<.

Fritz Engineering LaboratoryReport 223.10

LATERAL LOAD DISTRIBUTION IN MULTI-BEAM BRIDGES

by

Alfred Roesli

Progress Report 10 - Prestressed Concrete Bridge Members

Part of an Investigation Sponsored by:

PENNSYLVANIA DEPARTME~r OF HIGHWAYSBUREAU OF PUBLIC ROADSREINFORCED CONCRETE RESEARCH COUNCILJOHN A. ROEBLING'S SONS CORPORATIONAMERICAN STEEL AND WIRE DIVISION,

. U.S. STEEL CORPORATIONAMERICAN-MARIET.TA COMPANY,

CONCRETE PRODUCTS DIVISION

July 1955Lehigh University

Institute of ResearchBethlehem, Pennsylvania

~ ~. ~-...:'_*.- .... : • ..- - • -'_.. ,',,' ~ ..

(

TABLE OF CONTENTS

Page No.

•

I.

II.

III.

IV.

V.

VI.

ABSTRACT

INTRODUCTION1. General2. Introduction to the Theory of Orthotropic

Plates

DIFFERENTIAL EQUATION FOR THE MULTI=BEAM BRIDGE1. Derivation of Differential Equations2. Experi.mental Determination of the Relation

Between a and f33. Assumption for the Relation Between a and f34. Limiting Cases of Di.fferential Equation5. Boundary Conditions

INTEGRATION OF THE DIFFERENTIAL EQUATION1. General Considerations2. Plate Strip of Infinite Width Under Line

Loading3. Solution for the Bridge of Finite Width4. Explicit Form of Formulas5. Influence Surfaces

LATERAL LOAD DISTRIBUTION1. General2. Definition3. Approximation for the Lateral Load Distribution4. Properties of the Coefficient of Lateral Load

Distribution

NUMERICAL CALCULATIONS1. General2. Accuracy of Results3. Descri.ption of the Tables4. Coefficients of Lateral Load Distribution

COMPARISON OF THEORETICAL AND EXPERIMENTAL RESULTS1. General2. Deflecti.ons3. Coefficients of Lateral Load Distri.bution

1

33

7

1717

19212324

2525

26334647

51515154

57

6060636566

69696971

VII. USE OF THE COEFFICIENTS OF LATERAL LOAD DISTRIBUTION 74

NOTATIONS

LIST OF REFERENCES

110

113

LIST OF TABLES

Page No.\

Table I-Formulas for the plate Strip of InfiniteWidth

Table 11- Formulas for General Case

Table III - Formulas for Articulated Plate

Tables IV through IX

Tables X through XV

Tables XVI through XXI

Tables XXII through XXVII

81

82

83

84

85

86

87

•

•

Table XXVIII - Coefficients of Lateral Load Distribution(a/h= 1.7) 88

Table XXIX - Coefficients of Lateral Load Distribution(a/h = 1.0) 89

Figure

LIST OF FIGlJRES

Page No.

•

Multi~Be&~ Bridge With Lateral Prestress 90Schematic Cross-Sectiol)l of a I\fu.1ti=Beam Bridge 90Corrugated Steel Sheet as Orthotropic Plate 90Gridwork System 91Forces and Moments on Dif,ferential Element 91AS3umed Re,lations Betw'ee,n a a.nd J3 92Deformation Produced by Constant Twisting Mbments 93Equivalent Corner Loading 93Plate Strip of Infi.nite Width Under Line Loading 94Bridge of Finite Width 94Assumption for the Longitudinal Distribution of

a Wheel Load 95Distribution of Load Applied at Center for Q = 0 96Distribution of Load Applied at Center for Q = 0.1 97Distribution of Load Applie·d, at Center for Q = 0.5 98Distribution of Load Applied at Center for

Isotropic Plate Q = 1.0 99Distribution of Load Applied at Edge for Q = 0 100Distribution of Load Applied at Edge for Q = 0.1 101Distribution of Load Applied, at Edge for Q = 0.5 102Distribution of Load Appli.ed at Edge for

Isotropic plate Q = 1.0 103Comparison Between Experimental and Theoretical

Deflections for Load Applied in. Bridge Axis 104Comparison Between. Experimental and Theoretical

Deflections for Load Applied at Edge 105Compari.son Between Experimental and Theoretical

Load Distribution Coefficients for Load atBridge Axis 106

23. Comparison Between Experimental and TheoreticalLoad Distribution Coefficients for Load at Edge 107

24. Determinati.on of Load Carried by Edge Beam 10825. Percent of Wheel Loads Carried by Center - and

Edge Beams in a 27 ft. Wide Bridge of VariableSpan 109

•

..

ACKNOWLEDGEMENTS

This report concerns a part of a research program on

prestressed concrete, carried out at the Fritz Engineering

Laboratory, Lehigh university, Bethlehem, Pennsylvania of

which Professor W.J. Eney is Director.

This research program is guided and sponsored by the

Lehigh Prestressed Concrete Committee, Hr. A.E. Cummings,*

chairman, and is represented by the following organi.zations:

Pennsylvania Department of HighwaysU.S. Bureau of Public RoadsAmerican Steel & Wire Div., U.S. SteelJohn A. Roebling's Sons CorporationReinforced Concrete Research CouncilConcrete Products Company of America

The author is greatly indebted to Dr. C.E. Ekberg, Director

of this research program and Professor in charge of the pre-

sent report. His h~lp and advice, as well as that given by

the entire special committee for this doctoral work with

Dr. F.W. Schutz, Jr. as chairman, is sincerely appreciated.

Deep appreciation is expressed for the help given by

Dr. E. Bareiss who prepared the program for computation of the

numerical values. Tabular values were compiled on a contract

basis by Remington Rand, Inc., New York on their Univac Com-

puter.

The many suggestions and the untiring help received from

the entire staff of Fritz Engineering Labcratory and especially

* Deceased

•

•

from the author's friends, Dr. B. Thllrlimann, who furnished

much technical advice, and Messrs. A.N. Sherbourne and

A. Smislova for help on the editing of this paper, are

sincerely acknowledged. The cooperation of Mr. I. Scott,

who prepared the figures, and Mrs. V. Olanovich, who typed

the original manuscript, is gratefully appreciated •

..

•

•

ABSTRACT

A method is derived to analyze multi-beam bridges and"

especially to determine for what portion of the live load

each beam must be designed.

Present design procedure being rather conservative in

its scope, this analysis will represent a more realistic

solution for the problem and may lead to a more economical

design of such bridges'.

The method is based on the theory of orthotropic plates

and ',on the main assumption that the interaction of the beams'

provided by shear keys and lateral prest~ess exclude any

slip between the beams.

Formulas for deflections, moments and forces are

derived for bridges with various degrees of lateral prestress.

A limiting case was found in the "articulated plate", a

bridge with no lateral bending stiffness, but with a beam

connection which transmits the full shear force.

Numerical values are presented for the most important

loading conditions and for bridges of various sizes • For

design purposes the coefficients of lateral load distri-

bution may be used •

Among other results, it was found, that for two

standard trucks placed side by side on a 27 -f,t. wide bridge,

•

•

-2-

the maximum load carried by a 3-ft. wide beam is 55% of

a wheel load, as compared to 80% recommended by the speci­

fications, and is almost independent of the amount of

lateral prestress •

..

..

~3-

I. INTRODUCTION

1. General:

The design of modex'n structures tends to utilize

.construction materials to an opti.m·...'m while an appropriate

factor of safety is sti.ll me.intained. This requi.res how­

ever an accurate investigation of the stresses in the

structures. In many cases) the ~)i.mplified analysis based

upon the beam theory has to be replaced by a more exact

one, considering the structure as a two or three dimensional

one, such as a plate or a shell. It i.s h:>ped that the

following study is a contribution to this development.

The investigation described herein de,sd.s 'with the

lateral load distribution in multi··'bearn bridges. This

type of bridge is conetructed from precast beams made of

reinforced or prestressed concrete. They are placed side

by side on the abutments and joined together laterally

by steel rods -y;hich mayor may not be prestressed.. In order

to incr.ease the interaction hetvJeem the beams, continuous·

longitudinal shear keys are formed at the joints, e.g., by

dry packed mortar in a recess formed at the sides of each

preca~t unit. Fig. 1 shows an isometric view and Fig. 2

a cross-section of such a multi-be&'1l bri.dge.

.,

•

•

-4-

The problem to be i.nvesti.gated i.(..; the interaction of

the' beams and the determination. of the portion of the load

each beam must carry should the load be applied to one of

the beams. Various degrees of lateral prestress will have

to be considered.

In the present design proce~~re of these bridges a

limited interaction of the beam~, is considered. It is

assumed that a loaded beam c81.rr.'ies 80% of the applied load

and that the remaining portion is distributed among the

adjacent beams. This assumption is based on an inter­

pretation of design section 3.3,lb of the 1949 AASHO

Specification. (1)*

Past experience has shown, th.e,t bridges d.esigned

accordingly, a.re stiffer than expected and that: the in­

dividual beams are unnecessarily heavy. (2) The application

of this bridge system is thus limited to short span lengths.

In order to increa.se the span length and to desi.gn such

bridges more economically, a more exact analysis is

necessaty.

Extensive investigations have been made on similar

bridge systems. Of most interest for the present study

* Numbers refer to List of References .

'..,...

•

i

•

are the ones made on bridge~ formed by tv.10 systems of

parallel beams spaced at equ.al inte1:vals i.n the longi­

tudinal and lateral direction. The beams are rigidly

connected at their points of intersection and support the

bridge deck. This type of constru,ction will hereafter be

referred to as a gridwork and is ~'5ih0wn in Fi.g. 4.

Several rn'ethods, some of 't\yhich a.:re only approximate,

have been developed to analyze gridwork systems. Massonnet

presents in Reference (3) a very helpful survey of these

methods. It appears that thB analY8i.s of a gridwork as

an orthotropic plate is 'very F.":fficient. The gridwork with

its discontinuous elastic properties is replaced by an

equivalent plate, having the same average bending stiff­

nesses in the two directi'...~ns as existi.ng in the gridwork.

Y. Guyon applied this method to investigate the lateral

load distribution of beam and girder prestressed concrete

bridges. (4) To simplify the an81ysis he neglected the

torsional resistance of the beams. Based on this work

Massonnet extended the i.nvestigation and included the

torsional resistance of the beams as well as that of the

bridge deck. (5 )

Both of the above mentioned authors derived general methods

to analyze these structures and p:r:ep.:t.:r.'ed design ta.bles. The

'.

•

+,

-6-

latter were set up for a live load of sinusoidal nature.

They considered this type of loa.ding as a sufficiently

close approximation of any loads encountered in the design

of bridges.

P.B. Morice and G. Little described some laboratory

tests on models of gridwork systems. (6) They concluded from

their experiments that the analysis of gridworks can well

be based on the theory of the orthotropic plate.

A bridge formed by beams with the sides in continuous

contact to each other could be considered as a limiting

case of a gridwork and the methods mentioned above could

be used for its analysis. However, it is shown in this

investigation that the assumptions on whi.ch these methods

are based do not generally hold for multi-beam bridges.,

A method is therefore derived in this work which will

generally hold for the- latter systems, and which is also

based on the theory of orthotropic plates. The theory is

first described and. then modifi.ed for multi-·beam bridges.

The resulting differential equation is solved, considering

the structure as a plate with two opposite edges simply

supported and the other two edges free. Ntlmerical cal­

culations are presented for the most i.mportant loading

conditions, for bridges with different spans and widths, and

•

also for va.rious degrees of l,eteral prestress an.d dif-

ferent sizes of beams. Using the coefficients of the

lateral loa.d distribution the rE.:i;';jult~S% are presented in a

form suitable for the practical design of such bridges.

2. Introduc tion to the-Th.~ory of Orthotrop}.c _Plates:

An orthotropic plate is defined a~ a plete with

different bending stiffnesses D = EI in two orthogonal

directions x and y in the plai.!'.e. of the plate. These may

result either from different moduli of elasticity Ex and

Ey of the material i.n the t,IiTO direetions, or from different

moments of inertia Ixand I y per unit: width of the plate.

An example for the first kind. of Grthotropic plates

is a timber plate. Assuming the X~a:Ki3 par:allel and the

y-axis perpendicular to the grai.n ll the modulus of elasticity

in the y-direction Ey is ~ a.ccording to E. Seydel, (7) only

about 1/10 of that in the x·-direction.

A corrugated steel sheet is; a typical. example of an

orthotropic material of the secon.d kin.d. The modulus of

elasticity is the same in every direction; the material

itself is isotropic. Here the dtfferent bending stiffnesses

are functions of the shape of the shE.~et. The a.ve,rage

moment of inertia with respect to the neutral plane in the

•

=8~

direction perpendicular to the generator is much smaller

than in the direction parallel to the generator. Fig. 3

shows this case.

The basic assumptions in the theory of the orthotropic

plate are identical with those i.n the theory of the homo-

, genous plate, namely:

(1) The thickness of the plate is small compared

with its other dimensions.

(2) The deflections ware small compared with the

thickness of the plate.

(3) The transverse stresses oz:: are small and their

influence on the deformation c.an be neglected.

For a right hand coordinate system (x,y,z) where x

and yare in the plane of thE: plate and parallel to' the

two distinct directions of the orthotropic plate, the" ...:.

differential equation for the deflection w parallel to the

z-direction is given by~ !

(la)

• L

(EI)xand (EI)y represent the two bending "stii:fnesses per

unit width of the plate. 2H is a coefficient containing

two parts. The first part is the twisting resistance of the

plate and a,second, smaller part is a function of the two

bending stiffnesses modified by the Poisson's ratio. p(xy)

,

denotes the intensity of the load at the point (xy)

parallel to z.

Dividing equation (1a) by (EI)x, the following

differential equation results which i.s mainly used here-

after:

P (:K1Y)= ·--(F~'fr~ (lb)

2{3 =(2) (3)

•

2{3 shall be called the c.oeffi.ci.(~nt of torsional

rigidity and (l' the ratio of the bencirJI.g stiffnesses.

The derivation of equation (1) i.s attributed to

J 0 Boussinesq (1874) and is given, fen: example) in

Reference 8, page 188. M.T. Huber (9~10) applied it for

various plate structures of techrdcal i.mportance: Con-

crete slabs reinforced in two directions for cracked

...'-

cross-sections, beam and slab construction.~ gri.dwork systems;

corrugated steel sheets and plywood plates.

For the discussion of a multi-bearn bridge a coordinate

system (x,y,z) is assumed as shown. in Fig. 1 wi.th x and y-

axis respectively parallel and perpend.icular to the beam

in the middle plane of the bridge 0 The z'~axis is positively

directed downward and the origin is. located at midspan of

•

•

•

-10-

one support. This type of bridge can be treated as an

·orthotropic plate of the second kind Ex = Ey if, in addition

to the basic assumptions of the general theory, the following

assumptions are made:

1. The connections between the beams are such, that

the points of contact in two adjacent beams de-

form equally.

2. The number of beams is large enough, such that

the real structure can be rep1aarl by an idealized

one with continuous properties, in order that the

differential calculus can be applied.

Th~ first assumption implies that the connections.

prevent the beams from slipping against one another, and

that they transmit the full shear force. This can be

accomplished either by lateral prestress and/or by shear

keys as shown in Fig. 2. With lateral prestress only the

shear forces are transmitted by, the ·friction produced by

the prestress. If;4 is the coefficient of static friction

between two concrete surfaces and erp.dA is the lateral

'prestress applied on the"di~ferentia1 area dA a maximum

shear force of !/f,o-:podA can be transmitted by the frictionA

forces without a slip.

•

1'..- ..

.'-

-11-

The second assumption requires that the width of the

beams is small compared with the width of the bridge such

that the actual lateral bending stiffness Ely and the co­

efficient 2H, which are both discontinuous, can be replaced

by average values. This will be discussed in the following

for Ely or a and in a later paragraph for 2H or 2~.

If, for a sufficiently high lateral prestress and

for any given load, the entire beam sides remain in com-

pression; the bridge will behave as a homogenous slab. The

benqing stiffness Ely in the lateral direction will be

constant and will b~ the same in the direction of the

beams; or a = 1.

Assuming a smaller prestress, the bending moments My

in the loaded slab may produce tensile stresses in the

Joints between the bearns. Since the joints do not have any

or only minor tensile strength (from the shear keys) these

tensile stresses will have to be carried by.the prestressing

~lements. The joints tend the~efore to open up, thus.

reducing the bending stiffness i.n this cross-section; or a ~l.

A limiting case with no lateral bending stiffness in

the joints may be reached if no lateral prestress is applied., .

In this case 'tpe joints are unable to transmit bending

mo~ents, but according to assumpti.on ~, the shear keys trans-

mit the full shear forces. Such a structure can be thought

..

•

,. It

...

-12-''- ..-

of being formed of beams with the adjacent sides connected

by continuous hinges along the beams 0 Herea.fter this type

of structure will be called an articulated plate.

Fig. 2 shows a schematic cross-section of the bridge

with the joints having a smaller lateral bending stiffness

than the beams.

The simplest way of obtaining average values for the

lateral bending stiffness would be by experiment on a

model bridge. This could be ac~ompli~hed by supporting

the bridge along the two edge beams instead of the supports

at the ends of 'the beams. Subjecting the plate to a load

which does not vary along the length of the plate, pro­

duces a deformed surfa.ce of cylindrical shape which can

easily be measured o Comparing these deflections with

the theoretica~ ones of an isotropic plate with the same

boundary and loading conditions, an average values for Ely

can be obtained. This test can be done for various magni­

tudes and locations of the lateral prestressing elements.

Dividing Ely by the flexural rigidity Elx per unit width

of the beams gives the coefficient a. This may vary as

shown previously between the two limiting values 0 and 1.

For design purposes it should be accurate enough to assume

..

•

•

-13-

first an appropriate value of a and later to check whether

the lateral bending moment can safely be transmitted by

the reduced section of the joints o

In the application of the theory of the orthotropic

plate the main difficulty is not so much the determination

of the coefficient a but the determination of the coefficient

of the torsional rididity 2~o

Various simplified assumptions can be found in the

literature for different structures. Most of these can be

traced back to Huber (9, 10,'11).

For two-way rein~orced concrete slabs, Huber recom­

mended the following approximation:

~ =~ (4)

Massonnet (3) used a similar expression in his investigation

of gridwork systems viz:

~ =;v;- (5 )

t••

·..

where f is a parameter of torsion varying between 0 and 1.

This approximation allowed him to consider any proportion

of torsional resistance of the beams and the bridge deck.

The most commonly applied expression for a gridwork

system as shown in Fig o 4 is:

'.-14-

(6)

where Cxand Cy are the torsional rigidities of the individual

stringers and beams, respectively, spaced at distances Cx

and cy apart. Dx is the average bendi.ng stiffness per unit

width of the gridwork in the x-direction.

The following relation is in general used to determine

the torsional rigidity of beams of rectangular cross-section

a.h:

."

•

?C := KGh..Ja (7a)

where G = E is the shearing modulus, K a cross-section2 (1+lJ)

factor depending only on the ratiQ a/h. This factor is

tabulated in several engi.neering handbooks and textbooks,

for example, Reference 12, page 277. In general for con-

crete structures the effect: of Poisson v s ratio lJ is neglected

since its influence is small. (13)

'Using I, the moment of Inertia for rectangular sections,

the above formula can then be expressed as:

C = 6KEI (7b)

• It

'..Substituting this expression in equation (6) and using the

average bending stiffnesses per unit width of the gridwork

D -_ EI~Y Cy

~15-

one obtains:

f3 = 3 (Kx + a Ky ) (8)

'.•

A multi-beam bridge may be considered as a limiting

case of a gridwork system where the beams would correspond

to stringers and the transverse members, normally represente4

by beams in a gridwork system, would be replaced by a co~-

tinuous plate with average bending stiffness Dy • In equation

(8) Kx would then assume a value corresponding to a

particular a/h ratio of the beams and Ky the value cor­

respondin~ to a plate. With a later value, Ky = 0.333,

the above equation would then take the form:

f3 = 3 (Kx + 0.333 a) (~)

••

The two functions (4) and (9) of f3 are plotted in

Fig. 6 for a varying between 0 and 1. Equation (4) re-

presents a parabola through the origin and the point a = 1,

f3 = 1, the axis coinciding with the a~axis. Equation (9)

represents a series of parallel lines, intersecting a =0,

such that the intercepts on the f3-axis are three times the

Kx values.

Considering now the limiting cases of the multi-beam

. bridge, it is apparent that for a = 1 the differential

.,

•

'"

I . ..

.. .

~16~

equation must be identical with that of an isotropic

pla,te, which is:

d 4w + 2 a4w + 44,w;:;: p(xlY)() x~ dx2Jy2 dy4 EI (10)

This implies that ~ as well as a must be 1. For

the case of the articulated plate, or ~ = 0, the structure

maintains the torsional rigidity of the beams, which means

that ~ must be equal to' 3Kx .

Each of the two approximations for ~ given by (4) .and

(9) satisfies one of the above conditions. Equation (4)

yields the correct value for a = 1 but gives zero for a = o.

From equation (9), for a = lone obtains a value larger

than 1, and the correct value for a = O. The true function

for a lies therefore somewhere between these two approximat-

ions.

In the following chapter a,n attempt. is made to establish

a relation between a and ~.

,"

-17-"0'-..-_.

II. DIFFERENTIAL EQUATION FOR THE MULTI ~BEAM BRIDGE

1. Derivation of the Differentia.l Equation

The second assumption made previously for the analysis

of a multi-beam bridge as an orthotropic plate implies that

the number of beams making up the bridge is of such magni-

tude, that the real structure can be replaced py one of a

continuous nature which exhibits equivalent average elastic

properties. Hence a differential element, cut out of the

latter structure by two pairs of planes parallel to the xz

and yx planes, will have bending stiffnesses Elx and Ely

per unit width in the directions x and y respectively.

Fig. 5 shows the middle plane of such an element with

the forces and moments acting per unit width of the element.

The moments are designated by arrows with double heads and

their directions are defined by the right hand rule. p(xy)

represents the intensity of the applied load.

Three independent equations can be written, expressing

. the equilibrium of the forces and moments. These are:

_ 'dMy; "dMyxQx - dx + d Y (lla)

••

...

_~ ()MxyQy - dY + a x (llb)

(llc)

-18-""""'.

By substituting in equation (lIe) the expressions for

Qx and Qy as given in equations (lla) and (lIb) the

equilibrium equation for the entire element results:

(12)

..

•

To represent this equation in terms of the deflection

of the plate w, the relation between the moments and the

deformations are used. In the following the effect of

Poisson's ratio is neglected. This agrees with general

practice for the investigation of reinforced concrete slabs .

This simplification allows the use of the relation

between moments and deformations from the elementary

beam theory, giving:

Q2Z a22'My = -Ely dY = -OEIx dY

(13a)

.. (13b)

Similar expressions can be written for the twisting

moments:

••'Mxy (14a) .

...Myx = - (14b)

•

..

-19-

where ~x and ~y are two constants to be determined. By

substituting these expressions in equation (12) the

differential equation for the deflection of the multi-

beam bridge is obtained:

with ~x + a ~y = 2~ equation (lb) results.

2. Ex erimental Determination of the Relation Betweena and

In order to establish an experimental relation between

a and ~ the homogenous differential equation is considered:

A particular integral satisfying this equation is (Reference

15, page 164):

......

w = C·x·y

with C an arbitrary constant.

This solution yie~ds from equations (13)·and (14):

~2w= - EIx ax2 ·= 0

My = - aEI h.. = 0x ay2

(17)

(lBa)

(lBb)

-20-and

'd 2wMxy = - ~x EIx dXdY = - ~X Elx·C

Myx = -af3y , EI d 2w =x dX'Jy

Adding equations (19a) and (19b) the constant C can be

determined as:

(19a)

(19b)

1C = -­EIx

Mxy + Mvxf3x + a ~y , (20)

and the deflection w results as

• 1 Mxy + M~xw = - EIx xyf3x + a y

.. or

.. 1 Mxy + Myxw = - EIx 2f3 xy

fFrom equations (18) and (19) it is evident that only

(2la)

(2lb)

....

.,.

constant twisting moments are acting on the bridge, deforming

the middle plane to the anticlastic surface given by

equation (21). This loading and deformation condition is

shown in Fig. 7.

In each element dy of the edges parallel to the y-

axis, the twisting moment Mxydy acting on it, can be

considered as being formed by two vertical and opposite\ .

shearing forces of magnitude Mxy at a distance dy apart,

(see Reference 8, page 47). It is apparent that these

"1'

•

•.. 41

...

-21-

shearing forces balance the ones in the adjacent elements

and only the forces in the corners of the plate remain.

These are, ± Mxy from the edges parallel to the y-axis

and, with the same reason + Myx from the edges parallel

to the y-axts, giving equal resulting corner forces of .

shown in Fig. 8.

If such corner loads are applied to a small bridge,I

in the laboratory, an experimental relation between a and

P can be found from the observed deflections and equadbn

(2lb). Also the effect of lateral prestressing on the

value of ~ may be established.

3, Assumption for the Relation Between a and~: .

Since these test data are· not yet available an

assumption for P as a function of a must be made. This

assumed relation mUst be chosen so as to satisfy at least

the'cond!tions for the limiting cases.

(a) '; For a ~. 0 the twisting moments are carried only. by .

the beams along their longitudinal axes. With a torsional

rigidity of the beams as given in equation (7b), the

twisting moments which results from the deformation for

this case are:

Mxy = -~'2w

6KElx ~x~y

-22-

Myx = 0

(b) For'Q ~ 1 the bridge can be considered as an isotropic

plate with Mxy =Myx and 2f3 = 2 or for a = 1: f3x = f3y = 1.'j/~ t:J.Assuming that f3x ~ 6 K(l - a ) + a and tJy = 1 and

substituting these expressions in 2f3 = f3x + a f3y the

following approximate relation between a and~ is obtained:

.. .}'.t..f3 => 3K (1 - a ) -+ a (22)

•

This expression yields exact values for a = 0 and a =1 and

reasonably correct values in the intermediate range. It

represents a family of curves with K, the constant of

torsional rigidity for rectangular beams, as parameter.

In Fig. 6 some of the curves for variousa/h ratios are

plotted.

With this assumption and with Elx => EI the moments·and

forces result from equations (11), (13), (14), as follows:

Bending Moments:

... Mx => - q2w EIdX2

••

My ~ - 22wEI aay2

(23a)

(23b)

•-23-

Twisting Moments:

Shear Forces:

a2wMxy = - ;)x~Y EI(2J3

J2wMyx = - ~~y EI 0'

Qy =

0')

(2fl - <>J EI

(24a)

(24b)

(25a)

(25b)

•

•

Boundary Shear Forces: (Ref. 15, page 154)

Rx - Qx + ~;L -[~ + 2fl ~;;~ EI

Ry = Qy + ~yx = _[~ 0' + 213 '/3wl EI~x dy3 dx2d~

4. Limiting Cases of the Differential Equation:

(26b)

(26b)

As mentioned before for 0' = 1 the differential equation

of the orthotropic plate becomes the differential equation

of the isotropic plate:

'J4w ~4w p4w. 4 + 2 ax20 y2 + " 4 =JX .. dY

P (xlY)EI (10)

.--.

For the case when 0' is zero the differential equation reduces

to the one for the articulated plate:

P(X1Y)EI (27)

•

-24-

where 130 = 3 K.

5. Boundary Conditions

In order to solve the differential equation for a

given bridge deck, the boundary conditions (hereafter

called B.C.) for simple supports and free edges h~ve to

be formulated.

(a) Simply Supported Edges

For a simply supported edge parallel to the y-axis the

deflection w and. the bending moment Mx is zero. The fol~

lowing conditions are resulting:

w = 0

and'/lw = 0~x2

(b) Free Edge

(28a)

(28b)

...

Along a free edge parallel to the x-axis the bending

moment My and the boundary shear

must be zero. These conditions are fulfilled if

..,

~2w = 0~y2

(29a)

= 0 (29b)

-25-

III. INTEGRATION OF THE DIFFERENTIAL EQUATION

1. General Considerations

In the general theory of linear differential equations,

the solution of the given non-homogenous equation (lb)

can be obtained by adding to the general solution

of the homogenous equation a particular

solution of the non-homogenous equation, or (14):

(30)

•

•

I ~

!

....

••

WI is a particular solution of the complete differential

equation taking into consideration the effect of the loading .

In general this solution does not satisfy all the B.C. Hence

Wo has to be superimposed on wI to give the exact solution.

Wo is the solution of the homogenous equation:

d4w + 2f3 "d 4w + Q! d4w = 0dX4 d x 2J y 2 dy4 (31)

which represents the differential equation of the unloaded

plate with boundary forces and boundary moments acting on

it.

For wI the solution of the infinitely long plate strip

is used which satisfies the B.C. along the supports. This

solution will be derived in the following for the general

"

-26-

case 0 c::Q< 1. The limiting cases occur with Q = 1, the

isotropic plate, and with Q = 0, the articulate plat~.

2. plate Strip of Infinite Width Under Line Loading:

(a) General Case:

Fig. 9 shows a plate strip of infinite width simply

supported along the edges parallel to the y-direction. A

line load, applied along the x-axis, and representing the

live load, is given by the following sine series:

•

p (x) = 2P 2:.1Tc fTl::l

1 . muus~n --

m 1• m'Tr c • mrrxs~n-- s~n--

1 I (32a)

1 sin rnlJ sin m'6 si'n m~m )

or with 1!..u....=lJ1

p(x) = 2P !1'( m=l

1!C1 =(5 1LL- F

1 - \

(32b)

••

where P is the total load uniformly distributed over the

length 2c.

Since the deflection is symmetrical with respect

to the x-axis, only the portion with positive y will be

considered. This is done by cutting the plate along the

x-axis. The resulting parts can now be analyzed as unloaded

plates on which boundary forces and moments are acting of

•-27-

such magnitude as to restore the continuity. The deflection

w has therefore to satisfy the homogenous differential

e'quation (31) and the B.C.

The Levy solution (Ref. 8, page 125) may be assumed

for this equation and is written as follows:

w =~ Ym • sin rn ~m;~ )

=~ Wmm::l

(33)

•

where Y is a function of y only. As each term of the seriesm

mus~ satisfy the differential equation and all the boundary

conditions, it is sufficient to consider only one term.

The B.C. along x = 0, and x = 1 , expressed in equation

(28) are satisfied by sin m f. In order to determine Ym

the deflection Wm has to be inserted in the differential

equation (31), yielding the following fourth order linear

differential equation for Ym:

Ym = 0 (34)

Taking Ym = ery the characteristic equation results as:,

• • with the roots:

= 0 (35)

rl 2 3 4 = +~-. / {3 [1 + -V1 - "'J i (36), '. , - 1 V C¥ - {32

..-28-

On the basis of the assumed relation between a and {3 these

roots will always be real.

Taking:

•

and

k1 = -V ~ [1

k2 = -V ~ ~ --,; 1a' J'- {32

(37a)

(37b)

(38)

•

...

the general solution of the plate equation becomes:

where Am, Bm, Cm, Om, are arbitrary constants to be

determined from the B.C ..

Observing that the deflection and its derivatives

approach ze+,o at a large distance from the x-axis, it may

be concluded that the constants Am and Cm are zero.

From the conditions of symmetry:

•-29-

yielding:

(40)

•

A second equation results from the fact that the load at

y.= 0 is equally divided between the two halves of the

plate and that these loads have to be in equilibrium with

the appropriate shearing forces, or:

[QymJ y=o = - P2m

= ~ 1't ; sin musin lll"$ sin mt (41) .

with the expression (25) for Qy this equation reduces to:

Em [-kl a + kl (213-a~ + Dm [-k~ a + k2 (213-a~

_ P12 1 . 11'- EI'6 1/3 ITi4 s~n m s~nm~ (42)

Solving the two equations (40), (42) simultaneously one

finds:

= -1. . v.Iii4 s~D.m1f s~n mo (43a)

Dm = EI 't 11 3 (43b)

••

and the deflection results:

(44)

..

•

..

..

..

-30-

Using equations (23), (26) expressions can be found

for the following: Mx ' My, Mxy ' Myx , Qy. These are listed in

Table. (I). These expressions hold for values:

and

For the limiting values of a and ~, however, one must

proceed to the limit and this will be shown in the following

for the deflection w.

(b) Isotropic Plate:

For the case of the isotropic plate with a and ~ equal

to l? ~f i,~::easy tb seetha,t kl = k2 = 1. $ubstituting

these value,s in equation (44) results in indeterminate

expressions ~ for each term of the series. These are

examined as follows. Using the identity:

kle-k2m~-k2e-k1m~

a klk2(kf-k~ =

e-k2m~ (kl

-k2e{q -k2)m~)

a klk2 (kl+k2) (kl - k2)

a.

the right hand side can be resolved into the following

partial fractions:

e-k2m~ (l+e- (kl-k2)rn~ l-e- (kl -kt)m?)a klk2 2 (kl+k2) + 2 (kl-k2)

•

•

•

-31-

~he transition to the limit yields:

and the deflection of the isotropic plate is obtained

(Reference 8, page 169):

the formulas for the moments and forces are given in

Reference 16, page 26 •

(c) Articulated Plate:

For the case of 0:-.0 and (3-.{3o it is noted that:

'..

lim kl = lim ~ [1+ ~ l-fil CD 0()

0:... 0 ~{3 -{3o

For:

lim k2 = lim \I~ rl-~l - ~l(~..., ~ V l ~ J~~;Jdo

the Binom;nal Expansion is used:

i~ 1·0: 1 a2V1 - ~ = 1 -2 {32 + 8 ~ - ...

(464)

•-32-

and bearing in mind that, since ;2 is very small, higher

order terms can be neglected, the limiting value of k2

results as:

Furthermore, one ob~ains:

(46b)

..

•

..

(47)

Now, upon investigating equation (44) the general terms

can be partially rewritten as follows:

kle-k2m~ - k2e-klm~ e-k2m~ e-klml?a klk2 (k12. - k2'a..) - == a k2 (kt-kt>- a kl (k{..kt) .

Using the limiting:values as derived above, the second

expression becomes zero, the first becomes:

and the deflection w of the articulated plate results:

p~2w =.-EI-~-1i- J---~-r:2:"'?f3'"""o,...1 sin l1J\Isin ~}{ sin m~

(48)

... The expressions for Mx , Qy, and Mxy are derived from

equ~tions (23) to (26) and are given in Table (I). My

•

o •

••

-33-

for this case is zero. Equations 44, 45, and 48 hold for

the complete plate if instead of y the absolute values,'Y',

are used.

3. Solution ~or the Bridge of Finite Width:

(a) Orthotropic Plate:

The bridge of span 1 and width 2b is loaded with a

line loading as shown in Fig. 10.

Two different coordinate systems are used to simplify

the numerical calculations.

(x,y,z) is the coordinate system used for the solution

of the homogenous equation with the origin at the middle

of one support. All the values, such as deflections, forces

and moments related to this solution will be marked by an

index O. For the solution of the infinitely wide bridge

the coordinate system (xl, Yl' zl) is used with the x,axis

coinciding with the line of the applied load. An index 1

will designate the values in this second system. The

distances of the free edges from the two x-axes are denoted

as y = +b and Yl = e'b, and Yl = -e"b, where e'+e" = 2.

The midpoint of the applied load is given by the coordinates

(u, v), where v = (l-e 10

) • b is the distance from the x-

axis. The equations of transformation between the two

•-34-

coordinate systems are:

xl = x·

Yl = y-v

zl = z (49)

In later equations, similar abbreviations as in the

derivation of the particular solution, are used. These

abbreviations are given by:

£ _1-- b

•

A ='lI.£• 1

~ = k£

~I = ,.(~E+ e I~ I I (50)

As mentioned before, the solution of the non-homogenous

differential equation (lb) can be formed by adding the

general solution of th~ homogenous equation Wo to the

solution wl (equation (44) ) of the infinitely wide plate

strip in such a way that the B.C. at the free edges are

..... fulfilled.

Considering only one term of the series, the solution

of the homogenous equations as given in equation (39) will

•

•

•

..

-..

-35-

be used in the following transformation, using hyper~

bolic functions:

Where Am, Bm~ Cm, and Dm are a new set of constants. The

applications of this solution permit simplifications which

far outweigh the disadvantage of the more elaborate trans-

itions to the limit.

One term of the complete solution now takes the form

of

Wm = (Am coshklm~ + Ib:J coshk2m? + Gns:inhktm~ +D:nsinhk2~) sinm~+ Wim

(52)The requirement that each term of w satisfies the

B.C. permits the determination of the constants. The B.C.

as given by equation (29a and 29b) are:

for y = +b:

My = [My~ + [My~ Yl=e'b= 0 (a)y=b

Ry = [Ry~ y=b + [Ry~ yl=e'b = 0 (b)

My = [My~ y=-b +[ My~ yl=e"b = 0 (c)

Ry = ~y~ y=-b + [Ry~ yl=e".b = 0 (d)

(

(53)

1-"

-36-

substituting equation (52) in equation (53) the following

simultaneous equations can be formulated for each value of

m:

from equation (53a):

=

from equation (53c):

~y:l'yl=e 'b____~-!J'-.':'""-"---- = S1m'a EI sin m~ (54a)

•

from equation (53b):

[M.Yi! yl=-e"b. aEI sin m~ (54b)

Am k12 sinhlqm,(+Bmk22sinhk2mA+ Cnl<\cosh klmA + Dmk22 c.osh k2 m,(

= _ ~ [axIl yl=e'b _m3 ttr3 a klk2 E,I SinmT - S3m "(54c)

from equation (53d):

.. (54d)

•

••

-37-

Determinants will be used to solve this system of

equations. Observing the sYIIlmetrical and skew-symmetrical

properties, the determinant of the denominator reduces to

the following: (15)

al bl cl dl

al bl -cl -dl= 4 (albZ - aZbl) (cldZ - CZdl)

aZ bZ Cz dZ

dt(55)

-az -bZ cz where:

al = klZ cosh kl mA , etc •

For the same reason, simple expressions are obtained for

the third order determinants Upg , where p stands for the

eliminated column and g for the eliminated row.

Ull ~ U1Z = +ZbZ(cldZ -- cZdl) .

(56)

-38-

With these values the constant Am is a function

of the boundary values .Slm' S2m' S3m' S4m' is-found thus:

Ull Slm + U12 S2m +U13 + 83m +.U14 S4mAm = 4 (alb2 ~ a2b]) (Cld2 ... c2dl)

or:

and similarly:

Slm + S2m2 .

S3m - S4m2 (57a)

'.•

Bm =-a2 Slm + S2m a~ S3m - S4m

alb2-a2bl Z + alb2-a2bl 2

Cmd2 Slm S2m dl S3m + S4m

= cld2-c 2dl 2 cldZ- c2dl 2

-c2 Slm-S2m cl S3m + S4mOm = cld2-c2dl 2 + Cld2- c2dl 2,

As is shown below, Am and Bm contain the symmetrical part

(57b)

(57c)

(57d)

••

of wl' whereas Cm and Dm contain the skew-symmetrical.part

of wl- If wl is symmetrical with respect to the x-axis

the following identities hold:

-39-

and

or

and

which gives

= [RylJyl==b

Sm3 = - Sm4

em = Dm = O.

For wi being skew-symmetrical with respect to the x-axis and

• rMYl]L yl=b- ~M J- L YJ yl=-b rRylJ = fRylJ .L yl=b L· yl=-b

it follows that Am = Bm = O.

Replacing the coefficients of the determinants by the

terms given in equation (57) and using the following

abbreviations:

-40-1. .

jlmbl k2coshk2rn~

= alb2-a2bl = ktcosb~lmAsinhk2mA-k!sinhklmAcoshk2m~ (58e)... _.

.. al ktcoshklm,,(j2m =

alb2~a2bl - klcoshk:lmAsinhk2mA- kisinhklm,(coshk2mA (58f)

"'It-

dl k2s~_n.hk2mA

j3m = cld2- c 2dl - k~sinhklm~ coshk2mA- k.~coshklrnA sinhk2mA (58g)

Cl k~sinhk1m,(j4~ = cld2-c 2dl =- ktsin.hkJY,il~coshk2mA- k2coshk1mA sinhk2ql,{ (58h)

the constants can be written as;

Slm + SZm • 83 .. S4mJ mAm = ilm - 1Ji'i -• 2 2 (a)

Slm + S2m j S" - S4m,jID

Bm =-iZm 2 +'?"fifll -_._--2 (b)

..Slm ~ S2m . S3m + S4m

em = i3m 2 J3m 2 (c)

=-i4mSlm - S2m e S3m -} 84m

Dm 2 -+ J4~ 2 (cD, (59)

and the deflection of the bridge results:

.'.e.

w = Zl(t\ncoshkim? + Bmcoshkzm? + Cmsinhkim~ + IlmSinhkZm?)Sinmr+ wi

(60)

(b) Isotropic Plate:

As proved before, the deflections and therefore the

moments and shear forces of the infinitely long isotropic

".

•

e

•

-4l~

plate result from the equations of the orthotropic plate

by a transition to the limit a~l. In the following the

same will be done for the solution of the homogenous

equation (32). Since the analysis is the same, the

transition will be shown in more detail for the first two

terms of equation (51) and only the resulting expressions

will be given for the last terms.

The first two terms:

rewritten with equation (59) as:

Slm + S2m. S3m ~. S4m)(ilm . 2 -Jlm ----2 . coshklm?

(Slm + S2m S3m ; S4m]

+ -i2m 2 + j2m / coshk2m?can be transformed into:

and with the values given 1n equation (58) into:

....

e.

_ k~coshk2mAcoshklm? klcoshk1mAcoshk2m~

klcoshklmAsinhklm~ k~sinhklmxcoshk2mAS3m-84m

2

•

~ ..

..

...

".

-42-

Using the boundary values Myl and Ryl of the iso­

.tropic plate it is easy to see that Slm + S2m as well2

as 83m - S4m are of finite magnitude. One needs there­2

fore to investigate only the two remaining fractions.

The following steps are necessary:

1. Rewrite the fractions as functions of (kl + k2) and

2. Divide the numerator and denominator by (kl - k2).

3. Proceed to the limit (kl - k2)... O.

In order to clarify the operation the numerators Nl

and N2 and the denominator D will be investigated separately:

The first numerator becomes:

and with the following identities:

sinh (a+b) = sinhacoshb + coshasinhb·

cosh (a+b) = coshacoshb + sinhasinhb

it can be arranged into:

[kl +k2 kl -k2 kl +k2 kl -k2 ]

(kl-k2) sinhm~ 2 coshmA 2coshm~ .2 coshm~ 2·

•

•

'""

~43­

+(kl+k2) [ sinhm,< k.l:+k2cosh.mt< kl -k2sinhmn kl +k2sinhmrykl-k;l2 2 or 2 2 J

[kl+1<2 kl -k2 k +k k -k ~- (kl+k2) coshrnA sinhmA coshmn 1 2coshmr> 1 2

2 2 .( 2 I 2

Dividing by (kl - k2) and proceeding to the limit (kl -k2)... O

yields:

Applying exactly the same operations to the second

numerator N2 and to the denominator D one obtains:

N2 : -(2coshmA + mtsinhm~)coshm? + m?cosh m~sinhm?

D: 3coshmA sinhmA -m~

Substituting these values in equation (32) and rearranging

the expresgion yields:

with the values A*m and B*m as given below. Similar

expressions are obtained for the later terms.

,.

•

. ... . .

-44-

The deflection of the homogenous bridge thu~ is

obta1n~d as:

The coefficients A*m to D*m are given by equation (59) if

i m and jm ~~e replaced by i*m and j*m as follows:

(pI)

.*~lm

m~coshmA- sinhmA=~--=--~---:----:--:-~

m~ - 3coshm~ sinhm~

•

•

".".

.* sinhmA~2m = mA - 3coshmA sinhmA

.* . mA s inhm~ - co shmA~3m =-mA+3coshmAsinhmA

.* coshmA~4m =-mA+3cosl:l.InA sinhm~

.* mAsinhm~+2coshm~. JIm = mA - 3coshmA sinluriA

.* coshmAJ2m = mA-3coshmAsinhmA

.* mAcoshmA+2sinhmAJ3m mA+3coshmAsinhmA

.'J< sinhm~J4m ~~mA+3coshm~sinhm~ (62)

-45-

and~

•

1,Q~2'l- '= 1.:__'-JI ''''1 2 IEJ[

IlliI m 11 .

r;A~

33m

(c) Articulat,e.:! .?lat~_~

rM,p·:tl vJI -~e"b.l':..L~:'="__

Sir..,m~

._L~11E=e~ bsinm~

_hI] YL=-~siri\1Jll

(a)

(b)

(c)

(d)(63)

'0

•

o.

/

For the li.u:d.ting c:~t:3:e a~O 'which includes kr'~oO,

k2-ko = ,,\~i and ~l;'7 ."= 0 >, it is easy to see from equation"Vz.(3~ .f •

(58) that i.n equation (59) Am and em v,anish and that amreduces to:

and re,arranging the expre,~sion, we obtaiR,

1Em = a kTilk2Sill\hk~t<rJA ~:ak2' tanhklm,{coshk2ID .

~. tq

[~!L.x.l.=~..Ub.:.....~yJ !l=a'e"b2EI sin ror

-46-

For a-O the second term in the denominator approaches

zero and:

It follows that:

yl=-e"b

(64a)

and similarly:

yl=e 'b+ [Ryl] yl=-e"b(64b)

."The deflection of the articulated plate can now be

written as:

QC

W = L (Bg coshkom~ + ng sinhkoml?) sin m~ + wIm:l \

4. Explicit Form of Formulas

(65)

...

a.

Equations (60), (61) and (65) can be written in explicit

form by replacing the boundary values Myl and Ryl by their

values obtained from equation (24) and (26) with equation

(44). The resulting expression is listed in Table II for

the general case of the orthotropic plate together with the

important formulas for moments and forces used in the design

~' ,

..

•

•

•

a.

-47-

of bridges. Table III contains these expressions for

the articulate plate., For the,case of the isotropic

plate, reference is made to the extensive study by

H. Olsen and F. Reinitzhuber of the rectangular plate with

two opposite edges simply supported and the two other

edges free. (16). This study contains all the formulas,

as well as influence surfaces for deflection and moments

for plates of various sizes.

5. Influence Surfaces:

In beam statics influence lines are commonly used in

studying the effect of concentrated loads or load systems •

The influence line for the moment, for example, describes

the variation of the moment at a given point of a beam

due to the passage of a single load across the beam.

In a similar way influence surfaces can be defined for

plate structures. The influence surface, for example ,for'

the moment at a fixed point (u,v) for a unit load is given

by a surface with the ordinate z in the point (x,y) equal

to the moment'MX produced at the point (u,v) by the unit

load placed at (x,y).

The following discussion of influence surfaces holds

as well for the isotropic as for the orthotropic plate, and

covers only those items which will be used in the next

-48-

chapter to derive the properties of the lateral load

distribution coefficientis. For i:l [!]ore complete investi-

gation includi.ng the derivation of the basic rules, one

is referred to K. Gi.rkmcu.!~~ pCil.ge 225. (11)

d . I. fThe two rna,i.n rules for the . er~v,at~on of in luence

surfaces are~

1 0 The deflection su.rface W = 'VI (x,y;u,v) due to

p = 1 a.t the point (1.1, v) i.i:,~ equal to the influence surface

..for the deflection at point

•

•

....

-".

2. The i.nfluence Si.l.rf:aces for arn.y derivative of the

deflection at the point (u.~v) if; obtained by differentiating

the deflection surface w = w (x,y,u,v) due. to P = 1 at

(u,v) with re.spect to the coordinates u and v .

It C~1n be proved that the above rules hold also for

the case in 'which the unit load is replaced by a line load,

applied over the di.stance 2;;-; parallel to the xg·axis and

with the mid-poi.nt coordin.'ites u and v. The proof is based

on the fact that, for the gi:\,7en li.ne loa.d at (u, v), the

deflect:i..on at (x,y) is equ$.l to the deflection at (u,v)

due to the line load at (x,y). Thts can b~ seen from

equation (1), Table II, by il'.~teT.'ch.anging the coordinates

(x,y) and (u~v)o

"

"

As the expression for the line load in the equation

for the deflection·is not affected by the differentiation

of the latter with respect to x and y or to u and v, it

follows that the second rule holds also ,for the line load.

Obviously one obtains in this case the influence surface

for this type of loading.

Fr.om the first rule it follows that equation (1) in

Table II also represents the function of the influence sur-

face for deflection at the point (u,v) due to the given

line load. This conclusion holds generally since it

follows directly from Maxwell's law of reciprical deflection§.

Applying the second rule to derive the influence line

for the moment Mxat the point (u,v) which is denoted by

[Mx ] u,v one obtains:

[Mot ] uiit =- EI ~~~ =_ EI 02w

dX2 (66)

It is easy to see that the resulting expression is identic~l.

with equation (2) in Table 1;1. It follows that this

(

equation also represents the function

the influence surface for the bending moment Mx at the

of

"'. point (u,v) for th~ given loading. This result will be

used in the following chapter to establish an important

property of the coefficients of lateral load distribution.

•

!

-50-

It must be noted that the above identity holds only

for the bending moment Mx and only because of the simple

boundary conditions of the plate for x = 0, and x =1 ,

and the fact that the effect of Poisson's ratio is

neglec ted. If needed, the func t ions "of the influence

surfaces for other moments and forces may be found with­

out difficulty by differentiating equation (1) in Table ~~

according to the second rule. The same results can be

obtained from equation (3), (4), and (5) of Tab Le· II by

considering the coordinate (x,y) as fixed and the co­

ordinates.of the midpoint of the loading (u,v) as variable.

In this case the equations represent the influence sur­

faces for the point (x,y) •

...

I>

•

••

-51-

IV. LATERAl. LOAD DISTRIBUTION

1. General:

The formulas for any location of a live loading derived

in the foregoing chapter a.nd summarized in Tables I and II

permit an accurate design of multi-beam bridges. To

eliminate the extensive amount of calculation, numerical

values ar.e given in the later part of this study for de­

flections, moments and shear forces at points along, and

for loading positions ~n the midspan cross-section of the

bridge. They were computed for bridges of different span

and width and also for various degrees of lateral prestress.

These loading positions at midspan of the bridge are,

in general, the most unfavorable position$ encountered in

the design of such bridges. However, in many cases, other

loading positions have to be considered for which only the

principal bending moment Mx would be required. The deter­

mination of Mx ' however, may involve much numerical work.

In this case, an approximate determination of the bending

moment Mx can be made using coefficients of lateral load

distribution, introduced ~y Y. Guyon. (4)'

2. Definition

Considering again the bridge with the live load applied

as shown in Fig. 10, the average bending ~oment in a cross-

of

-52-

section x = Xs of the bridge is given by:

If+bMxav :: 2b-b

Mx dy (67a)

..

For the case where the effect of Poisson's ratio is neg-

lected this average moment is equal to the bending moment

per unit width obtained by replacing the applied live load,

which acts over distance 2c, by an equivalent uniformly

distributed load, acting over an ar~a 2c.2b, where 2bis

the width of the bridge.

By integrating equation (2) 1.n Table II the average

bending moment is obtained as :

•

+b

ib J Mx dy =-b

1 sinm1Jsin m~sinmfm3 )

(67b)

This expression developed in a sine-series is identical

with the average bending moment obtained from conventional

beam statics. This is true because the above mentioned

loading, which does not vary over the width of the bridge,

caus~s a deformation of cylindri.cal shape.

The coefficient o~ the lateral load distribution Sxy

for the point (x,y) subjected to the actual bending moment

Mx , may now be defined as:

,S~y - -::l~-i""'~~:----

2b Mx dy

-b

-53-

(68)

'.•

This is a non-dimensional coefficent and indicates the

portion of the average bending moments of the sections x = xs

which exist at the point (xs,y).

F'or the design of a multi-beam hridge one is especially

interested in the bending moment associated with each

beam incorporated in the bridge. For the beam i, for ex-

ample, this moment is s4bstantially the bending moment per

unit width at the mid-point of the beam, multiplied by its

width a; from equation (67a):

a Mxi = sxi tb JTb

-b f+b

sxiMxdY=n

-b

IVlx dy

(69)

as a percentage of the total cross-sectional moment

where n = 2b is the number of beams.a

one obtains:

Expressing this moment+b. (Mx dy

)-b

s*·x~sxin

100(70)

...Assuming now that the actual l~ve load is not directly

applied upon beam i, the bending moment at beam i may be

obtained from a consideration of the interaction of the beam

-...

-54-

as expressed by equation (60). This moment can be thought

of as being produced by a proportion of the actual load­

ing, directly applied on an i.ndependent beam. It is

apparent that the facto~ of proportionality is identical

with the factor given by equation (70). S*xi also indicates

the distribution of the actual load applied on. one beam

among the other bearps. Hence, in the determi.nation of the

magnitude of the beam bending moments, each beam is con­

sidered as being independent and subjected to a load

depending upon its value s*xi.

Similarly the coefficient sxi may be considered as

a measure of the lateral load distribution, and since it

is independent of the number of beams u, it is more suit­

able for non-dimensional representation.

It is to be understood that the coefficient Sxy for the

point (xy) varies with the type and the location of the

applied live loading. Its introduction, therefore, does

not yield any advantage in the accurate determination of

the bending moments. However, it is very useful in approxi­

mating the bending moments.

3. Approximation for the Lateral Load Distribution:

Y. Guyon (4) and Massonnet(5) computed the coefficients

of the lateral load distribution for a live load of~nusoidal

.\

-55-

type given by:

p(x) _. Po(71)

where Po is the load intensity at midspan of the bridge.

This expressi.on is i.dentical ~Jith the first term of the

sine-series of equation (32b) where P is replaced by:

Po_ D__2sinvsinl(

•

..

For this loading the moments and forces are given by the

formulas derived in Chapter III with rn = 1 .

With equation (2) in Table II and equation (67b)

the coefficient of the lateral load distribution for the

above loading results as:

SxyMx= _ .._.:

MA av (72-)

.~

where the expression in the bracket above is identical

with the expression in the bracket of equation (2) in

Table II.

It is easy to see that this coefficient of the lateral

load distribution depends only on the eccentricity of the

line live load wi.th respect to the bridge axis and on the

•

•

,-

~56-

ordinate y of the refer.pp~p. noint (xy). It is inde-

dependent of the abscissa x of the crosB-section under

consideration. It follows that for a particular loading

position, the moment curves of all beams are similar in

shape, with the moment ordi.nates dl~pending upon Sy. The

sarne relation can be derived for the deflection curves

and it follo\'iT8 that the cOE';;fficient Sy also represents the

ratio of the deflection at a pairl."!:; y of any crosse-section

to the average deflection of the cross-section.

Because of these consi.derable advantages and the fact

that their values can be more readily computed) Guyon and

Massonnet considered the coefficients for the sinusoidal

loading as a practical approximation to be used for any

type and location of loading encountered in the design of

bridges.

The application of thi.s approximation may be justified

for loading conditionG pr.escribed in some European speci-

fication. According to the .American specification for

l'l)highway bridges'" ,however, one must consider the effect

of concentrated wheel loads or truck axle loads. For such

types of loading whi.ch differ considerably from the sinusoi-

dal load, the above approximation coefficients lead to a

very favorable distribtltion of the load but induce a rather

small moment i.n. the directly loaded beam.

\

,.

e.

••

For a safer design~ especially for concentrated loads

and wheel loads, the following approximatiO!l is proposed.

This is based on the exact distribution coefficients com­

puted for the midspan cross=section of the bridge correspond­

ing to a wheel load placed at midspan. It i.s considered

that this posi.tion is generally the most unfavorable en­

countered in bridge design. It may therefore be assumed

that this lateral load distribution is representative for

all cross-sections of the bri.dge subject to a similar load

application in each crosS-6ect~on. It is apparent that

variations might exist in the curve of lateral loa.d distri­

bution from section to section along the bridge. However,

it may also be ass~~ed that, having derived the curve for

the worst condition of loading, that i.s, at midspan, it may

be applied to any section to yield a moment which would be

higher than the actu.al moment existing at that section.

Coefficients for lateral load distribution corresponding

to the most important loading conditions are computed and

given in the following chapter.

4. Properties of the Coeffi.cient of Lateral Load Distribution:

It was stated in Chapter III that equati.on (2) in Table II

representing the moment Mx at point (x,y) due to a load at

(u,v) also described the function of the influence surface

'\

~58-

for the moment Mx at the point (u ~ v) . F'rom this and using

equation (68) an important property of the lateral load

distribution coefficients follo,tJ'S: "The curve of the co-

efficients of the lateral load distribution in the midspan

cross-section for a load applied at the point v of the same

section, is also the influence line for the coefficient at

midspan and for the point v. n This means that the reciprocal

relation holds:

Syv = SV'lj (73)

or that the coefficient of the Lateral load di.stribution at

y due to a lo,a.d at v is equal to the coefficient at v due

to the load at y.

The se.cond property of these coefficients, though

obvious, is still worth mentioning, since it is useful for

checking purposes.

Keeping the definition of the coefficient in mind

(equation (68) ) and integrating Syv over the cross-section

one obtains:

..,

'..:; 2b

(74)

-59-

In performing this operation one may replace the integral

by the value of the area. under the syv-curve determined

using Simpsonvs rule:. In the next chapter the coefficients

are computed for the following points~

(-b ~ 3b = b = b, -4-~ 2 ~ 4'

• b..,- --2' +b)

Applying Sirnpsonos; rule to each point a.nd designating the

coeffici.ents for the va.rious points as 511' 52 .•.• 59

yields:

or:

This condition may be used to check the accuracy of

the distribution coefficients.

,'"

.,

'I,

0,

.,

'.

~60-

v. NUMERICAL CALCULATIONS

1. General:

The extensi.ve amoun.t of numerical work to compute exact

values for all formulas listed in Tables II and III was

reduced by making use of an electronic high speed computer.

Numerical values were computed for 9 points at equal inter­

vals along the midspan cross·asection. The values for Mxy

were determined for points along the support, x = 0. Two

loading positions in the same cross-section were investi­

gated, namely the load placed in the bridge axis with v = 0,

or e r = 1 (Fig. 10) and the load placed at one edge with

v =-b or e r = 2.

The following parameters were considered:

(a) The ratio of the bending stiffnesses a.

The computati.ons were made for a:::: 0, the case of

the arti.culated plate, and for Ol::: 0.1 and a = 0.5.

With these several a - values and the values given by

Olsen and Reinitzhuber(16) for the isotropic plate

a = 1, an interpolation for any intermediate a is

possible.

(b) The size of the beams •

The size of the beams, or more precisely the width

to depth ratio a/h of the beam making up the bridge

.,

..

-.

••

affects the coefficient of the torsional rigidity of

the orthotropic pi,ate, (equation 22). The following

two ratios which are used in present bridge design

were' considered in the cornputati.on~

a/h = 1.00 and a/h ~ 1.7

(c) The size of the bri.dge.

With span 1 and b the half width of the bridge,

computations were made for:

b~ = 0.5 0.375 0.25 0.125

For any intermediate bit ratio, moments and forces may be

obtained with sufficient practical accuracy by interpolating

between the given values.

(d) Line Loading

For the represe.ntation of a wheel load in the

form of a line load, the distribution of the load

in the longitudinal direction ~sho'Wn in Fig. 11.

It is assumed that the length of contact between

the wheel and the wearin~ surface is 4 inches in

the longitudinal direction and that the load is

then distributed at an angle of 45° through a

2-inch wearing surface plus the half depth h/2

of the beams. Thus a total longitudinal dist-

tribution of 2c = 8 in. + h. results.

..

•

•

-62-

This distribution, depending on the depth of the beams, is

in turn a function of the span of the bridge.

The present computations were made for one c~ ratio.

This was chosen as cll = 0.0318 or -rrc/l = 0.1, which

corresponds approximately to the value of c obtained from

the above distributio~ and for beam depths and spans

currently used for multi-beam bridges.

The writer is aware of the fact that the above assumption

for the distribution of a wheel load is not included in the

American Specifications for Highway Bridges(l). These

specifications do not consider any longitudinal distribution

of wheel loads. In bridges analyzed according to the

derived method however, a concentrated load would cause

infinitely large moments directly under the load. To avoid

these infinitely large moments, which are of a theoretical

nature only, the above longitudinal distribution over the

diStance 2c is assumed.

Any distribution of the load in the .lateral direction,

caused by the width of the tires and the wearing surface,

as well as the beam depth may be taken into account by

evaluating the prepared influence lines.

..

..

•

..

••

••

-63-

2. Accuracy of the Results:

The calculation of the values for the series expression

includes the terms up to and including ill = 19. This

limitation was set by the capaeity of the computing machine

in handling certain large parts of the series.

The accuracy of the results is therefore a function of

the convergence of the series in Table II, ~vhich in turn

depends mainly on the power of the factor directly after

the summation sign. Furthermore, it also depends on the

location of the points, for which series were computed, wi.th

respect to the applied load .

Check calculations have revealed that for points not

coinciding with the load location, the computed values, with

the exception of those for Qy' are exact to within three

significant figures. For points. directly under the load,

E= 0 (location of point) for a' = 1 (location of load) and

£= -1.00 for e' = 2 (see Tables IV - XXVII) the accuracy

is as follows:

(a) Deflections:

The convergence of the series with 1/m4 is excellent,

the calculated values check at least to within three

significant figures .

..

(b)

~64-

Moments. (l1x ' My, Mxy ): . ..The ser~es converge sat~sfactor~ly w~th 11m2 and the

•

•

..

•

given values are accurate to two significant figures.

(c) Shear Force Qy :

Since the power of m in the series expres~ion for Qy

is 1, the series converges very poorly. For points not

directly under the lo~d, ttle values ~ay be accurate to

within two significant figures. For the peak values, occurring

in the points directly under the loaq., only the first figure

is correct. A better valuernay be obtained by considering

equilibrium between ~he applied load and the shear force

(neglecting the influ,ence of twisting moments) yielding:

for e' = 1:

p- - 2.2y

e' = 2:

= -

~. 7085

p2C

'.

= ~. 15.7

•

..

-65-

In general, the accuracy of the results obtained may be

considered sufficient for design purposes, since only

slide rule precision is required.

3. DescriE,Sion of the Tables :.

The results of the numerical calculati.ons, are given

in Tables IV-XXVII (as obtained fro~ the computer). Each

table includes values determined for one combination of

parameters a, alh, bll, as listed in the heading of the

table. The left h~nd column shows the Q4antities w, Mx '

My, Qy and Mxy ' where each one is multiplied by a coefficient

which renders it dimensionless; the negative sign for My

and Qy should also be noted. The secop,d column indicates

the position of the load: e' = 1 (E' = 1)* represents

the load placed in the bridge axis and e' = 2 (E' = 2)

the load placed at the edge.

The top line shows the coeffici~nt £ = t determining

the distance of the points from the bridge axis for which

values were computed. The sign of £ is indicated in the

third column. For e' = 1 the values are given only for £

positive, since under tnis loading the values w, Mx , My are

symmetrical with respect tq the bridge axis; similarly Qy

* Note: Only capital letters are printed by the computer .••e' = E'

-66-

and MXy are skew-symmetrical to this axis.

For e' = 2 the values are given in the second and

third line for each qu~ntity. The second line includes,

as indicated by the sign in the third column, the values

for £ positive, whereas in the third line the values for

£ negative are listed. The values for £ = 0 are given in

both lines.

The computed values are represented in the following

form:

where

+ X • 10n

l""X~O.l

..

and the exponent is a positive or negative integer.

The tables include the numbers for x and n only and

have to be read as shown by the following example.

The value for wEI for e' gland £ = 0 in Table IVPfZ

is given by:

.33822 -1

which stands for:

4. Coefficients of Lateral Load Distribution:

The coefficients of lateral load distribution, as defined

by equation (68), were calculated and are given in Tables XXVIII

•

..

•

••

-67-

and XXIX for bridges composed of beams with a ratio

alh = 1.7 and a/h = 1.00 respectively. Included are the

coefficients for the isotropic plate or a = 1.0, which were

calculated from the bending moments given by Olsen and

Reintzhuber (16) .

The same results are plotted in Figs. 12-19. Figs. 12-15

show the coefficients for the load applied in ,the bridge axis,

or e' = 1. The upper graph includes the values for alh = 1.00,

whereas the lower graph shows those for alh = 1.70. For the

case of the isotropic plate (a = 1.0) the ratio alh is

meaningless .

Figs. 16-19 give the coefficients for alh = 1.7 with

the load applied at the edge £ = -1.00, and those for alh = 1.00

with the 10b1d at E= +1.00.

Each graph includes coefficient values for one a -value

and shows the effect of the various bridge sizes, described

by the parameter b~. This representation had to be chosen,

because of the unexpectedly small differences in the coeffi­

cients for a fixed b1 -ratio and variable a-values. With

the exception of the peak values directly under the load the

coefficients are only slightly affected by a variation of a.

The effect of this variation is somewhat larger in wide

bridges (bh = 0.500) than in narrow and long bridges

(bit. = 0.125).

•

•

-68-

Similarly one may conclude that the coefficients

calculated for the two a/h ratios differ only by a

small amount. Significant differen~es are obtained for

small a -values and then only for the points directly

under the load.

It follows therefore that the size of the beams and

the amount of lateral prestress in a multi-beam bridge

affects the coefficien~s of lateral load ~istributton only

in the points near the applied load.

For the design of multi-beam bridges, it is recalled,

that Figs. 12-15 represent also the influence lines for

the coefficients of lateral load distribution at the

midpoint£ = 0 and Figs. 16-19 those at the edge points

of the bridge. The application will be shown in Chapter VII.

•

•

-69-

VI. COMPARISON OF THEORETICAL AND EXPERIMENTAL RESULTS-~ - -- - ......;.;;-.-;.....,.,;.---

1. Gener:=l.l

A field test on a multi~beam bridge is described in

Reference 2. The following data is taken from this

report.

The tested highway bridge with a span of 34 ft. and a

width of 27 ft. is composed of 9 pr~fabricated) pretensioned

concrete beams 36 in. wide and 21. in. deep. placed side

by side the peams are connected together by a steel bolt

at midspan (without significant prestress) and dry-packed

shear keys. The geometrical parmoeters for this bridge are:

b 27f == 2(34)

1.7

= 0.4

o.

ft~ong other tests a single axle load of 47,000 lbs.

was con-:;entrated over the 'i.l7idth of one beam, first in the

bridge axis and secondly, 4 ft. 9 1/2 in. from the edge.

The results for these loading positions will be compared

with the theoretical values.

2. Deflections:

Fig. 20 shovJS the measured and theoretic8.1. deflection

curves for the load placed in the bridge axis. The,cross-

t.

..

•

n70~

section of the bridge and the loading position are

schematically indicated in the top of the graph.

The theoretical curves for a = 0, 0.1, and 0.5 were

obtained for b1 = 0.4 by interpolating between the values

for bit = 0.5 and bit = 0.375. The deflecti.ons were com­

puted with a modulus of Elasticity of 6.68.106 psi, the

approximate value determined in the fi.eld test. Similarly

Fig. 21 shows the measured deflections for the second

actual loading and the theoretical deflections for the edge

loading.

Comparing the experimental and theoretical deflection

curves, one cannot completely exclude the possibility that

slip may have occurred between the beams in the tested

bridge. Thus the main assumption for the theoretical analysis

(no slip· between the beams)wQuld have been violated. From

observation of the curves this fact is more apparent for

the load applied in the bridge axis. On the other hand, the

maximum measured deflection, for the same loading position

is approached by the theoretical one, for a = 0, to within

12.5%.

With these facts in mind, one may conclude that the

results of the theoretical study are in satsifactory agree-

ment with the experimental ones ..

•

3. Coefficients of Lateral Load Distribution:

In Reference 2, the difficulties arising for determining,

the exact coefficients of lateral load distribution from the

results of the field tests are discussed. For the reasons in­

dicated, only the approximate coefficients based on the dis­

tribution of the deflections could be obtained.

These approximate coefficients are defined by equations_

(68) and (70) when Mx is replaced by the deflection w of the

same point and Mxav by the average deflection wav of the cross­

section under consideration. The S~ coefficients so deter­

mined are given again in Fig. 22 for the load applied in the

centerline and in Fig. 23 for the load applied on the edge.

The same:figures include the approximate coefficients com­

puted from the theoretical deflections as well as the exact

S~i coefficients obtained from the theoretical moment dis­

tribution. The given theoretical results correspond to a

value of a = o.

A comparison of the curves plotted in Fig. 22 reveals

a satisfactory agreement between the approximate coefficients

resulting from the experimental and tbetheoretical investi­

gations. Both of these curves show a similar trend. The

curve representing the exact coefficients deviates from the

trend mentioned above to a larger extent. The deviation is

especially significant for the sections of the curves corres~

ponding to the directly loaded beam.

•

•

,

-72-

Similarly in the case of the edge loading (Fig. 23)

one observes a close agreement between the two approximate

values for the loaded edge-beams. Again, the corresponding

e:c.act curve shews a substantial difference.

To permit a better comparison the following table

includes the three different S*xi coefficients for the

loaded beam in both loading cases.

Approximate Load Exact LoadDistr:Lbution Distribution

Experimental Theoretical Theoretical

Center Loading 16.6% 15.2% 19.6%

Edge Loading 20.8% 22.6% 28.3%

First in the case of center loading, according to the

approximate coefficient obtained from the field tests, tl:\e

center-beam would have to be designed for 16.6% of the

applied load. However the exact coefficient indicates that

this beam carries 19. 6% of the load. The beam 'Ovould there-

fore be under-designed, whereby approximately one-fifth of

the load would have been neglected.

Secondly in the case of edge loading, the experimentally. .

determined portion.of the load to be carried by the edge

beam was fOurld to be 20.8%. The corresponding exact value

."

•

•

~73-

.~

•

•

.'

-74- '

VII. USE OF THE COEFFICIENTS OF LATERAL LOAD DISTRIBUTION

The following is an example showing how the coefficients

of lateral load distribution may ~e used in the design of a

multi-beam bridge.

It is assumed that the bridge has a span of 36 ft. and

otherwise is identical with the bri4ge described in Chapter

VI and in Reference 2. The overall width of the bridge is

27 ft. and includes two 8 in. curbs cast to the edge beams.

The nominal roadway is therefore 25 ft. 8 in. The geometrical

parameters for this bridge are:

b = 27 = 0.375r 2(36)

a = 36 1.7- =h 21

The live load to be considered in the design of the

bridge is given by the AASHO Specifications (1) and con-

sists of H20-s16-44 truck loadings.

For the determination of the design load of the edge

beam it is assumed that a = 0 or that no lateral prestress

is applied, The coefficients of lateral load distribution

for this case can be taken from Fig. 16 or Table XXVIII .

They are plotted to the correct scale in Fig. 24 below the

schematic cross-section of the bridge •.

-75-

On the latter, two truck axles are placed side by side

with the minimum spacing between, as given by the specifi-

cations and such that the sum or the areas Al' A2' A3 , A4

in Fig. 24 is a maximum.

Utilizing now the properties of the eurve plotted in

Fig. 24 as an influence line for the coefficient of lateral

load distribution of the edge beam, the portion of the load

P carried by this beam is given by equation (70) with n = 9,

the number of the beams of width a, as:

= 50.3%

•

.'

..,r.

This means that the edge beam carries 1.50.3% of a wheel load .

From the assu,rnption that the same ratio may be used for

other cross-sections of the bridge it follows, that the

edge beam has to be designed for 50.3% of the wheel loads

positioned for maximum effect.

Similarly it was found that the maxi.mum portion of

the wheel loads carried by the center beam is 50.2% or

practically the same as for the edge beams.

The same evaluation of the coefficient of lateral load

distribution was made for other a-values, as well as for

different bit ratios and also for alh = 1.,00. In all cases

it was assumed that the width of the bridge would be 27 ft.

" -;:"." " ...

-76-

•

The various bit ratios ~7ould then correspond to the

following bridge span$~

bh = 0.5 1= 27 ft.bit = 0.375 1 = 36 ft.bit = 0.25 1= 54 ft.b!l = 0.125 1 =108 ft.

For the case of a/h ='1.00 it was assumed that the

bridge would be formed 'by 15 beams, 21· inches wide.. The

re'sults of this evaluatio.n are summarized in Fig. 25.

The graph on: the left-hand side shows the portion of

the wheel load carried by the center beam as a function

of a and with b1 as parameter. The upper group of curves

includes the values obtained for a bridge formed :by 9

beams, whereas the. lower graph ~hows those for a bridge

composed of.15 beams.

In the right-hand graph the portion of the wheel

load carried by the edge beam is plotted in a similar

manner.

Both figures confirm the interesting result, that for .

the loading positions under consideration, the portions ,of

the load carried by the center beam and the edge beam are

practically the same. They vary only slight~y with the,

degree of lateral prestre.ss, described by the coefficient a.

In the case of a bridge with a span of 27 ft., composed of

i

•

•

-77-

9 beams, the largest percentage of a wheel load carried

by one beam amounts to 53.2%. For the same bridge com­

posed of 15 beams it amounts to 32.9%. The percentages

are smaller for longer bridges and approach in the

limits the values corresponding to a uniform load dis­

tribution, which are 44.5% and 26.7%.

A comparison of these results with the 80% prescribed

by the present specification reveals the obvious advantage

of an exact investigation .

1.

2.'.

•

•

-78-

VIII. CONCLUSIONS AND RECOMMENDATIONS

The main purpose of this theoretical study has been

~o provide information for the more economical design

of multi-beam bridges. The main results are summarized

below:

There was found to be a satisfactory agreement between

the theoretical and the available experimental results.

This may be considered as a confirmation of the estab­

lished assumptions and the applied methods of analyzing

this type of structure.

The approximate distribution coefficients based on the

deflections are especially insufficient for the design

of the directly loaded beam. A safe design for wheel

and axle-loads must be based on the given exact load

distribution coefficients.

3. The size of ,the beams and the amount of lateral pre­

stress affect the coefficient of lateral load distri-

bution only in the points near the applied load.

4. The curves representing the coefficients of lateral

load distribution for the midpoint - and edge - loading

can be considered as influence lines for the load­

portion to be carried by the center - and the edge beam,.

Their evaluation, similar to those used in the beam -

7 •

6 .

5.

•

•

-79-

statics yield the design load for one beam expressed

as the percentage of the left or right wheel loads

of the standard truck.

For a 27 ft. wide bridge with two trucks placed side

by side, 'it is shown that the maximum load carried by

a 3 ft. wide beam is 55% of the right or left wheel

loads of one truck. This value is almost independent

of the amount of lateral prestress and varies only

slightly with the span of the bridge. It has to be

compared with the 80% recommended by the present

specification.

This significant reduction of the design load can

fully be utilized only if the connections completely

prevent the beams from slipping against one another.

The condition mentioned is mainly a technological one

to be examined by tests, which are strongly recommended.

In the meantime, the results of the theoretical study

may be used for the design of bridges if the interaction

of the beam is guaranteed by sufficieQtly high lateral

prestress.

8. The relatively small influence on the load distribution

which is caused by a variation of a shows clearly that

..

..

•

. 9.

-80-

the main advantage of laterally prestressing a mu~ti-

beam bridge is an increase in the shear resistance of

the connections between the beams .

Further experimental investigations on multi-beam

bridges by means of field and laboratory test are

recommended. The scope of these tests should incor-

porate the following points:

(a) Investigati.on of the influence of the magnitudeand the locations of the lateral prestressingelements on the coefficient a.

(b) Verification of the assumptions, especially theo~e con~erning the relation between a and ~.

(c) Investigation of the behavior of multi-beambridges under higher loads, determination of thecollapse load and the factor of safety •

....a) General Case

-81-

..

..

'. Mxy = 2{3- c( . 00. -L (-~ITV? kl m7J) .p 2"yP VI p....

2:., m2 e - e Sin mv sin my cos m{

(I)

(2)

(3)

(5)

b) Articulated Plate My :: 0

f,.I"r

.9.pt! =_ .L ~ -L eko m'11r m=1 m sin mv sin my sin m!

~= i. ~ I ;komT] . .P "'y m;' m2'" . 510 mv 510 my COS m€

(6)

(7)

(8)

(9)

• •

. ,i,.;.. '. ..

..:,;

•¥.- I 'f *sinmUSlnmySOnmC[:t(rellz""h - 1011,""',,_ t (-U,~IS sonh"'"",+ Uz"2rsiMbz!'kl-V'~ISCOsll~m"+ ~llztcoshhzmll~2.,1J V'-;t ..., ,J . n

~, 12S-c()! ·~slf1mU!lll'Imycoamc[±(·ekz"""-e~''''''')-.L(-U,ll,sonh~,..... + Uz~zsinhllz"'" -~k,CO$hll,"'" + Vz~hhzli'i·P 21l'yfj l/i _.a "'" m 2 •¥ Sz .

o Note the upper S'lin IS valid far .~lI-e'). Ih3 Io<oct s'gn " valod for 0<(1-0'),

(1)

(2)

(3)

(4)

(5 )

3

N. ' cosh to, mH'M "'.lm~ - (-r.) s,nh k,m~ cosh IczmA

(~3Nz' "nh k,m~ cosh I<zm~ - \ll,Jcash II,m~ 'IGlh kzm~ Ila· J~~-FJ'] ,

k,(-Ilz"Wl' -lo.zmAe") I-ktm"e' k,mU")o ' -.. e + e - \e .. •, k,

k'I:IL..~· -IL..~") (-.,m~e' ,.m~e·)DZ 'k,'\e" - e--' - f' - e"

.. _ I"'z"",e' -kz~"l (~2\eI<,m~' -kt"",e")"t- e +e - e +e

. " r • [p (I.+V'- p~) - 011

s' [$ (, -Hz) - ~I

'I' ..!.l.b

Notation:

• ' .l.b

see Table III & Fig. 10

Table II Formulas for General Case

Ico~I

-83-

M)ty • ..L I ~ sin mv sin my cos me [t tom~_ t (u,slnh kom'l) +~ cosh ko m'))]P "''1 "'_, m \.. (4)

•

M~. 0 • Note: the upper sign Is valid for .t (I - e')

the lower sign is valid for fe: (I - et)

Formulas for Articulated Plate

. £~

;•...i~t....~:'1 ;, '

li.

i';:~~,'Table m

• III j.OJ b

~ I; 1- el

b

AU'H4 .0TAIILE IV

BETA IUJ. .f»O

~/h a 1,7

'l'AIlUl 9

liTA lU..l. .6)lI

tlo/'h. 1."IJISII.Ql '," .00 0.25 0.50 0.75

-~I;~~~i--- ---_ ..__._._. .. ~ ..... ~._-_. ---_._.~ ----1----------;. .,. i' 1 • .JJlW " 1 .~" 1 .1905,} " 1 .161W1 • I

i' ~ • • 151~ " 1 .1\2s;l • 1 .lIIlQll " 2 •74IM • 2 •'10\27 • a!a' :I. .IS1~ " 1 .;al61 . 1 .0I9JQ4 " 1 .41713 • II .606)0 • I j"". . ---_. . -..._- ...,-"-- ._ ...._.- ---

"-.'~'67"--O1M.. I' 1 .J9lIIl4 0 .;l8S74 il .:41J27 0 •162'l"1 0,..I' 2 •. IS167 0 .1119S 0 .&7196 • 1 • '7)617 - 1 .6'l288 • 1•I' :I. ,1511>7 0 .21J~ 0 .lU33 0 .4m5 0 .1128) • 1

'.- .'-EPSILOIl .,.- .. 0.25 o.~ Go'1'5 ,..

-- .'''' -, .'_. ,Hi _, I + ,2'lIIll - I ~-1 .19h6. ,

.. 2 + ....11 - , .1,. -, ..,,,,.. ._.. ....". a_, 1 - .'fI61' •, ,.ata)6 • , ..... , ...... .....

_ . ..,.,. .;a -' , + ..... 0 .-m 0 .... • ..- •_, a· + .1569'l1 0 .11'JS) 0 oWm-1 .-., ......._, a .. .1569'l1 0 ,21IlIO 0 .J'499 0' .- 0 .......

-;r _, I • -.1"" 0 .''7»1 •) .1OMB - , ...--a ..... '1

I' 2 + .11. - I ..,,, • a .6II'l6 - a """.1 .~.'l\

.' I - .11527 - , .''''75 • , .~2lIl • , ...... ,•..". tr--'- ........ a-:n I' I + .1IUCl'7 + I .1»41 • , .B'J89 0 .-

f' a • .4865J 0 ,)11'" 0 ."'" 0 ..'"' • '1·.456ft ••

.' a - o4Il654 0 •.".50 0 .1ll6a9 • I .-ra +' .,"" t a

~I' I • .00ll00 ."''16 0 .1*, 0 o6JOOt - 1 ..... 1

f. a I • .148'l\l 0 .Wip· I ,6tan - I .J)01) • , ...- -,I' 2 i - .~ \) .210"11 0 .211661 0 .»615 0 ".",., iI

TAilLE VIa£TA~ .6JO

~/ll • 1.7

'UILI VIIBETA IIU ,4.U

~/h • 1.0

-.

~=._-.....,...--,-----.,...'-- ---.,...---~~--~_._-

i.t - 1.00 o.as 0.50 0.7SJ--~~

"!-i i " 1 • .W7'1· 1 .23047.1 .iIJJ629. 1 .18565" '1 .169?5- I

PI I";j. • .1(,97{.1 .1'281·1 .10734 - I .90\43· 2\.7'179?" a.' 2. • •W/?S "' .:al9S· I .289S7· I ,)'JlU· 1 0488'19" I1---1---+--1----+----- 1--.----- e. --- .. -.

II" I" I • I .40921 0 .m67 () .2al4J 0 .18993 0 .17126 0

r "'.:1. • 1."'n2/> 0 .1)25"" 0!.,Cl64? () .89192.1, .76911 - 1~l~r-:- .~'~~2~ __~ .~~ OJ ..31~ .. 0 .4Hfl2 0 .•~31 _ oj-;r ," 1 .. "~IOS~. 0 .16960· '1.1)397. 1 .16llO6 - 1 -.19929 - h

" 2 I • .Cil!)l:S - 1 .J62.i\. 1 I .26798 • 1 .15'.l44" 1 -.)llj'l6 - I>

I' 2 i - .Cil!36· lj.S5m • 1 1.6500'1 . 1 .?!T/M - I .4)~ - S

• " I I. .601IJ'7' 1 • 1)Q}6 1 il.-ij"r#.-O 1-;1917:1-'0 :;65Q06~i-'l-I I ~, 'i " 2 1," .41170 0 .2.0/183 ~ i .14)55 () .471J33· I -.~5 - Ii I! " 3.! - i .41168 0' .64m u i .10627 .. 1 '2263' ~_,,~~=a :

~ :" 1 I... ! ,oooao ,4J445". I .58j5Q - 1 .~ - '1 .io/:DXJ. I

II ' .' 2 I .!.,3816 0 .<f12fJfl. 1 .b8U15 - 1 ~11. '1.J7SllQ - I ,

~16 il! .181~5 () .24BO 0 .25"'1 01 .2SJ5J "

- - __ ·_e"···EPSIUlti ...- .00 0.2S 0.'" 0.'15 '.00._._--- ----W o' 1 I • .,~" - 1 .251\44 - 1 •Ul3Q4 • I .W". I .1Jtat. - ,

~ al • .131:14 • I .89195 _ 2 .6J767 _ I .!IOt" • a .4J927 • a,. a - .131:14 - , .19790 - I .)0236 • I ~-1 •'D'III9 - I

1\

---,1,1. 0' .. .71o:lll 0 .;rjJ92 0 .18'MI 0 .14411 0 .1llP7S 0r

I' 2 • .1307' 0 .1IlU68 - 1 .6JQ:lI9 • I .49562 - 1 04UU. I

0' 2 - .1)075 0 .'9986 0 .31592 0 .SJ8;l'I 0 .1-"'1). I

-';J 1" .. ooסס0. .00ll00 ooסס0• ooסס0. . .lIIlIllioI' 2 • .00ll00 .oooao ooסס0. .oooao ooסס0.

I' 01 - .0000Il .oooao ooסס0. .00ll00 .oooao

-~!I' I • .603iJ'1 .. 1 .1Ql!4;Z • 1 .451~ 0 .18561 0 .~9)" 6

I' 2 • .J61'19 0 .2Olll95 Q .1268'1 0 .S7868 - 1I-.~ _ 6

.0' 2 · ,;)61'19 0 .Sif.!IS 0 .1Cl289 t·' .:.lZlI61 • 1 ,l6Ofi .• a...... _.~ ..............~-'"' ....... '.''-''~'" --~"':.- --

.!iQlll4 ---,-,.--=M..,

.' 1 • .24641 0 .17439 0 .'0'722 0 1-.421'71 - 6I

.68946 - 1 .4lXIU • I l.lm2 - I I-.J'-'9 ~ 6

I " 2 .. .11011 0

I' 2 - ."011 0 ,169?~ 0 ~2544J 0 i .J6710 0 .,49a1 0

AU'HA .1TABLIl VI II

II£TA IlLl .4Zl

~l\ • l.0

!AaU: UlETA IIJ.L .4AlI

&/tI a 1.0

IPSllQIl .,- ,OIl 0.2' 0.50 0,75 . "OO~--~----

'Sl " I .. .2'N.bl - 1 .2:;2" - 1 .1989'1 • 1 ,16221 - 1 .14O'1ll - IPI'

,4~7" 2'.' . • .140?8 - 1 .97912 - 2 .714'2- 2 .)6271 - 2

0' 2 - .14D17 - I .a:no~ - 1 .30657 " 1 .4S1S7 - 1 .i>5i!64 - I

"ll I' 1 • .S22Jl () .29-"90 0 .20711 0 • 1!>lOll 0 .I4Q(,.< 0r- .1' 2 .I4Q42 0 .9?<>'IJ - I .706Q/, • I .S5S76 - 1 .47361 • 1•

f' 2 - .14Q<l2 0 .2D162 0 .J.!213 0 .~ 0 .12169 • 1

.~ " 1 • -.1:;212 o -.1S464 - ~ .115JOl - 1 .10541· I -."J9'7'Z7 ... 7I

0' 2 t .14&27. 1 .10359 • 1 .723;lS - 2 .~_2 ··~-710' 2 - .1~Z7 - 1 .21H2 - 1 .29117 • 1 .36761 - I -.2Il1074 - S-1----_..__.- .- _. --,--_._ ... 1----._--- --._--

.~ " 1 • .0030'7 .. 1 .11111 .. 1 .4671/7 0 .18~ 0-.14581 • 1

[I ~ t .)56/02 0 .21an 0 .~a~ 0 .!i4741 • 1 -.54.)4;l • 2

" 2 - .35641 0 .~'/l!71 0 .997'J8 0 .~4Jll .. 1 .14857 • 2._----..., " I • .0Q:XIll ,938'iJ - 1 .9J'I9O - I .605J2 • 1 .4QJ69 - 1

" 2 .. I .12062 0 .7'118) - 1 .461~ _ 1 .2S06J. 1 .1Sl148 - 1

•, a • i .1206Z o . .17'190 0 .1"" 0 .~ 01 .JJa44_~

R"1.Ql +,. .00 0.25 0.'" 0.75 '.lIO----'-- .----~ rl I • .l!lOt2 • , .23a'75 - I .aD652 - I .1&:l~ - I .16Sa4 - I

I' 2 • .16$14 • 1 .12462 - I .966DO • 01 .7745' - .6lll' . aI' 2 - .16~ - 1 .2211lt • I .3911'6 - , ,)9'05 - I .5Zl9J - ,

;. I' , • .41_ 0 .2'l'il8I 0 .21954 \) •Ul6'r5 c .16m 0

II 2 • .16m 0 ,1~ 0 .~-1 .'765016 • I .62556 • I

I' :I. · ,16575 0 .2Zrl7 0 .Jan4 0 ..48434 c .968~ 0

.~ 0' 1 • _.:141)' o -.197"J6 _ , •12M - I ,16671 - , .WIII. 6

.' 2 • .51368 • 1 .3911IO " I .28682 " 1 .'6420 - I .,"', • 6

0' a - .,U'IO - I .~"'.7J605 _ 1 .80148 _ I ,aB65a ~ 4._--_. ----- 1-------

-~ " 1 t .603iJ'1 • I .1Ill68 .. 1 ,476J7 0 .162'72 o -.'1415a • I

I' :I • .);l'lll' Q .1'16'Jb 0 .101~ 0 •.nOIla _ 1 _.Jill•• 1

I' I - .J2'Al6 0 .1l1llO 0 .9Cl9'71 0 .1'D~ • , .1:asD • "-~ "' I • .ooaoo .JfH14 • , .51409 - 1 .4'mO. , .... ·'1:

~, 2 • .1~ II • ~'7oUl> - I .601Q3 • , "''107 • , .,.,.. .' ...

_, :I. · • 'a<i41 0 .1624D 0 .~ 0 .~ c .326Jt 01'

EPSIUlII +,- .00 0.2S" O.SO .I o. 7~ 1.00..

~, " I ..- .JeI211 - 1 .J09'4 • 1 .1:hJtIJ - 1 T;ul!lY' - 1 .22'lllJ -

.229lU - 1 .\IH17 - , .1<009 - 1.1478J - 1 .14286 - 1" Z. T

1.4704J3 - 1" 2- - .ZI9ll4 • 1 .2861,0 - 1 .3641) - 1 .(,19~7 - 1

..__..- ---- _.. -~ " i 1" .6415J U •.>6712- 0 .2IIJ1' 0 .24560 0 .~ 0

.' 2- t .;u..SS u .19071 0 .16Z~ 0 .I4/>72 0 .14161 0

.' 2-. .Z34~ 0 .)00t:6 0 .JJJJ7'1 0 .5t!d'l~ 0 .11416 + 1

----

-;: " 1 1- ooסס0. ooסס0. ooסס0. ooסס0. ooסס0.

" 2- ~ ooסס0. .OOOCO ooסס0. .OOOCO ooסס0.

" 2- - ooסס0. .WOOl) ooסס0. ooסס0.

:~-~'s-1.~..._..- --......~-,_.-., ..- .............,- ..

-%1 " 1 + .-.:JJ07' 1 .17187' 1 .7371' 0 .311112. 0

" 2. T .7111if- 0 .41'J89 0 .2'J044 U .13&0 0 .6le71 • 5 i" 1- - .71'~ 0 .10l'i?2. >I .,..,. Y'''''' 1 • 16016~ , 2.. ,\r---"-" ..._.,c

..... ... i'

1Ij1 I' 1 + .~I· 0 .Ie~ o .122U1 0 .60441- I .12JlI4 - ,~

(i Z. ~I·I~ 01 .0389 "I .S~I)ll - 1 .41J24- 1 .1916U - 5'!I

.:15<176 'J! .)~t;, oi .. lZU. oJ! .492112. 0:,,2.. -j .I~ 01 ,

IlUI BEMllI\l~

f~ll

BETA ILL.L .6JQ

a/ll 1"•I ~IL.OII +,- .00 0.15 0.50 0.75 "lID

-Et " I + .)2660 - I .)0)69 - 1 .Z7JQ4 • I .25025 • , ..u"'. If'YW

" 2 .-U~' - 1 .19693'· I .non· I .15113 • , .\4JD • I+

" 2 - .-U~~ • 1 .~., .~7- I ..um - I .t'IMI - ,

~ " 1 + .5)0)6 0 .36'S5 0 ._fJ 0 .2fJ05 0 .:141" 0

0' 2 + .24121 0 .191l19 0 .1'l'OO6 0 .Isago 0 .142'16 0

" ;l - ."4121 0 .JO"'I 0 ~19 0 .~J 0 .106QJ. 1

-(t " 1 + -.1))01 o .,41)72 • 2 .'I'Ill9' .2 .IQJ72 - I -.74149 • .,

(' 2 + .17249 - , .14\)4- I .11361 _ 1 .~.2 _.2lI012 • .,

" 2 - .11'44'1 - ., .2'061! - 1 .25347 - I .aMIl7. 1 -.IS01? • 4

I-~0' 1 + ,i$D)07 .. I .17A1Xl .. I .14J2'I 0 .)OJ-U o -.t4680 • I

c' 2 .. .6<;0aa6 01 04S990 0\ .2llO>S 0 .1296' o -.91'180 • 2

I " 2 - .6982' 0: .10570 • 1 .171lJ .. I .,)4S9tl ., I ."137" a

l~ " I + ooסס0. .16.u'l - 1 .Cl4920 • 1 .6ll6IlI • I ,"~. I

, I " 2.. .1'lO96 0 .I)6IlJ 0 .'JO'01 - 1 .S4J71 - I .,J'7llXJ • I

! ; " 1 - .19096 0 .2'lJ8 I) .)IJO'Ij I) .36126 0 """" 0-----

Bt\- .175'I'ABU! X

BETA N.L1. •<{So

.;n .. 1.?

ALPHA .0

8t\- .)7~' .... lIU

BETA ILU .4Z<

a/l> • 1. 0

M.PHA .0

[J-;IL;A>j '.- .00 O.2~ 0.50 0.7S 1.00

WII " 1 -t" .~., .3214il • 1 .2~7~ _ I .222402 • 1 .21122 _ I

PI'" 2 .ll1l'" • 1 .'6289 • I .IJ'67. 1 .1\.'6 - 1 .101l~ - I....

Ift 2 - ,2111't - 1 .211190. 1 ,JllJ2? - I .~I _I .7ioJY5 - I

II .. ~ 1 • .744~~ 0 .J71065 0 .27'24J 0 .22h~ 0 .212\/8 0r

f' 2 • .21298 0 .1~ 0 .1)060 0 .11295 0 .107)0 0

.' 2 - .2129ll 0 ."J!f0h8 0 0414'!6 C .i.I,ojS 0 .U"8 .. I

II " 1 01- .1))000 ooסס0. ooסס0. ooסס0. -oo,,:rסס0," 2 + ooסס0. ooסס0. .OOUDO .OOCIJO ooסס0.

" 2 · ooסס0. ooסס0. ooסס0. ooסס0. ooסס0.

-~!(' 1 .~ .&)JIY1 ... I .1429) , 1 ,~JO 0 ,2~14Ol 0 .4377'1 - 6

" d- + .~~N 0 .")61~ 0 .218It 0 .10282 0-.17130.6

" J. · .~W70 0 .~1ll 0 .14280 • 1 ,2%U t I .16061 , 2--_.'-

I liT " 1 .. .lo4641 I) .182" 0 .1111lS 0 .rtnJO - I -.:«1361 • 6

i " 2~, .1W43 0 .101l44 0 .6t6I4iiJ • I .)IIlT.! - I -.24171 - 6

! [' ;l · .I~ 0 .;z;:)'/o 0 .)OJI" 0 .NTaJ 0 .492111 0"

o

o

1.00

.25715

v .149'11'

Bt\- .)7''I'ABU! XII

BETA IUL .1.30alb, 1.1

.oouoo .2&432..·1 .;,;;67f!J- 1 .4611.;1 - 1 ...WI - 1

.1~5113 C .12'/,.. 0 .~21- 1 .76ez;' - I .6BOI~ - 1

.~1)7 • 1 .I"lOO5"' I .109'/4 U .2«>55 0 -.80.ti'4 • 1

.511:w> 0 .36"1) .~3Jt u .%<0 - 1 -.~)<\Q( - I

.sgJ;.4 V .~ Q .141>12.·' 1 .~+ 1 .1~'"

[I 1

" ~

-.PHA .,

EPSIUlII '.- .00 O•.2.S" I U.Si' I G.'6

J.-W-,...-~"'l';-;-~-'-2.'+--:-+....,,:::--~"'7O---.-::+-:-~c=,~~::=~T=:~ :::: :c' ~ <-5016 - 1 .<''1632.- 1 .35421 - 1 .1I47s - 1 .o;prf2-- 1

II.. (, I {o .I,(J.!~ U -.;-;);.4-'0 -:jif;;3····0 -:-;;:.:J!-"cr ;[, ~+ ~2S7I" C .1..ll.i.iJ 0 .18'/.1 0 .1~

I-_-+~(':'":.~:+..::"":,+-~.I~.$71~'~~O~.~3~1q$~r-_::o~.4UOB~~1~U1'r·~~'100::::"-T·~~

,'I 't -.urI4 0 -.~56 I ,$1/{$ - 2. .14;0) - 1 -.~120 - ,

c':l ... • ~9416 • 1 .5OCI4S .'3&3? - I .2.."'9:14 - I -.,..,131 - 6

"... .~14 - 1 .iih88. 1 .7:"U.t.12-- 1 .,31',7 - 1 -.~- 4

•

TA8U1:llYIITA IILl. ,4U

a/tl .. 1.0IEPSIUlII +.- .00. O• .l~ o.m 0.71 lao 1:~

( .. ,JQUQ • .7.!/J3'j - 1 .27m - 1 .261rr _ 1 .14m • , 1I o' ~ • .14731 - I .al69J • 1 .176\J - I .1S2~ - , .IJJOS. I I

" ;l - .447JI - I .m21 - I ,.J641t - I ,44,Jjl • , .5:164' • ,

IIx " 1 + ,41)71 0 .'~7 0 .JO;l16 0 .;n,60 0 .'~ 0p-o' 2 .. .2~ 0 .207lIO 0 • \7"" 0 .1~IJ\ 0 .1)1'1'6 0

" 2 - .2"'J<I 0 .;1"04 0 .4V1'dI> 0 .~789 0 .9SJ~ 0- .

-~ " I • ••-U~J4 o -.2?lS/j _ I .'/07'J~ • 2 ,IAIlO'I • I • •.lOJ75 • S

.' 1 • .6~'n~ • 1 .~'111 _ I .41;:79 _ I .4021. I •• 10~)~.J

" ~ - .65714 • 1 .74601 - , .lll*l) - 1 .~., -.1119:1 - ..t------~._- --

"~ .' I+ ,lKIJ07 • I ,15'7, + 1 .CU840 0 .Zl4OlJ o -.~O - ,

c' ~ • ....17IfJ 0 .~5 0 .1J,II44 0 .2.01161 _ 1 -.W164 - I

c' a · ....7781 0 .75181 0 .14Jj", .2S62l> ., I .12liO"/ ., :I_._--- ...

~ " I • .0lXI00 .lJ'IA - , .J7n6. , ....,001, - I ~·'I

I .' 2 • .14'H1 0 ,,,.,. 0 •.",711 • , ..,. -, "'14'1·' Ic' a - .14'~ 0 .'";<1, 0 .2D6llO 0 •.:tIOfJ 0 ,zaul 0....

tABU ~lV

IIrTA IUL .4<Ualb • ~.O

..-' -1IPSlL.OII ; +,•. .00 0.2' O.S:O 0.7) 1.0e

~I' 1 + ,';3'112-. 1 .Jl0" . ,

.27)1' - I .~70 _ 1 .2Z))JI • I

e' +. .~~-I .1?5"roJ. , .14~b6 - 1 .121,'1 _ 1 .1\011 • 1 i

o' II - .~.I .29:J& - 1 .1i'319 - 1 .ljl)5i,I - 1 .f#7i - I

;. ", + .l6470 0 .~71~ C .l.~I<l5" 0 .~UI 0 .22St/P "

" Ol. t" ,aa~ 0 .\7!l45" 0 ,142bO C .1~ 0 .IOB'JO 0

" 2I

.Z2~~ 0 .301011 C .417'2i 0 .62.7.:&1 0 .IVJ.i'" 1-> " 1 • -.I4ka \I -.68671 - <- .103'19 - 1 .11"90 - 1-,I?9~-b

" ;Z ... .aa~9. , .17'6~ - 1 .1j):' .. 1 .~.1- .; _.1~~. ,

.' , · .U"'9 -, .:IS:lJl - 1 .")L,e~-- 1 .38591 - 1 -·'JIJ.J:'(l- r-

-~! " 1 .. .•!kJ)07." 1 • 14111~, I .624bS u .24~t<J u -.2l.$)- I

" 2- 1" ,~o 0 .3I.9~ 0 .'2D'153 0 .9189S" - 1 -.'2451 - I

.1' 2 - .~ 0 .~p 0 .1318J + 1 '~'I • lJ,8~~ .. l......, i' 1 +- ooסס0. .1069' - I .14524 - I .66ID _ I SJJ96 - I

i'

2 1" .167/1 0 .117/1 0 .77494 - 1 .4&.)) - , .:l6U)) • ,

" 2- · .11>711 0 ."2Z~? o . :'~'h?. 0 .J296ll 0 .J3579 0

---'-" _L-.

taaul nIllETA IIU .6JO

alb • loT~------,.----,-----

I-O>SIUlNi';::r---'OO . 0.25 o.~ 0.." ,..

I .1 j" 1 . + ~ - 1 -.4JWJ - 1 04\4'lO - 1 .J999I -, ~ - ,

1 PI" ,'2 +,"- - 1 ",.. - 1 .J8l2 -, .Jt:o' -, .." - ,

_ I' 2\ -: ..., - 1 ..uJ06 - 1 .... - 1 !.,J'lQIl ~_'~ ~~r-~ I"' 1 I + i .~)9 0 .52* 0 ~5 0 .4M4 0 ,jI1MlJ 0

I ,'2 +! ,409" 0 .J6W 0 .J)499 0 .J1496 0 ...25 0

i :" 2 I -I ,409S2 0 047)91 0.J7* 0 .7J'I\O 0 .'"'' + 1

.l-;x I"'T '+ i~·'2W O!-=I-·G49: 1~:2 .996" - 2 .1964'7 ~6ii" 2 t' • I .26702 - 1 .2)410 - 1 .191)7 - 1 I .1a.66 -, .:wu4 - ,

I 1'~_2..~+-.~ -__1~~~ ~~.~)ll4.~.~+~)IU'7 -, 046522 - ,r;;--rll'.' 1 + Il1JJO'7 + 11 24515 + 1 ' .1lJ9,U + 1 045000 0 -.2192'7 - ,i -=J-! i . I' i • , I

; ! I' 2 1 + I .10715 + II' .7129'/ 0 'I' 04J65) 0 i.3Illm 0\-.1?'a4'7 - 1i J I I

: !" 2, - .10716 + 11 .16013 +_1 .~'il4 + 1 i~ + Ii ...~~"". 2

~TI·-·.'_--tl.,.- - 1 11 .'lIl89O - 1 ;, .66216:1'1 .~~i('1 -.! [' 2, + I .22555 01 .17391 0 i .12'736 0 j .91M - 1I •'I(Ua6 - 1

L ...._ ~~ 2.:....~l·22SS6 01 .27103 __~_±:_S'21 ui~~,J1 0

.'

.'

~ .'"

-.-----.--r..---r.. h•••-. ---.: - •••• _. 'T'" I ~.------~

"--~O>SI\.,* _~~~j ....~~-i.. ~"' _:_~~~_+-=~75 -t- .~.~.I !lEI ,'1 +: .4Il1:l - 1 : .4J~5 - I . .;.tJ6)7 - 1 Jl'T.J1 - l' )lI101l - 1

: PI"" + : .J8101l _ 1 ; .J45S'1 - 1 i.J211S - I! .J06l13 _ 1 : :30211 - ~: .' 2 _: .J81O'f - 1 : .J;H06 - 1.•·.4~~~:.~ •.~1~.-__1~.6~.~ 1t·> h_"~-:';';;;2- ~ :-:-»OI~ 0 .448~ 0 i .4115'7 0 i.~ 0 i

.' 2 +:..4lIOS4 0 .)551" 0 .J'/555 0 i Jl1!79 0 I .)OJ,s o!i ." 2 ..IoOll" 0 .46IlO, 0 SnS2 0 ! .75162 0 ' .11m + 1 i

::-2":~ :=.-:::. := ::= := '-1.' 2 ' .oooao .•00000 .00000 I .0lIl00 .0lIl00 I

+ ~ ;a;.m-~·-I-~:~~ ~'~'0ll62 ~"r 04SWJ 0-.~~::..~.1100 + 1 .•~19~ 0 04S019J 0' .2D3 0 .1:U"O - 4 I

: i.1109J + 1 ••16~ + 1 ,.263;/4 + I, .~ +_ 1 •• V.o&.4_+~

.~1 0.1871'1 0 .1<1471 0;.~-1 .1)224-5j

•.01\4 0 .1'102ll 0 ,11194 0: .55480 - 1 i .)OllJO; - 5 J.:U115 0 .~ 0 .)61)7 O· .42'Il!'/ 0.1 .4~21!2 0

1_ .

;-;x

&IlIA .0

",.

"" -, o4dCO O· o4&lI6Il 0 .~511" 0 .70528 0 .~18 O'

.250

O;.~j-u__ ....

: .CIlIODO.0lJ!l00

o . ,46111

.0lIl00

.0

[, 2

. "

T4ILI UKlETA IIU .422

alb • 1.0

EPSllOll . T0·---~~-·l----O;25l-·-o.;o·p.--O'1S --i'-,:citi,• B 1 r., 1 ~ • I.SZII) - 1 j~_ - I I .-.n - l.j14;o-:-i -:~'~-1lPi' : " ii, *'" II .~ - 1 ! .JI9lll - 1 ; .2Il710 - 1 i • -4',.~-1 !

I I . I !__•.'_2~,_~..-.J6/:IJ__5_-_1 , .43069 - .~ ; .51~.~6mS - 1 .~_:_~

II x.' +.lWoD96 0'.SJ9')2 0 ~ 0: .39251 0.J1896 0

r ,[. 2 + •.l~ 0 1.Ja)92 0 :... o;.~ o! ....151 0I I. I

01 ••" ... Oi.\4aOS+I'-t- j---

.C:O::D) : .OODDO.'~-- ".-"-'.-••• - -- •._-- ._-----_ +-"

0'-.)):<62 - 1 .24<>40 - 2 .11410 - I .0'1"00 - 6

TAlII.! xv IIIIl£TA I8ll ,1:>.)0

alb. " :'7

EPSILClll -;+,~T .00 -.. , .-. O.l: 0.50 O.~ r 1:00--1~-·~----"·---..I·-· t--.---- -+- t·..-- -------<,m .' 1 + .~Jl - 1: .4'076 - I ' .41m - 1 .40126 - II .40129 - 1: PI" ': " + o4IllJO - 1 .J71~ - 1 •.>41>'/4 - 1 .J2s'/E - 1, .:pm - I '

, .' ~ .401)0 - 1 j .4J676 - 1 ' ,4'If>:JJ - 1-' ,~1~ - 1 i .5'1'1S2 - 1 '1-- ,.__ .. ----+- - 'lilt" +' ."h)~J U.•51'1S6 0 .";'7'1< 0' .44145 0' .42li'll! 0

r .'., + ,J,;:YI'f 0 .J848s 0 .Js392 0 .)~ 0 .J0951 0

'.

(, " ' .7.llJ'/ - 1 .78'n4 - 1 .dOl.l~ - 1 .75010 - 1, .17785 - 4

I! .CXIJOO

: .QOOQO..-t-----

0'.10%9-4

o' .9~ - 5

: .CIlIODO

: .00000

O .•J75~

o .17186

; .oooao1 .ODOOO

o •.l71~

.0lIl00

.0lIl00

o .6OOIl2

.0lIl00

.90216 0 .IJ5"O' 1 ; .216015 + 1 .42950 + 1 ' .16065 + 2

.0lIl00

.'10212

,IQ507 + 1 • .lOSll4 + 1 . .1M6il+

.' 2

; II 2 .

..'.~ ,'I

: [I 2

oo

.~544 0 .1J~'''''''' 1 .. ;;\~~ ... 1 .42005'" 1; .12'-11(, ... tI.

.7)l:J4 - I •.-.~7'I~ - 1 04970J - I , .&19 - 1. .10<.51 - 5

"

. (I 2. +

......._.. _-- - --,..•_-~ ._.....:.~ , " 1 .lAJy;:n + 1 .~:N5O + 1 .1(51) + 1 .)'16'17 0 -.10560

" ~: + .t«lS4D C .55829 0 .~"flj52 O .•""'J17 - 1 -.1 S6n

--~.

~:"1

,I ~

.00000

.' .I<lOTJ

.1J.,'i) - 1 .l40~5 - 1 .295OJ - 1; .JIOI'I - 1

O' .15d08 0 • U'OO 0: .IUlI 0; .11;;21, 0

" 1

(I 2, +

.~1

.21444

o .18673 o .1:u9'f

o .10011

o .1)1716 - 1 .135"0 - 5

o 049S~ - 1 ' .262f>l! - 5

.' .ll!O7.l 0 .20160 o ,.<17~ 0 i .~J6 0: .22~'/_.1 ....1..._.....

o._.- .' 2

-_.j..2\444 o .2'11*11 o .•4:u03 o : 049282 o'

"2j .7)~ 0 .I\41b+l .1II7Jl+1 ,J684O+1 .125)4+4

" 2· - .167J.i 0 .1847', 0 .19'161 O'.~ 0!.aQ)ZJ 0" __ 1. _1.. _..1 1.... • __

O! .104l14 + 1....

.~167 - 1 .601., - 1t-

O' .4217) 0

o .72513o S13S4

.'''47 - 1 .~ - 1 .83571 - Ii .1807) - 4

TABLR IXIlIETA IIU .422

a/h • 1.0

0: .48.WI

.11105 - 1 .ll1tll - 1 • .uMJ> - 1 .....136 - 1

o .'14114.> 0 .IJO'19 0 .11?86 0;.". 0

.7'fll0 -

.421"14

.~:J07 + l' .a2J7' 1 .~51)5 O,.J4SllO 0 -.11'JQ.) "

.V594 0 ~ 0' .21819 0 .1)~ - 1 -.18/21 0

+ . -.21ot'l 0·-.:lE'176 - .•1),144 - 2 .I(l1J111 - 1 .6471~ - ~

+

+

+

(I 2 .

[, 1 .

.' 2 .

.'! "I 2

.•' ~l(I 1 !

" 2

i [I .i!,+-

AIJ'HA .5

-~ .._--..--_._--~ "1 + .0lIl00

(, 2 - + •167411

lilt,..

~-;s;u;- '+'-1 .00 I 0.25 I o.~ I O.75 ~ 1.00

~ ~ (' 11 •. .42'f1) - l' .424JJ - 11.41606 - 1 .40771 - 1: ~ - 1

r l 2' + .40010 - f .J6~79 - 1 .3'" - 1 .". - 1i .2JI9D - 1i

,4IYJ:H - 11 .440'6 - 1 ! .Jk1#fl - 1+-._ .. _._.. ..

.60051 0 .51%) 0;.46698 0 i .438"1tiI ,

,4.l174 0 •.l7I,J'J 0 i .34165 0; .3138' 0, .l.8974 0

fAIlLE IIUMA .1 Il£TA IlW. .J.U &A. .250

a/h • 1.0

I. :ls;~1+~~7;E;~~2:~ i·:~ll;~~·T.~~~~:II.J:-;;;~1-1I Pi", I', I' i ,,;, +: .3~121 - 1 •.l.lf,S9 - 1 : .J0335 - l' •.!7'J04 _ 1 .<1,011 _ 1i . • i ',.

~_~_~~-1..-; ..J~~~I ...:~~J-=-_1 ~ .!~1~"_-_~.~_-_~:~.:'~j-'IIX " 1 I +' .6'1"'-5 0, .5JI7~ 0: .45'/00 0 .4181) ,)i .J9594 0, ,..' I I , , :

I" 'I +! .)')~ 01 .)4,151 0: •.10512 0 .2'189ll 0, .2~ 0

, .'" - I .J~S'1~ 0' .4725'1 0 I .5874'1 0 .78574 0' .1;<6/;< + 1I 1 ----t-o - -- .. - -.+ .. - .• ---=-t--- _. --.--. . -- ...-1

'-1Ay " 1: + i-·14D72 J,-.I.J8'f5 - 1; .6S236 - 2, .'1'17.11 - 2:-.£)89. - 61

'

r I! Ii: ', .'" +: .J41S2 - I .~l5J - 1; .lJQ47 - 1 ,.14'08 - 11-.51855 - 6 II I' • : : ! Ii ; .' .! ~j .J4151.-=-j.~1~..::~.~7)- I ~ .~01'f - 1:-.469% - 5

i ~I i .' 1, + 'I .IJJJJtY/ + l' .21)41 + 1 '.92'J69 0' .Yt'2a6 01-.2981~-1-p'! I 1 '

: , .' <' +, .86117 0 .~ O'.J4QIl4 0 .14872 oi-."1I9 - 1I . It' I I

~.;; I -~..:.~~~1.,~:1~~.~_~.:.2.-l-.411~~_~~~ + 21

, ~' (,~ ,.0lIl00 !.4D7S.l .. 1 •.;0010 - 1 .~ - 1i .""-6~i !., 2 I + 1.21212 Oi .1655< J '.1~ 0 i·~;s,;7 - 1~ ,8'141 _IJ'

. • 1 I .' ~l -, .21212 "I .~~ 0 ,.Hl'IO 0: .J~15J 01 .JZlI" 0~. .L__, 1.-.- ...1 _

-S7-

IlIPllWIiI I,. oClII o.z O;,fl) eo" ,.~

II 1 • o849lI9 - , oIWfi - I ..an61. t ~., .... ,II 2 • .,ll31llG .. 1 .IlIlliOlI- • .-.. , •'itWlO. , ._., .,.. a .. olla1'79 .. , .JJ47i1fl - , o8'li'lJ ., , ..-., ,"·'i-- _._--- -, ".

fl ... •;a SI 1 .. .1Cl6W1 .. , .102$" 1 .9\I:I'l"P 0 •~1llI

"'2 + ~~/o AlDil ' ",.'0CW7 e ~ II .- t, ;, ..( '; '..' ,; ~ "

al ~ .. .~ 0 '.9'llIt6 ' 0: ~,~'" .'WtilI ... ....,'1' ,.

;a. SO , .. ~.ttlDa ,O-.~-l ·.,~_:-a .'lS:J&. a~.,

a' 2 .. ~'" ~ .. , .~". 1 .~.t ~'.,

&02 - "'iml- , 061170 - 1 ollilU1 .. , a!D679 .. , ........~

&0 1 + ~+1 '-111. I .~+1 ~ , -..JNf.,.' 2 • o2IIllfJI + , .15W1 ., oIImi 0 ~ 0 ...... '

I' 2 - .al718 • 1 .JIO'llI4 • I .H:IJ49 • I .~+1 .1Sl1•• '..- oJI''' • ,

~(, 1 + .0lI000 .16100 • I .a9165. 1 ~.,

I' a .. .~14 .. 0 olSY111J 1/ ,lW. 0 .~ 0 ..., t

.1 a · .:0413 0 .26OIIll 1/ .lIil1.u, 0 o2lM1 C ..... 0

~ .•AIJ!lfA .1. TABLE 1111

lIiITA IIU.l. .6JOalb .. 1.'7

.,IUII +,- .. 0.2~ 0.'" Ii, o.?S '.00

W.. , • .e1lW ~ , .&Il146O. , .Il26?S - " .816i11 '" , .B1:m - 1

I' 2 • 08'm - I .7ilI!8 • , .'n66a - I .~., .'l'6aIS • 1

I' 2 - 08,~ - I .~~1 o8'l8iO • 1 .9;oon - I .9'llI55 - I

';J .. I • .11S'O. 1 .tenlS. 1 ._'''4 0 .llO'UlI 0 .&lmV 0

II 2 , .~ 0 .a:r;u7 0 .a;o12 0 .~ 1/ olI0107 1/

., 2 - .ll9?lU 0 .96OlI6O 0 .~.,"~" .1~' I

-~1" 1 • .00llll0 .llllOIIO .0CGlID oGlIlllIO .GlllIl!9

s' a • .GlllllIO .GlllllIO .GlllllIO .ClOl!l!lD •aml!O

.. a - .lIlllIllO .CllIDIID .GlllllIO .CIlIlllID .--~

II , • ".1llIJO'I , I ..Ji"lQ + I .1972'r , I .~ o • .J69)l1 - S

cl 2 + .211:Q' I .14646 , , .'i1a>,) 0 04,J891 0 • 'ClIU'l - 'sc' a - .217U' I .Jt6S7· , o4llsn, I ,111469 t , .161l61 + 2

~~. 1 ,

.~1 1/ .11161l9 1/ .12490 1/ .6.168J - 1 oJ~· 7

II a + .<l4'1l102 1/ .111691 0 .12460 0 •~-I •'2~20 -6

I' 2. - o<l4llU 0 .7119"" 0 .J7~ 0 o4J60 0 04\lOIII2 0

.'

TASlZ IIIVll£rA tau ,630

alb • 1.'7

aA: .•",

/

,",Ul!l'.~

.. o.4~ o.§O 0.8 .,..fti u' , • ,--1 0851174 - , ~ .. , ~., ..... ,

U' 2 1- .t'CIlJ'.O - , •7'l2n6 - • .." ... .~., ...,.,.0 a · .~., ~-, .\WIt... t .~-, ... •

> ,0 , • .1:nw • , 01lll!O1 • 1 .\UZ$ 0 ~.o ... •.0 2 • oS'l'4llS 0 ~ 0 .7Mll1 0 .~ 0 0"," 0

I' 2 - .a7eG9 0 .9mJ 0 .~., 0Uil1G· • ..... ,..;x .0 1 • .0IIll00 ~ClllDIllI .iIll§GD Al!DJ ...

I' 2 • .0IllIlIll .ll:l/IIIO ~CIIllIlt3 olmllllll ...,0 a - .Clllllllll .lIlllllllI .Cl!lIllllI ocm!nll .-

-~10 1 .. .SIUO'1+ , ..JtoaJ + 1 .'esl8 .. I .'15'Il 0 .~'·4

I' 2 • .179'/11" 1 •'2Ila1 • 1 .7"" 0 ...J6tll& 0 ._-4I' 2 - .1797)' , .~,., 041171 • t .72t116 • , .1lIGIl4 • a

~I' , , .~1 0 .1,- 0 .1• 0 ~., .n.,.· •II a ,+ 0347'1'1 0 .1IIS63 0 .';U~ 0 .617J1) .. , oJIII'I. 5

I' a · 024'lll11 .'0 J10fA 0 J7J:l,l) 0 o6JM • ...,. 0

"

8'1IL.GII +.- .00 O."~ 0.50 o.?S 1,00

.1 ,I 1 • .A:I4lIi • 1 .~.I .1lJ177 - I .829?0 - 1 ,azffl· IPii

I' 2- • ,,&;nllQ - " .tlCH71 ~ 1 ,~-1 .777)0 ~ , .16224 • 1: ' .

.a6Il~ - 1,I·.;z. · $181 ~ I .-'i64 ~', .a'11~ - 1 ,'11)6) - I

~(I 1 • ,,~~. 1 .~ 0 ,'1~)ll 0 .Y)1II! 0 .'1:046 0

I' :t • .9"Ool/$ .. 0 .IlllJ19 0 .~1'!6 0 .82679 ~ .~911 0."II 2 · .~?JlJO 1/ ,W11.D (/ .1\lSJo. 1 .11658 • 1 .1))57' + 1

-;.. II I • ,..1(\ll8y O-~-I -.19567 - ~ •90761 - 2 oJOO7lI- - ,

., :z. • .79~ - 1 .~-1 .~-1 •)21711 .. I .J2'1J) • ,

I' c - .79:nf - 1 .107111 - 1 .1l1'l2J - 1 .•6J!iOJ9 ~ I • .1865$ - 4

_'JI I' 1 • .QlIJD7 • , 03'~+ 1 .18978 • 1 .7"30 .0 -.18/,41) 0

I' 2- • .m17., .1Clti~. 1 .~ 0 .11066 o -.;Dt\42 0

I' 2 - .17219' I .11IbIJaI • 1 JIJtTn ., .eli75/. " .14l9?'1 + 2

~t l I • .0lIll00 •17a!Plt • a ~·l .'7670 - II :,~-a

t' 2 + .1S13? 0 .1771& 0 ,17))2 0 .17070 0 .1€1#'l1ll 0

I' 'a - .tB1J8 0 .1a".s 0 .181\6) 0 .18'180 0 •las" 0

•

.,.. .'

~ +.- .00 0.2' o.lJO O.?S 1.00

.1 " •• "A15"'..... , .lU186 - ,. .82S ~ 1 .tJ2tJtT1 - I

~ ;,.1: ' ...... _~CO 2 • .....#1Im - 1 .'I9CO' - I .,..10 -I .74326 - , •72171 -,

.. a - ..... 1 ..,.,. - 1 "')49 -1 .9J'T.IO ~ I •-.am - 1------

;a s· 1.,

t .1GI64' I .'0108 • 1 .9S024. 0 .91~9 0 .avm 0

"It 2, ,+ .llJffl) 0 ~ 0 .8OIlJ7 1/ ,7'1m 0 .~9 0

~, l '- oS9'I'1S 0 -.•9'1160 0 .10'l8J + 1 .12i087 t 1 .1SlB1 • I'-:-'---"~T ~-~- -

-~cl 1 , • ,..'Zl44 ' 0 _.uz~ - , 08JJ86 - J .'lOO67 - 2 .sUllO - 6

. '

.9880S • 6-10 .. " :.4720J • I 041926 ,. 1 J)lY7 - , .1'195',. I

" .~-, • '0ll70 - 4I' 2 · '04721l6 _ 1 04'1413 • , 046m - ,- J»18 • •71YSl o -04947J • 1

-~ I' I , •tJJrJ#I + 1 I .1?27S • ,!

8' 2 + .1'lQ61:. + I .110166. 1 .674034 0 .lSS2/l o -.'1'1260 • I

I' ii - .17041 + I .2SJ68 + 1 ~~'6t '+.~'".~+2

._....~--- ._---'!fr Ii' I • .llOliDO .na'll - I .m~· I •.26'1llS-1 .ilflfiiJf. I

IIt' 2 ~ oZlTI? 0 .'lSTIO':; 1/ ... 0 1.'161ii 0 .171J7 0

••:2 - ~ 0 ..w.14 0 .26116 0 Io1JR19 0 .1llWilO It

IP8IUIl .,. .GII o...~ OOS 0.." 'A.i-- .---_... ,

Ii all 1 • .--99 -. .a»t6 •• .831'10 •• ~ .. , ..".. ,.. a .. 082752 - • ~-, ••." .. 1 076ln1" • .ftIla ...

sl 2 - .82752. t o85ll26 - , .lI'74'D .. 1 o~'" ...,... ,f--- r---- ..

;a SO , +1 .1CU401 + 1 :-mStZ 0 .~ 0 .~ 0 .JtIt1 0

.0 2 +1 .91\15) 0 ..~ 0 .Q86 0 .~ 0 .'IiD'I 0I

i

10 2 - .919Sl 0 .'Yi86? 0' .10616" 1 .t1ll6S. 1 •.,..,. t

-~1 0 1 + .1. O-~'" ro.,27m .. 2 .-... -.--,I' 2 .! olIl41Q1- 1 .'W?1 - 1 .""'iIi! .. , oJ2'DO - , ..._..,I' 2

I084163 - 1 .~-1 .Ila5Oll- , • 'I6ll11 - 1~-.-

-~I' 1 + '-+1 .J61J8 • 1 .1'W19. , .." 0-"" •U' 2 + .~+1 ...,., 0 04'" 0 .)U6J. a.....aO l - .144a4" 1 ""+1 .JS!IIIh • 1 .-., .~.a

~

~III 1 .' .QQIlIlO .112" • a ~-a 041.·. o41/li- •

ao a + .1fillIiI 0 .166JZ.. 0 .~ 0 .16_ 0 .- ..&02 - .1($)7 0 .~ 0 .1~ CO .,.,.. 0 .1_ •

•

-88-

~ 0<. ~ -1.00 -0.75 -0.50 -0.25 0 +0.25 +0.50 +0.75 +1.00

0' 1 0.618 0.663 0.813 1.163 2.435 1.163 0.813 0.663 0.6182 4.585 2.024 1.272 0.870 0.518 0.455 0.354 0.301 0.280

0.1 1 0.649 0.71'03 0.!'372 , 1.182 1.996 1.182 0.872 0.719 0.6491

0.643 0.484 0.381 0.324 0.2912 4.2f-.37 2.008 1.291: 'J.893

0.5 0.5 1 0.710 0.788 0.913 1.154 1.697 1.154 0.913 0.788 0.7102 3.728 1.922 1.296 0.938 0.704 0.547 0.436 0.366 0.317

1.0 1 0.740 0.810 0.921 1.132 1.645 1.132 0.921 0.810 0.7402 3.479 1.880 1.298 0.962 0.740 0.578 0.463 0.380 0.314

0 1 (').723 0.757 0.871 1.132 1.975 1.132 0.871 0.751 0.7232 :3.514 1.812 1.243 0.926 0.723 0.588 0.502 0.452 0.437

o. ~~,75 0.1 1 O.74E3 0.804 0.916 1.137 1.646 1.137 0.916 0.804 0.7482 3.272 1.784 1.247 0.941 0.744 0.611 0.525 0.472 0.441

0.5 1 U.SOI 0.854 0.944 1.112 1.449 1.112 0.944 0 • .954 0.8012, 2.864 1. 700 1.~;41 0.975 0.796 O.SGg 0.579 0.514 0.464

0 1 0.827 0.849 0.926 1.093 1.548 1.093 0.926 0.849 0.8272 2.470 1.551 1.17~ 0.965 0.827 0.732,0.672 0.637 0.625

.:. I0.1 1 C.D49 0.884 0.952 1.089 1.342 1.089 iO.952 0.884 0.849

2 2.314 1. :523 1.H30 0.979 O.F.347 0.754 0.692 0.651 0.622

o t"'r;: 0.5 1 ~.i .832 U• 91:5 (J .. 969 1.072 1.228 1.072 0.969 0.913 0.882.<:::0

2 12.058 1.457 1.169 0.998 0.880 0.795 0.731 0.682 0.640

1.0 1 0.895 0.921 0.971 1.062 1.229 1.062 0.971 0.921 0.8952 2.025 1.432 1.165 1.004 0.893 0.810 0.748 0.690 0.640

0 1 0.928 0.938 0.9'74 1.047 1.190 1.047 0.974 0.938 0.9282 1.552 1.2.59 L096 1.g96 0.928 0.881 0.851 0.834 0.828

:).1 1 0.941 0.957 0.987 1.039 1.106 1.039 0.987 0.957 0.941fa 1.480 1.239 1.096 1.004 0.940 0.895 0.862 0.839 0.820

0.125 0.5 1 0.954 0.969 0.992 1.029 1.065 1.029 0.992 0.969 0.9542 1.382 1.206 1.089 LOI0 0.954 0.913 0.881 0.855 0.832

1.0 1 0 •.962 0.974 0.997 1.020 1.069 1.020 0.997 0.974 0.9622 1.317 1.194 1.092 1.014 0.962 0.920 0.889 0.861 0.837

Table XXVIII Coefficients of Lateral LoadD1str1but1on(a/h = 1.7)

•

•

.'

-89-

~ ~ ~ -1.00 -0.75 -0.50 '-0.25 0 +0.25 +0.50 +0.75 +1.00

0 1 0.528 0.581 0.762 1.185 2.863 1.185 0.762 0.581 0.5282 .5.544 2.169 1.274 0.806 0.528 0.355 0.254 0.202 0.181

0.1 1 0.579 0.674 0.855 1.211 2.157 1.211 0.855 0.674 0.5792 4.947 2.146 1.309 0.854 0.569 0.394 0.289 0'.228 0.191

0.5 0.5 1 0.689 0.776 0.913 1.162 1.743 1.162 0.913 0.776 0.6892 3.971 1.984 1.316 0.930 0.680 0.508 0.393 0.316 0.258

1.0 1 0.740 0.810 0.921 1.132 1.649 1.132 0.921 0.810 0.7402 3.479 1 0 880 1 0 298 0.962 0.740 0.579 0.463 0.380 0.314

0 1 0.651 0.694 0.832 1.153 2.281 1.153 0.832 0.694 0.6512 4.213 1.951 1.262 0.887 0.649 0.494 0.399 0.345 0.326

0.375 0.1 1 0.704 0.776 0.907 1.159 1.760 1.159 0.907 0.776 0.7042 3.781 1 0920 1.287 0.929 0.698 0.540 0.441 0.377 0.336

0.5 1 0.785 0.847 0.941 1.118 1.477 1.118 0.941 0.847 0.7852 3.046 1.759 1.263 0.975 0.780 0.644 0.542 0.467 0.409

0 1 0.780 0.809 00912 1.111 1.730 1.111 0.912 0.809 0.7802 2.924 1.671 1.210 0.949 0.780 00667 0.593 0.551 0.539

0.1 1 0.822 0.867 0.948 1.104 1.402 1.104 0.948 0.867 0.8222 2.620 1.622 1.213 0.977 0.818 0.709 0.630 0.576 0.535

0.25 0.5 1 0.874 0.909 0.967 1.077 1.244 1.077 0.967 0.909 0.8742 2.165 1.498 10186 0.998 0.872 0.777 0.707 0.649 0.599

1.0 1 0.895 0.921 0.971 1.062 1.229 1.062 0.971 0.921 0.8952 2.025 1.432 1.165 1.004 0.893 0.810 0.748 0.690 0.640

0 1 0.906 0.919 0.964 1.058 1.262 1.058 0.964 0.919 0.9062 1.738 1.324 1.116 0.989 00905 0.848 0.811 0.790 0.783

0.1 1 0.930 0.948 0.983 1.047 1.124 10047 0.983 0.948 0.9302 1.610 1.292 1.115 1.004 0.929 0.873 0.831 0.800 0.774

0.125 0.5 1 0.952 0.966 00992 1.032 1.070 1.032 0.992 0.966 0.9522 0.424 1 0228 1.098 1.012 0 0 952 0.904 0.868 0.838 0.812

1.0 1 0.962 0.974 0.997 1.020 1.069 1.020 0.997 0.974 0.9622 1 0317 1.194 1.092 1.014 0.962 0.920 0.889 0.8f)1 0.837

Table XXIX Coefficients of Lateral Load Distribution(a!h = 1.0)

.. -90-

o

~ t Loterol Prestress

zFIG. I Multi-Beam Bridge With Lateral Prestress

..

• ---- ----

,...--Lateral· Prestressing Rod

----

FIG.2 Schematic Cross-Section of a Multi-Beam Bridge

FIG. 3 Corrugated Steer Sheet as Orthotroplc Plate

..y

..

=91·~

stringer. Moment of Inertia Ix

BecJm: Moment of Inert.la Iy

c

-92-

*=1.7 0.6*= 1.5• *=1.2

~=VCX-----• *=1.0 0.4~ = 3(K + 0.3330()/ --_.-

/f3 = 3 K(I - 0(12 )+ 0(

//

• // a0.2 I- .,

// {O K = f (t>I

00 0.2 0.4 0.6 0.8 1.00<

Fig. 6 Assumed Relations Between 0< and ~

"..

II

•

•

-93-

FIG. 7 Deformation Produced by Constant TWisting Moments

1" A;: + (Mxy +Myx)

-A.-(Mtty+My~ "'A~ -(Mxy+ My~

!tA.tIMxy+My~l

FIG, 8 Equivalent Corner Loading

-94~

•

• rI

C!.)R. i pI u 'p=-_1-... ._ 2c

Y zSimple support

....-

I

Ip

I)IL.

!! FIG. 9 Plate Strip of Infinite Width Under Line Loading

•

.. -b b

e"b - e'b-- v ~.

pt}c--.

u

II___00_ ..

o 01

e'+e"=2V" b(l - e')

--.__e._

YY.

FIG. 10 Bridge of Finite Width

-95-

45°-+-- --- --f--JJ--- -- -- -----~- . - ~-

1_ 2c =(8+h) Inches ~

--C'J

E• 0

OJCO

.. Y-o

..c.

..a-C-O)

0II

..c

FIG.11 Assumption for the Longitudinal Distributionof a Wheel Load

3.0

~.o .co~

:::J-0

~ 1.0 E====+===::::::::f~~o

lath = 1.00 I

-0o3 0.0 '-__---l- ..l--__~ ..J..______l_ _.J.._____l. _l

I\00'I

too

Ia Ih = 1.70 I

0.25 0.50 0.150.0

b/L = 0.500.3750.2.50.12.5

-0.50 -0.25-0.75

0.0 '--__-'- ...!.-__--L- ....L-__---l- --L-__--L -J

E -1.00

'0c..~o-J 2.0CJ­o

FIG.12 DislTibution of Load Applied at Center for oc= a.' . ..."

•~;

Ia/h =1.00 IbtL = 0.500.3750.250.125

2.0

co

...L-

~1.0~~~~~~~f~~i:::'~-1--..::::::r~""""i~~~~~~~o-g 0.0 ~__--'- ---'---__---' ---i- -'--__---'- ---'--__-'

.soL

E~ 2.0 I a Ih =1. 70 I

1.000.750.500.250.0-0.25-0.50-0.75

0.0~__--L.. -!.-__--I'--__-.L- -!--__---'- ---'--__----J

£ -1.00

FIG.13 Dis~ribu~ion of Load Applied at Center for ex: = 0.1

•-

Ia/h =1.00 IbIt = 0.500.3750.250.12.5

2.0

c

i1.0~~=~===:=::~i"....-.:::a'O=~-1----'=r""--=='~~~~~~3Co­

..L­V)

Q 0.0 '---__--L...,---__--'--__--JI.--__---L- ....l.....-__----1 --'--__-----'

co'-

2ro 2.0-JC&-o

0.500.3750.250.125

Ia/h =1.70 I

I\0OJ

•1.000.150.500.250.0- 0.2.5"-0.50- 0.75

0.0 --L.. --'-- L..--__~ ....L-__---' -.&.-__-----I

E -1.00

FIG. 14 Distribution of Load Applied at Center for 0(= 0.5

1.000.150.500.250.0

bIt =0.500.250.125

-0.50 -0.25-0.75

2.0 -

0.0 L-__-L L-__-l..-__--' --L-__---l -L-__---l

e -1.00

I\0\0I

FIG. 15 Distribution of Load Applied at Center for Isotropie Plate ex = 1.0

6.0

5.0 J 5.0

c c0 0.- :.L:-"- :J::J

. ..0 -0.

-c::: 4.0 Ia/h = 1.70 I Iajh =1.00 I4.0 'C:

..L-...L- V)\n .-Cl Cl

-0b/L = 0.50 b/L=0.50

"U0 00 3.0 3.0 0.....J 0.375 O.~75 . .....J

0 0.25 0.250L 0.125 0.125 '-.E OJ

...L-a 0--J2..0 2.0 --'

u... e.-o 0

-"-....l.- ec~.JOJ0-

u 1.0 1.0 ~u... u-CL.Q)QJ00

u u

0.0 0.0 I......0e -1.00 -0.75 -0.50 -0.25 0.0 0.25 0.50 0.75 1.00 0I

FIG. 16 Distri bution of Load Applied at Edge for ex:;:: 0

5.0

c:o

':.L::3

~4.0..1­Ul

a'"0oo :3.0

.-:J

Ialh= 1.70IbIt : O~50

0 ..3750.250.125

.. ,

I alh :: 1.00 1

.. bit = 0.500.3750.2.50,.12.5

, .~

5D. co

..1-

:J...0

4.01:U)

o\j\)

. . 0-3.0--1

0.0 L....-__---I.. --L- ~___..L. ...L_ L..____.L______' O~O

E -1.00 - 0.15 -0.50 -0.2.5 0.0 G.lS 0,50 0.75 1.00t.....o.....J

FJG..."11 Distribution of Load Applied at Edge for 0. :: 0.1

..

c c00

4.0 4.0~:.L: Ia/h = 1.70 I Iajh = 1.00 I::l..0 -0'C:: C-

..L-..L- (J)en --0 b/L = o.so b/L= 0.50 £:)

-0 3.0 0.375 0.375 3.0 -00.u 0.25 0.25 00 -J-J 0.125 0.125

0 0C-c.. (1)

.E 2.0 2.0-00-l ---J

e....- Ci.-a 0

..L- .J-

C1.0

cQJ 1.0 QJ

!J Ut.L-. CJ-CL- CL..OJ (lJ0 0u c..J

0.0 0,0..

C -1.00 -0.75 -0.50 -0.25 0.0 0.2.5 0.50 0.75 1.00

It-'0NI

FIG. 18 Distribution of Load Applied at Edge for ex. =0.5

..

1.000.750.500.250.0-0.25-0.50- 0.75

4.0c:0

:i:::J

..0c..

...L-V) 3,,0a-000~

0 2.0 bit 0.50r.. =OJ....-..0.250

--I 0.125t.&-o

C 1.0OJu.-

c.&-C&-alo

U 0.0 '--__-----I --1.. ---l... --1- -l.-__----'~______l....____---J

e -1.00

FIG.19 Distribution of Load Applied at Edge for Isotropie Plate ex =1.0

I.....oWI

.' •

p:. 47.700 # a~ Midspan

C p. Jj ~(

I I II I

a II I I I I II I

I II I I I I

II I I I I

I I II I, r

I I I II t

- f I I I20 I I I I

0 - I I I r I I~

0 "'- I I I I..J

0 I I~

I - - II I I I

"..../ rI -~

t II ....-r""" I

c I I ....... I oc = O.S" I /' t....... I I /' I-- ........

40 I oc = 0.1 I ./

C I0 OC=O

"/

..£- , /uC1J

, /

- \ I /c.L- \. I /QJ 60 \\

I

0 v" ""- /

" - I ./ ""-.J...--80 - ----Experimen~ar DePlecHons (Cenrerport Bridge)

---Theare~ic.al Deflec~ions (with E= 6.68 •106 psi)

FlG.20 Comparison Between Experimental and TheoreHcal Deflectionsfor Load Applied in Bridge Axis

I.....o.p-I

."

o

'.

p= 41~700# a~ Midspan

[1~fl, f?

--- -~Experiment-al DePlecHons (Cenrerport' Bridge)---Theore~ical DePlecHons (wit-h E=8.68 -10spsi)

20

-a

~ 00~

40c

c0--..L- 60uOJ-u-(J)

C)

80

100

I/"Y

//

/I /I ./L.../

;',./

--J_---

0< = 0.5oc = 0.1DC:: 0

IIIII'­-----'

'­----I,I---

I~

oVII

FIG.21 Comparison Between Experimental and Theoretical Deflectionsfor Load Applied at Edge

.. •

•.....o0',

981

Experimental Results (based on Deflec~ions)

Theoretical Resul~s for oc = 0 :(based on DePlecHons)(based on Moments)

4 5 6Beam Number

. 32

EoQJ

a::J.c.~ 2.4 t------;----+----+--\-----,

lLJ

~

. ..a 20 t------;----+----+--+------1---.,----r----.,A----~--____r-0

CIJ·C~ 16 r------+----+---------i;;;............-:,,-=---j----=~~~e:.---+---_+_--____t

U

"'0g 12 1-------+-----'-'--'--+-::-4''-----:zlfII'=-t-----'=::::.....-----I------'-==-+--='''''~~-::+----_+_--___I

--ItL- 9.3o B ~~::;;::~=__-~~_~---~---+--~:#---~~:::==........j8.0ID 83'---,W U u

-E 4 t-----+----+----+-----1I-----t-----+-------f------IOJ~~ 0 '--__....1-__--L.-__--.L-__---J~__~__.....L-__--L..__-.t

1

FIG. 22 Comparison Between Experimental and Theoretical Load Distrib\Jt~·v~

. Coefficients for Load at Bridge Axis

..

E}(perimen~al Resulf-s Cbased on DefiecHons1

Theoret-ical Results roroc: 0 :.(based on Deflec~ions)

(based on Momenrs)

6.'35.1

65 s..r -. 5.44.9 4.6

3.5-- .... - 2.9

I

4 5t-'

6 7 8 9 0'-l

Beam NumberI

32o

1

OJQ")

2 8 1-----+-------+---_f------------"""""""=--o7"-...=::::a,-=l--'-'~ .----1-=-----+-----.4

C(]Ju~ 40-

Eccu

a:l

-5 2B tt-2B_.3__-l-__-+-__--J ~__~_____L .l_______J

t3~

..c 24 t----J~--+----+--------l

-0cu

ot 20 ~~~::-I-1-9-.6--+----_____tyC

U

~ 16 ~-----+---~,_____;_--+--_'_-I___--__.__--__r__~______;,L..._,_---.____--____ta-J

c::; 12. I---~--+--~~~~:--+-__.r_r~--------+-

L

FIG. 23 Comparison Between Experimental and TheoreHcal Load DistributionCoefficients for Load at Edge

p= wheel loadall 2' 6'

..

4' 6' 3'

c tee 3'...1

0.:.c 4.0::J..0-c:...L..(f)

0

'-0 3~O00

--J

0£...

Jl2 2.0c.....JC&-o

..L.C 1.0Q).-u

(£:e-OJ0

CJ Q

I2.71 ftI f II l IIf't

I J II It I II II I II f IJ t II I I I I,

InPluence Line Par load DisrribtJ~ion

I CoePPicien~ for' Edge Beam:(cx.=Oj bIL=o.31s j o/h=1r70)

I I I II I J I

I II

f Areas:II A -, 2.73 +1.60 :1 - 6 50[1 - 2. .~ -. .

:. A2. =1.18 1°.93 ·3 =, 3..16'

t A" =.-0.85 +0.69 •~ ;; 2, ~'1'13 ~ ~ ~~,

t A4:;0.58 ~ 0.51 •. 3 = 1.64-

! 13..61I * 13.61 . ,al F'Pr S -= 3.9."10Q: 50.3 100 '

IIIIII

It-'ocoI

FrG.24 Determination of Load Carried by Edge Beam

.. •

1.00.5

l=5~1108

b) Edge Beam

/

L=~/36

a/h =1.70 Bri dge formed of 9. Beams

- L= 27 L/4-';36 108II /

"

I--

a!h=1-00 Bridge formed of 15 Beams

~-0

QJ

t.. 60c-oo

-0oo

...J 40

100 r­

EoQJ

a:l

i 80,- r- __ .!.Pr.:sent:-SpecifkaHon~_LU

OJQJ. -c3:

Beams

Beams

C&-

o 20 r--'-CQ,)

Fbridge in n. ~OJ0-

o0.5 1.0 ()( 0 0.1

Coerricien~' of Lateral Bending Stiffness oc0.1

..--

a) Cen~er Beam

f---~__ LPresen~SpeciPi~tion~_

_a/h =1.70 Bridge Pormed oP 9

f-- L=27 l= 54"/36 "!JOB

1/

I--

a/h=1.00 Bridge formed or 15

L= 2.~! L=5~/r- 36 - 108

t = Span 0

.1oOC 0

OJQJ

-Co~

c....0

2.0-'-cOJ\oJc...oJ

a...

E 100oQJ

CD

'­QJ

-'-

~ BOC,.)

~-0

Q).-t 60o

U

-0ao~ 40

FIG.25 Percent of Wheel Loads Carried by Center - and Edge Beams in a 21ft. Wide Bridgeof Variable Span

Ito-'o\0I

=110=

NOTATIONS

Roman Alpha.betc~,~"~ _~'._""--__

.i

\',;-

,Ii

•

..

£,,,,

Am~A*m2b

'j\' !iiiBm~Bm,Brn2cCx,CyCx.,CyC~Cm

D,Dx~Dyn DC"'~, ill

eJ "e ,e

EEx

EY

GhHiilm,i4rnifm' i.tmII x

K

Width of the be~us

Constants of integJ:'ationWidth of the bridgeConstants of illtegx:'ationLength of the applied line load (Fig. 10)Spacing of Btringer and beams (Fig. 4)Torsional rigidity of stringers and beams (Fig. 6)Constan.ts of integrationFle:x:ural ri.gidities of isotropic &: orthotropic platesConstants of integrationBase of natural logarithmEccentricity of applied load (Fig 0 10)Modulus of elasticityModulus of elasticity of material parallel to

x=di:r.ectionModulus of elasticity of material parallel to

y-directionShearing modulusDepth of plate or b~ams

Coefficient in equation 1Beam number i or·n beamsFu.nctions defined by equation 58Functions defined by equ.ation 62Moment of i.nertia per:' unit width of pla,teM~ment of inertia per unit width for a, section of

the orthotropic plate perpendicular to the x­dire.ctionand with respect to the y=axis

Moment of inertia per unit width for a section ofthe orthotropic plate perpendicular to the y­direction and with respect to the x~axis.

Functions defined by e.quation 58Functions defined by equation 62Modified roots of characteristic equa.tion, defined

by equations 37,460Cross=secti.on factor for torsional rigidity of

rectangular beams. .

Span of bridgeTerm of Fourier=seriesBending moments per unit length of sections of a

plate perpendicular to' x- and y-axes, respectively

-111-

NOTATIONS (continued)

t

•

•

Mxy,Myx

np (x,y)

rRX,RySxy

S~iS.J.m ,S~m

Slm'S4mU,vwWowlx,y,zYm

Twisting moments per unit length of sections ofa plate perpendicular to x- and y-axes,respectively 0

Number of beams in a multi=beam bridgeIntensity of the load at the point (x,y)

parallel to zTotal load uniformly distributed over the length 2cShear forces parallel to z-axis per unit length

of sections of a plate perpendacular to x- andy-axes, respectively 0

Roots of characteristic equations given by equation 36Boupdary shear forces defined by equation 260Coefficient of lateral load distribution defineq

by equation 68Percent 9f load carried by beam iFunctions defined by equation 54Functions defined by equation 63Coordinates of applied line load (Fig. 10)Vertical deflection of plateSolution of homogenous differential equationSolution of infinitely wide plate stripRectangular coordinates .Function of y, equation 33

~.

Greek Alphabet

a Ratio of bending stiffnesses (equation 3)

f3 Coefficient of torsional rigidity (equation 2)

f3x ,f3y Constants in equation 14

f30 Coefficient of torsional rigidity for a = 0

(} Abbreviation defined by equation 32

£ Abbreviation defined by equation 50

q,f"l, Abbreviation defined by equation 50

A Abbreviation defined by equation 50

-112-

NOTATIONS (continued)

~ ~pefficient of static friction

11 Poisson's ratio. Abbreviation defined by equation 3Z!i ' '"

, I

iI

~

/o

A9brevi,ation defined by equation 32

Parameter of torsion

Normal stress

I•

..

i" I1

-c Shear stress

.

•

"

~l13-

LIST OF REFERENCES.

1 0 "Standa.rd Specifications for Highway Bridges"(MSHO Specifications) Washington; the American

o Association of State Highway Officials, 1953

2. A. Roesli, A. Srnislova, C.E.Ekberg and W.J.o Eney"Field Tests on a Prestressed Concrete Multi-Beam Bridge"Lehigh University, Bethlehem, Pa., 1955 (In preparation)'

3. C. Massounett"Contributton au Calcul des Ponts a Poutres Multiples"Bruxelles; Annales des Travaux Publics de Belgique,1950, June, October. and Decernber$ page 377.

4. Y. GuyonIleale.ul des Ponts Larges A Poutres Multiples solidariseespar des Ent:cetoises"paris, Annales des Ponts et Chaussees, 1946, September,October, page 553.

5. C. Massonnet"Methode de Calcul des 'Ponts a poutres Multiples Tenantcompte de leur Resistance a la Torsion"Zurich, Publications, International Association forBridge and Structural Engineering, 1950, Vol.10, pp. 147-182.

6. P.B. Morice and G. LittlelII..oad Distribution in Prestressed Concrete Bridge Systems"London, Structural Engineer, 1954, Vol.32, Nc. 3, Ma+"ch,page 83-111.

7. E. Seydel"Ueber das Ausbeulen VOrl rechteckigen isotropen oderorthogonal anisotropen Platten bei Schubbeanspruchung"Berlin, Ingenieur Archiv, 1933, No.4, page 169.

8. S. Timoshenko"Theory of Plates a.nd Shells"New York, McGraw-Hill, 1940

9 • M.T 0 Huber"Die Theorie der kreuz\veise be~vehrten Eisenbetonplattenebst Anwendungen auf mehrere bautechnisch wichtigeAu£gaben Uber Rechteckplattenl!Berlin, Der Bauingenieur, 1923, page 354

-114-

10. M. T. Huber"Ueber die Biegung ei.ner Rechteckplatte von ung1eicherBiegungssteifigkeit in der L!ings - und Querrichtung"Ber.lin~ Der. Bauingenieur~ 1924~ page 2590

11. Ko rGirkmann"FUichentragwerke"Wi.en~ Springer Ver1.ag~ 1948 s Second Edition

120 S. Ti.moshenko"The Theor~y of Elastici.ty"New York, McGraw=Hill, 1951, Second Edition

13. A. Puche.r"Lehrbuch des Stahlbeton=bauses"Wien , Spri.nger Ver.lag~ 1949~ p.164.

140 A. Nadai"Elasti.sche pLatte.n"Berlin, Springer Verlag~ 1925

15. A. Pucher"Rechteck.p1atten mit zwei eingespannten RHndern"Berlin, Ingenieur Archiv, 1943, No. 14, pp. 246

16. H. Olsen and Fo Reinitzhuber"Die Zweiseitig gelagerte platte"Berlin ll Verlag Wilhelm Ern·st and Son ll 1950, Vol. I and II.

--------~

'.

•..... -... •. - .-.

.... - ~- "7"'"

, 1

/' \t~\tHfUNH!ERSj;'j

JAN ~a 1969A

~,!eJR£~,H.'l ~,,/.~.•---'