.
The total duration of p,(t) and p,(t) is considered to be t.
.
The time required for p2(t) to become practically zero is found
by using Equations (15) and (16), with the angle 9=0.
kd
This time is defined as:
t! » t + t = duration of initial triangular impulsed a r
mR f B sin p +
Ra ^
C U 2C ' r * t
fu2
The time t! defines t for the load p„(t) = 0.06 pd. am
4.3.2 120 Degree Arches
Approximations of the blast load acting on a 120 degree arch may
be derived in the 6ame manner as that used in finding the approximations
for the 180 degree arch, as described in Section k.3.1. It is necessary to
modify Equation (3) by substituting (cp + 30) for <$, since in Equation (3)
cp is measured from a horizontal radial line joining the arch abutment with
the center of the circular segment. Equation (3) becomes
Pr(<P,t) = [0.25 + 0.75 sin
2(cp + 30)] Pm
= t- + i*r sin cp + * n? sinCp coscp + *r cos cp Pm
Rdcp
Then the maximum possible values of p (t) are
«« 2*2P r— , fex r —~
~
/
Maximum p,(t) = ^-2 / \ T + ir sin <P +fl
sincp coscp
3
+ -4- cos cp) sin £ 9 Rdcp
= 1.103 P
1*9
2*
Maximum pJt) = — /5
( ^ + ^ sin q> + ^-^ sin<p cosqp
+ ^r cos 9 ) sinjcp Rdcp
= -0.505 Pm
2p r sTMaximum px(t) = -r-^ /
"( r + tr sin *P + o sinCP cosCp
5 I*R ^2* 4 10 a^flR ^2** 9
JL cos2cp ) sin 2 q
2
= -0.417 P.m
The harmonic components p (t) sin —~- would produce moment at the
haunch and crown as shown below, if the blast load is considered to act
statically. The moment coefficients are obtained from Table IV, for cp = 50
and 60 degrees.
M^ = - 0.009(l.l03Pm)R
2+ 0.125(0.505Pm
)R2
+ 0.050(0.4l7Pm)R
2
= - 0.094 P R2
m
Mc
= O.Ol4(l.l03Pm)R
2+ 0.051(0 Al7Pm)R
2
= 0.057 P R2
Using the moment coefficient for <P = 30 and 60 degrees in
Table IV, the analogous loads are found to be
.2
Maximum p (t) =0.094 P R
m
0.1547R2
0.61 Pmm
0.057 PmR
Maximum p (t) = *•
80.06556R
= O.56 P
50
m
These values are plotted in Figure 4.5 so that they may be
compared vith the' actual analogous loads found by harmonic analysis, as
given in Appendix B. Since they are conservative, the following arbitrary
reductions are made:
Maximum p (t) = 0.^5 P (21)
Maximum p (t) = 0.4l P (22)a HI
These equations are plotted in Figure k. 5.
These loads may be related to a simple beam load of 0.5 P by the
same method as that described in section U.3.1 for a 180 degree arch.
After the pressure wave has engulfed the arch, there is a uniform
all -around pressure acting which must be subtracted from the radial
pressure in order to determine the loading which causes flexural action.
This uniform pressure is
p = (0.25 + 0.75 sin2
50) P = A- P•ru x y " ^ ' m 16 m
The remaining radial pressure which is to be resolved harmonically is
Pr (<P) - Pu= ( 5 + ^ sin
2cp + 2_1 sinq> cosq> + ^ cos
2cp)P
m- ^ P
m
= ( - -4 + -fr sin 9 + * ^ sincp cos<P + ^ cos <p) P
Then
51
P-^t) » ^ / 3 (_ 3 + 9 sin2cp + ^__2 8ln<p cos<p + ^ cos
2q)) sin|<P Rd<p
3
0.5^6 Pm
2P2*
P5(t) = --£ / ,
5 (- ^ + -^ Bin29 + 2_^2 Sincp cosq) + ^ cos
2<p)sin|<P Rdqp
—XR o
= -0.020 Pm
These loads vould produce the following moment at the haunch and
crown of the arch if they acted statically:
^ = - O.OC^CO^^PjR2 + 0.050(0.020Pm)R
2
= -0.006 P R2
m
M = 0.014(0. 5k6? )R2
+ 0.05l(0.02OP )R2
= 0.009 P R2
The analogous antisymmetrical load for the haunch is
0.006 P R2
mPa
(t) =
0.155R
0.0U Pm (23)
The analogous load for the crown is
0.009P R2
».W- HO.O6556R
O.lU Pm
{2k)
I
52
The time t that defines the analogous loads in Equations (23)
and (2k) is the time required for the pressure wave to engulf the arch and
rise to its maximum value at the abutments. This time is expressed by-
Equations (15) and (l6). The angle 9 in both of these equations is thirty
degrees
.
^.5.3 Summary
The blast load acting on the arch may be defined by numerically
evaluating at specific times the coefficient p (t) for the distributed load
p (t) sin —7~ . The load-time curve for the analogous loads is obtained
by equating the static effects of the distributed load \ p (t) sin -^- to
those of the analogous loads. This is the procedure employed in Appendix B,
The procedure is extremely tedious, and, in fact, unnecessary, since the
blast load is not accurately known. Therefore, a more convenient method
such as that established in Subsections k.J.l and k.3.2 is desirable.
In these subsections, the blast load acting on the arch has been
idealized to two initially peaked triangular force pulses. For investigat-
ing the effect of the blast load on the haunch of a 180 degree arch, the
peak value of the first pulse is 0.37P> and the peak value of the second
pulse is O.06P . For similar investigation of a 120 degree arch the peak
value of the first pulse is 0.1*5 P , whereas the peak value of the second
pulse is 0.04 P . In each instance the duration of the first pulse is the
time required for the pressure wave to engulf the arch, and the duration
of the second pulse is equal to the effective duration of the blast loading.
p 8-
p? a•H
^ 8-•P iA
Pi a
^ 8-P CVJ«-* a
CVJ -HPi a
-^s 8-+>
Pi 0}
-P 8-•» IA
Pi a
8-tA
H09
-p e-
Pi 09
8-CVJ
d•H09
Pi 09
e-
a
09
+>
Pi OO
Pi
% CM -5 t— O rA^OCO onH H H H HQ C— lAt— CO CO rA OOvC— LfN CVJ 0\\5-=* CVJ
^H rH rH rH
OrACVJOCVJrAOrACVJNOHOJOJOJH H CVJ CMI I I I I
O-*00 H rArAHCO-* OI I r-\ H H r-1 I I
I i I I
CVJ LfNCO O CVJ -* IAV0 <Q
81IOO TvQjvO CVJ *H
I
rHI
r-i r-i H r* rt r-i r-\
-*COr-jrArArjCOJ- O
-d-NO .4- t— t— rAC— ONOvO ir\3- CM Q CO irs CVJa
O O O O H t-HvOvO
~> IfN IfN HCO u-\0O
I H HI I
cvj iao itnj- ifN o t~- olr—UNCO vO O\C0 rA liAHI kvj I
I I •
*
v5 Q o o o^oLfNCO O CO LfN lAOO O• •••• • • •OOOHOOOOOH
I I I
O O O ON O rA.* Ov H Ot-KNKNOvOrH »~i rH
8ia
8:ia<
8LfN
OOOOOHOOOI I
Q t—OCO C--* rA^tM3 Q rA CVJ o r- CVJ
I H H rHI
H I
rA\Oa
LfN irvCO CO^O\0N<
S3vO rA
CVJ KVrAvO <
IfN LTVOOCOvONOV
rACVJ
vO rA
OOOOOOOOOO
O O O LfNvp rA-* Q\Q Hr-i rl r4 rt r4
OOOOOOOOOl l I I l l l I
P^O O Q O CVJ CVJ CVJ
lArAO t—-* HH H H H
-* OJ Q rAvQ '
H rAlfN^O f-<
odoodooori
Q lf\-tf COON ON
LfN Q'CO.*'ON ON f-vO IfNrAH
ododododo
OOOOHLfMTNOONrHH r— KMfNC— t>-coH i-{ r4 r-i r4
\Q\Q O O OvlA LArAHf~-vO LArA OCO rAH » 3H H rH H H CO CO
OHH
+>e-•o sia +>•H09 •5?
^—
^
•»
-PrA 8-H rA 1>1 Pi a
•H+ 09
8- «H 1
098-
'p1
CVJ 09
PirA
+ HVOvO 8- 1H 5 8-
09 a•H
«^^ 09P"^H SPi Ha a
H|ON ^^. ^-^4* p
X •» _«s8- 8-
Pi
O K-NCOCOLfN f— KNKNCVJ C\J rA-sl- LfNvD CO
KN KNOO i
ON ON
CO HN KN
On On CO VO LfN
CO CO {A O_ lA rA t— IAIA-* rA CVJ CVJ
OOOOOOOOOH OOOOOOOOO
O O O OvOHvOKNH O QJ -=t CO -* IA IAVO tAon onco co co a5 cqI co:
co on on q ^FH r4 r4 t-{ rl r4 rl r4 r4 r4 rt r-{ r4
OOOOQOQOH CVJ rA-5 LfNvO C—COC^ O H CVJ rAQ O P Q O O
-3- LfNNOH rH H
Pi
55
B
«
?3CO
o
3
I
o
P
<D
+ d•H
^-^ COP /-**-.» Pft C
ft
W
PpO$VI
5-. CO
ft
crQcCM
H03
+> 9--£» *>8- du to
ft
KN
•H00
tr^kvi
fc to
ft
CO
• -p
ft
o r— o cvj -* itn .* trv^t cvj o r— oH t-K\vO t-h- t— C— C— vO KNt—
H
HHHH HHHHH
O CO UNCO O CO li"NCO OCO ITNCO OHWH H CVJH H CM HI I I
O ON ir\-* -* K\ CT\ rr\-# -* ir\ ON OJ" ON r<N.vQ CO CO CO vo K"N ON-3"
r-\ r-t r4 r-i r4 rl r-t
OKMTNKAOON Ht- OQ>OOOlKN IT\
• I I 1 +
r— O r— t— Of" f-ONOQO OQO OQOt— O r— r— o r— r— Or—• •• • •• •••OOHOOO H O O O H O O
I I I
O H CO l/N OMTN O KNUNQ J"CQ OlOiCO (A-* o-sFtc\0 -=T K^CO
r-{ r* M
lT\CO O CO ITS• • • • •
O O O H O O
I H H HI I I
8V0U-NCO
8IfN
O O O H O O OI I I I I
0tf\tc\0\tc\ H Q i/n ON QvO OIltn CT\r—<0 k>0\<5 cvj k dorlvOO K"NMD
t-\ r-\ H rt r-\ r4
8r—vo vo o vo v
O C5 vS Q \D ^
i/n r— co on O ovON Q r—1AOOCM ltn r—
8 ON
IPs CM• •••• • •«•••OOOOOO HOOOOOO
OHIAIAQO\ H\OvOQOO\.VO CVJ-d-vOvO t—vD if\--* CM ONIHHHH r-i r-{ r-i H t-i
HKOX
CQ CO
O H 1AIAO ON HvOvOOOQNtfNH r— hmtn c— r— co c— vq itn itn O <X>H r-i r-\ r-i H r-i H H r-\ H
O O O O Q OH CM KN.3- ITS $ C-CO ONOHW
ONlOJ
to
ft
9-
fl
(0
CMft
+
rr\jCM
q•HCO
Hft
ft
II
+^
€?
$-
ft
sII
-p
^ OnKCM
aHCO
LfNCM
K^VJ
a•H(0
s
5*
55
£
a
COHCO
O
a
o
p.
p«
§
s
Oh
1
3
o
o
<8
8HIt
CM•
o
CO
P4
P»
P.
P<
HII
P<
cm*
oII
>P4
Pi
to
Pi
Pi
CO
CJ
a co
CMH
COt—
l/N U"\
3 21
CM
OON
H CVJ
o
&
$
8 8 8 8 8H <M H CVJ CVJ
o
CM CM
COCM
CO £ HITN
£ J* £ HCO
H .* ir> ^o ^-* CO CM v£> CO
CM ITS K\ vO COK> ir\ -4" t— l/\
H -=t IA W JO <nj* CO f^ t*- CO vO
Lf\ ITS CM Q CO -*CM -4" K"\ \5 -* vO
H CM
56
Figure ^.1 Geometry of Blast Loading on a 180 Degree Central AngleCircular Segment Arch
57
a.
< a.E
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
.
r- 100 f)Si.
^-200 p1
5i.^,,**,~,
0.1 0.2 0.3 0.4 0.5 0.6
Attw
0.7 0.8 0.9 1.0
Figure k,2 Overpressure as a Function of Time
58
isd 'ajnss9Jdl
D !P Dd
o<£
V .c
o o•—
a. <O -e- a>0>
a>
o> o>c a>
< QO00
OCJ o
isd( 9jnssaJd |D;po^
-8
3olf\H
o «)
IO*g
.c «!
a>
u<
TJ0)
oo „ a> a>
<0
-e-a>
<u O «o> Oc O 3< CJ^™
^^^ §ao
a> S
59
(0
Q- 200
0>W
0) I
>O
O9Q_
00
0—9 QHHi—f-
Adopted Value =0.37 P.
Maximax Value = 0.46 P„
20 40 60 80 100
Maximum Value Pa(t), psi
wa. 200
#»
a>k_
3</>
(/>
a>k.Q.k.<V 100>Ojc:
o<D
Q_
100
Maximum Value pc (t), psi.9
CODE
c s= iooo C s
=2000 Ph
Pv
V
t 0.25 1.0
1.00 1.0
V f 0.25 0.5
A 1.00 0.5
Figure k.k Peak Values of Analogous Loads: Antisymmetrical p (t) anda
Symmetrical p (t) for a 180 Central Angle Archs
6o
CL
to
a>>O
o
a.
200
100
to
a)
10CO
>O
Da>
0_
200
100
Maximum Value Pa(t), psi.
Adopted Value = 0.41 P.
Maximum Value Ps(t), psi.
CODE
c s= iooo C s
=2000 J&L Vft
• 0.25 1.0
a 1.00 1.0
7 0.25 0.5
A 1.00 0.5
100
Figure 4.5 Peak Values of Analogous Loads: Antisymmetrical P&("t) and
Symmetrical p (t) for a 120 Degree Central Angle Archs
61
CHAPTER V
PASSIVE EARTB PRESSURE LOADING
If there were no soil surrounding the arch, it would deflect into
the dotted configuration shown in Figure 5«1 under the unsymmetrical blast
loading. However, the soil restricts this deflection by passive earth
pressure and the inertia of the soil mass. To evaluate the passive earth
pressure, the segment of the arch on the leeward side from the springing line
to the crown is assumed to act as a retaining wall. The soil volume and
the relative plane of slip are shown in Figure 5«1»
Under actual conditions, the body of soil located on the windward
side of the arch would most certainly be affected by the blast pressure and
would have to move downward and laterally into the arch to load it. The
energy required to accelerate this mass would lower the energy level in the
blast pressure. In addition, after the surface pressure wave passes the
arch, there would be a vertical pressure of varying intensity acting downward
on the soil body on the leeward side of the arch. This would increase the
shear strength of the soil, the passive earth pressure, and the inertia
force. Evaluation of these later effects would be extremely difficult and
is omitted in this study. This results in a conservative approach.
5.1 Unit Passive Earth Pressure
Rankine's formula for the total horizontal passive earth pressure
force P acting on the vertical plane AV with a height h in Figure 5*1 is
P = iwh^
1 + sin cp'
where K = •? 7—^7 , <P' being the angle of internal* friction of the soil
and w = unit weight of the soil.
62
The unit horizontal pressure along the plane AV in Figure 5*1
varies directly as y varies, y being the distance belov the ground surface.
This pressure then may be expressed as
pp
« wyK (25)
5.1.1 l80 Degree Arches
For a semicircular arch with no cover on the crovn and <p' = 30
degrees, since y = R(l - sincp), Equation (25) becomes
p - vR(l - sincp) J*"*" °*jj = 5vR(l - sincp) (26)p J. - u.5
The angle 9 is measured from the horizontal radial line which joins the
abutment with the center of the circular segment.
In addition to the horizontal passive pressure, it is necessary
to consider the weight of the soil above the arch rib. This soil exerts a
unit pressure
py= wR(l - sinq>) (27)
Using Equations (l), (26), and (27), the resultant radial earth pressure is
Pr (cP) = i (Pv + P
p) + i (P
p- Pv ) cos 29
- i [wR (1 - sinqp) + 3wR (l - sin<p) ]
+ 2 b*R (1 - sin9) - wR (1 - sinq>)3 cos 29
= 2wR (1 - sin9) + wR (1 - sin9) cos 29 (28)
In Figure 5.1, the angle 0, which defines the inclination of the
plane of shear AS with the vertical plane AV, is
6 = i(90°+9')> <P' being the angle of internal frictionof the soil
= i (90° + 30°) = 60° for 9' = 30°
65
From the geometry of Figure 5.1, the total weight of the soil
associated vith the passive earth pressure is
.2
Ws
= § R(R tan 60) + R2
- ^p- 1 v = 1.08 wR2
(29)
Equation (28) is for an arch with no cover on the crovn. If
there is a depth of cover H on the crown, a uniformly distributed hori-
zontal pressure 3*H , as derived from Equation (25), and a uniformly dis-
tributed vertical pressure wH , must be added to the radial pressure actingc
on the arch. For an arch with a central angle of 180 degrees, the resulting
radial pressure due to H , as obtained by Equation (l), is
Pr (cP) - 2 (Pv + P
p) + i (P
p- Pv ) cos 29
= i (wH + JwH ) + i (3vH - wH ) cos 29c c c c
= 2 wH + wH cos 2q> (30)c c x
The total radial pressure on a 180 degree arch with a depth of cover HC
is found by adding the pressures derived from Equations (28) and (30)
•
p (9) = 2wR (1 - sincp) + wR (1 - sin<p) cos 29
+ 2wH + wH cos 29c c
5.1.2 120 Degree Arches
Equation (28) is valid only for a semi-circular arch with the
angle 9 measured from the horizontal radial line. For an arch with no
cover on the crown and a central angle of 120 degrees, Equation (28) is
modified by substituting (9 + 30°) for 9. This yields
Pr(9)= 2wR [l - sin (9 + 30) ] + wR [l - sin(9 + 30)] cos 2(9 + 30)
= wR ( 2 - \ ? sin9 - £ cos9 - ^ cos 39
" "f sin 29 + i cos 29) (31)
6h
If there is a depth of cover H on the crown of the 120 degree arch, the
resulting radial pressure due solely to H is
Pr(9)= i (ph
+ Py) + i (ph- Pv) cos 29
= i (5wH + \fi ) + i (Jvfl - vH ) cos 2(9 + 50)c c c c
= 2 wH + i wH cos 29 - -^ wH sin 29 (32)C C c c
The angle 9 is measured from the radial line Joining the abutment vith the
center.
The total unit radial pressure of a 120 degree arch having a
depth of cover H on the crown is found by adding Equations (31) and (32).
The result is
p (9) = wR ( 2 - \ r sin9 - f cos9 - i cos 39
- -g sin 29 + 5 cos 29) + 2 wH
+ £ wH cos 29 - -i- wfl sin 29 (33)
The total weight of the soil associated with this force is
.2
Ws
= [i (R + Hc )
2tan 6Q - ^-
Jw
= 0.3^2 wR2
+ 1.732 wRH + 0.866 wH2
c c
5.2 Distributed Earth Pressure Load
In considering the unsymmetrical blast loading on the arch rib,
it is advantageous to resolve this loading into harmonic components since
Zi \ n«9p (t) sin —z~- are easier to treat
analytically than the actual irregular blast load. This advantage does not
exist for the passive earth pressure loading, since original analytical
65
expressions are available which may be studied in determining the effect
of the passive earth pressure on the arch. However, resolution of the
passive earth pressure into harmonic loadings does facilitate visualization
of the fact that the earth pressure may be thought of as a distributed
Zn«Pp sin -— in the arch rib. In addition, a desirable uni-
formity of approach is provided for the entire problem.
Equation (28) expresses the unit radial pressure acting on a
180 degree arch with zero depth of cover on the crown. The total force
acting on the arch rib may be resolved into harmonic components by
Equation (9)
•
p (9) = 2wR(l - sinqp) + *R(l - sin?) cos 29
*» - -hi *r<« 8in¥ Rd*
n
Pl=ITr /
2^2vR^ 1 - sinCp) + wR(l - sinq>) cos 2q>]sin |2 Rdqp
= 0.9W wR
=
a
2 [2 Pr(<P) sin2*96
Rd9 = 0.509 wR
2 [2* R Jo
PP(«P) sin 2*9e
Rd<p 1.180 wR
Equation (33) expresses the radial pressure acting on a 120 degree
arch having a depth of cover H on the crovn. The total radial force re-
sulting from this pressure may be resolved into harmonic loading components
as follows:
66
p (9) = vR (2 - 2—-2. sin9 - \ coscp - - cos 39
- «^ sin 2Cp + | cos 29)
+ vHc(2 + | cos 29 - -|- sin 29)
*n - FR /^ Sin *? Rd<P
?1 = ~- /5
[2 - ^j-^ sin9 - £ cos9 - t> cos 39
- -^ sin 29 + | cos 293 sin |9 Rd9
*r
+ vH /5 (2 + | cos 29 - -^ sin 29) sin |9 Rd9
o
= 0.111 wR + 0.864 wHc
p2= ~-
J3pr(9)
sin 39 Rd9 = 0.l8l vR + O.987 vHc
p_ = ~-/
5 p (9) sin §9 Rd9= 0.198 vR + 0.557 ^-> |*R J o
r d c
3
5.3 Analogous Earth Pressure Load
In order that the effect of the unsymmetrical blast load on the
arch rib may be determined conveniently, the blast load is represented as
an analogous uniformly distributed radial load acting in compression on the
•windward side of the arch and in tension on the leeward side. This greatly
facilitates the study of the dynamic effects of this load since the problem
67
of dealing with a time-dependent unsymmetrical load is reduced to that of
dealing with a time-dependent uniform load. The problem may be simplified
further by letting this uniform load define the parameters of the dynamic
load acting on the single-degree-of-freedom system.
It is logical to extend the concept of analogous loads to the
unsymmetrical passive earth pressure acting on the arch. In the first place,
a desirable uniformity of approach to the entire problem is obtained. In
addition, by defining the earth pressure load as an antisymmetrical radial
load acting in compression on the leeward side of the arch and in tension
on the windward side, the value of this pressure can be considered as inte-
gral part of the resistance of the arch rib, since the direction of the
earth pressure is directly opposite to that of the blast pressure.
Furthermore, this resistance provided to the arch by the soil may be used
to define the resistance of the single-degree-of-freedom system.
p sin -~ defined in Section 5.2 ,
coupled with the moment coefficients from Table IV, are used to define the
analogous loads for ISO and 120 degree arches. The resulting analogous loads
define the maximum, or yield, resistance given the arch rib by the soil,
since the passive earth pressure is evaluated at its peak value. The
resistance is evaluated as a function of the moment in the arch rib at a
point located by Qjk, the same point used to define the analogous load for
the blast pressure.
For a 180 degree arch with no cover on the crown, the resisting
moment due to the soil is
M= -0.023(0.9^8 wR)R2
+ 0.335(0.509 wR)R2
+ 0.088(1.180 wR)R2
» 0.252 wR^
68
The analogous load Is
0.252 wR^
a0.43A R
2
= 0.611 wR (3k)
For a 120 degree arch with a depth of cover H on the crown, the
resisting moment due to the soil pressure is
M - -0.009(0.111wR + 0.861fw5 )R2
+ 0. 125(0. l8lwR + O.987WH )R2
+ 0.050(0. 198wR + 0.557WHJR2
= 0.03^wR5 + O.lMfwH R2
The analogous load is
0.03ivwR5 + O.lWfwH R2
p = g.
0.155R
0.2l9wR + 0.929WH (35)
69
t V S
»
1 Hc
/ // /////
3 a'
r ""*
RAN Pp\
/y /
//
h /
Figure 5.1 Passive Earth Pressure on a 180 Degree Central Angle Arch
70
CHAPTER VI
ANALYSIS OF THE RESPONSE OF THE ARCH TO THE DYNAMIC LOAD
The arch rib is idealized as a single-degree-of-freedom system,
and the flexural resistance and mass of the arch rib per se are ignored. A
method of dynamic analysis developed by Dr. N. M. Newmark, References (8)
and (9)> is used to predict the response of the arch rib to the given
pressure pulse.
For the arches investigated in this thesis, it is found that the
response of each is in the plastic, or yielded, part of the assumed elasto-
plastic resistance diagram for the soil. Had the response been less than
the deflection required to establish the passive earth pressure, the dif-
ferential equation of motion stated in the introductory chapter of this study
would have been used to find the maximum response.
6.1 Method of Analysis
For an initially peaked triangular force pulse acting on a single-
degree-of-freedom system having an elasto-plastic resistance function,
Figure 6.1, Dr. Newmark presented in Reference (8) the following approximate
equation to predict the maximum response of the structure:
^ . _£_ Vi^TT + i-li_ ( 56)% nt
l 1 + 0.7 T~
where Pm** = maximum value of the pulse loading
a yield resistance of the structural element
u = ductility ratio = maximum deflection/yield deflection
= x /xm' e
71
mx
T = natural period of vibration of the structure = 2« \j-
—
t.. = duration of the loading
Another technique developed by Dr. Newmark, Reference (9)> can
be used to define the dynamic behavior of the structure for other pulse
shapes which may be represented by several initially peaked triangular force
pulses, Figure 6.-1. The relationship between the peak pressure and the
maximum response of the structure is defined approximately as
****[%* 8-
*
where F, , F2
= damage pressure levels for initially peakedtriangular force pulses of duration t,/T andtp/T, respectively, acting separately
c., c2
= ratio of the peak pressure of the first andand second replacement triangles to the totalpeak pressure
fl'
f2
= cl^ pn/qe^
and M 1^^' respectively, the
maximum value of the individual force pulses
To apply the above described methods of dynamic analysis to the
situation at hand it is assumed that:
(1) The configuration of the arch rib is defined by the radial
deflection of the haunch at a point located at &/k. The structure is
idealized as a single-degree-of-freedom system.
(2) The blast loading on the arch is considered to be a uniformly
distributed antisymmetrical loading. The unit value of this loading defines
the forcing function acting on the single-degree-of-freedom system.
(3) The shape of the blast loading force pulse consists of twoi
initially peaked triangular force pulses. The duration of the first pulse,
t_, is defined as the time required for the blast wave to engulf the arch
72
rib and rise to its maximum pressure at the leeward abutment, a time equal
to t plus t . The total duration of both pulses is defined as the effective
duration of the blast loading, t..
(k) The resistance provided to the arch rib by the passive earth
pressure is considered to be a uniformly distributed antisymmetrical load.
The unit value of this resistance defines the resistance function of the
single-degree-of-freedom system.
(5) The weight of the soil associated with the passive earth
pressure is uniformly distributed along the arch rib. This unit weight
defines the unit mass of the single-degree-of-freedom system.
(6) The passive earth pressure reaches its maximum value when
the radial deflection of the arch at a point d/k is O.l^R. The resistance
diagram of this pressure is elasto-plastic and defines the resistance func-
tion of the idealized single-degree-of-freedom system.
(7) The period of vibration of the arch is defined as
T = 2* \/mx /q~ , where m results from the unit weight found in (5)> x is
O.l^R, and a is the unit passive earth pressure resistance found in (h) .
The general nature of assumptions (l) through (7) above is shown
in Figure 6.2.
Numerical values of the parameters in Equation (36) are presented
in Table VIII. Using these parameters, the maximum responses of arches
with various radii subjected to 100 and 200 psi overpressures have been
determined and are plotted in Figure 6.3. It i6 assumed that the arch is
buried in soil with a seismic velocity of 2000 fps. The resistance and
mass of the arch rib have not been included, and soil arching effects are
not considered.
73
6.2 Sample Calculation
A sample calculation of the response of an arch under an assumed
set of conditions follows.
Let: C = 2000 fps W = 1 MT P = 100 psi U = 3000 fpss m
» 2 ft/ms = 3 ft/ms
w 120* pcf R 200 in. 180 degrees- 16.67 ft.
tr =it
tH =c
The parameters are:
pmax= °*57 Pm
= ° -57(10)) = 37 psi
p'.
= 0.06 P = 0.06(100) = 6 psi
R - R sin(9 - P) R - R sin?Zl
=C U 2Cs s
H^2^
l6.67[l - sin(0 - arctan |) ] l6 6_
=' 6
+ ~t~
= 20 ms
t . o.k (M)2/5
w 1/5
m
= UOO ms
T - 0.0076 Vi = 0.0076 n/200 = 0.107 sec.
107 ms
x = 0.001 R - 0.001(200) = 0.2 in.
% „ 0.6H vR = °'y 200 - 8.5 psi
7*
The force pulse is shown in Figure 6.2. Equations (36) and (57)
must be evaluated by trial and error. Reference (9), Chart I, may also be
used to evaluate Equation (36).
Assume that u = 12. For the first initially peaked triangular
force pulse having a duration of 20 ms, the damage pressure level necessary
for a maximum deflection of x = 12x ism e
— -xt^^' 1 +1 + 0.7(1/^)
1 m 1
-sS**- 1 +1 + o.7( w/20) - 8-*-'i
For the second initially peaked triangular force pulse having a
duration of 400 ms, the damage pressure level necessary for a m«.yimnm de-
flection of x = 12x ism e
1 m1
-qf=^i ViirTT +1 + 0.7(1^/1*00) - ^ ' F
2
The small increase in the peak value of the second pulse due to
the extension of the pressure line from t = 20 ms to t = ms, an increase
Of)of "zrrr (6) = 0.3 psi, is ignored.
The actual peak pressure in the first force pulse is 31 psi; the
actual peak pressure in the second force pulse is 6 psi; and the actual
ratio of the total peak force to the resistance is
^max 37 , „„
% 8.5
Equation (37) may now be evaluated to determine if the assumed
value of u 12 is satisfactory.
75
L F1
F2
5 = 1.0
where c. >Js • 0.84 c~ = T*7 = O' 1^
.,,_„ / 0.84 0.l6 N . „ . n
• • **«37 ( -gjp + j^-) = 1.02 = - 1.0
Therefore the assumed value of 12 for u is satisfactory, and the
maximum deflection is
Xm
s ^ Xe
o 12(0 .2)
s 2.4 inches
6.3 Discussion of Results
Figure 6.3 contains the quantitative results of the analyses of
the response of various sized arches. The results may be qualitatively
summarized as follovs:
(l) For any given overpressure level, the response of
the arch increases as the radius of the arch decreases.
This is partially explained by considering how the load
on the arch and the resistance given to the arch by the
earth pressure are defined. The blast pressure is de-
fined as a function of the peak overpressure only. The
value of the analogous, uniformly-distributed, anti-
symmetrical load is not a function of the radius of the
arch. The earth pressure is defined as a function of
the radius of the arch and varies directly as the radius
varies. Consequently, the response of the arch to any
76
given overpressure varies indirectly as the radius
of the arch varies.
Other factors which probably influence the re-
sponse of the arch have not been studied in this thesis.
One of these is the value of the resistance of the arch
per se. For arches of smaller radii, this resistance
may be relatively large compared to the resistance given
the arch by the earth pressure.
(2) For equal radius arches, the response of the arch
to any given overpressure level increases as the central
angle of the arch decreases.
(5) The effect of the second triangular force pulse is
very important since it has a relatively long duration.
Several analyses of the response of a 120 degree arch to
a 100 psi overpressure -were made, in which the effect of
this second force pulse was not considered. The response
of a 200 inch radius arch was found to be approximately
three inches, compared to a value of sixteen inches when
the effect of the second pulse is included.
(4) Increasing the depth of cover on the crown reduces
the effect of the blast load on the arch. The effect of
the depth of burial of the arch may be related conveniently
to the ratio of the depth of burial of the arch abutment to
one-half of the arch span. When this ratio is one, for
arches having radii of 200 inches or more and central angles
greater than 120 degrees, the flexural action caused by an
overpressure of 100 psi resulting from alHO weapon may
77
be ignored. For a 180 degree arch, this ratio is one
when there is no cover on the crovn, and the deflec-
tion is approximately tvo inches. For a 120 degree
arch, this ratio is approximately 0.9 "when there is
a depth of cover of 0.3R on the crovn, and the deflec-
tion is approximately one inch.
78
TABLE VIII
PARAMETERS FOR DYNAMIC ANALYSIS OF SOIL RESISTANCE
Parameter 180 Degree Arch 120 Degree Arch 120 Degree ArchH =c
H =c
H = 0.5Rc "
pmbeoc
0. 57Pa 0.^5PmmO.to
P' *0.06pm
O.O^Pm
O.C&Pm
*it + ta r
t + ta r *a
+ tr
*2 t.1 \ \
«e0.6llwR 0.219vR 0.h97yiR
Xe
O.OOIR O.OOIR O.OOIR
m 1.08vR2
* Rg0.3^2wR
20.9UOVR
2
1 *Rg
T 0.0076 >/r 0.0087 >/r O.OO96 >/r
79
q(x)
Mass -^ p(t)
'max.
(a) Single-Degree-of-FreedomSystem
(b) Initially PeakedTriangular Force Pulse
q(x)
1 1
'm
(c) Elasto-Plastic BilinearResistance
max
(d) Two Initially PeakedTriangular Force Pulses
Figure 6.1 Idealized Single-Degree-of-Freedom System, Force Pulses andElasto-Plastic Bilinear Resistance
80
Blast Loading Soil Resistance
q(x)
%
0.001 R m
Soil Resistance Diagram
Figure 6.2 Idealized Loading and Resistance Functions of an UndergroundArch Rib Subjected to Blast Pressure
81
40.0
<U|'d' 4.0
200
100 psi. , 9 = 120,
200 psi. , 9 = 180,
200psi., 0=120,
100 psi., 9 - 180,
100 psi., 9 =120,
Hc
=
Hc=0
HC =0.3R~
Hc =—
HC = 0.3R
400 600 800
Arch Radius , inches
1000 1200
Figure 6.3 Haunch Deflection vs Arch Radius
V » 1 MT C = 2000 fps8 *t
P as noted H as notedc
Q as noted
82
CHAPTER VII
QUALITATIVE DESIGN METHOD FOR ARCH RIBS
The problem of the structural design of the arch rib has not been
studied. However, the following qualitative design procedure, based on the
methods used in this thesis, is suggested:
(1) Idealize the arch as a single-degree-of-freedom system, the
configuration of which is defined by the radial deflection of the arch at
e/k.
(2) Define the fundamental period of vibration of the structure
by the mass and resistance of the soil, and the deflection of the arch at
6/k necessary to establish the full passive earth pressure.
(5) Define the forcing function as consisting of two initially-
peaked triangular force pulses, the first pulse having a duration of t -ft,& r
the second having a duration of t . . These forcing functions are uniformly-
distributed, antisymmetrical, radial loads analogous to the actual blast
loading.
(k) Assume a ductility factor u. The end objective is to
design the arch rib, not to determine the response of the soil surrounding
the arch. However, since the arch cannot deflect without the soil ac-
companying it, this ductility factor also defines the response of the soil.
The assumed value of u should be larger than that desired for the actual
arch rib.
(5) Evaluate the damage pressure level equation to determine
the total resistance required of the combined arch rib-soil system.
(6) Subtract the soil resistance from the total required
resistance and design the arch rib with a resistance equal to the remainder.
The usually recommended allowable stresses for dynamic design are to be used.
85
(7) Determine the yield deflection at d/k of the arch sectioni
designed in (6). The yield deflection must be stated in the same numerical
terms as those used in (2) for finding the period of vibration of the
structure
.
(8) Combine the resistance of the soil and arch rib into an
elasto-plastic resistance diagram. This is accomplished by making the area
of the idealized bilinear resistance diagram equal to the area of the
actual trilinear resistance diagram -when both diagrams extend to a numerical-
ly specified deflection. The ratio x /x is assumed; the initial slope of
the elastic part of the actual diagram is maintained; the value of a is
adjusted to make the areas equal. Determine a new period of vibration for
the combined arch-rib soil system, including the resistance and mass of the
arch section designed in ( 6)
.
(9) Using the elasto-plastic resistance function found in (8),
evaluate the damage pressure level equation to determine the maximum
response of the system. The resulting ratio of the nrnlrnnm deflection to
the yield deflection of the system is found and compared with the ratio
x /x used in combining the soil and arch resistances. If these ratios
are equal, the arch section designed in (6) is satisfactory. If they are
not equal, the arch section must be redesigned and (7) through (9) repeated.
The anticipated results of the design procedure outlined above
are discussed below without proof.
There are two principle modes of response of an arch to blast
loading, Reference (10). One of these is symmetrical, corresponding to
the compression in the arch, and the other is antisymmetrical, corresponding
to the flexural deformation of the arch. Consider an underground arch and
assume that the arch rib has been designed for the static load of the soil
cover and the compression load of the blast force. For the arch sizes,
overpressures, and veapon yield studied in this thesis, it is reasonable
to expect that such arches would have a yield deflection on the haunch of
at least one-half inch. For 100 psi overpressure, the maximum deflection
expected of the haunch, considering the resistance of the soil only, is
two inches for 180 degree arches with zero depth of cover on the crown, and
approximately one inch for 120 degree arches with 0.3R cover on the crown.
Therefore the ratio of the maximum deflection on the haunch to the allow-
able elastic deflection of the arch rib does not exceed k:l, even though
the reduction of the maximum deflection due to energy absorbed by elastic
or plastic deformation of the arch rib is disregarded.
For moderately ductile structures, the above ratio, ductility
factor u, ranges from 10 to 30, Reference (8). For designing flexural
members, including reinforced concrete members, the ductility factor usually
is assumed to be 20.
It is therefore concluded that if the arch rib is designed so that
it has a ductility factor of 20 for flexural action, and is covered as de-
scribed above, no increase in the required resistance for static and blast
compression load is necessary to accommodate the antisymmetrical mode of
response for an overpressure of 100 psi.
85
CHAPTER VIII
CONCLUSIONS AND RECOMMENDATIONS
8.1 Conclusions
The following conclusions are drawn from the results obtained
and the discussions thereof in the preceding chapters.
(1) Harmonic analysis provides a convenient means of analyzing
unsymmetrical loads acting on arches.
(2) The peak values of dynamic loads and their load-time
curves may be determined by harmonically analyzing the load at suitable
time intervals.
(3) The effects of the unsymmetrical blast load acting on the
arch may be studied conveniently through the use of analogous loads. When
studying the effect of the blast load on the haunch, the analogous load
employed is a uniformly distributed radial load which acts in compression
on the windward side of the arch and in tension on the leeward side. The
peak value of this load is approximately forty percent of the peak value
of the overpressure. When studying the effect of the blast load on the
crown, the analogous load employed is a uniformly distributed radial load
which acts in compression on the central third of the arch and in tension
on the outer thirds. The peak value of this load is approximately forty
percent of the maximum overpressure.
(k) The effect of the unsymmetrical passive earth pressure load
on the arch also may be studied through the use of an analogous load. This
analogous load is a uniformly distributed radial load acting in compression
on the leeward side of the arch and in tension on the windward side. It
directly opposes the blast pressure load. Its value varies directly with
the radius of the arch.
86
(5) The peak value of the analogous load employed in studying
the blast effects on the haunch of the arch increases as the seismic
velocity of the soil increases. The peak value of the analogous load used
in studying the crown of the arch decreases as the seismic velocity of the
soil increases.
(6) The relative severity of the dynamic load varies inversely
with the ratio of the horizontal to vertical pressure in the soil. It is
to be noted, hovever, that as this ratio increases, there may be a tendency
for the crown of the arch to move outward.
(7) The relative rise time of the blast pressure to its peak
value has little effect on the peak value of the analogous loads.
(8) The passive earth pressure contributes materially to the re-
sistance of the arch. An overpressure of 100 pounds per square inch from a
one megaton weapon causes a maximum response of approximately two inches
in the 180 degree arches studied. Subjected to the same overpressure, the
response of a 120 degree arch with a depth of cover on the crown equal to
three-tenths of the radius is approximately one inch. This response is
considered negligible.
(9) The passive earth pressure developed by the arch decreases as
the central angle of the arch decreases. It is doubtful if arches with
central angles of much less than 120 degrees would develop significant
passive earth pressure under blast loading.
(10) The rigidity of an arch interferes with the deflection of
the arch required to develop the passive earth pressure. This suggests that
flexible arches have certain advantages in blast resistant construction.
(11) The method of approach presented in this thesis may be
extended to consider soils with different angles of internal friction,
87
different unit weights, and varying vater table levels, as veil as to
cohesive soils, provided the cohesion of the soil can be stated in terms
of an equivalent angle of internal friction.
8.2 Recommendstions
It is recommended that the studies made in this thesis be
extended to:
(1) Consider how the blast loading en the arch is affected by
the soil on the windward side of the arch.
(2) Consider how the passive earth pressure loading on the arch
is affected by the increased shear strength of the soil due to the vertical
blast pressure acting on the leeward side of the arch.
(3) Compare the arch behavior predicted in this thesis with
available test data.
(k) Develop a quantitative procedure for design of the arch rib
under impulsive loading which includes the effect of the passive earth
pressure
.
88
APPENDIX A
BIBLIOGRAPHY
1. Anderson, F. A., "Blast Phenomena from a Nuclear Burst, " Journal ofthe Structural Division, Proceedings of the American Society of CivilEngineers, Paper 1856, St 7, November 195$
•
2. "Protective Construction Review Guide, " issued by Office of Secretaryof Defense, January 1959. Prepared under contract SD-52 by N. M.
Nevmark, R. J'. Hansen and others. Classified Secret Security Information.
5» Huntington, W. C, Earth Pressures and Retaining Walls, John Wileyand Sons, Inc., New York, 1957.
h. Seely, F. B., and Smith, J. 0., Advanced Mechanics of Materials,Second Edition, John Wiley and Sons, Inc., New York, October 1957.
5. Hendry, A. W., and Jaeger, L. G., The Analysis of Grid Frameworks andRelated Structures, Chatto and Windus, London, 1958; Prentice-Hall,Inc., New York, 1959.
6. Brode, H. L., "Numerical Solutions of Spherical Blast Waves," Journalof Applied Physics, v. 26, No. 6, pp. 766-775, June, 1955.
7. The Effects of Nuclear Weapons, Samuel Gladstone, Editor, United StatesDepartment of Defense, June, 1957*
8. Nevmark, N. M., "An Engineering Approach to Blast Resistant Design,"Transactions of the American Society of Civil Engineers, v. 121,Paper No. 2786, pp. 1*5-64, 1956.
9. Melin, J. W., and S. Sutcliffe, "Development of Procedures for RapidComputation of Dynamic Structural Response, " Civil Engineering Studies,Structural Research Series No. 171* University of Illinois, Urbana,Illinois, January, 1959*
10. Merritt, J. L. and N. M. Nevmark, "Design of Underground Structures toResist Nuclear Blast, " Civil Engineering Studies, Structural ResearchSeries No. 1**9, University of Illinois, Urbana, Illinois, April, 195^.
89
APPENDIX B
NUMERICAL CALCULATION RESULTS
This Appendix contains the results of the numerical harmonic
analyses made.
In Tables Bl through BIO are tabulated the vertical pressure-time
values of blast pressure on a 180 degree arch with zero depth of cover on
the crown for various combinations of the variables stated in Chapter II.
Since the 120 degree arch has the same radius, this table also provides the
vertical pressure-time values for the blast loading on a 120 degree arch by-
using the values between 30 and 150 degrees.
Tables Bll and B12 contain the values of the coefficients p (t)
for the harmonic components of loading, as found by numerical analysis.
These values were found by using the vertical pressures in Tables Bl through
BIO, applying the coefficients from Table I to convert the vertical pressure
to radial pressure, and then harmonically analyzing these radial pressures.
These tables also contain the values of the analogous loads p (t) and p (t).a s
The following sign convention is applicable to these loads:
(a) p (t) is positive when the load acts inwardly
(b) p (t) is positive when the uniformly distributed loadacts in compression on the windward half of the archand in tension on the leeward side
(c) p (t) is positive when the uniformly distributed loadacts in compression on the central third of the archand in tension on the outer thirds.
Originally it was intended to study only twenty-five foot radius
arches, some with a rise time equal to one-half the transit time and the
remainder with a rise time equal to the transit time. However, after numerous
calculations had been made, an error amounting to a multiple of two was found.
90
In order to vise the previous work it became necessary to decrease the arch
radius to 12.5 feet, and to increase the rise times of some of the arches
to equal the transit time, and of the remainder to twice the transit time.
This accounts for the data shoving t = 2t.. After the error was dis-
covered, only arches with a twenty-five foot radius and t = ^t were
studied. The results are of such a nature that general conclusions may be
drawn without regard to the disparity in the arch sizes.
Figures Bl through B6 are typical graphs of the values of the
coefficient p (t) and the analogous loads p (t) and p (t).XX fit s
91
I po +>
3 uCO -P
a3CO
1
•3 CO
S3?
p§££Pi
13•h ca+> COi o>u
§e(0to >
O
to
28+> H
S3
3
o
oH
OCM
OlA
3
9- otf\
4>
•s 5d•H
o5
c-
&(0
eg
k
3 oON
$aa 8•H Hy-^-Ps^ opT 3
u oCOCO a
& o4 3o•HPi 3s> H
OIfNH
5H
Or—H
sH
1,51
oj u\ do o\ o\ o\ co vo 4
OO OOOO0\i/\0Nt— l/\f^KNi«^fr\H-stcO (J\ 0\ C\ 0\ tO ^O 4
OOOOOONONCOvO-4-OJQOQQ(^ CD 0nOnON0\0\C0vO-4-
o o o o -4-4 CO CO VO 4 QJ O CO CO CO COCVICO ON ON ON 0\ ON CO ^ IfN KN
OO O O W CO ^ 4 Ol O CO vQ VO MD NOCO ON ON ON ON ON CO CO t— tfN K^r<\.
O O oo\Q\C— ir\K\HO>t— ir\ir\,mi/\iao\o\o\o\o\cococo t— u\ k\.
O o o g\t^iAK\rtc2\t-ir»iO!f\if\^O\O\ONO\O\COCO00CO b- IT\ K>
o o 800v04NQCO\n4wW0ICJo\o\o\0\5\cocococo t- ir\ kn
SONl-lAKNH^t-lAlOH H H HO\0\0\0\O\C0COC0C0C0 c— ir\ K>
80\C0^4fllOCO\04fllOOOOo\^chO>0NCF\C0cOcOfl0co t— in k\
8 0>C0vOJtCNQ0Q^0JtQJOOOOONONONONONOScOcocOcoco C— ir\ »r\
8Q00vO^l-C0OC0\Q-*0JOOOOO O\0\0\0\0\coc0cococo C*- itn mm
O Q 0>t— irNK\HQNt^ir\KNH H H Hitn O OnOnOnOnOncocooOcoco t— ir\. k*\
O CO QC0vfl40JOCO\04QlWWC\lmo ono\0\o\0\cocococo t— itn kn
o o-* 0\0\O\O\0\cococ0c0 F-iaka
° ° 3 £ & t-irvtCkHQxt— mmir\i/No\ ON ON OS CO CO CO t- l/\ ir»
O O O C—VO C2\^lAi^r|CT\t>-^t-f-K"\0O ON ON ON 0\ ON CO CO t* UN KN
O O O H 00\0>C-ifNKN,HO\0\ONO\H lf\ CO ON ON ON ON 0\ 00 t— IfN KN.
O O O 4 vo co QN £- if\ rf\ H d H HCVI i/\ On On On On CO vO -4-
HCVJ^-COOJVOO-4-COCMvOOQOCvJ
92
to P«-3+=
<M3to II
s 4>M
a 30} g3
1(0
(4
4 8o oa H•H
IIoITN 09H O
4) 5&'5VT» a
s 1H$
fc
5 s+» Ha a
! 0)
i §V 00
£094>
0) J_,
Q|0) £7
& >O
3 •Hu 09
•H P<
s> H
SI
0)
1
o
4>
05
&
&a
0909
4)
a\
o+»
I
OCVJ
oKN
O
8
o
8
§
a
o3
o
oH
5
OH
8H
a a
O O O O OCM -4-
O O O O
o o o o
o o o 3
o o
o o
o o
O H
» *C\J
£
8 &
SCO
CO vOOv ON
[— ITNON ON
<M
O -*cvj 3-
co voON ON
ON OS
^t cvj
ON ON
KN HON ON
KN K\.-4- VO
ON KNVO ON
£
ON ITN H ON H Q Htf\ t— ON ON ON VO -*
QJ Q CO v£> CO ON QCO O ON ON CO l/N -*
ON C— l/N KN UN t~ ONON ON ON ON CO ITN KN
O- UN KN H CT\ H UN COON ON ON ON CO CO UN KN
t— IfNON ON
-* CVJ
ON ON
KN H ON £- ON -** t~ON ON CO CO t— UN KN
Q 55 ^2 =t ^> cvj voON CO CO CO t- l/N KN
Cv»Qcovo-stcvj-*HirNON ON CO CO CO CO t— UN KN
ON co co co CO co t~- ir\ rr\
ON t— IA K> rl ON rl 0> -*CO CO CO CO CO C— f-3- KN
8C0^4WQSV04WOCO O ON -*ONONONONONcococOcocO f— c- j* KN
8cpvO^CMQCpvQJtOJQCOOQ>-*onononononcococococo t— t- j* KN
8
0\t^ir\K>H5\£-iAK\H On H ON -4-ONONONONONCOCOCOCOCO f- C- -* KN
3QCQSP J*CVJQ c3 ,iP.:5tfisl.SCV>OUN©ONONONONONCOCOCOCOCO t- UN KN
O -4 mt-lAiOH^fl^iOd KN H UN<M itnoSonOnOnOncocococOOO t— un kn
o-4- q\ p\ t- IA K> Hcvj vD On On ON On ON
<2> £r H> ^ "^ cu ^oco co co co t- un kn
OvOOOOQCOvp^tCVJOCO\OCOJtvOIO t- 5 CA ON ON ON ON CO CO F- UN KN
OOCOCVJt—HQNt"- IA A H OA rl 1T\ t-H Jt ^O OnOnONONONONCOCO UN KN
O O IT\ Jt KN KN QJ Q CO vo 4 CJ -gf C— ONCVJ -4 vO CO O ON ON ON ON CO UN KN
O O O CVICO-4 Q \0 QJ ON t- UN £- ON QH CVJ3-VO t- ON ON ON ON CO UN J*
HHCVJC\JC\JKNKN-*3-vOCVJOH CVJ
OOOOOOrrvirNt—QNONONirNt*-KN \0 ON ON t— IfN K>
93
3 OJ vO O <S 0\ t- IA KN
1oZ 4.
a*to n
tQ 3
iau
•51 5o o
o n
l/N CDH O
I*
si
fig
IsPi O
4) (0
b to
3 <uco u03 A4> ET
*3 «HU CO
•H A
I
Od
4>
to
1&d•H
Pi
&to
•51
o
*
ocvj
oN"N
s
5
£
£
oON
00000-*OJQONCOf-t— CVJVOHVO O ON ON ON t>- IfN KN
> H
a
o3
oKN.H
2H
OIfNH
OOOOOONQONCOf-vOvOHvOITN © ON On on On t*- ir\ r^N
OOOOK>QONQOt*-vOirNir\HifNO ONONONONONt— lfNK"NIfN
QOOlAQONCOt—K\ O ON ON ON
Vp l/\ Jt -* O IfNON ON ON t— IfN K\
OOOQONCOh-voirN^j-KN{<>oir\O ONONONONONONONC— IfNKN,
ooooNCor-votrN-4-KN.wcvjqN-i*
OOONCOt-vOUNJttAWHHqN^tONONONONONONONONONt'— J" K\
OQONCOC--vOlfN-si-»fN<NHHON.4'Q OnONONONONONONONOn[--j1- K"N
8gN00r-vOirN-4-KN0JHQOQ\^tONONONONONONONONONONt— -* r<"\
8Q\ 00 S VDirNj-KNOJHQOON-*OnONOnONONONONONONONC— -* K>
8 ON CO t— vOlfN-*tCvOJHOOON-=tOnOnOnOnOnOnOnOnonOnC— -3" k">fo
t-Q ONCO t- vo i/N-sfr N"N OJ HvO Q ONOnONONOnONOnOnON 3>
•*KN
OifNONCOt-vpifN.si-N"N<\iHHOJ*K\t-0\3\0\o\0\0\0\0\0\f-K\A
OQQONCOt— vOlfN.4^r^(^|<^JO-*H -* O ONONONONONONONONt^-irNKN
5H HVO O ONONOnONOnONOnS- ITNKN
OOCVJOQONOOt-vOirN-*J*HlAN"\ C— O ONONONONONONt«-lf\K"N
8 O O 00 2 £ 8QNCO t— VO IfN ITN OJ IfNON ON ON ON ON t— UN KN
a d HOJ-sl-MDCOOCM-*MDCOOQOQHHHHHCVJVOCVIO3
9*
a•H
4>
tt
8,
a
^
oCVJ
o
o
5
o
8
(0 o(0
0) a^Pi
H od r<"\
o H•H•PJ-i
ff 3H
Ol/\
5
oH
51
OOOOO0JCO^*-OvOCUC0Q\QJlfN00H K> IA vfl 00 <7\ ON CO lf\ KN
OOOOOKNKNfAK\ (^ If\ [f\ KN Q O^ OO H l/N COH K> 1ft h 0\ O 8\ ON CO 1T\K>KN
O 6S ON ON ON N IA K\K^N t/\ C—
OOOOH(MJ-CT\00C-^0irN^-f~K>viJK"\ MD O\0NO\0\O\0\0\hlrtKN
OOOvOOQQ\COt^vOl/\-sf»r\vOCVJvOCVJ t— O oSO\0\0\0\0\0\t-«NlO
OOf-hO\OOh-vO«N4KNWH4HUNH CO 0\0\0\0>ONO\0\0\0\t-lA|f\
OOQONCOt*-vOifN-^rACVIHQK\HirNO ONONONO\0\0\ONOSO\0\r-ir\r<^
OQQ\COt^-vOirN-*tCvCV»HQO\OJO-4-O 0\ONONONONO\ONONONO\CO t— \T\ tf\
OONC0t-^lf\4KNWHQCAC00Jq\4C>>0\0\0\0\0\0\0\.0\0\COC0 C— -3" KN.
8 0N CO t-VO U\ Jt KN CVJ H QO\CO H Q\^t0nOn0\O\0N0\0N0\0\OSCOC0 t— .* K"\
ON 00 t- \£ IT> 4 KN CVJ H Q CT\ CQ £- O ON-*caonoaononcaononc^cococOco t*- -3- KN
0> CO t— v£> u\ j* KN GJ H QOSCO t— O Q\ -*O\ONO\0\O\ON0\0\ONC^C0COC0 t^--* KN
8 0>CO t- vO ia 4 rr\ W H Q <7\ CO H <3\ -4"O\O\OnO\O\O\OnONO\O\CO0O Is- -3" KN
og>cor-vou>-*K>cviHQa\coHqN-*^C>i^CAO\OnO\OnCI\O\CT\C0CO t— -3" KN
t-H g>C0 t— vo if\-* KN CVJ HlA CO O\O\O\O\O\ON0\O\O\ Q GPv CVJ Q ITv
ON CO f- lA KN
CVJ l/N CO ONO\0\ONO\O\0\ONO\ON^UNK\
O-H UN ON Q On CO C— vO l/\ ^- KN CVJ lf\ CVJ MDKN IA t^ O ^O\O\O\O\O\0\0\[-UNKV
o f- i— r— r- t~H K\ UN t— COQ\ CO t— SO IT\ J* KN VO KN vO0\ONONONO\ONO\t'-IAlA
O -* O^O CVJCO^t Q ONCO t-VO lf\CO-*vr>CVJ KN ITN MD CO O O\O\0VO\O\t-lAi^
CVI-4-VOCO O CVJ J* VO CO O OJ 4 VO Q OQHHHHHCVJCVJCVICVJVOCMO21 CVJ
95
OOOOOOOOOCM^OCOQCMr>-vOOCQKN vO ON CM vo 0\ ON if\ CO 4A f-i r4 ri rA
o 3 co cvj ia c^ o\ o\ co cm co 3H H H H H H H
o oooooocmhoononunhc*-kno\cmvoCVJ Hvflr(lftO\0\CNCOCO(>-W^HHHHHHHHH
Oa
oKN
o o o o o
oUN
o o o o vo vo
o o o
CVIVQ O <T» UN H t— fC\ONUNHCO-4vOW CO IA o\ a\ o\ 00 CO h- t— t— H ^ 3
sOvO ON UN H t- |C\ fl\ ift H h- KN VO CM 3KN CM 5\ 0\ 0\ 00 CO t>- N t- VO VO H C— 3HHHHHHHHHHHH
CO vO \0 (\JCQ-4QvOCMCQ-4 Q vO O O 33 co o\ o\ co co co t- t- \5 vo ^5 ir\ ri t«- 3
CO
iCO
4)
COCOCD
h
o•H
5
£
8
&
8
o3
o21
oKN
O O t— ON UN H h- K\ ON lf\ H t— lO ON 1A H CO CO 4-4 ON ON ON CO 00 t~- t>- h VO vO m «\ lf\ O vD 3H H H H H H H H
OOCO4p\00|!C04OvQWC04Ov0^00CvlON ON ON CO CO t— t— f- \B ^O lf\ lA ift 4 O vO -4HHHHHHHHHHHHHHH
O Q\ IA rl t— KN ON UN H t«- KN ON UN H C— KN .4 VQ CMONONONCOCO t>- I— S- VO vO lf\IAU\44Ov04H H H H H H H H H H
KN ON UN H C— KN Q> UN H t— KN ON UN H -4 VO CMON ON CO CO CO t- h ^ VO ^ tf\ lA 4 4 4 O <D 3
CM H H H H H H H H H H
^^WCQ40VOWCQ4QvOOJCp4C)4vOC\|ON ON ON 00 CO CO r— C^ "3> ^O vO IA IA 4 4 4 O M> 3HHHHHHHHHHHHHHHHHQ0O-4 O vO W CO 4 OvQ CM CO -4 O^O CM 4 vO CMONONONCOCO t— C— t— VO >vO UNUNUn3-40vD-4CM H H H H H H H H rl ri r-{ H r4 rl
ON t— vO CM CO -4 5? ^ CVI OO 4 O VO W CO 4 4 vQ CMIA t^ 0\ ON CO CO CO f- N (5 vfl U) ir\ UN 4 4 O >5 4HHHHHHHHHHHHHHHHO CO vO \fl CM CO 4 O ^) CM CO -4 Q M3 CM CO VO 00 .43 CO ON ON CO CO 00 t— t— vO vO \0 IA IA 4 O VO 3Hr-4r-ir4f-\r4r-ir-ir-4r-4r4Hr4r-ir-i
3 O O t- MD GO -4 O\0CMC0-4O\0CMCQ-4OO.4;^O IA ON ON ON CO CO t— t— t— <D \0 UN UN H f- 3H H H H H H H
OUN
O O KN t— H UN VOC— H UN vO OJCO^hOvO CMOQ-4 Q CM O -4MD KN ON 0\ ON CO CO CO t— t— \0 vO vO H t>- 3rlr4r-it-i<-*Hr-{r-{r4f-ir-\.-*rA
3 O O O t— vOvO IT\ 4 Q ^ W CO 4 O vfl r\i oQ CO CM -4UNO UNO ON ON CO OO CO t— t~35 H t— 3rHHCMHHHHHHr-iHr-4O OOOOOO^O^CMQCOvOCMCO^tOvOCMJfvOt— 3cOCVJvOONONONCOcOcOf--CM.t~-3
8 O O O O O KN IA t- CT\ H KNUNt~-KN0NUN-4\0ON KN V0 ONONONCOCO CM t—r-ir4r4rlr-lr4r4AKN \0 3
H CM -4 vO CO OOCM3vOCOOCM-4\OCOOQOQHr-4HHHCMCMCMCMCMKNVOCMOH CM
96
O
OH
O<M
$ O
3rf\
o 9-4) -P"-J-P•9 W(0 II
« s•8
XI u. aO -P •H oU -d <D
u\
g a|
•H CO
is co 5*-• fa
8 8 & oS H co r-
ii3
o 3
afi
a> xl a2$8>2fi
*4> ovi a >aa^ ON
fillp."
«>
| 8«sg
09co
4>
H
me
plot
from
1 O3
•H CJ
1o
?& 21a> to >^Sress erpr
o
<3 <H so co H•H A+>
V O O> CM ITSH
* 5V HHXJco
tH Ot—H
£
s c
3 1•H vE-l
O o o o o O O O NOH CMrfN 3 vS <8& a
OO
21
NOt-H
CMONH
CO
9OON
CMIfN
o o o o o O ff\SI
H-4 8 ON
t—00ON
NOrfN
IfNIfNH H
KNOnH
rfN
HIfN00H <S
oIfN
o o o o o NO O IfNIfN
Onl>- * CO rf\.
IfNH ^ 8r4 CM
NO
2f&o8 .SIH H H
CO
HOH 8 3
o o o o drf\-4"
IfN
3ONrf\H
HH 8
CM
NOONH
CMONH 8H 33
NO
HCM
H 8 c3 1NO 2
o o o COH KNNO
-4ITNH 8
CM
NO
2?
CM 8 b?H H 1NO
HCM
HCONOH 3H 5 NO
IfNHCMlf\ S -4" 3
o o -4CM
rfNON
CMNOH
COONH
-4ONH
O2 8 H
00
H H $HCMNOH
COIfNH IfNH
OIfN
HCO g 5
o o 3 00
5.4
2O2 1 3
00
H H<5INOH
COlf\H IfNH
OIfNH 3H
CO
H-4«"NH
O3
OIT- 3
o -4"
CVJ
IfN
3H9 00H
IfNr—H
Hl>-
0-NOH
•0NOH
ONIfNH
IfNIfNH
HIfNH
5>H
ONKNH
IfN l>-
21
tfN
218 3
IfN
3H
COH HITN
HHH
r—
^ vOHONIfNH
IfNIfNH
HIfNH *H
Onrf\H
IfNKNH
HKNH 21
tf\
21
ON
3 8 3
0-ONH •-I
UN00H
HCOH
0-
H H HIfNNOH
HvOH IfNH
rfNIfNH
IfN H l>-tfNH KNH 3
IfN
21 218 3
t—ONH
IfNCOH
HCOH
t—
H H HIfN H
NOH IfNHIfNIfNH H
IfN H-*H H
ffNr<NH S
IfN
21 21
t-
d 8 3
ONONH
t/NONH COH COH
IfN HH NOH
ONIfNH
IfNIAH
HIfN
HON IfN
IfNHH
21
KN
21
ON 00NO 3
8 8CM
NOONH Sn s 5 8 NO sH 3 3H
NOIfNH
CMIfNH S 3 3H H H
NOKNH
CM CO
21
-4"
218 CM
-4
CVJ
KNON
CM
HCOONH
4"ONH
OOnH H 8 00
H HOH NOH
COIfN
H-4-
IfNHOIfNH
COKNH
J**f\
Ho5
o 3
o -4KN. S IfN
21
HH
IfNONH 00H ooH
ON
HIfN
Hl>-NOH
•0 ONIfNH
IfNIfNH
HIfNH H
ON CM 3
o CM 8 00ON
OtfNH
CMNOH $ ONH
KN
HIfNCOH 6^ H
IfNNOH
HNOH
IfNHKNtfNH
3>H
-4 3
o O j* COCM
ff\IfN §
NO
21
HIfNH
IfN
8CM
NOONH
CMONH 8 oS 08
NO
HCM
HCO
2 NOH 5 NO 3
o O o -4 KNCM 3 H
NO 8 & 00
HNOIfNH
IfN
H ONHl>-ONH
KN
HIfNCOH H H P fi 3
o O o o CM CO -4KN
oIfN v5 S 00
ONJ- 5H NOH
00
H-5f
2?
COOnH
-4
3OONH 8 CO
otfN
CM-4NOCO O CM-4NOCO O CM-4NOC0 O W 4 vO 00 Q CM O OHHHHHCMCMCMCMCMrOvrf\»fNrfNK\^-4CMCf3
O
OH
8
OKN
s
8-oP UN
m
«a 50)
3 o& t-(Q
1 8-3
$
a &d•H
4* 8s,,^ H>
Pi
P OS dCD(0
•& o
9do•H QP KN
5H
>
sH
OUNH
5H
Of-H
<8H
8(31
o o o o o o $ .* t— NOCM 00 OnH H £ 3
O O O O O ON no KN UN Q O CVJ J*kn h on o\ ch oj r-3-
rl rl H r-i rl
o o o o oO CO Q UN Q UN NO CM -*
H CM H H H H
O O O O t-
O O O
t— on -* on .* o\ o o cmON ON ON CD CO f~H C— -*
r-i r-i r4 r4 r4 rl
*o>-*ON-*ONJtooc\!NO On ON CO CO C— r*- r-\ C- .*
rl rl r-\ rl rA r-i rl
OOO0\4O\40\40\C0CQfl|ON ON 00 CO t- C— n£) O NO .*HHHHHHHH
O OCM
un o ir\ o m O ia co oo cvi
ON ON CO op C-r^^O O vo 4H H H H
OOnOHnOHnOHnO.HnOnOCM0\ ON CO CO h- t^- NO nQ O NO ifHHi-lrHr-IHHr-IH
O 0> -* ON .* On-* ON -* ON -* NO CMON ON CO CO C- t— nD NO UN O NO -*rtr4r-\r-\r-ir4r4rtr-ir-\
8t- W N CM t— CM t—CM C*- -* NOON ON CO CO t>- h- vO vO UN O NQ
CM
CM i-i r-i H r4 r-\ rA r4 r-\
ON NO H NO H NO H NO rl \0 4 \fl CMON ON ON CO CO t»- h- NO NO UN o >3 4H H H H H H H H
ON NO H no H no rl \flON ON ON CO CO t— C— NO NO UN OH H H H
H NO -* NO CVJ
H H H H H
8t—CM£-CMC^<\lC— 01 f- 4 vO WON ON 00 CO f- t— VO vO IA O vO 4<Ui-4r4r-*rtrlrlr4r4r-ir4
UN ON -* ON -* ON-* ON-* ON -* NOH ON ON CO CO t"— C— NO NO UN O NOi-lHHHHHHHHHHt— CO f— CM C— CMKN CM On ON CO coA r-\ r4 r-l r4
2
t- CM t— CM NO NO CMt— h- NO NO O NO -*r-i r4 r-i rt r4
OCMHnOHnOHnOHnOCQCOCMUNCO ON ON CO CO t^t— nOOnO-*H H
O tfN KNoH CM H H
rl r-i rl rH
UNO UNO UN O CO CO CVJ
ON ON CO CO t— t>- O NO -*r4 r-i r-i rl
O O NOKNQUNQUNQUNOJrCMOONONCOCOt^rHHCMHHHHHr-lO CVJ
t- -*
O O KN C— O -* UNQlfNO CM CM .*NO KN ON ON ON CO CO r-lt-3-H H H H H H i-l
CM-*N000OCM-*NOC0 QOQHHHHHnOCMOr-J CVJ
97
O
OH
O<M
OKN
s
9-
+> UNU
•8
•H 5os2, Oo* t-(0
b& 8Q
a &a•H
P 8«•_!-• H>
Pi
V ON ato
CO
<0
& oa
4o•H oP KNfc HB>
sH
OITSH
3H
Ot—H
<8H
•rH -H SH
OOOOOOCVJ-4vOCOKN ^ 0\ (M lf\
SI
Q C— vO vO CO
H H H
O O O O ON CO C— VO UN CO 1^ ON -* 4 vfl
H IA Q\ K\ f- ON ON 00 CVJ t- -*r4 r-t r-i H r-i A
OOOUN^KNCVJQUNHvOQJCVJCVJun b' un c5 0\ 0\ CO COH «H CVJ H H
O O O s aUN t— KN CO -4t— ON ON 00 COr-\ r4 r4 rt r4
ON IT\ VO CMC— t- H t—»H H H
OOHOQvOHC^WCOKNOnOOCVJO ON ON CO CO C— S- vO H t— ifKN21 CVJ S 3
O O -4 CO KN4-COKNON-4QUNHV0CVJOOCVJCVJ ON ON CO CO CO t— t>- vO vO H t>- .4rHHr-IHr-IHHHH.-lH
O O t- KN CO -4 ON UN O VOONONCOCO t>- t— t— vOrl r-i r-\ r4
H C— CO CO CVJ
\fl IfV O ^3 4r4 r4 rl
O VO KN^> KN fl\ 4 O IA H VO CVJ f- KN VO VOO OV CO CO CO t^ h VO vO UN UN O VO
CVJ
rl H r4 r4 H H
0\COQVOHC--CV|CQon On Ov co co t— o- vQ vo u> u>
r4 r* H r4 r4 r4
K\ ON 4 O -4- VQ CVJ
KN ITv O VO 4H H H
t*- "^ 93 -^t ON UN O vO H h- OJ CO 4 vO CUONONCOCO t—O-t— vOvO WN UN 4 O vO 4rlrtHr4Hr4r-it-ir-iHr-\rtr4
t— KN CO -4 ON UN O VO H t- CVJ CO .4 vO CVJ
ON ON CO CO t*- t— C*r VO VO UN UN 4F Q VO JTH H H H
f- *> 92 r3t ON UN O VQ H t- CVJ CO -4 VO CVJ
ON ON CQ CO ^ t- t- vO ^» UN UN 4 Q vO -4
H H
HHHHHHHHHH H H
^UNQVQHt--CVJCQ KN Ov 4 O -4 VO CVJ
ON ON ON CO CO t>-C— VOVO UN UN UN O vO -4H H H H H H H
.4 CO ^4
21
CO ^4 ONUNOvO H t>- KN CO -4 vO VQONONCOCOCO C— t— VQ VO UN UN O VOCVJ
r4 r-4 r-l rl r-l r4 r* ri
WrlC04t-U\OvOHNWCOCOCOWvO UN ON Ov CO CO CO C— O-vOvO UN O vO -4Hr*rlHr4r-\r-1rtr-i H rH
^^^&&&co, c3£Pv$SS£3r-\ t-\ r-i r-{
CVJ CO ^4 ON ON ON CO CO t—r-i t-i r-\ r4 i-i r-{
CVJ H Q Q\ CO vO CVJ
UN O -4" ON ONH H H Hf- KN OV 4 O
On co co i>- t— i>-r-\ r4 rt r* r-i r-* a
O CVJ
t- -4
OKNCVJHQ0NC0-40NUNOvOv0CVlC\|CVJVO O -4 t- ON ON CO 00 CO t- H t— J*
OCvl4vOCOOW4^0JtKN(MWKN vO ON KN VO ON ON ON CO CO CVJ t—r4r4Hr4HHHH
CVJ-4vOOOOCVJ-4vOCOOCVJ-4<2PQHHHHHCvJCVJCVJvOCVJOiH CVJ
98
IS
X!Ua
&CO
u
&CO
a
Pi
oCVJ
oKN
OIfN
$
Ot-
<8
4) o& da094)
b oPi
21
1o•H oP KNu H4)
>
oIfNH
5
OH
<8H
I
o o o o o o $
O O O O O ir\ OJ
O O O O O ON ONo\ t-
o o o o .4 NO NO
H H
o o o o g> itn iaON t~ IfN
r4 rl r-i
o o o o 00 -4 J*t— ir\ t<\H r-4 H
O O O ON H t- t-ON t- -4 CVJH H H H
O O NO CVJ -4 Q OON ON n£> -4 CVJ
r-i rt H r-i rl
t- C— KN ON H t— t—On On On CO NO rf\ Hrl r-{ t-\ r-\ r4 rl rA
g> CO -4 Q CVJ CO COON On On On no K"N HH H H H H H H
8 ON itn H KN 0\ ONON ON ON NO KN H
CVJ H H H H H H
O CO ON IfN t^- H"N KNH ON ON NO J* CVJH H H H H H
O O CO Q CVJ CO COON O t— -4 CVJ
CVJ H H H
O O O IfN O NO NO4 CO ITN lAH H H
OOOOCO NO "3-
H >-i H
O O O O Q NO NOO t— IfNCVJ H H
O O O O IfN CO COOn CO NO
O O o o ir\ Q oCVJ H
H CVJ -4 NO O CVJ CVJ
(M lA 4
IO -P4> P^HICMCO 11
9 a•H CO
u
sia cvj
•H
O co
KNO4) XJ4) P
<S wH £
S§P H3aPi O
g£-h 4)
4) CO
^SCO fc,
o a4> tr
a|<D -HO CO•H A
S> CVJ
OHPQ
4)H3
3x«oa•H
4>
CO
p\
co
aa
Pi
4>
&COca
4)
P\
o•H+»
I
o
orf\
oIfN
S
8
a
o
OIfNH
5
OH
O O O O O t— IfNON CO
O O O O O O COON t—H H
O O O O O -4 CVJ
CVJ CO t—H H
O O O O t— NO -4-
CO t— NOH H H
199
OOO O J- O COON t— IfNr-i t-i r-i
CMo o o o go -4CO NO IfNr-i r4 r-{
° ° ° ON £ £ £
O O O IfN On IfN K\ON f- UN 4H rH H H
O O CO O -4 O COON On t— IfN KNr^ H H r4 r4
§IfN
3 00HHH -4
IfNhf\
H
8CVJ
COONH 8 H
^4
H
8CVJ
CO
3-4-
3OH 5H
-4
8CVJ
00ONH
-4
3 8HOH 5 -4
rf\H
IfNCO 8
CVJ
NOONH 8H
CVJ
H 3HNOKNH
O CVJ IfN HIfNH
ONKNH
O O3
IfNONH
lf\IfNH H
O 8 00CVJ H
ONOH 3
OCO ON NOH H
-4IfNH
O O NO
3 PiH 5H
H CU -4 CO NOH COCVJ
J-(A
100
Pi
r-l t— WOir>ir»Wr)r|OOt-4rr\KNK\cur-ii-iHHr-iHH
II
Pi
4t-Or(Wir\HWWWWHOK\ KN (Ty rl
(M
oII
Pi
5
4)
i
8
Pi
Pi
CVJ
Pi
ri tf\ 4 4 4 K\ iT\ W H
-* vO UN i/N O i-H HOJ CVJ H rI I I
00 O UN 00 OCVJ K"N CVJ
I I I I
I I I I
cvj
H O O
r4 r4 r-4 H t-i
Piirv h o\ o ONin^t-sF-*^*-* cm h
(0
Pi
COOOHvOON-stHQQQCOH vo
Pi
o
05
HII
BO
CO
Pi
8
Pi
3
4)4)
8
aJ
Pi
Pi
Pi
Pi
On-* l(V On 4 CO in -* -* -tf -* CVJ HH CVJ CVJ r-i H
OOOOOOvO-* cvj ir\ m ir\ OCO CVJ CVJ H H
lfVir»C\IWrlif\4\0 ON O ON if\ KVN KV KV W rl I I I I H I I I
I I
4 t- COH H HI I I
OJ ON CVJ o H H H H
•3
JM
ao•H+>•H
aOo
HPi
I•HEH
UN (M t- Ht—COOCQt-CVJONOCOKNLAVOCOCOCOCOvO ITNITNJ* tC\ r-i
CVJ -* CO CVJ \Q O -* CO CVJ Q O QHCVJCVJCVJKN-3-CVJOCVJ
101
o
oII
pT
pf
II
0)
0)
&
oCMH
a)
Pi
aS
Pi
Pi
Pi
CMPi
ovo ov£>oo t— c— r-if\ |f\ |f\ H t- VO vo -* CVJ
r^ 00 CM O O H|T> lf\ 1^ rl
i-H H rH H O O
OOOOlTNl/NONCOC— ^ A 4" WH ir\ir\ir\i/\ir\irvK\CM
-4-mCMHt*-CMOOOOOOOCM CM H r-i H
I I I
CO CM CO OS OCM K> CM I
I I I
r-i rl r-{ rA
HPi
H-*K>t— HONHOONCOt— \DO\irvl^-0\OONK>KAK>kCMCMCMH
CO
HII
oO
Pi
aJ
Pi
gNQHJ*-^-vOK>CMCMCMHvOCM^-4-tOvCMf-IHHr-lHH
pUNCOCM-*t^^-CMCMCMCMHOCM CM CM CM H
•HCO
Pi
8H
Pi
3
Pi
Pi
CMPi
O O O OO OHC-rf\ovD O Orl Ol J UN 4 (f\ N
\0 -* CM CO K\W K\ l^ H I
I I I I
t— CM -* vO MD lA CMI I I I I
tf\ CO H t- O -4- r-iH H CM H H I I
i I I I I
rl H rl rl r4
ao•H+>
oo
HPi
vO A 0\ lf\ IA CO |T\ W ON H 0\ t^ ONKN tf\ vfl CO 0\ ON CO VO A K\ W H
HCM-r»-COCMvOO-*COCMQOHHW«Wlf\4Cj
102
int-
OII
«T
•P
II
CD
HII
CO
O
•H10
Pi
§
P-
CO
§
iOo
•3
ft)
Pi
Pi
Pi
Pi
Pi
Pi
A
(1>
8
Pi
Pi
Pi
A**Pa
CVJ
Pi
HPi
EH
& IfV t~- rT\ K\ CVJ |T\ K\ IA K\ KNKN (VI rl
ifNQK^OOJOOOOOOOI'N J K\ H I
OOOOOCVIt^vD^tCVJONlACVJ l>- h- t— t- t- vO .*
rAK"\C— vOOJtOvOOOOOOCVJ CVJ I H CVJ
I I
O i/N H H OKN K"\ K"\ HI I I I
t-i H rt rl
ia t- Ov rj CO CvlHr-jrHHH tf\
OnQH ir\-*j--*j*cviJ* K"V H H
O^DOirvtrvvOCVJOOOOOHCVJ CVJ KN CVJ H
o o o o o o c- H IA O t— -4" CO-si- VO c— VD 3- CVJ
\O-d-H-3-v0O\aOOHCVJCVJHW KV |f\ H H H H IIII I I I
rH CVJ CVJ CVJ H I I
I I I I I
H H
t-lf\WWWvOlf\lArlWHvDKVrA IT\ t>- ON Q O 00 ir\CVJHr-j
HWdtDW\004(OWQOQrlHWWWiAJWOrH CVJ
103
8 Pi
OlOt-CVICvltOHHHr-IHr-IHtOfOCVJHiiiililli
Pi
^>Pi
3
Pi
Pi
M3HlOCM-*CVJHHHHHHHK\ 4 A H I • I • » • I I I
o o o o -* go o-*9QOOOvQ.3tOOOw co CA co co « co ir\ to
n
&
HPi
CMPi
iOCVJvjDC— g\^CUC\l<vj CVJ CM CVI CVI
l i
H f- -* to O HtO to tO HIII! H H H H H H
Pi
tO COtO f- £ ^ & t-CVJCVJCVJCVJCVJCVJCVJ
oCO
(0
Pi
Pi
CVJ
cvi -3F -d- cvj h i i i i i i i
Ot~-CVJt— t— VOHCVJCVJCVJCVJCVJCVJW Cvl IA W H lllllll
03
Pi
8H
o
4)
o o o o o o
S3
Pi
Pi
Pi
CVJ
Pi
5J ^ VO -£ O O OCVJ IA CO <D CO lO toK\
VO -3" O 0\-=f 000\IT»lf\IAlAI/\OJ tO tO I H fO tO HI I I
vOrlt-lAht-HHWWWWWrl W Ol W H I I
I I I I I
ao
3oo
C~- lO lO t— O -* (J\ l>- 4 4 J 4 4H rOir\C"-OsHHCO-4-
rlW4(DWV004COWQOHHCVJCVJCVJtO-*CVJ
10^
oII
ft
>
sII
09
HII
wO
ft
8H
0,
ao•HP
oo
o
4)
IS
o3
CD
ft
a)
ft
ft
ft
CMft
ft
.dO
<D
8H
ft
05
ft
ft
ft
cvi
ft
Hft
CVI-*rr\CMHHr-|r-ir-|Hi-IH
* CMO-4--4CMCMCMCMCMCMCMCM
OOC50lT\0\lOCMHQQO\-4-H KN .4 -4 tOv KN
vO tA CM H O CO (OkCM KN CM I HI I I
H H O H H OII II
OCVlK%OCVIOHHHHrHHHCM CM CM .H I
I I I I
t— t— -3t Q Q\ 0\ Q\ \Q U\ -4 CM HMD*~1
v£> CO CM CO -4" h- -4" H 0\ C\ CO CO CO H t—CMrfNJ-rrvfrvCMCMCMHHHHrHr-i
IT\ O lA ON 41 O CO ITS IT\ -* 4- Jt 4 KN WH CM CM H H H
OOOOOOKNC— r| ^\ ON A A IA Or-| H H CM CM H H
-4- -4 ir\vQC0 ON vO .4 -4- vO t— CO OvO KNCM H*N K"S CM H I I I I I I IHI I
I I I I I I
O CM IT\ H t— -4 CMH <-{ H H I I I
i I I I
O O H H O O
Ot—HC— .4;vQKNt— H-4- t— 0-4- OS HKN-4-MD h- CO CO CO t— t- \0 UN IT\ -4" CM CM
HCM-4-COCMVOO-4-COCMVOQQOQr-lHCMCMCMeAK\3-<5CMOH CM
105
o
oII
rf?
II
Pi
P.
s 4^
4 **5*
Pi
3
& , ,
a •P
""nNo p<04H
/^N•p>—
'
CVJ
P*
p<
CO
UJ
o
a
Ph
§
oo
.ao
4)4>
Q
s
Pi
Pi
Pi
pT
cvj
Pi
Pi
O H IT\ O ONKN -* K> CVI
t— t^vOvOvO^DvO its
C— K> CVJ -* ros O OOl K\ ^ H H H H O
OOOOCOCOCOCvJHO\OOr-HcviJtv£)vo ir» ir\ ir\ ir\
\ocvjpifNi— cvjitnooooooCVI KN 0J r-l r-i
I I I
pi^ir\CV|K%OHrHCVJ CU CVJ H I
H HI I I I
4 ov tr -* ff\ ^ \D CO t-C-MD UN H4\O00O0\t-4WWWWWW
\00\WO\ONftlt-iriWr|0\OCOWlf\4AWWHHrirJ r-l
lAO^HUNiHr-^KNCVICMWWH CVI CVJ CVI H r-l
OOOOOOlTNKJCVJOOONO-^
-4- l/\^J- IA rl H-*MD-* CVI H -4 -4Cvl KN HN C« rl I
I I I I I
O CM NO -* O t— rAr4 A r-i rl r-i
I I I I I
I I I
I I I
O H H O O
tCv-4-VOCO OS ON ON f- VO lTN-4- IO CVI
H CVJ -* CO WV004COWVOQQ
106
u>
on
p."
>
CM
0>
Pi
P.
Pi
P«
OrHlfNt— irNrfNCMCMCMHrHCMCMIA ^ IO H
GOifNrrv^fHHHOHOHOOCM KN KN H II I I
OOOOOvOCMOMDirNKNCMrfNH rO vO 00 t- C— b- t— vO
\OW00Oif\\0f-OCM KN i-l H OJ H
I I I
Ht/Nt— K> rf\ O H HCM CM CM H I
I I I I
O H O O
rA rl r4
HPi
3" C— CT\ O O vO r^\HHH H
(0
Pi
^D ON CM CO t—W if\ 4 ir\ W t-Ol^K\«lAK\«WW
as
O aJ
Pi
l/\r-|ir\KN.VsOi-|vOfOOOHHHOOrH CM CM CM H rH III
Pi
HII
£
Pi
a
a>
ft)
8
OOOOOOCOQCM-^vOCOrcvKNPi
Pi
CMPi
O CM -4 VO CO rr\K\.0\CM tr\ J* irvvOvO^t CM
-d-if\^J-rC\VOCO-4-vO^*OirNOHHHN K\ lA W I t-i r-\ r-{ r-i IIII I I I
OrACOr-CMONi/NCMHHCMCMCMHOH H H H rH l I I
I I I I I
o
1oo
bO
O (J\ IA IA M) CM-4-Q-4-M? ON CM VO tf\ -*H rA-*vOOOO\0 0\COvO-*CMrH
Pi H
1 HCM-*COCM.vOO-=tCOCMvOQQOQ•H HHCMCMCMrr\rr\^vOCMO
107
8
ft
>
•p<M
OQ
o
•H(0
ft
HIt
ft
ao•H+>
oo
3
(0
ft
ft
ft
Pi
ft
HPi
3
H
ft
ft
KMft
CVJ
ft
Pr-
ft
HQK^ir\CMCMCMCMCMK\CMCMCMK\ 4 ^ rl I I I I I I I I
C044KNOKNlCiWCyWWW«W lA ffN H I I I I I I I I
o o o o ss^gcscscscgte
VO H VO -4 ON H ON -4" Jf -4" r<A -4" -4"CM KN H H CM CMI I I
d IA ON IA 4 O <M HCVJ CVJ H I
I I I I
H CVJ H H
ir\ H-4" t- o\ H q vo cm
K\ N"N CM CVJ KN KV
v£> CT\tOC~"-r»: H K"\ CM l/\\£> IA IA IA lA CICVJ KN -4- tc\ CVJ
ir\ H t— ir\ o\ H ir\H CVJ CM CVJ H r-i
I I I I I I I I
O K"N KN to to H HI I I I I I
r-l CM-* IP* f- CO C- -4 K\
-4 lf\ -4 O H t— Jt lA (A CO H <D 00 4 K"\CM^rOvCM lHCMCMCMHr-1I I I I
O^-OOMDHvOCMCMHCMCMCMrHOI I I I I I
HOC— QK\OQQO\C— mifNlfNrCNCMK\ U"\ v£> CTvO H O CO ia ia H
f*i rn rH
HCM-4-COCVJVOO-400CMVOQOOQHHCMCMCMKAKN^VOCMO
108
cvj
oI)
Pi
>
+>
.dO
i
oCVJH
Pi
<d
Pi
Pi
Pi
cvj
Pi
4 |T\ CO H ^ IA KVH CVJ CVJ H H H
ITv id O 4 KV CI iHNA -4 K"\ rH
O O O O O vO CVJ
CVJ -4
CO CVJ OV (Cv KV 1A HII H I
3 ir\
KA CO f- r-\ H rH_4 -3 CM r-i I
I I I I
CVJ
CVJ H
Pi
KA CVJ H CJ VQ fA CVJ-4" KO ON O O t— \T\H H
CO CO
to
CO
p.HWKN^lAWWCvlWOJWrlrlrf
to
oPi
4CJOKV^OO^irvif\lA4lf\WWCVJ K\ K\ CVJ H H
•H(0
Pi
8
P-.
o
<0<D
rH
Pi
Pi
CMPi
OOOOOOHONt— -4-^tO\tAO\H CVJ CV) H H
O 1A K\ O 4 00 4 ifNCO H rl CO \D 4CVJ rH I I I I rH rH I I IH H CVJ
I i I I I
ITN CVJ t— CO O -4"
CVJ KA CVJ H H I
I I I I I
I I
H H H H CVJ O
ao•H+>•H
aoo
ftH
-4 ON CVJ lANQ^f-CO^t-4 HHka-4- t>- co oo dSco t—vo ir\ ir\ 3- eocvj
CVJ-4MDCOOCVJ.-4-VOCOOQOQH CVJ vO CVJ OH H HCVJ
On K\ K\CM H CVJ O
I I
Pi
Pi
O O C- VO K\ -sfr CMin m kn H • i i
CMI
0>Pi
O O O O O ON -sfl/N On
-s|-
ll
Pi
CVJ
Pi
i/n On O t— ITN vOH H K\ H
CVJ Ql/N <oI
on t— CVJ CVJ CVJ
KN. H IIIICVJ
HPi
O O H ITN H O- CVJ
IA I- O H CVJ -st-3 21
cvj
cvi
ii
oCO
Pi
Pi
vo O m cm inH CVI CM CM Hl/NCV|mt-CMCMi<NCMCMCMr-lrHI I I I I I I I
CM -=! J* tr\ CM r-l I I I I I I I
(0
Pi
.£3
O
cu
0)
&
8
Pi
Pi
OOOOOOK>ifNt— HHOOn-s*rn vo On On t— -st kn.
COCO t>- H W 4 K\U"NKNK>KN.K\CM HIII H CM KN. CM rH
PiCM
Pi
CMvOVQt-t— CM-sJ-HCMCMCMCMHHm, .3 JF KN. CM H I
I I I I I I
aoP H
Pi
SON t— H OVO IA OV U\ |T» KN KN H HmOOOHr-JHI>-»n,
OO
HCM-st-NOOOOCM-shvOCO OQOH H H H H CM vO21
109
110
(M
Oii
P«<0 C— vQ-4-CMCMCMHHHH
Pi
> P4
l^OKNt-iOOIlAAWWOIWHVO t- ITt W
PII
•8
I
Pi
P4
CMPi
H \D CO C— t— C— -3" K\ <-\
OtfNCOrHl/NvOHHOOCVJHOi/\ tr\ w cvj ii iii i i
Hl/NrHQOKNr^fAK^VKACVJHHlf\ lf\ -* CM
I I I I
Pi0\ -=* CO O\C0 H CO CO t- t- IT\ lA CMH H H H r-l
(0
HII
CO
Pi
.4; ONO\lTkVO U*\ CM O 4 vOCO t— t/N -4- tT\ k> KN rr\ CM H
Oa5
Pi
irvOOvO r-OvD^D u\ l/\ if\4 K\ N H H
•H(0
Pi
0) Pi
Pi
r-l .3 JF K\ H
t- C- lf\ VO ON H vO CM -4" CVJCOI H H H IvO v£ ir\ CM I I
I I l I I l i
CM
II
PiCM
Pi
O On CM K> j4 H-=f K\rr\HHi<n.
I l l I
I l
ao•HP
Pi
8|T» rl \Q t~--=f lf\ CO -4/ K\ HKAIAVOVO IA |f\ 0\ <B vfl 4H H H H H H
OO
M 8 HCM-4-vOOJ-COCMvOOQPQ•H HHHCMCMKNVOCMOH H CM
in
o
OII
ft
ft
> ft
0000-*-*ir\OurNQO-4-co
+> ft
O -^ N^ O CO oIT\ IA W H K>» H
I I I
o o o o o cu
<Mft
OJ CO VQir\ i/\ .5
i i
iHK^t^N^tAfOCVJHH
ft H H OJ H
HM
CD
ft
CO^f-HHQCOl— VOJ-
Oa5
ft5 0\W<04lAWrlHWH
J- -4- CVJ H
o
ft
OOOOOvDcppHCVJCOH -=F CO ON vD t^
Wft 8 ft
CO vO H K\ KN O O N"\ CO t— -*M3 <0 lf\ H H OJ H I I I I
I I I I
II
CV1
ft
CVJ-d-CO ONCO O K\ -* -* OJl^ |0> CM H I
I I I I
o
Hft
CO On O 4 H -* if\»r\tf\-4-CO v£» HH H H H r-l
lA l^ f- 05 ^ HvO-4" KNCVJ
O
S HOJ-*vOO-*COCVJMD
112
un
oft
ft
ft
> ft
ptl
o
&oCMH
ft
ft
cmft
o o O O 00 CM
21
o CM NO O H -*•
UN1
UN1
H CM un H
O KNUN -4"
H Ht— O H l>- vOlA lA 0\ IA K>
H H H O O O
4Hr|C0W444lA va 1A (M I
I I I I
KN ^* CM
Hft
UN CM KN 00 Q UN UNJ" ON H ON NO H HH H OJ H 3 21
KN O NO
CO
HII
a
o
ft
ft
ON UN NO -4 CO UN H CM CM -3- KNt— CO C- 13- H
3 CMt— HKNHCMKNKNHHUN 4 lA H I I I I t
ft
H
ft
KNft
OOOOO-*CMOCQHv0CM C- CM -3F ON UNH H
s UN t— O -* CM t- t*- O CM HvO -* KN -4" CM II
I I I
ftCM
ft
4 © 4 UN rj rlKN KN KN CM r-i I
I I I I I
iA 4 4 CJ W
aO
dOo
p
ft
H
UN ^ C^ KN H OO -* ND ON O t-H H H CM HCM CM O O ONO KN H H
CM-d-NOO-4-aOCMNOOQOQHHrHCMCMKNNOCMOH CM
115
8 <n
P«
CVI -* K\004vO\O\OvOvO tc\ r-i H\D C— ifN OJ I I I • I I I I I
> Pi
Ov t— C— vO vO vQ vO K^HI i i i I i I I
PIt
.ao
a>
&o21
p<
ft
CMPi
O O O OSI
O 4 \0 co Q -2 ^> Qir\vo ir\ -4- -3- O vo iH H H H H H
& H Oi i
OsiTNCOOOOO-d-CVJ VO H H H H H CVJ CVJ
• I I I
CVJ H H
H O COO UA OS OJ O K>rH H H CVJ CVJ
-* us t- t- t- t— CVJCVJiA
10
to
Pi
0\ -4- -3 vO ifN H -* CVJ CVJ vo rovC— CO C— KN r-l H iH H I Itill
05
oPi
C\ UV O 4 KNH t- CO CO J- CVJ4 IT» lf\ K\ rl I I I I I I
•H(0
P«
J3O
4)0)
&
8
p^
Pi
OOOOOCVJMDOQQ^vO0\ -* -5 C> vOK\
rt ri H
s rr\ cvj if\ irv 4 K*\ H H On iaMD-* H u~\ vO -=* CVJ <M
I I l
toCVJ
Pi
vo cvj ov W vo tr\ ^t ir\tr\K\cvjKV 4 lO KV H I
I I I I I
§
ioo
HP«
C—0\vO t- 0\ 0\ O 4 4 vO K>O^tC— OHC-ONHHH H H CVJ CVJ H
HCVI-4-vOO^-COCVJvOOQOQHHHWWKVM5CJOH CVJ
U.k
CVJ
OII
03
Pi
K>OOOOOCVJOCVJr-|00-40\
rif
o
0)
4)
Pi
Pi
H H IT\ t— HCO CO K"\ I
«> W lA CM KA K>I I I I I I
o o o o *-\ <a on f- in iac— t— c— t—
r- o-4 rA
o21
p«
CVJ
Pi
H 0\ 4 W 1A (MCVJ CVJ I CVJ H I
• I
H H H H H rHI I I I I I
lA f- t- CV 4 4 4 10-4 KNH HCO h- CVJ i
i i i
Oo
CO irv f- irv ov 4 C\J CT\ \Q n^\0 itnH UA ON O -4 OV ON O) CO 00 lA (Ar-l H H CVJ H
CO
cvi
ii
CO
o
p.
03
Pi
vocooJOJOKVQcot"- t— ir\ ^l/\ vO t>-\0 iTi 4 4 KNN^rTNCVJ H
CVJ Q t— CVJ ON Hj4 rT\ H r-|
0\ vO C— C— .4 tr\
•HmPi
8cvj
P-.
H
P.
Pi
Pi
OOOOOOl^HC— i/NCOCOH lA 4 4 W rt
ir\ w \p4 in 4 wi i
0\ -4 ir\CO CVI 00 CO K"\ 00I I H H H H iII fill
t—rTNVO O tO H H -4 |C\ IfV lA4 lA lA CVI I I
I I I I
C3
O
Pi
\Dt*-irNCOCVJHcQirvQC*-u"\HC0r(UVV0f-f-3a|O0\^4
r)OJ4\OCOON4\OCOO21
115
HU
ft
C— UN CVJ IA vO 4 4- 4 4 4 H HKN -4 NN I I I I I I I I
ft
> ft
H KN ON H H. 0\ 0\ 5\ 0\ ON WON ON KN I H I I I I I I I
II
V
8
»
ft
ft
OOOOt-^OH^O rl vO 4 vOON t— t- vO \fl UN O vO
i~n f™i f"n f-1 *~i f
-1
o to 4 irNQvOvO^OvOvO CVJ HH i CVJ vO -4-
£I HO CVI t— l/N IA IA IA l/N UN W HO -4 I
I
I
ft
O H UNt— CVJ KNH CVJ CVI a
UN IT\ UN IA UN rH O
0)
cvi
it
uCO
CO
ft
ft
vOKN^OUN^fCVJOOOOCOKNHUN \Q vO KN H I H H I I I I
Hv0HCVJ-4--4vOOOCOf- HOvO t— t— -4 CM I I I I i i
CO
ft
CVJ
II
E
P-4
XIo
o«>
&
8H
ft
KNft
CVJ
ft
OOOOOOQ-^HvO-^-vOvO CVJ vO ifN O vOH H
tO H-4 -4
I I
ONVO UV \fl 4 O lA «N UN OlCVJ H -4- <5 ITN KN H H
00 4 CO OMD t— KN VO t— \D CVI
UN t— VO UN CVJ •
I I I I I
ao
aoo
ft
ft
•H
SO-4 t— O H^OCQ KN O ON KN CVJ0\iACOHWQl4vOHH H CVJ CVJ CO H
HW4vOCOOW4^00 QOH H H H H VO CVJ
116
CVJ
oii
ft
COIT\ C— VO
O 4 IA 4 r|H <-* H
ft
43O
ft
ft
00 Ht—
O0 -41
HI
0J1
O O o o\H cvj
+>
H<VJ
&oCJ
t<Nft
C— H CO4 K> (VI
I f
CVJ KN
CVJ
on cvj t— ir\ \r\ irsLT\ VO -4I I I
ft
cu -4 ir\ h t>- o-O -4 C— MD \D l/\r4 r-i rA r-\
mPt
Pt
to
$-i
HII
<D
ft
Pi
t>- ITN tf\ U\ Q\ GCO -* CVJ H
^53 CO 1A H
O
OJ
Pi
to
05
5
mft
CM
n
43
Pi
Pi
CMft
o o o o o
H -4 C— CVJ -* t-tr\ v£> vO H I
I I I I
0*^0 KN ON KN i/\lA KN K\ I
I I I
3
H
ao
oo
+>
ft
s
H KN COl>- o 2
l>- CVJ CVJ
UN -* t-H H
CVJ ^t vO O CVJ CVJ
CVI KN ^t
117
8
I
>ft
x:ft
+»
4™
o
ft
ft
ao
C!
oo
o
0)
o3
V
8
ft
ft
ft
ft
CMft
ft
ft
ft
ft
1
p s l/NI
On ON1
OnC— & CO
\o H1
O1
OHI
O O o 5 H00
3
CVJ
vOI
s1
l/NH
I
vOITNH
IA tf\ 41
vO t— VOI I I
t— t— t—
t- UN HO<-{ H r-i
1/N ON U\ H 3
0\ CO 4 ia OITN C- CO H CU
HCVJ
I I
CVJ ITN CO -* ON -3-
-* ITN ITN CVJ I HI
o o o o o t—
~kHH
ft
CVI
ITN•
vOI
•4"
1
CO ON CVI
+>
CMft
KN1
3I
ITNJ"
1
CVJ
1
KN 3
\Q CVJ CVJ Q CVJ CVI
t— h -* o on cvj
i-4 H CVJ H
CVJ -4 NO O CVJ CVJ
CVJ f<\ -*
118
CM
oII
wPi
-4 its h r— co co ^oH K"n i/\ _4 H H H
Pi
P<
N (ft 4\x> co r-
CVJ O O HKN I
4Ph
O O O O CO CO KNN \o ^)
oOJH
CMPi
HPi
vO ON1
OCM
1
KN1
O o H
t—1
KNON
1
<81
-4CM
1
vO \o MD
CMCO
KN.
3 LTv 5 £ P
(0
CM
II
OCO
Pi
Pi
i/n cm (A co -4tA IA VO \fl 4
4 00 4-4 tf\. vO S 2
-4- CM
-4 JO
CO
Pi
8CM
P«
S3O
4)
8
Pi
Pi
CMP<
O O O O O K"\ Q
vO ON ITN H vO ^4CM KN -4 -4 CO H H
I I I I I I I
KN IA ON CO CM VO vO4 ift lf\ A I
I I I I
o
aoo
P<
§
H CM C— -4 l/\ CM QvO 0\ H its vO ON ON
r-| H H
CM 4 CO vO CO 4H CM (O
+60
Ql
o
+ 40
,-T +20
-20
+ 120
119
+ 80
CO
,-* +40
Q.
-40
Time, ms.
20 80
100
y^ph-^=0.25
Ph- -5^=0.50
\\
r_^-
v^ — V-^-=1.00
10040 60
Time, ms.
Figure Bl Anti symmetrical p (t) and Symmetrical p (t) Analogous Loads -withQl 8
R 150 inches, C 1000fps, P = 200 psi, t - t.
,
s di r v
9 = 180 degrees, W = 1 MT, and PyVp SlB noted
120
10
E
o>
E
jsd '(*) d
CM
oII
-C >Q. Q.
Q. Fro
Q.
i
/
V,
E
0)
E
o o(0
o00
(VI + ++
jsd4
(*)ud
oCO
I
A
t
CO TOH"^
ti
Pi
K]|5
+>
•H
09
Jl 8•P CVJ
+>+>
(0
Pi
II
(3vi a
CO
HII
pa4)•HO
U CO
a> ooo •>
09
° o a«0 t/\
Pi
ICVJ
0>
H HII II
Pm
121
ooll
-cl >Q. |CL
Q.
m IQ. 1
v. 1
1
X &
CO
6
E
o
+
oCO
oCD
+
jsdl
{\) d
5•51
it
3c
CO
—. &
IXI
a
B
£
.3
II
+>
to
eufl
E
E
o1-
O oCO
CM
+ + +
isd4
(j)Ud
v.^ CO
«> o•H H
oco
1
fM
10
o
to
us
Pi1
^cjas p7
122
+ 80
+ 40
Q.
Q?
•40
1 1 1
VI
ph
/ Pv
/ p h
20 40 60
Time , ms.
80 100
+ 80
+ 40V)
toQ.
-40
1 1 \ ^=0/ p/ v
25
00
20 40 60
Time, ms.
80 100
Figure Bk Antisymmetrical p (t) and Symmetrical p (t) Analogous Loadsa s
with R = 150 inches, C = 2000 fps, P = 200 psi, t = t ,
Q = 180 degrees, W = 1 MT, and PjVp = as noted
123
E
CD
E
jsd4
(|) d
3
Pi
-a
U+>01
3p
HII
>
CO
<U<D
&
<srH
II
Q>
+>
II
•HCO
Pi
8CM
Ph
ooII
-C >
«
.
<t—*-—
_
€>*^.
CM
to
E•*
CD
E
o o<* COCM —+ +
oCO
+oCD
I
isd (Ou«
d)•HO•H<H<H0)
oO
I
OCO
s
0)
•H
CO
CVJ
II
wo
COCD
J3oa•H
oIT*HII
a
>
121*
lOCO
1I i
o K"\II
H / J>J Pi
-C >Q. Q
Pi /
r
o o o<fr (0 COCM
+ + +
jsd '(*)ud
E
E
oCO
I
5p
o
3
3
01
I
8
II
II
Q OJ
mCO
oII
x: >a, a.
oao
Oto
o oCM
(0
E
E
K
isd •(4)8d
sp*
mCO
on
x:| >a. |o.
oao
Ot©
o oC\J
10
E
0)
to
CJ
isd '(*) d
I
125
VITA
Lieutenant Commander Worthen Allen Walls, Civil Engineer Corps,
U. S. Navy, vas born near Little Rock, Arkansas, February 17, 1922.
Following his graduation from Stephens High School, Stephens, Arkansas, in
19^*0, he attended Magnolia A and M College, Magnolia, Arkansas, for two
years.
He entered the Naval Air Corps, U. S. Navy, in June of 19^2 and
after flight training vas designated a Naval Aviator and commissioned as an
Ensign, U. S. Naval Reserve, in 19^3 • Wartime assignments vere primarily
with the Pacific Fleet and included duty on the West Coast, the Aleutians,
Hawaii, Saipan, Peleliu, Okinawa, Iwo Jima, and then Japan in the early
days of the occupation.
After being released to inactive duty in 19k6, he entered the
University of Arkansas, Fayetteville, Arkansas, from which he graduated,
with honors, in 19kS with the degree of Bachelor of Science in Civil
Engineering. He was then employed as an engineer by Phillips Petroleum
Company, Okmulgee, Oklahoma, until April, 19**9 •
In April, 19^9> be returned to active duty and was commissioned
as a Lieutenant (junior grade), Civil Engineer Corps, U. S. Navy. His first
tour of duty was as Assistant Public Works Officer and Assistant Officer in
Charge of Construction, Bureau of Yards and Docks Contracts, U. S. Fleet
Activities, Yokosula, Japan. This duty was followed in 1951 by duty in
the Public Works Department, U. S. Naval Air Station, Corpus Christi, Texas,
until 1955.
In May, 1953> he entered the University of Illinois, for graduate
work in Civil Engineering under a program of postgraduate education
126
sponsored by the U. S. Navy and received the degree of Master of Science
in Civil Engineering in 195^.
Thereafter he vas assigned as Assistant Resident Officer in
Charge of Construction, Bureau of Yards and Docks Contracts, Seville,
Spain, for three years in connection with the Spanish Bases Program. From
1957 to 1958 he served as Public Works Officer and Officer in Charge of
Construction, U. S. Naval Ammunition Depot, Concord, California. He
returned to the University of Illinois in 1958 "to continue his advanced
study in Civil Engineering under Navy sponsorship.
He is a member of the American Society of Civil Engineers,
Society of American Military Engineers, and Tau Beta Pi
.