' Reinforced Concrete :

A n International Manual,

Written by a committee of
experts commissioned by UNESCO

Translated by
C.van Amerongen,M.Sc.,M.I.C.E.

London
Butterworths
 This manuai provides an engineering
code of practice for use primarily in the
developing countries.It has been
, drafted by an international working
group of experts who are leading
members of the European Committee
for Concrete and the American
Concrete Institute.
Developing countries do not require
highly sophisticated codes such as are
being produced in Europe and America.
W h a t is badly needed,and what this
manual offers,is a sound and simple
document which will be an efficient
guide to their growing engineering
cadres.
The prime importance of this manual
will be for those working in developing
economies but it should also be of
great interest to civil engineers and
architects in industrialised countries.
University and Technical College
lecturers,and in particular those teaching
in developing countries,will surely
welcome this initiative by UNESCO,
which w ill greatly assist their task.
Reinforced Concrete :
A n International Manual
Membership of the Committee
Chairman
Yves SAILLARD,Dr.Ing.,Technical Director of the Chambre
Syndicale Nationale des Constructeurs en Ciment Armé et
Béton Précontraint de France, Secretary to the Permanent
Comité Européen du Béton, 9,rue La Pérouse,Paris (le).
Members
Jean DESPEYROUX, Ing. Civil, Technical Director of the Société
de Contrôle Technique et d’Expertise de la Construction,4,rue
du Colonel-Driant,Paris (1“).
A. M.HAAS, Prof. Dr.Ing.,Department of Civil Engineering,
Technological University Delft, 25, Oostplantsoen, Delft
(Holland).
Telemaco VAN LANGENDONCK, Prof.Dr.Ing.,Escola Politecnica
de Universidade de Sao-Paulo,Caixa Postal, 30086,Sao-Paulo
(Brazil).
Franco LEVI, Prof.Dr.Ing.,Istituto Universitario di Architettura
di Venezia, Chairman of the Comité Européen du Béton and of
the Fédération Internationale de la Précontrainte,Corso Casale,
182, Turin (Italy).
Alan H. MATTOCK, Prof. Dr.Ing., Department of Civil En-
gineering, University of Washington, Seattle, Washington,
98105 (U.S.A.).
Jacques NASSER, Prof. Dr. Ing., Université Américaine de
Beyrouth,B.P. 2660,Beyrouth (Lebanon).
André PADUART, Prof. Dr.Ing.,Université Libre de Bruxelles,
49,square des Latins,Bruxelles-5(Belgium).
Raymond C.REESE, Consultant Engineer, Past President of the
American Concrete Institute, Member of the Committee
ACI-318 “Standard Building Code”, 743, South Byrne Road,
Toledo,Ohio 43609 (U.S.A.).
Stefan SORETZ,Dr.Ing.,Consultant Engineer,Klopsteinplatz, 1,
A. 1030,Vienna (Austria).
Secretary
Stéphane BERNAERT,MSc. (Illinois, U.S.A.), Ing. Civil, Pro-
fessor at the Ecole Spéciale des Travaux Publics, Paris, 9,rue
La Pérouse, Paris (le).
 THE BUTTERWORTH GROUP

ENGLAND Buttenvorth &C o (Publishers) Ltd

London : 88 Kingsway. W C 2 B 6 A B
AUSTRALIA Butterworth &C o (Australia)Ltd
Sydney: 586 Pacific Highway, Chatswood,N S W 2067
Melbourne: 343 Little Collins Street,3000
Brisbane: 240 Queen Street,4000
CANADA Buttenvorth &C o (Canada) Ltd
Toronto: 14 Curity Avenue, 374
NEW ZEALAND Butterworth &C o (New Zealand) Ltd
Wellington: 26-28 Waring Taylor Street, 1
Auckland: 35 High Street, 1
S O U T H AFRICA Buttenvorth &Co (South Africa) (Pty)Ltd
Durban: 152-154 Gale Street
First published in 1971
0Unesco 1967
English translation 0Unesco 1971
ISBN O 408 70175 7
Filmset by Photoprint Plates Ltd,Rayleigh,Essex
Printed in England by Camelot Press Ltd. Southampton
 PREFACE

In support of a number of technical assistance projects in developing

countries Unesco is carrying out extensive studies in fields where activities
appear to be most promising for the transfer oftechnologies to the developing
countries, for example in standardisation, engineering codes of practic,e,
and guides for the design and construction,or manufacture,of engineering
works.
The absence ofsuch standards,guides and recommendationsmay seriously
hamper economic progress. Foreign standards and codes are often used
although they may not be the most suitable owing to climatic conditions,
or the stage of economic and social development of the country concerned.
The drafting and development of the principal standards and codes is a
matter of great urgency and of prime importance to the economic progress
of developing countries.
There is an immense body of knowledge available in the industrial
countries in the field of standards and engineering codes, but in order to
apply this knowledge in developing countries,the available material must
be judiciously selected, sifted and modified by experienced engineers and
research workers,in cooperation with the internationalengineering associa-
tions and organizations for standardisation.This activity is an outstanding
example of technical assistance where existing knowledge can be applied to
development efficiently and with the least delay.
The developing countries do not require highly sophisticated codes,such
as are being produced in Europe and America. However, what is badly
needed is a sound and simple document which will be an efficient guide to
their growing engineering cadres.
This manual is one of the first attempts at providing an engineering code
for use in the developing countries.Reinforced Concrete: A n International
Manual has been drafted by an International Working Group consisting of:
Y.Saillard (France),Chairman,J. Despeyroux(France),A.M.Haas(Holland),
T.van Langendonck (Brazil), F.Levi (Italy), A. H. Mattock (U.S.A.),J.
Nasser (Lebanon), A. Paduart (Belgium), R. C.Reese (U.S.A.), S. Soretz
(Austria), S. Bernaert, Secretary.Unesco wishes to express its appreciation
and gratitude to all Members of the Group for their excellent work which
they have carried out with great expertise, efficiency and dispatch.
The document has been discussed extensively by this group of experts
who are some of the leading members of the European Committee for
Concrete and of the American Concrete Institute.
 CONTENTS

PART 1
Chapter 1 Scope And Object Of The Design 1
Calculations
1.1 Scope
1.2 Object Of The Design Calculations

Chapter 2 Units And Notation 3

2.1 Units
2.2 Notation

Chapter 3 Determination Of The Properties Of The Materials 16

3.1 Steel
3.2 Concrete

Chapter 4 Determination Of Safety 24

4.1 Principle Of Checking The Safety
4.2 Determination Of The Characteristic Loadings
4.3 Determination Of The Basic Strengths

Chapter 5 Determination OfThe Effects Of The Permanent Loads, 29

Superimposed Loads And Other Actions
5.1 Structures Composed Of Linear Members
5.2 Plane Structures

Chapter 6 Determination Of Sections 39

6.1 Normal Forces And Stresses
6.2 Tangential Actions And Stresses

Chapter 7 Constructional Arrangements 102

7.1 Agreement Between Constructional Arrangements And
Design Assumptions
 7.2 General Conditions Relating To The Reinforcement
7.3Arrangements Peculiar To Various Structural Members

Chapter 8 Preparation Of Designs 124

8.1 Calculations
8.2Drawings
8.3 Conditions Of Execution

Chapter 9 Execution Of Structures 128

9.1Requirements PertainingTo Formwork
9.2 Matters Pertaining To Reinforcement
9.3 Requirements Pertaining To Concrete
9.4Tolerances

PART 2
Chapter 1 Usual Values Of Superimposed Loads And Wind Actions 181
1.1 Preamble
1.2Definitions
1.3 Variable Superimposed Working Loads
1.4Wind Effects

Chapter 2 Determination Of Safety O n The Basis Of Probability 198

2.1 Preliminary Considerations
2.2 The PrinciplesOf The Probability TheoriesOf Safety
2.3 C.E.B./C.I.B. Semi-probabilityDesign Method
2.4 CharacteristicValues And Design Values OfThe Permanent
Loads,SuperimposedLoads And Other Actions
2.5 Characteristic Strengths And Design Strengths Of Steel
And Concrete
2.6 Method Of Checking The Safety
2.7 Relation Between The C.E.B./C.I.B.Semi-Probability
Method And The UNESCO Simplified Method

Chapter 3 Uniaxial Bending-Theoretical Analysis 21 1

3.1 Recapitulation Of The Fundamental Design Assumptions
3.2Analysis Of A Symmetrical Section Of Arbitrary Shape
3.3 Analysis Of A Rectangular Section

Chapter 4 Uniaxial Bending- Practical Design Calculations 231

4.1 Preamble
4.2Properties Of The Materials
4.3 Simple Uniaxial Bending
4.4CompositeUniaxial Bending With Compression
4.5 Composite Uniaxial Bending With Tension
Chapter 5 Analysis Of Tensile And Flexural Cracking 285
5.1 Preliminary Remarks
5.2 Analysis Of Cracking In Reinforced Concrete
5.3 PracticalChecking Of Cracking In Reinforced Concrete

Chapter 6 Calculation Of Flexural Deformations 311

6.1 RecapitulationOf The FundamentalAssumptions For The
CalculationOf Deformations
6.2 Determination Of The Basic Steel And Concrete Strains
6.3 General Calculation Oi Deflection Curves And Deflections
6.4 Simplified Calculation For Ordinary Buildings

Chapter 7 Shrinkage And Creep Of Concrete 320

7.1 Shrinkage
7.2 Creep

Chapter 8 Design Of Slabs And Plane Structures 329

8.1 SubjectAnd Field Of Application
8.2 Ultimate Limit State Corresponding To Failure By Exhaus-
tion Of The Flexural Capacity (FlexuralFailure)
8.3 Yield Line Theory
8.4 PracticalDesign Formulae For Simple Slabs
8.5 Practical Design Formulae For Flat-Slab Floors And
Mushroom Floors
Part 1
 1
SCOPE A N D OBJECT OF THE
DESIGN CALCULATIONS

1.1 SCOPE
The Manual applies to all reinforced concrete structures with the exception
of structures which are to be exposed to temperatures above 70°C and struc-
tures for which special design rules are necessary,namely:
(a) lightweight concrete structures;
(b)prestressed concrete structures;
(c)composite structures comprising reinforced concrete and structural
steelwork;
(d)concrete structures reinforced with rolled steel joists.
For residential buildings which are not exceptional in character and which
comprise not more than five storeys (i.e.,four upper floors) the building
owner may authorise the designer to employ a simplified design method,
e.g., a method based on the use of the modular ratio, provided that the
overall safety ofthe structure and the safety ofeach ofits component members
are checked to ensure that,in all circumstances and for any combinationsof
loads,superimposed loads and other actions,they are at least equal to the
structural safety that can be obtained by rigorous application of the code of
practice detailed in this Manual.

1.2 OBJECT O F THE DESIGN CALCULATIONS

1.2.1 N O T I O N OF ‘UNFITNESS’

All reinforced concrete structures or structural members should be so

designed and constructed that they are able, with appropriate safety, to
withstand all the loads,superimposed loads and other actions liable to occur
during construction and in use.
2
The object of the design calculations is to guarantee sufficient safety
against the structure being rendered ‘unfitfor service.’
A structure is considered to have become ‘unfit’when one or more of its
members ceases to perform the function for which it was designed,owing to
failure, buckling due to elastic, plastic or dynamic instability, excessive
cracking,excessive elastic or plastic deformation,etc.

1.2.2 N O T I O N O F ‘ L I M I TS T A T E ’

To each of the conditions in which a structure becomes ‘unfitfor service’

corresponds a particular state called a ‘limitstate’.These limit states are
respectively :
(a)the ultimate limit state (failure);
(b)the limit state of instability;
(c) the limit state of cracking;
(d)the limit state of deformation;etc.
The basic idea of the design method embodied in this Manual is to con-
sider each limit state respectively and to check that,for each of these limit
states,all the members of the structure as well as the structure as a whole are
able,with appropriate safety for each of them, to withstand all the loads,super-
imposed loads and other actions liable to occur during construction and in use.

1.2.3 GENERAL DESIGN PROCEDURE

The design method embodied in this Manual comprises the following

successive stages:
determination of the safety for each condition in which the structure
becomes unfit,i.e.,for each limit state;
determination of the effects of the loads, superimposed loads and other
actions, i.e.,determination of the internal forces for the structure as a
whole and for each of its members;
dimensional design of the sections for each limit state.
 2
UNITS AND NOTATION

2.1 UNITS
The system of measurement and units is the decimal metric system with
six basic units,as adopted by the General Conference on Weights and Mea-
sures and known as the ‘InternationalS.I.System’.
However,under the code of practice of this Manual it will be permissible
to use the ‘metre/kilogramme-force/second’ system on an interim basis.

2.1.1 B A S I C UNITS

The six basic units of the ‘InternationalS.I.System’are:

unit of length: the metre (m)
unit of mass : the kilogramme (kg)
unit of time: the second (s)
unit of electric current intensity: the ampere (A)
unit of temperature: the degree Kelvin (K)
unit of light intensity: the candela (cd)

2.1.2 S E C O N D A R Y UNITS

Of the secondary units the following more particularly concern reinforced

concrete design.
The unit offorce is the newton (N), this being the force which imparts to a
mass of one kilogrammean acceleration of one metre per second per second.
The unit of work and energy is the joule (J), this being the work done by a
force of one newton whose point of application is displaced a distance of one
metre in the direction of the force.
1 J = 1 N.m
The unit of pressure and stress is the pascal (Pa),this being the uniform

3
4
pressure which, acting upon a plane surface of one square metre, exerts a
total force of one newton perpendicularly to that surface.
1 Pa = 1 N/m2
A unit associated with the pascal is the bar,which is equal to pascal:
1 bar = Pa = N/m2.

2.1.3. R E L A T I O N S B E T W E E NT H E S.I. U N I T SA N D T H E
U N I T S O F THE ‘ M E T R E / K I L O G R A M M E - F O R C E /
S E C O N D ’S Y S T E M

A kilogramme-force(orkilogramme-weight)is equal to about 9.8 newtons

(9.8N)or 0.98 decanewton (0.98 daN):
1 kgf = 9.8 N = 0.98 daN,and conversely:
1 daN = 10N = 1.02kgf.
A decanewton therefore corresponds within 2% to a kilogramme-force.
A kilogramme-force (or kilogramme-weight) per square centimetre is
equal to 0.98 bar:
1 kgf/cm2 = 0.98 bar and conversely:
1 bar = 1.02kgf/cm2.

2.2 NOTATION
The notation uses Roman capitals and small letters,as well as small Greek
letters,with or without the addition of indices or subscripts.
The use of Greek capitals is not recommended.O n the other hand,the
use of small Greek letters is considered to Fit in with the tradition of standard
text-books on strength of materials and theory of structures and to be
compatible with the possibilities of present-daytypewriters.

2.2.1 S M A L L R O M A N LETTERS

These denote: lengths,forces and moments per unit length, and external
forces distributed per unit area.

2.2.2 R O M A N CAPITALS

These denote: geometrical and mechanical properties of the cross-sections

of prismatic members (areas,static moments, moments of inertia,section
moduli, applied external forces (total distributed forces or concentrated
forces)and their moments,and longitudinal strain moduli of materials.

2.2.3 S M A L L G R E E K LETTERS

These denote: stresses, strains, angles and slopes, and dimensionless co-
efficients.
 5
2.2.4 I N D I C E S
Compression is distinguished from tension by the addition of the prime (’)
to denote compression.
The symbols for characteristic loadings and basic strengths to which the
analysisand design calculations relateare provided with a bar (-)over the top.
In practice,however, the prime and the bar may be omitted in all cases
where there is no possibility of error of interpretation.More particularly,
there is no need to usc the prime when tension and compression are dis-
tinguished by the algebraic sign of their numerical value in the calculation:
+ for tension;
- for compression.

2.2.5 SUBSCRIPTS
The following subscriptsare used:
a -to denote the steel;
b -to denote the concrete;
c -to denote the critical state of buckling;
d -to denote the bond between concrete and steel;
e -to denote the (apparent or conventional) elastic limit of the steel;
m -to denote mean values;
r -to denote the failure characteristics of steel and concrete;
u -to denote the ultimate limit state (limitstate of failure)of a reinforced
concrete member.
Wherever there is no ambiguity the use of multiple subscripts should be
avoided.

2.2.6 P E R M I T T E D E X C E P T I O N S
The code of practice in this Manual allows some exceptions.These concern
the use of the following symbols,sanctioned by usage in the majority of
countries:
(a) diameter represented by 4 (instead of a small Roman letter);
(b) crack spacing represented by Al (insteadofa single small Roman letter);
(c) modular ratio (ratio of the moduli of elasticity of steel and concrete)
represented by m or n (insteadof a small Greek letter).

2.3 MAIN NOTATION U S E D IN THE CODE AND

THE MANUAL
2.3.1 R O M A N CAPITALS

A = cross-sectionalarea of reinforcement
A = cross-sectionalarea of main tensile reinforcement
6
A, = cross-sectionalarea of each individual layer of connector reinforce-
ment (ortransverse reinforcement)
A’= cross-sectionalarea of main compressivereinforcement
B = cross-sectionalarea of concrete
B = cross-sectionalarea of concrete;
more particularly : cross-sectionalarea of concrete in tension
B’= cross-sectionalarea of concrete in compression
Bi = cross-sectionalarea of the core of a member with binding reinforce-
ment
C = cement content
C = cement content (weight of cement used in making one cubic metre of
concrete)
E = strain modulus ofa material (or more specifically:modulus ofelasticity)
E, = longitudinal strain modulus (modulusof elasticity)of steel íexcep-
tion in common use)
Eb = longitudinal strain modulus of concrete
Ebo= longitudinal strain modulus of concrete for instantaneous (or
rapidly variable) loads
EA, = longitudinal strain modulus of concrete for long-term(sustained)
loads
F = load acting in any direction
G = permanent load,dead load
G, = characteristic value of a permanent load in the general case
G, = mean (oraverage)value of a permanent load in the general case
GL = characteristic value of a permanent load in the particular case
where a decrease in this load could endanger the stability of the
structure
Ck = mean (or average)value of a permanent load in the particular case
where a decrease in this load could endanger the stability of the
structure
other meaning:
gravel content
G = gravel content (weight of gravel used in making one cubic metre of
concrete as placed)
H = horizontal reaction
I = moment ofinertia (second moment of area) of a section
I, = moment of inertia of a section about any reference axis O,
I, = moment ofinertia ofa section about a reference axis U,perpendicular
to o,
I, = moment of inertia of a homogeneous section (uncracked:‘statei’)
M = bending moment
A4 = bendingmoment
M, = bending moment when the first cracks appear (‘stateI’)
M,,= difference between the total bending moment M and the partial
bending moment A 4 1
 7
M, = additional moment, used in analysing the buckling of a com-
pression member
M,,= additional moment,in the principal direction x, in analysing the
buckling of a plate
M,, = additional moment,in the principal direction y, in analysing the
buckling of a plate
M, = upper limit value adopted for bending moment in failure analysis
M u = bending moment at failure (ultimatemoment)
N = normal (ordirect) force
N = normal force (genericterm)
N, = resultant tensile force in main tensile reinforcement
N b= resultant compressiveforce in concrete
N; = component of normal compressive force,in principal direction x,
for a plate loaded parallel to its middle plane
Ni,= component of normal compressive force,in principal direction y,
for a plate loaded parallel to its middle plane
N u = normal force at failure íultimate state)
P = vertical load
Q = superimposedload,live load íthe symbolis also used to denote a loading
of random character)
Qk = characteristic value of a superimposed load (orof a loading other-
wise imposed) in the general case
Q;,= characteristic value of a superimposed load in the particular case
where a reduction of this load could endanger the stability of the
structure
Q, = most unfavourable value of the superimposed load (or loading
otherwise imposed) with 50% probability of being exceeded (in
the direction of abnormally high values)once during the anticipated
service life of the structure
Q; = most unfavourable value of the superimposed load (or loading
otherwise imposed) with 50% probability of being exceeded (in
the direction of abnormally low values) once during the anticipated
service life of the structure
Q* = design value of superimposed load (orloading otherwise imposed)
las conceived in the CIB/CEB semi-probabilisticmethod)
R = bearing reaction,reaction at support
R = bearing reaction in any particular direction
S = loading íin the most general sense)
S, = characteristic loading corresponding to the permanent loads
S,, = characteristic loading corresponding to the variable superimposed
working loads
S,, = characteristic loading corresponding to the dynamic superimposed
working loads
S, = characteristic loading corresponding to the effects of shrinkage,
creep and temperature variations
8
S, = characteristic loading corresponding to superimposed loads from
wind,snow and seismic effects
other meaning:
static moment,first moment of area of a section
S = static moment of a section
S, = static moment of a section about any reference axis O,
T = shear force
T = shear force (genericterm)
T, = contribution of the transverse reinforcement to the shear strength
(Morsch’sterm)
&= contribution of the concrete in the compressive zone to the shear
strength
T, = shear force at failure (ultimatestate)
other meaning:
overturning force
T = overturning force exerted on a structure by wind
other meaning (exceptionin common use):
temperature
U = lifting force
U = centrally acting lifting force exerted on a structure by wind
I/ = vertical reaction
other meaning (exceptionin common use):
volume
V = volume of an aggregate particle
other meaning (exceptionin common use):
velocity
V = wind velocity
W = section modulus

2.3.2 R O M A N S M A L L L E T T E R S

a = transverse dimension of a concrete section,or of a mesh in a grid or

lattice,or (moregenerally) of a side of a rectangle
a = transverse dimension of a concrete section (inmost cases ‘a’denotes
the largest transverse dimension)
a = edge length of a test cube
a = larger side of the area of application of a concentrated load
a = larger side of a mesh in welded fabric reinforcement
b = transverse dimension of a concrete section,or of a mesh in a grid or
lattice,or (moregenerally) of a side of a rectangle
b = transverse dimension of a concrete section (inmost cases ‘b‘denotes
the smallest transverse dimension)
b = side of square section
b = width of the section of a rectangular beam,floor slab or compression
flange of a T-beam
b = smaller side of the area of application of a concentrated load
 9
b = smaller side of a mesh in welded fabric reinforcement
b, = effective width ofthe compression flange of a T-beam
b, = fictitiouswidth ofthe rectangular section equivalent to a section of
any shape
bo = width of the web (orrib) of a T-beamat the level of the median axis
b,, = width of the binders (orlinks)forming the transverse reinforcement
in a member subjected to torsion
bo = thickness of the wall of a hollow member
b, = width of a haunch or splay or chamfer
b, = fictitiousdesign width of the rib of a T-beamprovided with chamfers
b‘ = width of the tension flange (or enlarged bottom of web) of a T-beam
c = concrete cover to steel
c,, = horizontal cover
c, = vertical cover
d = distance
d = distance from centroid of main tensile reinforcement to the extreme
fibre in greatest tension or least compression; or: distance from
centroid of a tensile reinforcing bar to the nearest concrete face
’ (calculationof anchorage length)
d’ = distance from centroid of main compressive reinforcement to the
extreme fibre in greatest compression
other meaning :
diameter of a circular section
d, = mean for average)diameter or fictitious thickness of a member
other meaning:
dimension of aggregate particles
do and d,, = minimum and maximum dimensions characterising the
class of an aggregate
e = eccentricity of a normal (ordirect) force
e = eccentricity of the normal force
e, = design value of initial eccentricity of the normal force
e, = design value of additional eccentricity of the normal force
other meaning:
distance
e = distance from centre of curvature of an anchorage based on curvature
to the nearest concrete face
e, = distance from centre of curvature of a curved bar to the concrete
face situated on the side to which the ‘radial’force (dueto the curva-
ture) is directed (unbalancedthrust)
f = deflection
f, = total instantaneous deflection
f, = total long-termdeflection
fi = partial deflection attained before cracking develops (‘stateI’)
fir = partial deflection which occurs after cracking develops (‘stateII’)
flI = limit value of the total deflection (‘stateII’)
h = depth (or height)
ho = depth of a slab or of the compression flange of a T-section
h, = depth of a haunch or splay or chamfer
10
h, = total geometric depth of the cross-sectionof a beam (ortotal thick-
ness of a slab or a plate)
h' = distance between the centroids of the main reinforcements
h" = distance from centroid of compressive reinforcement (or reinforce-
ment in greatest compression) to the tensile fibre (or the fibre in
least compression)
h,, = depth (or height) of the binders (or links) forming the transverse
reinforcement in a member subjected to torsion
h = difference in level between the base of a structure and the ridge of
the roof (fordetermining effects of wind)
i = radius of gyration
i, = radius of gyration of a section about the principal axis normal to the
plane of bending
i, = radius of gyration of a section about any reference axis x
k:see K
1 = length
I = clear span of a beam
1 = clear length of a structural member
1, = effective length (withregard to buckling) of a structural member
1, = straight bond length in tension
1, = straight bond length in compression
1, = effective bond length
m = bending moment per unit width of slab
mi = resisting moment per unit width ofslab,corresponding to reinforcing
bars parallel to the direction i
Note: if it is necessary to indicate the sign of the moment, the
addition of a prime (mi)denotes a negative moment
rn, = normal bending moment acting at a yield line per unit length of
that line
m, = torsional moment acting at a yield line per unit length of that line
m, and m,, = principal bending moments per unit width of slab at a
particular point of the slab
other meaning:
modular ratio in conventional reinforced concrete design
(exceptionin common use)
(thesymbol n is also used to designate the modular ratio)
n = modular ratio in conventional reinforced concrete design
(exceptionin common use)
(thesymbol m is also used to designate the modular ratio)
other meaning:
number or quantity (exceptionin common use)
n = number of bars making up the reinforcement
n = number of rigid elements into which a slab is divided by yield lines
p = perimeter
p = perimeter of the cross-sectionof a reinforcing bar or group of bars
p' = perimeter of the critical area in the analysis of a slab for punching
shear
q = superimposed distributed load per unit area or unit length
 11
r = radius
r = radius of a circular section
Y = radius of curvature of a structural element
r = radius of curvature of the axis of a bar
s = tangential force per unit area
s = tangentialforce per unit ofdeveloped length ofthe wall of a torsionally
loaded member
other meaning:
standard deviation
t = spacing of connector reinforcement
t = spacing of two consecutive layers of connector reinforcement
other meaning:
(exception in common use) time
At = interval of time
u = distance to centroid of a section
u = distance from the extreme fibre in greatest tension (or least com-
pression) to centroid of section
u’ = distance from the fibre in greatest compression to centroid of section
w = width of a crack
w = width of a crack
x = co-ordinate
x = distance from neutral axis to the face in greatest compression of a
member subjected to bending
xi = ordinate of any particular fibre in the section of a member subjected
to bending,with reference to the neutral fibre of that member
y = co-ordinate
y = depth of the rectangular diagram used in the simplified flexural
design method
z = lever arm
z = lever arm of the internal forces which form a couple to resist the
bending moment

2.3.3 G R E E K CAPITALS

A = change in quantity,interval (exceptionin common use)

AZ = crack spacing (distance between two consecutive cracks)
At = interval of time
= elastic strain of concrete produced by changes in the intensity of
the applied load

2.3.4 GREEK S M A L L LETTERS

ci = coefficient
ci = coefficient of thermal expansion
12
CI = coefficient to take account of dynamic increases in the superimposed
working loads
M, = coefficient representing effect of thickness of a member on creep of
concrete
CI, = coefficient representing effect of thickness of a member on shrinkage
of concrete
(also other uses as a coefficient,more particularly in 6.2.4and 9.4.2
of Part 1)
other meaning:
angle
CI = angle of inclination of connector reinforcement with respect to the
plane on which the tangential action is exerted,
or: angle of inclination of transverse reinforcement with respect to
centre-lineof member
CI = inclination of a roof (anglewith respect to the horizontal)
ß = coefficient
fl = coefficient representing effect of load arrangement (calculation of
deflections)
ß, = coefficient representing effect of concrete composition on creep
ß, = coefficient representing effect of concrete composition on shrinkage
ß = ratio between external normal force and normal resisting force
developed by the concrete section of a column (calculation of
minimum reinforcement percentage) (alsoother uses as a coefficient,
more particularly in 6.2.4of Part i)
y = factor of safety
Y = overall factor of safety
YWI = reduction coefficient for the strength of a material
ysteei = reduction coefficient for the strength of steel
yconcrete= reduction coefficient for the strength of concrete
YS = amplification coefficient for a loading or its effects
6 = relative mean square deviation (coefficientor variation)
E = strain of a material
E, = tensile strain of steel
E, = tensile strain of steel corresponding to the beginning of yielding
E: = compressivestrain of steel
&b = tensile strain of concrete
E; = compressivestrain of concrete
EO = maximum compressive strain of concrete
E;, or E: = instantaneous elastic compressive strain of concrete
AE; = variation of the elastic compressivestrain ofconcrete caused by
a variation in load intensity
Ebi = instantaneous plastic compressive strain of concrete
&gm = long-termplastic compressive strain of concrete
= creep of concrete (finalcreep coefficient)
&br = shrinkage of concrete (finalshrinkage coefficient)
c = coefficient
= coefficient representing influence of age at loading on creep of con-
crete
 13
8 = relative shear force
f the expression ‘reducedshear force’is alternatively used)
other meaning:
angle of rotation
û = angle at centre of curvature of a bent bar
IC = coefficient
IC = coefficient applicable to the coefficient of variation and depending
on the probability,accepted a priori,of obtaining test results which
fall short of the characteristic strength or of having a loading which
exceeds the characteristic value of the loading
IC, = regional coefficient (determinationof wind effects)
IC, = site coefficient (determinationof wind effects)
R = slenderness ratio of a structural member fR = l/i)
,u = relative bending moment
íthe expression ‘reducedbending moment’is alternatively used)
other meaning:
percentage of openings
,u = percentage of the area of openings in the total wall area of a building
(determinationof wind effects)
v = relative normal force
íthe expression ‘reducednormal force’is alternatively used)
other meaning :
Poisson’sratio
v = Poisson’sratio
other meaning:
percentage of cement paste in concrete (exception in common use; per-
centages are otherwise generally designated by the symbol w)
v E + c = percentage of cement paste in concrete
other meaning:
coefficient characterising wind actions
v o = coefficient characterising external wind actions on a structure
v 1 = coefficient characterising internal wind actions on a structure
n: = the number 3.1416....
w = percentage
w = mechanical percentage of longitudinal tensile reinforcement
wp = mechanical percentage of longitudinal tensile reinforcementstrictly
1 I I
necessary for resisting the fictitiousmoment M + 1.5 TI -0.5 N’
(shearanalysis)
wo = geometric percentage of longitudinal tensile reinforcement
wlo= geometric percentage of the ‘tie-bars’ in a member loaded in torsion
a, = mechanical percentage of connector reinforcement (or transverse
reinforcement)
wlo= geometric percentage of connector reinforcement lor transverse
reinforcement),
or:
geometric percentage of the binders forming the transverse rein-
forcement in a member loaded in torsion
=
w’ mechanical percentage of longitudinal compressive reinforcement
14
ab = geometric percentage of longitudinal compressive reinforcement
ai = geometric percentage of transverse binding reinforcement
other meaning:
specific gravity
p = coefficient
pt = coefficient representing the influence of time on the shrinkage and
creep of concrete
o = normal (ordirect) stress;
where necessary,a prime (’)may be added to the symbol in order to
denote compressive stress
om = mean strength of a material
ok = characteristic strength of a material
o* = design strength ofa material (CEB/CIB semi-probabilisticmethod)
o, = tensile stress in steel
o, = yield point (apparentelastic limit)for ordinary reinforcing steel
oo.2= 0.2% proof stress (conventional elastic limit) for cold-worked
reinforcing steel
oOm = mean tensile strength (elasticlimit)of steel
oak = characteristic tensile strength (elasticlimit)of steel
o,* = design tensile strength (elastic limit) of steel íCEB/CIB semi-
-o, = probabilistic method)
basic tensile strength of steel (UNESCOsimplified method)
o: = compressive stress in steel
o* = design compressive strength (elastic limit) of steel (CEB/CIB
:
semi-probabilisticmethod)
3, = compressive steel stress to be used in analysing the ultimate
-ot = basicstrength of a section (UNESCOsimplified method)
strength of transverse reinforcing steel (UNESCOsimplified
method)
ob = tensile stress in concrete
o. = tensile strength of concrete at age of 28 days
oj = tensile strength of concrete at age ofj days
obm = mean tensile strength of concrete
obk = characteristic tensile strength of concrete
:o = design tensile strength of concrete íCT.B/CIBsemi-probabilistic
-ob = basic method)
tensile strength of concrete (UNESCOsimplified method)
o; = compressive stress in concrete
ob = compressive strength of concrete at age of 28 days
o: = compressive strength of concrete at age ofj days
ob, = mean compressivestrength of concrete
oLk = characteristic compressivestrength of concrete
o;* = design compressive strength of concrete íCEB/CIBsemi-probabi-
listic method)
o; = basic compressive strength of concrete (UNESCO simplified
method)
obo = limit strength of concrete in a member under uniaxial compression
(UNESCOsimplified method)
 15
o; = Euler stress in a strut or column (bucklinganalysis)
oEx = Euler stress,in the principal direction x,in a plate
a;, = Euler stress,in the principal direction y,in a plate
o1 = largest extreme stress in a multiple state of stress
o2 = smallest extreme stress in a multiple state of stress
o3 = intermediate stress in a multiple state of stress
z = tangential stress (orshear stress)
z,,= bond stress between concrete and steel
rd = limit value of anchorage bond stress
zdl= limit value of flexural bond stress
I)= coefficient
$ = basic creep coefficient
I)= coefficient of friction between steel and concrete
I)= a quantity occurring in the expression forthe maximum diameter of
reinforcing bars (analysisof cracking)
4 = diameter ofreinforcing bar
I)ijor c # ~=~ angle
~ formed by the yield lines i and jin a slab
 3
DETERMINATION OF THE
PROPERTIES OF THE
MATERIALS

3.1 STEEL

3.1.1 D E F I N I T I O NO F T H E R E I N F O R C I N G B A R S U S E D

The reinforcing bars used are classified into four categories: plain bars,
deformed bars,welded fabric,rolled steel sections.
Plain bars are generally rolled from mild steel or from medium-tensile
steel.
Deformed bars (developinghigh bond strength in virtue of projections or
indentations) are generally rolled to a special geometrical profile;they are
of medium-tensile (or sometimes high-tensile)steel whose properties are
obtained either by appropriate composition (ordinary steels) or by cold-
working involving twisting or cold-drawing(cold-worked steels).
Guarantees must be given for these bars, more particularly with regard
to the geometrical and mechanical properties to be adopted in the design
calculation.These guarantees should be furnished by the manufacturers
and be checked by the representative of the building owner.
Fabric reinforcement generally consists of medium-tensile drawn steel
wires.It requires guarantees similar to those giveii for deformed reinforcing
bars.
Structural rolled steel sections can permissibly be used as reinforcement
subject to special justifications; the same applies to composite flexural
members having the tensile flange and the web made of steel, while the
compressive flange is of concrete. In the absence of regulations applicable
to this type of construction, the necessary justifications may consist of
experimental checks,comprising loading tests to failure, according to a
procedure agreed with the building owner.
16
 17
3.1.2 D E F I N I T I O NO F B A R D I A M E T E R S U S E D

The following bar diameters,expressed in millimetres,may be used :

4 5, 4 6,4 8,4 10,4 12,4 16,4 20,4 25, 4 32,4 40
The ten diameters indicated,forming a series serving as the basis for the
standardisationof reinforcing bars,have the important advantage that they
can be distinguished from one another by visual inspection on the site.
Besides,the cross-sectionalarea corresponding to each diameter is approxi-
mately equivalentto the sum ofthe cross-sectionalareasofthe two preceding
bar sizes. This facilitatesall combinations.
Five other diameters (414,q5 18,4 22, 4 28,4 30)are permitted,but it is
strongly recommendednot to use them,so as to avoid any possible confusion
with the next larger or smaller bar sizes on the site.

3.1.3 M E C H A N I C A L R E F E R E N C E P R O P E R T I E S O F THE S T E E L

Except in special cases,the only mechanical reference properties of the steel

are:first, the elastic limit;second,the stress-straindiagram up to a strain
of 1 %.
The elastic limit of the steel-apparent elastic limit for mild steel or
ordinary steels,0.2%proof stress forcold-workedsteels-requires a guaran-
tee as to its minimum value, which constitutes the reference value of the

1:
I Guaranteed minimum
I
I
I
I
B
CL
I
I
I
4/

Figure 3.1

mechanical strength of the steel. This guarantee must be given by the

manufacturer.
Available statistical information (statisticaldistribution of results,mean
value and standard deviation) may be used,subject to the building owner’s
consent.In that case the notion of ‘guaranteedminimum value’is replaced
by the notion of ‘characteristicvalue’,according to the procedure explained
in the second chapter of Part 2.
18
Stress-strain diagrams of standardised form (standard diagrams) for plain
bars and deformed bars,as defined in Section 3.1.1,for a strain of up to 1 %
should be used for the calculation.
The standard stress-straindiagram for ordinary steels is assumed to be
defined by (seeFigure 3.1):
(a)the straight line conforming to Hooke's law,extending from the origin
to the point whose ordinate corresponds to the elastic limit (whichis

Figure 3.2

assumed to coincide with the limit of proportionality and,for mild

steel,with the yield point);
(b)a straight line parallel to the axis of abscissae.
The standard stress-straindiagram for cold-worked steels is assumed to
be defined by (see Figure 3.2):
(a)the straight line conforming to Hooke's law,extending from the origin
to the point whose ordinate is equal to seventy-twohundredths of the
proof stress (0.7200.,);
(b)a curve determined by the following values of the permanent strain:
0.01%for a stress equal to 0.800go.,
0.03%for a stress equal to 0.880oO.,
0.07%for a stress equal to 0935 oO.,
0.10% for a stress equal to 0.960
0.15%for a stress equal to 0.985go.,
0.20%for a stress equal to oo.2
0.50%for a stress equal to 1.050oO.,
1.00% for a stress equal to 1.090
 19
This standard diagram is valid up to a proofstress of6O00 bars (o,,.2d 6O00
bars).
These standard stress-strain diagrams have been experimentally deter-
mined on the basis of a large number oftest resultscommunicated by various
manufacturers in several countries.They show a uniform value of 2 100 O00
bars for the modulus of elasticity E,.
In any case these standard diagrams are on the safe side and may,subject
to reversal of the algebraic signs,also be used as standard compressive
stress-straindiagrams.
For welded steel fabric and other special categories of reinforcement no
standard stress-straindiagram has been established:in such cases the designer
should use the diagrams supplied by the manufacturer, subject to prior
consent from the building owner.
If the designer does not know the precise nature of the steel, he should,
as a safety measure and basing himself on the guaranteed or measured
minimum elastic limit,adopt the diagram for cold-worked steels (up to
o, = oe or followed by a straight line parallel to the axis of abscissae
(upto the limiting strain E, = 10%).

3.2 CONCRETE

3.2.1 R E F E R E N C E V A L U E S O F THE M E C H A N I C A L STRENGTH

OF C O N C R E T E

Except in special cases the reference values of the mechanical strength of the
concrete in compression and in tension are defined by the minimum results
of preliminary crushing and splitting tests performed on cylindrical speci-
mens 28 days old.
The design methods embodied in this Manual are based on the above
definition of the compressive and the tensile strength of the concrete.If, at
the building owner’sexpress request,the tests are performed on other types
of specimens,according to other experimental procedures or at different
ages,then the strength values thus obtained must be adjusted by applying
the necessary corrections to them before introducing them into the calcula-
tion.
Unless special justification to follow a different procedure is submitted,
the compressive strength of the concrete should be determined from tests
performed at an age of 28 days on cylindrical specimens measuring 15 cm
in diameter and 30cm in height and subjected to crushing at the end faces,
which should be flat or trued.
If the crushing test is performed either on cylindrical specimens with
different geometrical dimensions or on prismatic specimens or on cubes,
the test results should be multiplied by the correction factors indicated in
Table 3.1.
If the crushing test is performed at an age other than 28 days,the test
20
results should be multiplied by the correction factors indicated in the
Table 3.2.
Table 3.1

Nature of the test specimen

Correction factor
(assumed to have parallel plane faces)

Cylinder 4 15cmx30cm 1.00

4 10cmx20cm 0.97
4 25 cm x 50 cm 1.05
Prism 15 cm x 15 cm x45 cm 1 .O5
20 cm x 20 cm x 60 cm 1.05
Cube 10cm x 10cmx 10cm 0.80
15 cm x 15 cm x 15 cm 0.80
20cm x 20 cm x 20 cm 0.83
30 cm x 30 cm x 30 cm 0.90

Table 3.2

A g e of the concrete (in days) 3 7 28 90 360

Ordinary Portland cement 2.50 1.50 1.00 0.85 0.75

Rapid-hardeningPortland cement 1.80 1.30 1.00 0.90 0.85

Unless special justification to follow a different procedure is submitted,

the tensile strength of the concrete should be determined from tests per-
formed at an age of 28 days on cylindrical specimens measuring 15 cm in
diameter and 30cm in height and subjected to splitting by application of
IP

IP Figure 3.3

two equal compressive forces along two generating lines diametrically

opposite each other (seeFigure 3.3).
The value of the tensile strength of the concrete is given by the formula:
2P
0 0 =-
ndl
where d denotes the diameter of the test specimen (d = 15 cm)and 1 denotes
the height thereof (d = 30 cm).
A similar splitting test may be performed on cube specimens (see Figure
3.4).
 21
In that case the value of the tensile strength of the concrete is given by the
formula:
2P
(To = ~

na2
As an alternative to the splitting test, the building owner may consent to
tests on prismatic specimens measuring 10 cm x 10 cm x 50cm which are
loaded in circular bending in the central part over a length of at least 15 cm.

Figure 3.4

Conventionally, the value ol the (direct)tensile strength of concrete is

taken as equal to three-fifths(0.60)of the flexural strength (i.e.,the tensile
strength in bending), namely :

where M u denotes the failure moment of the specimen in bending and b

denotes the side length of the cross-sectionof the specimen.
If the splitting test (orthe flexural test) is performed at an age other than
28 days,the test results should be multiplied by the correction factors indi-
cated in Table 3.3.
Table 3.3

A g e of the concrete (in days) 3 7 28 90 360

Ordinary Portland cement 2.00 1.40 1.00 0.95 0.90

Rapid-hardeningPortland cement 1.50 1.20 1.00 0.95 0.90

If no preliminary tests are performed on the concrete before it is used in

the structure,the reference value of the compressive strength of the concrete
may,subject to the building owner's approval,be taken as equal to one of
the values shown in Table 3.4.
Table 3.4

Compressive strength of the concrete

Cement content with constant without constant
supervision supervision

300 kg/m3 230 bars 150 bars

350 kg/m3 270 bars 180bars
400 kg/m3 300 bars 200 bars
22
These values,which have been derived from statistics of test results and are
on the safe side,do presuppose,however, that the quantity of water used
for making the concrete is the least that is compatible with the conditions of
placing the concrete and,furthermore,that the percentage of sand is between
30% and 50% of the total weight of the inert materials.
If no preliminary tests are performed on the concrete before it is used in
the structure,the reference value of the tensile strength of the concrete may,
subject to the building owner’sapprovai,be derived from the corresponding
compressive strength by applying the following empirical formula:
O,, = Job
where o. denotes the tensile strength and ob the compressive strength
of the concrete determined on cylindrical specimens,referred to an age of 28
days and expressed in bars.
This empirical formula is valid for compressive strengths in the range
between 150 and 400 bars.
If this formula is applied in a case where no preliminary tests are per-
formed on the concrete before it is used in the structure,the values to be
adopted for the tensile strength are as follows:
(a)with constant supervision:
for a cement content of 300 kg/m3: 15.2bars
for a cement content of 350 kg/m3: 16.4bars
for a cement content of 400kg/m3: 17.3bars
íb) without constant supervision:
for a cement content of 300kg/m3: 12.3bars
for a cement content of 350kg/m3: 13.4bars
for a cement content of 400kg/m3: 14.1bars
In comparison with the statistics of test results,these values are on the
safe side.

3.2.2 M O D U L U S OF E L A S T I C I T YO F C O N C R E T E

For Instantaneous Loads (Instantaneous Modulus)

For instantaneous (or rapidly changing) loads the modulus of elasticity
of the concrete at an age ofjdays can be calculated (inbars) from the follow-
ing empirical formula:
Ebo= 21 000JoS (bars)
where o; denotes the average compressive strength (inbars) of the concrete
at jdays.
This formula can be considered valid so long as the compressive stress
in the concrete in the state under consideration does not exceed one-third
of the corresponding compressive strength.Otherwise the designer should
refer to the compressive stress-strain diagram for the concrete and take
account of the effects of plastic behaviour and,possibly,hysteresis.
 23
For Sustained Loads (Long-Term Modulus)
For loads of long duration (morethan 24 hr) the modulus of elasticity of the
concrete can be derived from the instantaneous modulus by taking account
of long-termdeformations due to the combined effectofshrinkage and creep.
The basic data for this calculation,areindicated in Chapter 7,Part 2 of
this Manual.
In normal cases,however,the designer may use the followingapproximate
formula,which is valid if the stresses are not constant:
Eb, = 7000JoJ(bars)
where o; denotes the average compressive strength (in bars) of the concrete
at j days.

3.2.3 POISSON’S R A T I O F O R C O N C R E T E

Poisson’sratio for elastic strains may be taken as 0.15.

3.2.4 COEFFICIENT OF T H E R M A L EXPANSION OF C O N C R E T E

The coefficient of thermal expansion of concrete may,on average,be taken

as
This is merely an average value;tests show that the coefficient of thermal
expansion may vary within a wide range (ofthe order of I30%), depending
on the nature of the cement and of the aggregates,the cement content,the
humidity,and the dimensions of the sections. In cases where the influence
of temperature variations is of particular importance the .coefficient of
thermal expansion should therefore be determined experimentally from
measurements on members having the same dimensions and consisting of
exactly the same concrete as the actual members.

3.2.5 L O N G - T E R M L I N E A R D E F O R M A T I O N S OF THE
CONCRETE

The values of the long-term linear deformations of concrete (shrinkage

and creep) depend on a large number of parameters: dimensions of the
member, water/cement ratio of the concrete, relative humidity, etc. The
effect of these parameters may vary significantly from one region to another
and from one country to another. These values should therefore in each
case be determined experimentally from appropriate measurements on
members having the same dimensions and consisting of exactly the same
concrete as the actual members. Such measurements should be carried out
under conditions of temperature and humidity comparable to those on the
site.
A set ofexperimentaldata which can provide a basis for design calculations
is given for guidance in Chapter 7,Part 2 of this Manual.
 4
DETERMINATION OF
SAFETY

4.1 PRINCIPLE OF CHECKING THE SAFETY

Checking the safety of a structure should,in accordance with the results
of the probability theories and with the available statistical information,take
account of the scatter in the various kinds of loading and in the strengths of
the various constituent materials.This check should be carried out for the
various limit states corresponding to the respective conditions in which the
structure becomes unfit for service,including more particularly :the ultimate
limit state,the limitstate of instability,and the limitstate of deformation,etc.
Checking the safety consists in verifying that for each limit state the effects
of the ‘characteristicloadings’(definedin Section 4.2)do not exceed the load
capacity deduced from the ‘basicstrengths’(defined in Section 4.3)of the
steel and concrete.
This practical method of analysis is a simplificationof the semi-probability
method adopted by the Conseil International du Bâtiment (C.I.B.) and the
Comité Européen du Béton (C.E.B.). This simplified method is to be regarded
as sufficient for all ordinary structures not exceptional in character.
However,should it occur that the designer wishes to have more precise
information on the safety of the structure,and if he has at his disposal
sufficient statistical data concerning the values of the loadings and strengths,
he may apply,systematically and in full,the semi-probabilitymethod of
the C.I.B. and C.E.B. which is explained in Chapter 2,Part 2 of this Manual.

4.2 DETERMINATION OF THE CHARACTERISTIC

LOADINGS
The ‘characteristic loadings’to be introduced into the analysis for each
limit state comprise:
24
 25
4.2.1 P E R M A N E N T L O A D S A N D FIXED SUPERIMPOSED
W O R K I N G LOADS, s,.
The permanent loads are estimated from the volume of the materials and
their density under the conditions of use.
The fixed superimposed working loads should be taken as equal to the
corresponding ‘nominalsuperimposedloads’specified by the building owner.
If no such loads are specified, the designer may use the values given
for guidance in Chapter 1, Part 2 of this Manual.

4.2.2 VARIABLE SUPERIMPOSED W O R K I N G LOADS, S P I ,

MULTIPLIED BY A N AMPLIFICATION COEFFICIENT
E Q U A L T O 1.20 FOR RESIDENTIAL BUILDINGS
A N D TO 1.30 F O R O T H E R S T R U C T U R E S

The variable superimposed working loads should be taken as equal to the

corresponding ‘nominal superimposed loads’ specified by the building
owner.If no such loads are specified,the designer may use the values given
for guidance in Chapter 1, Part 2 of this Manual.
The increases of 20% or 30% applied to these nominal superimposed
loads take account of the scatter and furthermoreallow for the unfavourable
effect of the variation in direction and magnitude of the stresses,which is due
to the variability of the superimposed loads, irrespective of any dynamic
effect.

4.2.3 D Y N A M I C SUPERIMPOSED W O R K I N G LOADS, S,, ,

MULTIPLIED BY A ‘ D Y N A M I C COEFFICIENT’
SPECIFIED BY T H E BUILDING O W N E R

This amplification coefficient takes account of the unfavourable effect of

transient dynamic phenomena (as occur in bridges, overhead crane track
girders,etc.). If it is not specified by the building owner,the designer may use
the following empirical value:

0.50 0.80
a = 1.30+
+

l(metres)
l+ 5
1+ s,
SPI+ s,,

where 2 denotes the free (unrestrained)length of the element considered,

expressed in metres.Permanentfunctioningconditionsand cyclicphenomena
(e.g.,rotating machinery or machinery with reciprocating action) are not
allowed for in this coefficient and should be the subject ofspecialinvestigation
in each particular case.
26
4.2.4 C L I M A T I CS U P E R I M P O S E D L O A D S S,.D U E M O R E
PARTICULARLY TO W I N D A N D S N O W , A N D
E A R T H Q U A K E ACTIONS

In the absence of special regulations concerning climatic superimposed

loads the designer may use the values given for guidance in Chapter 1,
Part 2 of this Manual. These values,which implicitly take the scatter into
account,should be used without any increase of magnitude.

4.2.5 ACTIONS D U E TO SHRINKAGE, CREEP A N D

T E M P ERA T U RE VARI AT I O N S s,
In the absence of accurate experimental data for estimating the shrinkage
and creep,the designer may use the values given for guidance in Chapter 7,
Part 2 of this Manual.These values should be used without any increase of
magnitude.

4.2.6 THE E F F E C T OF THE C O N S T R U C T I O N P R O C E D U R E ,

W H E R E RELEVANT

The designer should examine whether,in the intermediate stages of con-

struction and erection of the structure, particular loadings or particular
combinations of loadings are liable to have an adverse effect on the safety
of the structure and therefore call for an additional check.
To sum up, the procedure for determining the ‘characteristic loadings’
to be introduced into the safety analysis for each limit state can be schemati-
cally represented by the followingtwo symbolic expressions:
(a) for residential buildings:
S, + 1.20s,,+S”+s,+ ...
(b)for other structures:
S,+ 1.30SPI+ U S,, + SV+&+. ..

4.3 DETERMINATION OF THE BASIC S T R E N G T H S

4.3.1 D E F I N I T I O NOF B A S I C S T R E N G T H S

The basic strength of the steel or the concrete,to be introduced into the
analysis for each limit state,is taken as equal to the guaranteed minimum
strength (referencestrength)divided by an appropriate reduction coefficient.
If the designer has at his disposal at least 20 results of preliminary tests
on the steel or the concrete to be used on the site,he may adopt instead of
the guaranteed minimum strength a ‘characteristicstrength’,which is taken
 27
as being equal to twice the mean value of half the results that fall short of
the median,minus the mean value of all the results.

4.3.2 B A S I C S T R E N G T H OF THE S T E E L

The basic strength of the steel is taken as equal to the guaranteed minimum
value of the elastic limit (referencestrength,cf. 3.1.3),divided by a reduction
coefficient whose value,for each limit,is as follows:
-o, = 6,
__
Ysteei

ultimate limit state Ysteei = 1.80

limit state of instability Ysteei = 1.80
limit state of cracking Ysteei = 1.80
limit state of deformation Ysteei = 1.80

4.3.3 BASIC S T R E N G T H OF T H E C O N C R E T E

The basic compressive strength Obof the concrete is taken as equal to the
guaranteed minimum value of the 28-day cylinder strength,i.e.,the com-
pressive strength determined on cylindrical specimens (reference strength,
cf. Section 3.2.1),divided by a reduction coefficient whose value, for each
limit state,is given in Table 4.1.
0'0
Table4.1 ob = -
Yconcrete

Ready-mixed concrete Concrete mixed on the site

Yconsreie (strictly controlled (with (without

batching and permanent permanent
permanent supervision) supervision) supervision)

Ultimate limit state 2.10 2.30 2.50

Limit state of
instability 2.50 2.75 300
Limit state of cracking 1.45 1.45 1.45
Limit state of
deformation 2.00 210 2.20

The values of the reduction coefficients for the ultimate limit state im-
plicitly presupposes that failure of the member or the structure under con-
sideration is not due to brittle fracture and is not of a violent character
marked by the absence of any warning signs.If, on the other hand,failure
does occur by brittle fracture,the values of the reduction coeficients for
the ultimate limit state should be increased by 20%, in which case they
become identical with the values of the reduction coefficients for the limit
28
state of instability; this is more particularly the case with load-bearing
walls and panels.
Similarly,the basic tensile strength Ob of the concrete,in all cases where
this strength has to be introduced into the structural calculation,is taken
as equal to the guaranteed minimum value of the 28-day tensile splitting
strength,i.e.,the tensile strength determined in the splitting test on cylin-
drical specimens (referencestrength,cf. Section 3.2.1),divided by the same
reduction coefficient as for the basic compressive strength (cf.Table 4.1):
-
fJb = ~
00
Yconcreie

Furthermore,in the case of a multiple state of stress characterised by the

extreme stresses o1 and o; (tensile and compressive stress respectively),
the designer should also check the following condition,under the effect of
the characteristic loadings:
f
J <2---
1 1
00 IfJi1
5
Yconcreie

This condition is obtained by considering the corresponding limit state,

for which :
- 01
Ob = -+-
1
4 I
2 lo
It is equivalent to specifying that,in the case of a multiple state of stress,
the designer must adopt asthe basic tensilestrengthofthe concretethe smaller
of the two values:
and 2--- 00 132 I
Yconcrete Yconcrete 5
 5
DETERMINATION OF THE
EFFECTS OF THE
PERMANENT LOADS,
SUPERIMPOSED LOADS AND
OTHER ACTIONS

5.1 STRUCTURES COMPOSED OF LINEAR M E M B E R S

In structures composed of linear members the effects of the loads and other
actions in the various sections may be calculated by applying the elastic
theory,in exact or approximate form.
A redistribution of bending moments may be taken into account without
necessitating a check for compatibility in the ultimate limit state.However,
to take account of such redistribution,it must be ensured that all the follow-
ing conditions are satisfied:

5.1.1 I N T H E E S T I M A T I O NO F T H E L O C A L S T R E N G T H O F
SECTIONS T H E STEEL STRESS T O B E T A K E N INTO
A C C O U N T M U S T N O T EXCEED T H E BASIC S T R E N G T H

This condition arises more particularly in the design of the section with
regard to the ultimate limit state.It means that the use of the basic diagram
for the tensile reinforcing steel (Section6.1)must be limited to the values
cra d O, in all cases where redistribution of the bending moments in relation
to their elastic distribution is taken into account in the calculation for a
statically indeterminate structure.For practical purposes this condition
applies only to cold-workedsteels whose stress-straindiagram has no definite
yield point.
29
30
5.1.2 N O R E D I S T R I B U T I O NM U S T B E A S S U M E D T O O C C U R
IN S T R U C T U R E S IN W H I C H C R A C K I N G IS LIABLE
T O H A V E H A R M F U L CONSEQUENCES

This condition is applicable more particularly to structures which have to

be watertight or which are exposed to a humid or corrosive atmosphere.
This condition must also be satisfied in the design calculations for statically
indeterminate structures which have to support brittle facings or claddings
for which excessive cracking and deformation may have harmful con-
sequences.

5.1.3 T H E M E C H A N I C A L P E R C E N T A G E OF R E I N F O R C E M E N T
IN T H E LINEAR M E M B E R S F O R M I N G T H E S T R U C T U R E
M U S T N O T EXCEED 0.20

The conditions of Sections 5.1.1 and 5.1.3derive from theoretical and experi-
mental considerations associated with the investigation of the equilibrium
of statically indeterminate structures beyond the elastic range.
In general, the first plastic deformations in the concrete appear only in
some of the critical sections.This results in a redistribution of the moments
which relieves precisely those regions which are most severely stressed and
delays the cracking thereof.
If the loads are further increased up to failure,fresh regions subjected to
positive or negative bending moments enter the elasto-plasticrange. The
distribution of the moments is then much more difficult to predict, since it
results from the algebraic sum of contrary effects. The danger will be greater
if the redistributions produced by the inelastic deformations of opposite
signs do not compensate one another,for failure is liable to occur in regions
having less capacity for adaptation, whereby the safety margin for the
structure as a whole could be significantly reduced.
This danger exists more particularly:
(a)If it is attempted to utilise to the full the strength capacity of certain
sections beyond the point corresponding to the start of large deforma-
tions (i.e.,beyond the basic strength of the steel,),,?i for it is not certain
that the other regions can continue to adapt themselves sufficiently.
This danger justifies the condition of Section 5.1.1.
(b)Ifcertain regions ofthe structure have a limited capacity for adaptation,
which is the case with heavily reinforced sections.This danger justifies
the condition of Section 5.1.3.

5.1.4 T H E E X T E N T O F THE R E D I S T R I B U T I O NO F T H E
M O M E N T S I N R E L A T I O N T O THEIR ‘ E L A S T I C ’
VALUES M U S T N O T E X C E E D 15%

Any value in excess ofthis should be justified by means ofa completeanalysis

of the behaviour of the statically indeterminate structure in the elasto-plastic
 31
range up to failure.For that analysis it is, in particular,necessary to know
the actual moment-curvaturediagrams applicable to each section.

5.1.5 THE B E N D I N G M O M E N T DIAGRA‘M T A K E N I N T O

A C C O U N T M U S T SATISFY T H E EQUILIBRIUM
CONDITIONS

In addition, it must in all cases be checked that the columns are able to
resist the flexural loads applied to them,both on the assumption of elastic
distribution of the moments and on that of moment redistribution.
There are at present various methods of analysis available whereby the
inelastic behaviour and the adaptation capacity of structures consisting of
linear members can be taken into account.However,the application of these
methods often calls for considerable computational effort; besides, their
development is not yet far enough advanced to enable them,except in special
cases,to be replaced by simple approximate rules.
Apart from design based on the elastic theory with a linear relationbetween
stress and strain,the following methods may be adopted.

Plastic Design

This is based on the hypothesis of the complete plastification of certain

sections of the linear members which together form the structure and of the
formation of ‘plastichinges’at those sections.These plastic hinges must be
sufficiently numerous and be so located that the structure is transformed
into a ‘mechanism’,i.e.,an articulated system with at least one degree of
freedom.
The ‘mechanism’and the corresponding failure load can be deduced from
the values of the plastification moments of the sections by the application of
‘staticcompatibility’conditions,which yield an upper limit for the failure
load,or ‘kinematiccompatibility’conditions,which yield a lower limit.The
actual failure configuration and failure load are those which simultaneously
satisfy the conditions of static compatibility and those of kinematic com-
patibility.
Generally speaking,however,reinforced concrete structuresdo not possess
sufficient deformation capacity to fulfil the hypotheses of plastic design.
Hence this design method can be applied only within a very limited range;
in particular,it must not be used if considerable redistributionsof moments
and forces in rrlation to the elastic distribution are necessary for attaining
the failure configuration assumed in the design calculation.

Elasto-Plastic Design With Limited Rotations

This is based on the consideration of complete plastification of certain
sections (‘plastichinges’)which are so located and in such number that the
structure is transformed into a statically determinate system.
32
It is assumed that the portions of members situated between the plastic
hinges retain their elastic behaviour, and a check is made to ascertain that
the values of the plastic hinge rotations which are necessary for actually
achieving the assumed configuration do not exceed the limiting values that
these rotations can attain.
The limiting values of the rotations in the plastic hinges are a function of the
section properties (geometricalshape,type and percentage ofreinforcement,
etc.) and of the kind of stress conditions to which those sections are sub-
jected. Experimental research with a view to determining these limiting
values is in progress.

Non-Linear Design
This is based on the adoption of non-linear relations between the stresses
and strains in the sections of reinforced concrete structural members, for
which purpose the corresponding moment-curvature diagrams are intro-
duced into the calculation.
In actual practice the overall results of this non-linearcalculation proce-
dure can be embodied in ‘redistributioncoefficients’.These coefficients give
the maximum alterations that can be introduced into the distribution of the
resistance moments of the sections,in relation to the elastic distribution,
without entailing any significant reduction in the load capacity of the struc-
ture and in the corresponding margin of safety.
In the present state of research,however, this analysis has as yet been
carried out only for a small number of structures and load arrangements.
Figure 5.1 gives some limiting values of the ‘redistribution coefficients’,
these being applicable to normal cases,subject to an additional check of the
conditions of static equilibrium.The following notation is used in the table:

p redistribution coefficient applicable to the elastic moment at the

- section (ïj
w o geometrical percentage of main tensile reinforcementin the section @
ce reference value of the elastic limit of the steel

5.2 PLANE STRUCTURES

5.2.1 P L A N E STRUCTURES LOADED PERPENDICULARLY

T O THEIR M I D D L E P L A N E

These rules relate more particularly to the analysis of the strength of slabs
and flat-slabfloors under flexural loading which is assumed to consist,in
the main,of forces acting perpendicularly to the middle plane thereof.They
do not comprise the analysis of slabs and flat-slabfloors with regard to
 33
O m O
ol I a
l m
- 0 O O
2
A
O O m
m W W
O O O
d
‘ O
al
L
3
Y
u
3
L
44
m
L
O
al
a
x
c

1.3
c \ I,
o O
34
punching shear due to concentrated forces;that analysis is envisaged in
Section 6.2.5.
In plane structures (plates,slabs and flat-slabfloors) loaded perpendicu-
larly to the middle plane the effects of the loads and other actions can be
calculated by means of the exact or approximate elastic theory, provided
that the actual support conditions and conditions of functioning of such
structures (moreparticularly,the stiffness of the supports,the effect of edge
beams,if any), as well as their more or less complex geometrical shapes,the
actual loading conditions and any loadings of an exceptional character are
taken into account.
By extending the elastic theory beyond its basic assumptions it is possible
in some cases to take account of the cracking and plastification of the con-
crete,particularly for the analysis of the limit state of cracking and the limit
state of deformation.Research is now in progress with a view to establishing
appropriate practical design methods.
For checking the ultimate limit state,methods which take account of the
statically indeterminate effect of plasticity,more particularly the so-called
yield-linetheory,can permissibly be employed,on condition that :
(a) the yield pattern of the structure under consideration is justified with
certainty or is determined by means of appropriate tests;
(b)the basic assumptions of these methods are really fulfilled;
(c) the set of loads under consideration corresponds to the most un-
favourable arrangement of these loads.
For practical application of the plastic theories it may be assumed that all
the loads undergo a proportional increase in magnitude (once their most
unfavourable arrangement has been determined) and that the steel and con-
crete strengths are proportionally reduced.
A precise analysis of the conditions of practical application of the plastic
theories,more particularly the yield-linetheory,has so far been carried out
only for a limited number of plane structures and methods of loading.The
corresponding recommendations, together with the examples known at
present,are given,as an appendix,in Chapter 8,Part 2 of this Manual.

5.2.2 P L A N E S T R U C T U R E S LOADED P A R A L L E L TO THEIR

MIDDLE PLANE

The structures under consideration are assumed to be loaded,in the main,

by forces acting parallel to the middle plane thereof.These are structures with
two dimensions large in comparison with the third and with a plane middle
surface.They include more particularly: load-bearingwalls and panels and
deep beams (girder walls). As a secondary feature these structures may be
loaded perpendicularly to the middle plane.
For the purpose of the present rules the structures under consideration
are assumed to be cast in situ. Prefabricated structures,more particularly
those constructed by assembling large precast panels,form the subject of a
chapter of this Manual.
 35
Load-bearing Partitions A n d Walls
Estimation Of The EffectsOf The Vertical Loads
Estimating the strength of a load-bearing partition or wall involves the
calculation of the magnitude and position of the resultant of the vertical
components of the forces acting upon the structure.
By ‘load-bearingpartitions or walls’ are understood piane structures
which are used in the vertical position and have continuous support along
the bottom edge.Apart from exceptional cases,in withstanding the vertical
forces acting on them these structures are not subjected to bending per-
pendicularly to the middle plane.
Load-bearing partitions and walls generally perform either or both of
the following functions:
(a) load-bearing function with regard to the vertical loads and superim-
posed loads and with regard to the vertical components of the forces
due to other loadings;
(b)wind-bracingfunction with regard to lateral loadings acting parallel
to the plane of the wall or partition.
The partitions and walls may be free along their vertical edges or be
secured by stiffeners along those edges.However,in order to be considered
structurally effective,the stiffener should have a width equal to a quarter of
the free (unrestrained)height of the panel considered.
The design should be done by the usual methods, derived from the elastic
theory of structures and,in particular,taking account of the actually existing
connections between the panel under consideration and the other compo-
nents of the structure. The risk of buckling should be taken into account
by introducing a complementary eccentricity.
Load-bearingpartitions and walls have not yet been exhaustively studied
on the basis of the fundamental concept of the limit states.Hence, in the
present state of knowledge, the design methods should be based on the
elastic theory,in accordance with the indications given in this Manual.
A concentrated load (ora load applied to an area of limited size) should
be assumed to be uniformly distributed within a zonedefined by two straight
lines inclined in relation to the vertical at a slope of 1 in 3 in the case of an
unreinforced wall or panel,and 2 in 3 in the case of a reinforced wall or panel
and starting from the point of application of the concentrated load (or from
the ends ofthe area of application ofthe load).
The same rule should be applied for calculating the disturbances that the
presence of openings (ifany)causes in the distribution of the forces.
The initial eccentricity is the resultant of
(a) structural eccentricities;
(b)accidental eccentricities;
(c) additional eccentricities.
Structural eccentricities are eccentricities arising,on the one hand,from
the eccentric positions of certain loads or superimposed loads (e.g., eccen-
tricity due to changes in the thickness of a gable wall) and,on the other hand,
36
from bending moments produced by other elements of the structure (e.g.,
transmission of bending moments from floors to walls). These structural
eccentricities should be taken into account with their appropriate algebraic
signs.
Accidental eccentricitiesare eccentricitiesdue to faultsofexecution (devia-
tions from true flatness,positional errors,etc.). For want of a more accurate
analysis, an overall accidental eccentricity, conventionally taken as 2 cm,
should be taken into account;this eccentricity may be reduced to 1.5cm in
the case of particularly accurate and careful workmanship,but it must be
increased to 2.5cm if the formwork is ofpoor workmanship,liableto deform,
or difficult to adjust.The overall accidental eccentricity should be taken
into account with its most unfavourable algebraic sign.
The designer should also take account of additional eccentricities due to
certain kinds of superimposed loads associated more particularly with the
transverse bending moment produced by wind forces (pressureor suction)
or with the transverse bending moment produced by the thermal gradient
that may exist between the two faces of the partition or wall considered.
These eccentricities should be taken into account with their appropriate
algebraic signs,but in such a manner as to obtain the most unfavourable
combination for the initial eccentricity.
The buckling risk of a load-bearingpartition or wall should be taken into
account by introducing a complementary eccentricity, e,, (or a comple-
mentary bending moment). The determination of this complementary
eccentricity will depend on the restraints existing along the edges of the
panel considered and also on the behaviour of the type of partition or wall
envisaged in the structure. In the absence of accurate experimental in-
formation as to this behaviour,the following hypotheses should be adopted :
(a)the Euler critical stress to be introduced into the expression for the
complementary eccentricity should be calculated on the assumption
of elastic behaviour,according to the methods of the theory of elastic
stability;
(b)floors should be conceived as free to move in relation to one another
in the case of one-waystructures,and as fixed in relation to one another
in the case of two-wayor multi-waystructures;
(c) load-bearing panels should be conceived as hinged at the top and
bottom edge; they may be regarded as rigidly restrained (built in)
only if such restraint is physically assured.
The overall eccentricity to be introduced into the calculation is taken as
equal to the resultant of the initial eccentricity and the complementary
eccentricity:
e = e,fe,

Estimation Of The Effects Of The Horizontal Loads

The forces corresponding to the wind-bracingfunction of the partitions and

walls should be calculated on the assumption of elastic behaviour,according
to methods derived from the theory of elasticity.
 37
The designer may, subject to justification, structurally interconnect a
number of panels situated in different planes but having common edges,with
a view to forming wind-bracingelements with a larger moment of inertia.
Such composite vertical ‘cantilevers’should be considered as ‘thin-walled
beams’.
These ‘cantilevers’may be subject to considerable torsional loads when
they undergo the effect of forces which do not pass through the centre of
torsion of the corresponding section.In certain cases,also,the ‘thin-walled
beams’ formed by such cantilevers may be subject to warping restraint
exercised by diaphragms (floors,in particular). Chapter 8,Part 1 of this
Manual indicates practical methods that may be employed for this analysis.
Finally,it is necessary to bestow particular attention on the manner in which
the actions tangential to the junction of two panels is equilibrated.
Simplifying assumptions may be applied to the analysis of panels provided
with openings:
(a)If the openings are of small size, separated by lintels of sufficient
stiffness,the panels may be analysed as flat panels. The tangential
actions that would occur in the panels if they were solid instead of
pierced are assumed to be transmitted to the lintels.
(b)If the openings are of large size,separated merely by lintels of low
stiffness,the parts of the panels separated by these openings should be
conceived as structurally inter-connectedso that they are compelled to
undergo the same horizontal deformations.Besides,the lintels should
be able to withstand the stressesdue to the deformations that they are
made to undergo.
(c) If the openings are of medium size,separated by lintels of medium
stiffness,the panels should be analysed by an appropriate method,
e.g.,the method proposed in Chapter 8,Part 1 of this Manual for
estimating the distribution of the forces in a pierced panel.
Chapter 8,Part 1 of this Manual also specifiesthe limits of the range of
validity ofthe foregoing simplifyingassumptions.

Deep Beams

Estimating the strength of a deep beam (or ‘girderwall’)involves the calcula-

tion of the magnitude and position of the forces acting upon the
structure.
By ‘deepbeams’are here understood plane structures which are used in
any position in order to balance the forces parallel to the middle plane
thereof and have discontinuous support along the edge opposite the forces.
Their structural behaviour involves bending about an axis perpendicular
to the middle plane. Such members may sometimes also be loaded per-
pendicularly to their plane.
The design should be done by the usual methods,derived from the elastic
theory of structures and from the theory relating to systems in which the
deformation due to shear force is not negligible. The designer should take
38
account ofthe risk ofbuckling by transversedeflection (twist-bendinstability)
by applying the usual methods derived from the theory of elastic stability,
taking account of such eccentricities as may be introduced by the lateral
loads.
A deep beam should be provided with a system of main reinforcing
bars and a system ofdistributed transversebars in the form of verticalbinders;
there should also be horizontal reinforcement distributed over both faces
0
of the beam.
If the loads or superimposed loads are applied to the bottom part of the
beam, the vertical reinforcement of the network of distributed bars should
be increased so that the additional steel cross-sectionalarea thus provided
corresponds to the amount of load to be suspended.
As regards the analysis for shear force,the shear stress to be introduced
into this analysis should be the value obtained by applying the theory of
homogeneousbeams and not that ofreinforced concretebeams.
Finally, if a deep beam is subjected not only to loads parallel to its
plane but also to flexural effects perpendicular thereto,the corresponding
stresses may, as a first approximation,be estimated by adding together
the stresses corresponding to each of those flexural effects considered
separately.Besides, the effect of such transverse bending on the buckling
conditions should be checked.
 6
DETERMINATION OF
SECTIONS

6.1 NORMAL FORCES AND STRESSES

6.1.1 U L T I M A T E L I M I TS T A T E

Uniaxial Bending (Simple Or Composite)

Fundamental Design Assumptions
The analysis of the ultimate strength for uniaxial bending should be based
on the following four fundamental assumptions:
1. The deformations (strains)of the elements of a cross-sectionare always
proportional to the distances of these elements to the neutral axis (the
‘Navier-Bernoullihypothesis’which states that piane sections remain plane).
This hypothesis is not strictly in agreement with reality,as experiments
show;because of (systematic or random) cracking phenomena under the
usual conditions or service the longitudinal strains and stresses in reinforced
concrete members do not conform to the theoretical laws of strength of
materials.This is what actually occurs:
(a) The longitudinal strains of the concrete and steel are not constant
along the member, even under the action of a constant bending
moment. In fact,these longitudinal strains are greater at the cracks
and smaller between the cracks.
(b)Progressive cracking of the tensile zone of the concrete produces,
between each pair of adjacent cracks,a sliding of the reinforcement in
relation to the concrete and,associated with this sliding,a local vari-
ation in the bond stress and the tensile stress in the steel.
(c) In consequence,the position of the neutral axis varies all along the
member.
T o take all these phenomena systematically into account would too
greatly complicate the designer’stask. Therefore the hypothesis that piane
39
40
sections remain plane can be regarded as a reasonable and sufficiently close
approximation.
2.The compressive stress distribution in the concrete in the compressive
zone of the member is assumed to conform to a standard stress-strain
diagram defined experimentally.
Theoretical and experimental research has shown that the relation be-
tween the compressive stresses and the compressive strains of the concrete
at a section depends not only on the quality of the concrete but also on a
large number of other parameters, including: the rate and duration of
application of the loads; the nature of the loadings; the position of the
neutral axis in the ultimate limit state;the geometrical shape of the section;
and the environment in which the member is situated.
To take all these parameterssystematically into account would too greatly
complicate the designer’stask.Therefore the main object of theoretical and
experimental research has been to determine simplified standard diagrams
which are easy to apply in the usual design calculations and which lead to
results that are close to the theoretically predicted and experimentally
measured values.
3.The tensile strength of the concrete is neglected.
4.The tensile (and compressive) steel stresses are deduced,on application
of the hypothesis that plane sections remain plane (see Assumption i), from
conventional stress-strain diagrams based on the standard stress-strain
diagrams for ordinary steels and cold-workedsteels and valid up to a strain
of 1%.

Taking Account Of The Concrete In Compression

The notion ‘widthof the compressive zone’has a real significance only in

the case of rectangular sections or sections similar to them.
In a rectangular section the width of the compressive zone obviously
corresponds to the geometrical width of the section.
In the case ofT-beamsand ribbed floorsa conventionalvalue,the ‘effective
width of the compressive flange’,is introduced into the calculation.This
effective width defines the width of the compressive zone that effectively
participates in the flexural load capacity of the member.
The effective width of the compressiveflange of a T-sectiondepends on a
large number of parameters,including:the support conditions of the beam
considered (freely supported or continuity at the supports); the method of
load application (distributed loads or locally concentrated loads); the ratio
of the length of the beam (between free supports or between points of zero
bending moment) to the width ofthe rib (or web) and to the distance between
consecutive ribs; the ratio of the thickness of the flange (or slab) to the
depth of the beam; and the presence (if any) of fillets or haunches at the
junction of flange and rib.
From an analysis of the effect of these various parameters it has been
possible to establish a practical method of determining the effective width
 41
which is easy to apply in the usual calculations and yields results that are
close to the theoretically predicted and experimentally measured values.
In freely supported T-beamshaving either a single rib or a succession of
parallel ribs joined to the came slab the effective width of the compressive
flange is given in Tables 6.1and 6.2which indicate,as a function of various
geometrical parameters, the flange width (b,-b,)/2 to be adopted on each
Table 6.1Single-RibT-Beam.Flange width (be-b0)/2 to be taken into account on each side of
the rib at mid-span,for a uniformlydistributed load applied to the beam

Table oJvalues of- be -bo

b-bo

Values Values 21
Values of ~

of of b-bo
-
ho -1
O 1 2 3 4 6 8 10121416 18
h bo
Flange = 48
without 0.99
flexural
O 0,18 0-36052 0.64 0.73 0.85 0.92 0.95 0.94 0.98
,48
stgfness 4.00
10 O 0.18 0.36 0.53 0.65 0.78 087 0.92 0.95 0.98 0.99 1.00
50 O 0.19 037 0.54 0.66 0.79 0.87 092 0.95 0.98 0.99 1.00
0.10 100 O 0.21 0.40 0.56 0.67 0.80 0.87 0.92 0.96 098 0.99 1.00
150 O 0.23 043 0.59 0.69 0.81 0.88 0.92 0.96 0.98 0.99 1.00
200 O 0.27 0.47 0.62 0.71 0.81 0.88 0.93 0.96 0.98 0.99 1.00
10 O 0.19 037 0.53 0.66 0.79 0.87 0.92 0.95 0.98 0.99 1.00
50 O 0.22 0.42 0.58 0.69 0.81 0.88 0.92 0.96 0.98 0.99 1.00
015 1O0 O 0.30 0.51 0.66 0.74 0.83 0.89 0.93 096 0.98 0.99 1.00
150 O 036 0.60 0.73 0.80 0.86 0.91 0.94 0.96 0.98 0.99 1.00
200 O 0.40 0.65 0.79 0.85 0.89 0-920.95 0.97 0.98 0-99 1-00
10 O 0.21 0.40 0.57 0.68 0.81 0.87 0.92 0.96 0.98 0.99 1.00
50 O 0.30 0.52 0.69 0.78 0.86 0.90 0.94 096 0.98 0.99 1.00
0.20 100 O 0.40 0.65 0.79 0.86 0.89 092 0.95 0.97 0.98 0.99 1.00
150 O 0.44 0.70 0.85 0.91 0.94 0.95 0.97 0.97 0.98 0.99 1.00
200 O 0.45 0.73 0.89 0.93 0.95 0.96 0.97 0.98 0.99 1.00 1.00
10 O 0.28 0.48 0.63 0.72 0.81 0.87 092 0.96 0.98 099 1.00
50 O 0.42 0.65 0.83 0.87 090 0.92 0.94 0.96 0.98 0.99 1.00
0.30 100 O 0.45 0.73 0.90 0.92 0.94 0.95 0.96 0.97 0.98 0.99 1.00
150 O 0.46 0.75 0.91 0.93 0.95 0.97 0.97 0.98 0.99 1.00 1.00
200 O 0.46 0.77 0.92 0.94 0.96 0.97 0.98 0.99 099 14û i.00
42
Table 6.2 Multiple-RibT-Beam.Flange width (be-b0)/2 to be taken into account on each side
of the ribs at mid-span,for a uniformly distributed load applied to all the beams

Table of values of be -bo

~

b-bo

21
Values of -
Values of Values of b-bo
h0
- 1
-
h bo o 1 2 3 4 6 8 2 10

Flnnye
without = 10 0.99
flexural O 0.19 0.33 0.57 071 088 0.96 , l.oo
stflness

10 O 0.19 0.38 0.57 0.72 089 0.96 1.00

50 O 0.19 0.39 058 0.73 0.89 096 1.o0
0.10 100 O 0.21 0.42 0.60 075 0.89 096 1.00
150 O 0.24 0.45 0.62 0.75 090 096 1.00
200 O 027 0.48 0.64 0.77 090 096 1.00
10 O 0.19 0.39 0.58 0.72 0.89 097 1.00
50 O 0.23 0.44 0.64 074 0.90 0.97 1 .o0
0,15 1O0 O 0.31 0.53 0.68 0.78 091 097 1.00
150 O 0.37 0.61 0.74 083 092 097 1.00
200 O 0.41 0.66 080 0.87 095 098 1-00
LO O 0.21 0.42 0.61074 090 097 1.00
50 O 030 0.54 071 082 092 097 1-00
0.20 1O0 O 0.41 0.66 0.80087 094 098 1.o0
150 O 0.44 0.71 0.86091 096 0.98 1.00
200 O 0.45 0.74 0.89093 097 099 1.00
10 O 0.28 0.50 0.65 077 0.91 097 1 .o0
50 O 0.42 0.69 083 088 093 0.97 1.00
0.30 100 O 0.45 0.74 0.90 094 096 098 1 .o0
150 O 0.46 076 092 095 097 099 1-00
200 O 0.47 0.77 092 096 098 0.99 1
 43
side of the rib.These numerical tables are valid in the case of distributed
loads (uniform or practically uniform,triangular,parabolic or sinusoidal
distribution).
In the case ofa locally concentrated load aptdied to a zone of width a the
values given in Tables 6.1and 6.2should be multiplied by a reduction coeffi-
cient as shown in Table 6.3.

Table 6.3

21
Values of-
Reduction coefficient for b-bo
be -bo
__ O 10 20
b-bo
~~ ~

as0 0.6 0.7 0.9

LocaI1y
concentrated O < a < li10 linear interpolation between the extreme values
load above and below

a 2 1/10 1 .o 1.0 1.0

Furthermore,in the vicinity of a free support the flange width (be-b0)/2

to be taken into account on each side of the rib must not exceed the distance
from the support to the section considered.
For continuousT-beamsand,more generally,for T-beamsin which,for a
particular loading condition, changes occur in the sign of the bending
moment,the method of determining the effective width of the compressive
flange is the same as that for simply-supported T-beams,provided that
instead of the free span length 1 the distance between the points of zero
bending moment is adopted.
If the rib of the T-beamconsidered is joined to the compressive flange by
haunches having a width b, and a depth h,, then the actual width bo of the
rib should be replaced in the calculation by a fictitious width b, which
should be taken as equal to:
b, = bo+2b,$b,<h,
b, = bo+2h,ifb,>h,
The maximum compressive strain of the concrete in uniaxial (simple or
composite)* bending is conventionally taken as equal to 0.2%.
Experimental results show that the value of the maximum compressive
strain (shortening) of concrete in the ultimate limit state may vary con-
siderably,depending on the rate and duration of the load application,the
environment,the geometrical shape of the section,the percentage of rein-
forcement,the eccentricity of the direct (or normal) force applied,and the

*‘Composite’bending is the combination of bending moment and direct force (longitudinal

force in the member); if there is no direct force acting,‘simple’bending occurs.
44
correspondingposition of the neutral axis.Figure 6.1. which relates to short-
term tests for simple bending,exemplifieshow large these variations can be.
It is thus seen that,other things being equal,the effect of the geometrical
shape and the reinforcement percentage of the section may cause variations
in the maximum compressive strain of concrete and that these may range
from:
0.19-0.21%for a T-section
0.21-0.34%for a rectangular section
0.37-0.52%for a triangular section
These values would be even larger in the case of long-termtests and could
even attain 1-1.5 %.
However,for practical purposes and for the sake of simplicity,it appears
reasonable to limit the maximum compressive strain of the concrete to a

Figure 6.1

uniform and conventional value of 0.2%applicable to all flexural and com-

pressive loadings,i.e.,to the entire range from simple bending to concentric
compression (without bending). Experimentally the value of 0.2% corres-
ponds to the case of sections entirely in compression or to the case of the
compression flange of a T-beam section loaded in bending,i.e.,they corres-
pond to the lowest possible value of the maximum compressivestrain of the
concrete in the ultimate limit state of bending.
The value of 0.2% is therefore always on the safe side.Besides,the effect
of variations in the maximum compressive strain of the concrete upon the
final result of the analysis for bending and compression is quite small in
practice,so that the error associated with adoptingthe conventional uniform
value of 0.2%can be regarded as acceptable.
The diagram representing the distribution of the stresses in the concrete
of the compressivezone is assumed to be a rectanglewhose width is taken as
equal to the basic strength Obof the concrete (defined in Section 43.3) and
 45

Figure 6.2

0;
100min
u)

2
4
t /20rni" I,

Figure 6.3
46
whose depth is a function of the distance x from the neutral axis to the most
compressed face of the member,namely:
$X if x 6 h (simplebending and composite bending)
x -ah
-. h if x 3 h (eccentriccompression)
x-3h
where h denotes the effective depth of the section.
Compressive stress-strain diagrams for concrete can be plotted, as a
function of the various parameters affecting the strain,from the results of
theoretical and experimentalresearch.Figures 6.2and 6.3,given for approxi-
mate guidance,show the effect of the age of the concreteat the time ofloading
(28days or 1 year), the rate of loading and the duration of loading.
Further research has also shown,however,that in the normally encounter-
ed cases the most unfavourable loading generally correspondsto the applica-
tion of the full design load during one day,when the concrete is 28 days old.

(rectangular section)

Neutral axis

m,, :A.Ü,
Figure 6.4

Under these conditions it becomes possible to replace the whole families of

diagrams taking account of the various parameters by a single diagram
which can be used for any grade of concrete,any position of the neutral axis,
and any geometrical shape of the section.This single diagram, which is
justified by the statistical interpretation of a large number of tests and com-
prises all the most unfavourable loading conditions,is formed by combining
a second-degreeparabolic diagram with a rectangular diagram whose width
is taken as eighty-two hundredths (0.82)of the basic strength 5; of the
concrete (parabolic-rectangulardiagram), as shown in Figure 6.4.
For practical purposes this diagram approximates closely enough to a
second-degreeparabolic diagram whose vertex is located at the most severely
compressed face of the member,the stress at that face being taken as equal
to the basic strength 6 ofthe concrete (parabolic diagram,see Figure 6.5).
The differences in the results obtained in the analysis of the ultimate limit
state with this diagram are slight and are on the safe side.
 47
Another simplificationconsists in transforming the parabolic-rectangular
diagram into a rectangular diagram (‘stress block’) whose depth is limited
to a certain fraction of the depth of the compressivezone.This fraction takes
account of the position ofthe neutral axis and makes possible a continuous
analysis for all cases ranging from sections subjected to bending,on the one

(rectangular sect ions)

- Neutral axis

L- Figure 6.5
N a = A.äa

(rec tang u Lar sect ions)

Figure 6.6

hand, to sections subjected to compression,on the other (truncated rect-

angular diagram,see Figure 6.6).
With the truncated rectangular diagram a major simplification in all the
ordinary design calculations for structural sections,which constitute the
bulk of the routine work in engineering design offices,can be effected.Yet
there are certain cases where the discontinuity of this diagram in the vicinity
48
of the neutral axis may present an obstacle to accurate analysis of the actual
behaviour of a member with a cross-sectionof complex geometrical shape.
Under these exceptional conditions it is advisable to use the parabolic-
rectangular or the parabolic diagram.The same applies to all cases where
the discontinuity of the truncated rectangular diagram is liable to cause
difficulty in establishing a programme for electronic computation.
The value of the resisting moment due to the compressive stresses in the
concrete (apart from the contribution of any compressive reinforcement
provided) is limited to the value of the moment (with respect to the re-
inforcing bars in tension or located nearest the least compressed face) of the
forces acting upon the total effective section assumed to be subjected to a
uniform stress equal to three-quarters(0.75)of the basic strength o b of the
concrete.
The introduction of an upper limit to the resisting moment is equivalent
to gradually reducing the concrete stress from the value corresponding to

Figure 6.7

the basic strength abto the value 0.75Zbaccording as the depth of the com-
pressive zone increases from a certain limit value (at which this upper limit
of the moment is reached) until it becomes equal to the effective depth
h (see Figure 6.7).
The upper limit of the bending moment is equal to NI,. z,so that,for a
rectangular section,this becomes :
h
0.7501,.b . h .- = 0.375b h2%
2
This upper limit value of the bending moment is assumed to be inde-
pendent of the quality of the concrete, although certain experimental
investigations show a variation in this value. This variation has been neg-
 49
lected for the sake of simplicity,but it has been taken into account in deter-
mining the values of the reduction coefficient yconcrete(see Section 4.3.3).

Taking The Tensile Reinforcement Into Account

The basic stress-strain diagram for the steel is derived from the standard
stress-strain diagram (Section 3.1.3)by affine transformation parallel to

Figure 6.8
ea I
eo.*------- _ _ _ _ _ _ ,---- ,I-- I
,r I 1 I
II‘ I I I
I I
191 /I I
‘II I I I
,I 1 / I

- III I I ,I / Basic /diagram

I

the ascending straight portion corresponding to Hooke’slaw,in the ratio:

1 - 1
= 0.556
Ysteei 1.80
The tensile strain of the steel in the ultimate limit state is assumed to have
an upper limit value of 1 %.
50
By way of example the basic stress-straindiagram for an ordinary steel is
shown in Figure 6.8.
Similarly,Figure 6.9is an example of the basic stress-straindiagram for a
cold-worked steel:
The 1% steel strain is regarded as corresponding to maximum plastic
deformation.For greatly under-reinforcedmembers,in which actual failure
x,
is reached only at steel strains exceeding 1 the ultimate limit state is
defined therefore,not by failure,but by excessive plastic deformation.The
corresponding reduction in the resisting moment is negligible in the case of
cold-workedsteels if, in the calculation,the steel stress is taken as the value
corresponding to the limit strain of 1 %.

Taking The Compressive Reinforcement Into Account

The compressive reinforcement can be taken into account only if the bars
of diameter 4 forming this reinforcement are secured by means of trans-
verse binders or stirrups of suitable section and spaced less than 12 4 apart.
Since the compressivestrain of the steel in the ultimate limit state is deter-
mined on the assumption that plane sections remain plane,the compressive
.steelstress to be introduced into the strength calculation is derived from the
basic compressive stress-straindiagram,which is assumed to be similar-
subject to reversal of the algebraic signs-to the basic tensile stress-strain
diagram as previously defined.
If d' denotes the distance from the centroid of the compressive reinforce-
ment to the most compressed face and x denotes the depth of the compressive
zone of the concrete in the ultimate limit state,then the compressive strain
of the steel,determined on the assumption that plane sections remain plane,
will be equal to :

The compressive stress corresponding to the strain E; can be read from

the basic stress-straindiagram for the steel.As this strain is always less than
0.2%, the use of the basic compressive stress-straindiagram is confined to
the zone comprised between the origin and a maximum strain of 0.2%.

Biaxial Bending (Simple Or Composite)

The analysis of the ultimate strength for biaxial bending,i.e.,the calculation

of the ultimate limiting value of the bending moment and the corresponding
direct (longitudinal)force,should be based on the same assumptions as the
analysis for uniaxial bending.
 51
Bending is called ‘biaxial’when the axis of the bending couple does not
coincide with one of the central axes of inertia of the section.
However,the analysis for biaxial bending can generally be avoided in the
frequently encountered case of edge beams of ribbed slabs. Because of the
asymmetry,these edge beams are subjected to biaxial bending,but for practi-
cal purposes their sections can be designed as though for uniaxial bending,
because the stiffness ofthe slab to which they are joined is generally sufficient
to prevent any significant deviation of the neutral axis.

Concentric Compression

Members Without Binding

The analysis of the ultimate strength for concentric (or axial) compression
should be based on the same fundamental assumptions as the analysis for
uniaxial bending.
These fundamental assumptions are valid in an entirely continuous
fashion throughout the range of simple or composite uniaxial bending,
i.e.,the whole range extending from simple bending to concentric com-
pression.
The load capacity analysis for concentric compression in the limit state is
considered to be adequate for all compressionmembers whose Euler slender-
ness ratio does not exceed 40.Otherwisethis analysis should be supplemented
by an analysis for the resistance capacity in the limit state of instability.
The limiting value for the concrete and steel strains in the ultimate limit
state is conventionally taken as 0.2%.
This corresponds to the strain hypothesis relating to uniaxial bending.
The compressive concrete stress in the ultimate limit state is taken as
seventy-fivehundredths (0.75)of the basic strength O;, of the concrete (as
defined in Section 4.3.3).
This corresponds to the stress hypothesis relating to uniaxial bending.
The compressive reinforcement can be taken into account only if the
bars of diameter q5 forming this reinforcement are secured by means oftrans-
verse binders or links of suitable section and spaced less than 124 apart.
The compressive steel strain in the ultimate limit state is taken as 0.2%.
The compressive steel stress to be adopted in the strength calculation is
derived from the basic compressive stress-straindiagram,which is assumed
to be similar -subject to reversal of the algebraic signs-to the basic tensile
stress-straindiagram as defined above.
This correspondsto the hypothesis relating to uniaxial bending.
O n application of this hypothesis and the preceding one,the value of the
ultimate limiting value of the force is obtained as the sum of:
1. a force equal to the product of the total cross-sectionalarea of the
concrete of the member and seventy-fivehundredths (0.75)of the basic
strength a;,of the concrete,and
2. a force equal to the product of the total cross-sectionalarea of the com-
52
pressive reinforcement and the stress O:, corresponding to a strain of
0.2% in the basic stress-straindiagram for the steel.
Therefore: NI = B x 0.756+A’%’

Members Reinforced By Binding

Binding can be employed for compression members with the object of
improving their strength with regard to local loads,their impact strength,
and their ductility at fracture,but it should strictly be confined to special
purposes.
The principle of ‘binding’is to provide lateral restraint and thus prevent
transverse expansion of the concrete loaded in compression.This restraint
is obtained by means of very tightly and closely applied reinforcement,with
the result that an apparent increasein the strengthofthe concrete is achieved.
In practice, binding should be used only for local reinforcement,e.g.,at
hinges and bearings having to transmit concentrated loads.O n no account
is it to be treated as a mere expedient of calculation by means of which it is
always possible to satisfy the requirements for checking the ultimate limit
state for compression.
The effect of binding can be taken into consideration in the analysis of the
ultimate limit state for concentric compression only if the following four
conditions are satisfied:
1. The height of the zoneprovided with binding in a compression member
should not exceed twice the least transverse dimension of that member.
Binding is efficient only in short members.In the case of long mem-
bers provided with binding,the concrete within the core enclosed by the
binding is in a state of plastification and thus creates a danger of
buckling which may bring about premature destruction by bursting.
2. The least transverse dimension of the zone provided with binding
should be not less than 0.25m.
If the transverse dimensions of the member are too small,it may not
be possible to construct the binding properly.
3. The binding should be formed by either of the following methods of
installing transversereinforcement,no other method being permissible.
(a) Binding by helical reinforcement or hoops:
The binding of a member of circular cross-sectionmay be in the
form of a continuous circular helix or of a series of closed hoops.
The pitch of the helix or the spacing of the hoops should not exceed
one-fifth(0.20) of the diameter of the core enclosed by the binding.
(b) Binding reinforcement in mats:
The binding in a member of rectangular cross-sectionmay con-
sist of mats formed of bars bent hairpin-wiseto and fro,these mats
being placed with the bars extending alternately in mutually per-
pendicular directions and suitably anchored into the concrete.
The spacing of these mats should not exceed one-fifth (0.20)of the
least dimension of the core provided with binding reinforcement.
 53
4.The geometrical percentage of binding reinforcement,referred to the
total volume of the zone provided with binding, should be at least
0.6%.
The calculation of the ultimate strength of a member provided with
binding should be based, not on the total cross-sectionof the member, but
only upon the cross-sectionof the core enclosed by the binding and bounded
by the outer edge thereof.
In a member provided with binding,the compressiveconcrete stress in the
limit state should be taken as seventy-five hundredths (0.75)of the basic
strengthobof the concrete (definedin Section 4.3.3),multiplied by an amplifi -
cation coefficient that takes the effect of the binding into account ('binding
coefficient').
Depending on the method of transverse reinforcement (binding)applied,
the binding coefficient is taken as one of the two following values,but must
never exceed the upper limit of 2.50:
(a) Binding by helical reinforcement or hoops:

(b)Binding reinforcementin mats:

In these expressions Ziodenotes the geometrical percentage of longitudinal

compressive reinforcement and Oidenotes the geometrical percentage of
transverse reinforcement employed as binding, both percentages being
referred to the total volume of the zone provided with binding.In the second
expression denotes the diameter (in centimetres) of the transverse rein-
forcementemployed as binding.

Concentric Tension

In the analysis ofthe ultimate strength for concentrictension only the strength
of the tensile reinforcement should be taken into account,ignoring the tensile
strength of the concrete.
This basic assumption applies not only to the case of concentric tension,
but also to tension occurring in combination with bending,insofar as the
entire section of the concrete is actually in tension.If not,the compressive
zone of the concrete section may be taken into account in calculating the
ultimate strength.
The analysis of the ultimate strength for concentric tension should be
supplemented by a check on the limit state of cracking (Section 6.1.3), having
due regard to the environmental and service conditions of the member
considered.
54
It may occur,however,that no cracking whatsoever is acceptable because
of special circumstances (e.g., structures enclosing nuclear reactors). In such
cases the analysis for the limit state of failure (ultimate strength analysis)
should be replaced by an analysisfor the limitstateofconcrete strainwhereby
it can be checked that the strains developed in the concrete remain below the
strain associated with tensile failure (which may, for example,be taken as
0.01%). In addition,it is necessary in such cases to take the unfavourable
effect of shrinkage into account,while the possible favourable effect of creep
should be ignored.Finally,despite these very severe conditions,the tensile
strengthofthe concrete-even though it is assumed to be uncracked -should
be neglected,just as is ordinarily done.

6.1.2 L I M I TS T A T E OF I N S T A B I L I T Y

Fundamental Design Assumptions

The analysis of the critical strength for uniaxial bending in the limit state of
instability,i.e.,the calculation of the limiting value of the buckling moment
and the corresponding direct (longitudinal)force,is reducible to the ultimate
strength analysis (Section 6.1.1)by conventionally introducing a com-
plementary eccentricity for the direct force.*
This assumption,which is applicable to the analysis of buckling in the
case of uniaxial bending,may be extended to the analysis of buckling in the
case of biaxial bending.It should not,however, be applied to the analysis
of twist-bend buckling, i.e.,checking the strength of a member with regard
to lateralinstabilitydue to buckling associated with bending;in that case the
designer should apply the standard theories of structural stability.
For practical purposes the above assumption consists in transforming the
analysis forthe limitstate ofinstability into the usual analysis of the ultimate
limit state for eccentric compression or composite bending.
Consider a member subjected to that longitudinal compression.At any
particular cross-sectionof that member the direct force has a total eccen-
tricity comprising:
(a) the known or intentional eccentricity due to deliberate structural
arrangements;
(b)the accidental and inevitable eccentricity due to inaccuracies of
construction;
(c) the deflectiondue to the flexural deformations arising from the two
foregoing eccentricity components (a)and (b).
It thus appears that, in terms of actual structural behaviour, concentric
compression is merely a hypothetical ideal condition; in reality,bending
always complicates the issue. For this reason the validity of applying the
standard theory ofelastic buckling ofreinforced concrete members may well
*‘Direct îorce’ (or ‘normal force’)denotes a force acting parallel to the longitudinal axis of a
member.If the line of action of this force coincides with the centroidal axis of the member,the
condition is called ‘concentric’(or‘axial’)compression or tension, as the case may be.
 55
be questioned.Accordingly,for checking a member with regard to the limit
state of instability,it appears advisable to turn to an analogy calculation
based on the knowledge of the maximum deformations of the member in the
ultimate limitstateand enabling the critical buckling strength to be estimated
in a simple manner by applying the usual ultimate strength analysis for
eccentric compression and composite bending.
In short, the analysis for buckling involves adding a bending moment
(referredto as the ‘additionalbending moment’)to the initial loading system
on which the design of the section for the ultimate limit state is based.
The analysis of the critical strength in the limit state of instability is
necessary for all compressionmembers whose Euler slendernessratio exceeds
40.Otherwise the ultimate strength analysis (i.e.,the ordinary analysis of the
strength of the member in the limit state of failure)will suffice.
In accordance with Euler’stheory the slenderness ratio A ofa member is
defined as the ratio of the effective length 1, to the radius of gyration i of the
(net)concrete cross-sectionin the corresponding plane :
A = y
1,
1

The effective length 1, depends upon the structuralfeaturesin any particular

case.If 1 denotes the geometrical length of the member under consideration,
the effective length 1, is equal to:
1, = 21 if the member is free at one end and fixed (rigidly restrained)
at the other;
1, = 1 if the member is hinged (pin-jointed)at both ends or, alter-
natively,if it can be considered fixed at both ends,but these
ends can move in relation to each other in a direction per-
pendicular to the longitudinalaxis ofthe member and situated
in the principal plane for which buckling is being investigated;
1, = l/ J 2 if the member is hinged at one end and fixed at the other;
1, = 112 if the member can be considered fixed at both ends and those
ends cannot move.
In the frequently encountered case of a multi-storey building in which
the continuity ofthe columns and their sections is assured,the effectivelength
1, may be equal to:
1, = l/ J 2 ifthe ends ofthe column are either rigidly fixed to a foundation
block or arejoined to floor beams which have at least the same
moment of inertia as the column in the direction considered
and which extend on both sides of it;
1, = 1/1.15 in all other cases.
In practice, one of the difficulties of the analysis consists in correctly
estimating the actual degree of fixity (degree of restraint) of the connection
between ‘themember under consideration and the other elements of the
structure.A convenient empiricalmethod consists in comparing,at eachjoint,
in a given plane,the stiffness of the supporting member with the sum of the
stiffnesses of the supported members:
56
If the ratio of the stiffness of the supporting member to the sum of the

-
stiffnessesof the supported members is less than 25,the joint may be regarded
as providing ‘full fixity’.O n the other hand, if that ratio is 25 or more, the
joint should be regarded as a ‘hinge’.
Other difficulties may also arise in estimating the geometrical length of
the member under consideration in the case of special constructional
features:
(a) Flat-slabfloors (seeFigure 6.10):

Figure 6.10 1
In this case the geometrical length 1 of the column is evidently the clear
distance between the floors.
(b)Mushroom floors (see Figure 6.11):

Figure 6.11 c
In this case the geometrical length 1 of the column is taken as the distance
between the top of the lower floor and the base of the column head.
(c) Ribbed floors (see Figure 6.12):
 57
Here the geometricallength 1 ofthe column is taken as the distancebetween
the top of the lower floor and the underside of the upper beam, which is
assumed to intersect the column and extend on both sides of it.
(d) Group of braced columns (see Figure 6.13):

Figure 6.13 D
The geometrical length 1 of the column is taken as the clear distance
between two consecutive cross-membersin each vertical plane,provided that
such cross-memberscan themselves resist lateral bending and are at right
angles to the columns (or do not deviate by more than 15" from the per-
pendicular thereto).
(e) Cross-memberswith haunches (seeFigure 6.14):

Figure 6.14

In this case the geometrical length 1 of the column is taken as the distance
from the top ofthe lower cross-memberto the bottom of the haunch of the
upper cross-member, provided that the transverse width of this haunch is at
least equal to that of the cross-memberand to half that of the column.

Limit State Of Instability Of Columns Loaded In Concentric

Compression (Axially Loaded Columns)

The analysis of the critical strength of columns for concentric compression

in the limit state of instability,i.e.,the calculation of the limiting value of the
58
force associated with instability,is reduced to the ultimate strength analysis
of such columns under the effect of the additional moment:
M, = N.- GO
3 O00 o; ht '

where N denotes the characteristic value of the compressive direct force

(assumedto be calculated according to the theory of first order), h, the total
geometrical depth of the section (measured parallel to the buckling plane),
the instantaneous modulus of elasticity of the concrete, and oE the
Euler critical stress.
The Euler critical stress for columns of constant cross-sectionis given by
the formula:

where 1, is the effective length,I is the moment of inertia of the (net)concrete

cross-sectionin the buckling plane,and B is the cross-sectionalarea of the
concrete.O n introducing the radius of gyration i and the slendernessratio A,
the expression for the Euler critical stress may be written in the following
simple form: n2
0
; = T. EbO
E.
The additional moment to be introduced in the ultimate strength analysis
is therefore:

M, = NI.-
A2
30 O00 ht '

The analysis of the critical strength for concentric compression in the limit
state of instability is thus reduced to the analysis of the ultimate strength for
eccentric compression by application of a conventionally introduced
eccentricity e,:
A2
e, =-
30 O00 '

to the characteristic compressivedirect force N',calculated according to the

theory of first order.
This calculation procedure is valid for instantaneous loads.Creep due to
the action of loads of long duration may be taken into account as indicated
above.

Limit State Of Instability Of Columns Loaded In Eccentric

Compression (Eccentrically Loaded Columns)

The analysis of the critical strength of columns for eccentric compression

in the limit state of instability,i.e.,the calculation of the limiting value of the
buckling moment and the corresponding direct (longitudinal) force, is
 59
reduced to the ultimate strength analysis ofsuch columnsby the introduction
of an additional moment:
Ek0
M, = N.- .(h,1-e,)
3 O00 o;
which should be added to the initial system ofloadings(Mand N')as obtained
by application of the theory of first order to the structure considered (cf.
notation above). This calculation procedure is applicable only on condition
that the initial eccentricity eo of the direct force N'does not exceed the total
geometrical depth h, of the section (measuredparallel to the buckling plane),
i.e.,eo < h,.
For columns of constant cross-sectionthe additional moment to be intro-
duced into the ultimate strength analysis is (withdue reference to the expres-
sion for the Euler critical stress given above):

The analysis of the critical strength for eccentric compression in the limit
state of instability is thus reduced to the analysis of the ultimate strength for
eccentric compression or composite bending by application of a conven-
tionally introduced complementary eccentricity e, :
/I2
e, =-
3o 000 (hf+eo)
'

to the characteristic compressivedirect force N',calculated according to the

theory of first order.
This calculation procedure is valid for instantaneous loads.Creep due to
the action of loads of long duration may be taken into account as indicated
below.

Limit State Of Instability Of Plates Loaded Parallel

To Their Plane
The analysis of the critical strength of plates loaded in compression parallel
to their plane in the limit state of instability,i.e.,the calculation of the
limiting value of the force associated with instability, is reduced to the
ultimate strength analysis of such plates in bending,this being achieved by
introducing an additional moment M,,acting in the principal direction
x and an additional moment M,,acting in the principal direction y:
EbO EbO
M,, = Nk. (ht+e,) +Ny. 3 ()oo (ht+e,)
3 O00 o;,

M,,= Nk. Ebû

3 O00 OEy
(h,+e,)+N'
y 3 O00 o
' ;,
(4+e,)
where the Euler critical stress o;, in the principal direction x is assumed to
be smaller than the Euler critical stress o;, in the principal direction y.
60
Taking The Effect Of Creep Into Account
The effect of creep due to the action of a load of long duration can be taken
into account in the analysis for the limit state of instability by increasing the
values of the additional moment M,as indicated above.These values should
be multiplied by the coefficient (1+$/3), where $ denotes the ratio of the
long-termload to the total load on the member considered.

6.1.3 LIMIT S T A T E OF C R A C K I N G
Fundamental Design Assumptions
Definition Of The Limit State Of Cracking

Cracking is a phenomenon specifically associated with reinforced concrete,

for the structural elements loaded in tension or in bending are normally
cracked in their usual conditions of service. However,for the sake of the
durability of the structures it is necessary to impose certain limits upon the
cracking or,to be more precise,certain maximum values of the crack widths
are not allowed to be exceeded.These maximum widths in each case define
the limit state of cracking of the member or structure concerned,having due
regard to environmentaland service conditions thereof.
The designers’attention is called to the fact that the process of corrosion
of the reinforcement does not depend solely upon the width of the cracks,
but.also on their direction (parallelor perpendicular to the direction of the
reinforcing bars), the quality of the concrete (and more particularly its
density),the thickness of the concrete cover to the steel,and the dimensions
of the bars themselves.For this reason the systematic limitation of the crack
width cannot in all cases be considered to provide absolute protection
against corrosion,especially in cases where the maximum crack width can
be reduced only by using very thin bars which in practice are more sensitive
to the effects of corrosion.
The theoretical method of analysis for the limit state of cracking in tension
or in bending is explained in Chapter 5,Part 2 of this Manual.
But this method has not an entirely general range of application,since it
only considers cracks perpendicular to the main reinforcement.It does not
consider cracks parallel to the main reinforcement (whichcracks may,in a
beam,produce tangential actions between the web and the zone containing
the main tensile reinforcement), nor inclined cracks (whichmay develop in
web under the action of shear force,whether or not in combination with the
action of other loadings). Yet such longitudinal and inclined cracks are often
more dangerous with regard to corrosion than are the transverse cracks in
the zone containing the main tensile reinforcement.

Practical Checking Of Cracking

The analysis of the limit state of cracking in bending and in tension should,
in normal cases,be confined to checking that the Rules of good construction
 61
have been applied in the design and distribution of the reinforcement,whose
quantity has first been determined by means of the analysis of the ultimate
limit state (failure).These practical rules for the design aad distribution of
the reinforcement define in each case,for a given total section,the maximum
diameter ofthe reinforcing bars as a function of the percentage,the elastic
limit and the bond properties of the steel,and of the tensile strength of the
concrete.
In normal cases it is not necessary to make a general and systematic
theoretical analysis of the limit state of cracking for all the elements of a
structure.A detailed analysis will be required only in certain cases that call
for specialjustifications.
For practical purposes the main requirement is that the designer should
adopt appropriate constructional arrangements (pertaining more particu-
larly to the diameter and distribution of the bars) and avoid certain gross
errors ofdesign which could result in concentrations ofcracksor dangerously
wide cracks.For this reason the practical analysis for cracking should confine
itself to verifying the due application of the Rules of good construction.
These rules do not explicitly state the limiting maximum values of the
crack widths,for the checking and the measurement of such widths would
encounter insurmountable difficulties and would, generally speaking,be of
no real value.O n the other hand,these rules do take account of the environ-
mental conditions and the conditions of service under which the structures
are used.

ClassiJication Of Structures According To The Consequences Of

Cracking
To take account of their varying conditions of environment and service,
structures are divided into three classes,according to the possible conse-
quences of cracking with regard to the behaviour and durability of the
structures:
Class 1. Structural members which must ensure watertightness or are exposed
to aggressive actions.These are members in which cracking is very harm-
ful, either because they have to be watertight(e.g.,
walls oftanks,shipping
locks or dry docks) or because they are exposed to a particularly
aggressive medium.
Class 2. Unprotected ordinary structural members. In these members
cracking of the tensile zones is harmful either because they are exposed
to the effects of the weather (asis the case with outdoor structures such
as bridges and other civil engineering works) or because they are
exposed to a humid and aggressive atmosphere (asin the case of certain
industrial structures,factory roofs, or workshop buildings in which
considerable quantities of water vapour are liable to be produced).
This class may also be taken to include members which have to support
very fragile claddings or facings which would suffer harmful conse-
quences from excessive cracking and deformation.
62
Class 3. Protected ordinary structural members. In these members cracking
is not harmful and does not have any seriously adverse effects upon the
preservation of the reinforcing steel nor upon the durability of the
structure. Interior structural members of buildings in normal atmo-
sphere are,for example,included in this class.
Agreement between the theoretical analysis of the limit state of cracking
and the verification of the practical design rules is achieved for the following
maximum values for the crack widths:
Class 1 : w d 0.1mm
Class 2:w <0.2mm
Class 3:w Q 0.3mm
These values define the limit state of cracking in all cases where, for
purposes of special justification, it is necessary to carry out a complete
theoretical analysis for this limit state.However,they are,in practice,merely
to be regarded as approximate indications of the order of magnitude of the
crack widths and must on no account be taken as reference values for an
in-situinspection.

General Rules For The Checking Of Cracking

Conditions As To The Validity Of The Rules

The validity of these rules is confined to the design of sections in which the
reinforcing steel is disposed in accordance with normal practice warranted
by the engineers’experience.
It is more particularly assumed that the members contain reinforcing bars
arranged normally in layers and fairly uniformly distributed and that the
concrete cover is at least equal to the bar diameter and does not exceed 4cm.
It is also assumed that the concrete surrounding the steel is of adequate
density,since it is this that in all circumstances provides the best guarantee
against the danger of corrosion of the reinforcement.

Definition Of The ‘Embedment Section’ Of The Concrete Surrounding

The Reinforcement
The rules for the checking of cracking are based on the consideration of the
local percentage of main tensile reinforcement,referred to the ‘embedment
section’of the concrete surrounding that reinforcement.
In members loaded in tension (tie-members)or in flexural members com-
prising a tensile flange in which the main reinforcement is installed, the
embedment section ofthe concrete surrounding the main tensile reinforce-
ment is taken as the total section of the tie-memberor the flange.
In flexuralmembers not provided with a tensile flange as envisaged above,
the embedment section of the concrete surrounding the main tensile rein-
forcement is, by definition, taken as the concrete section having the same
centroid as the reinforcement.
 63
This definition is equivalent to conceiving the embedment zone of the
reinforcement as an independent ‘tie-rod’ cut off from the rest of the member
by longitudinal cracks which cause tangential actions between the web and
the embedment zone of the main tensile reinforcement (‘tie-rod analogy’).
In flexuralmembers with a tensileflange a longitudinalcrack at thejunction
of this flange with the web really does occur because in such cases the tan-
gential actions are ofvery considerable magnitude on account ofthe thinness
of the web,so that the ‘tie-rod’ is actually always cut off from the rest of the
member.
O n the other hand,in flexuralmembers not provided with a tensile flange
there is longitudinal cracking and actual formation of a cut-off‘tie-rod’ only
if the member is sufficiently thin for this to occur -in practice this is so if the
width of the member does not exceed four times the sum of the diameters
of the main reinforcing bars (ifthese are plain bars) or twice that sum (ifthey
are deformed bars providing improved bond behaviour). If this is not so,
the rules for the checking of cracking are a little more severe than the theo-
retical analysis of the limit state of flexural cracking.

Determination Of The M a x i m u m Diameter Of The Reinforcing Bars

The maximum diameter 4 of the reinforcing bars should not exceed the
larger of the following values

4<*(yy
(4
750000 0 0
in mm,oein bars)

4<$- Oe ‘1+:00, (4in mm,oein bars)

where oedenotes the guaranteed minimum elastic limitof the steel (expressed
in bars) and Gothe geometrical percentage of main tensile reinforcement in
relation to the embedment section of the concrete.
For the various classes of structures (Classes 1, 2 and 3) the values to be
adopted for )I are as shown in Table 6.4
These practical rules for the checking of cracking constitute a simplified
application of the theoretical analysis of the limit state of cracking,which
Table 6.4
Values of $ Plain round Deformed
(units:mm,bars) reinforcing bars reinforcing bars

Class 1 1.0 1.8

Class 2 2.0 3.6
Class 3 3.0 5.4

analysis is explained in Chapter 5,Part 2 of this Manual.The first rule relates

more particularly to members which contain a low percentage of reinforce-
ment and in which cracking does not present a systematic character.The
64
second rule relates more particularly to members which contain a normal
percentage of reinforcement and in which cracking develops gradually and
in a systematic fashion.
The application of these rules to the design of the main tensile reinforce-
ment,for a given type of steel and reinforcement percentage,is facilitated by
the graphs shown in Figures 6.15,6.16and 6.17,which are given for guidance.

Taking The Effect Of The Swelling Of The Concrete Into Account

In the special case of structures belonging to Class 1 which are in permanent

contact with water the favourable effect of the swelling of the concrete may
be taken into account by substituting for the guaranteed minimum elastic
limit a fictitiouselastic limit which should be taken as:
oe-500 bars in the case of plain reinforcing bars
ge-800 bars in the case of deformed reinforcing bars

It has been experimentally verified,more particularly for the walls ofwater

tanks,that the phenomenon of swelling of the concrete does actually reduce
the width of the cracks.The foregoing rule is based on the interpretation of
the experimental measurements.
The practical application of this rule to the design of the reinforcing bars,
for a given type of steel and reinforcement percentage,is facilitated by the
graph shown in Figure 6.18,which is given for guidance.

Rules For Particular Types of Structural Members

Deep Beams

If the depth of the web (in metres) exceeds the value l-10-40, (in bars),
the designer should provide longitudinal web reinforcement-so-called‘skin
reinforcement’-of the same grade of steel as the main tensile reinforcement.
The geometrical percentage of such ‘skinreinforcement’,referred to the web
section excluding the embedment section of the main tensile reinforcement,
should be at least 0.05%at each of the two faces.Furthermore,the individual
bars of this reinforcement should not be spaced farther than 20cm apart.
In deep beams the concrete above the embedment zone of the main
tensile reinforcement is subject to complex shearing and tensile stresses.The
resulting inclined forces cause concentration of a number of cracks (pre-
viously formed in the main reinforcement zone)into a single crack in the web.
The width of this crack is significantly greater than that of each of the
elementary cracks.
One of the appropriate means of distributing the web cracking consists
in providing the web with longitudinal reinforcing bars of small diameter
which are disposed close to the faces (‘skinreinforcement’).Experience has
shown that if a sufficientpercentage of such reinforcement is provided,it is
certainly effective.
 65

Plain bars Deformed bars

'Lain bars Deformed bars

Figure 6.15 Class I

mm
I Plain bars I Deformed bars I
@

Figure 6.16 Class 2

 67

Figure 6.17 Class 3

68

+
rnaxirnur
rnrn Plain Deformed bars
bars
@O I ( I I l I l , , , ~ , l l lI I I I ( l I 1 I
3( IO !
I I I I I I I I I
61 8

Eiastic
limit
h bar
í kg/rnrn2ì
reference
value
Figure 6.18 Members in permanent contact with water
 69
Also, the bars constituting the group of main reinforcing bars may be
distributed in a graduated fashion over a fairly substantial proportion of
the bottom part ofthe beam.In that case it will be necessary to take the precise
positions of the bars into account in calculating the lever arm of the internal
forces and in the ultimate strength analysis for the beam.

Slabs A n d Plane Structures

In slabs and plane structures reinforced in one or in two directions and less
than 30 cm thick the spacing of the individual bars of the main reinforcement
should not exceed the values (which are given in centimetres) in Table 6.5.
Table 6.5

Guaranteed elastic 2000 3000 4000 5000 6000

limit of steel (I, bars bars bars bars bars

Plain round bars 20 20 20 20 20

Deformed bars 20 20 20 17.5 15
Plain drawn wires (welded fabric) 20 20 17.5 15 10

In addition,if the main reinforcement consists of welded fabric,the larger

sidelength b of the mesh should not exceed three times the smaller sidelength
a (hence:b d 3a).Furthermore,the average ofthese two dimensions [(a+b)/2]
should not exceed the values given in the above table.

6.1.4 L I M I TS T A T E OF D E F O R M A T I O N

Fundamental Design Assumptions

Definition Of The Limit State Of Deformation

Deformations are a phenomenon specifically associated with reinforced

concrete,since structuralmembers loaded in bending,compression or tension
are normally subjected to deformation under their usual conditions of
service.However,for the sake of serviceability and durability of structures,it
is necessary that certain deformation limits-or, more precisely, certain
maximum values of the deflections-should not be exceeded.In each case,
having due regard to the nature and the extent of possible damage due to
deformation,these maximum values define the limit state of deformationof
the member or structure considered.

Practical Calculation Of Deformations

In calculating the deformations of prismatic structural members subjected

to bending and compression,it is necessary to take account as accurately
70
as possible of the various physical and mechanical phenomena that charac-
terise the elasto-plastic behaviour of the concrete in compression and the
cracking of the concrete in tension.
In view of the uncertain and random character of many parameters,the
deformations can,with sufficient accuracy for practical purposes,be calcu-
lated by means ofa general method based on the three following fundamental
assumptions:
1. The geometrical cross-sectionsof the member under consideration
should first be made ‘homogeneous’by multiplying the steel cross-
sectional area by the modular ratio,i.e.,

E,
B+-.A
E;,
where A and B denote the steel and concrete cross-sectionalareas
respectively,E,denotes the modulus of elasticity of the steel (forwhich
a value of 2 100000 bars should be adopted,cf. Section 3.1.3)and
the modulus of elasticity of the concrete (for which the instantaneous
modulus Eboor the long-termmodulus Eb, should be adopted,accord-
ing as loads of short or long duration are involved,cf. Section 3.2.2.
2. The basic values of the tensile steel strain E, at the various sections
should, for the purpose of introducing them into the conventional
calculation of the deformations in the case of a member that is not
entirely in compression,take account of the cracking of the concrete
in tension and also allow for the corresponding effects of the bond of
the main tensile reinforcement.
3. The basic values of the compressive concrete strain E; at the various
sections should,for the purpose of introducing them into the conven-
tional calculationofthe deformations,take account ofthe instantaneous
plasticity,the long-termplasticity and the shrinkage of the concrete.

Permissible Maximum Deflections

In the case ofbridges and civilengineeringstructuresthe maximum deflections

defining the limit state of deformation should be specified by the building
owner. These values cannot be estimated by conventionalised general
methods, for they depend directly upon the nature and the conditions
of service of each particular structure.
If excessive deformation is liable to endanger the stability of the structure,
the designer should also investigate the consequential effects of the deforma-
tions upon the behaviour of the other elements of the structure.If need be,
he should modify the arrangements thereof so as to achieve a suitable
reduction of the magnitude of such deformations.This check should be made
more particularly in a case where the structure may,under service conditions,
be subjected to dynamic actions which can bring about a very great increase
in the deflections by resonance.
 71
In the case of buildings for public or private use the following maximum
values ofthe deflections defining thelimitstateofdeformationmay be adopted
by the designer, subject to prior consent from the building owner and
provided that excessive deformations can on no account endanger the overall
stability of the structure or the individual stability of certain parts thereof:
(a)If excessive deformations of a structural member may cause technical
damage to other,non-load-bearing,members of the structure (fragile
partitions or claddings), the maximum value ofthe deflection under the
action of the characteristic loadings (defined in Section 4.2)should
be taken as one three-hundredth(1/300)of the span.
(b)If excessive deformations of a structural member may cause psycho-
logical or aesthetic harm,the maximum value of the deflection under
the action of the characteristic loadings (definedin Section 4.2)should
be taken as one three-hundredth(1/300)of the span for service floors
in buildings for public use (schools,exhibition buildings,sports halls,
assembly halls, etc.) and as one hundred-and-fiftieth(1/150)of the
span for service floors in buildings for private use (residentialbuildings).
This latter value should also be adopted for the roofs (flator sloped)
for all buildings,whether for public or for private use.

General Rules For Calculation

The general calculation of deflections and deflection curves comprises the

following steps:
(a)establishing,for a sufficient number of sections distributed along the
member considered,the geometrical expression for the curvature as a
function of the basic strains of the steel and concrete in the limit state
of deformation;
(b)deducing therefrom the relationship for the variation of the curvature
over the entire length of the member;
(c) determining the deflection curve by means of a double integration.
The maximum ordinate of the deflection curve defines the deflection of the
member.

Cracked members
For members subjected to simple bending or composite bending which are
partially in tension,and cracked,the curvature at any particular section with
the abscissa x is equal to :

where (d2f)/(dx2) is the second derivative of the deflection curve with

respect to the abscissa of the section considered,while E, and E; are the basic
72
steel and concrete strains estimated for the limit state of deformation,and
h is the effective depth of the section.

Uncracked members

For members subjected to eccentric compression which are entirely in com-

pression,and not cracked,the curvature at any particular section with the
abscissa x is equal to:

1 - d2f- I ~ ; l - I ~ b i l
r dx2 - ht
where (d2f)/(dxZ) is the second derivative of the deflection curve with
respect to the abscissa of the section considered,E; is the basic compressive
strain of the concrete at the most compressed fibre (corresponding to the
basic strength obof the concrete,see Section 6.1.4), is the compressive
strain of the concrete at the least compressed fibre,and h, is the total geo-
metrical depth of the section.

Simplified Rules For Ordinary Buildings

For buildings intended for public or private use and not of an exceptional
character the analysis of the limit state of deformation may be dispensed
with by limiting the slendernessratiosofthe flexuralmembers ofthe structure,
i.e.,by not exceeding a maximum ratio between the span 1 and the effective
depth h of these members.
This maximum value of the slendernessratio should be taken as:

where oedenotes the guaranteed minimum elastic limit of the steel (inbars),
o the mechanical percentage of main tensile reinforcement,$the proportion
of permanent loads and Fixed superimposed loads in relation to the whole
the maximum value of the permis-
of the characteristic loadings,and (j,71)rnax
sible deflections (asspecified in Section 6.1.4).
This check is applicable only to flexural members whose mechanical
percentage of reinforcement o does not exceed 0.25.
A further simplificationmay be adopted for ordinary buildings not of an
exceptional character. For these it is permissible to adopt the following
values:
$ = i for service floors of buildings for public use
$ = i for service floors of buildings for private use
$ = f for roof structures of all buildings
 73
The interrelation of this condition of limiting slenderness and the detailed
methods of calculating the deflection is dealt with in Chapter 6,Part 2 of
this Manual.

6.2 TANGENTIAL ACTIONS AND STRESSES

6.2.1 G E N E R A L DESIGN R U L E S

Definition Of Connectors

The strength of a member subjected to tangentialactions should be ensured,

not by the concrete,but by transverse reinforcing bars which extend across
the surfaces on which these tangential actions are exerted and which are
suitably anchored on both sides ofthose surfaces.Such transversereinforcing
bars will be referred to as connectors or connector reinforcement.
Certain surfaces in the interior of structural members are subjected to
so-called‘tangentialactions’.For example,this occurs at the neutral plane
of a flexural member (thisplane is the locus of the neutral axes of the suc-
cessive cross-sections);also,at the junction plane between the flange and the
rib of a T-beam.In a more general sense,‘tangentialactions’are taken to
comprise all the forces and effects due to shear,bond and torsion.
Because of its low tensile strength and shear strength,concrete is not by
itselfable to resist the tangential actions,and cracking occurs.In the absence
of connectors,this cracking would tend to cause dislocation of the member
as a result of separation of the ‘blocks’on both sides of each crack.O n the
other hand,if connectors are provided,the interconnection of these ‘blocks’
will resist the dislocation tendency of the member.
But connectors are effective only if they are perfectly anchored in the
uncracked zones and can,by thus remaining secured to those parts of the
member which remain intact,ensure the general equilibrium of the forces
involved.

Fundamental Design Assumptions

The analysis of the strength with regard to tangentialactions in the ultimate

limitstate should be based on the following three fundamental assumptions:
1. The cracks that develop in the concrete as a result of the tangential
actions are assumed to be inclined at 45”.
In other words,the cracks which develop iri the vicinity of a plane surface
P subjected to a tangential action are assumed to be (see Figure 6.19):
(a) perpendicular to the plane of the tangential action,i.e.,perpendicular
to the plane N,which is itself perpendicular to the surface P under
consideration and contains the vector representing the tangential
stress z (orshear stress) prior to cracking;
(b)inclined at 45” in relation to the plane surface under consideration
74
and oriented in the direction in which the tangential action tends to
thrust the concrete 'blocks'separated by the cracks.
This assumed inclination of 45" is not always in agreement with reality,
as experimentally determined,more particularly in the case of shear force
loadings in the parts subjected to bending;this inclination is therefore only

an approximation-a safe one -which it may be possible to improve upon

in certain cases.
Furthermore, the normal components of the forces liable to act upon

'T
the surface under consideration may be neglected in the calculation of the
I
I

Figure 6.20

strength with regard to tangential actions.This is more particularly the case

at the flange root plane of a T-beam(see Figure 6.20).
In this plane may act a bending couple, which arises from the flange

Connectors
456 a-S90°
Figure 6.21
 75
itself and has its axis parallel to that of the beam,and a secondary direct
force produced by the curvature of the compressive stress trajectorieswithin
the flange; these influences may permissibly be neglected.
2. The connectors should be disposed either perpendicularly to the
plane surface under consideration or be inclined to that surface at an angle
or more than 45" measured in the direction opposite to that of the cracks
(see Figure 6.21).
With the assumptions 1 and 2 it is possible to calculate the tensile stress
onin the connector reinforcement.
Let a denote the inclination of the connectors in relation to the plane
surface P under consideration (abetween 45"and 90"),t the spacing of the
connectors, A, the individual cross-sectional area of each connector, b
the thickness ofthe member,and q0the geometrical percentage ofconnector

reinforcement referred to the volume of concrete comprised between two

consecutive connectors.Then:
At
'LTIto =
t.b.sina
Consider the forces acting upon the unit area of the plane surface P,
Figure 6.22.
The compressive force acting in the concrete 'struts',inclined at 45" and
compressed by the stress ob, is equal to:

The tensile force in the connector reinforcement,which is subjected to the

tensile stress on,is equal to:
o,,.A - q0.
-- on.sin CI
t.b
The resultant ofthese two forces must equilibrate the tangential force 7.
O n expressing this equilibrium condition by projection on P and on the
normal to P,we obtain:
7

i zuto .(sina +cos a) sin a

o;= .
27 sin a
sin a +cos a
76
If the connectors are perpendicular to P,we obtain
z

CI = 90"
o;,= 22
Similarly,in the extreme case where the cracks are inclined at 45" with
respect to P:

CI = 45" I"=-
o;,= 2
2

These relationships are called the Rule for connector reinforcement.

3. The concrete 'struts'separated by the cracks (assuming these to be in-
clined at 45"with respect to the plane considered)are assumed to be loaded
in concentric compression,i.e.,'axially'loaded.The connector reinforcement
(inclined at an angle of between 45"and 90" with respect to the same plane,
but leaning the other way) are assumed to be loaded in concentric tension.
By virtue of their interconnection the struts and the connectors form a multi-
ple lattice system which should be able to equilibrate the tangential actions
(latticehypothesis).
In the special case ofa beam loaded by a shear force the 'lattice hypothesis'
enables the cracked beam to be treated as a multiple lattice girder whose:
(a)compressivechord is formed by the compressive zone;
(b)tensile chord is formed by the main tensile reinforcement;
(c) web members are formed by the concrete struts in compression and
by the connector bars in tension.

6.2.2 SHEAR FORCE

Design Rule For Shear Force

The calculation of the shear strength in the ultimate limit state should be
based on the three fundamental design assumptions for determining the
strength with regard to tangential actions, more particularly the 'lattice
hypothesis' (Section 6.2.i). However, the statistical analysis of test results
shows that this hypothesis is often too conservative and that it should,in
many ordinary cases, be corrected by appropriately taking into account
the shear strength of the concrete in the compressive zone.
Tests show that in many cases the assumption of a lattice system with
members inclined at 45"(Ritter-Mörschtheory) does not properly represent
the actual behaviour of the member. The lattice hypothesis is always on the
safe side,but has the drawback that.itoften leads to over-designof the trans-
verse reinforcement and thus to a wasteful use of steel.
These are various ways to get round this difficulty and yet retain the lattice
hypothesis as the basis of design.
 77
One such way is to consider the inclined concrete struts as loaded in
eccentric compression(notin concentriccompressionas previously assumed),
which is equivalent to taking account of fixity (rigid end restraint) of the
struts in the compressive zone of the member and to considering this zone
as participating in resisting the shear force.
Another way is to add to the resistance capacity T, of the transverse
reinforcement (as determined from the lattice hypothesis) a contribution
of the concrete in the compressive zone.This is an empirical procedure,
the magnitude of Tbbeing based on experimentally determined data. It is
this second solution that has been adopted in this Manual.
With the aid of these assumptions it is possible to calculate the shear
strength of a member in the ultimate limit state.This calculation can be
considered a sufficient justification of the shear strength of a structural
member.Accordingly, no provision has been made for analysis of the limit
states of cracking and deformation as a result of shear force.

Minimum Percentage Of Transverse Reinforcement

Every structuralmember which has to resist a shear force should be provided

with transverse reinforcement.The mechanical percentage of such reinforce-
ment should be at least two per cent (0.02),and the successivelayers of bars
should,in a direction parallel to the centre-lineof the member,not be spaced
farther apart than eighty hundredths of the effective depth of the member
(t < 0.80h).
The need for providing transverse reinforcement arises from applying
the general design rule of connectors.
The only possible exceptions are shells and slabs (less than 25 cm thick)
concreted in one continuous operation (see below), in which, because of
their small thickness,it is not feasible to install transverse reinforcement.
It should be noted that the mechanical percentage of transverse reinforce-
ment is determined by:

where Zt denotes the basic tensile strength of the transverse reinforcing

steel (Section 4.3.2)and 0;I the basic compressive strength of the concrete
(Section 4.3.3).The other symbols are as already defined and explained
above.

Practical Design Rules

Design Of Beams And Ribs

The shear strength of beams and ribs in the ultimate limit state should be
determined by adding together the resistance capacity Taof the transverse
reinforcement (calculated according to the lattice hypothesis) and the
78
resistance capacity T of the concrete in the compressive zone (calculated
according to the formula in the Section on shear resistance capacity of the
compressivezone below).
This rule can be applied not only to the design of beam webs or the ribs
of T-beamsbut also to the ribs of ribbed floors (whichare similar to T-beam
ribs).
The shear resistance capacity T, of the transverse reinforcement should
be taken as:
h .
T, = 0.9Z,.A,.-(sinCI +cos CI)
t

where Ztdenotes the basic tensile strength of the transverse reinforcement

(Section4.3.2),A,the individualcross=sectionalarea of each layer of trans-
verse reinforcement, t the spacing of the transverse reinforcement layers
(measuredparallel to the centre-lineof the member), CI the inclination of the
transverse reinforcement (in relation ;o the centre-line of the member),
and h the effective depth of the section.
The expression for T, presupposes that the lever arm of the internal
forces is equal to 09 h.
In the case where the transverse reinforcement is perpendicular to the
centre-lineof the member (e.g., vertical stirrups), the formula can be simpli-
fied to:
h
T, = 09Zt.A,.-
t

The shear resistance capacity &of the concrete in the compressive zone is
allowed to be taken into account only on condition that no direct (normal)
tensile force acts upon the section considered and that the bending moment
M at that section is at least equal to 1.5Th.
The first condition rules out the possibility of taking the contribution of
the concrete into account in the case of a member subjected to bending in
combination with tension.The design of such a member should be based on
the direct application of the fundamental design assumptions for con-
nectors (Section6.2.1, rule for connector reinforcement).It can be shown that
those assumptions can indeed legitimately be applied to loadings consisting
of shear with tension as well as to those consisting ofshear with compression.
The second condition excludes the zones in the vicinity of the supports.
These zones should, in each case, be specially investigated (see section
below on design of slabs and shells); in particular,the mode of transmission
of the forces should be checked and,if need be,constructionalmodifications
should be made to ensure that this transmission is efficiently effected.
The shear resistance capacity T varies between a minimum ofi.o b . bo. h
and a maximum of
 79
Hence :

where the compressive direct force N (if any) is expressed in bars.

If the member is ‘over-reinforced,i.e., if the longitudinal tensile reinforce-
I I
ment is able, in the ultimate limit state,to resist a fictitious moment M +
I 1 I
1.5 TI h-0.5 N’ h (where the bending moment M, the shear force T,
and the direct compressive force N are expressed in absolute values), the
shear resistance capacity &of the concrete in the compressive zone should
be taken as :

subject to the following limiting condition:

O bo
~ ~ 2 0 . 0f.
5 .-
Ob

where w denotes the mechanical percentage of longitudinal tensile re-

inforcement (fully anchored beyond the section considered), w p the mech-
anical percentage of longitudinal tensile reinforcement strictly necessary
1 I I 1 1
to resist the fictitious moment M + 1.5 TI h-0.5 N h, b the width of
the compressive zone of the beam (or the effective width of the compressive
flange of a T-beam or ribbed floor), b, the width of the web (inthe case of a
beam) or the rib (in the case of a T-beam or ribbed floor), O, the basic tensile
strength of the longitudinal tensile reinforcement,o b the basic compressive
strength of the concrete, and Ob the basic tensile strength of the concrete.
If the longitudinal tensile reinforcement is not able, in the ultimate limit
1 I 1 I 1
state, to resist a fictitious moment M + 1.5 TI h-0.5 N h, the shear
resistance capacity & of the concrete in the compressive zone should be
taken as:
3 -
&= -.ob.
8
bo. h

These values and these limits have been deduced from the statistical
interpretation of existing test results.
They are applicable to all cases of bending combined with compression.
For simple bending (N= O) the design formulae become much simpler;
the resistance capacitity of the concrete in the compressive zone is in this
case comprised between the following limits:
T h e total resistance capacity for shear force in the ultimate limit state
should be taken as

for beams and ribs. This must, however, be limited to a maximum value of
80
2¿fbb0hin the case of a beam which has no compressive flange,and to a
maximum value of 2.5ó,b0h in the case of a beam with a compressiveflange
or a ribbed floor.
These upper limits are allowed to be increased by 40% if the transverse
reinforcement consists of:
(a)an orthogonal network of bars providing the same percentage of
reinforcement in both drections;or
(b) a mixed system of stirrups set perpendicular to the centre-lineof the
member and longitudinal bars bent up at an angle of about 45"
(cr=45")and suitably anchored;or
(c) inclined stirrups set at an angle of about 65" in relation to the centre-
line (aN 65") and suitably anchored.
These limiting values have been deduced from the statistical interpretation
of existing test results.

Local Conditions At The Supports Of Beams

If a load and a bearing reaction are applied to two opposite faces of a beam
and at a distance apart which is less than three-quartersof the effective
depth (0.75h) as shown in Figure 6.23,the fraction of the load equilibrated
by the reaction should not be taken into account in designing the transverse
reinforcement in the region of the beam comprised between those two forces.
The loads acting in the zone abc may be neglected in the calculation of the
shear resistance,in so far as they are equilibrated by the bearing reaction.
O n the other hand,it is necessary to check that the strength of the inclined

Figure 6.23

concrete struts is sufficient to ensure the direct transmission of the loads to

the supportand,furthermore,that the anchorage ofthe longitudinalreinforc-
ing bars is able to equilibrate the thrust from these struts.
It may also occur that the load or the reaction is applied at an intermediate
level.It will then not affect,directly or by compression of the concrete,that
fraction of the depth of the beam which,in that place,actually resists the
 81
shear force.It is therefore essential to transfer the load or the reaction to the
proper level by means of suitably anchored suspension reinforcement.

Junctions Between Flanges A n d Ribs

The sections through the compressive flange parallel to the centre-lineof

the member and,more particularly, the plane of the junction between the
flange and the rib should be checked with regard to shear.
The corresponding shear resistance capacity (shear strength) should be
determined in accordance with the fundamental design assumptions for
connectors (Section 6.2.1,rule for connector reinforcement), ignoring any

Figure 6.24

possible contribution of the concrete to that resistance.It should be limited

to a maximum value of 3 Ob.be. h, where be denotes the effective width of
the compressiveflange (Section6.1.1).
The effect of the bending couple of the flange or top slab itself (aboutan
axis parallel to the centre-lineof the beam or rib,see Figure 6.24)and also
the effect ofthe secondary direct force due to the curvature ofthe compressive
stresstrajectoriesin the flange can permissibly be neglected in the calculation.
The reinforcement of the flange or top slab itself is generally disposed at
right angles to the centre-lineof the beam and may be considered as con-
nector reinforcement, irrespective of the part it plays in determining the
flexural resistance capacity of the flange or slab.So it is merely necessary
to check that their anchorage conditions are such that these bars can indeed
efficiently function as ‘connectors’and that their contribution T, to the shear
resistance capacity is sufficient.
This contribution T, can be determined by applying the rule for con-
nector reinforcement (Section6.2.1) and is equal to:

T , = O , . ~ D ~ ~ . ~ , .-Z = CA*
T , ..be.z
-
t . be

hence:
h
T, = 0.9.¿7,.At.-
t

This value of T, must not exceed an upper limiting value of 3 ¿fbbeh;

82
hence:
h
T = T, = 0.9.O,. A,.- < 3 . ab.be. h
t
It should then be checked that this shear resistance capacity is sufficient.

Design Of Slabs A n d Shells

Notwithstanding the stipu1,ationof a minimum percentage of transverse

reinforcement in any member subjected to shear force,transverse reinforce-
ment may be omitted in slabs and shells,provided that the following con-
ditions are satisfied:
(a) The structural behaviour really is that of a slab or shell.
In other words, if a loading tends to produce a shear crack, the
strength in bending should, in the direction perpendicular to that
crack,prevent it from opening out.
(b)At each point the ratio of the bending moments of the same algebraic
sign acting upon two mutually perpendicular sections-i.e.,the ratio
of the moments producing curvatures of the same sign in two mutually
perpendicular directions-should, in the ultimate limit state, be at
least equal to four.
(c) Concreting should be performed as a continuous operation,without
any construction joints in the direction of the thickness of the slab or
shell.
Subject to fulfilment of these conditions,the shear resistance capacity of
slabs and shells in the ultimate limit state should be taken as equal to 0.8
Sb.h per unit width.

6.2.3 BOND

Definition O f Bond
The notion of 'bond'comprises two distinct functions: anchorage bond and
flexural bond.
Anchorage bond: At each end of a reinforcing bar the axial tensile or
compressive force applied to the bar has to be transmitted to the concrete
by bond.This bond at the ends constitutes bond by anchorage.
Flexural bond: This form of bond occurs in all intermediate parts of
the bar (i.e.,other than at the ends) and equilibrates the variations in the
axial tensile or compressiveforce applied to the bar.

Anchorage O f Reinforcing Bars

Basic Principles Of Anchorage Calculations
The anchorage should be checked for the ultimate limit state; the check
should be applied to each bar,considered individually,even ifthat bar belongs
to a group of bars.
 83
This check should be based on the three following fundamental assump-
tions :
(a) over the entire anchorage length of the bar the bond stress is constant
and equal to its limit value ? d ;
(b)in the curved parts of an anchorage a frictional force acts in addition
to the bond as envisaged in (a); this frictional force should be taken
as equal to the curvature reaction of the bar (whichfor this purpose is
conceived as a wire wrapped round a cylinder) multiplied by the co-
efficient of friction between steel and concrete;
(c) the anchorage of the reinforcement is considered to be ‘total’if, in the
ultimate limit state,the tensile force A.Fa in the bar under considera-
tion is equilibrated by the bond and frictional forces.
In the vicinity of the reinforcement the bond should be conceived as a
tangential action that can produce cracking at 45”in the covering concrete.
But the ‘struts’separated by the inclined cracks thus formed are only able to
equilibrate the compressive forces. Therefore, in order to equilibrate the
corresponding tensile components,it is necessary to provide transverse
connectors, so-calledfastening connectors. These are essential to the effi-
ciency of the anchorage.
The fastening connectors should surround the bar to be anchored (or the
corresponding group of bars); they should pass round the ‘outside’of the
bar (orbars), i.e.,on the sidenearest the exterior ofthe member,and should be
suitably anchored in the interior of the concrete.
Let Td denote the bond stress in the ultimate limit state, Zfthe basic
strength of the connector reinforcement,A,the total cross-sectionalarea of
the connector reinforcement,and p the perimeter of the bar to be anchored.
Consider the equilibrium condition:
p.t.Td=At.Zt
i.e.:

Transverse reinforcement already provided for a different purpose,e.g.,to

resist shear force,may be considered to function as fasteningconnectors for
the main tensile reinforcing bars at the points of curtailment or division of
those bars.However,it should be checked that such transverse reinforcement
is indeed sufficient for the purpose;otherwise it will have to be augmented.
Thin members (slabs and shells) present special problems,since it is not
possible, on account of their limited depth,to provide them with fastening
connectors.In such cases the designer will have to furnish special justifica-
tion of the efficiency of the anchorages.

Straight Anchorage

Over the anchorage length the limit value of the bond stress is assumed to
be constant and should be taken as:
84
for plain bars

and :
for deformed bars

where 4 denotes the nominal diameter of the bar,d the distance from the
centroid of the bar to the nearest concrete face, and Ob the basic tensile
strength of the concrete (Section4.3.3).
The corresponding limit value of the bond force is referred to the effective
perimeter p of the bar considered.
The effective perimeter p to be adopted in checking the anchorage should
be taken as equal to:
(a) for a single bar or a group of two bars:
the nominal perimeter,i.e.,for each bar: n$
(b)for a group of three bars:
the nominal perimeter less twice the internal perimeter arc,i.e.,for
each bar:$nd
The straight bond length 1, whereby total anchorage ofa tensile reinforcing
bar is ensured should be taken as:

where A denotes the cross-sectionalarea of the bar considered,p the effective

perimeter of that bar,O, the basic strength of the tensile reinforcement,and
-
z, the limit value of the bond stress.
The expression for the straight bond length is derived from the equilibrium
equation given in Section 6.2.3:

For a single bar or for a group of two bars the effective perimeter p is
taken as n4.Hence:

For a group of three bars the effective perimeter p is taken as 5

.4. Hence:

In a case where, irrespective of the loading conditions of the member

under consideration,the tensile reinforcement has its anchorage in a com-
pressive zone, the designer may, subject to justification of the general
equilibrium of the forces,reduce the straight bond length to the value:
 85
The straight bond length Id whereby total anchorage of a compressive
reinforcing bar is ensured should be taken as:
1'-- A' -
3,
d- P'Td
where 5 denotes the basic strength of the compressive reinforcement and
the other symbols are as defined above.

Anchorage By Curvature
The calculation of an anchorage by curvature is based on the following
differential equation conforming to the three assumptions stated in 'basic
principies of anchorage calculations'above.
A.da, = (p.r.Td+iI/.A.o,)d8
where A denotes the cross-sectionalarea of the bar considered individually,
p the effective perimeter of that bar, r the radius of curvature of that bar
(measured along its axis), Td the limit value of the bond stress (assumed
constant) a, the tensile stress in the reinforcement (zero at the end of the
bar, but attaining the limit value Z, at the point where total anchorage is

Figure 6.25 I +
achieved), 8 the central angle ofthe curve (thepositive direction of measure-
ment is that of the assumed sliding of the steel in relation to the concrete,
i.e.,in the direction of increasing values ofo=), the coefficient of friction of
steel on concrete (taken as 0.40for curved bars).
In Figure 6.25,for a curved portion IJ, integration of the equation yields
the following relation:

For 8 = O this relation is transformed into the expression for calculating

the anchorage of a straight length 1:

In a curved portion of a tensile reinforcing bar (irrespective of whether

the curvature is for the purpose of anchorage or due to any other change of
direction of the bar) the radius of curvature r (measured to the axis of the
bar) should satisfy one of the following conditions:
86
(a) for a single curved bar or a curved bar belonging to only one layer of
bars :
r20.20.4(1+;). 5
(b) for a curved bar belonging to a set of two layers of bars:

(c) for a curved bar belonging to a set of three layers of bars:

where 4 denotes the nominal diameter of the bar considered, e the

distance from its centre of curvature to the nearest face of the member,
O, the basic tensile strength of the steel,and O;, the basic compressive
strength of the concrete.
The pressure that a curved bar exerts on its concave side may entail a
danger of fracturing the concrete.This danger is negligible in a case where
such pressure is exerted deep down in the concrete; on the other hand,
near the surface of the concrete it may constitute a more serious hazard,
especially if the middle plane of the bar is parallel to that surface.Hence it is
necessary to limit the pressure exerted by the bar,and this can be done by
limiting its radius of curvature,i.e.,not permitting radii of less than a certain
acceptable value. It is also recommended that the plane of the curved an-
chorage portion of the bar be inclined inwards into the member.
A ‘standard hook’ (also called a U-hook,see Figure 6.26)comprises a
semicircularportion whose inside diameter is equal to five times the diameter

Figure 6.26

of the bar (correspondingto a radius of curvature equal to three times the

bar diameter), followed by a straight terminating portion whose length is
twice the diameter of the bar.
A hook of this kind must be regarded as always necessary on plain round
bars; on the other hand,hooks can often be dispensed with on deformed
(high-bond)bars,for which bond along the straight bar generally ensures a
sufficientlyeffective anchorage,provided that the necessary fastening con-
nectors are installed.
 87
Application of the calculation for anchorage by curvature gives the
following results for the standard hook (seeFigure 6.27):

The anchorage is considered to be total if:

which means that a standard hook is equivalent to a straight embedded bar

length of 26.34.If the anchorage is not total at A,it should be supplemented
by a straight anchorage of appropriate length.
The limit value of the tensile force which is available in the bar at the
beginning of a hpok (pointA) is approximately equal to :
60420bfor a plain bar
120q520bfor a deformed bar
Unless specialproof is provided tojustifya departure from this stipulation,
the radius of curvature for binders and stirrups should be three times their
bar diameter q5. This means that a bar-bending forming mandrel with a
diameter equal to 54 should be used on the site.
In addition,anchorage within the interior ofthe concretecan be considered
to be total only if the curved portions of the binders and stirrups are pro-
vided with straight terminating portions which:
(a) extend back at 180" and are at least 54 in length for plain bars (no
other form of anchorage being permissible for these);
(b)extend back at 180"and are at least 54in length,or slope back at 135"

I I 5 Ø I

Figure 6.28
and are at least 104 in length,or are set at 90" and are at least i54 in
length in the case of deformed (high-bond)bars (see Figure 6.28).
It is not permissible to anchor compressive reinforcement by making use
of curvature of the bars.
When a curved anchorage is loaded in compression,it tends to cause
bending of the bar at the point where sudden variation of the curvature
88
occurs. The reactions set up by this bending can produce ‘unbalanced
thrust’which is liable to cause spalling of the concrete cover to the bar.
The only possible exception to this rule is formed by members which
are alternately subjected to eccentric compression and composite bending,
in which the main reinforcing bars have to resist tension as well as com-
pression and must therefore be provided with anchorage by curvature. In
that case it should be checked that the corresponding ‘unbalanced thrust’
is equilibrated by the fastening connectors,which should be of appropriate
shape and section to serve the purpose.
A particularly dangerous anchorage is one which has a straight terminating
portion extending parallel to a concrete surface and situated close to that
surface. However, in that case a single tie bar (with a diameter of about a
quarter of that of the anchored bar) connecting this straight terminating
portion to the interior of the concrete will often suffice to obviate the danger
of spalling (Figure6.29).
Another solution,strongly recommended, consists in sloping this straight
terminating portion back into the interlor of the concrete. Transverse
reinforcement already provided for other purposes will,in this case,generally
suffice to equilibrate the ‘unbalance thrust’(Figure6.30).
Generally speaking, a curved bar will develop ‘unbalanced thrust’ when
its curvature reaction is directed towards the outside of the member instead

!-.-.___._.i
Figure 6.29 Figure 6.30

of inwards, i.e., into the interior of the concrete. The curvature reaction
per unit length is equal to the direct force in the bar divided by the radius of
curvature of the bar; the reaction is situated in the plane of curvature and is
directed towards the convex side of the curve if the bar is under compression
(orto the concave side if the bar is in tension).
If the load applied to a curved bar produces an ‘unbalanced thrust’
directed towards the outside of the member, this bar should be tied back
into the concrete by means of fastening connectors (Section 6.2.3).These
connectors should be disposed at right angles to the bar,completely surround
it, and be provided with total anchorage in the interior of the concrete.
The design of the fastening connectors for equilibrating the ‘unbalanced
thrust’ should be carried out according to the following procedure (see
Figure 6.31):
 89
Let Ft denote the basic strength of the connector reinforcement, A,the
total cross-sectionalarea of the connector (orof the two legs thereof),t the
spacing of the Connectors (the distance IJ between two consecutive con-

Figure 6.31

nectors). Then,as a first approximation,the following equilibrium equation

can be written down:
A .IJ;
t.-
r
= A,.Ft
In a case where the mechanical properties of the main reinforcement
and of the connector reinforcement are identical, the design equation
becomes,as a first approximation:
t
A, = - . A
r
Certain thin structures-shells, in particular -give rise to difficulties.
Because of their small thickness,these structures cannot be provided with
the necessary fastening connectors.For this reason,the ‘unbalancedthrust’
developed by a curved bar can be considered permissible if its radius of
curvature satisfies the following condition,which is valid for tension as well
as for compression:

where 4 denotes the nominal diameter of the bar considered,ea the distance
from the axis of this bar to the face of the concrete (onthe side where the
‘unbalancedthrust’ is acting), 0, the basic tensile strength or compressive
strength of the steel (whicheveris applicable to the case), and Ob the basic
tensile strength of the concrete.

Curtailment Of Longitudinal Reinforcing Bars

The checking of the anchorage conditions at the ends of the longitudinal

reinforcing bars should be based on the bending moment diagram,which
should be suitably displaced to take account of the need to absorb the
horizontal components of the forces in the ‘struts’(compression members)
of the fictitiouslattice system.
90
This ‘displaced’diagram (see Figure 6.32),which serves as the basis for
designing the longitudinal reinforcement,is obtained by shifting the en-
veloping curve of the bending moments parallel to the centre-line of the
member in the most unfavourable direction by an amount equal to the

,Referenceline

Figure 6.32

effective depth h of the section.The longitudinal bars should be anchored

outside the ‘displaced’diagram.
The amount ofdisplacement actually required may vary between h and h,
according to the efficiency of the transverse reinforcement:the value of
h is thus certainly on the safe side.

Splicing Of Bars B y Lapping

Check the efficiency ofa splice formed by the lapping oftwo identicalparallel
bars should be done in accordance with the three fundamental assumptions
stated in Section 6.2.3.The fastening connectors should be designed in
accordance with the rules given in the same section.
If the centre-to-centredistance (distancebetween the axes) of bars without
anchorage devices (straightbars) does not exceed five times their nominal
diameter,the lap length should be at least equal to the straight embedded
length (see Section 6.2.3).Otherwise the lap length should be at least equal
to the sum of the straight bond length and the centre-to-centredistance of
the bars.
In the specialcase of welded fabric reinforcementthe lap length should be
at least equal to:
2a + 1040
where a denotes the centre-to-centrespacing of the distribution wires and
q5a the diameter of the carrier wires. In addition, the lap should comprise
at least three welds in each layer of fabric.
 91
If the centre-to-centredistance (distance between the axes) of bars with
anchorage devices (hooks)does not exceed five times their nominal diameter,
the lap length should be at least equal to six-tenths(0.60)
of the straight bond
length. Otherwise the lap length should be at least equal to the sum of
six-tenths(0.60)of the straight bond length and the centre-to-centredistance
of the bars.
The lap length in compressive reinforcing bars should be at least equal
to six-tenths(0.60)of the straight bond length (see Section 6.2.3).In the
particular case where the structure is subjected to vibrations or to impact
effects,the lap length should be taken as equal to the straight bond length.
Anchorage devices relying on curvature should not be employed, as
stipulated in Section 6.2.3.

Flexural Bond

Overall Flexural Bond Developed By The Reinforcing Bars As A

Whole

Checking the flexural bond of the tensile reinforcement may be performed

for all the bars of this reinforcement as a whole, whether they are single
bars or groups of bars.
It should be noted that this check concerns the bond of those parts of the
bars which are situated outside the anchorage zones (i.e.,away from the ends)
and relates essentially to the transmission of the tangential actions which
cause the longitudinal force exerted by the reinforcement to vary.

Checking The Flexural Bond In The Ultimate Limit State

For checking the flexural bond developed by the tensile reinforcement

in the ultimate limit state by shear action all along the bars the following
assumptions should be made :
Outside the anchorage zone the limit value of the bond stress is constant
and has these values:
for plain bars in beams:

for plain bars in shells:

92
for deformed bars :

where 4 denotes the nominal bar diameter and d the distance from the axis
of the bar to the nearest face of the concrete.
The corresponding limit value of the bond force is referred to the effective
perimeter p ofthe bar or of the group of bars considered.
The effective perimeter p to be introduced into the check calculation for
flexural bond should be taken as:
(a) the nominal perimeter in the case of a single bar,i.e.,7c4;
(b) the minimum circumscribed perimeter of the cross-sectionin the case
ofa group of two bars,i.e.,(7c +2)4;
(c) the nominal circumscribed perimeter of the cross-section less the
perimeter of the space enclosed in the case of a group of three bars,
i.e.,(3. + 3)4.
The flexural bond of the reinforcement is effective if, in the ultimate
limit state,the shear force is equilibrated by the bond forces of all the bars
of which this reinforcementis composed.
For an elementary length dx ofthe reinforcementthe equilibrium equation
is expressed by the relation:
A.dZa = n.p.ïd,
where n denotes the number ofbars comprised in the reinforcementwith a
total cross-sectionalarea A.
O n introducing the shear force T,the lever arm z and the effective depth
h of the section,this equilibrium equation can be written as follows:

or:

6.2.4 TORSION

Basis For The Calculation Of Torsional Strength

The calculation for determining the torsional strength in the ultimate limit
state should be based on the ‘latticehypothesis’ (Section 6.2.1). Analysis of
test results has shown,however,that the lattice hypothesis is often too con-
servativeand that it should be corrected by appropriately taking into account
the torsional strength of the concrete in the compressive zone.
Tests have shown that in many cases the lattice hypothesis,which is on the
safe side,has the drawback that it results in wasteful over-designingof the
torsion reinforcement.
 93
The method adopted in this Manual for obviating this drawback is an
empirical one and consists in adding to the resistance capacity M,,of the
torsion reinforcement (calculatedon the basis of the lattice hypothesis and
multiplied by an experimentally determined adjusting coefficient)the resist-
ance capacity M,bof the concrete in the compressive zone.

Minimum Percentage O f Transverse Reinforcement

Every structuralmember which is subjected to a torsional moment should be
provided with transverse reinforcement.The mechanical percentage of such
reinforcement should be at least:
~-
h, 10
b,, +hat ob
'

where Obdenotes the basic compressive strength of the concrete (expressed

in bars), h, the total geometrical depth of the section,ha, the height of the
binders,and batthe width of the binders.
The transverse reinforcement should consist of closed binders placed
perpendicularly to the longitudinal axis of the member.Their spacing should
not exceed b,, nor iha,. The ends of each binder should either be provided

Closed binder
\

Figure 6.33

with anchorage hooks suitably bent back round the longitudinal reinforcing
bar,or be joined together by a weld capable of resisting the force developed
by the binder steel at a stress equal to its elastic limit.
The condition for the minimum mechanical percentage of transverse
reinforcement can be written as follows:

with the usual notation,the stresses being expressed in bars or in kg/cm2.

The need for providing a minimum mechanical percentage of reinforce-
ment arises from the fact that the torsional resisting moment of the concrete
94
in the compressive zone is much lower than the torsional moment corre-
sponding to the development of inclined cracks.So,in order to ensure that
the total torsional resisting moment will exceed the torsional moment corre-
sponding to cracking,a sufficient percentage of transverse torsion reinforce-
ment must be provided. This condition obviates the dangerous possibility
of brittle fracture occurring suddenly and without warning signs (see Figure
6.33).
In a structural member loaded in torsion the inclined cracks are liable
to appear on all the faces.For this reason it is essential to provide closed
binders, so as to have connector reinforcement across all cracks that will
possibly develop.It is for this same reason,too,that the minimum spacing
(t,<bat and t ,<+hat)has been laid down for the binders.

Practical Design Rules

The torsional resistance capacity (torsional strength)of structural members

in the ultimate limit state should be determined by adding the contribution
M,,of the transverse reinforcement (calculated according to the lattice
hypothesis,Section 6.2.1)and the contribution Mtbof the concrete in the
compressive zone (calculatedfrom the empirical formula given below).

Contribution Of The Transverse Reinforcement To The Torsional

Strength

The contribution M,,of the transverse reinforcement to the torsional

strength of a structural member is :
A
M,,= CI.at.L.
t
bat.hat

where :
u is a coefficient equal to 0.33+0.16! k but not exceeding 0.75;
bat
O, is the basic tensile strength of the transverse reinforcing steel,subject
to an upper limit of 2 500bars;
A, is the total cross-sectionalarea of the two legs of a binder;
t is the spacing of the binders (measured parallel to the centre-lineof the
member);
bat is the width of the binders;
ha, is the height of the binders.
The expression for M,,conforms to the application of the ‘latticehypo-
thesis’ in conjunction with the introduction of a reduction coefficient u
which is justified by the results of tests carried out in the Laboratories of the
Portland Cement Association,Illinois,U.S.A.
 95
The torsional reinforcement should not consist solely oftransversebinders.
There should,in addition, always be a longitudinal reinforcement with a
total cross-sectionalarea:
A
AT = bat.hat
This longitudinal reinforcement should comprise at least four bars placed
respectively at the four corners of the transverse binders. But if hat> 2bat,
then it should also comprise longitudinal bars at the longer sides of the
binders; these bars should not be spaced farther apart than $bat. Besides,
the diameter of the longitudinal reinforcing bars should be at least equal to
the diameter of the binder reinforcement and not less than 10mm.
The transverse binders and the corresponding longitudinal bars together
form the torsion reinforcement.
For the ‘lattice’mechanism’tobe effective,the reactionsfrom the concrete
‘struts’separated by the torsional cracks should be equilibrated in the trans-
verse and in the longitudinal direction.Hence the need for both longitudinal
and transverse reinforcement.
Thisdouble system of‘torsionreinforcement’is additional to the reinforce-
ment required for giving the member its strength to resist bending (longi-
tudinal flexural reinforcement, cf. Section 6.1) and to resist shear force
(transverseshear reinforcement,cf. Section 6.2.2).
The longitudinal reinforcing bars should be distributed as uniformly
as possible around the perimeter of the transverse binders,in order to en-
sure efficient arrangement of connector reinforcement across the torsional
cracks.

Contribution Of The Concrete In The Compressive Z o n e T o The

Torsional Strength

The contribution kf,, of the concrete in the compressivezone to the torsional

strength of a structural member can be taken into account only on con-
dition that no direct tensile force acts upon the section considered.This
contribution should be taken as:

where :
M,,= B.Zb.J!h,-J!
4
b2( 3)
fl is a reduction coefficient equal to

2)$1 ’}
1
J {1 + [
1.3.: (h, - .

T is the total shear force (Section6.2.2)taken into account in checking

the ultimate limit state ( T = T,+ TJ;
96
M ,is the total torsional moment taken into account in checking the ulti-
= MI,+Mth);
mate limit state ( M I
bo is the width of the compressivezone (widthof web);
h, is the total geometrical depth of the section;
-h is the effective depth of the section;
ob is the basic tensile strength of the concrete.
Furthermore,if a section is subjected to a torsional moment,the contribu-
tion q of the concrete in the compressive zone to the shear strength (Section
6.2.2) should be multiplied by a reduction coefficient ß as defined below.
Tests have shown that the contribution of the concrete in the compressive
zone to the ultimate strength of a member subjected to pure torsion is
approximately equal to half the torsional moment corresponding to the
formation of inclined cracks. The value of the resisting moment of the
concrete alone is assumed to correspond to a torsional shear stress equal
to 0.5ab, calculated on the assumption that plastic stress distribution
occurs and that,in the case of a T-beam,only the rib (orweb) resists torsion.
These assumptions lead to the following expression :

M t b= 0.5.abx 2 h , - O
2
b2( ,)
But iftorsionis combined with shear force,the contribution ofthe concrete
to the torsional strength is reduced.Tests have shown that the curve of inter-
action between the torsional moment and the shear force corresponds to a
circular arc with its centre at the origin.O n adopting this approximation,the
following expression is obtained for the reduction coefficient ß:
1

Similarly,the contribution q of the concrete to the shear strength of a

member is reduced if the section is also subjected to a torsional moment in
addition to shear force. In that case the following reduction coefficient fi,
obtained by adopting the above-mentioned curve of interaction,should be
applied to G:
r
-41 ht-
1
h M,23
bo'

Total Torsional Resistance Capacity

The total torsional resistance capacity of a member in the limit state should
be taken as:
=
but should be limited to a maximum value of 5 Mtb.
Hence:
 97
The object of imposing this upper limit is to ensure that,in conformity
with the design assumptions,the transverse reinforcement will be able to
attain its basic strength in the ultimate limit state.The coefficient 5 emerged
from the interpretation of the results of the tests carried out in the Labora-
tories of the Portland Cement Association,Illinois,U.S.A.

6.2.5 P U N C H I N G SHEAR

Assumptions Relating To The Applicability Of The Analysis

The analysis presented here is applicable to checking the punching shear

strength of a slab (or other plane structure) under the effect of a locally
concentrated force acting perpendicularly to the middle plane of the slab

Outline of area of
load application

Figure 6.34

Outline of area of
application of
column reaction

Figure 6.35 -
and applied to a small area of its surface,bounded by an outline which is
assumed to be convex (i.e.,having no re-entrantangles).
The locally concentrated load may be either a superimposed load or a
support reaction.
The analysis is therefore applicable not only to floor slabs,flat-slabfloors
98
and mushroom floors subjected to locally concentrated superimposed loads
(punching shear at the columns) but also to foundation slabs under single
points of support.
If the locally concentrated force acts through a surfaciqg on the slab,the
outline to be adopted for the area of application of that force upon the
surface of the slab itself should be parallel to the outline of the area of
application to the surfacing and situated at a distance therefrom equal to
the thickness of the surfacing (ifthe latter is of concrete or a similar material)
or equal to three-quartersof the thickness of the surfacing (if the latter is a
material with a lower strength than concrete,e.g.,a mastic asphalt or stone-
filled bituminous surfacing). This rule is also applicable in a case where the
slab is thickened at the locally concentrated force,as is more particularly

Equivalent convex outlines

Figure 6.36

the case with mushroom floors and column footings (Figures6.34and 6.35
respectively). The approximation due to applying this rule is always on the
safe side.
If the outline of the area of application of the locally concentrated force is
not convex, the analysis is still applicable,provided that this outline is
replaced by a fictitious convex outline C which envelops the actual area of
application but has the minimum perimeter (see Figure 6.36for examples).

Determination Of T h e Punching Shear Strength

The limit value of the resistance to punching shear due to a locally concen-
trated force acting uniformly upon a small area (with a convex outline
C,actual or equivalent,cf. above section) of the surface of a slab should be
taken as:

where Zbdenotes the basic tensile strength of the concrete;h the effective
 99

Figure 6.37

/
/ Apert Ure
/ /

,
’
- ,/
/

-h2 Outline C
of perimeter p

Figure 6.38

Outline C
Unsupported
edge

I
J.
Figure 6.39
1O0
depth of the slab at the outline C ;p’ the perimeter of an outline C outside,
and parallel to,the outline C,at a horizontal distance 3h therefrom.

Special Case: Outline Of Elongated Shape (Figure 6.37)

Ifthe larger dimension a ofthe outline C is more than three times the smaller
dimension b(a > 3b), the value of the perimeter p‘ should be limited to 8b +4 h
(p’< 8b +4h).
This condition is equivalent to limiting the larger dimension a to the value
3b, i.e.:
p’ = 2(a+b+2h)d8b+4h

Special Case: Presence Of An Aperture (Figure6.38)

If the slab contains an aperture in the vicinity of the outline C,the effect of
the aperture upon the punching shear strength should be taken into account
by an appropriate reduction of the perimeter p’. This reduction is taken
as equal to the length of the intercept between two lines which are drawn

Unsupported edge
h

I l

-
h
Outline C of 2
oerimeter P I
I
I
I
Unsupported edge

G
Figure 6.40

from the centroid of the area bounded by the outline C and are tangential
to the aperture in question.

Special Case: Vicinity Of A n Unsupported Edge (Figure 6.39)

If the locally concentrated force acts in the vicinity of an unsupported edge,

 101
the latter should be treated as an aperture of infinite length in the direction
parallel to this edge.

Special Case: Vicinity Of A n Unsupported Corner (Figure 6.40)

If the locally concentrated force acts in the vicinity of an unsupported corner,

the latter should be treated as an aperture of infinite dimensions in the two
directions parallel to this corner.
 7
CONSTRUCTIONAL
ARRANGEMENTS

7.1 AGREEMENT BETWEEN C O N S T R U C T I O N A L

A R R A N G E M E N T S AND DESIGN ASSUMPTIONS
The constructional arrangements should be compatible with the basic
data and with the fundamental design assumptions,more particularly with
regard to :
(a) the physical and mechanical behaviour of the materials;
(b) the nature ofthe various connectionsbetween the component elements
of the structure (hinged joints, fixed connections, continuity con-
ditions,etc.);
(c) the sequence of the various stages of construction (which should
be anticipated and clearly indicated in the design calculations).
All relevant information concerning these data and assumptions should
be made available to the contractor by the design engineer.If the contractor
considers such information to be insufficient,he should request the designer
for additional information.

7.2 GENERAL CONDITIONS RELATING TO THE

REINFORCEMENT
7.2.1 SIMULTANEOUS USE OF DIFFERENT GRADES OR
TYPES O F B A R S

The use of different grades or different types of reinforcing bars in one and
the same structural member should be avoided as far as possible.It can be
allowed only if there is no real danger of confusing the bars of different
grades or types.
In practice the simultaneous use of two different grades or two types of
I02
 103
reinforcing bars,one for the main reinforcement and the other for the stirrups
and connector reinforcement of the same structuralmember,can be allowed.
But in that case the designer should:
(a) introduce each of these two grades or each of these two types of bars
with its proper reference strength and its proper basic strength into
the design calculation;
(b) take account of the possible consequential effects of the respective
properties of these various bars upon the verification of the condition
of strain compatibility at each section.
The simultaneoususe of two grades or two types of reinforcing bars is on
no account permissible in all cases where there is even the slightest possible
risk of confusion.This prohibition is more particularly applicable to the
bars of which the main reinforcement of a structural member is composed.
O n the other hand,the resultsofmany tests and the experience ofdesigners
extending over several decades have shown that it is allowable,in flexural
members, to use simultaneously: deformed (high-bond)medium-tensile
bars for the longitudinal reinforcement (tensile reinforcement,compressive
reinforcement,short ‘cap’bars) and plain bars for the transverse reinforce-
ment (stirrups,binders, connectors) and also for bars used for fixing the
reinforcement and for bars left projecting from the concrete (theseusually
have to undergo successive bending and forming operations and should,
for this reason,have a higher capacity for to-and-frobending than the longi-
tudinal reinforcing bars). Although the geometrical conditions of strain
compatibility cannot be strictly satisfied in all cases,this practice is justified
by the fact that,for one thing,no major objections to it have ever been put
forward and that,furthermore,abandonment of this practice would, in a
great many simple cases,be disadvantageousfrom the viewpoint ofeconomy.

7.2.2 P E R M I S S I B L E C U R V A T U R E OF R E I N F O R C I N G B A R S

The radii of curvature of reinforcing bars should be determined with refer-

ence to :
(a) the risk of crushing of the concrete under the effect of local pressures
in the curve;
(b)the ductility properties of the steel and its possibilities for bending
and shaping without any abnormal risks of immediate fracture or the
formation of incipient fracturezones which are difficult to detect;
(c) the manner in which the bar-bending operations in the workshop
and the steel-fixingoperations on the construction site are carried out
(Section9.2.2).

Condition To Ensure That No Crushing Of The Concrete Will Occur

In preparing the working drawings for the various members of a structure
the designer should, for all the curved portions of the reinforcing bars,
104
check that the condition to ensure that no crushing of the concrete will
occur (Section 6.2.3) is satisfied.
This check is necessary in actual practice only if all the bars of the same
hyer (or all the bars of the several layers)in the section under consideration
have to be bent in the same place,e.g.,at the corner of a portal frame.In the
case of a single curved bar,too,it is generally unnecessary to check the con-
dition for crushing of the concrete if the radius of curvature is less than
106.

Condition For Bar-Bending

The permissible nominal values for the radii of curvature r of the reinforcing
bars (measured to the bar axis)and the corresponding minimum permissible
diameters of the bar-bendingformers are given in the Tables 7.1 and 7.2.
Table 7.1 Minimum radii of curvature

Mild steel Medium-tensile steel High-tensile steel

(3O00 bars < u, <
Minimum radii of (u, c 3 O00 bars) (u,> 5 O00 bars)
5 O00 bars)
curvature

Stirrups and binders 24 34

Anchorages 34 4.54 5.54 5.54
Bends 4-54 4.54 84 84 10.54 1054

Table 7.2 Minimum diameters of bar-bendingformers

Mild steel Medium-tensile steel High-tensile steel

Minimum
(u e 3000bars) (3O00 bars < ue Q (u,> 5000bars)
diameters of 5 O00 bars)
forming mandrels
@< 1 2 m m 4 1 2 m m 4 < 1 2 m m 4 > 1 2 m m 4 < 1 2 m m 4 >12mm
Stirrups and binders 34 54
Anchorages 54 84 104 104
Bends 84 104 154 154 204 204

In Tables 7.1and 7.2the term ‘anchorages’must be taken to include all

devices utilising curvature to effect anchorage of the end of a bar into the
concrete,while the term ‘bends’refers to all changes of direction of a re-
inforcing bar (forexample,when a bar is bent up to assist in resisting shear
force). The arrangement ofthe steel in large cantileverssuch as are encounter-
ed in the roofs of grand-stands or in certain structures of daring design
calls for particular attention,since the bends in the bars of such structures
are often located in the most severely stressed zones.
According to Tables 7.1and 7.2,reinforcing bars having an elastic limit of
more than 5 000 bars are not allowed to have anchorages based on curvature
 105
and are not allowed to be used for binders and stirrups.Apart from being
used in straight lengths,they are only allowed to be used as bent bars, and
then only if they do not exceed 25 m m in diameter (4< 25 mm).

7.2.3 C U R T A I L M E N T O F R E I N F O R C I N GB A R S
Devices Used At The Ends Of Reinforcing Bars

The devices used at the ends of reinforcing bars-hooks, in particular-

should be so shaped that bending the bar involves no abnormal risk of
fracture or initiation of fracture (Section7.2.2)and that,when load is applied
to the reinforcement,there is no risk of ‘unbalancedthrust’nor of crushing
of the concrete (Section6.2.3).
The positions of such end devices in relation to the adjacent faces of the
member should satisfy the same conditions.

Sudden Changes Of Section

The simultaneous curtailment of a considerable proportion of the bars at

the same cross-sectionof the member should be avoided.T o fulfil this require-
ment it is recommended to use a larger number of thinner bars (with the
same total cross-sectionalarea), so that more satisfactory staggering of the
ends of the bars can be achieved.

7.2.4 S P L I C E S I N R E I N F O R C I N GB A R S

All splices in reinforcing bars should be envisaged in the design and shown on
the working drawings and be executed in accordance with those drawings.
There should be as few of these splices as possible. If they are essential,
they should be located outside the zones where the severest stress con-
ditions occur.
T w o types of splicing are normally permitted:
(a) ‘lapped splices’,for bars not exceeding 32 m m in diameter;
(b)‘welded splices’,for bars of any diameter.

Lapped Splices

Laps In Tensile Reinforcement

Laps in tensile reinforcement should satisfy the requirements which specify

the lap lengths,and those by which the corresponding fastening connectors
can be designed or checked (Section 6.2.3).
Besides, at any one particular cross-section the transmission of tensile
106
forces by laps in bars should be effected by not more than half the total cross-
sectional area of the reinforcement (in the case of bending with or without
compression) or not more than one-third of the total cross-sectionalarea
of the reinforcement (in the case of tension with or without bending).
Ifthe reinforcement consists ofmany layers ofbars,it may be a structurally
advantageousarrangement to let the same bars serve both as main reinforce-
ment and as splice bars (staggered splice),as in Figure 7.1,which relates to a
group of eight layers of bars with two simultaneoussplices at each section.
In this example the resisting cross-sectionalarea of the steel corresponds
to 8-2 = 6 times the cross-sectionalarea of one bar.
As for the fastening connectors, these should enclose the bar (on the
outer side thereof in relation to its position in the member) and be suitably

Figure 7.1

anchored in the interior of the concrete.Transverse reinforcement already

required for another purpose (e.g., binders and stirrups) may at the same
time serve as fastening connectors for the main tensile reinforcement at the
points of curtailment of the bars in that reinforcement;but it must then be
checked that such transverse reinforcement is indeed sufficient for the pur-
pose,otherwise it will have to be augmented.

Laps In Compressive Reinforcement

Laps in compressivereinforcement should satisfy the requirements shown in

Section 6.2.3,which more particularly specifies the lap lengths.
These lapped splices in compressive reinforcement should always be
formed by straight bonded lengths.Anchorage devices relying on curvature
(hooks,in particular) are permitted only in a case where the member,gener-
ally loaded in compression,may nevertheless exceptionally be subjected
to compositebending as a result of certain transverseactions such as wind or
 107
earthquake effects. In that case, however, the designer should take pre-
cautionsagainst ‘unbalancedthrust’by the provision offasteningconnectors
of appropriate shape and cross-section(Section 6.2.3).
In other words, the exception envisaged here concerns the case where
the compressive reinforcement may, in exceptional circumstances,be sub-
jected to tension by the action of wind or earthquakes.

Welded Splices
Unless specialjustification for a departure from this stipulation is provided,
splices formed by welding are permissible only for reinforcing bars with an
elastic limit of not more than 5 O00 bars.In addition,the method of welding
should not cause any impairment of the mechanical properties of the steel.
This clause more particularly rules out the use of electric arc tack-welding
on the site to fix the transverse reinforcement (binders,stirrups,fastening
connectors) to longitudinal reinforcing bars with an elastic limit exceeding
5 O00 bars.The tack welds are liable to initiate brittle fracture and are there-
fore especially dangerous.
As regards cold-workedreinforcing bars in particular,welded splices are
permissible only on condition that tests are performed to verify that the
method of welding employed in no way impairs the mechanical properties
of the bars (and especially the elastic limit and the ultimate strength). The
manufacturers of such reinforcing steel often supply practical information
with regard to this.

Nature Of Welded Splices

Provided that the reinforcing bars possess the properties that make them
suitably weldable, welded splices may be formed in one of the following
ways :
(a) by means of butt welds produced by flash welding;
(b) by means of butt welds, with formed edges,produced by electric arc
welding ;
(c) by means of lapjoints produced by electric arc welding with longitudi-
nal weld beads.
In general,welded butt joints (a)or (b)should be used,except in the case
of bar splices at connections between precast reinforced concrete members
and of splices inside formwork for which lap joints with longitudinal weld
beads (c) are considered to be preferable.
In this last-mentionedcase the strength of the lap welds can be calculated
on the assumption that the shear strength of the welds (lengthx thickness
of weld beads x 65% of the tensile strength of the deposited weld metal as
indicated by the supplier of the electrodes) should be at least 1.5 times the
guaranteed ultimate strength of the bars to be welded and of any extra
splice bars used. In addition, the length of the longitudinal weld beads
should not exceed five times the diameter of the bar.
108
Finally,whichever type of welded splice is adopted the splice should be
exactly symmetricaland be as convenient as possible to execute.

Positioning Of Welded Splices

Welded splices should be confined to straight portions of reinforcing bars.
Besides,they should be staggered in the longitudinal direction by a length
equal to at least twenty times the bar diameter (204)in the case of plain
round bars and at least ten times the bar diameter (104)in the case of
deformed bars.
Finally, in any one particular cross-sectionthe transmission of tensile
forces by welded splices in bars should be effected by not more than half
the total cross-sectionalarea of the reinforcement (in the case of bending
with or without compression)or not more than one-thirdof the total cross-
sectional area of the reinforcement (in the case of tension with or without
bending).

Strength Of Welded Splices

Provided that the quality of the welds is strictly supervised,welded splices
in tensile reinforcement and in compressive reinforcement can permissibly
be utilised up to 100% of the strength of the bars they connect.
This rule is allowed to be applied only ifthe welding operations are carried
out by a properly qualified welder and are under constant and strict
supervision.

7.2.5 S P A C I N G OF R E I N F O R C I N GB A R S

The spacing of reinforcing bars -i.e., the distances between adjacent bars
within the section -should be sufficient to enable the concreting to be done
in an entirely satisfactory manner. In particular, the bars should be so
spaced that the freshly mixed concrete can be properly placed without risk
ofsegregation and that the concrete surrounding the reinforcement can be
efficiently vibrated.
The values stated below correspond to normal ‘insitu’concreting.Subject
to special justification,they may be reduced in the case of factory-made
prec;astmembers or in the case of temporary structures.

Spacing Of Bars In The Same Vertical Line

The clear vertical distance between two bars in the same vertical line in
the cross-sectionof a member should be at least equal to the largest of the
following values:
(a) 1 cm
(b) three-quartersof the diameter (0.754)of the thicker bar .
 1o9
(c) 0.5times the maximum size ofthe aggregates (inthe case of rounded
aggregates) or 0.6times (inthe case of crushed aggregates).

Spacing Of Bars In The Same Horizontal Layer

The clear horizontal distance between two bars in the same layer should be
at least equal to the largest of the following values:
(a) 2 cm
(b) the diameter of the thicker bar
(c) 1.2times the maximum size of the aggregates (inthe case of rounded
aggregates) or 1.4times (inthe case of crushed aggregates), if there is
only one horizontal layer of reinforcement,
or
1.4times the maximum size of the aggregates (inthe case of rounded
aggregates) or 1.6times (in the case of crushed aggregates), if there
are several horizontal layers of reinforcement.

Groups Of Bars In Contact

In any particular vertical line in the cross-sectionthe designer may always
permissibly provide two bars which are in contact with each other.O n the
other hand,he should not provide more than two bars in contact with one
another in one and the same vertical line,unless he makes special arrange-
ments to enable the freshly mixed concrete to fill all cavities perfectly.
In any particular horizontal layer the designer should not provide two
bars which are in contact with each other, unless there is sufficient space
Sufficient clear space
to insert a vibrator

Figure 7.2

on each side ofeach group of two bars to insert a vibrator.O n no account

should more than two bars be placed in contact with one another in one and
the same horizontal layer.
O n these conditions the ab,ove requirements are applicable to groups of
bars in contact,provided that each group is replaced by a single fictitious
bar with the same centroid as the group and with a cross-sectional area
equal to the total cross-sectionalarea of the bars in the group.
To facilitate the placing of the concrete it may sometimes be advantageous
to form groups of three bars (see Figure 7.2).In this way very good embed-
ment of the steel and good quality of the concrete can be ensured.
110
Laboratory tests have shown that when ribbed deformed bars (high-
bond bars with ribbed surface profile) are used,the space inside such a group
of three bars is filled with mortar from the concrete.
Also,to facilitate the placing and vibration of the concrete,it is sometimes
helpful to install binders or stirrups in twinned pairs.

Bar Spacings At IntersectionsO f Beams

The designer should, as far as possible,avoid excessive concentration of

reinforcing bars in certain zones of the structure.But such concentration
is hard to avoid at intersections of beams if the bars are placed at levels
that are close together and especially if, in such places,suspender bars are
essential for transmitting the forces. Such concentration of reinforcement
may interfere with the proper placing of the concrete and impair the quality
thereof.
To facilitate the placing of the concrete and improve its quality in zones
where there is considerable concentration of bars,the designer may,if need
be, specify a special concrete,with a smaller maximum size of aggregate,
for such zones. In that case the maximum size of the aggregate particles
should not exceed the ratio of the volume of the mould to the total surface
area of its walls (calculatedas the sum of the areas of the concrete faces and
of the surfaces of the reinforcing bars).

7.2.6 C O N C R E T E C O V E R TO R E I N F O R C E M E N T

The concrete cover to the reinforcement-i.e.,the distance from the bars to

the walls of the formwork or to the free surface of the concrete-should be
sufficient to enable concreting to be done in an entirely satisfactory manner,
so that, in particular, all risk of segregation is obviated and the concrete
can be compacted to the density that is essential to providing suitable
protection of the reinforcement against corrosive agents.

Arrangements C o m m o n To All Reinforcement Bars

The clear amount ofcover between any point of the externalgenerating lines
of any bar (longitudinalreinforcement,transverse reinforcement,connector
reinforcement or steel-fixing bars) and the nearest face of the concrete
should be at least equal to:
(a) 1 cm,if the concrete faces are protected not only from all chemical
attack, but also from all atmospheric influences and condensation
phenomena.
(b) 2cm,if the concrete faces,even though they are not exposed to any
particular chemical attack,are nevertheless exposed to atmospheric
influences(externalmembers)or to condensationphenomena (kitchens,
 111
bathrooms,etc.) or if they are in permanent contact with water (tanks,
pipes,etc.).
(c) 4cm, if the concrete faces are exposed to marine atmosphere or a
particularly corrosive atmosphere.
O n the other hand,the cover to the reinforcementshould not exceed 4cm.
If it does exceed this value in exceptional cases,the designer should provide
an additional reinforcing network (‘skinreinforcement’)within the thickness
of the concrete cover;this reinforcement should conform to Section 6.1.3.

Arrangements Peculiar T o The Main Reinforcing Bars

The clear amount of cover between any point of the external generating lines

t t

Figure 7.3
of a main reinforcing bar and the nearest face of the concrete should be at
least equal to 1.5times the diameter of that bar.
This rule is complementary to the rule stated above.
A practical example of cover in a structure is shown in Figure 7.3 (two
bars crossing each other at right angles,and a stirrup)

Groups Of Bars In Contact

The rules given in clauses 7.2.6are applicable also to groups of bars in

contact with one another.In a group of this kind it is the bar nearest to the
walls ofthe formwork or to the free surface ofthe concretethat must conform
to the rules.

7.3 A R R A N G E M E N T S PECULIAR TO VARIOUS STRUCTURAL

MEMBERS
7.3.1 COLUMNS

Minimum Section
The least transverse dimension of a column should be not less than 25 cm.
If it is not possible to fulfil this requirement, e.g.,if the columns have to
112
be accommodated within the thickness of walls or partitions, the least
transverse dimension may be reduced below 25cm on the following
conditions:
(a)the reference strength of the concrete (Section3.2.1)should be at least
200 bars;
(b) the geometrical percentage of longitudinal reinforcement should be at
least one per cent (1 %).

Longitudinal Reinforcement

Minimum Elastic Limit

The longitudinal reinforcing bars should have a reference elastic limit o,

of at least 4 o00 bars (o,2 4O00 bars); they may,at the designer’s option,be
plain round bars or deformed bars (high-bondbars).
If it is not possible to fulfil this requirement,i.e.,if the longitudinal rein-
forcing bars have a reference elastic limitoeof less than 4 O00 bars (o,<4 O00
bars),the basic compressivestrength must be reduced in the ratio of o,/4 000,
i.e.,by multiplying it by a reduction coefficient o,/4 000.

Minimum Percentage

The minimum geometrical percentage of longitudinal reinforcement,

referred to the total cross-sectionalarea of the column should be:
(a)for a corner column:
3
(
4 2 -1 O00 .ß. l+- 4y)

(b) for an edge column:

ab 2 -2.5 .ß .(1
1 O00
)?+
(c) for an internal column:

where oe denotes the reference elastic limit of the steel (above 4000 bars,
apart from exceptional cases,see above) and ß denotes the ratio between
the external direct force (as determined from the characteristic loadings)
and the resistance to direct force that the concrete section can develop (equal
to 0.750b. B’).
Introduction of the coefficientfl correspondsto the case where the column
has an excessive concrete section.Applying this coefficient is equivalent to
 113
referring the minimum percentage of reinforcement to the concrete section
that is strictly necessary for the equilibrium of the external direct force.

Examples Of Application

It will be assumed that the steel has a reference elastic limit of 4200 bars
and that the concrete section of the columns is 10% ‘excessive’(asenvisaged
in the preceding paragraph, p = (1/1.15).
The basic compressive strength of the steel should be taken as:
4200
-= 2335 bars
1.80
with a minimum geometrical percentage of:
0.003O. (1/1.15).[i +(4000/4200)] = 0.534% for the corner columns
0.0025.(1/1.15). [i +(4 000/4200)] = 0446% for the edge columns
0.002O .(1/1.15)[i. +(4 000/4200)] = 0.357%for the internal columns
If, on the other hand, the steel is assumed to have a reference elastic limit
of 3 600 bars, the basic compressive strength of the steel should be taken as:
--=--
600 6oo 240 - 1 800bars (insteadof 2 O00 bars)
1.80 ‘4000 1.80
with a minimum geometrical percentage of:
0~0030(1/1~15) [i +(4000/3240)] = 0.583%for the corner columns
0.0025(1p.15)[ 1 +(4000/3240)]= 0.486%for the edge columns
0.0020(1/1.15 [i)+(4000/3240)] = 0.389%for the internal columns
If, for compelling structural reasons, the least transverse dimension of the
columns will have to be made less than 25 cm, these minimum geometrical
percentages should be increased to 1 %for all the columns.

ConstructionalArrangements

The longitudinal reinforcing bars should be so distributed within the section

in the vicinity of the sides of the column as to give the latter the best possible
flexural resistance in the most unfavourable directions. In particular, in a
column with an elongated rectangular section, the distance between two
adjacent longitudinal bars should not exceed the width of the section (i.e.,
the dimension of the shorter side).

TransverseReinforcement
Minimum Diameter

The diameter of the transverse reinforcing bars should be at least equal to

5 m m and not less than one-quarter of the diameter of the thickest longi-
tudinal reinforcing bar.
114
M i n i m u m Percentage

The minimum geometrical percentage of transverse reinforcement,referred

to the total concrete cross-sectionalof the column,should be 0.50%.

M a x i m u m Spacing

The spacing of the transverse reinforcement,i.e.,the distance between two

consecutive planes in which transversereinforcement is disposed,should not
exceed twenty-fivecentimetres nor twelve times the diameter of the thinnest
longitudinal reinforcing bar.

Constructional Arrangements

Each layer of transverse reinforcement should be so disposed as to :

(a) form a continuous girdle around the outside of the member, en-
closing all the longitudinal reinforcing bars within it;
(b)secure each longitudinal reinforcing bar and prevent any movement
thereof towards the nearest face or faces of the member.
This second condition can be satisfied in practice only if the possible
outward movement of a longitudinal bar is restrained by direct tension in a
straight element of the transverse reinforcement.In the case of columns of

Bar not s e c u r e d l Correct arrangement

Figure 7.4

polygonal or circular cross-sectionalshape this condition is considered to

be satisfiedby the hoops or helicalbindings that form the transversereinforce-
ment. O n the other hand,in square or rectangular columns this condition
requires the designer to check that all the longitudinal bars are located
either at one corner of the binders or within the loop of a hairpin bar or a
link specially provided for the purpose.

Placing The Reinforcement In Position

The reinforcement cage, comprising the longitudinal and the transverse

bars, should be sufficiently rigid to ensure that,while it is being installed
 115
and also during the subsequent concreting operation,there is no risk of the
bars being displaced from their theoretical positions.

7.3.2 M E M B E R S R E I N F O R C E D B Y B I N D I N G
Geometrical Dimensions Of The Zone Provided With Binding
Itis pointed out that,in accordance with the rules laid down in Section 6.1.1,
the height of the zone provided with binding in a compression member
should not exceed twice the least transverse dimension of that zone.Besides,
this least transverse dimension should be not less than twenty-fivecenti-
metres.

Minimum Percentage O f Binding

It is pointed out that,in accordance with the rules laid down in Section 6.1.1,
the geometrical percentage of binding reinforcement,referred to the total
volume of the zone provided with binding, should be at least six per mille
(0.60x).
Arrangement Of Binding Reinforcement
Helical Binding Or Hoops
It is pointed out that,in accordance with the rules laid down in Section 6.1.1,

A
-. -A Section A A

rigure 7.5

the pitch of the helix or the spacing of the hoops should not exceed one-
fifthof the diameter of the core enclosed by the binding.
Furthermore,splices in the coils of circular helices should be formed,
not by merely lapping the bars, but by an anchorage device comprising a
minimum lap length of twenty bar diameters followed by two anchorages
116
formed by curvature and having their ends bent so as to point to the middle
of the core.
The end anchorages of the coils should terminate in straight portions
bent back parallel to the axis of the helix (see Figure 7.5).

Binding Reinforcement In M a t s

It is pointed out that,in accordance with the rules laid down in Section 6.1.1,
the spacing of the mats should not exceed one-fifth of the least dimension
of the core provided with binding.
Also, the mats should consist of bars bent hairpin-wise to and fro and

Figure 7.6

placed with the bars extending alternately in mutually perpendicular direc-

tions.The ends of the bars of each mat should be suitably anchored into the
interior of the concrete or,alternatively,welded to the preceding loop of the
mat (seeFigure 7.6).

7.3.3 F L E X U R A L MEMBERS

Tensile Longitudinal Reinforcement

It is pointed out that,in accordance with the rules laid down in Section 6.2.3,
the tensile longitudinal reinforcing bars should be designed on the basis of
a diagram obtained by shifting the enveloping curve of the bending moments
parallel to the centre-lineof the member in the most unfavourable direction
by an amount equal to the effective depth of the section.These bars should
be anchored outside this ‘displaceddiagram.
Furthermore,at the supports the designer should also extend and anchor
a sufficient proportion of the longitudinal reinforcing bars to be able to
absorb a tensile force equal to T+ M/zat the supports.
If the moment M at the support is zero (asin a simply-supportedbeam),
 117
the longitudinal bars should be able to absorb a tensile force equal to the
shear force T.
O n the other hand,if there is a positive bending moment M at the support
(or if there is a negative moment,but smaller in absolute value than T.z),
the longitudinal bars should be able to absorb a tensile force equal to
T+ M/Tat the supports.

Compressive Longitudinal Reinforcement

It is pointed out that,in accordance with the rules laid down in Section 6.1.1,
the bars of diameter 4 forming the compressive longitudinal reinforcement
should be secured by means of binders or stirrups of suitable section and
spaced less than 124 apart.

Longitudinal Distribution Reinforcement

It is pointed out that,in accordance with the rules laid down in Section 6.1.1,
if the depth of the web (in metres) of a flexural member exceeds the value
1 -10-40, (in bars), the designer should provide longitudinal distribution
reinforcement at each face of the web. This longitudinal distribution re-
inforcement,called skin reinjorcement, should be ofthe same grade of steel
as the main tensile reinforcement.Its geometrical percentage,referred to the
web section excluding the embedment section of the main tensile reinforce-
ment, should be at least 0.05% at each of the two faces.Furthermore,the
individual bars of this reinforcement should not be spaced farther than
20cm apart.
Also, the designer may distribute the main tensile reinforcing bars in a
graduated fashion over a fairly substantial proportion of the bottom part
of the beam,taking due account of the precise positions of the bars in cal-
culating the lever arm of the internal forces and in the ultimate strength
analysis of the beam.

Transverse Reinforcement

The transverse reinforcement of a flexural member has to perform a number

of functions at one and the same time:securing the reinforcing bars and
restraining them against buckling, providing the web with strength to
resist shear force,connecting the web to the compression flange,ensuring
the efficiency of the anchorage of the tensile reinforcing bars and,in a more
general sense,fully complying -for all loadings and possible combinations
ofloadings-with the fundamentalrule called the ‘ruleofconnector reinforce-
ment’.
In addition,all the bars of the transverse reinforcement should be totally
anchored.This can be achieved by looping the binders and stirrups around
118
the main reinforcing bars, provided that the angle that these binders and
stirrups make with the longitudinal bars is not less than 65".
In other words,the anchorage of binders and stirrups inclined at an angle
of more than 65"in relation to the longitudinal axis of the member should
be specially investigated and designed accordingly.
Finally,the spacing of the successive layers of transverse reinforcement
should not exceed twenty centimetres (for deep beams) nor exceed eighty-
five hundredths ofthe effective depth (forbeams oflow and ofmedium depth).
In the frequently encountered case of a rectangular rib joined to a top
slab and containing transverse reinforcement which is perpendicular to the

centre-lineof the rib and anchored by looping or hooking round the longi-
tudinal bars,it is not sufficient merely to provide stirrups in separate rows
around each longitudinal bar. The bottom (tensile)face of the rib should
also be reinforced transversely,either by means of general binders, or by
means of short links,or by other means (see Figure 7.7).

Changes In The Geometrical Shapes Of Sections

In the design calculations for a flexural member it is often found necessary

to vary the depth or the width of the cross-sections.The designer may make
the depth or the width of the section vary continuously all along the member
or he may alternatively,and more conveniently,increase the depth locally
in the form of haunches at the supports.The slope of such haunches in rela-
tion to the longitudinal axis of the member should not exceed one-third.
 119
7.3.4 SLABS A N D P L A N E S T R U C T U R E S

These rules and requirements relate to slabs and plane structures loaded
perpendicularly to the middle plane and not more than thirty centimetres
thick.

Mid-Span Reinforcement

M a x i m u m Diameter

The diameter of the bars of the mid-span reinforcement should not exceed
one-tenth(l/lO)of the thickness of the slab or structure.

M a x i m u m Spacing

In accordance with Section 6.1.3, the spacing of the mid-span reinforcing

bars should not exceed the values (statedin centimetres) shown in Table 7.3:

Table 7.3

Guaranteed elastic limit 2000 3000 4000 5000 6000

of the steel o, bars bars bars bars bars

Plain round bars 20 20

Deformed bars (high-bond) 20 20 20 17.5 15
Welded fabric of drawn wires 17.5 15 10

In addition,if the reinforcement consists of weld steel fabric,the larger

side length b of the mesh should be not more than three times the smaller
side length a ( b < 3a). Also,(a+b)/2 should not exceed the values given in the
above table.

Ratio Of Sections In Two Mutually Perpendicular Directions

The cross-sectionalarea of the reinforcing bars corresponding to the span-

ning direction in which the lesser values of the bending moment occur
should be at least one-quarter of that of the bars corresponding to the
spanning direction in which the larger values of the bending moment occur.

Edge Reinforcement

Along the supports and edges of slabs and of the panels of which plane
structures are composed the designer should provide edge reinforcement
120
whose local percentage should be at least one-quarterof that of the mid-
span reinforcement corresponding to the spanning direction in which the
larger values of the bending moment occur.

Corner Reinforcement
The designer should give due consideration to the possible risks of cracking
at the cornersand should,ifneed be,provide appropriate connector reinforce-
ment.

Punching Shear Reinforcement

In the vicinity of columns, without enlarged heads, supporting flat-slab
floors the designer should provide punching shear reinforcement consisting
of plain round bars and formed in one of the following ways:

Vertical Or Inclined Binders

The punching shear reinforcement may consist of vertical or inclined binders
s 0.5h- I ~0115h

Figure 7.8

having the arrangement,shape and size shown in Figure 7.8(verticalbinders)

and Figure 7.9(inclined binders).
The binders should be located outside the perimeter of the area to which
the locally concentrated force is applied and within a zone whose width is
approximately 1.5h. They should be spaced not farther than a distance of
 121

Figure 7.9

~0.5h
4
\ I
h

+ -?Oh c
T

Figure 7.12
122
0.75h apart. Besides, in order to ensure suitable anchorage, the binders
should completely surround the horizontal tensile reinforcing bars.

Bent-Up Bars
The punching shear reinforcement may consist of bent-upbars arranged in
one or in two layers,as shown in Figure 7.10(barsbent up in one layer)and
Figure 7.11 (barsbent up in two layers).
An equal number ofbars should be bent up in both directions.The bent-up
bars should be located over the perimeter of the area to which the locally
concentrated force is applied and should also extend outside this area to a
distance of approximately0.25hfrom it. If the bars are bent up in two layers,
these should each have approximately the same number of bars (seeFigure
7.12).
If the area to which the locally concentrated force is applied has a square
or rectangular outline with large dimensions in relation to the effective
depth of the slab (a+b > 6h), the punching shear reinforcement should be
concentrated towards the corners of that area,these being the zones where
the concentration of forces occurs.

Other Systems OfReinforcement

Among the other systemsof reinforcement with plain round bars which may
be used for punching shear reinforcement, subject to proper justification,
mention may be made of systems comprising bars bent to a ‘castellated’
profile as shown in Figure 7.13.
Finally,punching shear reinforcement consisting of steel plates or rolled
sections (‘shearheads’)may be employed, but the arrangements adopted
Flexural reinforcement

Ca st e li át io n s

Figure 7.13

for such reinforcement should first be properly justified by means of tests

with regard to the effectiveness of its contribution to the punching shear
strength of the slab.Also,the building owner’sapproval will be required.
The total cross-sectionalarea of the punching shear reinforcement con-
sisting of plain round bars should be such that the sum of the vertical com-
ponents of the forces developed in them corresponds to at least 75 %of the
 123
value of the locally concentrated force applied to the slab. These forces
developed in the bars should be calculated on the assumption that the stress
in the punching shear reinforcement is equal to its basic strength.
The purpose of punching shear reinforcement is to ensure better ductility
of the slab with regard to deformation ofthe slab in the vicinity of the locally
concentrated force.However,the presence of such reinforcement does not
warrant taking into account a value of the locally concentrated force that
is larger than has been defined in Section 6.2.5.In fact,tests have shown the
effectiveness of this kind of reinforcement to vary greatly with the various
parameters that affect it, especially the quantity of flexural reinforcement
with which the slab is provided. In the present state of knowledge it is not
possible to make a reliable assessment of the influence of these parameters.
 8
PREPARATION OF DESIGNS

8.1 CALCULATIONS
8.1.1 BASIC D A T A F O R DESIGN C A L C U L A T I O N S

Before preparing any detailed calculations the designer should obtain the
building owner’s(or his representative’s)agreement on the basic data to be
adopted for these calculations.
These data must be conformed to by those entrusted with the construction
and by the users of the structure,more particularly with regard to:
(a)the bearing pressure of the foundations on the soil;
(b) the nominal working loads,i.e. fixed or mobile superimposed loads
(live loads);
(c) the reference value of the compressive (and,where relevant,the tensile)
strength of the concrete;
(d) the reference value of the elastic limit of the steel and also the other
mechanical properties specified in the conditions of approval.

8.1.2 A R I T H M E T I C A L A C C U R A C Y OF THE C A L C U L A T I O N S

Assuming the basic data for the design calculations to have been accepted
by the building owner or his representative,the arithmetical check of these
calculations is to be considered satisfactory if the deviations found in them
do not exceed 3%(plusor minus). If this requirement is fulfilled,the building
owner will not be entitled to use such deviations as grounds for demanding
a modification of the design; in the contrary case, however, he will be
entitled to demand such modification.

8.1.3 S U B M I S S I O NOF C A L C U L A T I O N S

If the design is prepared by or on behalf of the contractor,the contract

documents should state whether the contractor is required to supply the
I24
 125
building owner or his representative with a complete set of typewritten
design calculations for all the component members of the structure.Other-
wise the contractor need merely, if so required, make available to the
building owner or his representativethe handwritten draft calculations.

8.2 DRAWINGS
8.2.1 P R E L I M I N A R Y D E S I G ND R A W I N G S

Preliminary design drawings accompanying the contractor’stender should

be drawn to a scale of 1 :50;they may be of a schematic character and need
not show reinforcement details. However, on putting in his tender the
contractor is deemed thereby to undertake the obligation,if he is awarded
the contract,subsequently to conform to all the requirements of the code of
practice as embodied in this Manual.

8.2.2 WORKING DRAWINGS

The working drawings should show,with the exactness and precision con-
sidered necessary by the building owner or his representative,all the geo-
metrical shapes of the component members of the structure and all the
reinforcement details.

Arrangements Common To All Working Drawings

All working drawings should give ;he following information in a special
boxed-inspace on the title page of the folded document:
(a) Name of the structure;
(b)Name of the building owner;
(c) Name of the building owner’srepresentative (architector consulting
engineer);
(d)Name of the contractor (if the design is prepared by or on behalf of the
contractor);
(e) Name of the reinforced concrete designer;
(f) Names of the draughtsmen;
(g) Title of the drawing;
(h)Number of the drawing;
(i) Date of drawing;
ci) Scale;
(k) Amendments;
(i) Name and signature of the person responsible for the reinforced
concrete design.
The following information should also be very clearly shown on all work-
ing drawings:
(a)the most-unfavourableloads transmitted to each of the foundations;
126
(b)the nominal working loads (fixed superimposed loads.and live loads)
and the weight of top and bottom surfacingsand finishes;
(c) the guaranteed minimum compressive strengths (and,where relevant,
tensile strengths) of the concrete;
(d)the guaranteed mechanical properties of the steel, including more
particularly : the guaranteed elastic limit and,for all bars that have to
be bent, the permissible radius of curvature and the corresponding
minimum diameter of the bar-bendingformer.
Finally, the working drawings should show all construction joints, all
fixing holes and sockets to be subsequently concreted or grouted up, and
all openings,and should,in the case of a floor,indicate whether it is pro-
vided with underfloor heating.

Formwork Drawings
The formwork drawings should show the various planes, sections and
elevations of the actual structural surfaces, not including any finishes or
coatings. They should, in particular, show all the dimensions necessary
for the correct setting-outand complete execution of all the components of
the structure.As regards heights and thicknesses,the formwork drawings
should show the total heights (or depths) and thicknesses of the concrete,
not including the various surfacingsand finishes.

Reinforcement Drawings
The reinforcement drawings should show all the details necessary for the
correct execution and accurate fixing of the reinforcement. They should,
in particular,clearly indicate the guaranteed minimum elastic limit of the
steel,the length of each bar, the geometrical features of curves and bends,
the diameter of the bar-bendingformer,the bar spacings and the distances
between the bars and the concrete faces,more particularly at intersections
of beams and at the junctions of slabs and columns.
In addition, for those parts of the structure where the reinforcement is
particularly dense the drawings should comprise large-scaledetails clearly
showing the interlacing of the bars (and,if need be, the essential arrange-
ments to enable the concrete to be properly placed).
Finally, if the simultaneous use of several different grades or types of
reinforcing bars is to be allowed,the reinforcement drawings should clearly
show the grade or type of steel of the various bars. If symbols are employed
for distinguishing these various steel grades or types,the meanings of such
symbols should be clearly explained in a very prominent boxed-in space
on the drawing.

8.3 CONDITIONS OF EXECUTION

Of the conditions of execution which have received the building owner’s
approval,the designs should define and justify those which are liable to
 127
affect the strength or the stability of the structure and which,in more general
terms,may affect its behaviour during the period of construction or during
its subsequent service life.
In particular,the designs should define and justify:
(a)the conditions of construction and stability of the formwork and its
ability to withstand the pressure ofthe freshly placed concreteand such
consequential effects as this may have on the concreting procedure;
(b)the treatment to be applied to the exposed concrete surfaces and the
possible consequential effects thereof upon the design and treatment
ofthe formwork surfaces;
(c) the devices for fixing the reinforcing bars in relation to the formwork;
(d)the procedure for construction in successive portions, and the justi-
fication of the strength and stability of the structure in each of the
successivestages of construction;
(e) the constructionjoints and their possible effects upon the strength and
stability of the structure;
(f) the conditions of removing the propping and formwork;
(g)the temporary shrinkagejoints;
etc.
 9
EXECUTION OF STRUCTURES

9.1 R E Q U I R E M E N T S PERTAINING TO FORMWORK

9.1.1 C L A S S I F I C A T I O NA N D C O N S T R U C T I O N O F
FORMWORK

Classification Of Formwork

Four categories of formwork (shuttering)or faces of moulds are to be dis-

tinguished,classified as follows in ascending order of quality:
(a) ordinary formwork;
(b)carefully finished formwork;
(c) fine-facedformwork;
(d) special formwork.
For each concrete surface ofa structure the category of formwork to be
applied should be specified in the contract documents.

Ordinary Formwork

If ordinary formwork consists of sawn boards merely assembled side by side,

these should be properly joined edge to edge.
If ordinary formwork consists of fibreboard or plywood panels merely
assembled side by side,these should be properly joined edge to edge.

Carefully Finished Formwork

If carefully finished formwork consists of sawn boards planed (wrought)on

all four surfaces and merely assembled side by side,the boards should be
properly joined edge to edge.
128
 129
If carefully finished formwork consists of fibreboard or plywood panels,
these should present a surface equivalent to that of planed timber. The
tightness (againstleakage) of the joints should be ensured by appropriate
means.
Adhesive strips should preferably consist of cellular material.They should
be installed within the thickness of the joint,because tapes that are simply
stuck to the surface tend to become detached under the combined action of
the formwork release agent and the vibration of the concrete.
Carefully finished formwork may be made of steel.The steel plate surfaces
in contact with the concrete must be quite free from projections and from
buckling. The tightness (againstleakage) of the joints should be ensured by
appropriate means.
The joints of steel formwork may, for example, be sealed by means of
adhesive tapes applied flat to the internal formwork surfaces,or by means of
resilient foamed material placed in the joints, or by means of a mastic
compound.

Fine-Faced Formwork

If fine-faced formwork consists of boards planed on all four surfaces,the

arrangement of the boards and the method of forming thejoints should,for
any one and the same concrete face,be specially designed from the aesthetic
point ofview.These boards may simply be assembled side by side or they may
be fitted together with tongue-and-groovejoints;in both cases the faces in
contact with the concrete should,if necessary,be smoothed by planing after
assembly.
Fine-faced formwork consisting offibreboard or plywood panels should
conform to the requirements laid down for carefully finished formwork of
similar construction.But,in addition,the arrangement of the joints should
be designed from the aesthetic point of view.These joints should be caulked
and smoothed with a mastic compound.Adhesive tape must not be used.
If fine-facedformwork is made of steel,the steel plate surfaces in contact
with the concrete must be quite free from projections and from buckling.
The joints should be sealed by filling them with packings of resilient foamed
material or any other equivalent method.
The means employed for sealing the joints must not protrude from the
internal surface of the formwork.In addition,the joints should be caulked
and smoothed with a mastic compound.Adhesive tape must not be used.
The pattern of the joints should be designed from the aesthetic point
of view.

Special Formwork

Where necessary,the requirements applicable to special formwork should

be specified in the contract documents.
130
Special formwork may be necessary for producing particular shapes and
for making precast units.

Joints In Formwork

If adhesive tapes are used for sealing the joints in formwork,they should
adhere so firmly that there is no risk of their becoming detached during
concreting,even if formwork release agents are employed.
Detachment of adhesive tapes is very objectionable for the aesthetic
appearance of the concrete surfaces.It is possible to obtain adhesive tapes
which adhere firmly to their base if the latter is not greasy to start with and
which remain adhering when formwork release oil is subsequently applied,
i.e.,after the tape has been stuck to the formwork.
O n the other hand,there are as yet no tapes on the market which can adhere
to a greasy base. For this reason fibreboard or plywood panels that have
been oiled cannot be re-usedwith adhesive tape for sealing the joints.
For the same reason,steel formwork should be degreased before adhesive
tape is applied to it.

Tightness Of Formwork

Formwork should be sufficiently tight to ensure that no harmful loss of

grout from the concrete can occur during concreting.

9.1.2 M E C H A N I C A L P R O P E R T I E S O F F O R M W O R K

Mechanical Strength

Formwork should be sufficiently rigid to resist -without undergoing

settlement or harmful deformations -the loads, superimposed loads and
forces of any kind to which it is subjected during the execution of the work,
including more particularly the forces set up by the compactionand vibration
of the concrete.
The stresseswhich develop in the formwork and in the parts ofthe structure
serving as support in consequence of the action of the forces to which they
are subjected during the execution of the work should remain below the
working stresses of the materials.
For example, if superimposed load has to be applied to a floor before
removal of the formwork,the latter and also the props supporting it should
be designed with due regard to this possibility.
Similarly,in multi-storey buildings it is necessary, when concreting is
done,to satisfy oneself of the strength of the formwork and propping which
support each floor.
The formwork designers’attention is furthermore called to the need to
 131
provide efficient bracing of formwork and props in order to obviate any risk
of buckling.

Sag A n d Camber
The amounts of sag and camber to be given to formwork,centring,etc.,
should be determined with reference to the sag (downward deflection) or
camber (upwarddeflection) envisaged for the completed structure.
For long-span beams it is recommended that the formwork be given a
camber which is so determined that after removal of the formwork the
aesthetic appearance of the structure is to be considered satisfactory.

9.1.3 P R E P A R A T I O N OF F O R M W O R K

Cleaning
Immediately before concreting,the formwork should be carefully cleaned
so as to remove all dust and rubbish.
If need be,separately closable apertures should be provided in the form-
work to facilitate cleaning with the aid of compressed air.
If the site has a compressed air supply -which is usually the case if internal
vibration is used for compacting the concrete-it is most advisable to finish
off the formwork cleaning operations by using compressed air.

Wetting
Before the concrete is placed, the following types of formwork should be
abundantly sprayed with water :
(a) ordinary formwork made ofsawn boards;
(b) ordinary formwork made of fibreboard or plywood panels;
(c) carefully finished formwork made of sawn boards.
Spraying may have to be done in several successivestages so as to wet the
timber as thoroughly as possible. However,the wetted surfaces should not
be streaming wet,and excess water should be carefully removed,preferably
with the aid of compressed air.
The object of wetting the formwork is to make the joints close up and to
prevent the concrete at these surfaces from drying out too rapidly.Wetting
is particularly important during dry and warm spells.

Oiling
In order to facilitate the subsequent striking (removal)of the formwork,
it is necessary,before placing the concrete,to apply a coating of oil to:
(a)all steel formwork;
132
(b)carefully finished formwork made of plywood or fibreboard panels,
as well as all fine-facedformwork in so far as it is not treated with a
special release agent;excess oil which collects at the bottom of the
formwork should be mopped up before concreting.
The oils employed for the purpose should be special ‘mouldrelease oils’.
They should be clean (i.e., they must not leave any stains on the concrete
surfaces) and not have an acid reaction.
The oiling of fine-facedformwork made of sawn boards,plywood or fibre-
board should be done by the successive application of at least two successive
coatings,so as to achieve proper impregnation of the wood.
Acid oils react with the concrete and cause powdering of the concrete
surfaces.
Furthermore,if no intermediate coat of plaster is to be applied to the con-
crete,it is advisable to check that any paint that will subsequently be applied
to floors,wails or ceilingsis not incompatible with the oil or other formwork
release agent employed.

Maintenance

Ifa number of re-usesare to be obtained from the same formwork,the latter

should be thoroughly cleaned and reconditioned before each re-use.

9.1.4 S T R I K I N GT H E F O R M W O R K

Striking (removing)the formwork should be done carefully,without jolts

and by the exertion of purely static forces.
Attention is called to the advisability of measuring the deflections when
striking the formwork of certain structures of an exceptional character
(vaults,cantilevers,large-spanstructures,etc.).
In many cases it can be judged from an inspection of the deflection dia-
grams whether removal of the formwork should be continued or whether
loading tests should be performed.
In determining the time that must elapse before removal of formwork it is
necessary to take account of the retardation that a fall in temperature or
exposure to wind may cause in the hardening of the concrete,especially if
cement with a relatively high content of blast-furnaceslag is used.
As an approximate indication,the following formwork striking times may
be adopted in the case of a normal concrete (containingno settingaccelerator
or retarder):
(a)2-3 days if the loading exerted by the concrete on the formwork is
small (shells,walls,etc.);
(b)6-8 days for formwork of members carrying only their own weight
(e.g.,floor slabs);
(c) 12-15 days for formwork and propping of cantilevered members or
load-bearingmembers of the structure.
 133
Shrinkagejoints and expansion joints should be free from anything that
may adversely affect their functioning.
In general,the removal of the formwork should be done in such a way as
not to produce any stress exceeding the normal working stresses of the struc-
ture. In particular, when removing the formwork of large canopies or
awnings, it is necessary always to start at the free end, the props being
released progressively asthe removalofthe.formworkproceeds.Furthermore,
if not all the formwork is removed at once and an intermediate row of props
is left in position,these should be so designed as to enable the formwork
removal to continue.It is not permissible to strip off all the formwork and
then replace temporary struts.
If it should occur that some reinforcing bars are accidentally stripped bare
of concrete as a result of removing the formwork,it is advisable to make a
careful inspection of the defective zone before making good the damage.

9.2 M A T T E R S PERTAINING TO REINFORCEMENT

9.2.1 TESTS ON T H E REINFORCING S T E E L

Tests In T h e Factory Before Delivery

Unless particularly required by the building owner and specified in the con-
tract documents of the structure,no acceptance tests in the factory are re-
quired for reinforcing steel which is supplied with the manufacturer’s
guarantee.If, on the other hand,such tests are considered to be necessary,
they should be performed for each nominal diameter -and by batches of
not more than twenty tons -on test specimens selected at random.

Tests O n T h e Site After Delivery

In all cases check tests should be performed after delivery of the reinforcing
bars to the construction site. In order to yield results for discussion (ifneed
be) with the manufacturer,these tests should be performed on specimens
selected at random and tested in a laboratory approved by the building
owner,the contractor and the manufacturer.

Checking The Reference Values Of The Mechanical Properties

Unless otherwise specified,these tests should be confined to checking the
guaranteed minimum elastic limit of the steel,as obtained from the experi-
mentally determined stress-strain diagram by dividing the tensile force
corresponding to a residual strainof0.2%by the nominal cross-sectionalarea
of the test specimen.
Unless otherwise specified,the number of test specimens should be one per
batch.
134
If the result obtained reaches or exceeds the guaranteed minimum elastic
limit,the test should be considered satisfactory and the batch of reinforcing
bars be accepted as conforming to the manufacturer's guarantee.
If not,then,for each batch as defined in Section 9.2.1,three more speci-
mens will have to be taken and tested.The following values should be cal-
culated for the total of fourtest results obtained:
the arithmetical mean o,:

the correspondingstandard deviation 6 :

6 = -rC(om-ai)23~ (where i = 1,2,...... 10)
Co,
the experimental characteristic value om(l -26); if this value exceeds
the guaranteed elastic limit,the test should be considered satisfactory
and the consignment be accepted as conforming to the manufacturer's
guarantee.
Otherwise the consignment must either be rejected or it should be used
with a reference value of o,(l-26) for the elastic limit,provided that the
cross-sectional areas and diameters of the reinforcement are amended
accordingly on the working drawings.

Checking The Bar-bending Properties

Unless otherwise specified,this check should comprise a series of to-and-fro

bend tests performed at a temperature of about 20°C.The bars should
successively be :
(a)bent through an angle of 45"round a forming mandrel whose diameter
is determined by the permissible limitofcurvature ofthe bar,according
to the values stated in Section 9.2.2;
(b)immersed in boiling water for half an hour;
(c) bent back to an angle of 22'30.
To satisfy the requirements,the bars should,at the end of these successive
operations,neither fracture nor exhibit any cracks or similar flaws.
Unless otherwise specified,the number of test specimens should be one
per batch.
If the to-and-frobend test yields a favourable result,the batch should be
accepted as conforming to the manufacturer's guarantee.
If not,then three more specimens will have to be taken and tested.If each
of the three additional to-and-frobend tests is favourable,the batch should
be accepted.Otherwise it should be rejected.

9.2.2 B A R - B E N D I N G
The reinforcing bars should be cut and bent in accordance with the working
drawings.
 135
Method Of Bending The Bars
Bars with diameters up to 4 12mm may be bent by hand.O n the other hand,
bars of more than 4 12mm should always be bent by mechanical means,in a
single operation,with the aid of a power bending machine equipped with a
forining mandrel and approved by the building owner or his representative.
It is strongly advised not to use bar-bendingmachines ofthe type equipped
with three rollers,as these machines have the seriousdrawback ofperforming
the bending operation in several passes.This results in discontinuities in the
curvature and damages the indentations or ribs of deformed bars,with the
attendant risk of initiation of fracture of the bars.

M i n i m u m Diameter Of The Forming Mandrel

Bending the bars to the required shapes should be done with a forming
mandrel having a diameter appropriate to the kind of steel concerned.The
minimum diameters that should be adopted for the mandrel in each case
(inaccordance with Section 7.2.2).
are given,in millimetres,in Tables 9.1-9.3
Table 9.1 For bars with a reference elastic limit of less than 3 O00 bars (mildsteel bars:
u, < 3 O00 bars):

Nominal diameter
oibars 4< 1 2 m m
Minimum diameter of Stirrups and binders 34
forming mandrel Anchorages 54
Bends 84

Table 9.2 For bars with a reference elastic limit of not less than 3 O00 bars and not more
than 5 O00 bars (medium-tensilesteel bars: 3 O00 bars f ue < 5 000 bars):

Nominal diameter
of bars
4< 12mm 4> 12mm

Minimum diameter of Stirrups and binders 54 -

forming mandrel Anchorages 104 104
Bends 154 154

Table 9.3 For bars with a reference elastic limit of more than 5 O00 bars
(high-tensilesteel bars: 6, > 5 O00 bars):

Nominal diameter
of bars 4 f 12mm 4 > 12mm
Minimum diameter of Stirrups and binders - -
forming mandrel Anchorages - -
Bends 204 204
136
Table 9.3does not envisage the use of these steelsfor anchoragesformed by
bending nor for stirrups and binders.For these steels the minimum mandrel
diameter can permissibly be reduced if the supplier of the steel explicitly
guarantees the steel as being suitable for this, but the mandrel diameters
adopted must on no account be less than those in Table 9.2.
These diameters may in each case be rounded off to the actual diameter
of the available forming mandrel, provided that this is larger than the
minimum indicated.

Speed Of Bending
The speed of bending applied to the bars should take account of the nature
of the steel and of the ambient temperature.It should be determined by
means of preliminary tests, especially in the case of medium-tensile and
high-tensilebars.
If the ambient temperature is below + 5"C,additional precautions should
be taken:the speed of bending should then not only be greatly reduced,but
the diameters ofthe forming mandrels should be increased as well (inrelation
to the values stated in the tables in Section 9.2.2).
If the ambient temperature is below -5" C,no bar-bending should be
done.

Prohibition To Straighten Bent Bars

Straightening bent bars is always very risky.Hence no systematic straighten-
ing of bends in bars should be allowed.
If a curve or bend has to undergo a correction in situ,this should be done
by increasing the curvature,never by decreasing it, not even partially.
The fact that a medium-tensileor high-tensilesteel bar satisfiesthe to-and-
fro bend test (Section9.2.1)does not authorise the contractor to perform such
operations on the reinforcing bars actually used in the job.

9.2.3 W E L D I N G OF R E I N F O R C E M E N T

Welding of the reinforcement can be envisaged only for bars having an

elastic limit of less than 5 O00 bars and on condition that welding will not
reduce the strength nor entail a risk of brittle fracture of the bars.

Method Of Welding
The method of welding should be agreed among the building owner,the
contractor and the manufacturer of the steel. The latter should provide all
references and justifications that may be considered necessary.
In any case the method of welding should conform to the requirements
of Section 7.2.4concerning welded splices:
(a)either by means of butt welds produced by flash welding;
 &
137
(b)or by means of butt welds,with formed edges,produced by means of
arc welding;
(c) or by means of lap joints produced by electric arc welding with
longitudinal weld beads.
N o gas welding ofany kind is allowed.

Execution Of Welding
The welding of reinforcement should be carried out,in the workshop or on
site,by expert welders,to the exclusion of all other operatives.

9.2.4 P O S I T I O N I N GTHE R E I N F O R C E M E N T

At the time of being installed in position the reinforcing bars should be clean
and free from loose rust,traces of earth,paint,grease or any other harmful
substance. They should be positioned in accordance with the working
drawings and accurately secured by means of mortar or plastic spacers so
as not to undergo any appreciable displacement before or during the placing
and vibration of the concrete.
Steel spacers of which certain parts might remain visible on removal of
the formwork and thus be exposed to corrosion should not be used.In fact,
the use of such spacers might not only spoil the aesthetic appearance
of the structure,but might also endanger its stability.
The use of mortar spacers is permissible only if their presence will in no
way impair the quality and appearance of the structure.
It is strongly recommended to use spacers made of plastic.

9.3 REQUIREMENTS PERTAINING TO CONCRETE

9.3.1 C O M P O S I T I O NOF T H E C O N C R E T E

Definition
Concrete consists of an intimate mixture of inert materials called ‘aggregates’
(sand,gravel,crushed stone,etc.) with cement and water. Because of the
action of the cement, the mixture thus obtained, called ‘freshly mixed
concrete’,begins to harden after a few hours and gradually acquires its
strength properties.

Cements
ClassificationA n d Quality Of Cement

The cementused is generally ofthe ‘Portlandclass,with or without secondary

constituents.The quality of the cement should be defined by stating the
138
28 day compressive strength obtained with ‘RILEM’standard mortar and
expressed in bars (orin kg/cm2).
In general,the following cements are to be distinguished:
(a)‘Portlandcements without secondary constituents;
(b)‘Portland’cements with secondary constituents such as slag,fly-ash,
pozzolanas,etc.;
(c) special cements such as supersulphated cements and high-alumina
cements.
The qualities of the different types of cement are variable;the average
28 day compressive strengths determined on ‘RILEM’standard mortar
test specimens are of the following orders of magnitude:
(a)rapid-hardeningPortland cement: 400 bars
(b)normal Portland cement with or
without secondary constituents:
high early strength: 325 bars
ordinary: 250 bars
(c) supersulphated cements:
high early strength: 325 bars
rapid-hardening: 400 bars
(d)high-aluminacements: 575 bars

Choice Of Cement
The class and quality of the cement should be chosen with due regard to the
nature of the structure to be built,its structural characteristics,its purpose
and the various requirements that it will have to fulfil more particularly
with reference to climatic and local conditions: warm weather,cold weather,
presence of aggressive water,etc.
For ordinary reinforced concrete or prestressed concrete structures a
high early strength Portland cement of class 325 may be used. However,
for prestressed concrete structures in which the tendons are to be tensioned
while the concrete is still young or for structures necessitating early removal
of formwork the use of rapid-hardeningcement may be considered.O n the
other hand,in the case of structures requiring only relatively low mechanical
strength,‘ordinary’cement of class 250 can suitably be employed.
For structures to be built in aggressive surroundings (inthe presence of
water with a high content of calcium sulphate) a cement containing a suit-
ably high proportion of slag (upwardsof 80%)should preferably be used.
For maritime structures a suitable special cement capable of proper
setting and hardening on exposure to sea-water should preferably be used.
In cases where concrete is used in the form of large masses,cements with
very high early strengths should not be used,nor too rich mixes (withmore
than 350 kg of cement per m3), as these could cause considerable evolution
of heat of hydration.
High-aluminacements may be used for refractory concretes.
 139
For structures whose exposed concrete surfaces are required to have an
architectural or decorative character it is advisable to use white Portland
cement (normalor rapid-hardening).

Storage Of Cement

Cement may be stored either in bags or drums or,alternatively,in silos or

bins (storagein bulk). It should be kept protected from the weather and
more particularly from moisture. Storage in bags or drums should be sys-
tematically organised, so as to make sure that the cement in some of the
bags or drums will not be kept in storage for an excessively long time before
use and thus undergo too much ageing.

Aggregates

Granulometric Classijìcation

The granulometric class of an aggregate (sand,gravel, crushed stone) is

defined by two dimensions do and d, corresponding respectively to the
smallest and the largest of the particles of which that aggregate consists.
An aggregate is,by definition,of the class do/d, if, ford, > 2do,the following
values are obtained:
(a) a residue of less than 10%on the sieve with apertures dM ;
(b)less than 10% passing the sieve with apertures do;
(c) less than 3 %passing the sieve with apertures $do.
Aggregates can be subdivided into the categories shown in Table 9.4
and Figures 9.1 and 9.2.

Table 9.4

Classijìcation Sieve mesh Diameter of apertures

of aggregates apertures (in mm) of round-hole screen (in mm)

Fines (orpowders or fillers) <0.08

Sands : fine 0.08-0.315
medium 0.315-1.25
coarse 1.25-5

Gravels: fine 6.3-10

medium 10-16
coarse 16-25

Crushed stone:fine 25-40

medium 40-63
coarse 63-100
140

Figure 9.1 Various sizes of sand: (left) fine sand (d,,<0-5mm) (centre) mediuni sand
(0.5<d,,<1-6mm)(right)coarse sand (I6<d,<5mm)

Figure 9.3. Various sizes of gravel (lefr) fine gravel (6.3< d, < d, < IO mm) (centre)
medium gravel (IO < d, < i6 mm) (right) coarse gravel (I6 < d, < I5 mm)
 141
Grading

The grading of the aggregate is defined by its grading curve as determined

by sieve analysis.This curve should lie within a predetermined zone which
defines the permissible grading of the aggregate.
In drawing the grading curve the percentage of aggregate passing each of
the sieves (or round-hole screens) is plotted, as the ordinate,against the
corresponding sieve mesh aperture (ordiameter of aperture in the case of a
round-holescreen), as the abscissa.
It is not usual to adopt a linear scale for the abscissae;instead,the values of
v d A fmay suitably be adopted as the scale.
The apertures of the sieves or screens to be used for ordinary particle size
analyses may be as follows:
(a) Sieves (forsand):
meshes (inmm):0.08, 0.16,0.315,0.625,1.25, 2.50,5 mm.
(b)Round-holescreens (forgravel and pebbles):
holes (dia.in mm):6.3,8, 10,12.5,16,20,25,31.5,40,50,62.5,80,
100mm.
It is assumed that the result obtained by sieving on a 5 mm sieve is approxi-
mately the same as that obtained on a 6.3mm round-holescreen.
The shape of the grading curve gives information on the granulometric
composition of an aggregate of the class d,/d,, which may have a higher or
lower content oflarge or small particles (cf.Figure 9.3).
For making high-strengthconcretesit is recommended-except in particu-
lar technical or economic circumstances-to use sands having a grading
curve situated within the zone indicated by hatching in the graph.

M a x i m u m Dimension

The maximum dimension d, of the aggregates employed should,on the one

hand, be less than the horizontal clear distance between two reinforcing
bars (orbetween a bar and the formwork)and,on the other,be less than one-
quarter (25%)of the thickness of the member to be concreted.
It is always advantageous.to use aggregates with a large maximum
dimension, having due regard to the dimensions of the member to be
concreted,the shape of the formwork,the congestion of the reinforcement,
etc. For a given cement content the concrete will be denser and stronger
according as the aggregate dimension is larger. But this must not be
overdone,or else a concrete will be obtained which is difficult to place and
work and which will not ensure satisfactory filling of the formwork.

Cleanliness

The aggregates employed should be clean and free from all foreign matter
such as cinders,coal,gypsum,wood waste,dead leaves,organic matter,etc.
142
-
Ln
m
Lo ct
N ct
O
N
u3 N
r U
x
r

r
U
3
0 U
a3
m m
w m
<
n
u3
m
m
m
n N
r m
m o)
6 N
m
N
U W
3 N
N F
I
3 N
r 3
6 N
ul
u3
O
 143

Clear water

Flocculate

Sand

Figure 9.4

Figure 9.5. Measuring the cleanliness of sand (‘sand

equivalent’ test). In the right-hand cylinder,which has
just been agitated,theflocculatehas not yet formed
144
Gravel and pebbles should be free from clayey matter or adhering soil.
Sand should be free from fine clayey matter and should satisfy the so-called
‘sandequivalent’test. The desirable optimum value of the sand equivalent
(SE)if visual measurement is employed is:75<SE< 85.
Measuring the cleanliness of the sand should be done by means of the
so-called‘sandequivalent’(SE)test (see Figures 9.4and 9.5).
A certain quantity of sand is put into a glass cylinder with a washing
solution based on calcium chloride, glycerine and formaldehyde and
agitated.The liquid is then allowed to stand for 20 minutes. Next, with the
aid of a graduated scale,the height to the top surface of the flocculate (turbid
water) and the height to the top of the sand deposit are measured.If these
heights are h, and h, respectively,the sand equivalent is:
h
SE = 1002
hl
The height h, may be measured either ‘byeye’ directly with the aid of a
rule (graduated scale) or ‘by piston’,i.e.,on the rod of a weighted piston
which is lowered on to the sand at the bottom of the cylinder and whereby
generally a slightly smaller value for h, is obtained than ‘byeye’.
If the sand contains fine clayey matter, the values in Table 9.5can be
adopted for guidance.
Table 9.5

SE ‘by eye’ SE ‘by piston’ Nature and quality of the sand

SE<65 SE < 60 Clayey sand: risk of shrinkage or swelling;

not suitable for high-strengthconcretes.
65GSE <75 60<SE<70 Slightly clayey sand, sufficiently clean for
making concretes of ordinary quality,unless
shrinkage is something particularly to be
avoided.
75<SE < 8 5 70s SE <80 Clean sand with low percentage of fine
clayey matter,quite suitablefor high-strength
concretes (optimum value of SE = 75 by
piston or 80 by eye).
SE 2 85 SE280 Very clean sand: because of the almost total
absence of fine clayey particles the concrete
may be deficient in plasticity;this should be
compensated by increasing the water con-
tent.

Nature And Shape

The aggregates used should either be natural sands,gravels and pebbles,or

crushed materials produced from appropriate types of rock (seeFigure 9.6).
In particular,rock which is too friable or too soft (suchas certain limestones)
or which decomposes in contact with the air (such as certain porphyries)
or by hydration (such as certain shales) must not be employed. O n the
other hand certain blast furnaceslags can,when crushed,permissibly be used.
 145

Figure 9.6.Aggregates of various mineralogical kinds (left) calcareous aggregates (right)

siliceous aggregates

Figure 9.7. Concrete test specimens after failure

(bottom: prism fractures in tension; top: cylinder
fractured in compression). Concrete with cal-
careous aggregates of low strength but good
adhesion. (Failure has occurred in consequence
of failure of the aggregate particles themselves)
146
With regard to the geometrical shape of the particles,gravel having flat
particles (‘plates’)
or excessively elongated ones (‘needles’)should not be used.
In choosing the kind of aggregates,it should be endeavoured,as far as
possible, to obtain material of sufficient hardness (which determines the
hardness of the concrete) and also sufficient adhesiveness of the cement
paste (which is likewise essential to the development of strength). In fact,
failureof concrete is generally due either to failure(fracture)ofthe aggregates
or to failure of the adhesion of the cement paste to the aggregate particles
(seeFigures 9.7and 9.8).
The geometrical shape of the aggregate particles is also very important.
They may be ‘crushedstone aggregates’(obtainedby the crushing and screen-
ing of suitable rocks)or ‘roundedaggregates’(obtainedmerely by the screen-
ing of natural alluvial materials). Crushed stone aggregates sometimes have ’

Figure 9.8. Concrete test specimens after failure.

Concrete with siliceous aggregates of high strength
but poor adhesion. (Failure has occurred in
consequence of detachment of the aggregate par-
ticles from the cement paste)

particles of unsuitable shape (flat‘plates’or elongated ‘needles’)and should,

in that case,not be used,since the concrete made with them has poor worka-
bility,is deficient in density,and often presents a bad surface on removal of
the formwork.
The shape coefficient(orvolumetric coefficient) of an aggregate is defined
as the ratio of the sum of the volumes V of the particles and the sum of the
 147

Figure 9.9. Graoel 16/25 m m of differentshapes. (left) grauel with high proportion offlat
and elongated particles: shape coefficient = 0-12 (centre) ordinary grawl: shape coef
ficieiit = 0.19 (right) rounded gravel: shape coefficient = 0-38

volumes of the spheres circumscribed around each particle,the diameter of

each sphere being equal to the maximum dimension d of the corresponding
particle,i.e.:
CV

A normal value of this coefficient is 0.20;it is recommended never to go

below 0.15(seeFigure 9.9).

Mixing Water

The water used for mixing the concrete should be clean and not contain
more than 5 g ofmatter in suspension per litre (mud,silt,etc.)nor more than
35 g of soluble matter and salts per litre, provided that these dissolved salts
entail no risk of harming the durability of the concrete (acids,sulphates,
corrosive salts,organic matter).
It is not permissible to use sea water,unless there are special reasons to
justify it and the building owner’sconsent is obtained.
Any water of doubtful quality should be analysed.
The use ofsea water for the mixing ofconcrete generally causes a significant
lowering of the subsequent strength of the concrete;besides, it promotes
148
corrosion of the reinforcement and is liable to be particularly dangerous in
heavily reinforced or in prestressed concrete. In any case it must be taken
into consideration that the amount of calcium chloride introduced with the
sea water is equivalent to about 2%(byweight) of the cement.

Additives
Additives are special substances which are added in small quantities to
mortars and concretes at the beginning of mixing and whose purpose is to
modify some of the properties thereof. Special reasons must be adduced
to justify the use of additives, and the building owner’s consent must be
obtained.
The method of using the additive and the amount to be added should be
specified by the building owner and strictly conformed to. Special precau-
tions must be taken to ensure that the additive is uniformly distributed
throughout the mix.
Additives are available in the form of powders or liquids which are added
at the start of mixing, in order to ensure uniform distribution of these
substances.
In the main,the following kinds of additives are to be distinguished:
Plasticisers and wetting agents: By using these additives it is possible to
reduce the quantity of water and yet obtain the same plasticity of the mix or,
alternatively,to increase the plasticity (and thus achieve better workability
of the concrete)without increasing the water content.
Air-entraining agents: These additives, which are sometimes mixed with
the cement in advance (air-entrainingcement) cause large numbers of very
small air bubbles to be incorporated into the concrete;these bubbles increase
the plasticity of the freshly mixed concrete and improve the frost resistance
of the hardened concrete.
Setting retarders:These may be necessary in cases where the setting of the
concrete has to be retarded (e.g., at construction joints where cbncreting is
temporarily stopped, exposing the aggregate by scrubbing the concrete
surface,concreting in very warm weather).
Setting accelerators: Additives of this type may have to be used in cases
where the formwork must be removed soon or where concreting is done in
cold weather. A commonly used accelerator is calcium chloride,but as it
entails a serious risk of corrosion of the reinforcement, its use is strictly
limited.
Some additives may:
(a)either entail a danger of corroding the reinforcement or other metal
objects embedded in the concrete (heatingcoils,plumbing,etc.);
(b)or adversely affect other properties (hardening accelerators cause
increased shrinkage,antifreeze agents cause reduction in strength,etc.).
These dangers should be duly taken into consideration when choosing
an additive.
Besides,it must always be remembered that an additive must be added
 149
only in small quantity and that it must be uniformly distributed. Most
adverse effects attributable to the use of additives are in fact due to excessive
amounts being employed (which is something that is often difficult to
supervise on the site) or to lack ofperfecthomogeneity in mixing the additive
with the concrete.
Because of the serious risk of corrosion of the reinforcement,calcium
chloride and additives containing chlorides are not allowed to be used for
making mortars and concretes used for the following structures or structural
components:
(a)prestressed concrete structures;
(b)tanks and containers;
(c) floors in which the main or the secondary reinforcement of the joists
consists of steel strips or sheets;
(d)floors in which heating coils are incorporated;
(e) concrete members treated by steam curing.
For other reinforced concrete structures the use of calcium chloride or
additives containingchloridesis permissible only on the followingconditions:
(1) Calcium chloride and additives containing chlorides must not be used
with lime-based slag cements,nor with high-alumina cements, nor with
supersulphated cements.
For cements other than Portland cements it is necessary first to carry
out a setting test and,if necessary on account of the conditions of use ofthe
concrete,a short-termcompressive test,because the results vary according
to the nature and proportions of the constituents and also according to the
proportion of chlorides that may have been incorporated with the cement
at manufacture.
(2)The maximum permissible amounts of calcium chloride-ratios of
the weight of commercial calcium chloride (flakes containing 75-77% of
CaCl,) to the weight of cement used -are as follows:
(a)2% for mortars,for plain (unreinforced)concretes,and for reinforced
concretes with at least 4cm of cover to the reinforcement;
(b)1 %for reinforced concrete with at least 2 cm of cover to the reinforce-
ment.
The maximum percentages indicated relate to the total quantity of
chlorides,i.e.,they include the quantity of chlorides (if any) that is already
present in the cement or in some other constituent of the concrete or in any
other additive used in conjunction with calcium chloride.
So if another additive is used together with calcium chloride,it is necessary
to satisfy oneself that this other additive does not contain chlorides or,if it
does,to take this into account and check that the total quantity of chlorides
will not -having regard to other possible sources ofintroduction ofchlorides
into the mix -exceed the permitted maximum percentage.
Calcium chloride and additives containing chlorides should be protected
from moisture during storage.These substances should never be incorporated
.
directly into mortars and concretes.Whichever method of introducing them
1 50
into the mix is employed (firstdissolving the chloride or additive in water
and then adding this solution to the mixing water, or directly putting the
chloride or additive into the mixing water), it is necessary always to satisfy
oneself that the substance is completely dissolved and to take the necessary
steps to ensure that the concentration is as uniform as possible.In particular,
a stirrer should be used for homogenising the solution at all stages,both
when a preliminary solution is prepared and when the substances are mixed
with the water for making the concrete.
The following method is recommended for using calcium chloride as the
sole additive containing chloride:
First,a solution ofcalcium chloridein water should be prepared as follows:
(a)obtain a tank of more than 100litres capacity;
(b)put in 80 litres of water and mark the level of this water with an
indelible mark;
(c) then slowly add the contents of a bag of calcium chloride while vigor-
ously and continuously stirring the water to ensure that the flakes will
dissolve completely:in this way 100 litres of a solution is obtained
whose strength corresponds to 5 kg of flakes per litre;
(d)now put into the concrete mixer,which should be running and already
have had at least half the required quantity of mixing water put into it,
as many times 2 litres of solution as there are kilogrammes of chloride
to be introduced into the concrete;for example:
1 litre of solution per 50kg bag of cement if the specified content of
chloride is 1 % of the weight of the cement (i.e.,0.5kg of chloride
per 50kg of cement);
(e) add the rest of the mixing water;
(f) then run the mixer for a sufficient length of time to obtain a good mix
(atleast 13 minutes or more,depending on the efficiency of the mixer
employed).
If calcium chloride or an additive containing chlorides is used in conjunc-
tion with another additive,it is necessary to satisfy oneself that the two
substances are not incompatible,i.e.,that the two additives will not,on being
mixed together,give rise to chemical reactions liable to cancel the desired
effect;in particular,they must not cause any insoluble salt to be precipitated.

Batching The Cement

The concrete mix should normally contain between 250 and 450kg of
cement per cubic metre of concrete as finally present in the structure.
Departure from this requirement will call for special justification and the
building owner’sapproval.For ordinary reinforced concrete structures the
concrete should contain 350 kg of cement per cubic metre. For reinforced
concrete structures which are required to have special properties of water-
 151
tightness and density,and for prestressed concrete structures,the concrete
should contain 450kg of cement per cubic metre.
The mix proportions of a concrete can be determined by a number of
theoretical or experimental methods which yield mix formulae from which,
in each case,the quantities of each of the constituents (includingthe water)
making up the composition of one cubic metre of concrete as finally present
in the structure can be calculated.
The concrete is said to be batched (orproportioned) ‘byvolume’if those
quantities are measured by volume, and batched ‘by weight’ if they are
measured by weight,the latter being the preferable method.
The cement content most commonly adopted for reinforced concrete
structures exposed to the weather is 350 kg/m3.However,this can permissibly
be varied on the basis of the following considerations:
(a) the strength of a concrete will be higher according as the cement
content is higher;
(b)increasing the cement content will increase the risk of shrinkage and
cracking of the concrete and will cause more heat of hydration to be
evolved during setting;
(c) for equal strength of the concrete,the cement content can be reduced
if the dimension d of the aggregates is increased;this reduction may
be applied proportionally to y d .
As an empirical rule,the cement content of reinforced concrete should not
be less than (550/7d) (inkg/m3ofconcrete as finallypresent in the structure),
i.e.:
350 kg/m3for d = 10mm
315 kg/m3for d = 16mm
290 kg/m3for d = 25 mm
250 kg/m3for d = 50mm
220 kg/m3for d = 100mm

Batching The Aggregates

The quantities of aggregates in a concrete mix are defined in terms of the

proportions by volume or,preferably,by weight of the various aggregates
making up the composition of one cubic metre of concrete as finally present
in the structure. These proportions should be specially determined by
experimental methods (trial and error), unless the contractor has at his
disposal, for similar materials used under identical conditions, a set of
reliable practical rules confirmed by long experience with those materials.
The relative proportions of sand and gravel should be such that the
concrete has satisfactory homogeneity without any risk of segregation.
In most cases the choice and proportions of the aggregates will be deter-
mined by means of investigations in the laboratory,with due regard to the
nature of the structure,its structural features, the required strengths,the
nature,shape and grading of the available aggregates,etc.For guidance the
152
following method,called the ‘GIScoefficient method’may be employed in
normally encountered cases.

G/SCoefficient Method (seeFigure 9.10)

This method ofconcrete mix design is purely empiricaland is based on a large
number of tests performed on a wide variety of concretes made in the
laboratory.
In the most frequent case of a concrete mix comprising coarse aggregate
(gravel,weight G)and fine aggregate (sand,weight S) the proportions are
determined by the ratio GIS,once the cement content C has been decided on
the basis of the information given in the preceding section.
In general,the value normally adopted for G/Sis 2.0.It may, however,
be varied between 1.5 and 2.4,having regard to the following considerations:
(a) The higher G/S is, the higher will be the mechanical strength of the
concrete;against this,however,the concrete is more liable to segregate
and present difficulties in placing because of deficient workability or
because of considerable wall effect.
(b)For a very plastic concrete mix, with high content of mortar and good
workability,giving surfacesthat present a good appearanceon removal
of the formwork,but not producing a concrete of exceptional strength,
the following values may be adopted:
1.5< G/S< 1.7
(c) For normal concrete,as used in ordinary reinforced concrete construc-
tion,with plasticity that may be varied according to the nature of the
structure by varying the amount ofwater in the mix,of fair workability
and giving good strengths,the following values may be adopted:
1.8<G/S<2.0
(d) For high-density concrete, of ‘stiff consistency when freshly mixed,
producing high strengths but liable to segregate and requiring certain
precautions at the time of concreting (compaction by powerful vibra-
tion,in particular), the following values may be adopted:
2.0< G/S< 2.2(and exceptionally:2.4)
(e) In the case of rounded aggregates the above values of G/S are valid,
but if crushed stone aggregates are employed, slightly lower values
should be adopted to GIS:for example,the above values should be
reduced by: 0.1 in the case of rounded sand and crushed coarse
aggregate;0.2in the case of crushed fine aggregate and crushed coarse
aggregate.
When the value of G/Shas been chosen,as also the quantity of cement in
the mix, the respective quantities of gravel and sand can be determined
by considering that the total weight (C+G+S+water) must be equal to
the weight of one cubic metre of concrete as finally present in the structure.
153
154
With the aid of the accompanying graph these quantities can readily be
calculated for the normal case where two aggregates (coarseand fine) are
used in the mix: sand 0/5mm and gravel 5/25mm, both with a specific
gravity o = 2.6.If the aggregates have a specific gravity differing from 2.6,
the weights read from the graph should be corrected by multiplying them
by o/26
The weights indicatedare dry weights,i.e.,for the completely dry materials.
However,if the water contents of the aggregates have been measured (or
estimated), the actual moist weights of the materials to be put into the
concrete mixer can be read directly from the oblique scales.
The total weight G of gravel, thus determined, remains approximately
valid also in a case where several coarse aggregates are employed in the mix.
The respective proportions of each of these (G,,G,, G,) must be such that
G,+ G,+ G, = G.The weights G,,G,,G, should be so distributed that the
grading curve of the mixture G,+ G,+ G, is a smoothly continuous curve.
To proceed from batching by weight to batching by volume,the weights
of the aggregates employed should be divided by the respective bulk densities
of these aggregates.
When the theoretical ‘mix design’(determination of the proportions of
constituents in the mix) has thus been carried out,the following should be
determined by means of a preliminary test:
(a) the quantity of water needed for obtaining the desired plasticity;
(b) the density of the freshly placed concrete, whereby the necessary
corrections can be applied to the theoretical weights of the aggregates,
so that the mix formula,corrected in this way,does indeed correspond
to a cubic metre of concrete as finally present in the structure;the
weight of this cubic metre must be equal to the sum of the weights of
the constituents (includingthe water) as given by the mix formula;
(c) the strength of the concrete made in this way.

Quantity Of Water A n d Consistency Of The M i x

The quantity of water (total water) is determined for aggregates assumed to

be dry and for a cubic metre of concrete as finally present in the structure.
If the aggregates employed contain a certain amount of water, their water
content (additional water) should be estimated and deducted from the
required total water: thus the quantity of water to be added during mixing
is found.
The mix should contain enough water to give the plasticity compatible
with good workability, but not too much water, as the strength of the
concrete becomes lower with increasing water content of the mix. O n no
account must water be added to a concrete mix that is considered to be too
dry after leaving the mixer.
The desired consistency can be defined on the basis of the slump test
(see Figure 9.11).
The water content of a concrete can,and must,be defined only in terms of
 155
the desired consistency ofthe mix.The theoretically correct quantity of water
can be calculated in advance,but there is no justification for insisting on
using that quantity of water if infuct it results in a concrete of unsatisfactory
consistency (eithertoo plastic or too stiff).So the actual water content must
always be determined by means of a preliminary test. Nevertheless it is

Figure 9.11.Equipment for determining the consistency by means of the slump test: plate,
conical,mould, rod, scoop, trowel and measuring bridge

necessary to be able to make an approximate estimate of the required

quantity of water in advance.This can,in general,be done on the basis of
the following principles :
For mixes containing a quantity of cement C = 300 to 400kg/m3,a total
water content (assumingdry aggregates) may be adopted so as to obtain a
waterfcementratio ( W / C within
) the following limits:
0.4< W / C<0.6,with an average value W / C = 0.5
A value of W/C<O.5is adopted if it is desired to have a stiffor very stif
mix, or if the sand is somewhat deficient in fine particles, or if the gravel
consists for the most part of large particles and is very porous,or for values
of G/S>2,or if an additive is used (plasticiser or wetting agent). In other
cases a value of W / C> 0.5 should be adopted.
When the quantity of water has thus been approximately estimated, a
small batch of concrete is prepared for preliminary testing,and the water
needed for obtaining the desired consistency in actual practice is added to
the mix.
156
The consistency can be measured by various methods, the simplest of
which is the slump test (see Figure 9.12).
Concrete is placed, in three successive layers, in a sheet-metal mould
shaped like a truncated cone.Each layer is punned 25 times with the aid of a
steel rod, 16mm in diameter and 600mm in length,provided with hemi-
spherically rounded ends.
After the concrete has been struck offflush with the top of the mould,
the latter should immediately be removed by slowly raising it vertically,

H Figure 9.12

with care and without jerking it. Then the slump should be measured at the
highest point of the slumped concrete.
In general,the values in Table 9.6 can be taken as applicable to concretes
ordinarily employed:
Table 9.6

Slump as measured in the slump test Consistency of the concrete mix

0-2 c m stiff
3-7 cm plastic
8-15 c m very plastic

A consistencycorrespondingto a slump of 5-7 cm is generally very suitable

for normally vibrated concretes.
When the total quantity of water has thus been established,it is necessary,
during execution of the concreting,to take account of the additional water
contained in the aggregates and deduct it from the total water in order to
find the quantity of water to be put into the mixer.This water in the aggre-
gates may be quite a significant quantity in certain cases. For example,in
the case of silico-calcareousaggregates, the values in Table 9.7 can be
adopted for guidance.
 157

Figure 9.13. Consistency measurement for a very still concrete (practically zero slump)
158

Figure 9.15.Consistency measurement for a very plastic concrete (12 c m slump)

Figure 9.16.Judging the consistency by the feel. With plastic consistency it should still he
possible tofòrm a ball in the hand (7 c m slump)
 159

Figure 9.17.Judging rhe consistericy by the feel. With w r y plastic corisisteiicy it is not
possible to /orin ci hall; ihe corici'etc runs between the$ngers (17ctn sliirnp)

The most effective method to ensure a constant water content of the mix
consists in checking the consistency at very frequent intervals by means of
the slump test.
Table 9.7
Apparent Additional water (litreslm')
moisture
content Sand Gravel Gravel Gravel
015 5/16 16/25 5/25

Appearance:
Dry 10-20 negligible negligible
Wet 30-60 20-30 5-15 10-20
Very wet 80-120 40-60 20-40 30-50
Saturated 120-150 70-90 50-60 60-75

9.3.2 TESTINGTHE STRENGTH OF C O N C R E T E

Nature Of The Tests

The object of the tests is to estimate and check the compressive strength
and the tensile strength of concrete. Such tests may serve two different
purposes:
160
Control

The object of ‘controltests’ is to control,i.e.,keep a check on,the intrinsic

strength of the concrete as manufactured,independently of the subsequent
conditions of handling,placing,vibration,curing and storage.
Control tests are performed on specimens taken at the mixer and stored
under standardised conditions.These tests are more particularly used for
checking that the intrinsicproperties ofthe concrete conform to the specifica-
tions and to detect any defects in the constituents (cement,aggregates,
additives) or even possible errors in batching.

informat ion

The object of ‘information tests’ is to estimate as closely as possible the

strength of the concrete in the actual structure.
Information tests are performed on specimens taken at the site during the
placing of the concrete in the formwork and stored under conditions which
are as similar as possible to those of the structure itself. These tests are used
more particularly for deciding whether it is suitable or safe to remove
formwork or centring,to handle a concrete member,or to apply loading or
prestress to a structure.
All operations necessitated by the tests should be performed by suitably
skilled personnel.

Sampling

Sampling should be carried out in such a way that samples which are truly
representativeof the concrete are obtained.
For control tests, sampling should be done at the mixer, the concrete
being taken as it leaves the mixer half-waythrough the discharge of a batch.
For information tests, sampling should be done at the point of placing
the concrete,at the moment when it is being discharged into the formwork,
and should comprise at least three samples taken at different points far
from :he edges of the mass,where partial segregation is liable to occur.
The amount ofconcrete taken in the samples should correspond to at least
one and a half times the volume of the test specimens to be made. The
samples may be remixed,if necessary,on a non-absorbentsurface so as to
ensure the homogeneity of the concrete.They should be carefully protected
from sun and rain;the test specimens should be made as soon as possible
after the samples have been taken.

Moulds For Specimens

Cylindrical moulds for making test specimens should be of non-absorbent

material and be sufficiently rigid to undergo no deformation during the
 161
making of the specimens.The bottom and the upper edge of such a mould
must not deviate by more than 0.05mm from a contact plane. The angle
formed by the bottom and sides of the mould must not deviate by more
than 0.5"from a right angle.
The joints should be tight and be lightly coated with mineral oil (or any
other appropriate substance that will not react with cement) in order to
prevent bond between the concrete and the mould.
The use of other types and dimensions of test specimens is allowed for
measuring the compressive strength (crushing test performed on cube or
prism) and tensile strength (splittingtest on cube or flexural test on prism).
If there are compelling reasons why it is not possible to use cylindrical test
specimens,the following alternatives are recommended:
(a) For the crushing test: prisms having a square cross-sectionalshape
with a side length equal to at least four times the maximum dimension
of the aggregate and having a length equal to three times that side
length,or cubes with an edge length equal to at least five times the
maximum dimension of the aggregate.
(b) For the splitting test: cubes with an edge length equal to at least five
times the maximum dimension of the aggregate.
(c) For the flexural test: prisms having a square cross-sectional shape
with a side length equal to at least three times the maximum dimension
of the aggregate and having a length equal to five times that side
length.
For practical purposes, and subject to the building owner's approval,
the dimensions shown in Table 9.8can be recommended:

Table 9.8
Maximum dimension of dA,$35mm 35mm<dMg45mm d,,>45mm
the aggregate

Crushing Cylinder diameter = 15 cm diameter = 20 cm diameter = 25 cm

test height = 30cm height = 40cm height = 50cm
Prism side = 15cm side = 20cm side = 25cm
height = 45 cm height = 60cm height = 75cm
Cube edge = 15cm edge = 20cm edge = 30cm
Splitting Cylinder diameter = 15 cm diameter = 20 cm diameter = 25 cm
test height = 30cm height = 40cm height = 50cm
Cube edge = 15cm edge = 20cm edge = 30cm
Flexural Prism side = 10cm side = 15cm side = 20cm
test height = 50cm height = 75cm height = 100cm

The given coefficients (correctionfactors) are applicable to the results of

these various types of test so as to obtain values directly comparable with
one another:see Table 3.1.
With regard to tolerances the internal faces of the cube moulds must not
deviate by more than 0.05mm from a contact plane. Besides, as in the
162
case of the cylindrical moulds, the angle formed by the bottom and the
sides must not deviate by more than 0.5"from a right angle.

Number Of Test Specimens

Each set of control test specimens should comprise six cylindrical specimens,
three of which should be subjected to the crushing test (formeasuring the
compressivestrength)and three to the splittingtest (formeasuring the tensile
strength).The number ofcontroltest specimensto be made and the frequency
of taking the samples of concrete for making them should be specified by
the building owner,having regard to the volume of concrete in the structure,
the rate of concreting,the properties that the concrete is required to have,
and the difficulties encountered in obtaining them.
In most cases a minimum of six cylindrical specimens (three for crushing
and three for splitting)should be required for each class of concrete per day
and per 200 m3.

Manufacture And Storage Of The Specimens (See Figure 9.18)

In the case of control tests the specimens made from concrete taken at the
mixer should be kept for 24 hours in their moulds,provided with covers and
stored in an enclosed space at a temperature of 20"f4°C.After removal
from the moulds the specimens should be stored in water at a temperature
of 2Oo&2"C (or,failing this, in air at a temperature of 2O0&2"C and a
relative humidity of at least 95%).
In the case of information tests the concrete should be taken from the
point where it is actually being used,and the specimens made from it should
be stored under conditions as closely similar as possible to those of the
actual structure.
In both cases the compaction of the concrete in the moulds in one or more
layers should be effected by punning, ramming or vibration,depending on
the method of compaction used for the structure itself.If the concrete has to
be vibrated,the specimens may be compacted on a vibrating table or,failing
this,a pneumatic or electric internal vibrator may be used. The diameter
of the internal vibrator must not exceed one-fifthof the transversedimension
of the specimens;if it does,the vibrator should be applied to the outside
of the wall of the mould.During vibration,more concrete should be added
to the mould,so as to keep it well filled all the time (see Figure 9.19).
As an example,the following procedure may be adopted for making the
specimens:
(a)Cylinders;15 cm diameter,in two layers.
(b)Cubes:edge length = 15 cm,in one layer;
edge length> 20 cm,in two layers.
(c) Prisms:side length (ofcross-section)= 10cm,in one layer;
side length > 15 cm,in two layers.
-
.

Figure 9.18.Moulds for niaking test specimens

Figure 9.19.Making a cylindrical test specimen of 15 c m diameter. T h e concrete is vibrated

by means o r a small internal vibrator of 25 m m diameter and more concrete is put into the
mould as the uibiator is slowly withdrawn
164
If compaction of the concrete in the structure is effected by vibration,
the test specimens should also be compacted by vibration-on a vibrating
table or,failing this,with the aid of an internal vibrator of 25 mm diameter,
30-40cm long, and having a frequency of more than 10000 cycles/min.
The vibration time for each 1ayer.should be 18 s in the case of stiff concretes
(0-2cm slump) and 12s in the case of plastic concretes (3-7cm slump).
For very plastic concretes (slumpexceeding 8 cm)no vibration is, in general,
used in concreting the structure,and the concrete in the moulds for making
the specimens should be punned with a steel rod,60cm long and of 16mm
diameter,at a rate of 12 strokes per layer and per 100cmzof surface area.
Immediately after casting,the test specimens should be provided with
distinctive markings in chalk on the moulds. Marking by scratching the sur-
face of the concrete is not allowed.Merely a serial number will be sufficient,
provided that all the other necessary information is properly recorded :
date,mix proportions, structure or part of structure in which the concrete
was used,etc.

Test Procedure
The tests should be performed on a duly checked and calibrated testing
machine, continuously and without shock.The error in the loads within
the loading range must not exceed _+ 1 %.
The testing machine should be equipped with a ball-mounted platen
whose centre should approximately coincide with the centre of the upper
loaded face of the specimen.The other platen on which the specimen rests
should consist of a very rigid solid block.
If the accuracy of the moulds is not satisfactory,the dimensions of all
specimens made in them should be measured with an axcuracy of within
1 mm.

Age Of T h e Concrete At Testing

For control tests the age of the concrete at the time of testing should be
28 days.
For information tests the age should be the age at which it is desired
to know the strength attained by the concrete in the actual structure,for
example,with a view to removing the formwork or centring,loading or pre-
stressing the structure.For such purposes it will be necessary to make an
appropriate number of test specimens.

Crushing Test (see Figures 9.20and 9.21)

(Measuringthe compressivestrength of the concrete)
If the loading faces of the test specimens deviate by more than 0.05mm
from a contact plane,they should be capped or trued in such a manner as
 165
to present a plane surface that is perpendicular to the axis with an accuracy
of within 0.5".The capping layers,applied as thinly as possible,must not flow
nor crack during the testing of the specimen.
The specimen should be well centred between the platens of the press,
and loading should be applied at a constant rate so that the stress increase

Figure 9.20. Crushing test on a cylindrical specimen. for

measuring the compressive strength

is 6 f 4kg/cm2per second.During the first half of the loading operation a

higher loading rate can permissibly be applied,however. Loading should
continue until failure of the specimen occurs.The maximum load reached
should be recorded.
Cylinders and prisms should be tested in the vertical position between
the platens of the press. The upper face of the specimen should be capped
or trued. Cubes should be tested preferably by placing the lateral faces
(whichwere in contact with the mould) in contact with the platens;if these
faces are sufficientlyflat (<0.05mm),no capping or trueing is necessary.
Capping may be done at the time of casting the specimens,within a period
of 2-6 h,by means of a thin layer of 'neat'Portland cement paste.The cement
used for the purpose should be mixed so as to obtain a thick paste 1-4h before
it is required for use and subsequently remixed.Application of the capping
layer should be done with the aid of a glass plate at least 6mm thick or a
166
trued metal plate at least 1 2 m m thick. The dimensions of the plate used
should exceed those of the mould by more than 25 mm.
The cement paste can be prevented from sticking to the glass or metal
plate by giving the latter a thin coating of oil or grease.The cement paste
should be worked until the underside of the plate is everywhere in close

Figure 9.21. Crushing test on a cvlindrical specimen, after

removal ojthe cracked and spalled concrete

contact with the top edge of the mould.The plate should be left in position
until the specimen is demoulded.
Alternatively,if the test specimens are capped at the time of testing,a
mixture ofmelted sulphur,lamp-blackand finesand may be used.Themixture
may, for example, be composed as follows: sulphur (50kg), lamp-black
(1.6kg)and fine sand (30kg).

Splitting Test (seeFigure 9.22)

(Measuringthe tensile strength of the concrete)
The specimen to be tested should be placed between the platens of the press
so that the force is applied to two diametrically opposite generating lines
(in the case of a cylindrical specimen) or to the upper and lower face (in
the case of a cube). The load may be applied quickly up to 50%of the antici-
 167
pated failure load;thereafter it should be increased more slowly,at such a
rate that the increase in stress at the extreme fibre never exceeds 0.5bar/s.
Approximately 15 mm wide and 5 mm thick strips of plywood,cardboard
or some similar material should be inserted between the platens of the press
and the generating lines to which the loading is applied in the case of a
cylindricalspecimen.
Similar strips, whose length should be equal to the edge length of the
specimen,should be inserted if the splitting test is performed on cubes.

Flexural Test
(Measuringthe tensile strength of the concrete)
The prismatic specimens are tested with a span equal to three timesthe side
length of the square cross-sectionof the specimens.They should preferably

F i p i e 9.32.Splitting test on n cvlinilrical specimen for merisuring the tensile .sti-riiyth

be placed on their sides in relation to the position they had during casting
the mould.The loading should be applied according to the same procedure
as that for the splitting test.

9.3.3 MAKING THE CONCRETE

Putting The Materials Into The Mixer

The constituent materials of the concrete should be put into the mixer in the
following order:gravel,cement,sand.Water must not be added until these
three materials have first been mixed in the dry condition for a time.
168
In certain cases it is to be recommended that first a portion of the coarse
aggregates and water be put into the mixer and to run the latter briefly so as
to wet the walls of the mixing drum or pan and thus prevent possible sticking
of the mortar to them.

Mixing Procedure

Mixing should preferably be done in a vertical-axismixer. For a medium-

size mixer running at 15-20rev/min the minimum mixing time can be taken
as 2 min.

9.3.4 H A N D L I N G A N D P L A C I N G THE CONCRETE

Checks Prior T o Concreting
Before concreting is done,it is necessary to check:
(a)that the formwork has been properly installed;
(b)that the reinforcement has been fixed in position in accordance with
the drawings (moreparticularly with regard to the minimum concrete
cover to the bars) and that there is no risk of its being displaced during
concreting or vibration of the concrete.

Handling The Concrete

The concrete should be handled and transported in such a way as not to

entail any risk of segregation nor of setting before it has been placed in the
formwork.Every precaution should be taken to ensure that,during handling
and transporting of the concrete,no excessive loss of ingredientsor intrusion
of foreign matter will occur.
The supervisorin charge ofsite work should be alive to the risks ofsegrega-
tion associated with some methods of handling the concrete. He should
ascertain and obviate the causes thereof.
If the transporting time exceeds 30min,it is advisable,especially in warm
weather, to check by means of laboratory tests whether this is indeed a
permissible length of time.

Placing The Concrete

Unless special justification to waive this requirement is provided, concrete
should always be placed and compacted by vibration.
Before any concrete is placed,a concreting programme should be estab-
lished in advance and should indicate the mixing and handling methods and
also the procedure and rate of placing of the concrete.Interruptions oí the
concreting should be as few as possible.
 169
Before a structural member is concreted,the supervisor should check the
formwork (dimensions,strength,tightness,cleanliness,wetting, oiling) and
satisfy himself that the specified amount of concrete cover (distancefrom
reinforcement to formwork)is everywhere assured. He should draw up in
advance a concreting scheme,having regard to the dimensions and shapes
of the members to be concreted,the capacity of the mixer,the construction
joints to be provided,the arrangement of the reinforcing steel,etc.
If there is congestion of reinforcement over a considerable height, it is
necessary to provide chutes to carry the concrete to the bottom of the form-
work and thus prevent any ‘cascading’through the reinforcement (withthe
attendant serious risk of segregation). In that case it is desirable that the
designer should,at the time of designing the reinforcement,already have
allowed for the possibility of introducing such chutes.
If the consistency of the concrete is not constant, because of difficulties
in proportioning the water,a look at the concrete in its skip should enable
the supervisor to judge its consistency and stop a batch which happens to be
too dry from being deposited in a heavily reinforced part of the structure
where the wall effect would be particularly pronounced,with the attendant
risk of clogging of the concrete.
In some cases (e.g.,at the bottom of a beam with closely spaced reinforcing
bars) it is preferable to ask the men at the mixer to send along a few batches
of a more plastic consistency.The too dry batches (provided,of course,they
are acceptable anyway) may be kept for concreting the compression flanges
and other parts where there is less congestion of reinforcement.

Vibrating The Concrete

Internal Vibration

The internal vibrators (immersion vibrators) employed should,be able to

penetrate into every part of the formwork,so that, having regard to their
radius of action,they can reach and compact the whole of the concrete.The
vibrators should be held vertically, moved axially and withdrawn very
slowly,in such a way that the cavity left in the concrete can suitably fill up.

Surface Vibration

The thickness of layers of concrete compacted by surface vibration with the

aid of vibrating screeds or floats should not exceed 20 cm.
Vibration gives the concrete its maximum density by eliminating air
voids and ensuring perfect filling of the formwork.It considerably reduces
the internal friction of the particles of the constituents of the concrete and
tends to give them liquid properties.
The following methods can be distinguished:
(a)External vibration:the vibrators are attached to the formwork,which
170
should therefore be of strong construction. This method is seldom
used on construction sites.
(bì Internal vibration:internal vibrators (alsoknown as immersion vibra-
tors or poker vibrators) of varying thicknecs are inserted into the
concrete.A vibrator of this kind consists of a tube containing a pneu-
matically powered high-speedturbine driving a slightly eccentrically
mounted vibrating unit.
(c) Surfacevibration:this form ofvibration is effected by means ofvibrating
floats, screeds or surface compactors, generally on relatively large
areas of concrete:precast panels,slabs,pavings,etc.
Vibration should not be overdone,especially with concrete of very plastic
consistency,because the liquefaction of the concrete causes the larger aggre-
gate particles to sink to the bottom,with the result that at the surface the mix
will contain an excess of fine constituents(mortar)and water.It is preferable
to apply vibration for short periods,but at a large number of points located
sufficiently close together. The internal vibrators should be withdrawn
slowly from the concrete,before vibration is stopped,so as to avoid leaving
holes in the concrete, as these would subsequently fill up with mortar,
laitance or water.

9.3.5 INTERRUPTION A N D R E S U M P T I O N OF C O N C R E T I N G

Interruptions in the concreting of a structural member should as far as

possible be avoided.If they cannot be avoided,it is necessary to take pre-
cautions in order to ensure good adhesion of the new to the old concrete:
in particular, the surface of the construction joint should be roughened
by hacking,and thoroughly cleaned,so as to expose the coarse aggregate
particles;this surface should then be long and copiously wetted in order to
saturate the old concrete with water,the excess water finally being eliminated
with the aid of compressed air before concreting is resumed.Construction
joints (i.e.,joints where concreting is interrupted for some considerable
length of time) should be located only in compressive zones and their
positions should be explicitly indicated in the initial concreting scheme.
If a member,because of its size,cannot be concreted in a single operation,
construction joints should be planned in advance and not be left to chance,
depending on the rate of progress of concreting or the end of a day’swork.
Construction joints should not be formed with more or less irregular sur-
faces but should,instead, be arranged according to planes perpendicular
to the direction of the stresses. In large volumes of concrete the construction
joints should not be arranged in too large planes but should be distributed
over a number of planes (instepped or staggered arrangement).
Vertical constructionjoint faces should be formed by means of temporary
formwork (so-calledstop ends or stunt ends). Alternatively, fine-meshed
wire netting supported by a rigid grating can be used. The netting is left
embedded in the concrete,and in this way a rough surface providing a good
key is obtained.However,in that case concrete of too plastic a consistency
 171
(or with a low content of coarse aggregate) should not be placed against
the netting,nor should the concrete placed against it be vibrated too close
to the joint or for too long a time.Thereafter,as soon as the concrete has set,
the laitance that has collected at the bottom ofthe wire netting through which
it has trickled should be removed.
Horizontalconstructionjoint faces(orfaces sloped at so low an angle as to
enable the concrete to be placed without requiring top formwork)should not
be too smooth,as is often the case as a result of the ‘fat’(finemortar or grout)
in the mix coming to the surface during vibration of the concrete.At the
start of setting,the surface can be roughened by pitting it with a sharp tool.
Failing this,the more or less hardened concrete surface should be hacked
and thoroughly cleaned before concreting is resumed.
O n the outer faces of the concrete a construction joint (whether vertical,
horizontal or inclined)must not appear as a more or less wavy line,but should
present a properly straight line.To achieve this,a small fillet may be placed
against the formwork at the termination of concreting,so as to form a clear-
cut stop to the concrete over a thickness of a few centimetres.
The first fewbatches of concrete placed against the construction joint on
resumption of concreting may be given a higher mortar content (byreducing
the quantity of coarse aggregates put into the mixer), more particularly if
the concrete used for the job has a gravel/sand ratio of more than 2.It is
not good practice to apply a cement slurry to the construction jpint surface
before resumption of concreting.

9.3.6 S T E A M C U R I N G O F C O N C R E T E

If it is necessary to accelerate the setting and hardening of concrete,steam

curing may be employed.The equipment provided for the purpose should
enable the concrete to be heated to a temperature of around 80°C,but the
rate of heating-upmust not exceed 20°Cper hour.Every precaution should
be taken to prevent drying-outof the concrete;also, the exposed surfaces
should be constantly in contact with steam.
Steam curing procedures call for special study,both with regard to the
composition and mix proportions of the concrete and with regard to the
equipment employed. With steam curing it is possible to speed up setting
and hardening very considerably and to remove some parts of the formwork
only a few hours after concreting.

9.3.7 C U R I N G OF C O N C R E T E

The object of curing is to keep the concrete sufficiently moist for satisfactory
hardening to take place;it is indispensable in dry and warm weather.
Curing should start as soon as initial setting of the concrete occurs,for a
delay of several hours may significantly impair its effectiveness: it should
be continued for a week in normal cases and for two weeks if the weather
is very dry and very warm.
172
Curing can be done either by wetting the concrete or by means of an
impermeable temporary coating applied to it.
Wet curing consists in spraying the exposed concrete surfaces and the
timber formwork with water two or three times daily, depending on the
temperature and the atmospheric humidity. Exposed surfaces (i.e.,not
covered by theformwork)aremost vulnerable,and it is advisableto coverthem
with straw mats or sacking which should be kept permanently moist by
spraying at appropriate intervals.Alternatively,horizontal surfaces may be
covered with a layer of sand.
If impermeable formwork is used,such as steel or sheet-metal-facedform-
work, there is no need to wet the concrete surfaces in contact with such
formwork so long as the latter has not been removed.
When concrete is cured by means of an impermeable temporary coating
(membrane curing), a special compound is sprayed on to the concrete
surfaces to be protected. The compound forms a coating on the concrete
and, being impermeable,prevents the evaporation of water from the con-
crete. These curing compounds are generally resin emulsions which break
up as soon as they come into contact with the freshly placed concrete.The
thin film of resin which is thus formed is the protective ‘membrane’.The
compound should be slightly coloured so as to enable the continuity and
evenness of application to be judged.

9.3.8 CONCRETING I N COLD WEATHER

Generally speaking,in normal cases where Portland cement is used without

an additive,concreting should be stopped when there is a likelihood of the
ambient temperature falling below freezing point (OOC) within the next
48 hours -i.e., for practical purposes,if the temperaturereading at 9 o’clock
in the morning (solartime)is below 5°C.
If pozzolanic or slag cements are employed,these minimum values should
be increased by 5°C.O n the other hand, for mass concrete or if calcium
chloride is used as an additive (in a proportion not exceeding 2% or 1 %,
as the case may be,in accordance with Section 9.3.1),these minimum values
may be reduced by 3°C.
If it is absolutely necessary to continue concreting at temperatures below
the limits specified here, special precautions -for which the contractor
must produce proper justification-should be taken so as to keep the whole
mass ofconcrete at a temperature above 0°Cthroughout the period of con-
creting,setting and hardening.
Subject to such special precautions being taken, concreting can, as an
exception,be carried out at temperatures below 0°C.These precautions may
be :
(a) keep the aggregates under protection where they can be heated a
little,or heat the stockpile by injecting steam into it through a nozzle;
(b)use a rapid-hardening cement which evolves considerable heat of
hydration;
(c) use a sufficientlyhigh cement content:350-400 kg/m3of concrete;
 173
(d)use a setting accelerator such as calcium chloride (in an amount not
exceeding the limit laid down in clause 9.3.1);
(e) use the least possible amount of mixing water compatible with the
desired consistency;
(f) use a plasticiser,wetting agent or air-entrainingagent;
(g) heat the mixing water so that its temperature on arrival at the mixer
does not exceed 70°C;
(h) avoid long transporting or waiting times before placing the concrete
and do not use long chutes;
(i) keep the mixer protected from the cold;
(j) use formwork made of fairly thick timber;thin steel formwork should
be heat-insulated,and this is all the more important according as the
concrete members concerned are thinner;
(k)effectively protect the exposed concrete surfaces(i.e.,
those not covered
by formwork)as soon as the concreting operations have ended.
If these various recommendations are strictly applied, it is generally
possible to continue concreting at temperatures between O" and -5°C.
At lower temperatures,down to -lWC,it is advisable not to do any con-
creting.
In frosty weather it is more than ever desirable to make control specimens
and store these under the same conditions as the concrete in the actual
structure (heat insulation of the moulds, protection of the demoulded
specimens,etc.). These specimens are very useful in cases where there is
doubt as to the possible effect of frost on the concrete:with them it can be
determined to what extent hardening may have been retarded by the cold
and whether certain operations(removalofformwork,prestressing,handling,
etc.) can be carried out as intended or whether they should be deferred.

9.3.9 C O N C R E T I N G IN W A R M WEATHER

In warm weather it is necessary to make sure that the concrete will not lose
too much of its water by evaporation. Special precautions should be taken
in handling, transporting and placing the concrete and during its setting
and hardening:in particular,curing is to be considered indispensable.
Special precautions which may have to be taken, depending on the
temperature and humidity of the ambient air,are the following:
(a)stop concreting during the hottest hours of the day; if necessary,
concreting may have to be done at night;
(b)use cold water or,possibly,water which has been cooled by having ice
added to it in advance (ice must never be put directly into the mixer);
(c) protect the stockpiles of aggregates from the sun and spray them with
water;
(d) never use hot cements;
(e) do not make too dry mixes;
(f) cover the skips in which the concrete is transported;
174
(g) spray and copiously wet the external surfaces of the formwork before
and after concreting;
(h) place the concrete in the formwork as quickly as possible after mixing;
(i) organise the curing of the concrete under the best possible conditions;
(j) use a setting retarder (inexceptional circumstances).

9.4 T O L E R A N C E S

The building owner may specify dimensional tolerances and alignment

tolerances for structures in the contract documents.
In certain cases it may be advantageous to relax or tighten the general
and normally accepted rules regarding tolerances.For example,the toler-
ances may be increased for massive structural members which will remain
concealed from view.O n the other hand,closer tolerances may be specified
for precast members whose assembly by fitting them together demands
a high degree of accuracy or for members whose alignment is important to
structural stability,such as slender arches or vaults.
In the absence of special requirements laid down by the building owner in
the contract documents,the following tolerances should be adopted.

9.4.1 DIMENSIONAL TOLERANCES

The tolerance on any dimension d-measured between opposite faces or

between edges or between edge intersections-is as follows:
yd (cm)for ordinary structures
and :
5 y d (cm)for structures requiring exceptionalaccuracy (moreparticularly in
the case of members precast in a factory and assembled on the site).
For a horizontalbeam the dimensions in question are the length,the width
(orthickness) of the web,and the depth (orthicknessin the vertical direction).
For a slab or a panel the dimensionsin question includenot only the length,
width and thickness,but also the lengths of the diagonals.

9.4.2 TOLERANCES ON PERPENDICULARITY

The tolerance on the perpendicularity of a vertical member with height h is

as follows:
M l h (cm)
where ci has the value stated in Table 9.9.
By load-bearing member is understood a member which essentially has to
a column,pillar,bridge pier,façade panel). If such a
carry vertical loads (e.g.,
member has vertical faces on two sides and inclined (‘battered’)on the two
other sides,the tolerances given in the first row of the above table should be
 175
adopted in the direction perpendicular to the vertical faces and the toler-
ances given in the second row should be adopted in the other transverse
direction.
Table 9.9
Ordinary structures Structures requiring exceptional
accuracy (e.g., prefabricated)

Load-bearingmembers with
vertical faces c( = 0.33 m = 0.20
Load-bearingmembers with
non-vertical faces c( = 040 u = 0.25
Non-load-bearingmembers m = 0.50 m = 0.33

By non-load-bearing member is understood a member which does not

essentially have to carry vertical loads.However,this does not necessarily
mean that such a member is not subjected to loading:it may,for example,
be a retaining wall.

9.4.3 T O L E R A N C E S O N STRAIGHTNESS

The tolerance on straightness along an edge or arris (or along any straight
generating line of a plane or ruled surface)is characterised by the permissible
maximum deviation in any segment oflength 1 ofthat edge (or generating
line). The following values are to adopted for this tolerance:
1
-
300
(with a minimum of 1 cm)for ordinary structures;and:
1
-
500
(witha minimum of 0.5cm)for structures requiring an exceptional degree of
accuracy.

9.4.4 STEEL FIXING T O L E R A N C E S

Tolerances O n The Minimum Concrete Cover To Reinforcement

(Distance T o Face Of Formwork)

For concrete faces formed against the (horizontal or inclined) bottoms of

formwork the tolerance on the minimum distance from each reinforcing
bar to the face is to be one-tenth (0.1)of that distance.To conform to this
tolerance it will be necessary to use spacers having the precise dimensions.
For concrete faces formed against lateral formwork faces (or against
176
sloping top formwork faces) the tolerance on the minimum distance from
each reinforcing bar to the face is to be one-fifth(0.2)of that distance.
For top concrete faces which are levelled and not formed againstformwork
the tolerance on the minimum distance from each reinforcing bar to the
face is to be one-quarter(0.25) of that distance.

Tolerances O n The Position O f The Main Reinforcement

In the direction in which the deviation from the correct position has the
most unfavourable effect upon the strength of the member, the tolerance
on the positioning of the main reinforcing bars (which are subjected to the
direct stresses acting upon the cross-sectionsof the member: beam, slab,
plate,shell,etc.)in relation to the positions shown on the working drawings
should be taken as one-tenth(0.1) of the total thickness of the concrete in
that same direction, with a limiting maximum value of 1 cm for beams
and 0.5cm for slabs,plate,shells,etc.
In the direction perpendicular to that envisaged above, the tolerance
should be taken as half (0.5)the distance to the nearest adjacent reinforcing
bar (if there is one), with a limiting maximum value of 1 cm in ail cases.

Tolerances O n The Position O f Transverse Reinforcement

For transversereinforcementin prismatic structuralmembers,such asbinders

and stirrups,the tolerance on the positioning of the bars in relation to the
positions shown on the working drawings should be taken as one-tenth
(0.1)of the distance between consecutive transverse reinforcing bars, with
a limiting maximum value of 2 cm.

9.4.5 SIMULTANEITY O F M O R E T H A N ONE TOLERANCE

If more than one tolerance is applicable,the severest should be adopted.

Part 2
 1

USUAL VALUES OF
SUPERIMPOSED LOADS AND
WIND ACTIONS

1.1 PREAMBLE
The nominal values of the superimposed working loads should be specified
by the building owner.The information contained in this chapter is given
as an indication,for guidance,and states the usual values adopted for these
superimposed loads for various types of structures.

1.2 DEFINITIONS
Four types of superimposed loading are to be distinguished.

1.2.1 FIXED S U P E R I M P O S E D W O R K I N G L O A D S

Which correspond to the weights of additional components or features

permanently attached to,or incorporated with,the load-bearing members,
e.g.: ceilings or finishes applied to ceilings, floor coverings,coatings and
surfacings applied to load-bearing members, fixed partitions,smoke flues,
ventilation ducts,etc.
The values of such superimposed loads should be calculated from the
volume and the highest density of the relevant materials under the conditions
of use.

1.2.2 VARIABLE S U P E R I M P O S E D W O R K I N G L O A D S

These comprise:
(a) static superimposed loads, which vary only in a gradual manner,
e.g.,furniture,stored objects or materials,etc.;
182
(b)superimposed loads due to persons occupying or moving about in
the rooms of the building.
T o simplify the design calculations, these superimposed loads are assumed
to be uniformly distributed. Their usual values are given in Section 1.3.
They include the ‘dynamiccoefficient’(see Section 1.2.3)for taking account
of moving loads due to persons walking about.

1.2.3 D Y N A M I C S U P E R I M P O S E D W O R K I N G L O A D S

These are superimposed loads which produce a dynamic effect in the struc-
ture in consequence of movements and variation of the forces involved:
persons, machinery, mobile equipment such as overhead travelling cranes,
etc.
They are introduced into the calculations with their nominal values
multiplied by an amplification coefficient of 1.20 or 1.30 (see Section 4.2.2
of Part 1) and furthermore by a dynamic coefficient as defined in Section
4.2.3 of Part 1.
If the dynamic effect causes vibrations in the structure, the dynamic
superimposed load should be considered both with the positive and with the
negative algebraic sign.These two possibilities should be taken into account
in the calculations.
In many cases the vibrations are long sustained or are frequently repeated,
whereby fatigue effects may arise which should be taken into account.
It is furthermore advisable to compare the fundamental vibration fre-
quencies and the harmonics of the structure and of its components with the
frequencies of the oscillating loads in order to obviate dangerous resonance
phenomena.

1.2.4 C L I M A T I CS U P E R I M P O S E D L O A D S

These are due to the effects of wind, snow and earthquake actions. Loads
due to wind are dealt with in Section 1.4 of this chapter.Earthquake loads
should be the subject of special investigation.

1.3 VARIABLE SUPERIMPOSED WORKING L O A D S

1.3.1 S P H E R E O F A P P L I C A T I O N

The values for variable superimposed working loads as indicated in Section

1.3.2,are applicable to the design of ordinary residential buildings, schools,
industrial buildings, office buildings or farm buildings, but not to civil
engineering structures,exhibition halls, railway stations,etc.
It is recommended that the following values.in which the dynamic co-
efficient is included, should be adopted in the calculations.
 183
1.3.2 N O M I N A L V A L U E S FOR SUPERIMPOSED LOADS

Flat roofs

kg/m2
(a) Flat roofs not accessible except for maintenance 100
(b) Privately accessible flat roofs 175
(c) Flat roofs accessible to the public 500
Instead of these superimposed loads, the climatic superimposed
loads or the water load up to overflow level should be taken into ac-
count if any of these loads is of greater magnitude.

Dwellings

(a) Rooms 175

(b) Stairs 250
(c) Balconies 350

Offices

(a) Private rooms (except for storage of office records) 200

(b) Public rooms (except for storage of office records) 250
(c) Stairs 400

Hospitals

(a) Individual wards and treatment rooms 175

(b) Stairs 400
(c) Balconies 350
(d)Multi-bed wards 350

Schools

(a) Classrooms 350

(b) Stairs,ambulatories 400

Shops And Stores

(a) Shops 400

(b) Department stores 500
 184
subject to the condition that special requirements may have to
be fulfilled in the case of particularly heavy goods

kg/m2
Theatres And Public Entertainment Buildings 500

Dance Halls 500

Warehouses And Workshops

The user should determine the actual superimposed loading upon the various
structural members, having regard to the materials stored, the method of
storage and the machines to be installed and taking due account of the
coefficientsjustified by experience.

Garages, Passages Or Courtyards Accessible T o Vehicles

The superimposed loads to be adopted should be laid down in the special

specifications, which should state the maximum load per axle, the wheel-
base and minimum distance between axles, and the minimum space (on
plan) occupied by a vehicle.
The secondary floor members should be checked for a concentrated load
equal to the heaviest axle load applied to an area of 10 c m x 10 cm = 100 cmz
(axle supported by a jack), without taking the dynamic coefficient into
account.
With regard to the design of columns and main beams the specifications
\
may take account of the loading due to the average axle load and spacing
of the various types of vehicle that are likely to be parked on one and the
same floor. If a garage must be designed to accommodate a particular type
of commercial vehicle, this should be stated in the specifications.
Unless otherwise specified,avalue of 1.15 should be adopted for the dynamic
coefficient.

1.3.3 R U L E S RELATING P A R T I C U L A R L Y TO V A R I A B L E
SUPERIMPOSED L O A D S

Progressive Reduction Of Superimposed Loads In Multi-Storey

Buildings

In a case where points of support carry the loads of several floors on which
the maximum superimposed loads are unlikely to be acting simultaneously
(residential buildings, office buildings, etc.), appropriately reduced values -
 185
as defined below -can permissibly be adopted in the design calculations
for the load-bearingmembers:
Let So be the superimposed load upon the roof and let SI,S2,S, ... S,,
be the respective superimposed loads upon the floors numbered 1, 2, 3 ... n,
starting from the top of the building.
For the design of the points of support (columns,etc.) the following
superimposed loads should then be adopted:
supports under roof So
supports under top
floor (floor 1) SO+S,
supports under floor 2 So+0.95(S1+ S,)
supports under floor 3 So+090(S,+S, +S,)
supports under floor 4 So+0.85(S1+S2+S, +S,)
..................................
3+n
supports under floor n So+--(SI..
2n
.)(SI+S, +S3+ ...Sn)
the coefficient (3+n)/2n is valid for n > 5.
If the superimposed load is the same for all the floors of the building,
the above rule of progressive reduction is equivalent to the usual rule
whereby the superimposed loads for the successive floors are reduced in the
following proportions:
for the roof SO
for the top floor S
for the floor directly below 0.90s
for the next floor below this 0.80s
and so on, applying a reduction of 10% per floor until a value of 050s is
reached,which is thereafter retained for all the following floors below.
In residential buildings the reduction of the superimposed loads should
not be applied to any shop or office floors that may be present; for these
the full superimposed load should be adopted for designing the points of
support.
In principle, no reduction of the superimposed loads is allowed for the
floors of warehouses, shops, schools or workshops.

Lightweight Demountable Partitions

To take account of the possibilities of altering the positions of lightweight

demountable partitions weighing approximately 300 kg/m, their effect
upon the joists should -on the assumption that the partitions are normally
distributed and not installed particularly close together*-be allowed for
by adding 75 kg/m2 to the uniformly.distributed static superimposed load,
provided that it is justified by calculation and, if necessary,by experimental
means that the transverse interconnection of the joists enables the weight
*E.g.shower-bath and dressing cubicles.
186
of the partition to be distributed over a number of them and that no deforma-
tions liable to cause cracking of division walls and partitions will occur.

Roofs
If the total climatic superimposed load on the members of the supporting
framework (purlin,rafter) is less than 200 kg, such members should be de-
signed for a concentrated superimposed load of 100 kg applied at any point,
in order to take account of the weight of persons walking about on the roof,
in addition to the permanent load (dead weight) of the roof covering.
For the same reason the secondary roof components, such as laths,
boarding,etc. should be able to carry a uniformly distributed loading of 100
kg/m2.
The actual roof covering units (other than of the traditional kind), such
as concrete slabs of more than 1 m span between supports,etc.,should satisfy
the same requirement.
However,this requirement does not apply to brittle roofcovering materials
such as slate,glass, or asbestos cement,on condition that special precautions
are taken to protect these in the event of persons having access to the roof.

Mobile Equipment
The superimposed loads due to mobile equipment (overhead travelling
cranes,other cranes,lifts) -i.e., ‘live loads’in the true sense -should include
the weight of such equipment,together with the reactions due to the move-
ment thereof (inertialforces,impact effects, transverse forces,swaying, etc.).
The building owner, in consultation with the manufacturers of the equip-
ment concerned, should state these loads accurately in the special speci-
fications,which should also indicate the appropriate coefficientsto be applied
in order to take account of dynamic effects.

Horizontal Forces On Parapets

Balcony parapets, railings, handrails and their fixings should be designed to

withstand a horizontal force applied at the top, this force being 60 kg/m
for private premises and 100 kg/m for those accessible to the public.
The static and elastic stability should be checked for this force multiplied
by an amplification coefficient equal to 513.

Test Loads

If structural members have to be subjected to loading tests and if the en-

visaged test loads are larger than the superimposed working loads, taking
into account the amplification coefficients for dynamic effect in terms of its
 187
algebraic value, then the test loads in lieu of the actual superimposed loads
should be adopted in the design calculations for such members.

1.4 WIND EFFECTS

1.4.1 C H E C K I N G P R O C E D U R E

The strength of the structures should be checked for the following conditions:
(a) the effect of the nominal wind pressures,in conjunction with the most
unfavourable combination of vertical loads, having due regard to the
basic strengths of the materials;
(b)the effect of the exceptional wind pressures, in conjunction with the
most unfavourable combination of vertical loads, on the assumption
that the basic strengths of the materials are increased by 50%.

1.4.2 SPHERE OF APPLICATION

The simplified rules given in this clause are applicable only to ordinary
residential buildings, office buildings, warehouses or factories comprising

Figure 1.1.Qfh22"<u<40° or 0.404Q tanuG0.839

parallelepipedal blocks rectangular on plan and of normal height, con-

sisting of substantially identical storeys,with masonry walls and partitions.
These structures should conform to the following requirements:
(a) The structure should consist of a single block or of blocks combined
under a single roof.
(b)The base at ground level should be rectangular, with length a and
width b.
(c) The height h-i.e., the difference between the level of the base of the
structure and the level of the top of the roof-should not exceed 30 m.
(d)The dimensions should satisfy the following conditions (see Figure
1.1)
h/b>0.25;
h/a < 2.5,with the additional condition blu < 0.4if h(b > 2.5;
f<;h for a roof with two plane sloping surfaces,and
f<+h for a vaulted roof.
188
(e) The roof should be:
either a flat roof;
or a pitched roof of height f; with one or two plane surfaces inclined
at not more than 40"to the horizontal;
or a curved (vaulted)roof for which the tangent plane at the springing
of the vault is inclined at not more than 40" and not less than 20" to
the horizontal.
(f) The walls should:
rest directly on the ground;
be flat, without discontinuities or indentations;
have a permeability p of not more than 5 or, for only one of them,a
permeability p of not less than 35.*
(g) The period of the fundamental mode of oscillation should be not more
than 1 second for reinforced concrete frameworks and not more than
0.75 second for steel frameworks.
(h)The structure should be situated on ground which is substantially
horizontal for some considerabledistance around.

1.4.3 DEFINITIONSA N D G E N E R A L P R I N C I P L E S

Wind Direction

For the purpose of structuraldesign the average overall wind direction should
be assumed horizontal.

Exposure O f Surfaces

Suppose the structure to be illuminated by a beam of light parallel to the

overall direction of the wind:

'bulent

Figure 1.2

(a) the lighted surfaces (exposed to the wind) are referred to as the 'wind-
ward' surfaces;
*A wail has a permeability to wind of p% if it comprises openings (of whatever dimensions)
whose total area,i.e.,the sum of the areas of the openings,is p %of the overall area of the wall.
 189
(b)the surfaces receiving no light (not exposed to the wind) or parallel to
the light beam (parallel to the wind direction) are referred to as ‘lee-
ward’surfaces.
In aerodynamics the windward surfaces are subject to a regular air flow
pattern without separation of the streamline flow.
Leeward surfaces are subject to turbulent flow. They are separated from
one another by lines of separation of the air filaments (see Figure 1.2).

Projected Area

The ‘projectedarea’ is the orthogonal projection of the surface under con-

sideration,or of the structureas a whole,upon a plane normal to the direction

Figure 1.3

of the wind or, in optical terms,the shadow area projected on to a plane at

right angles to the beam of light (see Figure 1.3).

Wind Action On O n e Of The Faces Of A Wall Element

The action exerted by wind on one of the faces of a wall element is regarded
as being directed at right angles to that element and is a function of:
(a) the wind velocity;
(b)the class of structure and its overall proportions;
(c) the location of that particular element in the structure and its orienta-
tion with regard to the wind;
(d) the dimensions of the particular element considered;
(e) the shape of the wall (flat or curved) to which that element belongs.

Dynamic Pressure And Pressure Coefficient

The elementary unit action exerted by wind on one of the faces of a wall
element is given by a product cq,
where: q denotes the dynamic pressure, which is a function of the wind
velocity (cf.Section 1.4.4);
c is a pressure coefficient, which is a function of the structural
arrangements.
A face of an element belonging to a structure is said to be subjected to a
190
positive pressure if the force perpendicular to the face tends to push it in-
wards. If the force acts in the opposite direction, i.e., tends to pull the face
outwards, it is described as negative pressure (or suction).

1.4.4 D Y N A M I C PRESSURE

Definition
The dynamic pressure q in decanewtons per m2 (daN/m2)is given, as a
function of the wind velocity v (in m/s), by the formula:
q = - VZ
16.3

Nominal And Exceptional Dynamic Pressure

According to Section 1.4.1it is necessary, in the structural design calcula-
tions,to consider a ‘nominal’and also an ‘exceptional’dynamic wind pressure;
the ratio of the latter to the former is taken as 1.75.

Regions
According to the magnitude of the nominal and the exceptional wind
velocities,as measured by weather stations,the various regions may broadly
be classified into three categories (seeTable 1.1).
Table 1.1
Nominal values Exceptional values

Region I 100km/h 130 km/h

Region II 120 km/h 160km/h
Region III 140 km/h 190 km/h

Dynamic Pressure Values

The dynamic pressures are constant over the entire height of the structure
and are given by the formula:
q = (48+O.6h)krk,
where the ‘region coefficient’k, has the values shown in Table 1.2.
Table 1.2

Nominal pressure Exceptional pressure

Region I 1.00 1.75

Region II 1.40 245
Region III 1.80 3.15
 191
and the ‘site coefficient’k, has the values shown in Table 1.3.
Table 1.3
Region I Region II Region 111

Protected site 0.80 0.80 0.80

Normal site 1.00 1.00 1.00
Exposed site 1.35 1.30 1.25

Three types of site are considered in the rules:

Protected site, e.g.:Bottom of a basin (low-lyingregion) surrounded by hills
and thus protected from wind in all directions.
Normal site, e.g.: Large plain or plateau with only minor differences in level
and with slopes of less than 1 in 10 (undulations).
Exposed site, e.g.:Near the sea:the coastal region in general (upto a distance
of about 6 km inland); cliff tops;islands or narrow peninsulas.
Inland: narrow valleys through which the wind rushes; isolated or high
mountains and some mountain passes.

Reductions
There is a screening effect when a structure is partly or completely screened
by other structures which can in all probability be expected to last for a long
time see Figure 1.4.The screening effect may manifest itself:

Figure 1.4.Examples of completely and partially protected structures. Building D is

completely protected by building A. Building B is completely protected over the height
h, by building A. Only the facades a and b of the buildings A and B are protected over
the length 1, and height h,

(a) either in an increase in the action of the wind, if the structure situated
behind the ‘screen’happens to be in a zone of turbulence,in which
case it is not possible to give rules;accurate informationcan be obtained
only from wind tunnel tests;
(b) or in a reduction in wind action in other cases:the structure in question
192
is thus protected,and the dynamic pressures relating to the protected
surfaces can permissibly be reduced by 25 %.
In this latter situation the surfaces concerned should satisfy both of the
following conditions:
(a) they should be entirely protected by the ‘screen’with regard to all wind
directions in the horizontal piane;
(b)they should be situated below the surface formed by a generating line
which slopes at a gradient of 1 in 5 (20%) in relation to the ground
inwards from the ‘screen’ (protecting structure) and is in contact
with the apparent outline thereof.
Since every case is in effect a special case, it is difficult to give general
rules,and the reduction in respect ofthe screening effect should only be applied

Largest dimension of surface presented to the wind

Figure 1.5.Dynamic pressure reduction coeficientfor large surfaces

with caution because of the possibility of a turbulent ‘wake’behind the

screening structure.
The dynamic pressures determined above can permissibly be reduced by
applying a reduction coefficient 6 as obtained from Figure 1.5, where 6 is
indicated as a function of the largest dimension of the surface presented to
the oncoming wind by the element considered in the calculation.

1.4.5 EXTERNAL A N D I N T E R N A L ACTIONS

Definition
The external wall surfaces of any structure are subjected to :
(a) negative pressures (suction)in the case of leeward surfaces;
(b)positive or negative pressures in the case of windward surfaces.
These actions are referred to as ‘external’actions.
 193
In closed, open or partly open structures the internal spaces enclosed
within the walls may be in a state of positive or of negative pressure (in
relation to the normal ambient air pressure), depending on the orientation
of the openings with regard to the wind direction and their relative size.
As a result, the internal surfaces of such structures are subject to what are
referred to as 'internal' actions.
The external actions are characterised by a coefficient ce,and the internal
actions are characterised by a coefficient ci.
The internal actions are essentially a function of the permeability of the
structure.
Thus, if all the walls and the roof are wind-tight and completely enclose
the interior, there will be no internal wind action, and the structure will,
in this respect,behave as if it were solid.
Otherwise it has been found that in isolated buildings of normal shape
and proportions an internal negative pressure of between -0.2q and -0.4q
will occur if the percentage p of the openings in relation to the total wall
surface area varies between 0.03 and 5. In such structures the average
internal pressure conditions are not affected in any practically significant
way when a door or a window is opened (or broken), even though this may
considerably affect the pressure within the room concerned by suddenly
raising it to the maximum pressure corresponding to that in an open structure.
O n the other hand, in open structures the internal pressure is significantly
altered when an opening is formed in one of the walls. The interior of the
structure has a positive pressure if the opening is on the windward side;it
has a negative pressure if the opening is on the side facing away from the
wind or in a wall parallel to the wind direction or,in some cases,on the sloped
roof surface exposed to the wind (in the case of slightly sloping roofs).

External Actions

O n the assumption that the wind direction is perpendicular to the vertical

wall surfaces of the structure, the following coefficients should be adopted:

Average Actions

Vertical walls :
windward: ce = + O 8
leeward: ce = - 0 5
Roof:
Wind normal to the generating lines (see Figure 1.6).The values of e,,
denoting the average pressure coefficient (plane sloping surfaces) or the
pressure coefficient at a particular point (vault or curved roof)are given in the
194
Table 1.4,where u denotes the angle of the plane slope (in degrees), or of the
tangent to the vault, in relation to the horizontal.
Table 1.4

Windward Leeward

IaI YO "O

Flat
sided
o" < la1 < 10" -2 (20.25 +gj
roof lO"<IaI <4o"

o"< la(< 10" -1-8(040+~j -1.8(04I-$j

Vaulted With minimum = -0.8

roof
10" < IC(( < 40" -2k50-i)
( !;A)
- 1.8 0.40- -

With maximum = -0.27

Figure 1.6. Roofs with plane sloped surfaces or curved surfaces (vaults).
Coefficient ce.B = curved roofs, A = roofs with plane sloped surfaces

Wind parallel to the generating lines:

In this case the value to be adopted for ce is that corresponding to u = O
for plane sloped surfaces.
 195
Local Actions

Along roof edges and vertical arrises,over a distance (measured from the
edge or arris)equal to one-tenthof the minor horizontal dimension b of the
structure:c = 2c, (suction).
At corners,where the above-mentionedzones along roofedges and arrises
overlap:c = 3c, (suction).
Additional to these local actionsare other externalactions such as average
actions on the eaves (undersides of overhanging roofs) or internal actions,

-b
Figure 1.7. Example of limitation of the -2,.0,7
resultant action. In the zone A B the = -i.&
resultant action of -2.2q is limited to A
-2 q. In the zone BC the resultant action
is -1.7 q. In the zone CD the resultant +

action is 1.0q

but on the understanding that the resultant action cannot exceed -2q and
-3q respectively (see Figure 1.7).
These local actions are applicable only to the design of roof covering
elements (tiles, slates,slabs), cladding elements or flashings and to their
fastenings and supports in the zones defined.
Observations made after storms and in wind tunnel tests have clearly
shown that in a great many cases the damage,and often the most serious

Figure 1.8

damage, sustained by buildings and other structures as a result of wind

action is due to suction effects or to combined pressure and suction effects.
Certain small roof covering elements (tiles,panes of glass,flashings,etc.)
which are not, or only insufficiently,secured are often in danger of being
lifted and even torn off by the action of positive internal and negative
external pressures.
Such effects are locally intensified along arrises, edges, gables, eaves,
cornices,chimneys, etc. and wherever high local wind velocities develop
in consequence of particular air flow conditions,which are moreover often
very unstable in these parts of a structure (see Figure 1.8).
196
These local effects frequently produce sudden irregular actions which may
initiate the detachment of relatively light cladding elements.
Clips and other fastenings for these roof cladding elements should be
designed with considerable care, calling for extensive technological
experience.

Internal Actions (See Figure 1.9)

Closed structures: ci = f0.3

Open structures: with opening on windward side: ci = + O 4
with opening on leeward side: ci = -0.5

- 0.72

Figure 1.9.Examples of application. Closed structure with two plane sloped roof
surfaces. Open structure with curved roof (vault) with circular generating line

Unit Values Of Resultant Actions O n Walls And Roof Slopes

These values are determined by finding, for each element, the most un-
favourable combination of the average external actions and internal actions.
They are expressed by (ce-ci)q.
For example, for the vertical walls the average external actions combined
with the internal actions give the following unit values of the resultant
actions:

closed structures open structures

+ 1.lq f 1.3q
-0.8q
Since the wind may blow from any direction in relation to the structure,
in many cases it is possible to confine oneself,for roofs,only to the maximum
values of the actions upon the roof slopes (e.g.,for a roof with a slope of 30"
on an open structure the two slopes can be designed for a suction equal to
(-0.45-0.8)q = -1.25q).But both values (windward slope,leeward slope)
should be taken into account in structures (e.g.,lattice trusses,etc.)for which
the combination of different actions upon the two slopes of the roof would
 197
produce more unfavourable results in certain members (web members of the
trusses,etc.).

Overall Actions

These are obtained by geometrically compounding the total resultant

actions upon the various walls of the structure.
For example, for a structure with a rectangular base and a flat roof the
overturning force is expressed by T = 1.3qha (or b), irrespective of whether
the structure is an open or a closed one, and the centrally acting force
tending to lift the structure is expressed by:
U 5 O&& for closed structures
U 5 1.3qSUfor open structures
S, denotes the area of the horizontal projection of the structure.

Blocks Joined Together In A Single Row And Covered By O n e Roof

The simplified method may be extended to the case where several blocks
are joined together in a single row under one roof, provided that the blocks
as a whole and each block separately satisfy the conditions stated above.
Independently of the overall calculation,all the intermediate blocks should
be considered as closed and isolated and be checked with regard to their
ability to withstand overall actions equal to six-tenths (6/10)of those cal-
culated in accordance with the previous section assuming the wind to be
blowing in a direction normal to the plane of the joints.

1.4.6 S P E C I A L CASES

Structures which do not conform to the definitions given above should be

the subject of special investigation.

e
 2
DETERMINATION OF SAFETY
ON THE BASIS OF
PROBABILITY

2.1 PRELIMINARY C O N S I D E R A T I O N S
W h e n a structure collapses or, in more general terms,becomes unfit for ser-
vice, this depends on a large number of factors that render it unsafe.
These factors include: inaccuracy in knowing the values of the super-
imposed (or live) loads and other actions that the structure will have to
support throughout its service life; inaccuracy in the basic hypotheses and
in the design calculations;the degree of skill put into the design;inaccuracy
in the design engineer’s assumptions as to the basic properties of the con-
crete and steel or the failure of these materials to comply with those pro-
perties;the degree of strictnesswith which the materials are checked and the
quality of execution of the work is supervised;and the possible deterioration
of the structure in course of time.
Since these factors which make a structure unsafe are ofa random character,
it appears rational to introduce the conception of probability in establishing
methods for the determination of structural safety.Actually, the problem
consists in keeping the probability of collapse,or unfitness for service,within
a permissible limit which must take various factors into account, such as:
cost of construction,maintenance of the structure.insurance against various
risks,psychological considerations,etc.,and must do this in accordance with
criteria comparable to those used in assessing insurance risks.
Obviously, the conventional notion of the factor of safety as an arbitrary
overall figure makes only very inaccurate allowance for all these parameters
and their effects on structural safety.
However, for a probability analysis to have the requisite reliability,it is
essential to know the corresponding laws of statistical distribution. U n -
fortunately,the statisticaldata at present available are still very inadequate.
Indeed, in many cases they are non-existent.Hence it appears difficult tq

I98
 199
envisage so rigorous and reliable an application of the probability theory
to the practical determination of structural safety.
This difficulty prompted the Comité Européen du Béton and the Conseil
International du Bâtiment to develop a semi-probability method, which con-
sists in defining characteristic values of the mechanical strengths and the
loadings, calculated from the average values thereof by introducing the
standard deviation and assuming a normal statistical distribution for
the experimental results.T o these characteristic values are applied a reduc-
tion coefficient (forthe mechanical strength properties)and an amplification
coefficient (for the loadings) which may take account of the greater or less
degree of uncertainty in our knowledge and in the actual functioning con-
ditions of a structural member considered in the structure as a whole. T o
each limit state of the structure corresponds a set of Coefficients whereby
the safety appropriate to that particular state can be determined.
The rules given in the UNESCO Code are based on similar principles,
but, with a view to simplifying the procedure, the following modifications
have been introduced for most normal purposes of design :
(a) The characteristic value of the steel strength and concrete strength
has been replaced by the guaranteed minimum value.
(b) Instead of applying an amplification coefficient to the loadings, a
further reduction of the strengths is applied, with the result that the
amplification and reduction coefficients are reduced to two in all,
namely,a single factor of safety for each of the two materials,and that
the characteristic loadings are used directly in the calculation.

2.2 THE PRINCIPLES OF THE PROBABILITY THEORIES

OF SAFETY
The object of the analysis of strlictural safety is to ensure that the probability
of reaching the envisaged limit state will remain below a certain value est-
ablished in advance for the type of structure considered.
In so far as is compatible with moral considerations (such as respect for
human life) and, possibly, with psychological considerations (e.g.,possible
reactions of public opinion following an accident), the permissible limiting
value could be so determined as to achieve a minimum for the total of the
initial cost of the structure,the capital sum needed to maintain it during its
service life, and the amount of fictitious insurance premium covering the
risks of material damage and personal injury during construction and
throughout its service life. This fictitious premium should take account of
the possibility of the structure being rendered unfit for service, this being
dependent upon the probability relationships deduced from the collected
statisticaldata.
A complete probability calculation would call for a knowledge of the
pattern of distribution of the most unfavourable loadings liable to occur
during the construction and throughout the service lives of similar structures,
and also for a knowledge of the capacity of the structure or of its various
members to resist such loadings.It would therefore be necessary to take into
200
account all the random factors that could affect the probability of reaching
the envisaged limit state:
(a) The degree of approximationof the actual calculationand.in particular,
of the basic assumptions,having due regard to the nature of the struc-
ture,its value as a design,and its constructional features.
(b)The strength values of the materials in the most severely loaded zones,
particularly with regard to the choice of the construction methods
and the quality of the workmanship, the strictness of control and
supervision applied,and the deterioration of the structure in the course
of its service life.
(c) The values of the loadings which are most unfavourable with regard
to stability.
However, the necessary statistical data are at present inadequate or en-
tirely lacking.Under these conditionsthe systematic and general application
of the probability principles presents serious difficulties. It is therefore
advisable to confine oneself to making the best possible use of the statistical
results already obtained and estimating as well as possible the dispersions
concerning which little valid information is available at the present time.
Later, corrections will have to be made as more knowledge is acquired. It
should, finally, be noted that certain phenomena cannot be regarded as
being entirely of a random character.

2.3 C.E.BJC.1.B. SEMI-PROBABILITYDESIGN METHOD

The semi-probability concept adopted by the Comité Européen du Béton
(C.E.B.) and the Conseil International du Bâtiment (C.I.B.)consists in limit-
ing -as a first approximation-the actual statistical analysis to the study of
the variation of the strengths and the intensity of the loads and other actions.
For these quantities the dispersion (orscatter)of the data is taken into account
by the introduction of characteristic values G, which are estimated by means
of expressions having the following general form :
G, = Gm(1 fk6)
where G, is the mean value of the quantity considered,6 is its relativestandard
deviation, and k is a variable coefficient whose value depends on the pro-
bability -accepted a priori -that the data will be outside the interval
( G m ,GJ
These considerations lead to the adoption of a characteristic value for the
strengths (‘characteristicstrengths’),which defines the mechanical properties
of the materials, and a characteristic value for the permanent loads, super-
imposed loads and other actions (‘characteristicloadings’).
As for the remaining uncertainties, which are not amenable to statistical
analysis,these are taken into account by the application of a series of design
coefficientsy to the characteristic quantities.Thus,for the materials so-called
‘designstrengths’are defined,which are equal to the characteristic strengths
divided by a reduction coefficienty,,, which allows for the fact that the actual
mechanical properties of the steel and concrete may be inferior to those
assumed in the design. Similarly, for the permanent loads, superimposed
 201
loads and other actions so-called ‘designloadings’ are used as the basic
design values, these being equal to the characteristic loadings multiplied
by an amplification coefficient y, which allows for the fact that the loadings
actually applied to the structure may be higher or may give rise to effects
that are more unfavourable than those assumed in the design.
Under these conditions,determining the safety of a structure consists in
checking that,in all the limit states,the effects of the design loadings do not
exceed the values that can permissibly be attained with reference to the
design strengths of the materials.In other words,for a given loading and a
given material,the overall factor of safety is equal to the product y,. y,.
This semi-probabilitymethod offers many advantages.
In the first place,and despite any appearance to the contrary,it is not a
complicated method,the product of abstract speculationsof novelty-seeking
experts. By way of example it is of interest to call attention to a.specific
problem to which the semi-probabilitymethod provides a simple solution:
Consider the ultimate strength of a reinforced concrete section subjected to
bending. Obviously, it will be necessary to adopt a different margin of
safety according as failure of the concrete or failure of the steel causes failure
of the member. Since the failure of the concrete is, generally speaking,less
precisely predictable than that of the steel, it stands to reason that a larger
reduction coefficient should be applied to the concrete strength than to the
steel strength.In this way a greater measure of safety can be obtained for
sections which fail as a result of concrete failure than for those which fail
as a result of steel failure.This approach to the problem is in fact precisely
the approach that the semi-probabilitymethod adopts.
Besides,it is a perfectly logical method,inasmuch as it assigns to each of
the uncertainty factors a share of the overall safety of the structure. The
introduction of characteristic strengths and characteristic loadings takes
account of the variability of the mechanical properties of the materials and
the variability ofthe various kinds ofsuperimposed loads,while the reduction
coefficients y, for the characteristic strengths and applification coefficients
y, for the characteristic loadings take the other uncertainty factors into
account,such as faults of construction,the inaccuracy of the design assump-
tions, errors in the interpretation of the behaviour of the structure,etc.
Evidently,this procedure enables all those who share responsibility for the
structure -manufacturers of the materials,designer,contractor-to form a
precise idea of the safety margin available to them and thus to assess their
own respective responsibilities.

2.4 CHARACTERISTIC VALUES AND DESIGN VALUES OF

THE PERMANENT LOADS, SUPERIMPOSED L O A D S
AND OTHER ACTIONS
2.4.1 GENERAL DEFINITION

For loadings that can be considered to be random in character the charac-

teristic value Qkis defined by the relation:
Q k = QJ1+ kd)
202
where Q,, which is based on the statistical analysis of all structures of the
same type and the same durability of the structure being designed,denotes
the value of the most unfavourable loading which has a 50% probability
of being exceeded (in the direction of abnormally high values) only once
during the anticipated service life of the structure,while 6 denotes the
relative standard deviation (or‘dispersioncoefficient’)of the distribution of
the maximum loadings,and k is a coefficient depending on the probability -
accepted a priori -of having loadings in excess of the value Qk.
As an indicationit can be said that the dispersionsfound to occur in certain
countries varied from O to 0.15 for the permanent loads (closeto O for very
large beams; between 0.08and 0.10for slabs;and of the order of 0.15 for
materials whose specific gravity is not accurately known or varies with
moisture conditions), from 0.10to 0.20for superimposed loads (around
0.10for residential buildings and offices; of the order of 0.15 for certain
industrial superimposed loads).
It should be noted that in certain cases a reduction of the loading may
endanger the stability of the structure.However,special cases of this kind
can relate only to the permanent loads,since,for superimposed loads,the
least possible loading corresponds always to the absence of superimposed
load. On the other hand, for permanent loads, obvious examples are the
chimney or retaining wall whose own weight contributes to its stability.In
such cases the characteristic value Qk is defined by the relation:
Q; = Qh(1 -k6)
where QL denotes the value of the most unfavourable loading which has a
50%probability ofbeing exceeded (inthe direction ofabnormally low values)
only once during the anticipated service life of the structure,while 6 denotes
the relative standard deviation (or‘dispersioncoefficient’)of the distribution
of the minimum loadings,and k is a Coefficient depending on the probability
-accepted a priori -of having loadings that fall short of the value Qk.
For superimposed loads (or live loads) which cannot be considered to
present a random character the values to adopt in the design may be estab-
lished a priori by a decision (standard highway loading,standard railway
loading,convoys of military vehicles) or be chosen with reference to the
intended use to which the structure will be put. In such cases the values
thus established or chosen should be taken as the characteristic values.
The design value ofthe permanent loads and superimposed loads is derived
from the characteristic value thereof by means of a relation of the type:
Q* = Ys&
where the amplification coefficient y, enables the following to be taken into
account:
(a)The probability of reaching a limit state,which may be chosen as a
function of the extent of the harm resulting from a possible accident.
The higher the probability of reaching a limit state,the higher should
be the value of ys; at the same time,however,this probability should
not be too high, otherwise the cost of the structure would become
prohibitive.
 203
(b)The possible increase of the permanent loads or superimposed loads
and of their effects beyond the anticipated values,not only as a result
of statistical deviations (given by 6) of the permanent loads or the
superimposed loads themselves,but also as a result of the inaccuracy
of the design assumptions and possible inaccuracies of calculation
and because certain influences are sometimes neglected.The coefficient
y, should,inter alia,allow for the effect of the degree of approximation
of the design method and other factors upon the behaviour of the
structure to be designed.
(c) Possible errors of construction,which in turn affect the magnitude
of the forces that the concrete and the reinforcement have to resist.
Examples of this are afforded by incorrect positioning of the reinforce-
ment,defective functioning of a hinge,an error in setting out the curve
of an arch,the out-of-plumbnessof a column,etc. This also happens
when an increase or a decrease of loading is caused by deformation
of formwork and affects the magnitude of the permanent loads, or
indeed when any error occurs that is liable to alter significantly the
geometrical dimensions of the components of the structure (beam
spans,column heights,etc.) or their relative positions (e.g., accidental
eccentricities in the point of application of a load). The additional
forces arising from these errors and defects of execution on the site
have not been taken into account in the statistical evaluation of the
superimposed loads and therefore have to be catered for by an increase
in the loadings.
The designer should consider each problem with care.For example:
y, should have a higher value for thin slabs,in which relative errors
in the thickness ofthe concrete and in the positioning of the reinforcing
bars are more harmful and may have more serious consequences
than in thick slabs.
Similarly,if the application of a given load or superimposed load
corresponds to a low or zero stress due to taking account of positive
and negative areas of the influence line (lattice member of a roof
truss,for example), the given coefficient y, will not be sufficient (since
y, x O = O). In this type of structure the stress should be increased in
accordance with criteria which are rather difficult to lay down in a
set of regulations.
Of course, the coefficient y, does not take account of mistakes
in design,nor of serious defects in construction.
(d)The probability of various loads or superimposed loads not occurring
simultaneously.The coefficient y, must also take account of the com-
bination of the various categories of loads or superimposed loads
(deadweight,working loads,wind,snow,etc.). Each of these loads or
superimposed loads very rarely attains its maximum value, and the
probability of their simultaneous occurrence is low. Accordingly,
the value of ys is allowed to be reduced if it is applied to the whole set of
loads and superimposed loads of various kinds (in some countries a
reduction of as much as 30%appears to be permissible).
(e) The possibility of redistribution of forces and moments. The value of
204
ys should likewise depend on the type of structure. This value could
be lower for structures or structural members in which a redistribution
of forces and bending moments is possible than for structures in which
the failure of a single member may bring about total collapse or for
members which are liable to fracture suddenly and without warning.
As regards the actions arising from the effects of shrinkage,creep,
temperature,etc.,these should also be multiplied by an amplification
coefficient ys whose magnitude should be determined in accordance
with the same principles.

2.4.2 CHARACTERISTIC V A L U E S A N D DESIGN V A L U E S

OF T H E P E R M A N E N T L O A D S

Apart from exceptional cases, no dispersion should be assumed to exist

in the values of the permanent loads estimated from the volume and density
of the materials in the conditions in which they are used (i.e.,6 = O). The
exceptional cases include more particularly those where the stability against
overturning is concerned and where the density of the materials is not
accurately known or is liable to vary (6# O). In such cases a value k = 1
may be adopted,so that the relation for the characteristic permanent loads
then becomes:
G, = G,(1 f6)

with the sign corresponding to the most unfavourable effect.

Ultimate Limit State

The design values of the permanent loads are derived from the characteristic
permanent loads by the application ofthe amplification coefficient ys = 1.40.
However, in the exceptional limiting case where all the most favourable
conditions of design,construction or use of the structure are simultaneously
satisfied (accuracyof the basic assumptions,exactness of the design methods
and calculations, very careful design of anchorages and structural con-
nections,carefulworkmanship,constantsupervision,only material damage-
not personal injury -resulting from a possible structural mishap) this
amplification coefficient for the ultimate limit state may be reduced to :
ys = 1.25.In fact,it is very rare for these ideal conditions to be all satisfied
simultaneously.And if this is not so,the following additionsto,or deductions
from,the value of 1.25should be made:
+0.15 in the case of uncertain assumptions,design and calculations carried
out with moderate care;
+0.15in the case of moderately good workmanship;
+0.15in the case of very considerable risks of damage;
-0.15 in the case of very slight risks of damage.
 205
Finally,on adding together these increases or reductions,the following
value for the ultimate limit state is obtained in almost every case:
y, = 1.40

Limit State Of Instability

For the limit state of instability,which generally corresponds to the case of
failure without warning and is liable to bring about the total collapse of the
structure,the coefficient y, relating to the member under consideration
should be suitably increased with regard to the value ys = 1.40.It is advisable
to adopt:y, = 1.70.

Limit States Of Cracking And Deformation

For the limit state of cracking and the limit state of deformation,which
correspond practically to the normal working load conditions, the co-
efficient y, may,on the other hand,be reduced to:ys = 1.00.

Summary
Ultimate limit state: G* = 1.40Gk= 1.40Gm(1 6)
Limit state of cracking: G* = G, = Gm(l+a)
Limit state of deformation: G* = G, = Gm(1 f6)
Limit state of instability: G* = 1.70Gk= 1.70Gm(1 6)

2.4.3 CHARACTERISTIC VALUES A N D DESIGN VALUES

OF T H E SUPERIMPOSED L O A D S A N D O T H E R ACTIONS

The superimposed loads and other actions comprise:

(a)the fixed superimposed loads and the working loads (live loads,
variable superimposed loads);
(b)the superimposed loads due to climatic conditions: snow and wind;
(c) the actions that may arise from the working loads,such as braking
and swaying effects,centrifugal forces,vibration phenomena ;
(d)the displacements,frictional effects and resistances of supports and
bearings;
(e) the effects of shrinkage,creep and temperature variations ;
206
(f) the effects of earthquakes(if any);
(g)the influence of the method of construction (if at all).
The characteristic superimposed loads should be determined by means
of the expression:
Qk = QJ1+ 6)
where the value of the coefficient k by which 6 is multiplied is taken as equal
to 1 and the value of the dispersion coefficient 6 is laid down in the special
specifications.It should also be noted that, contrary to the procedure for
certain types of permanent loads,the dispersion 6 should always be added
(‘plus’sign,never the ‘minus’sign), for if the ‘minus’really had to be con-
sidered as applicable to certain types of superimposed loads,in any case the
most unfavourable effect would always correspond to a zero value of such
loads.
The recommendations of the Comité Européen du Béton also lay down a
15% increase for all variable superimposed loads and live loads whose
effect is much more dangerous than that of the fixed superimposed loads.
This increase of 15 % takes account of the variation in the direction and
magnitude of the stresses;it should not be confused with the usual dynamic
coefficients.In fact,if the variable superimposed loads and live loads are
additionally accompanied by dynamic effects, these loads must, over and
above the 15 % increase referred to just now,undergo a further increase by
the application of so-called dynamic coefficients of suitably chosen value
to take account of the influence of inertial forces,impact effects or vibration
and their resultant phenomena.
The design values of the superimposed loads are derived from the charac-
teristic superimposed loads by application of the same amplification co-
efficient as to the permanent loads (seesummary above). Just as for these last-
mentioned loads,the ys = 1.40relating to the ultimate limit state may, in
exceptional cases,be reduced to ys = 1.25.
Furthermore, special rules have been laid down for certain cases of
superposition of superimposed loads.
Thus,when the effect of wind has to be superimposed upon the effect
of the permanent loads and working loads,the amplification coefficient ys
is allowed to be reduced by 10% (i.e.,ys = 0.90x 1.40= 1.26),provided
that the strength of the structure under these conditions is not less than the
strengthappropriate to the design value ofthe permanent load (i.e.,multiplied
by ys = 1.40)and to the variable superimposed loads increased by 15 %(i.e.,
multiplied by ys = 1.40x 1.15 = 1.61).But if there is no working load,as in
the case ofa chimney,no reduction coefficient should be applied.
Similarly,when the effect of an exceptional superimposed load (suchas an
earthquake) has to be added to the effect of the permanent loads,working
loads and other actions, the amplification coefficient ys is allowed to be
reduced to ys = 1, provided that the 15% increase of the variable super-
imposed loads is retained.Furthermore,the design strengths are allowed to
be multiplied by 1.15. In other words, in this exceptional case,the overall
factorofsafety can permissibly be reduced to 1 for the steel and to 1.50/1.15=
 207
1.30for the concrete,provided that the combination considered really is the
most unfavourable.

2.5 CHARACTERISTIC S T R E N G T H S AND DESIGN

S T R E N G T H S OF STEEL AND CONCRETE
2.5.1 G E N E R A L DEFINITION

The characteristic strength ok of a material is obtained from a statistical

analysis of test results;it is defined by the relation:
[Tk = om(l-k6)
where om denotes the arithmetical mean of the test results,6 denotes the
relative standard deviation (or ‘dispersion coefficient’), and k denotes a
coefficient depending on the number of test results defining om and on the
probability,accepted a priori,ofthe test results being below the characteristic
value ok.
The design strength o* of a material is defined by the relation:
o* = -
1
Ok
Y,
where the strength reduction coefficient y m is a function of the statisticallaws
relating to the mistakes or shortcomingscommitted during construction and
resulting in a reduction of the strength of the cross-sections.
For the steel, for example,this may be due to inadequacy of the cross-
sectional areas of the reinforcement,which in turn may be due either to a
fault in the manufacture of the bars or to a mistake in the bar-bendingshop
or a mistake in fixing the steel on the site.
For the concrete the risks of deviations are much greater than for the
reinforcement:
(a)Whereas the strength of the steel can be accurately determined by
tests on specimens, the tests performed on the concrete may not
correspond to the actual strength of the concrete in the structure.
This actual strength may be affected by: small accidental errors in
batching the cement or the water in a batch of concrete;by handling
and transport;by climatic conditions at the time of concreting (heat,
frost,rain,wind); or by inadequate compaction.
(b)The concrete cross-section may be inadequate on account of the
presence of voids (gravel pockets) or a reduction in the formwork.
dimensions as a result of a mistake in constructing the formwork or of
an accidental deformation thereof.
(c) Short-termtests hardly ever show up the unfavourable effects of loads
of very long duration.
(d) In addition, in many cases failure of the concrete may, much more
frequently than failure of the steel, bring about total failure of the
structure,without any warning signs.
208
For these various reasons the value of ym for the concrete should, in
practice,be 1.25to 1.50times as high as that for the steel.

2.5.2 C H A R A C T E R I S T I C S T R E N G T H A N D DESIGN
S T R E N G T H OF T H E STEEL

The characteristic strength of the steel is the value corresponding to the

probability that,in a normal statistical distribution of the results of tests for
measuring the elastic limit,5 %ofthe results wili be below the specified value.
This probability of 5 %entails:k = 1.64;hence:

The design strength of the steel is derived from the characteristicstrength

by application of the reduction coefficient y, = 1.15 for the ultimate limit
state and the limit state of instability and y, = 1.60 for the limit state of
cracking and the limit state of deformation,i.e.

1.15 1.15
(ultimatelimit state and limit state of instability)
O,,,,(1 - 1.646)
.a*=-&= O
1.60
(limitstate ofcracking and limit state of deformation)

2.5.3 C H A R A C T E R I S T I CS T R E N G T H A N D D E S I G N
S T R E N G T H OF T H E C O N C R E T E

The characteristicstrength of the concrete is the value corresponding to the

probability that,in a normal statistical distribution of the results of tests for
measuring the strength (compressiveor tensile strength), 5 % of the results
will be below the specified value. This probability of 5% entails:k = 1.64.
However,on no account must a relative standard deviation of less than 7%
be considered;hence:

obk = Obrn(1 -1.646)< 0'8850b,

An approximation which yields very much the same value but which
avoids having to calculate the standard deviations consists in taking for the
characteristic strength twice the mean value of half the results below the
median,minus the mean value of all the results.
The design strength of the concrete is derived from the characteristic
strength by application of the reduction coefficient Y b = 1.50for the ultimate
 209
limit state and the limit state of instability, Y b = 1.30 for the limit state of
cracking,and Y b = 1 for the limit state of deformation,i.e.:

(ultimatelimit state and limit state of instability)

(limitstate of cracking)

(limitstate of deformation)

These values are valid for concrete made under supervision on the con-
struction site. The value Y b = 1.50 applicable to the ultimate limit state is
allowed to be reduced to 1.40for closely supervised,carefully batched factory-
made concrete, but it should be increased to 1.60 for concrete made with
only a small amount of supervision.
Besides, the design strengths should be reduced by 10% for members
of small cross-sectional size which are concreted in the vertical position
without special precautions, as it has been found that the quality of the
concrete is liable to vary considerably throughout the height of such members.

2.6 METHOD O F CHECKING THE SAFETY

This check should obviously be carried out in the various limit states en-
visaged, in particular: the ultimate limit state (failure), the limit state of
cracking, the limit state of deformation, the limit state of instability and,
possibly, other limit states to be considered in each particular case. It con-
sists in verifying that the effects of the design loadings do not exceed the
load-carryingcapacity as deduced from the design strengths of the steel and
concrete.
This procedure of checking the safety by means of the C.E.BJC.1.B. semi-
probability method can be symbolically expressed by :

function o f s a n d -
YO Yb

where s k denotes the various characteristic loadings to be taken into account

and R denotes the design strength of the member in the limit state,this being
210
a function of the design value of the mechanical strength of the steel and
concrete.

2.7 RELATION BETWEEN THE C.E.B./C.I.B.

SEMI-
PROBABILITY METHOD AND THE UNESCO SIMPLIFIED
METHOD
The two essential simplifications introduced in the UNESCO method and
valid for the design calculations most commonly encountered consist in:
(a)replacing the characteristic strength of the steel and of the concrete by
the guaranteed minimum strength;
(b)shifting the amplijìcation coeficient y, for the characteristic loadings
to the reduction coeficients for the characteristic strengths of the steel
and the concrete, which thus become respectively:y,. y, and ? b . ys
For practical purposes the simplified calculation is carried out with
‘characteristic loadings’ which are not increased and with ‘characteristic
strengths’ which are reduced by y,. y, and Y b . y, respectively.For the notion
of ‘designstrengths’is thus substituted the notion of ‘basic strengths’accord-
ing to the relation:
basic strength = design strength -
- characteristicstrength
Ys (Ya .Ys) Or ( Y b . Ys)
This overall reduction coefficient (y,. y,) or ( Y b . y,) is therefore comparable
to a singlefactor of safety definedfor each of the materials steel and concrete.
Its values are laid down in Sections 4.3.2and 4.3.3of Part 1 of this Manual,
according to the relations:
Ysieei = Y a .Y s
Yconcrete = Y b .Y s
The corresponding basic strengths are,respectively,equal to:
steel: Ca = ce~

Ysieei

concrete:ab= ~
aó
Yconcrete

In the most frequently encountered cases this UNESCO simplified method

gives results which are close to those obtained with the C.E.B./C.I.B.
semi-
probability method;the margin of error is on the safe side.
In relevant cases the designer should judge whether it is advantageous
to reduce this margin of error and to benefit by the additional saving that
he may achieve by applying the semi-probabilitymethod systematically
and in full.
 3
UNIAXIAL BENDING-
THEORETICAL ANALYSIS

3.1 RECAPITULATION O F THE FUNDAMENTAL DESIGN

ASSUMPTIONS

In addition to the two equations of equilibrium of the forces and moments,

the following basic assumptions are applied in the analysis for the limit
state of a section subjected to simple or composite uniaxial bending:

3.1.1 C O N D I T I O N OF STRAIN COMPATIBILITY

The strains of the elements of a cross-section are assumed always to be

proportional to the distance of these elements to the neutral axis (the‘Navier-
Bernoulli hypothesis’ which states that plane sections remain plane). The
corresponding condition constitutes the strain compatibility equation.

3.1.2 T A K I N G A C C O U N T O F T H E C O N C R E T E IN
COMPRESSION

The stress distribution diagram j o y the concrete in the compressive zone is

assumed to be a rectangle whose width is taken as equal to the basic strength
Obof the concrete and whose depth is a function of the distance x from the
neutral axis to the most compressed face of the member, namely:

$x if x < h (simplebending and composite bending)

.h if x 2 h (eccentriccompression)
x-4h
where h denotes the effective depth of the section (see Figure 3.1).

21 1
212
However,the value ofthe resisting moment due to the compressivestresses
in the concrete (apartfrom the contribution of any compressive reinforce-
ment that may be provided) is limited to the value of the moment (withregard
to the reinforcing bars in tension or located near the least compressed face)
of the forces acting upon the total effectivesection assumed to be subjected

fi: -0.75bx.ü‘b
(rectangular sections
-625~

--+- -I-$--- L
Neutral axis

Figure 3.1

to a uniform stress equal to three-quarters(0.75)of the basic strength O;,

of the concrete.
Finally,the m a x i m u m compressive strain of the concrete in the ultimate
limit state is conventionally taken as equal to 0.2%.

3.1.3 T A K I N G THE T E N S I L E STEEL I N T O A C C O U N T

The basic stress-strain diagram for the steel is derived from the standard
stress-strain diagram by affine transformation parallel to the ascending
straight portion correspondingto Hooke’slaw,in the ratio:
1 - - 0.556
Ysteei 1.80
The tensile strain of the steel in the ultimate limit state is assumed to have
an upper limit value of 1 %.

3.1.4 T A K I N G T H E COMPRESSIVE STEEL INTO A C C O U N T

The compressive strain of the steel of the compressive reinforcement in the

ultimate limit state is determined on the assumption that plane sections
remain plane.Hence the compressive steel stress to be introduced into the
 213
strength calculation is derived from the basic compressive stress-strain
diagram,which is assumed to be similar -subject to reversal of the algebraic
signs-to the basic tensile stress-straindiagram.In normal cases as encoun-
tered in actual practice this compressive stress can be taken as equal to the
basic compressive strength O:, of the steel.
For the purpose of simplifying the present treatment of the subject,the
complete theoretical analysis has been worked out only in the case of
ordinary reinforcing steels having a stress-strain diagram comprising a
definite yield stress range,i.e.,in which the tensile stress remains constant
throughout the stagr in which plastic behaviour develops (‘plastification’).
Similar considerations would, however, also be applicable to cold-worked
steels,provided that,in the plastification stage,due account is taken of the
law of variation of tensile stress as a function of the strains in accordance
with the standard stress-straindiagram for the steel (referenceFigure 3.2,
Part 1 or Figure 6.8,Part 1 of this Manual) or,possibly,in accordance with
its actual experimentally determined stress-straindiagram,if this is known
with sufficient accuracy.

3.2 ANALYSIS OF A S Y M M E T R I C A L SECTION OF

ARBITRARY SHAPE
3.2.1 D E T E R M I N A T I O N OF THE T Y P E O F FAILURE

It is known that,depending on the percentage of reinforcement,the ultimate

limit state (limitstate of failure)is reached either by crushing of the concrete
in compression (for high steel percentages) or by yielding of the tensile
reinforcement (fornormal steel percentages). These two types of failurewill
occur simultaneously for a limit value of the percentage of reinforcement for
which it is easy to determine a corresponding limit value ofy/h.
When these limit values are known,it is possible to predict the nature of
the failure.

Expression Of y/h As A Function Of The Section Properties

And External Loadings

In the most general case of a section subjected to compression and bending,

the general equation of equilibrium of the forces can,on the assumption that
the tensile reinforcement has attained its yield stress at failure of the member,
be written as follows:
-
N = O;, j:b. dy-A. a,+A’. a:,

where A’.O:, denotes the contribution of the compressive reinforcement

(ifany).
214
The integral can be replaced by a conventional expression :

lOyb.dy = b, .y,

where b, denotes the fictitious width of a rectangular section equivalent to

the section considered.We thus obtain:
Y - N + A . 0,-A’.<
- (3.2)
h- $.b,.h
It is merely necessary to compare this value of of y/h with the limit value
ofy/h in order to determine whether or not failure will occur as a result of
yielding of the steel.

Expression For (y/h),,,,,

The limit value of y/h is reached when the concrete attains its ultimate com-
pressive strain EO and the reinforcing steel simultaneously attains the tensile
strain E, corresponding to its yield stress oe.Hence, on applying the strain
compatibility equation :

(3.3)

i.e.:
(3.4)

O n replacing E, and E; by their respective values E, = 2.1 x lo6 bars and

EO = 0.2%,a simple expression for (y/h)limit
is obtained:

0.75
(i),,,¡,= + 1 2.38. oe
(3.5)

This expression, in which o, should be expressed in bars (kg/cm2),is valid

irrespective of the geometrical shape of the section and depends only on the
elastic limit of the steel.

Practical Determination Of The Type Of Failure

T o determine the type of failure,it is merely necessary to compare the value
of y/h obtained from equation 3.2 with the value of (y/h)limit obtained from
 215
equation (3.5). O n application of this latter expression the values shown in
Table 3.1 are found:
Table 3.1

CT, (bars or kg/cm2) (Y/hhimii

2 100 0,500

2 500 0.470
3000 0,438
3 500 0.409
4000 0,384
4 500 0.362
5 O00 0.342
5 500 0.325
6000 0.309

If y/h does not exceed (y/h),imi,,it means that failure will indeed occur by
yielding of the steel.
Otherwise failure will be due to crushing of the concrete in the compressive
zone.

3.2.2 D E T E R M I N A T I O NO F THE FAILURE MOMENT

Expressions For The Equilibrium Equations

In the most general case of a section subject to compression and bending,
the general equation of equilibrium of the moments can be written as
follows :
-
so
N.e = 3, b(h-y)dy+A.%.h

which can also be written:

(3.6)

N.e = $.b,.y(h-hy)+A'.<.h'
where 6 denotes a coefficient (less than unity: 6 < 1) corresponding to the
relative ordinate,with respect to the most compressed edge, of the centroid
of the compressive forces.This coefficient is equal to f in the case of a rect-
angular diagram and to i in the case of a parabolic diagram.
O n combining this general equation of equilibrium of the moments
with the equation 3.2 derived from the general equation of equilibrium of
the grces, two basic equations for the analysis of the section are obtained:

J := N+ A- .O,-A'. 3
o;. b, .h
(3.2)

I fi = N .e = Zb.b,. hZ .?
h
+A'.3,.h' (3.7)
216
With these two equations it is therefore possible to calculate the failure
moment directly,but they are valid only if

i.e.,if failure occurs by yielding of the steel.

In the contrary case,i.e.,if

it is necessary first to estimate the stress on of the tensile reinforcement,by

bringing in the strain compatibility equation in addition to the two general
equations of equilibrium.Thus the three following equations are obtained,
which then constitute the basic equations for the analysis of the section:

II y - 0.75.
h
EO
EO
+-an
En

Ñ = og.b,,,.y-A.o,+A'.Ti:,]
M = N . e = %.b,.hZ.x h
1 (3.8)

(3.9)
(3.7)
With these three equations it is therefore possible to calculate the failure
moment directly,in the case where

(i)
2 (i)iirnii
i.e., failure occurs by crushing of the concrete.
However, the problem is complicated by the existence of an upper limit
value of the bending moment.

Expression For Upper Limit Of The Moment

According to the fundamental assumptions on which the method is based,

we know that the maximum value of the moment that can be developed by a
concrete section is equal to:

J:
M, = 0.75.% b(h -y) dy
Furthermore the value of (y/h) for which the moment reaches its upper
(3.10)

limit is obtained by equating the general expression for the moment with
the upper limit value thereof,i.e.:

(3.11)
 217
It is therefore essential to compare the value of y/h not only with (y/h)limit
but also with (y/h),,,,, calculated from equation 3.11.
Thus, with the aid of equation 3.11 we can determine (y/h)upper. For the
ordinarily encountered case of rectangular sections (or sections which can
for practical purposes be treated as rectangular) we obtain :

(i)"ppe.=
N o w Table 3.1shows that for all the steels normally employed in reinforced
concrete (a,< 2 100 bars) the value of (y/h)limitis always less than 0.50.
Hence it can be inferred that for all rectangular sections (or sections which
can be treated as rectangular):

(i) limit < (;)upper

This greatly simplifies the analysis in practical use.
However, if the more general case of a symmetrical section of arbitrary
shape is considered,it is necessary, for defining the appropriate procedure
for analysing the section in the ultimate limit state,to compare the value of
y/h with and also with (y/h)upper. The procedure is as shown below.

Recapitulation:General Procedure Of The Analysis

First Case (Y/h)iimit < (Y/h)upper

As indicated above, this is more particularly the case of all rectangular
sections or sections which can for practical purposes be treated as rectangular.
(a)If
-< -
1 (i) (i)limit < upper

use the equations 3.2 and 3.1

(Case of failure by yielding of the steel,with no upper limit applied to
the moment)
(b)If

use the equations 3.1-3.9.

(Case of failure by crushing of the concrete, with no upper limit
applied to the moment)
(4If
Y

use equation 3.10.

(i) (i)
iimit < upper <
218
(Case of failure by crushing of the concrete, with upper limit applied
to the moment)

use equations 3.2 and 3.7.

(Caseof failure by yielding of the steel,with no upper limit applied to
the moment)
(b)If

use equation 3.10.

(Case of failure by yielding of the steel, with upper limit applied to the
moment)
(cl If

use equation 3.10

(Caseof failure by crushing of the concrete, with upper limit applied
to the moment)

3.3 A N A L Y S I S OF A RECTANGULAR S E C T I O N

3.3.1 GENERAL ANALYSIS

Determination Of The Failure Moment

General Expression For The Failure M o m e n t

For a rectangular section the relations in Section 3.2.2 become,with b, = b

and 6 = 05:
(a)In the case of failure by yielding of the steel:

(3.12)

I = N . e = &.b.hZ.x h +A‘.<.h’ (3.13)

 219
(b)In the case of failure by crushing of the concrete:

II
y 0.75 .EO
h-
EO
oll
+-
Ea

J
N = Zb.b.y-A.o,+A’.3:,
M=N.e=Zb.b.h2x
h
I
.;fi (i) limii
(3.14)

(3.15)
(3.13)

(c) Taking no account of the compressive reinforcement A’,the expression

for the upper limit value of the moment is:
M p= 0.75X O . ~ O X Zb.~ h. 2 = 0.3755.b . hZ (3.16)
Introducing the notion of ‘relativemoment’,we can write:
-
p =- M P - 0.375 (3.17)
oL.b.h2-
Furthermore,the value (y/h)upper, for which the moment reaches its upper
limit value,can readily be calculated from equation 3.11,giving:

( =
(3.18)

Expression For The Relative Moment As A Function Of yfh

It should be noted first of all that,in the general expression for the failure
moment (equation 3.13), the first term represents the contribution of the
compressivezone of the concrete,while the second term corresponds to the
contribution made by the compressive reinforcement.Obviously,this second
term,which remains practically invariable (subjectto certain constructional
conditions being satisfied), cannot in any way alter the overall pattern of the
phenomenon.For this reason only the relationship governing the variation
of the first term as a function of y/h will be considered,for which purpose
the first term can,on introducing the relative moment,be written as follows
(see Figure 3.2):

i( h)
p = - 1-0’50- G0.375 (3.19)

Designing And Checking The Reinforcement

Data
The following are assumed to be known :
(a)the external loadings, characterised by the direct force and the
eccentricity e ;
220
(b)the geometrical properties of the section,i.e.,b, h and h';
(c) the basic compressive strength obof the concrete;
(d)the basic tensile strength Za and basic compressive strength 3 of the
steel.
O n the assumption that the cross-sectionalarea A' of the compressive
reinforcementis known a priori or has been determined by a suitable method,

P
Reiat ive
moment A
Upper limi1 moment = 0.375

0.300-

O O1 0.2 0.3 0 4 0.5

Figure 3.2

the value of y/h can be calculated by solving the second-degreeequation

3.13ofwhich it is the lesser root:
2(1 -0.50;) = N .e-A'. <.
h
h 6.b.h2
Y N.e-A'.if,.h
i= 1 4 1 - 2 6.b.h2
From the value of y/h can be determined the cross-sectionalarea A of the
tensile reinforcement. However,the calculation procedure may differ accord-
ing as y/h is below or above the upper limit value 0.50.

First Case: y/h < 0.50

If
 22 1
then the strength of the section is limited by the risk of yielding of the steel,
and the steel area A can be determined from equation 3.12 in the following
form :
(y/h).&,.b.h-(N-A’.<)
A= - (3.20)
On

If

then the strength of the section is limited by the risk of failure (crushing)of
the concrete, and the steel area A can be determined by application of
equations 3.14and 3.15.
0’75-(y/h)
o, = E,. EO .
Y/h
(y/h).ob.b . h-(N-A’. O,)
A=
00

Second Case: 0.50<ylhG0.75

In this case, in accordance with the fundamental design assumptions,in

establishing the ‘truncated’rectangular stress diagram for the analysis of the
section only a reduced value of the basic strength of the concrete,obtained
by decreasing this basic strength in a proportion ranging from 1 to 0.75,
should be taken into account,i.e.:
(3.21)

But since y/h, exceeding the value 0.50,itself depends upon the value of
the tensile stress o, of the steel (which stress is below the basic strength
,)?i it is necessary to associate equation 3.21 with the strain compatibility
equation in the following form:
(3.14)
h oa
EO+-
Ea
and to use, for determining the cross-sectionalarea A of the tensile re-
inforcement,the relation:
Y .O;.b . h -(N-A’.O,)
A . na = -
h
0.75 -
A.0, = ~ .O;. b . h- (N-A’. O,)
2 -Y/h
222
This expression is valid for
Y X
0.50<-<0.75or $<-< 1
h h
i.e.,for all sections in which the neutral axis is located within the section or,
in other words,for the whole range of composite bending.
When y/h approaches the value 0.75,the expression for A assumes the
indeterminateform O/O,but since the numerator is an infinitely small quan-
tity of second order and the denominator is an infinitely small quantity of
first order,this expression in reality tends to zero,which does indeed repre-
sent the mechanical behaviour of the section.
When y/h exceeds the value 0.75,this expression for A converges to a
negative result, which is normal, since the concrete section then comes
entirely under compression (notensile zone) and the reinforcement A then
represents,not tensile reinforcement,but the ‘leastcompressed’ reinforce-
ment. The result obtained remains significant,for it merely interprets the
conditions of equilibrium, compatibility and upper limit.Nevertheless,
for convenience in the practical design and analysis of sections,it is pre-
ferable,in the case of eccentric compression,to adopt a more appropriate
expression for the conditions of equilibrium,compatibility and upper limit,
as explained in the following.

Third Case :0.75< y/h < 1

In Figure 3.3, let e denote the eccentricity of the direct force (normal
force) N with respect to the centroid of the least compressed reinforcement

I J I
I I Neutral
I axis
--. _.-.-.-.I.-
O O
Strains St resse5
Figure 3.3

A.With due regard to the upper limit condition,the conditions of equili-

brium of the forces and moments and the strain compatibility conditions
can be written as follows:
Equation of equilibrium of forces:
N = ab.b.y+A.oL+A’.< (3.23)
 223
Equation of equilibrium of moments:
N .e = 0.375.ab.
b . h2 +A' .a,.
h' (3.24)
or,on considering the eccentricity (h'-e) of the direct force N with respect
to the centroid of the most compressed reinforcementA':
Ñ .(h'-e) = 0.375. O;, .b . h'" + A .a:.h' (3.25)
Strain compatibility equation:
_ -
X - EO (3.26)
h &-(o:lE,)
or,taking account of the upper limit condition

(3.27)

(3.28)

The expressions for determining the steel cross-sectionalareas A and A'

are determined directly from the equilibrium equations 3.24 and 3.25:

A' = N .e-0.375.ob.
b . h2
0;.h
N(h' -e) -0.375. o;,
.b .h"'
A=
3,.h'
For analysing the section and calculating its load capacity (ultimate
strength) it is necessary to determine y/h. This can be done by solving the
equilibrium equation 3.23 in which the unknown stresses o: and ob are,
for this purpose, respectively replaced by equations 3.27 and 3.28 in terms
of ylh. In this way a second-degreeequation in ylh is obtained,one of the
two roots of which is between 0.75and 1. The values of o:, ob and N are
obtained from it.
It should also be noted that equations 3.24and 3.25can conversely be used
for determining the direct forces N, and N 2 corresponding to a given
distribution of the loadings and producing maximum compression at one
or the other edge (Ñ, < N2):
- 0.375.O;, .b . h2 +A' .O: .h'
N, = (3.29)
e
- 0.375.$. b . h f f 2 + A .0;.h'
N 2= (3.30)
h'-e
224
Similarly,if fil or N2is given,the permissible limit eccentricity can be
defined by means of these expressions :
0 . 3 7 5 . 6 . b . h2 +A’.ZL.h’
e< (3.31)
Nl
0.375 .Ob.b .-h”’ +A .O:, . h‘ (3.32)
h‘-e<
N2
Also, with these expressions the formulae applicable to concentric com-
-
pression,conceived as a limitcase,can be established.In that case Ñ,= N 2
and the point of application of the resulting loading coincides with the
centroid of the section.It follows that e and (h’-e) are linked by the two
relations:
--e - 0.375 .i?; . b .h2 +A‘ . ¿?
.h’
:
h’-e 0.375.6 . h
b . h’I2+A.¿?L.
From this the value of e can be determined,which can then be substituted
into equation 3.29 for NI,whence the limit value N of the direct force for
concentric compression is obtained:

N = Ñ l= N2 = 0.375.6.b.- h 2 + h ” 2 + A$!+A,,
,
h’
Hence :
N = 0.75 .ob. b . h,+A .O:,+A‘ .O:, (3.33)

3.3.2 A N A L Y S I S OF A R E C T A N G U L A R S E C T I O N F O R
SIMPLE BENDING

Design Of The Section

First Case: The Effective Depth Is A Predetermined Value

This is not a desirable case,for it often results in uneconomical design. It
may nevertheless occur that architectural or operational requirements
necessitate a very low construction depth for a flexural member.
The design procedure consists in first calculating the relative ultimate
moment (relativefailure moment)

and comparing it with the upper limit moment 0.375 : if < 0 3 7 5 , then there
is no point in providing compressive reinforcement;
it ji>0.375, then the designer may either increase the width b of the flexural
 225
member concerned (e.g.,
by giving it a wider compressive flange)or provide
compressivereinforcement with a cross-sectionalarea:

(a) Rectangular section without compressive reinforcement

If the relative ultimate moment is known, the relative depth y/h of the
rectangular compressive stress diagram of the concrete can be derived
from it:
= !(1-0.50$
h
*
therefore:
= 141-2jj)
h
Two possibilities may then occur:

First possibility: -<

i.e.:
Y 075
-< (o,in bars)
h 1 +2.38x 10-40,
In this case the ultimate resisting moment of the section is limited by the
risk of yielding of the steel and is defined by the condition:

ITa =- Ge
Ga = -
Ysieei

The cross-sectional area A of the tensile reinforcement can readily be

determined:
since
!!--
A .O,
h -b . h . 6
we obtain:

Second possibility: ->

($limit

i.e.:
Y 0.75
-> (o,in bars)
h 1 +2.38 x 10-4~,
226
In this case the ultimate resisting moment of the section is limited by the
risk of crushing of the concrete before the tensile reinforcement has been
utilised to its full resistance capacity:o,<O,.
The tensile steel stress oato be introduced in the calculation is defined by
the relation:
Y -
_ 0.75.EO
h - EO + [(oa/Ea) x ~steel]
hence:

or,more specifically:
J(l-2j4-i
= (3- )
1 x2333 bars =
1-41-29
x 2 333 bars

The cross-sectionalarea A of the tensile reinforcement can be directly

determined from the stress oa according to equation 3.15
Z b . b . y = A.0,
whence:
A = b.h.--obY
Ga h
(b)Rectangular section with compressive reinforcement
Assuming that the difference between the relative ultimate moment ji
and upper limitmoment 0.375can be taken up by compressivereinforcement
with a cross-sectionalarea:
Gbh2 -
A’= r-(/.i-0.375)
o: h’
it remains to design the section for the relative upper limit value 0.375 of
the moment, to which corresponds the upper (limit)value y/h = i of the
relative depth of the compressivestress diagram of the concrete.
N o w it is known that for all types of steel normally employed in reinforced
concrete construction this upper value y/h = 3 exceeds the corresponding
limit value,i.e.,

In other words, in all normally encountered cases the ultimate resisting

moment of the section is limited by the risk of crushing of the concrete
before the tensile reinforcement has been utilised to its full capacity:o,<O,.
As before,this stress o, is defined by the relation:
= 1)
(x-2333 bars
i.e.,for y/h = 3,we have:o, = 1 167 bars
 227
The cross-sectionalarea A of the tensile reinforcement is determined from
equation 3.17 for the equilibrium of forces
A=-x
b.h a:, a:,
2 1 167 barsfA’ 1 167 bars
It is at once apparent that this section is uneconomical and that the
arbitrary limitation of the effective depth h results in wasteful use of steel.
This can readily be shown by means ofa worked example.Suppose:jï = 0.450,
o, = a; = 4O00 bars, ¿Fa = 0,= 2 222 bars, Zó= 350 bars, 3,= 152 bars,
h’ = 0.9h. The total quantity of steel required is then:
76 b.h+- 3 3 8 9
A+A’ = - 152 x!!!x0.0’75b.h
1167 1167’m 9
= 6.512~10-~+(2.904~0.571~ 10-2)b.h
= 8.17 x low2b . h
O n the other hand, if the designer could increase the effective depth by
50%, he would obtain: = 0.200and y/h = 0.225, which value is not only
below the upper value (y/h)upper, but also below the limit value (y/h)ii,it,
so that it would not only be possible to dispense with the compressive
reinforcement,but also to utilise the tensile reinforcement to its full capacity.
Then,other things being equal,the quantity of steel required would be:
A = 153 x x b x 1.5h = 2.30 x 1 O P 2 b .h
So by increasing the effective depth by 50% (and therefore increasing the
quantity of concrete by 50%) it is possible to effect a saving of more than
70%in reinforcing steel.

Second Case: The Designer Is Free’ToChoose T h e Eflective Depth

In this case considerations of economy will obviously require the designer

to remain not only below the upper (limit)value (y/h)Fqp,,but also below the
because only then can he fully utilise the capacity of the
limit value (y/h)li,,,it,
steel and take the best possible advantage of designing the section for the
limit state.
Table 3.2

6, (YIh),imi1 Piimi, f .Piimii Ylh

2 100 0500 0,375 0,125 0134

2 500 0.470 0.360 0.120 0128

3 O00 O 438 0,342 0114 0121
3 500 0409 0325 0.108 0115
4O00 0.384 0.310 0.103 0.109
4 500 0.362 0296 0.099 0104
5 O00 0.342 0.283 0094 0.099
5 500 0.325 0.272 0.091 0095
6 O00 0,309 0,261 0.087 0.091
228
In the absence of design charts or graphs a reasonable and convenient
solution(butnot always the most economical solution) consists in arbitrarily
basing oneselfon one-third ofthe limit moment (see Table 3.2).
The effective depth h and the cross-sectionalarea of the tensile reinforce-
ment A can then directly be determined:

A=b.h.:.-6 Y
Ga h

For example: suppose that for a moment M = 50000kgm we have

= 350 bars,ob= 152 bars,b = 0.60m;
ce = 4O00 bars,Ca = 2 222 bars,ob
we then obtain :

h= J(0.1035xx 11520 6x 60)= 73cm

152
A = 60x 73 x -x 0.109= 32.6cm2
2 222
i.e.,a geometrical percentage of 0.75%.

Analysis Of The Section

The analysis of a rectangular section with or without compressive reinforce-

ment is carried out by applying the general equations 3.12-3.18in which
the direct forceis substituted as of zero value,i.e.N = O.
For practical purposes the procedure is as follows:

First Attempt
Calculate the relative depth y/h of the rectangular diagram by means of
the formula:
y A.ZO-A'.CL
_
h-
- Zb.b.h (equilibrium equation)

If the value obtained for y/h is below the value of (y/h)ii,it

corresponding
to the grade of steel used,we have the case where failure is by yielding of the
steel,with no upper limitapplied to the moment and we must apply equation
3.13,which immediately gives the resisting moment :

= 3,.b . h2 .?
h
! +A'. O:, .h' (momentequation)
 229
If not, then it means that failure occurs as a result of crushing of the
concrete,not yielding of the steel.It will then be necessary to make a second
attempt.

Second Attempt
Calculate the relative depth y/h of the rectangle diagram by solving the
following set of equations:
y
_ - 0.75.EO (compatibilityequation)
h - EO +(oa/EJ
y - A.a,-A'.Z,
- (equilibriumequation)
h- Zb.b.h
In this way we find the value of oa (lowerthan O,) and a fresh value of
Ylh.
If the value thus obtained for y/h, while exceeding ( y / Q i m i t ,is below
the value (y/h)upper= 3,we have the case where failure is by crushing of the
concrete,with no upper limit applied to the moment) and we must then,
as in the previous case,apply equation 3.13, which immediately gives the
resisting moment :
fi=%.b.hZ.? +A'.%.h' (momentequation)
h
If,on the other hand,the value obtained for y/h exceeds the value (y/h)upper
= t,the upper limit must be introduced and a third attempt made.

Third Attempt
W e now have the case where failure is by crushing of the concrete, with
upper limitapplied to the moment). This moment can be directly calculated:
M = 0.375.b.h2.Zb+A'.Za.h'

3.3.3 ANALYSIS OF A R E C T A N G U L A R SECTION F O R

COMPOSITE B E N D I N G W I T H COMPRESSION

Design Of The Section

First Case: The EffectiveDepth Is A Predetermined Value
Just as in the case ofsimple bending,the design procedure consists in first
calculating the relative ultimate moment
-
j¿=-
N.e
ob.b . h2
and comparing it with the upper limit value of the moment,i.e.,0.375:
230
if ji<0.375,then there is no point in providing compressive reinforcement;
if jï> 0.375,then the designer may either increase the width b of the flexural
member concerned (e.g.> by giving it a wider compressiveflange)or provide
compressivereinforcement with a cross-sectionalarea:

When the compressive steel area A’ is known,the value of y/h can be

calculated by application of equation 3.13:

(3.13)

The value obtained for y/h will then depend upon the value of

N .e -A’.K .h - - Za.A’.h
-
6.b.h2 ’-Oh b . h2
.

If the value of,uis below the upper limit 0.375 or if the difference (p-0.375)
is completely equilibrated by appropriate compressive reinforcement,
the value obtained for y/h will be below 0.50and the strength of the section
will be limited either by the risk of failure of the tensile reinforcement if

or by the risk of failure (crushing)of the concrete if

In the alternative case,i.e.,if the value of exceeds the upper limit 0.375
while the difference (g-0.375) is not completely equilibrated by the com-
pressive reinforcement,the value obtained for y/h will exceed 0.50and the
upper limit condition will have to be applied (0.75. 3,< ob < ob).

i,(;)
First possibility: -<
limi1
G0.50

In this case the ultimate resisting moment of the section is limited by the
risk of yielding of the steel;it is defined by the condition:
 23 1
The cross-sectional area A of the tensile reinforcement can readily be
determined:
(y/h)& .b .h(N .A’. ¿fL)
A=
6
Second possibility: (i) limit
6i60.50

In this case the ultimate resisting moment of the section is limited by the risk
of crushing of the concrete before the tensile reinforcement has been utilised
to its full capacity (0,6i?,):
0.75-(y/h)
O, = E,. EO.
(Y/h) (3.14)
The cross-sectionalarea A of the tensile reinforcement can be directly
calculated :
(y/h).%.b. h-(N-A’.&) - y/h (y/h).ab .b . h -(N-A‘.<)
A=
Ca 0,75-(y/h)’ E,.EO
(3.15)
Third possibility: 0.50<!<0.75
‘ h
In this case the ultimate resisting moment of the section,determined by the
risk of crushing of the concrete,is limited to the upper limit value of the
moment.Then,in accordance with the fundamental assumptions,a reduced
value of the basic strength Obof the concrete should be taken into account in
determining the ‘truncated’rectangular stress distribution diagram. This
reduced value is obtained by reducing the basic strength in a proportion
varying from 1 to 0.75.Hence:

But since the upper limitmoment is reached before the full capacity of the
tensile reinforcement has been utilised,we also have,in accordance with the
equation of equilibrium of forces:
(y/h).üb.b.h-(N-A’.O,)
0, =
A
The cross-sectionalarea A of the tensile reinforcement is therefore defined
by the relation:
Y/h b . h- [2-(y/h)] (N-A’.3)
075.Zb.
A = [2-(Y/h)l -(Y/h)l E,. EO
This value of A tends to zero when y/h approaches 0.75,i.e., when x/h
approaches the value 1 or, in other words, when the neutral axis of the
bending couple approaches the centroid of the tensile reinforcement.This
232
limit constitutes the borderline between composite bending (section parti-
ally in compression) to eccentric compression (sectionentirely in compres-
sion).
Fourth possibility: 0.75<-<
Y 1
‘h
The sectionis loaded in eccentriccompression and is subjected to compressive
stress over its entire area.It should be designed as such in accordance with
Section 3.3.4.

Second Case: T h e Designer Is Free To Choose T h e Effective Depth

In this case it is evident that,for reasons of economy,the designer should

remain,as far as possible,not only below the upper (limit)value (y/h),,,,,
but also below the limit value (y/h),imit. In the absence of design charts or
graphs a reasonable and convenient solution consists in arbitrarily basing
oneself-just as in the case of simple bending-on one-third of the limit
moment (accordingto Table 3.2).
The effective depth h of the section and the cross-sectionalarea A of the
tensile steel can then directly be determined :

-I

A = b.h.:.- ‘b Y
‘a h
Another case which is very frequently encountered in designing for
composite bending is that of the symmetrically reinforced section (A = A’).
If failure occurs as a result of yielding of the tensile reinforcement,the
equilibrium and compatibility equations will then be :

(equilibrium of forces)

Ñ .e = ob.b .h2 .f + A .o; .h‘ (equilibrium of moments)

h

I----h’ 4 -y
- h 3’h
da= -6, (compatibility)
I-- 4 -Y
3’h
If one-thirdof the limit moment is adopted,as above,the effective depth
h can at once be written down.The cross-sectionalareas of the tensile and
 233
the compressive reinforcement can then be determined with the aid of the
set of compatibility and equilibrium equations by elimination of y/h and o;.

Analysis Of The Section

The analysis of a rectangular section with or without compressive re-

inforcement is carried out by applying the general equations 3.12-3.18.
For practical purposes the procedure is as follows:

First Attempt
Calculate the relative depth y/h of the rectangular diagram by elimination
of N from the two equilibrium equations 3.12 and 3.13:
13.12)

N . e = Zb.b.h2.1
h
(3.13)

If the value obtained for y/h is less than ( ~ / h ) ,corresponding

~~~, to the
grade of steel used,we have the case where failure is by yielding of the steel,
with no upper limit applied to the moment and we must apply equation
3.13, which immediately gives the resisting moment N .e, i.e., the resisting
direct force (if its eccentricity is predetermined) or the permissible eccen-
tricity (if the magnitude of the direct force is predetermined).
If not, then it means that failure occurs as a result of crushing of the
concrete,not yielding of the steel.It will then be necessary to make a second
attempt.

Second Attempt
Calculate the relative depth y/h of the rectangular diagram by elimination
of N and 6, from the three compatibility and equilibrium equations 3.13-
3.15

(3.14)

N = %.b.y-A.oa+A’.% (3.15)

N . e = Zb.b . h2.J (3.13)

h
If the new value obtained for y/h, while exceeding (~/h),¡~¡~,is below the
value (y/h),,,,, = 0.50,we have the case where failure is by crushing of the
concrete,with no upper limit applied to the moment and we must then,as in
the previous case,apply equation 3.13,which immediately gives the resisting
234
moment Ñ .e,i.e.,the resisting direct force(ifits eccentricityis predetermined)
or the permissible eccentricity (if the magnitude of the direct force is pre-
determined).
If,on the other hand,the value obtained for y/h exceeds (y/h)upper= 0.50,
the upper limit must be introduced and a third attempt made.

Third Attempt
Calculate the relative depth y/h of the rectangular diagram by elimination
of Ñ,O, and o: from the four equations of compatibility,equilibrium and
upper limit (3.14,3.34,3.35 and 3.21):
y - 0.75.EO
_ (3.14)
h - E O +(o,/&)
N = r&.b.y-A.o,+A'.& (3.34)
Ñ.e = 0.375.6.b.h2+A'.&.h (3.35)
075
(3.21)
Ob = (y/h)[2-(y/h)] ob
'

If the new value obtained for y/h, while exceeding 0.50,is less than 0.75,
we have the case where failure in compound bending by crushing of the
concrete, with upper limit applied to the moment, and equation 3.35 im-
mediately gives the resisting moment N .e, i.e., the resisting direct force
(if its eccentricity is predetermined) or the permissible eccentricity (if the
magnitude of the direct force is predetermined).
If,on the other hand,the new value obtained for y/h exceeds 0.75,we have
eccentric compression,and the section,which is then subjected to com-
pressive stress over its entire area,should be analysed in accordance with
Section 3.3.4.

3.3.4 ANALYSIS OF A R E C T A N G U L A R SECTION F O R

ECCENTRIC COMPRESSION

Design Of The Section

First Case: The EffectiveDepth Is A Predetermined Value

The cross-sectionalareas A and A' of the least and the most compressed
reinforcement, respectively, should be determined, in accordance with
Section 3.3.1,by means of the two equations:

A' = N .e-0.375.6. b.h2 (3.24)

3,. h
 235
Ñ(h‘-e)-0.375. O;, . b . h”
A= (3.25)
6á.h

Second Case: T h e Designer Is Free To Choose T h e Effective Depth

In this case a reasonable and convenient solution consists in equilibrating

the moment Ñ .e by the concrete section alone.According to equation 3.24
this gives:
Ñ.e=0375.0;,.b.h2
Hence:
Ñ.e
= dû.375. ob.b
A‘ = O

So now only the proportion [

1 -(e/h’).(h2+h”’/h2)] of the direct force
N remains to be equilibrated by the least compressed reinforcement.

Analysis Of The Section

The analysis can be done directly by means of equation 3.24 (equilibrium

of moments).
The equations 3.29 and 3.30 can be used for determining the direct forces
NIand Nzcorrespondingto a given disposition ofthe loadiq and producing
the maximum compression at one or the other edge (Ni <IV2):
- 0.375.¿?b. b. h2+A‘. O,.h
N, = e
(3.29)

- 0.375.ob. b . h”’+A. 8 .h
N 2= (3.30)
h’-e

Alternatively,if NIor Nzis predetermined,the permissible limit value of

the eccentricity can be determined:

e<
0.375. Zb.
b. h2+A’. &. h‘
(3.31)
NI
0.375 .zb,b . h“’ +A .¿.?
h‘ :
h-e (3.32)
N2
236
3.3.5 ANALYSIS OF A R E C T A N G U L A R SECTION FOR
C O N C E N T R I C COMPRESSION

Design Of The Section

The design procedure is based on the application of the equilibrium equation

3.33:
N = 0.75.Zb.b.h,+A.<+A’.% (3.33)
W h e n the minimum percentage of longitudinal reinforcement required
has been adopted,the concrete section is defined by:

b a h ,=
N -( A+A’)%
0.75.Z b

Analysis Of The Section

This is done directly by application of the above equation 3.33.

 4
UNIAXIAL BENDING-
PRACTICAL DESIGN
CALCULATIONS

4.1 PREAMBLE
The object of this chapter is to show,by means of a series of examples,how
the general theoretical analysis developed in Chapter 3 can be applied to
the various problems that are likely to be encountered by the designer in
actual practice.
For this purpose the author has prepared a number of graphs which
enable the calculations to be simplified and the required solution to be found
very rapidly.

4.2 PROPERTIES OF THE MATERIALS

4.2.1 P R O P E R T I E S OF THE C O N C R E T E

Ten particular values of the basic strength abof the concrete will be con-
sidered.For each of these the corresponding ultimate limit strength %o for
concentric compression,taken as equal to 0.75Zb, will be calculated.These
values are given in the Table 4.1(expressed in bars):
Table 4.1
Zb 40 50 60 70 80 90 100 120 140 160
-
4. 30 37.5 45 52.5 60 67.5 75 90 105 120

4.2.2 P R O P E R T I E S OF THE S T E E L

Five types of steel, corresponding to the grades most commonly used in

reinforced concrete,will be considered.These types are defined in Table 4.2,
237
238
which in each case gives the reference value of the elastic limit oe,the basic
tensile strength O,, the basic compressive strength 3,,and the maximum
basic tensile stress Ziocorresponding to a limit strain of 1 % (Section 6.1.1,
Part 1 of this Manual).
The stress 0,is obtained by dividing oeby the overall factor of safety 1.80.
The stress 3, represents the compressive stress that can permissibly be
applied to the reinforcement A and A’.These values correspond to a com-
pressive strain (shortening)of 0.2%.
Table 4.2 Types of steel and basic stresses (in bars)
- - - -
Type of steel ce 00 O10 4
(1)ordinary steel 2 400 1335 1335 1335
(2)ordinary steel 4 O00 2 225 2 225 2 225
(3)cold-worked steel 4 O00 2 225 2 425 2 110
(4)cold-worked steel 5 O00 2 780 3 025 2 585
(5)cold-worked steel 6 O00 3 335 3 625 2 960

For each value of the strain E, defined in the ultimate limit state, Figure
4.1indicates the basic stress oa of each type of steel. This stress has been
obtained from the actual stress-straindiagram of the steel by affine trans-
formation parallel to the ascending straight portion corresponding to
Hooke’s law, in the ratio 0.556 (see Section 6.1.1,Part 1). The maximum
strain E, is limited to i%, this value being considered the maximum per-
missible plastic strain in the ultimate limit state.

4.3 SIMPLE UNIAXIAL BENDING

4.3.1 A R B I T R A R Y S E C T I O N S Y M M E T R I C A L W I T H RESPECT
T O T H E P L A N E OF B E N D I N G

Recapitulation Of The Fundamental Assumptions :

The analysis of the ultimate limit state for simple bending is based on the
following assumptions :
(a) The longitudinal strains in a section are proportional to the distances
to the neutral axis.
(b)The tensile strength of the concrete is neglected.
(c) The maximum compressive strain E; of the concrete in the ultimate
limit state is taken as 0.2%.
(d)The compressive stress distribution diagram of the concrete is assumed
to be a rectangle whose width is taken as equal to the basic strength
aband whose depth y is taken as equal to,&$, times the depth x of the
zone in which compressive strain (shortening)occurs.
(e) The value of the ultimate bending moment has an upper limit M,
which it cannot exceed.This upper limit is equal to the static moment
of the concrete section (with respect to the centroid of the tensile
239
240
reinforcement)multiplied by 6 of the basic strength of the concrete
(0’750b= 0bo).
(f) The basic stress in the tensile reinforcement is indicated as a function
of E, by the curves in Figure 4.1.The values of E, are linked to the
values of y by the relation :
E,- h- 1.33~
(4.la)
0002 - 1.33~
Figure 4.2gives E, as a function of ylh. The strain is limited to 1 %,
in
accordance with the condition laid down in Section 6.1.1,Part 1.

1~0%
0.9
0-8
0.7
0.6
0.5
04
0.3
o .2
0.1
O

Figure 4.2

A consequence of introducing this limit value for the steel strain is that,
for very low percentages of reinforcement,the ultimate limit state is
reached before the concrete strain attains a value of 0.2%. The reduc-
tion in the magnitude of the maximum moment that this limitation of
the steel strain causes is negligible for practical purposes, however.
(g)The basic stress in the compressive reinforcement corresponds to a
strain of 0.2%and is indicated in Section 4.2.2.

Problem No. 1

Data: the geometrical cross-sectionof the member, the cross-sectionalareas

of the reinforcement, the properties of the materials, the bending
moment M.
Determine whether the structural safety of the member is ensured.
 241
Let S denote the static moment of the concrete section,with effective depth
h, with respect to the axis u-u passing through the centroid of the tensile
reinforcement (Figure 4.3). The maximum bending moment that can be
equilibrated by the compressive reinforcement and the concrete is:
M,= 0.75.S.Zb+A’.Zá.h (4.1b)
The first condition of safety can therefore be written:
MGM,
It now remains to be checked whether the reinforcement A is sufficient.
To do this we first seek the level n-n forming the bottom edge of the
rectangular compressive stress diagram of the concrete.
Let S, and B’respectively be the static moment (with respect to u-u)
and the area of the section above the level n -n.
Then:
M -A‘.0;.h‘
s,= $I,
(4.1~)
whence the level n -n and the depth y can be found.
From Figures 4.2and 4.1we obtain E, and on.
The condition of equilibrium with regard to longitudinal translation gives
I

Figure 4.3
the minimum requisite cross-sectionalarea of tensile reinforcement:
(4.ld)

Example (a)
Data:the triangular section shown in Figure 4.4a.
A’= 1.13cm2(type 3) A = 11.3cm’ (type 1)
6 = 90bars M = 1.85t.m
Analyse the member.
 242

m
E,
a.
I,

N 2
O Ln O Ln O

Figure .1.5
I’
 243
We have :
40
s=- 40x40
2
x -= 10 667 c m 3
3
Hence: M, = (10667 x 67.5)+(1.13 x 2 110 x 32.5)
= 719 O00 kg.cm +78 O00 kg.c m = 7.97 t.m
The first condition is satisfied.

Furthermore:

s, = 185000-78000 = 190cm3
90
The diagram plotted in Figure 4.5 gives S, as a function of y ; for the abscissa
of 1 190cm3 this diagram gives y = 8.2cm. For y/h = 8.2/40 = 0.205,
Figure 4.2 gives: E, = 0.53%,and Figure 4.1 gives: o. = 1 335 bars.
We also have:

The minimum cross-sectional area of tensile reinforcement is therefore:

The safety of the section is therefore ensured.

Example (b)

Data: the section shown in Figure 4.4b.

A' = 1.13 c m 2 (type 3) A = 6.2c m 2 (type 3)
- M = 4.20t.m
0; = 90 bars

Analyse the member.

We have :

(3i5 2')=
S = 37.S -+-

Hence: M, = (22800 x 67.5)+(1.13 x 2 110 x 32.5)

22 800 em3

= 1540000kg.cm+78000kg.cm = 16.18t.m
The first condition is satisfied.
Furthermore:
-
s, = 420 O0090 78 O00 - 8oo c m 3
By trial and error we find: y = 2.4c m and B' = 105 cm'.
For y/h = 2.4/37.5= 0.064,Figure 4.2 gives E, = 1 %, and Figure 4.1 gives
o, = 2 425 bars.
244
The minimum cross-sectionalarea of tensile reinforcement is therefore:
. = (105x90)+(1~13~2110)
A min = 4.9cmz
2 425
The safety of the section is therefore ensured.

Problem No. 2

Data: the section of the member, the reinforcement,the properties of the

materials.
Determine the resisting moment.
The answer is provided by the smaller of the following two values M,
and M a :
(a) Criterion of the concrete:
Apply equation 4.lb:M,= 0.75.S.$b +A’.3,.h’.
(b) Criterion of the tensile reinforcement:
The maximum force that the concrete has to equilibrate if the reinforcement
is functioning at its basic stress is:
Nó = A.Z,-A‘.Z,
Hence,with the aid of equation 4.ld:

Whence we obtain S,and the second limit value:

M a= S,.Zb+A‘.&.h (4.le)
For cold-workedsteels (types3,4or 5) it is necessary to proceed by trial
and error in order to find the precise value of o,, which depends on E,.

Example (a)
Data:the section shown in Figure 4.4a.
A’= 1.13cmz(type3) A = 11.3 cm’(type 1)
-
ab = 90bars.
Determine the resisting moment.
(a)Criterion of the concrete:
From the previous problem we obtain:M,= 7.97t.m
(b) Criterion of the tensile reinforcement:
To begin with,let us adopt:a. = 1 335 bars
Nó = (11.3x 1335)-(1.13 ~2 110) = 12720 kg
BI=--12720 - 141cmZ
90
 245
y = J(2 x 141) = 16.8cm + Y
- = 0.42-* E, = 0.16%
h
W e can therefore indeed base ourselves on oa = 1 335 bars. For the given
triangular section we find :

Hence:M a= (4060x90)+(1.13~2 110~32.5)

= 366 O00 kg.cm + 78 O00 kg.cm = 4.44t.m
This last value is the required answer.

Example (b)
Data:the section shown in Figure 4.4b
A’= 1.13cm2(type3) A = 6.2cm2(type 3)
ob= 90bars
Determine the resisting moment.
(a) Criterion of the concrete:
From example (b)of problem No. 1 we have:M, = 16.18t.m
(b) Criterion of the tensile reinforcement:
To begin with,let us adopt the value:oa = 2 425 bars

Nó = (6.2x 2425)-(1.13 x 2 110)= 12660kg

Y
y = 3.3cm+-=O.O88+~,= 1%
h
W e can therefore indeed base ourselves on o, = 2 425 bars
W e have:S, = 141 x 35.9= 5 060cm3
Hence:
M a= (5060~90)+(1.13~2 110~32.5)
= 455 O00 kg.cm +78 O00 kg.cm = 5.33t.m
This last value is the required answer.

Problem No.3

Data:the section of the member,the compressivereinforcement,the proper-

ties of the materials,the bending moment.
Determine the requisite tensile reinforcement.
First,it must be checked that the upper limit moment M, given by
246
equation 4.lbis not exceeded.The cross-sectionalarea of the tensile rein-
forcement can then be calculated with the aid of equations 4.lcand 4.ld.

Example (a)

Data:the section shown in Fig.4.4a

A’= 1.13cm2 (type 3) tensile reinforcement of steel
type 1
8,= 90bars M = 3.85t.m
Determine A.
Referring to example (a) of problem No. 1 we find that A’ is excessive,
since M < M,.
Furthermorewe have,according to equation 4.IC
s, = 385 O0090-78 O00 = 410cm3
The diagram in Figure 4.5gives y = 15.2cm.
For (y/h) = (15.2/40)= 0.38we obtain from Figure 4.2:E, = 0.195%, and
from Figure 4.1:o, = 1 335 bars.
W e also have:
B’= 116cm2
Hence:
(116x 90)+(1’13 x 2 Ilo) = 9.6cm2
A=
1335

Example (b)

Data:the section shown in Figure 4.4b

A’= 1.13cm2(type3) tensile reinforcement of steel
type 3
Zb= 90bars M = 12.3t.m
Referring to example (b) of problem No. 1 we find that A’is excessive,
since M < M,.
Furthermore,we have according to equation 4.IC :
s, = 1 230O0090-78O00 = 12 cm3
By trial and error we find:
y = 9.8cm and B’= 393 cm2
For (y/h) = (9.8/37.5)= 0.261we obtain from Figure 4.2:E, = 0.375%,and
from Figure 4.1: o, = 2 270 bars.
 247
Hence
A=
= 16,7cm2
(393~90)+(1.13~2110)
2 260

Problem No.4

Data:the section of the member,the compressive reinforcement,the proper-

ties of the reinforcement,the bending moment.
Determine the minimum strength of the concrete,the corresponding tensile
reinforcement.
Equation 4.1b can be transformed to :
A’.O:, . h‘
ob= M - 0.75s (4.19
The minimum requisite strength of the concrete can thus be directly
determined.
O n substituting this stress in equation 4.lcand using equation 4.ld,
we obtain the requisite cross-sectionalarea of the tensile reinforcement.
For certain cross-sectionalshapes it may occur that this calculation leads
to a value of y very close to 075h or even in excess of this limit.In that case
E, will have a very low value,which may even be negative, so that there
would be no reasonable solution to the problem considered in this way.
In order to obtain a fair efficiency ofthe tensile reinforcement,it is advisable
not to let E, go below 0.1%, which limits y/h to 0.50(cf.Figure 4.2).
With the aid of equation 4.lcwritten in the following form:

M -A’.Zá.
h’
=
sn

we can determine the minimum concrete strength corresponding to this

new position of the neutral axis.
It should be noted that it may be economically advantageous further to
increase E, if the resulting reduction in the steel area A compensates for the
extra cost of using a concrete of better quality (i.e.,higher strength).

Example (a)

Data:the section shown in Figure 4.4a

A’= 1.13cm2(type3) tensile reinforcement of steel
type 1
M = 3.85t.m
Determine o
b and A.
248
From equation 4.lfwe obtain:
- 385 O00 -(1.13 x 2 110x 32.5)
a; =
0.75x 10667
= 38.4bars
Equation 4.IC gives :
307O00
S, =
38.4
~ = 8 O00 cm3
Hence:y = 27 cm
This value exceeds 0.50h, so that the solution is of no interest.
Instead,we shall choose:E, = 0.1%, which corresponds to O, = 1 335 bars.
W e have:ylh = 0.50and hence:y = 20cm and S, = 5 333 cm3.
Hence :

-1 307O00 202
Ob=--
- 57.5bars and B' = -2 = 200cm2

110)=
(200~57.5)+(1.13~2 10,4cm2
A= 1350

Example (b)

Data:the section shown in Figure 4.4b

A' = 1.13cm2(type3) tensile reinforcement of steel type 3
M = 12.3t.m
Determine oband A.
From equation 4.lfwe obtain:
-0; = 1230OO0-(1~13~2 110~32.5)
0.75x 22 800
= 67.5bars
Equation 4.IC gives:

By trial and error we find:y = 15.4cm

Hence :

Figure 4.2gives:E, = 0.17%,

and Figure 4.1gives:O, = 2 080bars
W e also have:B'= 575 cm2
 249
Hence:
(575x 675)+(1.13 x 2 110)= 20,0cmz
A=
2 080
In conclusion it should be noted that in example (b)of problem No. 3
we found,for the same member, a value 16.7cmzfor A because a value of
90bars had been presupposed for #b. This serves to show the saving in steel
that can be effected by using a better quality (i.e.,
stronger) concrete.

Problem No. 5

Data: the section of the member, the strength of the concrete,the cross-
sectional area of the reinforcement,the bending moment.
Determine the minimum strength that the reinforcement should have.
W e shall decide upon the type of compressive reinforcement to use,taking
care not to exceed the upper limit value of the moment given by equation
4.1b.
The further procedure is as indicated in problem No.3,for which purpose
equation 4.ldis written in the form:
B' . %+A' . o:, (4.lh)
Gamin = A
With E, known,the suitable types of steel can be determined from Figure
4.1.

Example (a)

Data:the section shown in Figure 4.4a.

-
A' = 1.13cm2 o; = 90 bars
A = 5.3cmz M = 3.85t.m
Determine which types of steel are suitable.
Choose steel type 3 for the compressive reinforcement

M, = (10667x 67.5)+(143 x 2 110x 32.5)= 7.97t.m

Equation 4.lc gives:
78 O00 =
s, = 385 000-
90
410cm3
Hence:
y = 15.2cm ?
h
!= 0.38 B' = 116 cmz
Figure 4.2gives:E, = 0.195%
250
From equation 4.lhwe obtain:

Figure 4.1shows that the tensile reinforcement can suitably be of type 4 or 5.

Example (b)
Data:the section shown in Figure 4.4b
A’= 1.13cm2 ab= 90bars
A = 17.5cm2 M = 12.3t.m
Determine which types of steel are suitable.
Choose steel type 3 for the compressivereinforcement:
M, = (22800~67~5)+(1~13~2110~32~5) = 16.18t.m
Equation 4.lc gives:
1 230 O00 -78 O00 = 12
S” = cm3
90
By trial and error we find:y = 9.8cm and B = 393 cm2
For ylh = 9.8137.5= 0.261,Figure 4.2gives:E, = 0.375%.
Furthermore,equation 4.1h gives:
(393x 90)+(1.13 x 2 110)=
gamin = 150bars
17.5
Figure 4.1shows that all types of steel,except type 1, are suitable.

Problem No.6

Data:the properties of the materials,the bending moment.

Determine the requisite cross-sectionof the member and the cross-sectional
areas of the reinforcement.
A direct solution of this problem is not possible. It is necessary to begin
by assuming a concrete section which appears suitable and then finding
the reinforcement that is required.Next,it must be investigated whether the
reinforcement thus obtained presents a rational and economical solution.
In addition, it must be verified that the safety with regard to cracking
and with regard to deformation is duly ensured, i.e.,the limit states of
cracking and deformation must be checked.

4.3.2 R E C T A N G U L A R S E C T I O N w I T H O U T C O M P R E S SI V E
REINFORCEMENT

The general method set forth in Section 4.3.1will be applied,introducing

A‘ = O.
Let b denote the width of the rectangular section.
Figure 4.6
252
Then:
S = 0.5b . h2 and S, = b . y (h-0.5~)
The upper limit value of the moment is therefore:
= 0.375b . h 2 . aI,
M, = 0.5b . h 2 . Zbo (4.2a)
Knowing the actual bending moment M (which obviously must not
exceed this upper limit)we can apply equation 4.lc), which becomes:

Thissecond-degreeequation givesy ;we can then directly find the lever arm
of the internal forces:
z = h-0.5~
The problems can be solved very easily by means of the set of graphs in
Figure 4.6.
From these graphs we can so determine the coefficient 6 that:

h = 6J(T) (4.2b)

where the units employed are the c m and the kg.

The family of curves drawn as full lines correspond to a range of values of
o'b(inbars). The ratios z/h and y/h are obtained as ordinates associated with
these curves.
From the curvesdrawn as dotted lines in the right-handpart of Figure 4.6
the steel stress ca can be determined as a function of the type of steel and of
the ratio y/h. The values of oaare marked on the horizontal scale at the top
of the diagram.
The requisite cross-sectionalarea of tensile reinforcementis given by :

A=- M (4.2~)
z . ca
With Figure 4.6it is possible to solve the analysis and design problems
normally encountered in actual practice.

Problem No.7

Data:b, h, A,grade of steel,M .

Determinethe requisite minimum ab.
First we determine:

(4.2d)
 253
which is easy to do with the aid of the two scales of the slide rule,and:
M
(i)min= K h (4.2e)

Next,in Figure4.6we must find the point defined by these two co-ordinates.
The required answer is obtained from this point by interpolation between
the curves.
It should be noted that,except for steels of type 1, the stress oa varies as
a function of z/h.
With equation 4.2e it is therefore, in general, necessary to make some
successive approximations in which we each time introduce the value of
oa corresponding to the value of z/h that has been found in the previous
attempt.Satisfactory convergence is obtained.

Example
Data :
b = 40cm h = 90cm
A = 32.17cm2(typeI) M = 36 t.m
Determine the requisite minimum ob
W e find successively:
90J40
= ~36000000= 0.300
36
(i)min = 32.17x 1 335 x 0.90
= 0.993

From Figure 4.6the requisite minimum concrete strength is found to be:

-
ob = 90bars

Problem No.8
Data:b, h, A,properties of the materials.
Determine the permissible bending moment.
The method used for solvingproblem No.2 will be applied.The permissible
bending moment is equal to the smaller of the two following values M,
and M,:
(a) Criterion of the concrete:
M,= 0.315b . h 2 .6
254
(b) Criterion ofthe tensile reinforcement:
The maximum force that the concrete has to equilibrate if the reinforce-
ment is stressed to its permissible limit is:
Nó = A.0,
Hence :

and :
B’
y=-=-
A.o,
b b.Zb
Since:z = h-0.5y, we obtain:
M a = B’.z .Zb
A correction of o, as obtained from Figure 4.6as a function of z/h is some-
times necessary,just as in the previous problem.

Example

Data:
b = 40cm h = 90cm
-
A + 32.17cm2(type3) o; = 80 bars
Determine the permissible bending moment.
(a) Criterion of the concrete:
M, = 0.375x 40x 902x 80 = 97.2t.m
(b)Criterion of the tensile reinforcement:
To begin with,we shall adopt:o, = 2 100bars
32.17x 2 100
= 21.2cm
y = 40x80
Hence :
y 21.2
- = 0.236
h 90
Then,according to Figure 4.6:o, = 2 300bars
W e shall now adopt a corrected value for o,, say:2 270 bars. This time we
have :
32.17x 2 270
= 22.8cm
y = 40x 80
y 22’8- 0253 4 o, = 2 270bars
h - 90
 255
The value adopted for o, is therefore correct,so that:

B’= 32.17x 2 270 = 915cm2

80
z = 90- 11.4 = 78.6 c m
M, = 915 x 78.6 x 80 = 57.5 t.m
This last value provides the required answer.

Problem No. 9

Data: b, h, the properties of the materials.

Determine the permissible bending moment, the cross-sectional area A of
the reinforcement.
We shall take:M = M, = 0.375b.h2. 6
which can also be written as follows:
M = - b.hz
b2
For this value the lever arm z becomes 0.75 h, so that:
M
A=
0.75. h . oa

Example
Data: b = 40cm h = 90cm
Zb= 80 bars steel type 1
Determine M, and A.
We find:
M, = 0.375 x 40 x 90’ x 80 = 97.2t.m
97.2
A= = 108cm2
0.75 x 0.90x 1.33

Problem No. 10

Data: b, h, the properties of the materials, M.

Determine the cross-sectionalarea A of the reinforcement.
We can calculate:
(4.2d)

Taking this value of 6 as the abscissa in Figure 4.6, we determine the cor-
responding ordinate for the appropriate Z,curve and thus obtain z.
256
Since the type of steel is known, we can find o.
The following formula can then be applied :
A=- M (4.2~)
z.

Example
Data:
b = 40cm h=90m
5: = 80 bars M = 36 t.m
steel type 4
Determine A.
We have :
90 40
J = 0.300
= J 3 600000
From Fig.4.6 we obtain:z/h = 0-925and o, = 2 970 bars.
Hence:
36
A= = 14.6cm2
0.925 x 0.90x 2.97

Problem No. 11

Data b, h, grade of steel,M

Determine requisite minimum Zb,
cross-sectionalarea A of the reinforcement.
First,we determine:

(4.2d)

The minimum concrete strength is read from Figure 4.6 by taking the
abscissa corresponding to this value of 6 and interpolating between the
curves for ¿?íb where these intersect the bottom edge of the diagram.
In that case :
Z
-
h
= 075
We can then apply the formula:
A=- M
z .o, (4.2~)

It should be noted that it may be advantageous from the point of view of

 257
economy to choose a concrete of better quality, whereby zlh and o, can be
increased.As a result,the quantity of steel required will be reduced.

Example

Data :
b = 40cm h = 90cm
steel type 3 M = 81 t.m
b and A.
Determine 8
W e have:
90,/40
I3 = 4 8 100000
= 0.200
= 67.5bars.The corre-
(1) At the bottom edge of Figure 4.6we obtain: O;,
sponding requisite steel area is:
81
A= = 66.5cm2
0.75x 0.90x 1 800
(2)N o w let us choose a concrete of better quality (higherstrength), namely,
6 = 100bars.Then:
z
-
h
= 0854 and on = 2 230 bars
Hence:
81
A=
0.854x 0.90x 2.23
= 47.3cm2

So there is a substantial reduction in the quantity of steel required.

Problem No. 12

Data:b, the properties of the materials,M

Determine the minimum effective depth h, the cross-section area A of the
reinforcement.
The minimum value of h is given by:

bmin is read from Figure 4.6,at the intersection of the appropriate curve for
abwith the bottom edge of the diagram.In general,for reasons of economy
and stiffness,it should be endeavoured to give h a value exceeding this
minimum, in so far as is compatible with architectural requirements.
258
When h has thus been determined,the further treatment is as in problem
No. 10.

Example

Data:
b = 40cm Zb= 80bars
steel type 4 M = 36 t.m
Determine the requisite minimum effective depth h and A.
From Figure 4.6:hmi, = 0.184
Hence :
hmi, = 0,184 J(
3 600O00
4o )
= 55.2cm
W e shall adopt a rather higher value,say:h = 75cm
To this corresponds:
75
6 = 0.184~- = 0.250
55.2
O n intersecting the perpendicular to this value of the abscissa with the curve
-
o: = 80 in Figure 4.6we obtain (as the abscissa,on the left-hand vertical
scale):
Z
-
h
= 0.886and oa = 2 870 bars
Hence :
36
A=
0.886x 0.75x 2.87
= 18.9cm2

4.3.3 R E C T A N G U L A R
SECTION W I T H COMPRESSIVE
REINFORCEMENT

Having recourse to the use of compressive reinforcement to increase the

bending moment that can be resisted by a beam whose cross-section and
concrete quality are predetermined is always expensive in terms of the steel
required.
So it is, generally speaking, only in rather exceptional circumstances
that compressivesteel will be used.In actual practice there are only two cases
in which this is normally done:
(a) if for operational or other compelling reasons it is quite impossible to
increase the cross-sectionaldimensions of the member;
(b) if the compressivezone has to be provided with reinforcement anyway,
for some other reason,e.g.,if the member has to be designed for bending
in alternate directions.
The problems dealt with in the previous case (no compressive reinforce-
ment) will not be examined in detail for the member with compressive
 259
reinforcement.Only two cases will be considered,which will suffice to show
how such reinforcement should be taken into account.

Problem No. 13

All the section properties and material properties ofthe member are assumed
to be known,except the cross-sectionalarea A of the tensile reinforcement,
which has to be determined.
W e shall calculate the bending moment M athat the compressivereinforce-
ment can resist jointly with the tensile reinforcement. The difference,i.e.,
(M-Má),acts upon the concrete,as explained in Section 4.3.2.
W e can therefore write:M a= A’.h . ZR
M-Má ZR
A=- +A‘.-
z . (Ta cll
The stress Zahas been indicated in Section 4.2.2,while z and aacan be deter-
mined with the aid of Figure 4.6.
Of course,(M-Ma) must not exceed the upper limit value M, expressed by
equation 4.2a.

Example
Data: b = 40cm h = 90cm h‘ = 84cm
Zb= 60bars A’= 16.0cm2(type 3)
M = 81 t.m tensile reinforcement of steel type 3
Determine A.
W e have:
M a= 16.0x 0.84x 2 110 = 28.4t.m
M - M a = 81 -28.4 = 52.6t.m
6= 90x40 = 0.249
J 5 260000
From Figure 4.6we obtain:
Z
-= 0.842and (T, = 2210 bars
h
Hence :
52.6 2.11
A= + 16.0x -= 31.4+15.3 = 46.7cm2
0.842x 0.90x 2.21 2.14

Problem No. 14

All the section properties and material properties are known,except A’and
A,which have to be determined.
260
M,should be calculated with the aid of equation 4.2a.If M < M,,we Gan
take A’= O, whereby we revert to problem No. 10.
IfM > M,,we can write:
’ . =-M - M ,
A min
h . o:,
The stress Oais indicated in Section 4.2.2.
Hence:

W e have previously seen that for the upper limit value M, of the bending
moment we have:z = 0.75h. It may be advantageous to increaseA’slightly,
whereby the moment to be resisted by the concrete can be reduced;in that
case z/h and oa increase.A certain amount of trial and error is needed for
arriving at the most economical solution.

Example

Data:-b = 40cm h = 90cm h‘ = 84cm

ob = 60bars M = 81 t.m
Determine A’and A.
W e have :
M,= 0.375x 40x 90’x 60 = 72.9t.m
M -M,= 81 -72.9= 8.1t.m
Hence:

72.9 2.11
A=
0.75x 0.90x 1.800
+4.57x -
1.80
= 604+ 5.4 = 65.4cm2
W e shall now consider whether it is advantageous to increase A’.For
example,let us take A’= 17cm2. W e are then back to problem No.13.
Ma= 10.7x 0.84x 2.11 = 19 t.m
M - M ; = 62t.m
Z
6 = 0.228+ - = 0.80and = 2 100bars
h
Hence:
62 2.11
A= + 10.7x- = 41.0+10.7= 51.7cm2
0.80x 0.90x 2.10 2.10
In the example ofthe previous problem it was found that for A’= 16.0cm2
 261
we obtain A = 46.7cm’.The cost of the reinforcing steel will accordingly
be proportional to the following figures:
for A’ = ALin= 4.57cmz: 4.57+65.4= 69.97
A‘ = 10.7cm’: 10.7+ 51.7= 62.4
A‘ = 16.0cm2: 16.0+46.7= 62.7
The last two solutions are therefore practically equivalent from the point
of view of economy.

4.3.4T-SECTION
Determination Of The Effective Width Of The Compressive
Flange

For freely supportedT-beamswith a single rib (web)or comprisinga number

of parallel ribs joined to the same top slab the effective width be-i.e., the
flange width to be adopted in the structural calculations-can be determined
with the aid of the graphs in Figures 4.7and 4.8.
These graphs are valid for uniform, triangular,parabolic or sinusoidal
load distribution and also for the case of a constant bending moment.
O n the other hand,if the beam carries a locally concentrated load applied
to a zone of length a,the width to be taken into account on each side of the
rib, at the corresponding section, should be reduced in relation to the
values given by those graphs in the proportions stated in Table 4.3.
For continuous T-beamsand,in general,for T-beamsin which changes
in the sign of the bending moment occur, the effective width should be
calculated by adopting for 1 the distance between the points of zero bending
moment.
Table 4.3 Table of values to be adopted for the flange width on each side of the rib,referred
to 0.5 (be-bo)

1
__ a=O O < a <0.ll O.ll<a
b-bo

O 1 calculate by 0.6
5 1 linear 0.7
10 1 interpolation 0.9

In the vicinity of a ‘free’support the effective width must not exceed the
width bo plus twice the distance between the support and the section con-
sidered.
In the case of an asymmetricalbeam,e.g., an edge beam,the effective width
bemay be taken asthe arithmeticalmean ofthe valuesobtained by respectively
considering the slab portion on each side of the rib in conjunction with its
own mirror image.
If the compressive flange is joined to the rib by haunches having a width
b, and a depth h,, then the actual width bo of the rib should be replaced by a
262

I
3 0-4

Figure 4.7
 263

t
1.0 ' I

____ 1.
-
-1-
O
........
. ...
h
=0,30
0.25
0.20
-
yi
al
3 ho
; h (0.10
>

j
ò O O O

Figure 4.8
-
bo
b
264
fictitious width b, which should be taken as the smaller the two values:
bo + 2bsand bo +2hs.

Analysis Of A T-Section

T o begin width, the given section should be compared with a rectangular

section of the same depth and having a width equal to be.
For this section it is possible, with the aid of Figure 4.6,to determine
y/h and therefore y.
If y < h o , then the analysis will be an exact one, for the properties of the
given section and those of the assumed rectangular section are the same.
If y > ho (whichcase occurs only if the compressive flange is thin),then the
analysis will have to be modified.
In this latter case the compressive flange can be conceived as functioning
in the manner of compressive reinforcement,and the analysis can be carried
out as though for a rectangular section of width bo and provided with com-
pressive reinforcement.
The bending moment that the flange can thus absorb jointly with the
tensile reinforcement is :

M f= (be- bo)ho. O;, (4.4a)

The difference, i.e., (M-Mi),acts upon the rectangular beam of depth

h and width bo as dealt with in Sections 4.3.2 and 4.3.3.

Problem No. 15
All the properties are assumed to be known, except the cross-sectionalarea
A of the reinforcement,which has to be determined.
If y < ho, the treatment is the same as for a rectangular section.
If y > ho, then Mi should be calculated by means of equation 4.4a and then,
considering the rectangular section bo .h, we can write:

(4.4b)

Example 1
Data: be = 1.80m bo = 30cm
- = 12cm
ho h = 70cm
ob = 120 bars M = 106 t.m
steel type 2
Determine A.
Equation 4.2d gives :
 265
From Figure 4.6:
?! = 0.11 and hence: y = 7.7c m
h
We have: y<ho, so that it can indeed be treated as a rectangular section.
From Figure 4.6we furthermore obtain:

Z
-= 0.945and o, = 2 220 bars
h
Hence :
106
A=
0.945x 0.70x 2.22
= 72.3cmz

Example 2

The data are the same as in the previous example, but now 3,= 60bars.
For 6 = 0.288we now obtain from Figure 4.6:
?!= 0.225and hence: y = 15.7c m
h
Since y exceeds ho, the calculation can permissibly be carried out in this way.
Equation 4.4a gives :
M ; =(180-30)~12~60~(70-6) = 69.1t.m
The rectangular section bo. h must therefore absorb:
106-69.1= 36.9t.m
Hence,applying equation 4.2d:

From Figure 4.6it appears that there is no solution,since the perpendicular

erected at the abscissa 6 = 0.200does not intersect the curve 3,= 60.It is
therefore necessary to provide compressive reinforcement, whereby the
bending moment that the concrete must absorb can be reduced. We shall
so reduce this moment that, for example, y/h is equal to 0.40,which gives:
o,,= 2220 bars.
We find:6 = 0.228and,with the aid of equation 4.2f:
30x 702
O 228 = 28.3t.m
M l=
For these conditions the compressive reinforcement must absorb :
369-28.3 = 8.6t.m
For example, reinforcement of steel type 3 will be installed at 7 c m from
266
the top edge of the section.For Za= 2 110 bars we obtain:
A’ = 8.6
0.63 x 2.11
= 65cm2
and : 28.3 2.11 60
A= +6.5 x -+ 150 x 12 x -= 77.5 cmz
0.80x 0.70 x 2.22 2.22 2 220

Problem No. 16

All the properties are assumed to be known, except h and A. Determine the
requisite minimum depth, though bearing in mind that this is probably not
the most economical solution and that it may be deficient from the point of
view of stiffness.
If it can be assumed that h will exceed 2h0,the following method may be
employed, which is very quick and gives a slightly over-estimated result
very close to the precise value.
Assume: y = ho;this gives:

h= M h0
(4.4c)
be. ho.ob+?
and :
Zb
A = be. ho .- (4.4d)
Ga
If this calculation gives: hc2ho,the result is incorrect, for the upper limit
moment M, will have been exceeded. If so, the section should be treated as
a rectangular section of width be.
It should be noted that the effective width be itself sometimes depends on
the ratio ho/h,so that a readjustment may be necessary after a first calculation.

Example
Data: be = 1.80m bo = 30cm ho = 12cm
A’ = O Ob= 120 bars M = 106 t.m
steel type 4
Determine hmi, and A.
Equation 4.4~ gives :
10600000
h= +-12 = 41+6 = 47cm
18Ox12x 120 2
=-
Check: h 2h0;the result is therefore acceptable.
We thus obtain:
_
Y--ho_ --!.?= 0.256+ oa = 2 840 bars
h 47
 267
Equation 4.4d gives:
120
A = 180 x 12 x -
2 840
= 91.2c m z

4.4 C O M P O S I T E UNIAXIAL BENDING WITH COMPRESSION

4.4.1 A R B I T R A R Y S E C T I O N S Y M M E T R I C A L W I T H R E S P E C T
T O T H E P L A N E OF B E N D I N G

Recapitulation Of Fundamental Assumptions

In addition to the fundamental assumptions on which the analysis for

simple bending is based, the following is assumed:
(a) If the depth x of the zone subject to compressive strain (shortening)
exceeds the effective depth h, the depth y of the rectangular compressive
stress distribution diagram should be limited to :
X-075h
= x-067h .h if x>h
so that, when x tends to infinity (which corresponds to the case of
concentric compression), y will not exceed h.
(b)The bending moment M is calculated with respect to the level of
the centroid of the tensile reinforcement.

Problem No. 17
Data: the section of the member, the reinforcement,the properties of the
materials, the direct force N’,the bending moment M = N’.e.
Check the safety.
Just as for simple bending (cf.problem No. i), the first condition is:
MSM,
where:
M, = 0.75S.%+A’.K.h (cf.4.lb)
We can now write down:
M -A’.O:, .h
s, = -1
ob
(cf.4.1~)

Whence the level n-n and the depth y can be determined. With B and o,
thus known, we obtain:
. = B .O:b+ A’.O:, -N’
A min (4.5a)
=a

If the numerator of this quotient is negative, the section is entirely in com-

pression and the rectangular stress distribution diagram of the concrete
268
covers the whole depth of the section. In that case we must check that the
force N is not too large.The compressive stress in the concrete is then:
M -A’.3,.h
o;,= (4.5b)
S
and the total force that the given section is able to resist is equal to:
M -A‘.Za.h’
= B,.
Nmax +A’.Za+A.Za (4%)
S
where B,denotes the total cross-sectionalarea of the member. It remains to
be checked that N’< NLax.
If this condition is not satisfied,it is necessary to consider a change in the
sign of the bending moment: calculate the eccentricity of N‘with respect
to the reinforcement A’,which then functions as the tensile (or the least
compressed) reinforcement; the depth y should then be reckoned from the
edge closest to the reinforcement A.

Example (a)

Data: the section shown in Figure 4.4a.

A’= 1.13c m 2 (type3) A = 8.6cm’ (type 1)
Zb= 90bars N = 3.1 t
M = 1.85t.m
Analyse the member.
W e have :

Hence :
M, = (10667x 67.5)+(1.13 x 2 110x 32.5)= 7.97t.m
The first condition is satisfied.
Furthermore :
185000-(1.13x 2 110x 32.5)=
s, = 90
190cm3
From Figure 4.5 we obtain: y = 8.2c m

For
-8’2- 0.205we obtain from Figures 4.3 and 4.1:
h=z-
E, = 053% and o, = 1 335 bars
W e also have:
 269
Equation 4.5a gives the minimum cross-sectional area of the tensile re-
inforcement :
A . = (33.7~90)+(1.13~2110)-3 100 = 1.73cm2
min
1335
The safety of the section is therefore ensured.

Example (b)

The data are the same,except that N‘= 22.7t.

The previous calculation would now give:
(33.7x 90)+ (1.13x 2 110)-22 700
Amin = 1335
= -13’0t0
Since the result has a negative sign, we shall calculate Nmax
Equation 4.5~
gives :
452 185
Om) +(i43 x 2 110)+(8.6x i 335)
10667
= 10160+2 380+ 11 470 = 24.01t
The safety of the section is therefore ensured.

Example (c)

The data are the same,except that N = 42.3t.

This value exceeds NmaX as found above.N o w consider the sign of the bend-
ing reversed and replace the reinforcement A (steel type 1) by steel type 3.
We have :
1.85
e = 0.325-- = 0.281m
42.3
The bending moment, calculated with respect to the level of A’,is (in absolute
value):
M = 42.3x 0.281= 11.9t.m
O n going back to the given section,we thus find:

+7.5x 37.5’= 229oo c m 3

s=- 37.53
3 2
M, = (22900x 67.5)+(8.6 x 2 110x 32.5)= 21.4t.m
. So we have indeed: M < M,
<O. We must therefore apply equation
Applying equation 4.5a,we find:Ami,,
270
~ ~ gives :
4 . 5 which

Nmax = (2
452 1 190 O00 -8.6 x 2 110 x 32.5
x 22 900
)
+(8,6x 2 110)+(1.13 x 2 110)

= 26 450+ 18 100+2 380 = 46.93t

The safety of the section is therefore ensured.

Problem No. 18

Data: the section of the member, the reinforcement, the properties of the
materials,the direct force N’.
Determine the resisting moment, i.e., the maximum eccentricity that N
can be given with respect to the tensile reinforcement.
The bending moment is given by the smaller of the following two values
M, and Ma:
(a) Criterion of the concrete:
M, = 0.75s.Zb+A’.Za. h (cf.4.lb)
(b) Criterion of the tensile reinforcement:
The maximum force that the concrete has to absorb if the reinforcement
is stressed to its permissible limit is:
Nó = A.Za-A’.Za+N’
whence we obtain:

from which are obtained the static moment S, and the second limit:
M a= S,.Zb+A’.Za.h‘ (cf.4.lc)

Example

Data: the section shown in Figure 4.4a

A’ = 1.13 c m 2 (type 3) A = 8.6cm2
3,= 90 bars N‘ = 3.1t
Determine emax
(a) Criterion of the concrete:
From the previous problem we have: M, = 7.68 t.m
(b)Criterion of the tensile reinforcement:
We have: oa = 1335 bars
Nó = (8’6x1335)-(1.13~2110)+3 100 = 12200kg
 271
Hence:
BI=--12 200 - 136cm’
90
y = 16.5c m and Y-h = 0.413
Whence we obtain:
E,= 0.175% and from this:o, = 1 335 bars
For: y = 16.5cm. Figure 4.5 gives: S, = 4 570 c m 3
Therefore: M, = (4570 x 90)+(1.13 x 2 110 x 32.5) = 4.88 t.m
We finally obtain:

Problem No. 19
Data: the section of the member, the compressive reinforcement, the pro-
perties of the materials, the direct force,and the bending moment.
Determine the requisite tensile reinforcement.
It must first be checked that the compressive reinforcement is sufficient,
i.e., the following condition should be satisfied :
M-075S.Zb
Ata
3,.h
The cross-sectional area A of the tensile reinforcement can then be deter-
mined by means of equations 4.lc and 4.5a of problem No.17.
If the result is negative, the calculation will have to be modified, inasmuch
as the section is entirely in compression. In that case the requisite cross-
sectional area A should be determined on the assumption that the steel
is subjected to its permissible compressive stress. On transforming equation
4 . 5 we
~ thus obtain:
M - A’.3,.h
N-B, -A‘.3,
S
A= (4.5d)
3,
If this, too, gives a negative result, then it means that the reinforcement A
is not needed in tension nor in compression, but can be provided merely
as nominal reinforcement.

Example (a)

Data: the section shown in Figure 4.4a

A’ = 1.13 c m 2 (type 3) tensile reinforcement type 3
b = 90 bars
O N = 3.1 t
M = 1.85 t.m
Determine A
272
With reference to example (a) of problem No. 17 we check that A’ is
sufficient and find A = 1.73cm2.

Example (b)

The data are the same as above,except that N‘= 22.7t. The tensile reinforce-
ment is of type 3.
Having first found a negative value for A, as in example (b) of problem
No.17,we apply equation 4.5d:
185 000- 1.13 x 2 110 x 32.5
-(1’13 x 2 110)
10 667
A=
2 110
= 4.8 c m 2

4.4.2 R E c T A N G U L A R s E c TIO N

General Formulae

The following five cases which may be encountered in practice will be

successively considered:

h
First case: y,<- and A‘ = O
2
Having regard to the fundamental assumptions stated in the foregoing,
the following relations can be written down, where N b denotes the direct
compressive force and M b denotes the bending moment calculated with
respect to the centroid of the tensile reinforcement:

(4.6a)

and :
NIb = b.y.Zb-A.Ea (4.6b)
Putting:

(4.6~)

(4.6d)

(4.6e)
 273
the equations 4.6a and 4.6b can be written in the form:

y’ Y
= --m
h
These last two relations have been plotted as graphs in Figure 4.9 which
comprises, on the one hand, a family of curves for a range of values of m
plotted on the co-ordinates v’ and p, and, on the other hand, a number of
curves (in dotted lines) which give, for each type of steel,the value of oaas a
function of p.
For cold-worked steels it is necessary to apply iterative corrections to the
values of oaand m.

h
Second case: y <- and A’ # O
2
T o the values of M, and N, considered in the previous case must be
added the effect of the compressive reinforcement assumed to be subjected
to a stress equal to its basic strength ZA.
Therefore :
M = Mb+A’.za.h‘
N’ = Nb-k.4’.
h
Third case: y > - A‘ = O A =O
2
If the upper limit moment is reached, M, has a constant value which is
independent of N b and is equal to:
M, = 0.375b . h 2 .6
The maximum value N, is obtained by considering the state of concentric
compression of the section.
Assuming: h, = 1.05h,we find the maximum:
M g = 1.05b . h . 0.750,
Hence: vLax= 0.788
In Figure 4.9 this value is marked by the arrow at the top of the diagram (at
the ordinate p = 0.375).

h
Fourth case: y > - A‘ = O A #O
2
In the limit state corresponding to the maximum value of N’the tensile
reinforcement A functions in compression at a stress equal to Oaindicated
in Section 4.2.2.
 274

Figure 4.9
 275
To the value of vkax w e can therefore add the following:
A .Üa
WC = b.h.3,
~

h
Fgth case: y > - A’ # O
2
Just as in the second case w e can write down for the limit state:
M = M,+A’.$,.h
N = Nb+A’.Üa

Problem No. 20

Data: b, h, h, A’,A,the properties of the materials, N‘and M = N‘.e

Check the safety of the member.
Calculate:
Mb = M-A‘.ZL.h
Nb = N-A’.Ü,
Equations 4 . 6 and
~ 4.6d give p and d.
The values w and oa are obtained from Figure 4.9.
The requisite minimum steel area can n o w be determined:
A . = w.b . h . Üb
min
Da

The actual steel provided should be at least equal to this, i.e.,A 2 Ami,

Example

Data
b- = 5Ocm h = 65cm h’ = 6 0 c m
ob = 120 bars N = 74.5 t M = 80.3t.m
A’ = 10 c m 2 (type 3) A = 25.5 c m 2 (type3)
Analyse the member.
W e can successively write down:
M, = 8030000-(10x2110x60) = 6764000kg.cm
N b = 74 500- (10x 2 110)= 53 400 kg
6 764 O00
= 50 ~ 6 x 120
5 ~
= 0.266

v’ = 53 400
50 x 65 x 120
= 0.136
Figure 4.9 gives:w = 0.180and oa = 2 210 bars
276
Hence :
. = 0.180x 50x 65x 120 = 31.8cm2
A min
2210
The given cross-sectionalarea of tensile reinforcementis therefore insufficient.

Problem No. 21

Data: b, h, h,A’,A,the properties of the materials,N’

Determine the resisting moment, i.e.,the maximum eccentricity of N with
respect to the tensile reinforcement.
Calculate :Yb = N’-A’ .¿fa
Equations 4.6dand 4.6e give v’ and m.
Figure 4.9provides the maximum value of ,u and the corresponding value
of o, which may necessitate some correction of zu.
Finally: M
,, = ,u.6.h2.Zb+A’.Z,.h’

Example (a)
Data:
b = 50cm h = 65cm h’ = 60cm
- N‘ = 74.5t
o; = 120bars
A‘ = 10 c m 2 (type 3) A = 25.5c m 2 (type3)
Determine M,,,.
We can successively write down:
N b = 74500-(10 x 2 110)= 53 400kg
v‘ = 53400 = 0.136
50 x 65 x 120
25.5x 2250 = o.147
Adopting: o, = 2250 bars, we obtain:ar =
50x 65x 120
Figure 4.9gives: p = 0.242and o, = 2 250 bars
The value adopted for o, was therefore a suitable one.
Hence: M,,, = (0.242x 50 x 65’x 120)= 73.9t.m
and therefore:
73.9
emax =-74.5 = 0.99m

Example (b)

The data are the same as before, except that N’= 297 t.
In this case we have :
Nb= 297000-(10~2110)= 275900kg
 277
= 275 900
VI = 0.706
50 x 65 x 120
Y’ is thus found to be less than the maximum value of 0788.
Irrespective of the value of m,we can take:p = 0.375.
Hence: M,,, = (0.375x 50 x 652 x 120)+(10 x 2 110 x 60)= 107.7t.m
and therefore:
107.7
emax = - - 0.364m
297
The force N can therefore have its line of action at 64m m from the centre
of the section.

Problem No. 22

Data: b, h, h’, the properties of the materials,N,M

Determine A’and A.
First, the moment M, which can be resisted by the concrete alone will
be determined.This maximum value corresponds to: p = 0.375.
However, for higher grades of steel (higher strengths) it may be advan-
tageous to reduce p a little, so as to attain suitably high values for o,.
Having thus chosen a value for p, we can calculate:
M, = p . b . h 2 . <
First case: M > M,
It is necessary to provide compressive reinforcement whose cross-
sectional area is given by:

W e can then calculate:

v’ = N-A.3,
b.h.3,
From Figure 4.9we obtain m and o,; we can then determine:
m .b . h . 3,
A=
0
.

Second case: M <M,

No compressive reinforcement is needed, i.e., A‘ = O. We can calculate:
M and Y‘ =
N
’= b.h2.0,
-, ~

b.h.Zb
278
The further procedure is the same as for the first case.If a value of v' exceeding
v',, = 0.788were found, the excess would have to be taken up by the
reinforcement A in compression.Then :

Example (a)

Data:
b = 50cm h = 65cm h'= 6 0 c m
ob= 120bars N'= 74.5t M = 925t.m
compressive and tensile reinforcement of type 3
Determine A' and A.
For the limit of p we shall, for example, choose the value 0.34.Then:
Mb = 0.340X 50X 652X 120 = 86.2t.m
We find that: M >M b
Hence :
A' = 92.5-86.2= 5.0cm2
2.11x 0.60

v' = 74 500-(5.0x 2 110)= o,164

50x 65 x 120
From Figure 9:w = 0268
Hence :
0.268x 50x 65 x 120 = 51.0cm2
A=
2050

Example (b).

The data are the same as before, except that M = 58 t.m

Since M < M b , we shall take A' = O, i.e., no compressive reinforcement
(other than nominal bars for assembly of the reinforcement). We can then
calculate:

= 74500 = 0.191 5 800O00

= 50x 652x 120 = 0.299
y'
50x 65 x 120

From Figure 4.9:m = 0.072and on = 2 270 bars

 279
Hence :
0.072x 50 x 65 x 120 = 12.4cm2
A =
2 270

Example (c)
The data are the same as before, except that N' = 370 t and M = 120 t.m
Adopting the maximum value of 0.375 for p, we obtain:
M, = 0.375 x 50 x 65' x 120 = 95.0t.m

Hence :
y' = 370 000-(19.7 x 2 110)= o.843
50 x 65 x 120
Since v' exceeds the reinforcement A is loaded in compression.Equation
4.60gives the solution:
(0843x 0.788)x 50 x 65 x 120
A=
2 110
= 10.4cm2

Problem No.23
Data: the properties of the materials, N M
Determine:b, h, A', A
There is an infinite number of solutions to the problem and it is, in some
measure,possible to choose b and h in advance so as to revert to the previous
problem. Some trial and error will be necessary in order to find the most
suitable section with regard to the various factors to be considered:economy,
shear force,stiffness,cracking, architectural requirements.

4.5 COMPOSITE UNIAXIAL BENDING WITH TENSION

4.5.1 A R B I T R A R Y SECTION S Y M M E T R I C A L W I T H RESPECT
T O T H E P L A N E OF B E N D I N G

First case: The line of action of the tensile force N passes between the re-
inforcements A and A'.
The concrete plays no part in resisting the force. The conditions for the
limit equilibrium give :

(4.7a)

The values of 5, are indicated in Section 4.2.2.

280
The eccentricity e of the force N is measured with respect to A.
Second case: The line of action of the tensile force N passes outside the
reinforcement A.
The eccentricity e has a negative value. The formulae for bending with
compression remain valid, provided that N is replaced by -N.

4.5.2 R E C T A N G U L A R SECTION

First case :e 2 O
Equation 4.7a should be applied.
Second case: e < O

If A' = O, equation 4.6ashould be applied:

Instead of equation 4.6b we must now use:

(4.8a)

Putting:

(4.6~)

(4.8b)

(4.6e)

we obtain:
/.i
h(
- 1--
)h:
G0.375

These two relations have been plotted as graphs in Figure 4.10 which
comprises, on the one hand, a family of curves for a range of values of EI
plotted on the co-ordinates v and /.i, and, on the other hand, a number of
curves (indotted lines) which give, for each type of steel, the value of oaas a
function of /.i.
 281

-v
Figure 4.10
282
If A’ is not zero,the following formulae should be applied:
M=Mb+A‘.3a.h‘
and
N = Nb-A’.3*

Problem No. 24

Data:b, h, h’, A’,A, the properties of the materials,N, M = -N .e (e<O).

Check the safety of the member.
Calculate:
M, = M-A’.Za.h
and :
N b = N + A ’ .3,

Equations 4.6~ and 4.8bgive p and v.

The values ofm and oaare obtained from Figure 4.10.
The requisite steel area
can then be calculated:

. = w.b . h.3,
A min
oa

Example
Data: b = 50cm h = 65cm h’= 60cm
3,= 120bars N = 594t M 435 t.m
A’ = 10cm2(type3)
Analyse the member.
W e can successively write down:
M,= 4350000-(10x2110x60) = 3084000kg.cm
N, = 59400+(10~2110)= 80500kg

3 084O00
= 50 x 652x 120
= 0.122

80 500
’= 50x65~120= 0.207
Figure 4.10gives:m = 0.34and oa = 2420 bars
 283
Hence :
A min
. = 0.34x 50x 65 x 120 = 54.8cm2
2 420
The safety of the section is therefore ensured.

Problem No. 25

Data: b, h, h', the properties of the materials, N,M.

Determine A' and A.
First,we shall determine the moment M,that can be resisted by the con-
crete alone. This maximum value corresponds to :
p = 0.375.
However,for higher grades of steel (higherstrengths)it may be advantageous
to reduce p a little, so as to attain suitably high values for on.
Having thus chosen a value for p, we can calculate:
Mb=p.b.h2.&
First case :M >M,
It is necessary to provide compressive reinforcement whose cross-sectional
area is given by:

We can then calculate:

N+A'.3,
v =
b.h.3,
From Fig. 10 we obtain m and a,; we can then determine:
m .b . h. a
b
A=
On

<M,
Second case: M
No compressive reinforcement is needed, i.e.,A' = O. We can calculate:
M N
and v =
= b. h 2 .ob
~

b.h.3,
The further procedure is the same as for the first case.

Example

Data: b = 50cm h = 65cm h' = 6 0 c m

3,= 120 bars N = 59.4t M = 43.5t.m
Determine A' and A.
Choose: ,LL = 0.300
284
We thus obtain:M, = 0.300x 50 x 652x 120 = 76.0t.m
so that we can take: A’ = O
Therefore: M < M b ,
We can now calculate:
95 400
v = = 0.152
50 x 65 x 120

4 350 000
= o.172
!
-
¿ =50 x 65’ x 120
Figure 4.10gives: w = 0.342and o. = 2 340 bars
Hence :
0.342x 50 x 65 x 120 = 57,0,
’
A=
2 340
 5
ANALYSIS OF TENSILE AND
FLEXURAL CRACKING

5.1 PRELIMINARY REMARKS

A tensile crack will appear in a structural member subjected to tensile stress
when,at a given point,the strain of the member exceeds the maximum ulti-
mate strain and when, associated with this, the stress exceeds the tensile
strength.In an initially uncracked concrete member without reinforcement
these phenomena produce failure which concides with the development
of the first crack,perpendicularly to the direction of the tensile force.This
failure is not preceded by any warning signs; it occurs suddenly.Precise
measurements show that the strains which occur in a test specimen of this
kind remain substantially proportional to the tensile stresses almost up to
failure. The slight departure from proportionality that occurs just before
failure can be neglected,so that failure can be said to occur practically in the
elastic range of behaviour, without any plastification of the concrete in ten-
sion being involved.
In the case of concrete the origin of the cracks is attributable to one of the
following factors:
Action of shrinkage of the concrete,if that action encounters restraint either
as a result of structural arrangements (casewhere the member is unable to
expand freely,e.g.,if it is anchored at its ends) or as a result of bond of the
reinforcement (casewhere the member is reinforced and the development of
tension in the concrete is accompanied by the development of compression
in the steel).
Action of temperature variation, because of the difference between the co-
efficients of thermal expansion of concrete and steel respectively.However,if
the .variationin temperature does not exceed f2O0C,its action can be re-
garded negligible in the development of cracking.
Action of an external tensileforce applied to the ends of the member.

285
 286
5.2 ANALYSIS OF CRACKING IN REINFORCED CONCRETE
5.2.1 P U R P O S E OF A N A L Y S I S OF C R A C K I N G

The essential purpose of the theories of cracking of reinforced concrete is

to be able to predict -with due regard to the action of shrinkageand external
forces-the development of the distribution and widths of the cracks in the
various members of a structure. In a more general way they should enable
us to determine and verify, with a degree of accuracy compatible with the
random character of the phenomenon, the limit state of cracking of those
structural members.This limit state will itself be conditioned by the nature
of the structure,its environmental conditions,the possible risk of corrosion
ofthe reinforcement,and the need to safeguard the durability ofthe structure.
In their present state of development the theories of cracking of reinforced
concrete aim at calculating the distribution and the widths of the flexural and
tensile cracks,which are assumed to be perpendicular to the direction of the
reinforcement.
This is evidently a limited domain,since it excludes more particularly
the shear cracks,the theoretical calculation of which,if properly carried out,
would be of extreme complexity.In fact,such a calculation would have to
take account not only of the development of the ‘diagonal’stresses and of the
fundamental criteria of failure of concrete,but also of the type of reinforce-
ment and the other structural arrangements which decisively affect the
development of the phenomenon.
Also, the theories are unable to take the longitudinal cracks, extending
parallel to the reinforcement,into consideration,although such cracks are
very dangerous from the point of view of corrosion of the steel. The forma-
tion and spreading of these cracks depend essentially upon the density of the
concreteand the coverto thebars,particularly along theedges ofthe members,
and are therefore not amenable to systematic theoretical calculation.

5.2.2 B A S I C A S S U M P T I O N SF O R T H E CALCULATION OF
CRACKING

These basic assumptions concern:

(a)the reference definition of the crack width;
(b)the bond stress distribution along the longitudinal reinforcement;
(c) the effect of the presence of transverse reinforcement;
(d)the estimation of the tensile strength of the concrete;
(e) ignoring the effect of temperature variations.

Reference Definition Of Crack Width

A crack generally follows a sinuous path. For defining and measuring the
width of the crack it is therefore difficult to refer to a precise direction.
 287
The only reasonable assumptions consist in considering the width of a
crack parallel to the main reinforcing bars, at the level of their centroid,at the
external surface of the concrete.
These arbitrary assumptions are, however, each a source of inaccuracy
which accentuates the random character of the phenomenon.
These inaccuracies are confirmed by many tests which have clearly re-
vealed the great variations that may occur in the crack widths,both at the
external surface and in the interior of the concrete surrounding the re-
inforcement.

Distribution Of The Bond Stresses Between Concrete And Steel

The relationship z(x) which governs the distribution of the bond stresses
along the reinforcement is of fundamental importance in the development
of the theories of cracking ofreinforced concrete.In actualpractice,however,
this importance is more apparent than real, for the work of the Comité
Européen du Béton (C.E.B.) has shown that,while the theories of cracking
generally differ from one another in having very different basic hypotheses,
they can,by virtue of their mode of development,be brought into line with
one another in a single theory and yield practically equivalent results con-
forming to the usual experimental measurements.
This does not make it any less essential to adopt a hypothesis for the bond
stress distribution. The various ‘laws’or relationships governing the
phenomenon,as evolved by theoretical investigators,can for the most part
be classified into two principal types: the relationshipsbased on sinusoidal
variation (including Professor Saliger’s theory) and those based on uniform
friction (including Mr.L. P.Brice’s theory). Other investigators have en-
visaged, more or less implicitly, a linear variation (including Professor \

Wästlund’s theory). Others,again,have adopted formulae of experimental

origin,without theoretical development,and have thus been able to bypass
the need for making basic assumptions as to the stress distribution.
For practical purposes, in the case of members subjected to concentric
tension or simple bending, the tests appear to show that (at any rate,in the
mid-span zones of such members):
(a) if the steel stress remains below about three-quartersof the elastic
limit,the bond stress distribution can be considered to be sinusoidal;
(b)if the steel stress exceeds about three-quartersof the elastic limit,the
bond stress distribution can be taken gradually to approximate to a
relationship characterised by constant friction.

Effect Of Transverse Reinforcement

The presence of transverse reinforcement undeniably forms points of dis-
continuity in the bond between the concrete and the steel because,for one
thing,it is equivalent to a local reduction of the section and particularly to a
288
lowering of the quality of the concrete cover and, furthermore,it presents
an obstacle to the displacement of the reinforcing bar in relation to the
concrete. This local impairment therefore very considerably increases the
probability that a crack willform.

Estimation Of The Tensile Strength Of The Concrete

It is known that the tensile strength of concrete is highly sensitive to the

slightest local flaw in manufacture and that the minimum value o. of this
strength at a given point of the structure determines the development of the
first crack.
The progressive development of cracking is a function of other local flaws
of the concrete in tension, for respective strengths (oo+Aloa), (oo+Azoo),
...(oo+Anoo), where:
OtA,oo < Azoo< ...< Anoo
However, since the actual distribution of the local tensile strength values
of concrete is of an entirely random character, the only possible hypothesis
for the calculation of cracking consists in considering,over the whole length
of the member (or even throughout the entire structure), a uniform and
minimum value of the tensile strength, defined more or less arbitrarily in
terms of results yielded by tests on standard specimens.

Ignoring The Effect Of Temperature Variations

This effect can be neglected in the calculation of cracking because, for

temperature variations not exceeding f20°C,it corresponds practically
to the error in estimating the shrinkage of the concrete.

5.2.3 C A L C U L A T I O N OF C R A C K I N G I N
UNDER-REINFORCED MEMBERS

Range Of Validity

The members concerned are 'under-reinforced in the sense that the rein-
forcement, because of its percentage being too low, is unable, when the first
crack appears, suitably to resist the force that previously was resisted by the
concrete in tension.
This is more particularly the case with members in which the reinforcement
percentage is too low to enable them to withstand the action of the shrinkage
of the concrete, even in the absence of any external tensile force. Such
members crack systematically merely as a result of shrinkage,without the
reinforcement being able to ensure effective transmission of the forces.
Reinforcement present in so low a percentage will reach the range of plastic
strain as soon as the first shrinkage crack is formed.
 289
Expressing the corresponding value of the direct force No,we obtain:
N o = B . o o = A.o,>A.oe
W e can put

or :

W e are therefore dealing with members whose mechanical percentage of

reinforcement w is less than oo/oó:

i.e.

This category of members includes,for example,bracings and partitions.

More frequently,the members in question contain a sufficient percentage
of reinforcement to withstand the action of shrinkage of the concrete but
not sufficientto enable such members to resist the action ofan external tensile
force after the concrete in tension has cracked.When the first crack develops,
there is a sudden and sharp transfer of the resisting force in the concrete in
tension to the tensile reinforcement. As a result of this the stress in this
reinforcement is, on account of its percentage being too low, suddenly
increased beyond the elastic limit of the steel:

oa’oe
If the external tensile force continues to increase, the steel strain will
increase by plastic deformation without an attendance increase in the steel
stress:the first crack will remain the only crack and steadily widen.
Let oaOdenote the steel strain immediately before and oalthe steel strain
immediately after cracking.Then:A .oao+B .o. = A .oal> A .oe(equation
ofequilibrium offorces)(oao/&)= (oo/Eb) (straincompatibility equation).
W e can put:

or:

W e are therefore dealing with members whose mechanical percentage

of reinforcement w is below a certain limit percentage,which is called the
minimum effective percentage of reinforcement and is defined by the relation:
290
This category includes,for example,the walls of large tanks,dry-docksor
other massive structures in which the percentage of reinforcement, even
though a substantial amount of steel is installed in such structures,seldom
attains the minimum effective percentage.
In a general way the calculation of the cracking of under-reinforced
members is applicable to all cases of accidental cracking due to discontinuity
of the member or to the presence of a constructionjoint (i.e.,at a point where
concreting was temporarily stopped).

Purpose OfThe Calculation

In all cases referred to in the foregoing there cannot be any systematic

development of cracking.
The crack,whether accidental or not,remains the only one and becomes
wider as the loadings to which the member is subjected are increased.Under
these conditions the member functions as an assembly of two parts, on
each side of the crack,in each of which there is a build-upof bond along
the reinforcement,starting from the crack.
The calculation is thus confined to calculating the width of the crack.

Calculation Of The Crack Width

The problem consists in calculating the deformation of the above-mentioned
assembly.It is determined by the difference between the elongation of the
bare steel and the relative displacement of the bar in relation to the concrete
over the bond length 1, on each side of the crack.
Assuming a sinusoidal distribution of the bond stresses along the bar,we
have :
(a) elongation of the bare steel:
1, .-O R
ER
(b)displacement of the bar in relation to the concrete:
1, j:2 (1 -cos$,> dx = 1, (: -:)
1
The resultant deformation along the bond length of the bar is the differ-
ence of the two above quantities,i.e.,
-.lo.-
2 GR

= E R

Furthermore,the bond length i, can be determined by integration of the

equilibrium equation expressing the transmission of the tensile forces
between the concrete and the steel over the elementary distance dx:
A . do, = &d . Z(X) dx
 29 1
Since the bond stress distribution is taken to be sinusoidal,we obtain:
o, = - x . dx = En4
-. 1, . 7average
A
hence:
lo=-
On -A
average '

The deformation along the bond length is therefore

Since this deformation occurs on both sides of the crack,the width ofthe
crack is equal to:
A
w = -4 - (
0,)
'
71' Cn4'raverage ' Ea

The Effect Of Possible Swelling Of The Concrete

If,under normal conditions of service,one of the faces of the structure or of
the member under consideration is in permanent contact with water or with
an atmosphere saturated with water vapour, the resulting swelling of the
concrete has a favourable effect and tends to reduce the crack width. It is
therefore advisable to take this into account,in a more or less empirical
manner, in checking the cracking. This is more particularly the case with
tanks,dams,dry-docksand other hydraulic engineering structures.

5.2.4 C A L C U L A T I O N OF C R A C K I N G I N N O R M A L L Y
REINFORCED MEMBERS

Range Of Validity
In 'normally reinforced' members the reinforcement is sufficient to be able,
when the first crack appears,suitably to resist the force that was previously
resisted by the concrete in tension.More particularly,these are members in
which the reinforcement percentage is at least equal to the minimum effective
percentage, as defined in Section 5.2.3:

In such circumstances the combined action of shrinkage and of an ex-

ternal tensile force causes cracks to develop successively,these being distri-
buted in a random manner over the entire length of the member.
292
In other words, the calculation for the cracking of normally reinforced
members is applicable to all cases of systematic cracking.

Purpose Of The Calculation

The calculation of cracking in normally reinforced members consists in

determining the distribution and width of the flexural and tensile cracks,
which are assumed to be perpendicular to the direction of the reinforcement.
Not included in such calculation are 'diagonal'cracks due to shear force
(for which a limited and empirical extension of the calculation procedure
is envisaged,however) and longitudinal cracks (whichgenerally depend only
on structural arrangements and on the conditions of execution) (Section
5.2.1).
The complete analytical calculation of cracking has been developed for
ordinary reinforced concretemembers,conforming to the five basic assump
tions stated in Section 5.2.2,for concentric tension and for simple bending
respectively.
This calculation can be extended to prestressed concrete members by
using additional assumptions which,despite some extra inaccuracies(which
are generally conservative in character and therefore on the safe side),
yield results that provide a sufficiently good approximation.

Calculation Of Tensile Cracking

Calculation Of The M a x i m u m Crack Spacing

The analysis of the development of cracking shows that the spacing Al of

the cracks is comprised,in a random manner,between a minimum 1, corre-
sponding to the length of zone of disturbance due to the formation of the
initial crack or of a subsequentcrack and a m a x i m u m correspondingto twice
that length,i.e.,21,:
1,<A1<21,
The most unfavourable case is the one in which the cracks are the least
well distributed,i.e.,the case where the crack spacing has its maximum
value:
Al,,, = 21,
Hence :
Al,,, = 2 . -B - =o
En4 Taverage
'

Calculation Of The M a x i m u m Crack Width

The maximum crack width obviously corresponds to the case where the
cracks are least well distributed,i.e.,where the crack spacing is a maximum.
 293
Besides,this width is equal to the resultant deformation of the cracked
member,corresponding to the width Al ofthe block between two consecutive
cracks whose distance apart is equal to the maximum spacing.The analysis
of the behaviour of this member under the action of an external tensile force
shows this resultant deformation to be equal (per unit length)to:

where the stress on and the strain E, of the steel correspond to the action of
the external tensileforceupon the reinforcement(assumednot to be embedded
in concrete).
Upon the action of the external tensile force must be superimposed the
shrinkage E,,, which is likewise a function of the reinforcement percentage
w o according to the relation:

Hence:
wmax = ( E I +Era) . Alma,

i.e.:

Hence :

Worked Examples

(a) Calculation of cracking in a tie-member reinforced with plain mild steel bars
A tie-member which has to resist a pull of about 125 under working-
load conditions has a 22 cm squaresection and is reinforced with nine plain
mild steel bars of 32mm diameter.The concrete has a compressive strength
O; = 345 kg/cm2 (cylinder strength, i.e.,determined on a cylindrical test
specimen)and a tensile strength O, = 29.4kg/cm2.The steel has a measured
elastic limit ue = 2 830 kg/cm2;its working stress is O, = 1 700kg/cm2.
The average bond stress,having due regard to the conditions of embedment
of the bars,is taken as:zo = 050,.
294
For these conditions the maximum crack spacing calculated according
to equation 5.1 is:
484 1
Al,,, = 2 x - x-
90.43 0.54
= 19.82cm
Furthermore,according to equations 5.2and 5.3

E,, = 0.51~~
= 153 x
(for an average shrinkage of the concrete:E,= 3 x
Hence,for the maximum crack width (equation 5.4):
w,,, = ( E ~+E,,)A~,~, = (793+ 153)x lop6x 198.2= 0.189mm
It should be noted that,in this ordinary case,shrinkage accounts for about
16%of the maximum crack width.
(b) Calculation of cracking in a tie-member reinforced with medium-tensile
deformed bars
The tie-member is similar to the one in the previous example in respect
of its external dimensions and loading,but is reinforced with nine medium-
tensile ‘deformed’bars (i.e.,bars provided with specially formed projections
to produce high bond) of 22mm diameter, with a measured elastic limit
oe = 4600kg/cm2 and a working stress 0, = 2780kg/cm2.The concrete
has a cylinder strength o; = 345 kg/cm2 and a tensile strength o, = 32.6
kg/cm2.The average bond stress, having due regard to the conditions of
embedment of the bars,is taken as:T, = 1.350,.
For these conditions the maximum crack spacing calculated according to
equation 5.1 is:
484 1
Al,,, = 2x - x-
62.17 1.35
= 11.54cm
Furthermore,according to equations 5.2and 5.3:

E1 = E,--. 1 -
00
= (1 323-27) x = 1296x
4.m0 E a
E,, = 186x
= 0.62~~ (for average shrinkage:E, = 3 x
Hence,for the maximum crack width (equation 5.4)
w,,, = ( E +~E~,)A~,,, = (1 296+ 186)x x 115.4= 0.171mm
So,thanks to the high bond developed by the reinforcement,the maximum
width ofthe cracks remains of the same order of magnitude as in the previous
example.
O n the other hand,because of the substitutionof medium-tensilesteel for
mild steel and the corresponding reduction of the geometrical percentage of
 295
as against 16 %in the
reinforcement,shrinkage now accounts for only 13 %,
previous case.

Calculation Of Cracking In A Flexural Member

Calculation Of The M a x i m u m Crack Spacing

Consider an element comprised between two flexural cracks in a beam

(see Figure 5.1). The bending moment M is assumed to be of constant
magnitude over the length of this element.
At a section with abscissa x the equations of equilibrium of the forces and
moments can be written down as follows:
A . o a l= A . o , + C m j

A .Oal.z = A . z+-. ob
Iu
T(x)dx

I
where oaldenotes the tensile stress in the steel at the cracks (i.e.,oal=
[ M / a. z
(5.5)

]), while o, denotes the tensile steel stress at the section considered,

Figure 5.1

ob the maximum tensile stress in the concrete at that section,and I/u the
section modulus for tension.
From the two equilibrium equations we can determine the value of the
maximum tensile stress ob in the concrete at the extreme tensile fibre of the
section with abscissa x:

The minimum spacing 1, of the two cracks corresponds to the abkissa

x = 1, for which the maximum tensile stress ob in the concrete can attain the
tensile strength of the concrete;hence:
296
But the most unfavourable case evidently occurs if the cracks are least well
distributed,i.e.,if their spacing has its maximum value:
Al,, = 21,
The calculation of the maximum spacing of the cracks therefore consists
in solving the equation:

Hence:

Into this expression we can also introduce the resultant N of the com-
pressive forces in the section under the action of the cracking moment,
noting that:
1
Mcracking= N .z = 6 0 .i

Hence:
U V
N = -.O0
Z
and therefore:
N'
Al,, = 2.
En4 .zaverage

Calculation Of The Maximum Crack Width

The calculation of the width ofa flexural crack can be reduced to the calcula-
tion of the difference between the respective elongations of the steel and the
concrete in the block comprised between two consecutive cracks,taking
account of the sliding of the reinforcement in relation to the surrounding
concrete in accordance with the normal action of bond.
Consider a section with abscissa x.The steel strain can be taken as:

while the 'concrete strain is the result of the elastic elongation and the
shrinkage:
*

Hence :

w
,, = 2 JOA"' (5-3 + dx
Ea
 297
According to the equilibrium equations in 5.5 and 5.6we can write:

so that we obtain:

where the numerical coefficients are independent of the bond stress distribu-
tion along the reinforcement and where corresponds to the tensile stress
in the steel at each crack
M
(Tal =-
A.z
Hence:

Worked Examples

(a) Calculation of cracking in a T-beam reinforced with plain mild steel bars
Consider a T-beam,90 cm in depth,with a compressive flange 1 m wide
and 16 cm thick.The rib -ofthe beam is 18 cm wide. The main tensile rein-
forcement comprises eleven plain mild steel bars of 32 mm diameter,with a
measured elastic limit oe = 2 980 kg/cm2 and a working stress oa = 1990
kg/cm2.The concrete has a cylinder strength ab = 292 kgjcm2 and a tensile
strength o,,= 41.8 kgjcm2.The average bond stress, having due regard to
the conditions ofembedment of the bars is taken as:zo = 1.050,.
For these conditions the maximum crack spacing calculated according to
equation 5.7 is:
95 310 1
Al,, = 2 x X- = 22.90~~1
110.5 x 71.75 1.05
Furthermore,
-1 __-M
- 948 x lop6
E,,'A.z
(elasticelongation of the steel)
1 I o0
= 149~
E a ' ~2'
(apparent plastic elongation of the concrete)
298

(elasticelongation of the concrete)

E,, = 0.50~~
= 150x
(for average shrinkage of the concrete:E, = 3 x
Hence,for the maximum crack width (equation5.8):
w,,, = (948-149-68+ 150)x 10- x 229 = 0.202mm
It should be noted that,in this ordinary case,shrinkage accounts for about
17% of the maximum crack width.As for the elastic elongation of the con-
crete,this accounts for under 8%.
(b) Calculation of cracking in a T-beam reinforced with medium-tensile
deformed bars

The T-beamis ofthe same external dimensions as the one in the previous
example,but in this case the reinforcement consists of eleven medium-tensile
'deformed' (high-bond)bars of 22m m diameter, with a measured elastic
limit oe = 6 980kg/cm2 and a working stress o, = 4650kg/cm2.The con-
crete has a cylinder strength ob = 260kg/cm2and a tensile strength o, =
40.1kg/cm2.The average bond stress, having due regard to the conditions
of embedment of the bars,is taken as:zo = 2.670,.
For these conditions the maximum crack spacing calculated according
to equation 5.7is:
58710 1
Al,,, = 2~ 58.4x 77.28 2.67= 9.74cm
'
~

Furthermore,
- -- - 2 2 1 4 ~ 1 0 - ~
E, ' A . z
(elasticelongation of the steel)
1 I O,
= 239 x lop6
E,'v' 2
(apparent plastic elongation of the concrete)

(elasticelongation of the concrete)

E,, = 0.585~~= 175 x
(foraverage shrinkage:E, = 3 x lop4)
Hence,for the maximum crack width (equation 5.8):
w,,, = (2214-239-69+ 175)x x 97.4= 0.203mm
 299
So,thanks to the high bond developed by the reinforcement,the maximum
width ofthe cracks remainsof the same order of magnitude as in the previous
example.
O n the other hand,because of the substitution of medium-tensilesteel for
mild steel and the corresponding reduction of the geometrical percentage of
reinforcement, shrinkage now accounts for only 8% instead of 17% as in
the previous case.Also,the proportion associated with the elastic elongation
of the concrete has been reduced from 8%to a mere 3 %.

5.2.5 CONCLUSIONS

In the case of ‘normally reinforced’members a comparison between the

complete analytical calculation and experimental measurements of crack
widths yields various practical conclusions which can facilitate the applica-
tion of this calculation procedure by designers.
Cracking is a random and fundamentally dispersive phenomenon which it
would be illusory to pretend to define in precise terms.Anyway,the designer
is interested only in the order of magnitude of the crack widths.
The only purpose and only significanceof the systematic application of the
analytical formulae is to be able to compare the respective influences of the
various parameters of the phenomenon and to justify the approximations
that are indispensable to practical design calculations.
In the most commonly encountered cases the overall effect of shrinkage,
elastic deformation and apparent plastic deformation of the concrete can be
considered negligible, for the error associated with this approximation is
found in practice to be distinctly less than the dispersion (scatter)of the
experimental measurements.
This being so,the expressions for the maximum crack width given above
can be simplified as follows:
(a) Concentric tension:

where the bond stress z,,is the average ‘flexural’bond stress.

(b) Simple bending:

where the bond stress zo is the average ‘flexural’bond stress. Provided

that the width bo of the web of the beam considered is less than four times the
sum of the diameters of the bars forming the reinforcement (withplain bars)
or less than twice that sum (withdeformed bars), the calculation of flexural
cracking can be simplified and reduced to the calculation of tensile cracking
applied to the ‘embedment zone’ of the main tensile reinforcement (‘tie-rod
300
analogy’). The bond stress q,to be introduced into the calculation must
obviously be the ‘anchorage’bond stress and not the ‘flexural’bond stress.

5.3 PRACTICAL CHECKING OF CRACKING IN

REINFORCED CONCRETE
5.3.1 P R I N C I P L E SO F T H E C H E C K I N G O F C R A C K I N G

Nature And Validity Of The Practical Rules For The Checking Of

Cracking
Having regard to the basic assumptions, a calculation for checking the
cracking of a reinforced concrete member is possible only for transverse
flexural or tensile cracking in the ‘embedment zone’ of the main tensile
reinforcing bars and cannot systematically take account of the influence
(ifany) of shear force.
It must therefore be admitted that a check of this kind covers only a
relativelysmallproportion ofthe crackingphenomena that are liableto occur
in a reinforced concrete structure. More particularly, it does not include
longitudinal cracks which may give rise to tangential actions between the
web and the zone containing the main tensile reinforcement, nor does it
include the ‘diagonal’(inclined)cracks that develop over the entire depth of
the web or rib as a result of shear force,whether or not acting in combination
with other loadings.And yet these cracks are often more dangerous from the
point of view of corrosion than the transverse cracks occurring in the
embedment zone of the main tensile reinforcement.
So one must avoid attributing to the theoretical calculation of cracking
various virtues that it does not possess or regarding it as having in practice a
universal scope or major importance which are justified neither by its degree
of accuracy nor by its range of validity.
It would indeed be misleading to claim to be able to safeguard against the
possible effects ofcracking merely by limiting the theoretical maximum width
of the flexural and tensile transverse cracks.
The practical object is merely to give guidance to the designer in making
appropriate structural arrangements (concerningthe diameter and distribu-
tion of the bars,in particular) and in avoiding certain gross errors of design
which might result in concentrations and excessive widths of flexural and
tensile cracks.This being so,the calculation of cracking should be confined
to verijìcation of the ‘Rules of good design’.
For practical purposes these rules may be presented in the form of a
relation between the working stress,the percentage and the diameter of the
main tensile reinforcing bars,having due regard to the bond properties of
the steeland the tensilestrengthofthe concrete.These rulescan be established
and applied for various classes of structures based on certain conditions of
preservation and service. But on no account must these rules explicitly
state the maximum value of the crack widths, for the measurement and
 301
checking of these widths would run into insurmountable difficulties and
would be of entirely illusory value (Section 5.2.2).

Classification Of Structures According To T h e Permissible Crack

Widths
Class 1 :Structural Members Which M u s t Ensure Watertightness
Or Are Exposed To Aggressive Actions
These are members in which cracking is very harmful,either because they
have to be watertight (e.g., walls of tanks,locks or dry-docks)or because
they are exposed to a particularly aggressive medium.
For such members it is agreed to introduce implicitly into the cracking
calculation,for establishing appropriate rules for designing the reinforce-
ment, an upper limit of 0.1mm for the maximum width of the cracks.In
certain cases of walls provided with a low percentage of reinforcement this
limit may result in apparently severe rules,but this severity can be substan-
tially mitigated by taking account of the favourable effect of the swelling of
the concrete in all cases where the walls in question are permanently in
contact with water or with an atmosphere saturated with water vapour
(Section5.2.3).

Class 2: Unprotected Ordinary Structural Members

In these members cracking of the tensile zone is harmful either because they
are exposed to the effects ofthe weather (asis the case with outdoor structures
such as bridges and other civil engineering works)or because they are exposed
to a humid and aggressive atmosphere (asin the case of certain industrial
structures, factory roofs, or workshop buildings in which considerable
quantities of water vapour are liable to be produced). This class may also
be taken to include members which have to support very fragile claddings
or facings which would suffer harmful consequences from excessive cracking
and deformation.
For such members it is agreed to introduce implicitly into the cracking
calculation,for establishing appropriate rules for designing the reinforce-
ment,an upper limit of 0.2mm for the maximum width ofthe cracks.

Class 3 :Protected Ordinary Structural M e m b e r s

In these members cracking is not harmful and does not have any serious
adverse effects upon the preservation of the reinforcing steel nor upon the
durability of the structure.
For such members it is agreed to introduce implicitly into the cracking
302
calculation,for establishing appropriate rules for designing the reinforce-
ment,an upper limit of 0.3mm for the maximum width of the cracks.

5.3.2 D E S I G NR U L E S F O R T H E M A I N R E I N F O R C E M E N T

Preliminary Assumptions

These practical rules,which are applicable to all structural members loaded

in bending,have been established for each of the classes of structures defined
in the foregoing -i.e.,classes 1,2and 3 - on the basis of simplified assump-
tions and average values as indicated in the following:
For the practical application of the calculation it is assumed that the main
tensile reinforcing bars are satis$actorily embedded in concrete; this means
more particularly that the bars must have an amount of cover at least equal
to their diameter 4.
Consequently, the average anchorage bond stress zo is assumed to be
independent of the embedment and is taken as three-quartersof the tensile
strength o. of the concrete in the case of plain bars and to five-quarters
thereof in the case of deformed bars.
The tensile strength o. ofthe concrete is assumed to have an average value
of 40kg/cm2,but this value will in practice only enter into the design of
‘under-reinforcedmembers in which cracking is not of a systematic
character and can be calculated only by considering the bond of the steel
in the concrete.The practical scope of this assumption is therefore limited.

Design Of Under-Reinforced Members (Cracking Not O f A

Systematic Character)

Application of the preliminary assumptions to the expression at the end of

Section 5.2.3yields the following results:

Class 1 : Watertight Structures Or Structures Exposed To

Aggressive Atmosphere

The rules for checking the cracking for class 1 are based on the condition:
w,,, +O.1 mm
i.e.,with reference to the expression derived in Section 5.2.3:

N o w (AIC.rr4) = (4/4)and the average anchorage bond stress zo can

be replaced by its value as reasoned above.Then:
371 oo.E,.+1for plain bars)
4mJiF.m 10
 303
571 oo.E, 1
.-(for deformed bars)
@mrn> 4.0210
The maximum tensile stress o. in the steel at the crack under consideration
is, as a first approximation,taken as equal to the basic strength of the steel,
i.e.,equal to the guaranteed elastic limit divided by the reduction coefficient
relating to the limit state of cracking.Hence:

If the modulus of elasticity of the steel is taken as E, = 2.1x lo4 kg/mm2

and the tensile strength of the concrete is taken as go = 40kg/mm2(average
value), the check criteria for cracking can be presented in the following
simple form:
5 O00
4mm+ - (forplain bars)
(aJ2
Class 1
q5,, > (fordeformed bars)

where the elastic limit oeof the steel is expressed in kg/mm2.

If the concrete has a tensile strength differing from the assumed average
value of 40kg/cm2,the value of the maximum diameter should be corrected
in the ratio 00/40.

Class 2: Unprotected Ordinary Structures

The rules for checking the cracking for class 2 are based on the condition:
wmaX>0.2mm
whence the following criteria are obtained:

q5,, > =(for plain bars)

(oJ2
Class 2
i > E(fordeformed bars)
4,,,,,,
where the elastic limit 6, of the steel is expressed in kg/mm2.

Class 3 :Protected Ordinary Structures

The rules for checking the cracking for class 3 are based on the condition:
wmax >0.3mm
304
whence the following criteria are obtained:
4mm>-15 O00
T(for plain bars)
(ce)
Class 3 25 500
4,,,,,,>
(0,)
(for deformed bars)
where the elastic limit o, of the steel is expressed in kg/mm2.

Design O f Normally Reinforced Members (Systematic Cracking)

Application of the preliminary assumptions to the expression in Section 5.2.5

yields the following results:

Class 1 : Watertight Structures Or Structures Exposed To

Aggressive Atmosphere

The rules for checking the cracking for class 1 are based on the condition:
k O.1 mm
wmax
i.e., with reference to the expression derived in Section 5.2.5and assuming
the 'tie-rodanalogy' to be valid (Section5.2.5):

Now

and the average anchorage bond stress zo can be replaced by:

q5,,,,,, # ?.
3 mo .-
E, .-
1 (forplain bars)
u, 10
&,,, 5,wo.-
k,, E, .-
1 (fordeformed bars)
a, 10
The maximum tensile stress o,in the steel at the crack under consideration
is, as a first approximation,taken as equal to the basic strength of the steel,
i.e.,equal to the guaranteed elastic limit divided by the reduction coefficient
relating to the limit state of cracking.Hence:

If the modulus of elasticity of the steel is taken as E, = 2.1x lo4 kg/mm2,

 305
the check criteria for cracking can be presented in the following simple form:

I
5 O00
4mmk ~
. w o (forplain bars)
ce
Class 1
4m,+ F
ce
.
w o (fordeformed bars)

where the elastic limit ce of the steel is expressed in kg/mm2.

Statistical comparison of the results thus obtained with experimental
measurements obtained from a large number of tests confirm the validity
of this simple formula.
However,it has been found in actual practice that it is advantageous to
make this cracking criterion:
a little less severe for members with a low percentage of reinforcement, in
which there is no obvious reason for the designer to provide large-diameter
reinforcing bars ;
a little more severe for members with a high percentage of reinforcement, in
which the designer would tend to use bars of very large diameter in order to
simplify the steel-fixingand achieve economies of material and labour.
It has, in a way, been found advisable to make some adjustment to the
normal theoretical formula so as to give the designer a larger ‘safety margin’
in cases where high percentages of reinforcement are provided,for in such
cases the cracking phenomena are generally more dispersive in character
and liable to have more serious consequential effects upon the behaviour
ofthe structure.The significance ofsuch an adjustmentis moreover apparent
from the first investigations conducted by the Committee on Cracking of
the Comité Européen du Béton (1957-1959).
For practical purposes this ‘adjustment’consists in introducing into the
condition for cracking a homographic function of the local geometrical
percentage of reinforcement wo in the ‘tie-rod’in lieu of the initial linear
function,the coincidence between the ‘theoretical’and the ‘adjustedcondi-
tion being obtained for a percentage wo of the order of 0.05-0.06.
For these conditionsthe criteriafor checking the cracking can be presented
in the following form:

4,mk 7500‘l+l0wo
ge
wo (forplain bars)
Class 1
4mm+E w o (for deformed bars)
ce .1 + loa,
where the elastic limit ceof the steel is expressed in kg/mm2

Class 2 Unprotected ordinary structures

The rules for checking the cracking for class 2 are based on the condition:
w,,, k 0.2mm
 306
4
Maximum
mm Plain Deformed bars
bo rs
@ 20I I I I ) I I I I I I
30 30 I I ) I I l I
LO I l I I I I I I I
50' I
''l''I6\

e \\

Figure 5.2.Class 1
 307
d
maximum
mm

@3 -
20
Plain
bars
3(
Deformed bars

8 o

o o

@ @
o o
o o
@ @
@ Non-- -systemat
-
. -_ - -
c9
@ !O_ - _ -3c
. -
I

Piain
bars
-
I , , ,

Deformed bars
-
Elastic
limit
I bars
kg/mm2ì
:haraderistic
value
Figure 5.3Class 2
308

mm I Piain bars Deformed bars

@'

__---
Non-
systematic
r 0king -
.C

Eiastic
o - , -3- , Limit
*
h bars
Plain bars Deformed bars (kg /mm2)
Char ac ter is
value
Figure 5.4.Class 3
 309

maximun
mm Piain
Deformed bars
bars
l i i l ~ i i l l
!O 3(

!Lastic
.¡mit
Piain ]bars (kg/mm2ì
Deformed bars :characterist ic
bars
value)

Figure 5.5.Members in permanent contact with water

310
Hence:
'Oo
4 m m + l5-
ae .I + IOWo
(forpiain bars)

where the elastic limit ae of the steel is expressed in kg/mm2

Class 3: Protected Ordinary Structures

The rules for checking the cracking for class 3 are based on the condition:
w,,, +0.3 mm
Hence:

&,,,,+ a, '1+10mo
m g (for plain bars)
Clais 3

I 4,,+ -500 m g (fordeformed bars)

ae 1 + lomo
'

where the elastic limit a, of ihe steel is expressed in kg/mm2.

Practical Design Graphs

The conditions in Section 5.3.2 are represented in a simple and convenient

manner in the three sets of graphs (Figures 5.2-5.4) relating respectively
to the classes 1, 2 and 3.
It is important to be able to take the swelling of the concrete into account
in all cases where this swelling is of a systematic character,i.e.,where one of
the faces of the member under consideration is in permanent contact with
water or with an atmosphere saturated with water vapour. The swelling
ofthe concrete will then have a favourable effect in that it tends to reduce
the width of the cracks.It can be taken into account by substituting for the
characteristic elastic limit a fictitious elastic limit equal to:
a,-5 kg/mm2in the case of plain bars
{ o, -8 kg/mm2in the case of deformed bars
Generally speaking,this clause relates only to class 1. The corresponding
design rules are embodied in Figure 5.5.
 6
CALCULATIONS OF
FLEXURAL DEFORMATIONS

6.1 RECAPITULATION OF THE FUNDAMENTAL

ASSUMPTIONS FOR THE CALCULATION OF DEFORMATIONS
The calculation of the deformations of prismatic structural members loaded
in bending and compression should take account as accurately as possible
of the various physical and mechanical phenomena that characterise the
elasto-plasticbehaviour of the concrete in compression and the cracking
of the concrete in tension.
In view of the uncertain and random character of many parameters,the
deformations can, with sufficient accuracy for practical purposes,be cal-
culated by means of the conventional methods of Strength of Materials
based on the application of the elastic theory,subject to applying the three
following fundamental assumptions:
(a)The geometrical cross-sections of the member under consideration
should first be made ‘homogeneous’by multiplying the steel cross-
sectional area by the modular ratio,i.e.:
Ea
B+-.A
Eb
where A and B denote the steel and concrete cross-sectional areas
respectively, E, denotes the modulus of elasticity of the steel (for
which a value of 2 100O00 bars should be adopted)and Ebthe modulus
of elasticity of the concrete (for which the instantaneous modulus
Eboor the long-termmodulus EL, should be adopted,according as
loads of short or long duration are involved).
(b)The basic values of the tensile steel strain E, at the various sections
should,for the purpose of introducing them into the conventional
calculation of the deformations in the case of a member that is not
entirely in compression,take account of the cracking of the concrete
31 1
312
in tension and also allow for the corresponding effects of the bond of
the main tensile reinforcement.
(c) The basic values of the compressive concrete strain E; at the various
sections should,for the purpose of introducing them into the con-
ventional calculation of the deformations,take account of the instan-
taneous plasticity, the long-term plasticity and the shrinkage of the
concrete.

6.2 DETERMINATION OF THE BASIC STEEL AND

CONCRETE STRAINS
The basic values of the steel and concrete strains to be introduced into the
general analysis for the limit state of deformation of a member loaded in
bending and compression should be referred to the ‘basicstrength’of the
steel and the concrete,respectively, corresponding to that limit state (in
accordance with Sections 4.3.2and 4.3.3of Part 1).

6.2.1 D E T E R M I N A T I O N OF T H E S T E E L S T R A I N

The determination of the tensilesteel strainE, in the limitstate ofdeformation

of a member loaded in bending and compression,but not in compression
over the entire cross-section,should take into account as accurately as
possible the cracking of the concrete in tension and the corresponding
effects of the bond of the main tensile reinforcement.In practice, for each
section considered,the steel strain E, in the limit state of deformation should
be taken as:

where 0, denotes the basic tensile strength of the steel, Ob the basic tensile
strength of the concrete, w o the geometrical percentage of main tensile
reinforcement with reference to the embedment section of the concrete,
and E,the modulus of elasticity of the steel (takenas 2 100O00 bars).
The first term of this expression represents the elastic strain of the steel
assumed not to be embedded in concrete.The second term takes account
of the effects of the bond of main tensile reinforcement in the zone where
cracking occurs.This calculation is based on the general theory of cracking
as set forth in Chapter 5 of Part 2 of this Manual.

6.2.2 D E T E R M I N A T I O N OF T H E CONCRETE STRAIN

The determination of the compressive concrete strain E; in the limit state of

deformation of a member loaded in bending and compression should take
into account as accurately as possible the various physical and mechanical
phenomena that characterise the elasto-plasticbehaviour of the concrete
 313
in compression.In particular,it should take account of the instantaneous
plasticity,the long-termplasticity (creep)and the shrinkage of the concrete.
In practice,for each section considered,the concrete strain E; in the limit
state of deformation should be taken as:
E; = &;o +&;i +E;, +E;,
where the terms on the right-hand side have the following meanings :
(a) &bo is the instantaneous compressive elastic strain (shortening)of the
concrete:

where ób denotes the basic compressive strength of the concrete and

ELo the instantaneous modulus of elasticity of the concrete.
(b)&Li is the instantaneous plastic strain of the concrete,for which the
following overall value is to be adopted:
-=’
&Li = 0.15~;~
=0
.
1
5
:
E bO
The average value 0.15is derived from a consideration of the compressive
stress-straindiagram of the concrete (see Figure 6.1).
O n the assumption of a parabolic curve,the total instantaneous compres-
, +&Li under working load conditions can be estimated,in terms
sive strain &

Figure 6.1

of the ultimate maximum compressive strain EO of the concrete,by means

of an expression of the following type:
&;o+&;i = &u[1:.71]
where y denotes the reduction coefficient yconcretefor the ultimate limit
state.
Since &i0 = (4,/2y),we can determine the value of ELJ JE^^).
314
Depending on the value of the reduction coefficient yconcrete,which ranges
from 2.10to 2.50,the ratio &bi/&bO varies from 0.135to 0.160,which justifies
the adoption of the average value of 0.15.
(c) &Lm is the long-term compressive plastic strain of the concrete,which,
if no experimental creep measurements are available,should be taken
as the following average value:

&km = 2*.&Lo = 2*. ob

where i,b denotes the ratio of the long-term (sustained) load to the
total load on the member considered.
The total long-termcompressivestrain (creepshortening)of the concrete,
which is applicable only to the proportion $of the load,is equal to:

Hence :
-,
o
&bO+&b, = (1+2*).+ = (1+2$),4 0
E,,
and therefore :

&;, = 2*.&bo = 2*. o

b

The estimation of $ does not present any practical difficulty.With refer-

ence to the definition of characteristic loadings it can be said that the ratio
t,ú represents the proportion constituted by the permanent loads and fixed
superimposed loads S, in relation to the total characteristic loadings that
have to be considered in the calculation of the deformations.Hence we have,
as a first approximation:

s,
* = s,+ 1.20SP1 for buildings

s,
* = S, + 1.3OSp, for other structures

A further simplification can be adopted in the case of ordinary buildings,

not of an exceptional character,for which the following approximate values
may be used:

*$$=== a fffooorrr service floors of buildings for public use

service floors of buildings for private use
roof structures of all buildings
 315
(d)&ir is the shrinkage of the concrete,measured under conditions strictly
comparable to those on the construction site.
In all,for each section of a member loaded in bending or compression,
the compressive strain ofthe concrete in the limitstate of deformationshould
accordingly be taken as:

E; =

6.3 GENERAL CALCULATION OF DEFLECTION C U R V E S

AND DEFLECTIONS
The general calculation of deflection curves and deflections consists in:
(a)establishing,for a sufficient number of sections distributed along the
member considered,the geometrical expression for the curvature as
a function of the basic strains of the steel and concrete in the limit
state of deformation;
(b)deducing therefromthe relationship for the variation of the curvature
over the entire length of the member;
(c) determining the deflection curve by means of a double integration.
The maximum ordinate of the deflection curve defines the deflection of the
member.

6.3.1 C R A C K E D M E M B E R S

For members subjected to simple bending or composite bending which are

partially in tension,and cracked,the curvature at any particular section
with the abscissa x is equal to:

where d2f /dx2 is the second derivative of the deflection curve with respect
to the abscissa ofthe section considered,while E, and E; are the basic steel
and concrete strains estimated for the limit state of deformation,and h
denotes the effective depth of the section.
O n replacing E, and E; by the values derived for them earlier on in this
chapter,we obtain:
-
-
a,-- gb

2mo+(1.15+ 21))
--
d2f Ea
dx2 - h
316
and thence,by double integration,the function f(x)definingthe deflection
curve.
The deflection of the member is defined as the maximum value of f(x).

6.3.2 U N C R A C K E D MEMBERS

For members subjected to eccentric compression which are entirely in

compression,and not cracked,the curvature at any particular section with
the abscissa x is equal to:
1-
-
-~ 1 I 1 I
d2f - ~b - & L i
r dx2- h*
where d2fldx2is the second derivative of the deflection curve with respect
to the abscissa of the section considered,E; is the basic compressive strain
of the concrete at the most compressed fibre (corresponding to the basic
strength $b of the concrete;cf. Section 6.2.2),&Li is the compressive strain of
the concrete at the least compressed fibre (corresponding to the concrete
stress obiat that fibre,calculated by the same procedure as indicated in Sec-
tion 6.2.2),and h, denotes the total geometrical depth of the section.
O n replacing E; and &Li by the values derived for them,we obtain:
d'f (1.15+2$) (ob -abi)
a?= h, Go
'

where the value of obiis obtained from the stress distribution diagram of the
concrete at the section considered.
By double integration we then obtain the function f(x)defining the deflec-
tion curve.
The deflection of the member is defined by the maximum value of f(x).

6.4 SIMPLIFIED CALCULATION FOR ORDINARY

BUILDINGS
In the case of buildings for public or private use and not of an exceptional
character the general calculation of deflection curves and deflections can
be replaced by a simplified calculation of the maximum deflection in the
limitstate ofdeformation,applicableto allmembers loaded in simplebending
and with a mechanical percentage of tensile reinforcement not exceeding
0.25.
If these conditions are satisfied,the margin of approximation (inrelation
to the results obtained by applying the general calculation) will not exceed
f20 %. On the other hand,if they are not satisfied,e.g.,in members contain-
ing a very high percentage of reinforcement, the designer should use the
general calculation procedure set forth in the previous section.
The simplified calculation consists in calculating the total maximum
deflectionf as the sum of two partial deflectionsf,and fII:
(a)The first partial deflection fi relates to the uncracked state (called
 317
‘stateI’)and representsthe deflection attained at the timejust before cracking
of the tensile zone of the member occurred:

where MI denotes the value of the bending moment at cracking (estimated

for the ‘homogeneous’section and with reference to the tensile strength o.
of the concrete), Ek0 the instantaneous modulus of elasticity of the concrete,
I, the moment of iflertia of the homogeneous section,$the proportion of the
permanent loads and fixed superimposed loads in relation to the whole of
the characteristic loadings,1 the span of the member, and B the numerical
coefficient occurring in the standard deflection formulae based on the
theory of elastic behaviour.
The general expression for the deflection of an elastic member is:
f=ß.-.l M 2
E.I
The coefficient fi depends more particularly on the structuralarrangements
and on the nature and mode of application of the loads and superimposed
loads.In ordinary buildings the loads are often uniformly distributed,and
for such loads the coefficient fi has the following values:
& for a freely supported beam
& for a beam rigidly restrained at both ends
&for a beam freely supported at one end and rigidly restrained at the
other
i for a cantilever
(b) The second partial deflectionf, relates to the cracked state (‘state11’)
and represents the deflection attained after cracking of the tensile zone of
the member has occurred :

where M denotes the total bending moment corresponding to the whole of

the characteristic loadings,E,the modulus of elasticity of the steel (forwhich
a value of 2 100 O00 bars should be adopted), A the cross-sectionalarea and
Q the mechanical percentage of the main tensile reinforcement.The other
symbols are as previously defined.
(c) Finally, the total deflection f = fI +frImust not exceed the following
limit value:

An additional simplification may be adopted if M I<3M or if the member

under consideration has a compressiveflange whose effective width is more
than five times the width of the rib (web).In either case the deflectionfcan
be taken as equal to the following limit value:
M
f = B.(1+2$). .l2
E,. A . h2(1 -2Q) (i -+.I)
318
This last-mentioned simplification is applicable to the great majority
ofribbed floorsofordinary buildings.The approximate values for $indicated
in Section 6.2.2may be adopted for these.
However,an uncertainty may occur in the estimation of the coefficient ß
because of the possibly varying degree of continuity between adjacent floor
slabs or because of the incompleteness of their fixity in the external walls.
O n account of this the coefficient (ß)is generally less than (which is the
&
value relating to freely supported members) but remains above (the value
relating to members completely fixed at both ends). For practical purposes
the average intermediate value may be adopted :
ß
=
' 12

Adopting this value for ß and taking M = a,. A.z = (a,.A.z/1.80)we

obtain the following simplified expression for the maximum deflection:

or

or, approximately:
Ge 1
1.80.E,( 1 -2m) h '

For the check condition for the limit state of deformation f 11 <(flZ),imi, w e
can therefore substitute the conditionfor the limit value of the slenderness ratio:

where the guaranteed minimum elastic limit a, of the steel is expressed in

bars.This conditionis applicable only to flexuralmembers with a mechanical
percentage of reinforcement m not exceeding 0.25.
First example:
Service floor of a residentialbuilding,reinforced with mild steel bars:
-
1
max = 150

4 3 x 107 i
(i)max i
x-x-(~-~w)
= 2.4x lo3 1.50 150
= 83(1-2~~)
 319
Second example.
Service floor of a school building,reinforced with grade 40 steel bars:

(i)max 4.5x 107 i i

x-x-(~-~w)
= 4 x lo3 1.40 300
= 27(1-2~7)
 7
SHRINKAGE AND CREEP OF
CONCRETE

The shrinkage and creep values indicated in the following are given for
approximate guidance only,and they are valid only for concretes made with
ordinary Portland cement which harden under normal conditions and which,
under working load,are subjected to stresses not exceeding about 40-45%
of the compressive strength of the concrete.
Accordingly, these values and also the curves representing the effects of
the various shrinkage and creep parameters should be used with caution,
as significant differences are liable to occur,depending on:

(a)the geometrical properties and climatic conditions;

(b)the modulus ofelasticity of the aggregates;
(c) the nature of the cement;
(d)the density of the concrete;
(e) the treatment of the concrete (heating,curing,etc.).

7.1 SHRINKAGE

The shrinkage coefficientE: of the concrete at an arbitrary instant t can be

estimated by multiplication of the five following factors:

E: = lj . CI,. p,(l-O~lOaO)p,

where :
lj is the ‘basicshrinkage coefficient’,defined as a function of the relative
humidity ;
CI, representsthe influence ofthe leastdimension ofthe member considered;
/3, represents the influence of the composition of the concrete;
320
 321
(1 -O.1Oar0)represents the influence of the geometrical percentage of rein-
forcement of the member;
pt represents the influence of time.

7.1.1 BASIC SHRINKAGE COEFFICIENT

The ‘basicshrinkage coefficient’$ of concrete varies as a function of the

relative humidity,in accordance with the diagram in Figure 7.1, which is
valid for unreinforced concrete.
The values of the coefficient t,b may be taken as conforming to the average
curve in this diagram.
These values should not be applied to floorswith built-inheating elements
io54
70
60 95% fractile of
all the tests
50
LO
30 5% fractile of
20 all the tests

10
O
-10
100 90 80 70 60 50 LO 30%
Figure 7.1

and to concrete members exposed to the heat of furnaces.In such cases it is

necessary to base oneself on direct experimental measurements.

7.1.2 I N F L U E N C E OF THE L E A S T D I M E N S I O NO F T H E
MEMBER

To define the influence of the least dimension of the member upon the
shrinkage, the notion fictitious thickness d, will be introduced. This is
defined as the area B of the section divided by the halfperimeter assuming
p to be the perimeter in contact with the atmosphere.This definition,which
is essentially applicable to a circular section,may be extended to comprise
other cross-sectionalshapes as well (asshown in Figure 7.2).
It is seen that,if one of the dimensions ofthe section is very large in relation
to the other, the fictitious thickness is very nearly equal to the actual
thickness.
322
The diagram in Figure 7.3gives the average values of the coefficient CI,
as a
function of the fictitious thickness d, expressed in cm.The diagram also

b
re a

Figure 7.2

Fictitious thickness d m
o 10 20 30 LO I (in cmi
Figure 7.3

shows the associated range of dispersion (scatter)as experimentally estab-

lished.It appears that the shrinkage proceeds more rapidly as the fictitious
thickness of the member lessens.

7.1.3 INFLUENCE OF THE COMPOSITIONOF THE CONCRETE

Figure 7.4gives the average values of the coefficient B, as a function of the

water/cement ratio,for a cement content ranging from 250 to 450 kg/m3.
 323

1
3.0

95% fractile for

!I
2.0 C = 350 -L50 kg/m3

5% fractile for
1.0
-
C = 350 L50 kglm3

.6 0.8 cement
Water
O 0.2 0.L ’
Figure 7.4

‘i
p= .
(.

years
Figure 7.5

L.0

3.0

-3
2.o

1.0

O
100 90 80 70 60 50 LO 30 20 %
Figure 7.6
324
It also shows the associated range of dispersion (scatter),as experimentally
established,for a cement content of 350-450 kg/m3.

7.1.4 INFLUENCE OF TIME

The average values of the coefficient pt which expresses the variation of the
shrinkage as a function of time for constant climatic conditions are given
in Figure 7.5.
With this diagram it is also possible to determine the amount of shrinkage
deformation that occurs in any particular time interval (t, -ti), namely :
E‘ ?(Pin -Pii)
It should finally be noted that these values are valid in the case where the
concrete is not protected. If, on the other hand, suitable moist curing is
applied on the site,the shrinkageis reduced by about 50%while the concrete
is still very young.In this way it is possible to obviate cracking of this con-
crete, even though its tensile strength at this stage is still very low. This
reduction of the shrinkage diminishes in course of time,however;at an age
offour months it is no more than about 10% and tends gradually to disappear
entirely over a longer period of time.

7.2 CREEP
The creep coefficient E; of the concrete at an arbitrary instant t can be
estimated by multiplication of the six following factors:
E; = I).a/. . ßs(l -O.1OWo)~. pr
where
I) is the ‘basic creep coefficient’,defined as a function of the relative
humidity;
a,f representsthe influenceofthe leastdimension ofthe member considered;
ßJ represents the influence of the composition of the concrete;
(1 -0.10~~) represents the influence of the geometrical percentage of rein-
forcement of the member;
representsthe influence of the age of the concrete at the time of loading;
pi represents the influence of time.
The creep coefficient E> is applicable to the instantaneous strain (elastic
shortening) due to a compressive stress os, assumed to be of constant
magnitude,which causes creep of the concrete.Hence:
E;, = E; 0s
.y= ES. O’ (inbars or kg/cm2)
E,, 21 OOOJOJ

7.2.1 B A S I C C R E E P COEFFICIENT

The ‘basiccreep coefficient’I) of the concrete varies as a function of the

relative humidity,in accordance with the diagram in Figure 7.6,which is
valid for unreinforced concrete.
 325
The values of the coefficient iI/ may be taken as conforming to the average
curve in this diagram.

7.2.2 I N F L U E N C E O F T H E L E A S T D I M E N S I O NO F T H E
MEMBER

The diagram in Figure 7.7gives the average values of the coefficient as as a

function of the fictitious thickness d, (asdefined in Section 7.1.2and ex-
pressed in cm).It also shows the associated range of dispersion (scatter)as

1.2

1.0

O.8

0.6

Fictitious thickness dm
O 10 20 30 LO 50 (in crn)
Figure 7.7

experimentally established.In comparison with the diagram for CI, (Figure7.3)

it appears that creep is less sensitive than shrinkage to the influence of the
fictitiousthickness,i.e.,the influence of the least dimension of the member.

7.2.3 I N F L U E N C E O F T H E C O M P O S I T I O NO F T H E C O N C R E T E

Figure 7.8gives the average values of the coefficient Pr as a function of the

water/cement ratio,for a cement content ranging from 100 to 500kg/m3.It
also shows the range of scatter,as experimentally established,for a cement
content of 300 kg/m3.
For estimating ßf the following empirical formula, which is valid for
concretes as ordinarily used,can be employed:
Let W/Cdenote the water/cement ratio and v w + c the percentage (by
volume) of the cement paste contained in the mix.W e can put:

ßf = 7.-WC ’vw+c
326
and furthermore:

so that:

pf=7.- -+- -
o
ww1 c
c c 3 1000
where C denotes the cement content in kg/m3of concrete.

4
3.0

2.0

1.0

Water
Cement
O 0.2 04 0.6 0.8
Figure 7.8

Thus,for concrete with a cement content of 400kg/m3and a waterlcement

ratio of 0.5:
p -7.-1 -+-
f - o
1 1 400 1 5 2 7
-=7x-~-x-=-=1.17
2 2 3 1000 2 6 5 6

7.2.4 I N F L U E N C EO F T H E A G E O F T H E C O N C R E T E A T T H E
TIME OF L O A D I N G

The age of the concrete at the time of application of loading to it exercises

at least as great an influence as do the climatic conditions.This influence is
represented by the coefficient whose values, and the associated range of
dispersion (scatter),are indicated in Figure 7.9,which is valid for normal
conditions of hardening at an average ambient temperature of 20°C.
 327
If the average ambient temperature T differs from 20°C,the time t,,
calculated by linear extrapolation and expressed in days,should be adopted
as the age of the concrete at the time of loading (inlieu of the actual age):
T+ 10"
t, = CAt -.
30"
where CAt denotes the hardening time at the temperature T

7.2.5 I N F L U E N C EOF T I M E

The values ofthe coefficient ptrepresenting the variation ofcreep as a function

of timeare virtually identicalwith the corresponding coefficientfor shrinkage
(Section7.1.4) and are given by the same diagram (Figure7.5).
The shortening due to creep at an arbitrary intermediate instant t,
(preceding the instant t, which corresponds to the end of the creep pheno-
menon) under the influence of a sudden stress variation applied at an
arbitrary instant ti is equal to:
. $. pJ(1 -0.10.wo)lti.p(t, -ti)
.

In more general terms the total shortening &in (elastic shortening+

shrinkage shortening+creep shortening) at an arbitrary intermediate
instant t, (precedingthe instant t, which correspondsto the end of the creep

Age' at loading
1 3 7 28 90 360 days
Figure 7.9

phenomenon) under the influence of a load applied at the instant t, and

subject to sudden variations of intensity at arbitrary instants such as ti can
be expressed by :
E;" =Eqi+*. c(J. pJ(l-O.lO. w0)lt1 .p(t,-tl)]+ ...
...+ 1 + $.M~ .fis( 1 -O.10 .mTlg)[ti.p(t, -ti)] +C&:(P~,
-PtJ
328
where:
E:,, represents the elastic shortening caused by the load applied at the
instant t,
the variation of the elastic shortening caused by a sudden variation
in the intensity of the load at an arbitrary instant ti
(t, -ti) the intervalof time between application of the load and the instant
at which the creep shortening is estimated
(t, -ti) the interval of time between the arbitrary instant ti at which the
load undergoes a sudden variation in intensity and the instant at
which the creep shortening is estimated
This general expression can serveas a basis for estimating the deformations
of structuralmembers and structures.
 8
DESIGN OF SLABS AND
PLANE STRUCTURES

8.1 SUBJECT AND FIELD OF APPLICATION

8.1.1 DEFINITION O F T H E E N V I S A G E D LIMIT STATE

The safety of a structure should be analysed with regard to the various limit
states corresponding to the respective criteria of unserviceability (unfitness
forservice).
In the particular case of a plane structure loaded perpendicularly to its
middle plane the limit states to be considered are in general:
(a) the limit state of cracking;
(b)the limit state of deformation;
(c) the ultimate limit state (failure).
The purpose of the present chapter is to supplement the Sections relating
to plane structures in Part 1 of this Manual,more particularly with regard
to the analysis of the ultimate limit state. The limit states of cracking and of
deformation will not be considered here.

8.1.2 D E F I N I T I O NOF T H E E N V I S A G E D M O D E OF F A I L U R E

In the case of plane reinforced concrete structures the ultimate limit state
may correspond to either of the two following modes offailure:
(a)failure by punching shear;
(b)failure by exhaustion of the capacity to resist bending.
329
 330
The present chapter is concerned solely with the analysis of the ultimate
limit state corresponding to failure by exhaustion of the flexural capacity,
i.e.,the capacity to resist bending.
The ultimate limit state of failure by punching shear should be checked in all
cases where locally concentrated loads are acting (superimposed loads or
bearing reactions) by direct application of the rules given in Section 6.2.5of
Part I of this Manual. In the present chapter it willbe assumed that the analysis
for punching shear has been carried out and that therefore the safety with
regard to punching shear for the plane structures under consideration is
ensured.

8.2 ULTIMATE LIMIT STATE C O R R E S P O N D I N G TO FAILURE

BY E X H A U S T I O N OF THE FLEXURAL CAPACITY
(FLEXURAL FAILURE)
8.2.1 G E N E R A L D E S C R I P T I O NO F T H E B E H A V I O U R O F
P L A N E REINFORCED C O N C R E T E S T R U C T U R E S U P T O
F L E X U R A L FAILURE

If a plane reinforced concrete structure is subjected to gradually increasing

superimposed loads, it will, in general, exhibit the following successive
stages of behaviour.*

Stage Of Elastic Behaviour

In this first stage of loading,the distribution of the bending moments cor-
responds to the elastic distribution.

Stage Of Cracking
With further increase of the superimposed loads the cracking that develops
in the tensile zones of the concrete leads to a gradual reduction of the
moments of inertia of the cracked sections.This reduction is reflected in an
alteration of the distribution of the bending moments in that the moments
in the uncracked zones increase more rapidly,for equal load increments,
than they did before cracking occurred.
So long as the reinforcement remains within the range of elastic deforma-
tions,the width of the cracks will be limited.

Stage Of Plastification
Provided that the percentage of tensile reinforcement is low enough-or,to
be more precise,ifit is below the ‘uppercritical percentage’,which is generally

*In accordance with Section 8.1.2,the structure considered here is assumed to fail as a result of
exhaustion of the flexural capacity (the capacity to resist bending).
 331
the case in plane reinforced concrete structures -this reinforcement will,
with continuing increase of the superimposed loads, gradually develop
plastic behaviour i.e., yielding, in the zones where the largest bending
moments occur.
The sections where the reinforcement has reached its yield point continue
to deform,but the bending moment at such sections no longer undergoes
any appreciable increase,and because of this there occurs a greater amount
ofmoment redistribution than in the previous stage.
Plastification spreads gradually along narrow strips where the widest-
open cracksare concentrated.These stripsmay be regarded as lines,so-called
‘yield lines’.They develop according to a pattern which depends more
particularly upon the shape of the structure,its support conditions, the
distribution of its reinforcement,and on the method of loading.The yield
line pattern as a whole is composed of straight-linesegments.

Stage Of Failure
When the yield lines have developed to such an extent that the slab has
become a ‘mechanism’,any very slight further increase of the superimposed
loads will give rise to an unstable state of equilibrium.The structure will then
continue to deform by rotation about the yield lines,until the rotation in
certain zones reaches such a value as to cause destruction by crushing of the
concrete in compression.The failure of these zones results in gradual exten-
sion of the crushing of the concrete along the entire length of the yield lines,
with a corresponding loss of the load capacity of the structure.The yield
lines therefore also constitute the ‘fracturelines’of the structure.
In Figure 8.1are shown,by way ofexample,the experimentallydetermined
and the ‘idealised’yield line pattern of a simply-supportedsquare slab,not
anchored at the corners,subjected to four symmetrically placed concentrated
loads.

8.2.2 G E N E R A L D E S C R I P T I O NOF T H E M E T H O D S F O R
U L T I M A T E LIMIT STATE ANALYSIS

Introduction

Concerning the ultimate limit state of plane structures loaded perpendicu-

larly to the middle plane,Section 5.2.1of Part 1 of this Manual states:
‘For checking the ultimate limit state,methods which take account of
the staticallyindeterminate effect ofplasticity,more particularly the so-called
yield line theory,can permissibly be employed,on condition that:
(a)the yield line pattern of the structure under consideration is justified
with certainty or is determined by means of appropriate tests;
(b) the basic assumptions of these methods are really fulfilled;
332
(c) the set of loads under consideration correspondsto the most unfavour-
able arrangement of these loads.’

Elasto-PlasticTheory
For the sake of completeness, mention must be made of the elasto-plastic
theory,which is the most general method of analysing the behaviour of a
plane structure from the commencement of loading until failure.This theory
takes account ofthe effects of inelastic deformations (dueto cracking or to
plastic deformations properly so called) with a view to estimating the
correspondingredistributions.
The principle of a practical method of applying this theory was published
as far back as 1950by Professor Franco Levi.’ Since then the method has
been developed by Messrs. Franco Levi and C.E. 26 A similar
method is being elaborated under the direction of Professor Ch.Massonnet
at the University of Liège.31,32
With the aid ofthe elasto-plastictheory the behaviour of the structure can
be analysed not only at failure,but also under working loads.At the present

Figure 8.í(a)
 Figure 8.I(b). Yield line pattern of a simply-supported square slab, not anchored at the
corners, subjected to four symmetrically placed concentrated loads (test performed by J. C.
Maldugue at the Structural Testing Centre at Saint Remy-les-Chevreuse).(a) Photograph
of the slab after failure (underside).(b)idealised yield line pattern

time the theory is relatively complex in application,but the results obtained

are very promising and should lead to its practical application in structural
design in conjunction with the use of electronic computers.

General Theory Of Limit Analysis

The general theory of limit analysis is concerned solely with the ultimate
limit state.
For the purpose of this analysis the actual material is replaced by an
idealised material which is assumed to be ‘perfectlyplastic’.According to the
authors this material is considered as ‘elasticplastic’or ‘rigid plastic’ (see
Figure 8.2).
Whichever of these two hypotheses is adopted,the general theorems are
the same.However,in the case of plane reinforced concrete structures the
hypothesis of a ‘rigidplastic’material is,in general,implicitly adopted.
334
It should be noted that a distinctive feature of the limit analysis is that it
does not, generally speaking,result in the determination of one particular
ultimate load,Actually,the theory shows the existence of a lower limit and
an upper limit between which the actual ultimate load is situated.Only in
a certain number of cases is it possible to obtain a result in which these two

(a) 3 Ibl 3

Figure 8.2.(a) Elastic-plastic material, (b) Rigid-plastic material

limits of the ultimate load coincide,so that a so-called‘complete’or ‘exact’

solution emerges.
The limit analysis is applicable to any system consisting of a perfectly
plastic material.However,for the sake of clarity,the fundamental concepts
and theorems will,in the following,be stated for the particular case of the
analysis of slabs.
The following two fundamental concepts are of significance in connection
with the theory of limit analysis:
In the limit state of plastification:
(a)the bending moments in the slab form a ‘staticallypermissible’field,
i.e.,these moments are in internal equilibrium,equilibrate the applied
loads,and are such that at no point ofthe slab does the bending moment
exceed the plastification moment m ;
(b)the yield mechanism is ‘kinematicallypermissible’,i.e.,this mechanism
satisfies the support conditions of the slab and the forces in it do
positive work (whichis dissipated as heat in the plastic deformations).
For any particular slab there exists an infinitenumber offields of statically
permissible moments that can equilibrate a given system of loads.
Similarly,a given system ofloads may correspond to various kinematically
permissible mechanisms.
To each of the fields of statically permissible moments corresponds a
definite value P of the load,as defined by the equations of equilibrium at
the perimeter.
Also,to each kinematically permissible mechanism correspondsa definite
value P of the load,as defined by the condition that the work done by the
external loads must be equal to the work done by the internal forces.
The two fundamental theorems of limit analysis may be stated in the
following form:
(a) Lower bound theorem (or kinematic theorem): The limit load is the lower
 335
bound of all the loads P corresponding to the various kinematically
permissible mechanisms.
(b) Upper bound theorem (or static theorem): The limit load is the upper
bound of all the loads P corresponding to the various fields of statically
permissible moments.
These two theorems are completed by the following corollary (theoremof
uniqueness): If it is possible to combine a kinematically permissible yield
mechanism with a-fieldof statically permissible moments, the c o m m o n load that
corresponds to both of them will be the exact limit load.
The solution obtained in that case is called the complete solution or
exact solution.

Application Of The Limit Analysis Theory To The Case Of

Plane Reinforced Concrete Structures
Conditions Of Application

Although reinforced concrete does not strictly conform to the definition of a

‘perfectlyplastic’material, the theory for such a material can nevertheless
justifiably be applied to it with a fair degree of approximation,provided that
the percentage of reinforcement is sufficiently low.To put it more precisely,
this percentage should be below the ‘lowercritical percentage’which marks
the borderline between failure due to plastification (yielding)of the tensile
reinforcement and failure due to crushing of the compressivezone.
For practical purposes the tensile reinforcement percentages in plane
structures always satisfy this condition.
In that case the successive stages in the behaviour of a reinforced concrete
slab under loads which are increased up to failure will correspond to the
description relating to the particular case considered in Section 8.2.1.As
already explained,failure occurs when the slab has become transformed into
a ‘mechanism’following the gradual spreading of plastic behaviour (yielding)
of the reinforcement along the ‘yieldlines’.
Strictly speaking,the limit analysis theory is applicable only to structures
reinforced with steel that has a definite yield point.However,tests performed
on slabs” and on flat slab floors and mushroom floors” reinforced with
steel that exhibits no definite yield point in the stress-straindiagram (high-
tensile ordinary or cold-worked steels) have shown that an approximation
which is quite sufficient for practical purposes is obtained by using the limit
analysis method and adopting for the yield point the 0.2% proof stress,i.e.,
the stress at which the permanent strain is equal to 0.2%.

Criterion Of Plastic Behaviour

The criterion of plastic behaviour (‘plastification’)for reinforced concrete

slabs is, in the general case,more complex than for slabs consisting of a
homogeneous material.
336
As a rule,reinforced concrete slabs are provided with two layers of rein-
forcement,one of which is installed near the underside of the slab and is
referredto as the positive reinforcement,while the other is installed near the
top and is referred to as the negative reinforcement.
Neither ofthese layers of reinforcementneed necessarily be homogeneous
over the entire area of the slab,and each may comprise several different
systems ofreinforcement in different zones of the slab.For example,in the
case ofa slab continuousover its supportsthe layer ofnegative reinforcement
will normally contain differentpercentages of reinforcement at the supports
and in the mid-span region respectively. Similarly, the layer of positive
reinforcementmay contain differentpercentages at the centre of the slab and
at the supports respectively.
Hence it followsthat the yield lines of the same algebraic sign,whether
negative or positive,will not necessarily be crossed by the same system of
reinforcement. It may even occur that a given yield line will encounter
different systems of reinforcement along its alignment. However, for the
present purpose,it will be assumed that any particular yield line encounters
only one system of reinforcement,though it should be noted that Jonesz3
has given a method of dealing with the case where a yield line encounters
two different reinforcementsystems.
Furthermore,in the most general case,a system of reinforcement may
comprise bars extending in r arbitrary directions. Such a reinforcement
system is called anisotropic.To each direction i ofbars correspondsa resisting
moment miper unit width of the slab;this moment can be represented by a
vector with a direction perpendicular to the direction i. The system is
completely defined if the angles a,,az,... al ... a, of the r directions of the
bars in relation to a reference direction are given,together with the resisting
moments m,,m 2 ... mi... m,.
In the particular case where the system comprises only two orthogonal
(mutuallyperpendicular) directions of bars,it is called orthotropic.
Finally, in the more restrictedly particular case where an orthotropic
system of reinforcementis so designed that the resisting moments in the two
mutually perpendicular directions of the bars are equal,the system is called
isotropic.
A slab in which each of the two layers of reinforcement,top and bottom,
contains only one orthotropic system of reinforcement is itself called ortho-
tropic if the bars in each direction in the top layer are parallel to the bars
in each respective direction in the bottom layer.Furthermore,a slab in which
each of the two layers contains only one isotropic system of reinforcement is
itself called isotropic.

Criterion of ‘stepped‘yielding (or ‘square’yield criterion) (Johansen)

Consider a yield line segment crossed by a reinforcementsystem compris-

ing bars in various directions 1,2,3 ... i ... r (seeFigure 8.3).Suppose this to
be positive yield line and let m,,m, ... m i...m,denote the resisting moments
per unit width corresponding to the various bar directions.These moments
 331
are represented by vectors situated in the plane of the slab and perpendicular
to the corresponding directions of the reinforcing bars.
Let Bi denote the angle measured in the trigonometrically positive direc-
tion from the moment mito the yield line.
Professor Johansen assumes that each reinforcing bar yields,i.e.,develops
plastic behaviour, in the yield line according to its original direction. In

Y
Figure 8.3.Reinforcing bars in direction i
other words,the two parts of one and the same bar separated by the yield
line will retain their alignment when the two slab portions,situated on each
side of that line,undergo their rotation.
Hence,for each direction i of the bars, the yield line can be considered
as consisting of an infinitesuccession of straight lengths disposed stepwise,

I
Lines in direction i
Figure 8.4.Reinforcing bars in direction i

i.e., respectively at right angles and parallel to the direction considered

(see Figure 8.4).
Suppose,to begin with,that there are bars in one direction i only.Let m,,
and m, denote the bending moment acting at right angles to the yield line
and the torsional moment at the yield line respectively,at the instant when
 338
plastic rotation occurs in the yield line.These moments are in each case the
moment per unit length of the yield line.
If it is assumed that all the bars which cross the yield line of length AB
develop plastic behaviour,i.e.,that yielding of the steel occurs in them,the
following expressions are obtained if the moments are projected on to the
direction i and on to the perpendicular to that direction:
AB.m, sin ûi-ABm,cos 8,= O
and
ABm,cos ûi+ABm,sin Bi-ABm,cos Bi = O
where
m, = micos2Bi
m,= misin Bicos ûi
In the general case of a system comprising bars extending in r directions
(1,2... i.. .r) the expressionsfor the normalbending moment and the torsional
moment at the yield line (per unit length of that line) are as follows:
i=r
C micos2ei

1 m, =

m,=
i=l
i=r
C
i= 1
Bi
misinBicos
Note:O n transformation of these expressions by introducing the angle $
of the yield line in relation to an arbitrary reference axis x and the angles
CI^, u2,... ...CI,formed with this same axis by the moments m,,m2 ... mi...
m,,the following expressions are obtained (since ûi = $-ai):
i=r i=r i=r
m, =+cos24. 1 mi.cos2ai++sin24.i1 mi.sin2cri++C mi
i=l =l 1 i=
i=r i=r
m,=+cos24. 1 mi.sin2cri-+sin24.
i=l i=l
micos2cri
These expressions show that there are two mutually perpendicular
possible directions I and II of the yield lines for which m,= O.These direc-
tions are given by:
i=r
1 mi.sin2ai
tan24, =
1 mi.cos2ui
i=l
The values of the moments acting at right angles to these yield lines-
the principal moments-are given by :
m,orm,,=+
i=r
1 mi& J[(iirmi.cos2CIi>’+(
i= 1 1 i=
i=r
1 misin2ui
i=l I’)
I
 339
Conversely,the moments m, and m,can be determined from the principal
moments m, and m,, by means ofthe Mohr’s circle construction.
Particular case:System comprising reinforcing bars in two directions only.
In that case,if one of the two bar directions-e.g.,the direction i- is
taken as the direction of the reference axis, we obtain by substituting
CI,= O, CI’= CI, m, = m,m2 = pm in the above expressions:
p . sin CI
240 = l+p.cos2c!
m,orm,, = t m [ l + p ~ + ~ ( 1 + 2 p . c 0 ~ 2 ~ 1 + p ’ ) ]
If furthermore CI = 4 2 (orthotropic system of reinforcement), the
expressions become:
tan 240 = O
m, = m
m,, = Pm
In this case the directions of the bars are the directions of the principal
moments.
The normal bending moments and the torsional moments in a yield line
forming an angle û with the moment m (i.e., with the direction CI = 742) are
then respectively:
m, = m.cos2û+p.m.sin2û
m,= m(1-p)sinû.cosû

Finally, if furthermore p = 1 (isotropic system of reinforcement), we

obtain m, = m,, = m,and Mohr’s circle dwindles to a point. Irrespective
of the orientation of the yield line,we have:
m, = m
m,= O
Discussion of Johansen’s ‘stepped’criterion
The application of the criterion of ‘stepped’yielding (‘square’yield
criterion)entails certain difficulties in the vicinity of the p,ointsofintersection
of converging yield lines associated with different systems of reinforcement.
For example,consider ajunction point where three yield lines 1,2and 3-
associated with three different reinforcement systems-converge (see Figure
8.5).The normal bending moment and the torsional moment along each of
these three yield lines have the following values:
i=rl
m,, = 1 mli.cos2eli
i= 1
i = rl
m,,=
i=l
1 mli.sin oli.cos eli
340
i = r2
mn2=
i= 1
1 m2i.cos2ûzi
i = r2
mt2= 1 m,, .sin ûZi.cos
i= 1
i = r3
mn3= 1 m3i.cos2û3i
i= 1
i = r3
mr3=
i= 1
1 m3i.sine3i.cosû3i
The three normal moments m,,,m,, and mn3completely define (according
to Mohr’s circle) the two principal moments mi and m;, at the point of

- Direction of principal
moment m’,

3
Figure 8.5

intersection. Hence it follows that the torsional moments mnsl, mns2and

mns3 acting along the three yield lines at the point of intersection are com-
pletely defined.
The values of mnsl, mns2 and mns3 in terms of the moments m,,, mn2and
mn3will now be calculated.Let the three yield lines be numbered 1, 2 and 3
(proceeding in the usual positive trigonometrical direction of angular
measurement)and let $12, $23 and $31 denote the angles (measured in the
positive direction) between the yield lines corresponding to the first and the
second subscript respectively.
Furthermore,let pi,ß2 and ß3 be the angles (measured in the positive
direction) between the direction of the principal moment mi and the yield
lines 1,2and 3 respectively.
The following relations can be written down (Mohr’scircle):

2
-mi -mi, cos 2ß1
m,, =-mi -k mi,+
2
 341
m;+mi, m;-m;,
mn2 = 2
+ 2 cos 2ß,

O n eliminating (m;+m;,)/2, (m;-m;,)/2and ßi from the first four of the

above equations, and bearing in mind that c$31= 2 ~ - ( 4 ~ 2 + 4 1 3 ) , the
expression for mnslis obtained:

Similarly,the following are obtained:

The torsional moments mnsl, mns2and mns3are therefore completely

determined in terms of m,,,mn2and mn3.
Hence,in the general case:
mns1 f mt, mnr2 + mt2 mns3 f mí3
i=r
where m,,,mt2and mt3represent the values of 1 misin ûicos ûi at each of
the three yield lines. i= 1

Consequently, Johansen’s theorem of ‘stepped’yielding does not apply

at the point of intersection.
Only in the special case where the three converging yield lines are asso-
ciated with the same reinforcement system (i.e.,the principal moments
m; and mi, at the intersection coincide with the principal moments m, and
m,, of the common system)do the following relations apply:
mns1 = mt, mns2 = mt2 mns3 = 4 3
342
In this special case Johansen’scriterion of yielding is therefore satisfied
at the point of intersection.
In the general case the difficulty arising from the non-fulfilment of this
criterion at the intersections of yield lines can be resolved by two different
solutions,which will be explained in the following:
First solution: Retention of Johansen’s criterion. Concept of ‘disturbed zones
Johansen’scriterion is retained for those parts of the yield lines which are
not close to a junction point.The moments at a yield line 1 are therefore:
normal bending moment :
i = rl
m,, = 1 mlicos2eli
i= 1

torsional moment:
i = rl
m,, = 1 m l i.sin eli.cos eli
i= 1

Furthermore,in the vicinity of a junction point of yield lines there is

assumed to exist a very small disturbed zone in which Johansen’scriterion
is not satisfied.This disturbed zone is taken into account by introducing for
each of the yield lines 1, 2 and 3 an additional torsional moment at the
junction point.These torsional moments have the following values :
-mtlfor yield line 1
mnsl
-mi2for yield line 2
mnS2
-mt3for yield line 3
mnS3
This first solution is the one which was adopted implicitly by Johansen’
and Jones,21and explicitly by NielsenZ2
Solution solution: Adoption of the ‘normal‘criterion
It is assumed that over the entire length of a yield line only the normal
moment is specified,namely :
i = rl
m,, =
i=l
1 m l i.cos2ûli
The difference in relation to Johansen’scriterion is that now the torsional
moment mnsl at a yield line may have any value.
On the basis of this criterion the intersections of yield lines present no
problem at all because the torsional moments mnswhich occur there can be
given the values obtained from Mohr’scircle construction.
‘i Lojecond solution,which has been adopted more particularly by Kemp
and Morley,23appears to be in not so good agreement with reality,since the
‘disturbed’zones do indeed correspond to actual observations.Tests show
that in the immediate vicinity of yield line junction points there are in fact
such ‘disturbed zones where the converging lines ramify so that there is,
strictly speaking,no true ‘pointof intersection’.
However,the concept of ‘disturbedzone’does not readily lend itself to
mathematical treatment. If the hypothesis of straight yield lines,which is
the basis of the simple method of analysis embodying Johansen’stheory,is to
 343
be retained, then it would appear preferable,from the point of view of
mathematical treatment,to adopt only the ‘normalmoment criterion’.
In anticipation of the methods which will be dealt with further on it should
be noted that the ‘methodof work is equally suited to the application either
of Johansen’soriginal criterion or of the normal moment criterion.Indeed,
since the torsional moments at the yield lines do not as a whole do any
work,these moments play no part in the procedure.
On the other hand,the torsional moments do come into the ‘equilibrium
method’,for the determination of the ‘nodalforces’(forcesat the junctions
of yield lines) depends on this.In the present treatment of the subject,the
normal moment criterion,as adopted by Kemp and Morley,will be used for
establishing the expressions for the nodal forces.However,the expressions
for the nodal forces derived therefrom, according to Kemp, in the case
where Johansen’scriterion of plastification is strictly adhered to.
Remark: Kinking effect
Some authors do not accept Johansen’sfundamental hypothesis,namely,
that the bars crossing a yield line develop plastic behaviour (yielding)in their
initial direction. It is, in fact, possible to conceive a different criterion,

Lines of direction
Figure 8.6.Bars in direction i

according to which the bars undergo yielding perpendicularly to the yield

line.
The expression for the normal moment at the yield line will then be:
i=r
m, = micosei
i= 1
instead of:

m, = 1 m icos2ei
i= 1
344
The value of the normal moment that emerges from this hypothesis as
to the ‘kinkingeffect’ of the bars is higher than the value of the normal
moment corresponding to Johansen’shypothesis.
It would appear that the behaviour in reality is somewhere between these
two hypotheses. Basing himself on tests carried out at the University of
Wales,Swansea,Kwiecinski has formulated a criterion in which account is
taken ofthe ‘partialkinking’of the bars and whose results are intermediate
between those emerging from Johansen’scriterion and from the criterion
of ‘totalkinking’respectively (cf. refs 24,25,27). Kwiecinski’s tests as well
as the tests performed by Sozen and Lemschow show that the adoption of a
‘totalkinking’criterion would not be on the safe side.O n the other hand,it
would appear that Johansen’s criterion gives normal bending moments that
are smaller than the actual moments in all the cases investigated and is
therefore on the safe side.
This conclusion is confirmed by the first results of the systematic research
which has been undertaken by Massonnet at the University of Liège with a
view to arriving at the direct experimental determination of the criterion for
failure by simple bending in reinforced concrete slabs. The tests already
performed have indeed shown that by applying Johansen’s criterion the
results obtained are on the safe side in relation to the experimentally
determined results.

Yield Line Method

The so-called ‘yield line method’, which was introduced by Ingerslev and
developed by K.W.Johansen in his doctorate thesis in 1943,3embodies the
practical application of the kinematic theorem (lower bound theorem) to
reinforced concrete structures.For a given plane structure this theory results
in the determination of a load P which is larger than,or equal to,the actual
ultimate load. If the kinematically permissible mechanism adopted is the
precise failure mechanism, then this load P will indeed be equal to the
ultimate load.
In order to be certain that this last-mentioned condition is satisfied,it is
necessary (in the absence of direct experimental verification) to be able to
determine, for the plane structure under consideration,a field of statically
permissible moments giving the same limit load P. Only in that case is it
possible to assert- on the basis of the theorem of uniqueness -that the load
P is the exact ultimate load.
Unfortunately,at the present time only a small number of exact or com-
plete solutions (i.e.,
in which a kinematically permissible mechanism occurs
in conjunction with a statically permissible field of moments so that both
conditions give the same limitload)are available.These solutionscorrespond
only to slabs of the simplest shapes. Some known complete solutions are
indicated in Section 8.4.
In the next section (Section 8.3)the yield line method of Johansen will be
outlined. For the purpose of practical application of the method it will,
however, be necessary to consult publications giving a fuller treatment of
 345
the subject,including more particularly Professor K.W.Johansen’sbook3
and the publications of Steinmann,” Wood,I2Massonnet16 and Jones.14
Reference may also be made to ‘Recentdevelopments in yield line theory’
published by the Cement and Concrete Association,May 1965.23

8.3 YIELD LINE THEORY

8.3.1 P R I N C I P L E SO F T H E T H E O R Y

The yield line theory assumes that reinforced concrete can be considered
as a rigid-plasticmaterial.In the ultimate limit state the elastic deformations
of the various elements of the mechanism are therefore neglected in relation
to the plastic deformations. The failure mechanism thus comprises plane
rigid elements whose junctions are formed by so-calledyield lines,or ‘fracture
lines’,functioning as linear hinges, at which the plastic deformations are
concentrated.
The deformations of the structure are due solely to the rotations of these
rigid constituent elements about axes that are compatible with the support
conditions. The deformed surface is of a polyhedral shape (or possibly a
ruled surface in certain zones).
As indicated earlier,it is assumed that the normal bending moment at a
yield line crossed by a reinforcement system comprising bars disposed in the
directions 1, 2 ... i ... i-, to which correspond the resisting moments m,,
m, ...mi... m,,has the value:
i=r
m, = micos2ei
i= 1

The torsional moment at the yield line is not specified (cf.above 2nd
solution: Adoption of the normal criterion).
The application of the yield line theory comprises two successive stages:
The first stage consists in determining the various types of possible
mechanisms,having regard to the shape of the slab,the support conditions
and the loading.Each type of mechanism thus defined should of course be
compatible with the conditions of restraint of the structure.
In this way various ‘families’of possible mechanisms can be defined,each
ofthese familiesbeing dependent upon p geometricalparametersx,,x2 ...x,.
The second stage consists in seeking,for each of the families defined,the
particular mechanism that gives the lowest limit load value.This mechanism,
determined by particular values of the parameters xl, x2 ...p, constitutes
the ‘best’ mechanism for the family of possible mechanisms under
consideration.
This procedure of seeking the ‘best’mechanism is based on considering the
conditions of equilibrium of the various rigid elements of which the
mechanism is composed.
For practical purposes two methods can be applied.The first consists in
expressing the equilibrium conditions in an overall form for the slab as a
whole by applying the equation of work.O n the assumption that the ratios
346
of the various resisting moments of the slab to one of them-called the
‘referenceresisting moment’and designated by m -have been determined in
advance by the designer,the load P is thus obtained in the form of a function
of the p geometrical parameters of the mechanism and the reference resisting
moment m (or,alternatively,the moment m is obtained as function of the p
geometrical parameters and the load P).Next,the values of the p parameters
must be sought which make the function P (xi,x2 ... x,, m ) a minimum or,
what comes to the same thin? uliich make the function m (xl,x2 ... x,, P)
a maximum.
The second method consists in expressing the conditions of equilibrium
of each of the n rigid elements that constitute the mechanism. In applying
this method it is necessary to take into account the ‘nodalforces’which are
statically equivalent to the shear forces and torsional moments acting at the
yield lines.
One each of the ultimate loads that corresponds to the ‘best’mechanism
of each family envisaged has been obtained by applying either of the two
methods indicated above,the ‘best’mechanism for the whole ofthese families
can be deduced therefrom.The corresponding load is adopted as the upper
bound of the limit load (cf.lower bound theorem).
It should be borne in mind that we can be certain that this load is the exact
limitload only ifit is possible to find a field ofstatically permissible moments,
in the slab as a whole, which results in an equal load (cf. theorem of
uniqueness).

8.3.2 F I N D I N GA P R O B A B L E F A I L U R E M E C H A N I S M

The procedure for this is based on the application of the followine theorems,
which result from the fundamental hypothesis concerning the concentration
of the deformations at the yield lines.
Theorem 1: The yield line between the two rigid elements of a slab passes
through the point of intersection of their respective axes of rotation.
Hence it follows that,if the element considered is supported along one
of its edges,the axis of rotation will coincide with the line of support.For a
slab element resting on a point support the axis rotation will pass through
that support.
By applying this theorem it is possible completely to determine the yield
line pattern if the rotations Bi of the various slab elements are known.In the
ultimate limit state the structure is transformed into a ‘mechanism’,and if
the yield line pattern has been established,the angles of rotation Bi of the
various elements will be defined,except for a common factor.
Reciprocally :
Theorem 2:If the axes of rotation of the various rigid elements of the slab and
the ratios of the various angles of rotation ûito any one of these are given, the
yield line pattern is compìetely determined.
 347
Suppose the deformed slab to be cut by a plane parallel to the plane of the
supports and at an arbitrary distance h therefrom.The intersections of the
first-mentionedplane with the various elements of the slab (which form a
polyhedron or possibly a ruled surface if there are an infinite number of
infinitely closely spaced yield lines) are contour lines of the deformed slab.
These contour lines are situated at a distance h/ûifrom the axis of rotation
of the element considered (see Figure 8.7).
Furthermore,these contour lines intersect one another on yield Iines.The
latter are determined by joining the intersections of the axes of rotation to
the intersections of the contour lines.
From theorem 2 it follows that,if the yield lines cut the slab into n parts
and if all the axes ofrotation are known, the yield line pattern will be com-
pletely determined by the determination of n -1 geometrical parameters
(see Figure 8.8).
In the general case the axes ofrotation of the n slab elements forming the
mechanism are not always known in advance.Let r be the degree of indeter-
minacy in knowing these axes.Then,in order to define completely the yield

\contour iines

Figure 8.7

line pattern, there remains (n-1) +r geometrical parameters to be determined.

Hence,in the general case,the yield line pattern cannot be completely
determined merely from considerations of geometrical deformation.
Determining the yield line pattern must,as indicated in Section 8.2.1,
be
done either by applying the method of work (work of deformation) or the
348
p = n - 1 +r = 4-1

.€-

------
- Column
Unsupported edge
Axis of rotation
-MIC~~-.C.LH Yield line
Simple support D Number of parameters
Restrained support of the section

Figure 8.8.Example of determination of probable mechanisms

 349
method of nodal forces (equilibrium of the rigid elements of the slab). These
two methods are explained in Sections 8.3.3and 8.3.4.

8.3.3 APPLICATION OF THE METHOD OF W O R K

Suppose that a family ofpossible mechanisms,depending upon p geometrical

parameters xl,x2, ... xp,has been determined on the basis of the conditions
ofdeformation.Considera conveniently chosen point ofthe slab and suppose
that this point is given a displacement 6 compatible with the restraints to
which the slab is subject.
The equation of work is obtained by equating the work done by the
external forces to the work done by the internal forces in bringing about
the displacement under consideration. O n the assumption that the mech-
anism comprises only the straight yield lines,*this equation can be written
as follows:
CPihi+jjpj.hj.dx .dy = lm, .a.ds
where :
di = displacement of a concentrated force Pi
hj = displacement of a distributed force pj per unit area
a = rotation at a yield line
m, = normal bending moment at a yield line per unit length
The displacements diand dj as well as the rotations o! are all expressed in
terms of 6,which occurs as a factor on both sides of the equation and is
therefore cancelled out.
The normal moments m, are known in terms of the resisting moments of
the different directions of the reinforcing bars in the layer corresponding to
the yield line considered.Suppose that the ratios of the various positive and
negative resisting moments of the slab reinforcement to one of them -called
the ‘referencemoment’(m)-is determined in advance by the designer.That
being so,the equation of work will give a relation between the total load P,
the ‘referencemoment’m,and the p geometrical parameters relating to the
family under consideration.
P is therefore given by a function P (xl,x2 ... xp,m) or, alternatively,
m is given by a function m ( x ~x2, ... xp,P).
The next step consists in determining the values of the p geometrical
parameters which make the function P a minimum or,what amounts to the
same thing,make the function m a maximum. This determination can be
done either by successive numerical approximations or by solving one of
the following two sets of equations:
-ap= o -dm= o
8x1 ax1

*The case of fan-type mechanisms cannot be dealt with in this short treatment of the subject.
For this it will be necessary to consult the literature (more particularly refs. 3 and 12).
350

8.3.4 A P P L I C A T I O N O F T H E M E T H O D O F E Q U I L I B R I U MO F
RIGID ELEMENTS ( M E T H O D OF N O D A L FORCES)

Principie Of The Method. Need For Taking The Shear Forces

And Torsional Moments Into Account

This second method consists in expressing the conditions of equilibrium of

each of the II rigid elements constituting the mechanism.In the general case
3n equilibrium equations are thus obtained.
In applying the equation of work it was not necessary to know the shear
forces and torsional moments acting at the yield lines,inasmuch as their
total work is zero.O n the other hand,in order to write down the equilibrium
conditionsfor the various elements of the mechanism,it is necessary to know
D

Figure 8.9

the forces that are statically equivalent to the torsional moments and shear
forces acting at the yield lines.
Consider a rigid element i of a mechanism,bounded by yield lines AB,
AC,BD (see Figure 8.9).
Let k, and k,, be the forces which are statically equivalent to the shear
forces and to the torsional moments acting upon the element í at the yield
line AB,these statically equivalent forces being applied at the ends A and B
respectively.
Similarly,let kAc and k, be the forces statically equivalent to the shear
forces and torsional moments at the yield line AC, and let k, and k, be
such statically equivalent forces with regard to the yield line BD. In the
following treatment of the subject these forces acting at the ends of the yield
lines will be called ‘terminalforces’.
 351
At the node Cjunction ofyield lines)A a concentrated force K A acts upon
the element I,this force being the algebraic sum of the forces which are
statically equivalent to the shear forces and torsional moments acting at the
two yield lines converging at A. This concentrated force will be called the
‘nodalforce’.
Johansen determined the nodal forces by considering infinitely small
rotations of the yield lines about one of their points. The problem has
latterly again received attention more particularly from Kemp, Morley,
Nielsen, Wood and Jones,23who have tried to show that the method of
nodal forces does not differ much, except in its form,from the method of
work.Actually,the procedure employed in the method ofnodal forcesconsists
in endeavouring to establish the positions of the yield lines that correspond
to fixed values of the reference moment m. The work of these recent inves-
tigators has furthermore enabled the conditions for determining the nodal
forces to be established with greater precision.
The following treatment of the subject is based on these recent investiga-
tions,including more particularly Kemp’spaper in ‘Recentdevelopments in
the yield line theory’published by the Cement and Concrete Ass~ciation.~~

Determination Of The Nodal Forces

Sign Conventions A n d Graphical Representation

Forces: Forces are considered as positive if they act downwards.O n plan,

a downward vertical force is represented by a cross.An upward vertical
force is represented by a dot.

Positive yield Positive normal

moment
line

Figure 8.10
352
Moments: A bending moment is considered as positive if it produces
tensile stresses on the underside of the slab.
A torsional moment acting upon a rigid element of a slab along one of the
yield lines by which that element is bounded is considered as positive if it
acts in the clockwise direction with respect to the element.
Positive bending moments are represented by vectors pointing in the
clockwise direction around the rigid element considered.Positive torsional
moments are represented by vectors pointing towards the interior of that
element (see Figure 8.10).

Preliminary Investigation: Variation Of The Moments D u e To The

Rotation Of A Yield Line About O n e Of Its Points

Let AB be a yield line whose position is assumed to correspond to the fixed

value of the reference moment m (see Figure 8.11).

8’
\
\
\

8
Figure 8.11

At the line AB the value of the normal moment m, is therefore given by the
failure criterion:
i=r
m, = 1 m icos’ Bi
i= 1
N o w let the line AB undergo a rotation through an infinitely small angle
d$ about one of its points (pointO) and let A’B’ be the new position of the
line after undergoing such rotation.
The direction ofAB is defined by its angle,)I which is reckoned as positive
in the trigonometrical direction of angular measurement with respect to a
 353
fixed direction in the plane of the slab-for instance,one of the supported
edges of the slab.The origin of the co-ordinatesis located at O,and the axes
n,s are respectively normal to,and tangential to,the line AB and are situated
in the plane ofthe slab.Let a be the angle (measuredas positive in the trigono-
metrical direction) between the fixed direction X and the axis n.
The normal moment at the line A'B' no longer satisfiesthe failurecriterion.
Consider an arbitrary point S on AB,situated at a distance s from the point O.
After the rotation has taken place,this point S will have moved to S',and the
variation dm, of m, between S and S'is given by the expression:

Now
da = d4,dn = -s.d4.cosd+andds = -s.d$.sind4
Hence,when d 4 O:-f

dm, am, am,

__ - ~- S-
d4 aa an
Furthermore,if m; and mirare the principal moments whose directions
form angles y and y +$n, with the direction X:

m;+m;, I m;-
m,=- I cos 2(a -y)
2 2
mi -m;,
mns= ~ sin (2a- y)
2
Hence :

W e thus have the relation:

where :

Or :

But:
354
and therefore:
i=r
dm, = - 2 = -2m,
mi.sinûi.cosOi
d4 i= 1

where m, denotes the value of the torsional moment of Johansen's failure

criterion.
Finally :
am" = 2(m,-m,,)
s-
an
In the particular case of orthotropic reinforcement this relation remains
unchanged.For isotropic reinforcementwe have m, = O and therefore:

Calculation Of The Terminal Forces

This section is concerned with the calculation of the 'terminal' forces k,

and k, which are applied at A and B and are statically equivalent to the
system of forces transmitted through the yield line (i.e.,
on the one hand,the
force per unit length statically equivalent to the shear forces and torsional
moments,given by the general equation of Thomson and Tait:

and,on the other hand,the forces nSA mnsAand mnsBacting at the two ends
A and B and due to the torsional moments at these points).
For the purpose of this calculation,the moments of the forces about the
point O will be considered. Let sA and s, denote the distances 0.4 and OE.
Then:

The following relation has already been established

am, = 2(m,-mns)
s-
an
Hence:

k,.s,-kA.sA
[:s
= mnsB.sB-mnsA.sA+ 2m,-2 m,,+s-
( aA.)]ds
 355
and finally:
kB .SB -k, . SA = -(m,,,-2m,)S~+(IT,,,A-2m,)S~(8.4)
This relation remains unchanged for the particular case of orthotropic
reinforcement,and for isotropic reinforcement it becomes:
k, .S B - k A . SA = -mnsB.
Sg+mns~
SA.

From the form ofthe above expression it is apparent that various cases
have to be considered.

First case: The yield line is not compelled to pass through afixed point (two
degrees of freedom).
In this case it is possible to make the yield line successively undergo a
rotation about each of its two ends A and B.
O n applying equation 8.4by first supposing the point O to coincide with
B (se = O) and then by supposing it to coincide with A(sA = O), we obtain:
k, = -mnsA+2mt
kB = -mnsB+2mt
In this case the terminal forces can therefore be determined.
Second case: The yield line is compelled to pass through afixed point O not
coinciding with either of its ends (one degree of freedom)
In this case equation 8.4must be applied in its general form:
k, .sB -kA .SA -2m,)sB+(mnsA
= -(mnsB -2mt)sA
The two terminal forces kA and k, cannot be determined individually.
However,the sum of their moments with respect to an axis passing through
the fixed point O is known,and it is precisely this sum that will in this case
normally have to be considered in the equilibrium method.
A special case of this second case occurs when the fixed point O is situated
at infinity on the extension of AB.In other words,this is a yield line which is
compelled to remain parallel to a fixed direction.
Putting sB -S, = 1, we can in this case write equation 8.4as follows:

and for ,s +m this gives:

kB-kA = -mnsB+mnsA (8.6)
Here again the two terminal forces cannot be determined directly.Their
algebraic sum is known,however, and it is thus possible to calculate the
sum ofthe moments ofthese forcesabout an axis parallel to the fixed direction
of the yield line.Once again it is this sum that will in this case normally
have to be considered in the equilibrium method.
356
Third case: O n e ofthe two ends of the yield line isfixed (one degree offeedom)
Suppose that one ofthe two ends-the end A,for example - is fixed a priori
(e.g.,for reasons of symmetry or perhaps because a large concentrated load
or a discontinuity of the edge of the slab occurs at A). The end A of the yield
line can then be said to be ‘anchored’or ‘stationary’.
Putting S, = O in equation 8.4,we obtain:
k, = - m n s B + 2 m ,
The terminal force at the ‘movable’end B is therefore determined.
O n the other hand,the terminal force at the fixed end A is not known.
Attempting to calculate it by writing down the expression for the equilibrium
of the vertical forces acting upon the yield line,we obtain:
kB -k A = +mnsB -mnsA
where FA, denotes the total vertical force acting upon the length AB between
the points A and B.
O n substituting the value found above for k, into this expression,we find:
kA = - F A , + 2m,-2mnsB+mnsA
Hence it appears that the terminal force k, could be determined only if
FA, were known.
Fourth case: The two ends of the yield line are fixed
In this case it is not possible directly to calculate the terminal forces at
either of the ends.
To sum up,the general expression for the terminal force acting at one end
ofa yield line is : -m,, + 2m,.This expression can be used only in the following
cases :
(a)at both ends of a yield line of which no point is determined a priori;
(b) at both ends of a yield line which has to pass through a fixed point,
provided that this fixed point is not one of the ends of the yield line
and that the terminal forces calculated in this way are used only for
calculating the sum of their moments about an axis passing through
the fixed point;
(c) at the ‘movable’end of a yield line whose other end is fixed a priori.
The terminal force at a fixed end ofa yield linecannot be directly calculated
by means of the expression -mns+2m,.

Calculation OfNodal Forces

If a rigid element (12) (see Figure 8.12) forming part of a mechanism is

bounded by two yield lines 1 and 2 numbered in the trigonometricaldirection
of angular measurement,the nodal force K,,acting upon the element (12)
at the intersection A of the two yield linesis the algebraic sum of the terminal
forces corresponding,at A,to each of these two lines:
 357
The nodal forces are reckoned as positive if they act downwards.
Provided that A is not a fixed point,we have:
Ki2 = -mns2+2mt2+mnsi-2mtt
and hence:
Ki2 = mnsi -mnsZ-2(mrt -mt2) (8.7)
In this expression mnsland mns2represent the actual terminal torsional
moments at each of the two yield lines 1 and 2,while m,,and mt2represent

Figure 8.12 k2

the values of the torsional moments of Johnansen’sfailure criterion at each

of the two lines 1 and 2.
In order to arrive at an explicit expression for the nodal forces,the inter-
section of three yield lines numbered 1, 2 and 3 (in the trigonometrical
2

Figure 8.13 3

direction of angular measurement) will now be considered (see Figure 8.13).

The expressions for nodal forces acting at A upon each of the three rigid
elements 12, 23,31 are respectively:
Ki2 = mnsi-mns2-2(mti -mi21
K2 3= mnsl -mns2 -2(mr2 -mt3) (8.8)
K,i = mns3 -mnsi -2(mt,-mti)
358
mns2and mns3
mnsl, can be replaced by their values calculated in Section 8.2.2
namely :

(8.9)

Note 1:
Equations 8.9are not identicalwith the expressionsfor the nodal forces given
by Johansen’ or Jones.21The reason for this is that those authors adopt
Johansen’s ‘stepped’criterion of yielding (as defined in Section 8.2.2),
according to which a constant bending moment m, and a constant torsional
moment m,act at the yield line,namely:
i-* i=r
m, = 1 micos2ûiand m, =
i= 1 i=l
misin Bicos Bi
N o w equations 8.9are based on the assumption of the criterion for the
normal moment:
i=r
mn = 1 micos2ûi
i=1

while the torsional moment is not specified.Using these expressionstherefore

presupposes that,on writing down the equilibrium equations for the rigid
elements ofthe mechanism,the torsional moment m,embodied in Johansen’s
failure criterion is not taken into account.
For practical purposes it is more convenient to write the equilibrium
equations with due regard not only to the normal moment mn but also to
the torsional moment m,of Johansen’sfailure criterion.
Consider a yield line forming an angle ß with the axis of rotation of the
corresponding rigid element (see Figure 8.14)
Suppose that the yield line AB corresponds to an orthotropic reinforce-
ment system with principal moments m and ,um.Let CI be the angle that the
moment m forms with the axis of rotation of the element.
First of all we shall use the criterion of the normal moment.
The normal moment at the yield line,per unit length,is:
mn = m cos2(ß-CI)+ ,umsin2(ß-CI)
The component of this moment with respect to the axis of rotation is:
[m cos’ (b-CI)+ ,um sin2(ß-CI)]sin ß
 359
If, on the other hand,we adopt Johansen’s‘stepped’criterion,we merely
have to project the principal moments rn and prn on to the axis of rotation,
so that we obtain per unit length a moment of the following magnitude:
rn cos a +prn sin a,i.e.,a simpler expression.
If,for the sake of convenience,this latter procedure is adopted,it will be
necessary to modify the expressions for the nodal forces by deleting from the

Figure 8.14

second member the torsional moments directly taken into account.Values

Ki2,KL3 and Ki, are obtained,such as:

(8.10)

hence :
2 = (%l- cot413 -( m n 2 -m n 3 ) cot4 2 3 -(rntl -rnrZ)
K23 = (mn2-rnni) cot 4 2 I -(mn3-mní) cot 4 3 1 -(rnrz-mt3) (8.11)
Ki, = (mn3-mn2)cot 43z-(mn,-rnn,)cot4,z-(mr,-m,,)
It can be shown that these expressions are identical with the cyclic expres-
sions given by Jones,I4though in a somewhat different form.
Actually, Jones’s expression for the nodal force Q 1 2(cf.ref. 14, page 147,
equation 9.12):
Q I2 = (mn3- m n 1 ) 3 cot $1 cot 4 2 3 +(rnt2)3 -(fflf 1)3
3 -(mn3 -rnn2)3

where (m,,),, (mn2)3 and (rnn3)3 are the normal moments at the line 3 due to
the reinforcement system corresponding respectively to the lines 1, 2,3;and
(m,,),and (rnt2)3 are the torsional moments at the line 3 due to the reinforce-
ment system corresponding respectively to the lines 1 and 2.
The following relations can readily be established :
mm1 cot 413-mtl = (%1)3 cot4 1 3 + ( r n t 1 ) 3
-rnn2 cot423+mt2 = -(rnn2)3 cot423-(‘f2)3
360
by means of which the expression for K;, in equation 8.11 can be written
in the following form:
K;2 = (%1-%3)3 cot613-(mn2-mn3)3 cot 423+(mt1)3-(mt2)3
Furthermore:K12= -Q12,having regard to the different sign conven-
tion (Jonesreckons Q12as being positive in the upward direction). The two
expressions are therefore identical.
The expressions for the nodal forces, with Jones’s notation for the
moments, but using the sign convention adopted in the present treatment
of the subject,can now be written as follows:
K I 2 = ( m n 1 - m n 3 ) 3 cot613-(mn2-mn3)3 cot623+(mt1)3-(me2)3
K;3 = (mn2-mnl)i cot 621-(mn3-mn1)1cot 6 3 1 +(mt2)1-(mt3)1(8.12)
= (mn3-mn2)2 cot632-(mn1-mn2)2 cot612+(mt3)2-(mti)2
Note 2:
For the nodal forces to make equilibrium at the node,not more than three
yield lines associated with different reinforcement systems can converge
there. However, in the special case where all the converging yield lines
e Unsupported edge e

(2e)

Figure 8.15

correspond to one and the same reinforcement system, the principal

moments at each yield line will coincide in magnitude and direction at the
point of intersection.In that case any number of yield lines may converge
at the node.
That being so,the actual terminal torsional moments mnswill be equal to
the torsional moments of Johansen’sfailure criterion and therefore all the
nodal forces will be zero.Evidently,this conclusion is valid only in the case
where the node under consideration is not a fixed node.
Note 3:
The special case of the unsupported edge will now be considered.
If a yield line encounters an unsupported edge,the terminal force due to
If it
this edge is equal to the actual torsional moment at the edge,i.e.,mnse.
were a supported edge,it would not be possible to determine the terminal
force,as the bearing reaction would not be known.
 361
Consider two yield lines intersecting at a point A (nota fixed point) on
an unsupported edge (seeFigure 8.15).
The nodal force which acts at A upon the element (el)is:
K,, = k,-k, = -mnsl-mnse+2m,,
O n applying the expressions for the terminal torsional moments estab-
lished in Section 8.2.2 and bearing in mind that the three lines are here
numbered e, 1, 2 (inthe trigonometrical direction of angular measurement)
instead of 1, 2, 3, we obtain:
-mnsi-mnse = -(mne-mni)COt 4ei
And therefore,since mne= O:
Similarly :
(8.13)

or,if the torsional moments m,, and m,, of Johansen’scriterion are taken
into account in the equilibrium equations (cf.Note 1):
Keí = mní cot 4ei +mti
Ké2 = -m,, cot qîze -m,,
1 (8.14)

The forces K,, and K,,(or Ké, and Ké,)are called ‘edgenodal forces’.
The nodal force acting at A upon the slab element bounded by the two
yield lines 1 and 2 can be calculated by applying the general equation 8.9:
K12 = mn2 cot 4e2-mni cot 4 e i -2(mt,-mtz)
or,on applying the general equation 8.11 (takeninto account in the equili-
brium equations for the torsional moment of Johansen’scriterion):
K;2 = mn2cot -mnl cot 4ei -(mti- m d
In the specialcase ofisotropic reinforcement systems the above expressions
become :
K,, = Ké, = m,, cot 4el
K e 2= Ké, = - m n 2cot 4e2
KI,= Ki2 = mn2cot 4e2 - m n 1cot

Establishing And Solving The Set Of Equilibrium Equations

Once the nodal forces which can be directly calculated by means of the
expressions established above have been determined,it is possible to write
down the equilibrium equations for the n rigid elements forming the
mechanism under consideration.
In the general case there are 3n of these equations.For any particular
rigid element one equilibrium equation for the vertical forces and two
equilibrium equations for the moments can be written down.The number
ofequations is, however,reduced in the case of symmetry or in the case of
elements adjacent to the lines of support.
362
The unknowns occurring in the equilibrium equations are the following:
(a) The nodal forces which cannot be directly calculated by the method
explained in the previous section,i.e.,nodal forces at theJixednodes.
In the special case where a concentrated load P acts at a node A (intersec-
tion of yield lines)this load is equilibrated by forces P,,PI,,etc.,of opposite
algebraic sign which act respectively upon the rigid elements I, II, etc.
separated by the yield lines converging at A (see Figure 8.16).
By means of the equilibrium equations of the vertical forces for the
elements I,II,etc. it is possible to determine the respective values of the

1
3

2 Figure 8.16

forces PI,PI,and PI,,,which can then be introduced into the equilibrium

equations of the moments for the corresponding elements.
(b) T h e p geometrical parameters xl,x2 ...x p which determine the mechanism.
(c) The reference moment m.
(InSection 8.3.3it was assumed that the ratios of the various positive and
negative resisting moments of the slab reinforcement to one of them-the
reference moment m -had been determined in advance by the designer.)
The applied loads are here considered to be given quantities.
By solving the set of equations comprising the equilibrium equations of
the various rigid elements of the mechanism it is possible to determine the
values of the unknown nodal forces and also the values of the p geometrical
parameters defining the ‘best’mechanism of the family considered.Finally a
relation between the reference moment m and the applied loads is obtained
which corresponds to the ‘best’mechanism of the family.
If a number of families of possible mechanisms are investigated in this
way, the values of m obtained for the ‘best’mechanism of each of these
various families can be compared with one another. It is thus possible to
find the ‘very best’ mechanism,i.e.,the mechanism to which-for a given
set of loads-corresponds the largest reference moment m. O n the assump-
tion that the ratios of the various resisting moments of the reinforcement
of the slab as a function of this reference moment have been determined in
 363
advance,these various resisting moments can be calculated.The requisite
cross-sectionalareas of reinforcement can then be found.
It should be noted that the solution obtained by means of the equilibrium
method (i.e.,based on the conditions of equilibrium of the rigid elements of
the slab) is an upper bound solution and that only if a statically permis-
sible bending moment field which at no point conflicts with the plasticity
criterion,and which results in the same limit load,is established will it be
possible to affirm -thanks to the theorem of uniqueness -that the ‘exact’
solution has been obtained.
Note: The set of equilibrium equations can be directly solved by the
usual methods. Alternatively,a procedure of successive approximations
may be adopted, which consists in predetermining all the parameters xl,
x 2 ...xp.In this way,for each element of the mechanism,a relation between
the reference moment m and the applied loads is found.Comparison ofthe
values m,,m 2 ... m, thus obtained for the various elements of the mechanism
will provide an indication of the degree of approximation achieved and will
also indicate the direction in which the yield lines should be shifted-i.e.,
the direction in which the parameters xl,x2 ... xp should be varied in order
to obtain a better approximation. The calculation can then be repeated
with a fresh set of values xl,x2 ... xp,and so on,until the approximation
obtained is considered to be good enough.

8.3.5 OTHER PROBLEMS

The object of the foregoing treatment of the subject was to summarise the
essential aspects of the principles and the procedure of applying the yield
line theory to the analysis of slabs in the limit state of flexural failure.
With regard to all the points which it has not been possible to deal with
in the limited scope of this exposition it will be necessary to refer to the
bibliography,and especially to the references stated in the present chapter.
Such points are more particularly:
The method of afJine transformation whereby,subject to the fulfilment of
certain conditions,it is possible to avoid having to carry out a direct analysis
for an orthotropic slab and,instead,to replace the latter by an equivalent
isotropic slab.
Cf.in particular:ref. 4 (pp.67 to 74) and Ref.12 (pp.117 to 126).
The extension of the method of superposition of the elastic theory to limit
analysis in a case where several load systems are acting simultaneously.
Cf.in particular: ref. 4 (pp.74 to 81), ref. 12 (pp.41 to 45)and ref. 16
(pp.261 to 264).
The special mechanisms in the vicinity of corners and of concentrated loads,
and the general case of curved yield lines (fanwise mechanisms).
Cf.in particular:ref. 4 (pp.82 to 136),ref. 12 (pp.28 to 36 and 52 to 57).
The interaction of a slab and its edge beams in a case where the slab is asso-
ciated with such beams.
Cf.in particular: ref. 4 (pp.136 to 142), ref. 12,ref. 16 (pp.264 to 269)
and ref. 28.
364
Membrane effects and arch action.
Cf.in particular:ref. 12 (pp.225 to 261),ref. 16 (pp.302 to 308).

8.4 PRACTICAL DESIGN FORMULAE FOR SIMPLE SLABS

The design formulae for simple slabs of various shapes,as set forth in the
tables on pp.365-388,have forthe most part beenestablished by application
of the yield line theory.
For each formula indicated,the tables give the relevant bibliographical
reference so as to enable the user to refer to the original publication in order
to obtain the necessary information on the range of validity of the formula.
It should be noted that the yield line patterns given in these tables have
been simplified in that the yield lines are assumed to emerge from the corners
of the supported side$.In actual fact the yield lines diverge in the vicinity of
the corners,this being associated with some reduction of the load capacity
of the slab.For a given value of the angle formed by the corner under con-
sideration this reduction is greatest if no negative reinforcement is provided
at the corner. The reduction is less according as the negative resisting
moment at the corner is larger. It has not been possible to deal with this
matter in the limited scope of the present simplified treatment of the subject,
but the reader can obtain the relevant information from the publications of
Johansen3 and Wood.I2As an indication the percentage increases in the
ultimate moment that may be taken into account to allow for the corner
effect,for given values of the ratio m'/m,are indicated in the following tables.
Furthermore,in the last column of each table the existence of a lower
bound solution (if any) corresponding to the upper bound solution is
indicated (by stating the relevant biobliographical reference). It must be
pointed out that in all cases where no such indication is given,the solution
presented is an upper bound solution and should be used with due caution.

8.5 PRACTICAL DESIGN FORMULAE FOR FLAT-SLAB

F L O O R S AND M U S H R O O M F L O O R S
8.5.1 PURPOSE

The formulaegiven in the tables on pp.394-402 are intended for the analysis of
flat-slabfloors and mushroom floors in the limit state of flexural failure for
various loading conditions.This limitstate has been defined in Section 8.1.2.
It is assumed that the slabs are analysed with regard to punching shear in
accordance with Section 6.2.5of Part I of this manual.

8.5.2 B A S I S OF THE M E T H O D

The structural analysis is based on the yield line theory,whose hypotheses

and principle have been explained in Section 2.
 -
II
€
k a
all all
-s
"&I-
€
E
i. E. B.
d ' a 6
I1 II Il
.1 .L .~
E E E
f
II
.~
II
m m
h
Y I<
I
P
P
P
c c
@ m
 c
3
F E B aE
e s e
!!l.
! II
." .
II..
E € € €
P
4 P
 e
" +
:
:
II
E
O 0
Il II
E Y
II II
P
-
0 -
d oe
$2 P
Y
j
e
-
O 0
II II
T Y
II II
c i'
B
4
P
n
"
P
P
 w1
9
al
II
E
c
b
o
t-
 v
Il
E f E ."
Srt& Il
Il II II . -
E E E Il
D
9
N
 c
s
P
E C
D
P
 E o
o
II B Il
.+ I E
E
'E
P
P P
1
A
4
 w
+
v
-I
4N
?
-,w
II Il
i i
-I
- P
 c
3
E !
c å
II II
. . ."
E €
-3
L
E _ L
-3
D
 E.!! €
S S % Z k
II II Il II Il
E aE i' .-
E
E ?
o
u
4
 h
a
v
*- m
LI“
E
E
€4
II
h
E E
 o
II
s
l l
 '9
+
a
+
g
I
-
.s
?
1
Il
0
a
E O o
II II II
.~
'c E
O O
ll Il
E h
1
i
.a E
.e
.i II
.€
P
€g
I
I
 389
8.5.3 G E N E R A L
CONSIDERATIONS-RANGE OF
APPLICATION

Slab
Dimensions

The present formulae relate to floors consisting of continuous slabs,without

ribs,supported directly by the columns,except possibly at the slab edges,
where there may be load-bearingwalls or edge beams protruding above or
below the slabs.These slabs may sometimesbe cantilevered beyond the edge
columns.
In the middle plane of the slab the points corresponding to the vertical
axes of the columns are the intersections of two sets of mutually perpen-
dicular straight lines XI,X, .... YI,Y, ... yi which divide this plane into

Figure 8.17

.rectangular‘panels’with dimensions 1, and l, which dimensions are not

necessarily the same for all panels (see Figure 8.17).
If the dimensions I, and i, satisfy the under-mentionedconditions (a)and
(b), the failure mechanisms to be considered are those indicated in the tables
of formulae.If these conditions are not satisfied,it may be necessary to per-
form the analysis on the assumption that other mechanisms could develop,
such as the lifting of an unloaded panel due to the action of a superimposed
load acting upon the two adjacent panels.
Condition (a): The ratio of the longer to the shorter side of a panel must be
equal to 3.
390
Condition (b): In the direction X (or Y)the ratio of the larger to the smaller
of two spans must not exceed 3.

Reinforcement

The reinforcement is assumed to consist of a grillage of mutually perpen-

dicular bars disposed parallel to the directions X and Y.

Column-To-Slab Connection Zone

Definitions

The columns may or may not be provided with flared heads of truncated
conical or pyramidal shape.If such columns heads are provided, the term
‘mushroom floor’is employed;otherwise ‘flat-slabfloor’.*If the floor slab
does not cantilever beyond the edge columns,the flared heads which may be
provided on these columns and on the corner columns are considered to be
present only on the inner side or sides of such columns,i.e.,away from the
slab edges.
Furthermore,in some cases the slab may be thickened locally over the
columns.This thickening (on the underside) may be of parallelepipedal or
truncated conical or pyramidal shape and is called a ‘drop’.

M u s h r o o m Floors

Ifthe column heads and drops have dimensionsconforming to the conditions

stated below, it is possible to predetermine the point through which the
negative yield lines will pass (cf.Note I). In other cases it will be necessary
to determine that point by calculation.
Dimensions of the column heads
(a) Truncated pyramidal column head
In the direction X (or Y)the side length of the top face of the column head
a
should be between i and of the smaller of the two adjacent spans.
(b) Truncated conical column head
The sides of a square having an area equal to that of the circular top face of
the column head should satisfy the condition a above.
Dimensions of the drops
(a) Truncated pyramidal drop
The height (depth)h, of.the drop should be less than the normal thickness
of the slab (seeFigure 8.18).
*The term ‘flatslab’is frequently used to describe both types of floor envisaged here as belonging
to one general form of construction.However, in the present treatment it is convenient to retain
the distinction,
 39 I

t
Figure 8.19
392
The dimensions a, and b, of the top face of the drop should conform to
the following relations:

h
b,-b1>O.l8l,-'
h0
where 1, and 1, denote the smaller of the two adjacent spans in the respective
directions.
The other symbols are defined in Figure 8.18.
(b) Parallelepipedal drop
The height (depth)h, of the drop should be less than 0.6 times the normal
thickness of the slab.
The sides (a, and b,) of the drop should conform to the following relations:
h
a,-a, 20.361 2
y h0

h,
b, -b,> 0.361,-
h0

where 1, and 1, denote the smaller of the two adjacent spans in the respective
directions.
The other symbols are defined in Figure 8.19.
(c) Circular drop (cylinder,truncated cone)
An equivalent rectangular drop should be considered whose top and bottom
faces should have areas equal to the corresponding faces of the circular
drop.The dimensions of the equivalent drop should satisfy the conditions
indicated in points a and b above.
Determination of the moments at the negative yield lines crossing a drop
In a yield line which crosses a drop the compression zone of the concrete
may be situated wholly or partly within the drop, provided that the con-
nection between that drop and the slab is properly ensured.In that case it
can be assumed that the reinforcing bars correspondingto the compression
zone thus defined are situated within a strip whose centre-linecoincides with
the column and which has a width equal to a,+2ho (with the notation
indicated in Figures 8.18 and 8.19).

Loads

The method of analysis under consideration is valid for the case of dead
load and uniformly distributed superimposed load. However,if the super-
imposed load as a whole consists ofa large number ofconcentrated loads,the
method can still be applied,provided that the value of the largest individual
 393
concentrated load does not exceed 0.2times the value of the total load
on a panel.

8.5.4 TABLES FOR DESIGNA N D A N A L Y S I S

The tables on pp. 394-402 give the moment/ultimate load relations corre-
sponding to the failure mechanisms under consideration. Some additional
information which cannot be included in these tables is appended as notes.

Note 1 :Positions Of Negative Yield Lines Of Mechanism II

(a) Flat-slabfloor:
The negative yield lines are assumed to be straight lines that pass along the
faces of the columns if the latter are of rectangular section.If the columns
are circular in section,the negative yield lines are assumed to pass along
the faces offictitioussquare columns which are concentric with,and have the
same cross-sectionalareas as,the circular columns.
(b) Mushroomfloor:
In the case oftruncated pyramidal column heads the positions ofthe negative
yield lines,which are straight lines,should be determined as follows:
If the angle of the lateral faces of the column head with the vertical axis

yield lines

ai C1SL50 b) a >LSo
Figure 8.20

of the column is less than 45",the section in which the yield line develops is
situated over the top edge of the column head (Figure8.20a).
If the angle of the lateral faces of the column head with the vertical axis
of the column exceeds 45",the section in which the yield line develops is
determined by the intersection of the top face of the column head with a
straight line which passes through the bottom edge of the column head and
forms an angle of 45"with the vertical axis of the column (Figure8.20b).
-!+
o I-
h
%If
E-
 z
E"
-
E I
f I
d
-
.
II II
* ' - i
i
E
 :
P
o
II
Ê
4
 403
In the case of a truncated conical column head, a fictitious truncated
pyramidal column head should be considered whose top and bottom square
bases should have areas equal to the areas of the corresponding circular
bases of the actual column head. The negative yield line positions should
then be determined in the manner indicated for a truncated pyramidal
column head.

Note 2 :Slab-To-Column Connection

If the floor is loaded asymmetrically in relation to a row of columns,the slab
will transmit some bending moment to each of the columns.
It is assumed that to each column is transmittedthe moment from a section
of slab whose centre-linecoincides with the centre of the column and whose
width a,,is defined as follows (seeFigure 8.21):
a, = a+2h
where a is the-sidelength of the column cross-sectionor of the top face
of the column head;
h is the effective depth of the slab.

Note 3:Failure Moment Of An Edge Column (OrCorner Column)

In Composite Bending

Edge columns:
The load transmitted from the floor slab to a row of edge columns should be
taken as equal to the load acting upon that part of the slab which is bounded
by the yield line X‘X(see Figure 8.22):
Of the load acting upon this slab strip bounded by X’Xeach column
should be assumed to receive a proportion corresponding to the area
bounded by the centre-lines of the two panels adjacent to that column
(areashown hatched in Figure 8.22).
Corner columns:
A corner column is assumed to carry the load acting upon the corresponding
quarter panel.

Note 4:Local Failure Mechanism Around A n Edge Column Or

Corner Column

The local failure mechanism liable to develop around an edge column or a

corner column can be assumed to have the shapes shown in Figure 8.23:
The angle CI which serves to define the shape of the mechanism of pattern
(a)is calculated from the following expression:
tan’a =
404

.
-
. a,,= a +2h

Figure 8.21

O I O o
i

Figure 8.22

Figure 8.23
 405
The positions of the positive yield lines are defined by the distance from
the node A to the edge(s)of the floor in the case of pattern (a),and by the
distance from the positive yield line to the corner of the floor in the case of
pattern (b). These distances play no part in establishing the moment-load
relations corresponding to those mechanisms. However, as the yield lines
may pass through differently reinforced slab zones,it is necessary,in applying

Figure 8.24

Centre line of columns or

edge of f loor
-- -

Centre Line of
columns --E& y-
Figure 8.25

these relations, to take account of the positions adopted for the positive
yield lines.
The moment-load relations are as follows:
Edge column:
D
4s<l:
rD
@,> 1: = m,(4+ 1.144,)
1 +4,/s/l
406
Corner column :

4,< 1 (patternu): i +4PJsII = m,(1.57+0.436,)

P
1 (patternb): = 2m,
1 +4JsIl

Note 5: Length Of Reinforcing Bars

(a) Positive reinfoficement :

The lengths of the bottom reinforcing bars may be determined as follows:
Part of the reinforcement should have a length at least equal to I, (or.),I
The rest of the bars may have a length less than i, (or;),i the length of
these bars should then be so determined as to avoid giving rise to the failure
mechanism represented in Figure 8.24.
This mechanism is associated with the followingmoment-loadrelationship
(withthe notation as indicated in the diagram):
PIr4(L-4) = 2m[I~,(1+61)+~1(62-41)1
(b) Negative reinforcement:
The negative reinforcing bars should be of sufficient length to obviate the
formation of the failure mechanism represented in Figure 8.25.
In this mechanism, mi and ml, are the ultimate moments of the slab
sectionsin which all or some ofthe negative reinforcement(supportreinforce-
ment) ceases to be effective.
Thismechanism is associated with the following moment-loadrelationship:
PG
= 2[J(1+61)+J(1+62)]2
 407
REFERENCES
i. BACH, c. and GRAF, o., ‘Versuche mit allseitig aufliegenden quadratischen und recht-
eeckigen Eisenbetonplatton’ (‘Tests with square and rectangular reinforced concrete
slabs supported on all sides’), Deutscher Ausschuss fur Eisenbeton, No.30, Berlin (1915)
2. BACH, c. and GRAF, o.,‘Versuche mit zweiseitig aufliegenden Eisenbetonplatten mit kon-
zentrierter Belastung’ (‘Tests with reinforced concrete slabs supported on two sides and
subjected to concentrated loading’), Deutscher Ausschuss fur Eisenbeton, No. 44 and 52,
Berlin
3. IOHANSEN,K. w.,‘Yield line theory’.English translation of the original Danish publication
(1931), Cement and Concrete Association, London (1962)
4. JOHANSEN,K. w.,‘Pladeformler’(‘Slab formulae’), Copenhagen
5. GEHLER, w.and AMOS, H.,‘Versuche mit kreuzweise bewehrten Platten’ (‘Tests with two-
way reinforced slabs’), Deutscher Ausschuss fur Eisenbeton, No. 70,Berlin (1932)
6. IOHANSEN,K. w.,‘NoglePladeformler’ (‘More slab formulae’), Copenhagen
7. LEVI, F., ‘Superfici d’influenza e fenomeni di adattamento nelle lastre piane’ (‘Influence
surfaces and adaptation phenomena in flat slabs’), Giornale del Genio Civile N o . 5 (1950)
8. HOGNESTAD, E., ’Yield line theory for the ultimate flexural strength of reinforced concrete
slabs’,J. A C Z (March 1953)
9. HAYTHoRNTHwAtTE, R. M. and SHIELD, R. T., ‘Anote -onthe deformable region in a rigid-
plastic structure’,J. Mech. Phys. Solids, 6,127-131 (1958)
10. STEINMANN,G. A., ‘La théorie des lignes de rupture’ (‘The yield line theory’), Bulletin
d’lnjormation,C.E.B., No. 27 (September 1960)
i i. ‘Instructionssoviétiques concernant le calcul des structures hyperstatiques en béton armé‘
(‘Soviet instructions concerning the design of reinforced concrete statically indeterminate
structures’) (Chapter IV). French translation, Bulletin d’Information,C.E.B., No. 28
(October 1960)
12. WOOD, R. M., ‘Plasticand elastic design of slabs and plates, London (1961)
i 3. NIELSEN,M. P., ‘Plasticitetsteorien for Jernbetonplater’ (‘Plastic theory for reinforced
concrete slabs’), Danmarks Tekniske Hojskolr, Copenhagen (1 962)
14. IONES,L. L., Ultimate load analysis of reinforced and prestressed concrete structures, Chatto
and Windus, London (1962)
15. NIELSEN,M. P., ‘Exact solutions in the plastic plate theory’, Bygningstatiske Meddelelser,
34,NO.1, 1-28 (1963)
16. MASSONNET, C.and SAVE,M.,Calcul plastique des constructions (Plastic design of structures),
Vol. II, Brussels (1963)
17. SOBOTKAO, I. z.,‘Etude de la capacité de résistance des dalles biases en béton armë (‘In-
vestigation of the strength of reinforced concrete skew slabs’), Bulletin d’Information,
C.E.B., N o . 38, 84-133 (March 1963)
18. SAWCZUK, A. and JAEGER, T., Grenztragfahigkeitstheorie der Platten (Limitstrength theory
of slabs), Springer-Verlag,Berlin (1963)
19. University of Illinois: Structurai research series’. Bulletins No. 181, 200, 211, 228, 249,
265, 277; Also: A C Z J O U R N A L Proceedings, 60,No. 9 (September 1963)
20. NIELSEN,M. P.,‘Limitanalysis of reinforced concrete slabs’.Acta Polytechnica Scandinavica
Ci 26, Copenhagen (1964)
21. MALDAGUE, I. c. ‘Essaisde dalles simples ou continues armées de différents types d‘acier’
(‘Tests on single or continuous slabs reinforced with different types of steel’). Bulletin
d’Information, C.E.B., No.44,47-88 (Oct. 1964)
22. CALLARI,GI E. ‘Méthodegénérale de calcul des dalles dans le domaine anélastique’(‘General
method for the analysis of slabs in the inelastic range’. Annales de l’Institut Technique,
Paris (September 1964) XVIIe Année, No. 201, Série ‘Théorieset méthodes de calcul’.
23. ‘Recent developments in yield line theory’.M.C.R. specialpublication (May 1965)
24. KWIECYNSKI,M. w.,‘Yield criterion for initially isotropic reinforced slab’, Magazine of
Concrete Research, 17,N o . 51 (June 1965)
25. KWIECYNSKI,M. w.‘Yield criterion for an orthotropically reinforced slab’,International
Journal of Solids and Structures, 1, No. 4
26. LEVI,F., ‘Contrôledes conditions de fissurationet de déformation des dalles dimensionnées
à l’état limite ultime’(‘Checkingthe cracking and deformation conditions of slabs designed
for the ultimate limit state’), Bulletin d’Information, C.E.B., No. 50,201-226 (July 1965)
27. KWIECYNSKI,M. w.,‘Some tests on the yield criterion for a reinforced slab, Magazine of
Concrete Research. 17,No.52 (1965)
28. ROBINSON,I. R. and KARAESMEN,E., ‘Etude expérimentale de dalles bordées de poutres’
(‘Experimentalinvestigation of slabs with edge beams’), Bulletin d’Information,C.E.B.,
No.50 (July 1965)
29. GVOZDEV, A. A. and KRYLOV, s. M.,‘Recherchesexpérimentales sur les dalles et planchers-
dalles effectuées en Union Soviétique’ (‘Experimental research on slabs and flat-slab
floors carried out in the Soviet Union’), Bulletin d’lnformarion,C.E.B., No.50, 174200
(July 1965)
30. KUANG-HAN CH, R A M B. SINGH,‘Yield analysis of balcony floor slabs’,J. ACI Proceedings
63,No.5. 571-586 (May 1966)
3i. MASSONNET. CH., ‘Théorie générale des plaques élasto-plastiques’ (‘General theory of
elasto-plastic plates’), Bulletin d’lnformalion,C.E.B.,No. 56 (August 1966)
32. CORNELIS,A., ‘Etudeà l’aide d’une calculatriceélectronique du comportement élastique des
plaques’ (‘Investigation of the elastic behaviour of plates with the aid of an electronic
computer’), Bulletin d’Information,C.E.B.. No.56 (August 1966)
 INDEX

Additional moment, 58, 59 Batching, continued

Additives, 148-150 cement, 150
Adhesive tapes, 130 Beams, 91
Affine transformation, method of, 363 bar spacings at intersections of, i 10
Aggregates, 137, 139-147 deep, rules for, 64
batching, 15 1 design of, 77
cleanliness, 141 diaphragm, 37-38
crushed stone, 146 supports of, 80
grading, 141 thin-walled,37
granulometric classification, 139 Bending, biaxial, 50-51, 54
maximum dimension, 141 composite, 88
nature and shape, 144-147 with compression, rectangular section,
rounded, 146, 152 229-234
shape (or volumetric) coefficient, 146 of reinforcing bars, 104
silico-calcareous,156 simple, rectangular section, 224-229
Aggressive atmosphere, 301, 302, 304 ultimate limit state for, 238
Air-entraining agents, 148 transverse, 38
Amplification coefficient, 25, 53, 182, 201- uniaxial. See Uniaxial bending
204, 206 Bending machines, 135
Anchorage, bond by, 82 Bending moment diagram, 31, 89
by curvature, 85-89 Bending moments, 36, 48, 116, 352
calculations,82 additional,55
dangerous case, 88 permissible, problem on, 253, 254
devices, 106 redistribution of, 29
of reinforcing bars, 82-91, 104 Bent-up bars, 122
straight,83-85 Binders, vertical or inclined, 120
total, 87 Binding, geometrical dimensions of zone
Aperture, of infinite length, 101 provided with, 1 1 5
presence of, 100 helical, 115
members reinforced by, 52, 11.5-116
members without, 51
Bar (notation), 5 minimum percentage of, 115
Bars, reinforcing. See Reinforcing bars reinforcement, 115
Basic strengths, 210 in mats, 116
definition of, 26 Binding coefficient, 53
determination of, 26-28 Bond, 82-92
of concrete, 27 by anchorage, 82
of steel, 27 by interaction, 82, 91-92
Batching, aggregates, 151 definition of, 82
409
 410
Bond length, 85, 290 Compressive zone, 78-79, 94,21 1
deformation along, 291 and torsional strength, 95
Bond stress, 83-84, 91, 300 width of, 40
anchorage, 302, 304 Concentric compression, 51-54, 57, 224, 236
distribution between concrete and steel,287 Concrete, 19-23, 210
Braced columns, 57 basic strength, 27
Buckling, 35, 36, 38, 54, 55, 117, 131 characteristic strength, 208
Buckling moment, 58 composition, 137-159
compressive strain, 43, 44,212
compressive strength, 19, 27, 164166
consistency, 169
Calcium chloride, 148-150 control tests, 160, 162
Calculations. See Design calculations creep. See Creep
Cantilevers,vertical, 37 crushing. See Crushing; Crushing test
Capping, 165 curing. See Curing
Castellations, 122 definition, 137
Cement, 137-139 deformations of, long-term linear, 23
air-entraining, 148 design strength, 208
batching, 150 elasticity modulus, 22-23
choice of, 138 freshly mixed, 137
classification and quality of, 137 handling, 168
Portland, 137, 138 high-density, 152
rapid-hardening,138 in compression,40,211-212
storage of, 139 information tests, 160, 162
Characteristic loadings, 200, 201, 210 making, 167-168
determination of, 24 putting materials into mixer, 167-168
effect of, 28 mechanical strength, 19
Characteristic strengths, 200, 201, 207-210 minimum strength of, problems on, 247,
concrete, 208 252, 256
steel, 208 mixing procedure, 168
Characteristic values, 201-207 normal, 152
definition,201-204 placing, 168
permanent loads, 204 plastic, 152
superimposed loads, 205-207 Poisson’s ratio,23
Chlorides, 149 prestressed, 138
Circular drop, 392 properties of, 237
Climatic superimposed loads, 26, 182 reference values of mechanical strength, 19
Cold weather, concreting in, 172 requirements, 137-174
Column-to-slabconnection zone, 390 sampling, 160
Columns, 111-115 shrinkage. See Shrinkage
braced, 57 strength of, 207
corner, 403, 407 stress distribution diagram, 21 1
edge, 403, 405 stress-strain diagram, 46,313
longitudinal reinforcement, 112 tensile strength, 2CL22, 28, 166167, 288,
minimum section, 1 1 1 302
piacing reinforcement in position, 114 testing strength of, 159-167
transverse reinforcement, i 13 age at testing, 164
Comité Européen du Béton, 24, 199, 200, compressive strength, 164-166
206, 209, 210, 287, 305 procedure, 164167
Compatibility equation,229, 232, 233 specimens, 162
Compression, concentric, 51-54, 57, 224, 236 tensile strength, 166167
concrete in, 40,211-212 thermal expansion coefficient,23
eccentric, 58, 59, 72,,:b 222, 234 Concrete cover to reinforcing bars. See
notation, 5 under Reinforcing bars
Compressive flange, effective width of, 40 Concrete mix, consistency of, 154-159
Compressive strain of concrete, 43, 44, 212 Concrete mix design, 152
of steel, 212 Concrete strain, determination of, 312
Compressive strength of concrete, 19, 27, Concreting,checks prior to, 168
164-166 in cold weather, 172
 41 1
Concreting,continued Cracking shear,286
in warm weather, 173-174 Creep,26,58,59,313,324-328
interruption and resumption of,170 effect of,60
Conditions of execution,126 influence of age of concrete at time of
Connection zone,column-to-slab, 390 loading,326
Connectors,73-76, 81 influenceof composition of concrete, 325
definition of,73 influence of least dimension of member,
fastening,83,106,107 325
design of,88-89 influence of time,327
transverse,83 Creep coefficient,324
Conseil International du Bâtiment, 24, 199, basic, 324
200,209,210 Cross-memberswith haunches,57
Construction joints, 17C171 Cross-sectionof member,problem on,250
Construction procedure,effect of,26 Crushing of concrete,103,105
Constructional arrangements,102-123 Crushing test, 19,161,162,164-166
and design assumptions,102 Curing,171-172
Control tests on concrete,160,162 membrane, 172
Correction factors,20 steam, 171
Crack spacing,maximum, 292,294,295,298 wet,172
Crack width,calculation of,290
maximum, 292,294,296,298
expressions for,299
permissible,301 Dance halls, i84
reference definition of,286 Deflection curves,71
Cracked members,deflection,315 general calculation of,315-316
Cracking,30, 53-54, 60,73, 83,94,120, 330 Deflections,cracked members,71, 315
accidental,290 general calculation of,315-3 16
analysis of,285-310 general rules for calculation of,71
purpose of,286 maximum permissible,70
calculation of,286 simplified calculation for ordinary build-
in flexural member,295-299 ings,316-319
in normally reinforced members, 291- uncracked members,72,316
299 Deformation(s), 69
in T-beam reinforced with medium- along bond length,291
tensile deformed bars,298 flexural,calculations of, 311-319
in T-beam reinforced with plain mild limit state of. See Limit state of de-
steel bars,297 formation
in tie-member reinforced with medium- of concrete, long-term linear,23
tensile deformed bars,294 simplified calculation for ordinary build-
in tie-member reinforced with plain ings, 316-319
mild steel bars,293 Design calculations, arithmetical accuracy,
in under-reinforcedmembers,288-291 124
purpose of,292 basic data for, 124
worked examples,293,297 object of, 1-2
checking of,62,300-310 submission of, 124
classification of structures according to Design formulae,364
consequences of, 61-62 for flat-slabfloors and mushroom floors,
flexural,295-299 364
in concrete,285 range of application,389
in reinforced concrete,286-300 Design graphs,310
limit state of. See Limit state of cracking Design loadings,201
longitudinal,286 Design method,general procedure,2
non-systematic,302 semi-probability,24,199-201,209,210
practical checking of,60 Design strengths,200,207-209
progressive development of,288 concrete,208
systematic,292,304 steel,208
tensile,285 Design values, 201-207
calculation of,292 permanent loads,204
web, 64 superimposed loads,205-207
 412
Designs, fundamental assumptions, 21 1 Fictitious thickness, 321,325
preparation of, 12&127 Flanges and ribs,junctions between, 81
Diaphragm beams, 37-38 Flexural members, 116123
Direct force, 54 changes in geometrical shapes of sections,
Dispersion coefficient,202,206,207 118
Displaced diagram, 90,116 Flexural test, 21, 161,167
Disturbed zones, concept of, 342 Floors, 36
Draw effect, 293 flat-slab,56,393
Drawings, formwork, 126 design formulae, 364
preliminary design, 125 mushroom, 56,393
reinforcement, 126 design formulae. 364
working, 125-126 ribbed, 40,56,78. 79,318
Dwellings, 183 Formwork, carefully finished, 128
Dynamic coefficient,25. 182 checking, 169
Dynamic pressure, 189-190 classification,128
formula, 190 cleaning, 131
nominal and exceptional, 190 fine-faced, 129
values, 190 joints in, 130
maintenance, 132
mechanical properties, 130
Earthquake. 26, 107, 182,206 oiling, 131
Eccentric compression, 58, 59, 72, 88, 222, ordinary, 128
234 preparation, 131
Eccentricity, 35-36, 54, 58, 59, 235 requirements, 128-133
maximum, problems on, 270,276 sag and camber, 131
Effective length, 55 special, 129
Effective width of compressive flange, 40 striking, 132
Elastic behaviour, 330 tightness, 130
Elastic limit,fictitious, 310 wetting, 131
of reinforcing bars, 107.112 Formwork drawings, 126
of steel, 17 Frost, 172-173
Elastic theory, 34,35 Full fixity, 56
Elasticity modulus of concrete,22-23
Elasto-plastic design with limited rotations,
31
Elasto-plastictheory, 332 Garages, 184
Elongated shape, outline of, 100 Geometrical length, 56
Embedment section, 62
G/Scoefficient method, 152
Equilibrium, method of, 350-363
Equilibrium of forces, equation of, 222
Equilibrium of moments, equation of, 223 Haunches, 118
Equilibrium equations, 215, 228, 229, 232, cross-members with, 57
233,236, 350, 361-363 Helical binding, 115
Euler criticai stress, 36, 58 Hooke’s law, 18, 49.212
Euler slenderness ratio, 55 Hooks, 105, 106
Expansion. See Thermal expansion standard, 86,87
External tensile forces, 285,289 Hoops, 115
Hospitals, 183

Failure, 335
determination of moment of, 215, 218 Indices, 5
determination of type of, 214 Information tests on concrete, 160, 162
due to tensile cracking, 285 Instability, limit state of. See Limit state of
envisaged mode of, definition of, 329-330 instability
flexural, 33C~345 Interaction, bond by, 82
behaviour of plane reinforced concrete
structures up to, 330-331
prediction of nature of, 213 Kinematic theorem, 334,344
types of, 213 Kinking effect,343-344
 413
Lapping. splicing of bars, by 90, 105 Mortar, R I L E M standard, 138
Laps in compressive reinforcement, 106 Moulds for test specimens, 160-162
in tensile reinforcement, 105 Multi-storey buildings, 184
Lattice hypothesis, 76, 92, 94
Length, geometrical, 57
Limit analysis, general theory of, 333 Navier-Bernoulli hypothesis, 39, 21 1
application, 336345 Nodal forces, 343, 346, 351
Limit state, 2, 28, 54 calculation, 356-361
envisaged, definition of, 329 determination, 351-36 1
ultimate, 39, 204, 208. 209, 213, 33G345 edge, 361
analysis methods, 331-345 expressions for, 358-361
for simple bending, 238 method of, 350-363
Limit state of cracking, 60-69, 205, 208, 209, Non-linear design, 32
286 Normal criterion,342
definition,60 Notation, &15
fundamental design assumptions, 60
Limit state of deformation, 69-73, 205, 208,
209
definition of, 69 Offices, 183
fundamental design assumptions, 69 Orthotropic reinforcement, 354, 355
simplified rules for ordinary buildings, 72
Limit state of instability, 5&60, 205, 208, 209
columns loaded in concentric compression Parabolic diagram, 46, 48
57 Parallelepipedal drop, 392
in eccentric compression, 58 Parapets, horizontal forces on, 186
fundamental design assumptions, 54 Partitions, lightweight demountable, 185
of plates loaded parallel to plane, 59 load-bearing,35
Linear members, structures composed of, Permanent loads, 25, 202, 203
29-32 characteristic values, 204
Load, permanent. See Permanent loads design values, 204
superimposed. See Superimposed loads determination of effects of, 29-38
Load-bearingpartitions, 35 Plane structures, 69
Load capacity, 223 loaded parallel to middle plane, 34
Loading tests, 186 loaded perpendicularly to middle plane, 32
Loadings, characteristic. See Characteristic Plastic behaviour, 331
loadings criterion,336-337
Lower boundary theorem, 334, 344 Plastic design, 31
Plastic hinges, 32
Plastic theories, 34
Materials, properties of, 237 Plasticisers, 148
determination of, 16-23 Plasticity,instantaneous,313
Mats, binding reinforcement in, 116 long-term,313
Mechanical reference properties of steel, 17 Plastification, 330-33 i
Mechanical strength of concrete, 19 Poisson’s ratio for concrete, 23
Method of affine transformation, 363 Portland cement, 137, 138
Method of equilibrium, 350-363 Prime, use of, 5
Method of nodal forces, 350-363 Probability theories of safety, preliminary
Method of superposition, extension of, 363 considerations, 198-199
Method of work, 349-350 principles of, 199
Minimum effective percentage,289, 291 Proof stress, 19
Mobile equipment, 186 Properties of materials, 237
Modular ratio, 1 determination of, 1 6 2 3
Modulus of elasticity.See Elasticity modulus Protected ordinary structures, 301, 303. 310
Mohr’s circle, 339, 340, 342 Public entertainment buildings, 184
Moment-curvature diagrams, 31 Punching shear,97-101, 330
Moment equation,228, 229 limit value of resistance to, 98
Moment-load relations,405,407 Punching shear reinforcement, 120-123
Moment-ultimate load relations,393 Punching shear strength, 97
Moments, redistribution of, 30 determination of, 98-101
414
Rectangular diagram,47 Reinforcement coniinued
Rectangular section,280-284 positive,407
analysis of,218-236 punching shear,120-123
composite bending with compression, requirements,133-137
229-234 skin,64,117
concentric compression,236 suspension,81
eccentric compression,23&236 tensile,49
for simple bending,22&229 laps in, 105
procedure for,228,233-235 problems on. 245.247,271
determination of failure moment,218 tensile longitudinal,116
general formulae,272 tolerances, 175-176
reinforcement,219-224 torsional,95
with compressive reinforcement,226,258- transverse,78,80,82,83,88,96,106,113,
26I I17
without compressive reinforcement, 225. constructionalarrangements,114
250-258 contribution to torsional strength,94
Redistribution of forces and moments. 30, effect of,287
203-204 minimum percentage,77,93.114
Redistribution coefficients,32 under-reinforcedmembers,288,300-304
Reduction coefficients,27-28. 43,49,9496, welded fabric,90
192,200,201,207. 208,210,313, 314 see also Reinforcing bars
Reference moment,349,352 Reinforcement cage, 114
Reference resisting moment, 346 Reinforcement drawings, 126
Region coefficient,190 Reinforcing bars,62
Reinforced concrete structures, field of anchorage of,82-91,104
application of Manual, 1 bend tests, 134
Reinforcement,61,390 bending, 104,134
anisotropic,336 minimum diameter of forming mandrel
binding, 52,115 135
in mats, 116 bending method,135
compressive,50,258 bending speed,136
laps in, 106 bent-up,122
longitudinal, 117 classification,16
congestion of, 169 compressive,91
connector,73-76,81.95 concrete cover to,110-11 i
corner,120 all bars, 110
cross-sectional area of, problem on, 250, 'groupsof bars in contact,1 1 1
255-257,259.264,277,283 main bars, 1 1 1
design rules for,302-310 minimum, 175
normally reinforced members,304310 curtailment of,105
under-reinforcedmembers,302-304 definition,16
designing and checking,219-224 deformed,92
edge, 119 design of,64
general conditions relating to, 102-1 11 devices used at ends, 105
high percentage of,305 diameters, 17,63
isotropic,336,354,355 differentgrades or types of, 126
longitudinal,95.i 12 elastic limit,107,112
constructionalarrangements,iì3 groups of bars in contact,109-110
minimum percentage, 112 interaction bond,91-92
longitudinal distribution,117 length of,407
low percentage,305 longitudinal,95.1 i2
mechanical percentage,30 curtailment of,89-90
mid-span,119 mechanical properties reference values,133
minimum effective percentage,289,291 permissible curvature,103
minimum strength of,problem on,249 positioning,137
miscellaneous systems,122 spacers,137
negative,407 spacing,108-1 10
orthogonal,336 at intersectionsof beams, 110
positioning, 137 in same horizontal layer, 109
 41 5
Reinforcing bars continued Shear force continued
in same vertical line,108 design rule for,76
splices in. 104108 torsion combined with,96
lapped,90,105 total resistance capacity for,79
staggered,106 Shear heads, 122
welded, 107,108 Shear resistance capacity,78-79. 81-82
straightening, 136 Shear strength,calculation of,76
sudden changes of section,105 Shells,89,91
tensile longitudinal,i 16 design of,82
tests on, 133 Shops, 183
transverse,73,113 Shrinkage, 26, 54, 285, 293, 295, 298. 299,
use of different grades or different types, 315,32&324
102 influence of composition of concrete,322-
welding,107-108, 136-137 324
Relative failure moment,224 influence of least dimension of member.
Relative standard deviation,202,207,208 32i
Relative ultimate moment, 224 influence of time,324
Residential buildings, 25,26 Shrinkage coefficient,320
Resisting moment,48 basic,321
problems on,244,270,276 Site coefficient,191
Ribbed floors.See Floors Skin reinforcement,64,117
Ribs,design of,77 Slab-to-columnconnection. 403
junctions between flanges and,81 Slabs. 69
RILEM standard mortar, 138 analysis of,334
Ritter-Mörschtheory,76 design,82
Roofs, 186,193,195 design formulae,364
cladding elements, 196 punching shear strength,97
flat,183 Slenderness ratio,55,72,318
Rules of good construction,60-61 Slump test, 154,156, 159
Snow,26,182
Spalling,88
Splices.See under Reinforcing bars
Safety,checking for,method of,209-210 Splitting test, 19,21, 161,162,166-167
principle of,24 Staggered splice, 106
problems on,267-270, 275,282 Standard hook,86,87
determination of,24-28. 198-210 Static theorem,335
factor of, 198,201,210,238 Statically determinate system,31
preliminary considerations,198 Statistical data, 198,200
probability theories of, preliminary con- Steel,16-19,49,50,210
siderations,198-199 basic strength,27
principles of, 199 characteristic strength,208
structural,problem on,240 compressive strain,212
Sampling,concrete,160 design strength,208
Sand, 144 elastic limit,17
Sand equivalent (SE) test, 144 mechanical reference properties, i7
Schools,183 properties of,237
Sea water, 147 strength of,207
Sections. changes in geometrical shapes of, stress-strain diagram, 18-19,49,50,212
118 tensile strain,212
determination of,39-101 Steel strain,312
estimation of local strength of,29 Steel stress, 29,50
rectangular.See Rectangular section Steelfixing tolerances,175
sudden changes of, 105 Stiffeners,35
symmetrical.See Symmetrical section Stiffness,55-56
Semi-probabilitydesign method,24,199-201, Stores,183
209,210 Strain.concrete,determination of,3i2
Setting accelerators,148 steel,determination of,312
Setting retarders,148 Strain compatibility condition,211, 222
Shear force,38,76-82, 350 Strain compatibility equation,21 i, 223
 416
Strain hypothesis,51 Terminal forces,350
Strength,basic. See Basic strength calculation of,354-356
guaranteed minimum,26 Theatres,184
Stress,steel, 29,50 Thermal expansion coefficient of concrete,23
Stress distribution diagram for concrete,211 Tie-members,calculation of cracking in,293,
Stress hypothesis,51 294
Stress-straindiagram,basic compressive,213 Tie-rod analogy,63,299-300, 304
for concrete,46,313 Tolerances,174-1 76
for steel, 18-19, 49,50,212 dimensional,174
Subscripts,5 minimum concrete cover to reinforcement,
Superimposed loads, 181,202,203 175
characteristic values,205-207 more than one, 176
climatic,26,182 on perpendicularity,174
definitions,181 on straightness,175
design values,205-207 position of main reinforcement. 176
determination of effect of,29-38 position of transverse reinforcement,176
dynamic,25, 182 reinforcement, 175-176
exceptional,206 steelfixing,175
fixed,25, 181 Torsion,92-97
nominal values, 181 combined with shear force,96
superposition of,206 Torsional moments,350,352
variable,25, 181,182-187 Torsional resistancecapacity,94,96
rules relating,184-187 Torsional strength,and compressivezone,95
Superposition,method of, extension of, 363 calculation of,92
of superimposed loads,206 transverse reinforcement contribution, 94
Supports of beams,80 Total resistance capacity for shear force, 79
Suspension reinforcement,81 Truncated pyramidal drop,390
Swelling,64,291,310
Symbols,5
Symmetrical section, arbitrary, analysis of
213-218
analysis of, general procedure of, 217-
U-hook,86
228 Ultimate limit state, 39, 204. 208, 209, 213,
with respect to piane of bending, 238- 330-345
250,267-272, 279-280 analysis methods, 331-345
expression for upper limit of moment,216 for simple bending,238
expression for b/h)limit,214 Ultimate strength,analysis of,39
failure moment,215 Unbalanced thrust,88-89, 105,107
failure type,214 Under-reinforcedmembers,288,302-304
y/h as function of section properties and
UNESCO simplified design method,210
external loadings,213 Unfitness,notion of, 1-2
Uniaxial bending,39,43,51,54
composition,with compression, 267-279
with tension,279-284
T-beams,40,43,44,78,79,96,261,297,298 practical design calculations,237-284
T-section,261-267 simple,238-267
analysis of,264 theoretical analysis,211-236
problems on,264,266 Uniqueness,theorem of,335
effective width of compressive flange, Units,3 4
261-264 basic, 3
Tangential actions,73,83 relations between S.I. units and metre/
fundamental design assumptions,73 kilogramme-force/secondsystem,4
Temperature variations,26,285,288 secondary,3
Tensile forces,external,285,289 S.I.system,3
Tensile strain of steel,212 Unprotected ordinary structures, 301, 303,
Tensile strength of concrete, 20-22, 28, 305
166-167,288, 302 Unsupported corner, 101
Tension,concentric,53-54 Unsupported edge, 100
notation,5 Upper boundary, theorem of,335
 417
Vehicle access, 184 Wind effects continued
Vibration, 162,164,168-170 internal actions,193,196
external. 169 local actions, 195-196
internal,169,170 on one face of wall element, 189
surface,169,170 overall actions,197
Vibrations, 182 projected area, 189
Vibrator, 109,162,169,170 reductions,191
screening,191
sphere of application,187
suction effects, 195
Wall element,wind effect on one face of,189 unit values of resultant actions on walls
Walls, 35 and roof slopes,196
Warehouses,184 Work, method of,349-350
Warm weather,concreting in. 173-174 Workshops,184
Water, 147-148
permanent contact with, 64
quantity of, 154 Yield line method, 344
Watertight structures,61,301,302,304 Yield line patterns,346347,364
Welding of reinforcement, 107-108, 136 Yield line theory,34,345-364
Wetting agents, 148 application of,345
Wind-bracing,35, 36,37 failure mechanism,346-349
Wind effects,26,106-107,182, 187-197,206 principles of,345
blocks joined together in single row and Yield lines,335,337,338,342,343,345,355,
covered by one roof, 197 356. 358
checking procedure, 187 negative,393
definitions and general principles,188 positive,405
dynamic pressure,189-190 Yielding,331. 335,343
exposure of surfaces, 188 stepped criterion of (Johansen), 336-345,
external actions, 193 358