Designers' Guide To en 1992-1-1 and en 1992-1-2

DESIGNERS'
EUROCODES
GUIDES TO THE
an ice initiative
R. S. Narayanan & A. Beeby
E
DESIGNERS' GUIDE TO EN 1992 ... 1 ... 1 AND EN 1992 .. 1
EUROCODE 2: DESIGN OF CONCRETE STRUCTURES
GENERAL RULES AND RULES FOR BUILDINGS AND
STRUCTURAL FIRE DESIGN
Designers' Guide to EN 1990. Eurocode: Basis of Structural
M. HolickY-. 0 7277 30 I I 8. Published 2002.
H. Gulvanessian, Calgaro and
Designers' Guide to EN /994-1-1. Eurocode 4: Design of Composite Steel and Concrete Structures. Part 1.1:
General Rules and Rules for Buildings. R. P. johnson and D. Anderson. 0 7277 3151 3. Published 2004.
Designers' Guide to EN /997-1. Eurocode 7: Geotechnical Design General Rules. R. Frank, C. Bauduin,
R. Driscoll, M. Kavvadas, N. Krebs Ovesen, T. Orr and B. Schuppener. 07277 3154 8. Published 2004.
Designers' Guide to EN 1993-/-1. Eurocode 3: Design of Steel Structures. General Rules and Rules for Buildings.
L. Gardner and D. Nethercot. 0 7277 3! 63 7. Published 2004.
Designers' Guide to EN /998-1 and EN 1998-5. Eurocode 8: Design Structures for Earthquake Resistance.
General Rules, Seismic Actions and Rules for Buildings and Foundations. M. Fardis, E. Carvalho, A. Elnashai,
E. Faccioli, P. Pinto and A. Plumier. 0727733486. Published 2005.
Designers' Guide to EN 1992-1-1 and EN 1992-1-2. Eurocode 2: Design of Concrete Structures. General Rules
and Rules for Buildings and Structural Fire Design. A. W. Beeby and R. S. Narayanan. 07277 3105 X. Published
2005.
Designers' Guide to EN 199/-4. Eurocode I: Actions on Structures. Wind Actions. N. Cook. 0 7277 3152 I.
Forthcoming: 2005 (provisional).
Designers' Guide to EN 1996. Eurocode 6: Part 1./: Design of Masonry Structures. J. Morton. 0 7277 3155 6.
Forthcoming: 2005 (provisional).
Designers' Guide to EN 1995-1-/. Eurocode 5: Design of Timber Structures. Common Rules and Rules for
Buildings. C. Mettem. 0 7277 3162 9. Forthcoming: 2005 (provisional).
Designers' Guide to EN 1991-1-2,1992-1-2, /993-/-2 and EN 1994-/-2. Eurocode I: Actions on Structures.
Eurocode 3: Design of Steel Structures. Eurocode 4: Design of Composite Steel and Concrete Structures. Fire
Engineering (Actions on Steel and Composite Structures). Y. Wang, C. Bailey, T. lennon and D. Moore.
a 7277 3157 2. Forthcoming: 2005 (proviSional).
Designers' Guide to EN /992-2. Eurocode 2: Design of Concrete Structures. Bridges. D. Smith and C. Hendy.
0727731599. Forthcoming: 2005 (provisional).
Designers' Guide to EN 1993-2. Eurocode 3: Design of Steel Structures. Bridges. C. Murphy and C. Hendy.
07277 3160 2. Forthcoming: 2005 (provisional).
Designers' Guide to EN 1994-2. Eurocode 4: Design of Composite Steel and Concrete Structures. Bridges.
R. Johnson and C. Hendy. 0 7277 3161 O. Forthcoming: 2005 (provisional).
Designers' Guide to EN 1991-2,/991-1-1, 1991-f-3 and 199/-1-5 to /-7. Eurocode I: Actions on Structures.
Traffic Loads and Other Actions on Bridges. j.-A Calgaro, M. Tschumi, H. Gulvanessian and N. Shetty.
o 7277 31564. Forthcoming: 2005 (provisional).
Designers' Guide to EN 1991-1-1, EN 1991-/-3 and 199 f -1-5 to /-7. Eurocode I: Actions on Structures. General
Rules and Actions on Buildings (not Wind). H. Gulvanessian, J.-A. Calgaro, P. Formichi and G. Harding.
07277 3158 O. Forthcoming: 2005 (provisional).
www.
EE
DESIGNERS' GUIDE T EN 1992 I-I
D EN 1992-1-2
EUR C DE 2: DESIG C CRETE
STRU URES
GENERAL RULES D RULES F R
BUILDINGS A STRU U FIRE DESIGN
w. and
Series editor
Published by Thomas Telford Publishing, Thomas Telford
URL: www.thomastelford.com
1 Heron London E14 4JD
Distributors for Thomas Telford books are
USA: ASCE Press, 1801 Alexander Bell Drive, Reston, VA 20191-4400
Japan: Maruzen Co. Ltd, Book Department, 3-10 Nihonbashi 2-chome, Chuo-ku, Tokyo 103
Australia: DA Books and Journals, 648 Whitehorse Road, Mitcham 3132, Victoria
First published 2005
Eurocodes
Structural Eurocodes offer the opportunity of harmonized standards for the European
construction market and the rest of the world. To achieve this, the construction industry needs to
become acquainted with the Eurocodes so that the maximum advantage can be taken of these
opportunities
Eurocodes Expert is a new ICE and Thomas Telford initiative set up to assist in creating a greater
awareness of the and implementation of the Eurocodes within the UK construction industry
Eurocodes provides a range of products and services to aid and support the transition to
Eurocodes. For comprehensive and useful information on the adoption of the Eurocodes and their
implementation process please visit our website or email [email protected]
A catalogue record for this book is available from the British Library
ISBN: 07277 3105 X
The authors and Thomas Telford Limited 2005
All rights, including translation, reserved. Except as permitted by the Copyright, Designs and Patents
Act 1988, no part of this publication may be reproduced, stored in a retrieval system or transmitted in
any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written
permission of the Publishing Director, Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay,
London E14 4JD
This book is published on the understanding that the authors are solely responsible for the statements
made and opinions expressed in it and that its publication does not necessarily imply that such
statements and/or opinions are or reflect the views or opinions of the publishers. While every effort has
been made to ensure that the statements made and the opinions expressed in this publication provide a
safe and accurate guide, no liability or responsibility can be accepted in this respect by the authors or
publishers
Typeset by Brighton and Rochester
Printed and bound in Great Britain by MPG Books, Bodmin
Preface
I ntrod uction
EN 1992-1-1 (General Rules and Rules for Buildings) and EN 1992-1-2 (General Rules
Structural Fire Design) will replace BS 8110 Parts 1 and 2 in the near future. While the broad
requirements of EN 1992-1-1 and EN 1992-1-2 are not dissimilar to those in BS 8110,
designers will still need to get used to new terminology, set of new documents and the
interaction between them.
This guide has been written with the aim of providing practising civil engineers with some
insight into the background to EN 1992-1-1 and EN 1992-1-2. The authors have been
involved with the evolution of the codes from their ENV (pre-standard) status. The guide
starts with a brief outline of the Eurocode system and terminology. The code requirements
are illustrated by some local examples. Some design aids are also provided. The guide can be
used anywhere in Europe; but it should be noted that the UK values for the Nationally
Determined Parameters (set by the UK National Annex at the time of going out to print)
have been used in the handbook generally. Some adjustments may be required in this regard
when used outside the UK.
All the practical aspects of application of EN 1992-1-1 and EN 1992-1-2 to prestressed
concrete design are included in Chapter 11 of this guide. The depth of coverage is limited,
but the authors are indebted to Mr Keith Wilson of Faber Maunsell for drafting this chapter.
It is hoped that this guide will facilitate the effective use of Eurocode 2 by designers.
Layout
All cross-references in this guide to sections, clauses, subclauses, paragraphs, annexes,
figures, tables and expressions of EN 1992-1-1 are in italic type, which is also used where text
from EN 1992-1-1 has been directly reproduced (conversely, quotations from other sources,
including other Eurocodes, and cross-references to sections, etc., of this guide, are in roman
type). Expressions repeated from EN 1992-1-1 retain their numbering; other expressions
have numbers prefixed by D (for Designers' Guide), e.g. equation (DS.l) in Chapter S. The
foregoing also applies to cross-references to EN 1992-1-2, discussed in Chapter 12.
R. S. Narayanan
A. W Beeby
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Contents
Preface v
Introduction v
Layout of this guide v
Chapter 1 Introduction 1
1.1. Scope 1
1.2. Layout 1
1.3. Related documents 2
1.4. Terminology 2
Chapter 2 Basis of design 5
2.1. Notation 5
2.2. General 5
2.3. Fundamental requirements 5
2.4. Limit states 6
2.5. Actions 6
2.5.1. Classifications 6
2.5.2. Characteristic values of action 6
2.5.3. Design values of actions 7
2.5.4. Simplified load combinations 12
2.6. Material properties 12
2.6.1. Characteristic values 12
2.6.2. Design values 12
2.7. Geometric data 13
2.S. Verification 13
2.9. Durability 13
Example 2.1 14
Example 2.2 16
Example 2.3 16
Example 2.4 16
Chapter 3 21
3.1. Introduction 21
3.2. Load cases and combination 21
3.3. Imperfections 25
3.3.1. General 25
3.3.2. Global analysis 25
DESIGNERS' GUIDE TO EN 1992-1-1 AND EN 1992-1-2
Chapter 4
Chapter 5
viii
3.3.3. of slender elements
3.3.4. Members transferring forces to elements
3.4. Second-order effects
3.5. Time-dependent effects
3.6. Design by testing
3.7. Structural analysis
3.7.1. Elastic analysis with or without redistribution
3.7.2. Plastic analysis
3.7.3. Non-linear analysis
3.7.4. Strut-and-tie models
3.8. Design aids and simplifications
3.8.1. One-way spanning slabs and continuous beams
3.8.2. Two-way spanning slabs
3.8.3. Flat slabs
3.8.4. Beams
3.8.5. Simplifications
Materials and design data
4.1. Concrete
4.1.1. General
4.1.2. Strength
4.1.3. Elastic deformation
4.1.4. Creep and shrinkage
4.1.5. Stress-strain relationships
4.2. Lightweight concrete
4.2.1. General
4.2.2. Density classes
4.2.3. Other modifying coefficients
4.2.4. Design compressive strength
4.3. Reinforcing steel
4.3.1. Strength (fyk)
4.3.2. Ductility
4.4. Prestressing steel
V-",,.,,,.u of sections for "'''.,.u ... ~ and axial force
5.1. Basic assumptions
5.1.1. Stress-strain curves
5.1.2. Assumptions relating to the strains at the ultimate
limit state
5.2. Limiting compressive strains
5.2.1. Singly reinforced beams and slabs
Example 5.1: singly reinforced rectangular section
5.2.2. Doubly reinforced rectangular sections
5.2: rectangular beam with compression reinforcement
5.2.3. Design of flanged sections (T or L beams)
5.2.4. Checking the moment of resistance of more complex
section shapes
5.2.5. Design of rectangular column sections
Example 5.3: rectangular column section
5.2.6. Design for biaxial bending
Example 5.4: biaxially bent column section
5.2.7. of prestressed sections
26
26
26
27
27
28
28
34
35
36
38
38
38
43
49
49
51
51
51
51
51
52
52
56
56
56
57
57
57
57
58
58
59
59
59
63
63
64
66
66
67
67
67
68
78
78
81
83
CONTENTS
6 shear and torsion 85
6.1. Shear - general 85
6.2. Background to the code provisions 85
6.2.1. Members without shear reinforcement 85
6.2.2. Strength of sections with shear reinforcement 86
6.2.3. Maximum shear strength of a section 89
6.2.4. Shear capacity enhancement near supports 90
6.2.5. Summary 91
6.3. Summary of the provisions in clause 6.2 of EN 1992-1-1 91
Example 6.1: T section 93
6.4. Torsion 95
6.4.I. Introduction 95
6.4.2. Evaluation of torsional moments 95
6.4.3. Verification 96
6.5. Punching shear 98
6.5.I. General 98
6.5.2. Basic control perimeter 98
6.5.3. Design shear force 99
6.5.4. Punching shear resistance of slabs without shear
reinforcement 102
6.5.5. Reinforcement for punching shear 103
Example 6.2: lightly loaded slab-column connection 104
Example 6.3: heavily loaded slab-column connection requiring
shear reinforcement 105
Chapter 7 Slender columns and beams 107
7.1. Scope 107
7.2. Background to design of columns for slenderness effects 107
7.3. Design for slenderness effects 112
7.3.l. Basic approaches 112
7.3.2. First-order moments 112
7.3.3. Moment magnifier method 113
7.3.4. Nominal curvature method 115
Example 7.1: uniaxially bent rectangular column 119
7.3.5. Other factors 121
7.3.6. Global second-order effects 123
7.3.7. Walls 123
7.3.8. Lateral buckling of slender beams 124
Chapter 8 Serviceability limit states 127
8.l. General 127
8.1.1. Assessment of design action effects 127
8.1.2. Material properties 128
8.2. Limitation of stresses under serviceability conditions 129
8.2.1. General 129
8.2.2. Procedure for stress checks 130
Example 8.1 135
8.3. Control of cracking
,148
8.3.1. General 148
8.3.2. Minimum areas of reinforcement 150
8.3.3. of the phenomena 153
8.3.4. Derivation of crack prediction formulae 155
8.3.5. Checking without direct calculation 157
ix
x
~ " , U " U I - " V 8.2
8.3.6. Checking cracking by direct calculation
159
159
'-' .... IJ' . "'. 9 161
9.1. General 161
9.1.1. Historical perspective 161
9.1.2. Common mechanisms leading to the deterioration of
concrete structures 162
9.1.3. Relative importance of deterioration mechanisms 164
9.2. Design for durability 165
10.1. General
10.2. Discussion of the general requirements
10.2.1. Cover to bar reinforcement
10.2.2. Spacing of bars
10.2.3. Mandrel diameters for bars
10.2.4. Basic anchorage length
10.3. Anchorage of longitudinal bars
10.3.1. Design anchorage length
10.3.2. Transverse reinforcement at anchorage
10.4. Anchorage of links
10.5. Laps for bars
10.5.1. General
10.5.2. Design lap length
10.5.3. Transverse reinforcement
10.6. Additional requirements for large diameter bars
10.7. Requirements for weld mesh
10.8. Bundled bars
10.9. Detailing requirements for particular member types
10.9.1. Beams
10.9.2. Slabs
10.9.3. Columns
10.9.4. Walls
10.9.5. Corbels
10.9.6. Nibs
10.9.7. Reinforcement in flat slabs
Chapter 11 Prestressed concrete
11.1. General
11.2. Summary of main clauses
11.3. Durability
11.4. Design data
11.4.1. Concrete
11.4.2. Prestressing steel
11.4.3. Partial factors
11.5. Design of sections for flexure and axial load
11.5.1. Ultimate limit state
11.5.2. Serviceability limit state
11.6. Design of sections for shear and torsion
11.6.1. Shear
11.6.2. Torsion
169
169
169
170
170
170
170
170
173
173
175
175
175
175
175
175
176
177
177
177
181
182
183
184
185
185
187
187
187
187
190
190
190
192
192
192
193
195
195
197
CONTENTS
11.7. Prestress losses 197
11.7.1. General 197
11.7.2. Friction in jack and anchorages 197
11.7.3. Duct friction 197
11.7.4. Elastic deformation 197
l1.7.S. Anchorage draw-in or slip 198
11.7.6. Time-dependent losses 198
11.8. Anchorage zones 199
11.8.1. Pre tensioned members 199
11.8.2. Post-tensioned members 200
11.9. Detailing 201
11.9.1. Spacing of tendons and ducts 201
11.9.2. Anchorages and couplers 201
11.9.3. Minimum area of tendons 201
11.9.4. Tendon profiles 203
Chapter 12 Structural fire design 205
12.1. Aims of design 20S
12.2. Design procedure 205
12.3. Actions and partial factors 206
12.3.1. Actions 206
12.3.2. Material factors 206
12.4. Member analysis using tabular data 206
12.4.1. Scope 206
12.4.2. Basis for the tabulated data 207
12.4.3. Discussion of some features 207
12.S. Simple calculation methods 208
12.S.1. SOOC isotherm method 208
12.5.2. Zone method 209
12.5.3. Annex E method 209
References 211
Index 213
xi
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
CHAPTER I
I ntrod uction
1.1. Scope
Eurocode 2, Design of Concrete Structures, will apply to the design of building and civil
engineering structures in plain, reinforced and prestressed concrete. The code has been
written in several parts, namely:
EN 1992-1-1, General Rules and Rules for Buildings
EN 1992-1-2, General Rules - Structural Fire Design
EN 1992-2, Reinforced and Prestressed Concrete Bridges
EN 1992-3, Liquid and Containment Structures.
EN 1992-1-1 has been written in such a way that the principles of the code will generally
apply to all the parts. The specific rules, which only apply to building structures, are
identified as such. Under the CEN (the European standards body) rules, other parts of
Eurocode 2 are allowed to identify those clauses in Part 1.1 which do not apply to that part
and provide other information that will complement Part 1.1.
This guide is concerned primarily with Part 1.1. Some limited information on Part 1.2 is
also provided in Chapter 12.
Part 1.1 covers in situ and precast structures using normal-weight or lightweight concrete.
It applies to plain, reinforced and prestressed concrete structures. Thus, many of the
separate parts of the ENV versions of the code covering the above topics have been brought
into one document. Part 1.1 has 12 main chapters and 10 annexes; Part 1-2 has six main
chapters and five annexes.
Compliance with the code will satisfy the requirements of the Construction Products
Directive in respect of mechanical resistance.
layout
The code clauses are set out as Principles and Application Rules. Principles are identified by
the letter P following the paragraph number. Application Rules are identified by a number in
parentheses.
Principles are general statements and definitions for which there are no alternatives. In
addition, they also include some requirements and analytical models for which no alternative
is allowed unless specifically stated.
Application Rules are generally accepted methods, which follow the principles and satisfy
their requirements. It is permissible to use alternative design rules, provided that it can be
demonstrated that they comply with the relevant principles and are at least equivalent with
regard to structural safety, serviceability and durability to the rules in the code. This matter
should be approached with caution. A narrow interpretation of this will provide
DESIGNERS' GUIDE TO EN 1992-1-1 AND EN 1992- -2
2
no incentive to develop alternative rules. Equivalence could be defined more broadly as
meaning that the safety, serviceability and durability that may be expected from these
rules will be sufficient for the purpose. If this is accepted, procedures in the current national
codes are, by and large, likely to be acceptable, as the principles are likely to be similar.
Clearly, any alternative approach has to be acceptable to regulatory authorities. Provisions
of different codes should not be mixed without a thorough appraisal by responsible bodies. It
should also be noted that the design cannot be claimed to be wholly in accordance with the
Eurocode when an alternative rule is used.
Building regulations will not be harmonized across Europe, and safety in a country remains
the prerogative of individual nations. Therefore, in the Eurocode system, some parameters
and procedures are left open for national choice. These are referred to as Nationally
Determined Parameters (NDPs). These generally relate to safety factors, but not exclusively
so. Although at the outset of the conversion of ENVs into ENs, there was a desire by all
countries to keep the number of NDPs to a minimum, in practice it has proved difficult to
achieve this, and a number of parameters other than safety factors have also become NDPs.
The code provides recommended values for all NDPs. Each country is expected to state in
their National Annex to the code (which together with the code is likely to form the basis of
regulatory control in the country) whether the recommended value is to be changed. Where
the UK National Annex alters the recommended value of an ND P, it is identified in this guide.
Chapters are arranged generally by reference to phenomena rather than to the type of
element as in UK codes. For example, there are chapters on bending, shear, buckling, etc.,
but not on beams, slabs or columns. Such a layout is more efficient, as considerable
duplication is avoided. It also promotes a better understanding of structural behaviour.
However, some exceptions exist such as a chapter on detailing particular member types.
Eurocode 2 refers to a number of CEN standards and codes and some ISO standards where
relevant. The following are some of the more important:
e EN 1990, Basis of Structural Design
" EN 1991, Actions on Structures
" EN 206-1, Concrete Specification, Peiformance, Production and Conformity
EN 10,080, Steel Reinforcement of Concrete
EN 10,138, Prestressing Steels.
Eurocode 2 relies on EN 206-1 for the specification of concrete mixes to ensure durability
in various exposure conditions, which are also defined in the same document. In the a
complementary standard, BS 8500, has been developed. Part 1 of this British standard
is written to assist anyone wishing to specify concrete to BS EN 206-1. Part 2 of this
complementary standard contains specification for materials and procedures that are outside
of European standardization but within national experience. This part supplements the
requirements of BS EN 206 -1.
ENV 13670, Execution of Concrete Structures, deals with workmanship aspects. It is due to
be converted into an EN.
In addition to the above, a number of product standards have been developed for
particular products - hollow core units, ribbed elements, etc. While the product standards
rely on Eurocode 2 for design matters, some supplementary design rules have been given in
some of them.
the language and employed will be familiar to most engineers. They
do not differ from the ENY. EN 1990 defines a number which are to all
CHAPTER I. INTRODUCTION
materials (actions, types of analysis, etc.). In addition, EN 1992-1-1 also defines particular
terms specifically applicable to Eurocode 2 A few commonly occurring features are noted
below:
"
"
Loads are generally referred to as 'actions' to characterize the generalized format of the
code. Actions refer not only to the forces directly applied to the structure but also to
imposed deformations, such as temperature effect or settlement. These are referred to
as 'indirect actions', and the subscript 'IND' is used to identify this.
Actions are further subdivided as permanent (G) (dead loads), variable (Q) (live loads
or wind loads) and accidental (A). Prestressing (P) is treated as a permanent action in
most situations.
Action effects are the internal forces, stresses, bending moments, shear and deformations
caused by actions.
Characteristic values of any parameter are distinguished by the subscript 'k'. Design
values carry the subscript 'd', and take into account partial safety factors.
The term 'normative' is used for the text of the standards that forms the requirements.
The term 'informative' is used only in relation to annexes, which seek to inform rather
than require.
All the formulae and expressions in the code are in terms of the cylinder strength of
concrete. It is denoted by Jek' The relationship between the cylinder and cube strengths is
set out within the code. The strength class refers to both cylinder and cube strengths, e.g.
C25/30 refers to a cylinder strength of 25 N/mm2 and a corresponding cube strength of
30 N/mm2. It is not anticipated that the quality control procedures for the production of
concrete will switch to cylinder strength where the cube strength is now used (e.g. as in
the UK).
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
CHAPTER 2
Basis of design
I. Notation
In this manual symbols have been defined locally where they occur. However, the following
is a list of symbols, which occur throughout the document.
Gk,j
Gk,inf
Gk,sup
Qk,i
Ak
P
k
IG,j
IGA,j
IQ,i
Ip
characteristic value of the permanent actionj
lower characteristic value of a permanent action
upper characteristic value of a permanent action
characteristic value of the variable action i
characteristic value of an accidental action
characteristic value of prestressing force
partial safety factor for permanent actionj for persistent and transient design
situations
partial safety factor for permanent actionj for accidental design situations
partial safety factor for variable action i
partial safety factor for prestressing force
1jJ 0' 1jJ l' 1jJ 2
multipliers for the characteristic values of variable actions to produce
combination, frequent and quasi-permanent values, respectively, of variable
actions to be used in various verifications
characteristic value of a material property
partial safety factor for the material property, including model uncertainty.
General
All Eurocodes rely on EN 1990 for the basis of structural design. The provisions of that code
are applicable to an materials, and as such only the requirements which are independent of
material properties are noted. In the main it provides partial safety factors for actions,
including the values that should be used in a load combination.
Al! Eurocodes are drafted using limit state principles.
A brief resume of the main requirements of EN 1990 as they affect common designs in
concrete is noted below. For a fuller treatment of the subject, reference should be made to
Designers' Guide to EN 1990 in this series of guides to Eurocodes.
Fundamental requirements
Four basic requirements can be summarized. The structure should be designed and executed
in such a way that it will:
(1) during its intended life, with <c>nnr,w,,,.,11-p degrees of reliability and in an economical way
6
sustain all actions and influences that are likely to occur
remain fit for the intended use
(2) have adequate mechanical resistance, serviceability and durability
execution and use
(3) in the event of fire, have adequate resistance for the required period of fire exposure
(4) not be damaged by accidents (e.g. explosion, impact and consequences of human error)
to an extent disproportionate to the original cause.
According to EN 1990, a design using the partial factors given in its AnnexAl (for actions)
and those stated in the material design codes (e.g. Eurocode 2), is likely to lead to a structure
with a reliability index (3 greater than the target value of 3.8 stated in the code for a 50 year
reference period (see the note to Table B2 of EN 1990).
limit states
Limit states are defined as states beyond which the structure infringes an agreed performance
criterion. Two basic groups of limit states to be considered are (1) ultimate limit states and
(2) serviceability limit states.
Ultimate limit states are those associated with collapse or failure, and generally govern the
strength of the structure or components. They also include loss of equilibrium or stability
of the structure as a whole. As the structure will undergo severe deformation prior to
reaching collapse conditions (e.g. beams becoming catenaries), for simplicity these states are
also regarded as ultimate limit states, although this condition is between serviceability
and ultimate limit states; these states are equivalent to collapse, as they will necessitate
replacement of the structure or element.
Serviceability limit states generally correspond to conditions of the structure in use. They
include deformation, cracking and vibration which:
(1) damage the structure or non-structural elements (finishes, partitions, etc.) or the contents
of buildings (such as machinery)
(2) cause discomfort to the occupants of buildings
(3) affect adversely appearance, durability or water and weather tightness.
They will generally govern the stiffness of the structure and the detailing of reinforcement
within it.
Figure 2.1 illustrates a typical load-deformation relationship of reinforced-concrete
structures and the limit states.
2.S. Actions
2.5.1. Classifications
An action is a direct force (load) applied to a structure or an imposed deformation, such as
settlement or temperature effects. The latter is referred to as an indirect action. Accidental
actions are caused by unintended events which generally are of short duration and which
have a very low probability of occurrence.
The main classification of actions for common design is given in Table 2.1.
of .:U.l.lll'lI
Loads vary in time and space. In limit state design, the effects of loads, which are factored
suitably, are compared with the resistance of the structure, which is calculated by suitably
discounting the material properties. In theory, characteristic values are obtained statistically
from existing data. In practice, however, this is very rarely possible, particularly for imposed
loads whose nominal values, often specified by the are used as characteristic loads. In
countries where wind and snow data have been gathered over a period, it will be possible to
prescribe a statistically estimated characteristic value.
CHAPTER 2. BASIS OF DESIGN
u
co
o
..J
Phase 1
Uncracked
Ultimate load
Cracking load
Phase 2
Cracked
Phase 3
Inelastic phase
Phases 1 and 2: serviceability limit states
Phase 3: ultimate limit state
Deformation
2.1. Typical load relationship of reinforced concrete structures and the limit states
Table 2.1. Classification of actions
Permanent action
(a) Self-weight of structures,
fittings and fixed equipment
(b) Prestressing force
(c) Water and earth loads
(d) Indirect action, e.g. settlement
of supports
Variable action
(a) Imposed floor loads
(b) Snow loads
(c) Wind loads
(d) Indirect action, e.g. temperature
effects
Accidental action
(a) Explosions
(b) Fire
(c) Impact from vehicles
Characteristic values for loads are given EN 1991(Eurocode 1: Actions on Structures).
For permanent actions, which vary very little about their mean value (such as weights of
materials), the characteristic value corresponds to the mean value. When the variation is
likely to be large, e.g. walls or slabs cast against an earth surface with random variations in
thickness, or loads imposed by soil fill, upper and lower characteristic values (commonly
corresponding to 95tb and 5th percentiles) will need to be assessed. These values will apply
during the life of the structure. These are denoted by Ok, sup and Ok, in! respectively.
values actions
The values of actions to be used in design are governed by a number of factors. These
include:
(1) The nature of the load. Whether the action is permanent, variable or accidental, as the
confidence in the description of each will vary.
7
8
(2) The limit state being considered. Clearly, the value of an action governing design must
be higher for the ultimate limit state than for serviceability for persistent and transient
design situations. Further, under serviceability conditions, loads vary with time, and the
design load to be considered could vary substantially. Realistic serviceability loads
should be modelled appropriate to the aspect of the behaviour being checked (e.g.
deflection, cracking or settlement). For example, creep and settlement are functions of
permanent loads only.
(3) The number of variable loads acting simultaneously. Statistically, it is improbable that
all loads will act at their full characteristic value at the same time. To allow for this, the
characteristic values of actions will need modification.
Consider the case of permanent action (G
k
) and one variable action (Qk) only. For the
ultimate limit state the characteristic values should be magnified, and the load may be
represented as 1GG
k
+ 10Qk, where the 1 factors are the partial safety factors. The values of
IG and 10 will be different, and will be a reflection of the variabilities of the two loads being
different. The 1 factors account for:
(1) the possibility of unfavourable deviation of the loads from the characteristic values
(2) inaccuracies in the analyses
(3) unforeseen redistribution of stress
(4) variations in the geometry of the structure and its elements, as this affects the determination
of the action effects.
Now consider the case of a structure subject to variable actions Ql and Q2 simultaneously.
If Ql and Q
2
are independent, i.e. the occurrence and magnitude of Q
1
does not depend on
the occurrence and magnitude of Qz and vice versa, then it would be unrealistic to use
10, 1Qk, 1 + 10, ZQk, 2 as the two loads are unlikely to act at their maximum at the same time.
Joint probabilities will need to be considered to ensure that the probability of occurrence of
the two loads is the same as that of a single load. It will be more reasonable to consider one
load at its maximum in conjunction with a reduced value for the other load. Thus, we have
two possibilities:
10, lQk, 1 + 'l/Jo. z( 10, 2Qk, 2)
or
'1)0, lCrO.1Qk, 1) + 10,zQk,z
Multiplication by 'l/Jo is said to produce a combination value of the load. It should be noted
that the values of 1 and 'l/Jo vary with each load. EN 1990 provides recommended values, and
the UK National Annex will stipulate the values to be used in UK. See also Table 2.3 below.
The method of deriving 'l/J values is outlined in the addenda to ISO 2394: 1986. In practice, the
designer will not have sufficient information to vary the 'l/J values in most cases. Table 2.4,
below, summarizes the 'l/J values recommended by the UK National Annex.
The above discussion illustrates the thinking behind the method of combining loads for an
ultimate limit state check. Similar logic is applied to the estimation of loads for the different
serviceability checks.
Ultimate limit state
(1) Persistent and transient situations fundamental combinations. In the following
paragraphs, various generalized combinations of loads are expressed symbolically. It
should be noted that the '+' symbol in the expressions does not have the normal
mathematical meaning, as the directions of loads could be different. It is best to read it
as meaning 'combined with'.
EN 1990 gives three separate sets of load combinations, namely EQU (to check
against loss of equilibrium), STR (internal failure of the structure governed by the
strength of the construction materials) and GEO (failure of the ground, where the
strength of soil provides the significant resistance).
Equilibrium. Equilibrium is verified using the load combination Set A in the code, which
is as follows:
supGk,j, sup + IG,j, infGk,j, inf + IQ, lQk, 1 + IQ, ;1/;0, ;Qk,;
IG,j, supGk,j, sup is used when the permanent loads are unfavourable, and IG,j, infGk,j, inf is
used when the permanent actions are favourable. Numerically, IG,j, sup = 1.1, IG,j, inf = 0.9,
and IQ = 1.5 when unfavourable and when favourable.
The above format applies to the verification of the structure as a rigid body (e.g.
overturning of retaining walls). A separate verification of the limit state of rupture of
structural elements should normally be undertaken using the format given below for
strength. In cases where the verification of equilibrium also involves the resistance of the
structural member (e.g. overhanging cantilevers), the strength verification given below
without the above equilibrium check may be adopted. In such verifications, IG,j, in! = 1.15
should be used.
Strength. When a design does not involve geotechnical actions, the strength of elements
should be verified using load combination Set B. Two options are given. Either combination
(6.10) from EN 1990 or the less favourable of equations (6.10a) and (6.10b) maybe used:
(6.10)
IG,j, supGk,j, sup is used when the permanent loads are unfavourable, and IG,j, infGk,j, inf is
used when the permanent actions are favourable. Numerically, IG,j, sup = 1.35, IG,j, inf = 1.0,
and IQ = 1.5 when unfavourable and Owhen favourable (EN 1990, UK National Annex).
(6.lOa)
(6.lOb)
Numerically, ~ = 0.925, IG,j, sup = 1.35, IG,j, in! = 1.0, and IQ 1.5 when unfavourable and
when favourable (EN 1990, UK National Annex).
The above combinations assume that a number of variable actions are present at the
same time. Qk, 1 is the dominant load if it is obvious, otherwise each load is in turn treated
as a dominant load and the others as secondary. The dominant load is then combined
with the combination value of the secondary loads. Both are multiplied by their respective
I values.
The magnitude of the load resulting from equations (6.lOa) and (6.10b) will always be
less than that from equation (6.10). The size of the reduction will depend on the ratio
X = Gk/(G
k
+ Qk)' Table 2.2 gives the reduction factors for different values of X.
Now turning to the factors IG, in! and IG, sup' it will be noted that the numerical values
are different in the verification of equilibrium and that of strength. For instance, in an
overhanging cantilever beam, the multiplier for self-weight in the cantilever section will
be 1.1 (IG, sup) and that in the anchor span will be 0.9 (IG, cnf)' The possible explanation for
I G, sup being 1.1 and not 1.35 as in the strength check is that
(a) the variability in self-weight of the element is unlikely to be large
(b) the factor 1.35 has built into it an allowance for structural performance (which is
necessary only for strength checks)
(c) the loading in the cantilever will also generally include variable actions, partial
safety factors for which will ensure a reasonable overall safety factor.
When a design involves geotechnical action, a number of approaches are given in
EN 1990, and the choice of the method is a Nationally Determined Parameter. In the
UK, it will generally be necessary to carry out two calculations using the load
9
10
combinations Set C and Set B and the resistances given in EN 1997. For details,
reference should be made to EN 1990 and EN 1997.
(2) Accidental design situation. The load combination recommended is
Gk,j,sup + Gk,j,inf+Ad + 1/J),;Qk,) + ;Qk,;
where Ad is the design value of accidental action, Qk,l is the main variable action
accompanying the accidental action and Qk, i are other variable actions.
Accidents are unintended events such as explosions, fire or vehicular impact, which
are of short duration and which have a low probability of occurrence. Also, a degree of
damage is generally acceptable in the event of an accident. The loading model should
attempt to describe the magnitude of other variable loads which are likely to occur in
conjunction with the accidental load. Accidents generally occur in structures in use.
Therefore, the values of variable actions will be less than those used for the fundamental
combination of loads in (1) above. To provide a realistic variable load combining with
the accidental load, the variable actions are multiplied by different (and generally lower)
1/J factors. Multiplier 1/J] is applied to the dominant action, and to the others, Where
the dominant action is not obvious, each variable action present is in turn treated as
dominant. 10 for accidental situations is unity.
Multiplication by 1/Jl is said to produce a frequent value of the load, and multiplication
by 1/J2 the quasi-permanent value. Numerical values forlj;l and are given in EN 1990,
and confirmed in the UK National Annex. Table 2.4 below summarizes the values.
Serviceability limit state
(3) Characteristic combination.
lGk,j(+P)+Qk,l + l1/JO,iQk,i
i>l
This represents a combination of service loads, which can be considered rather
infrequent. It might be appropriate for checking states such as micro cracking or
possible local non-catastrophic failure of reinforcement leading to large cracks in
sections,
Table 2.2. Reduction factors for different values of X
Load from equation (6. lOa) Load from equation (6.1 Ob)
---_._'._._--- -----------
X
Load from equation (6.10) Load from equation (6, 10)
0,0 0.70 1.00
0,1 0.73 0.99
0.2 0.76 0.99
0.3 0.78 0.98
0.4 0.81 0.117
0,5 0.84 0.96
0,6 0.87 0.96
0.7 0.90 0.95
0,8 0.93 0.94
0.9 0.97 0.93
1.0 1.00 0.93
(I) The reduction factors shown in bold should be used,
(2) The table assumes lG == 1.35, lQ 1,5, '1/)0 == 0.7 and ~ == 0,925.
Table 2.3. Partial safety factors for actions in building structures ultimate limit state (in
accordance with the UK National Annex)
Action Combination
Fundamental
Permanent actions caused by structural
and non-structural components
Stability check
Unfavourable 1.10
Favourable 0.90
Other checks
Unfavourable 1.35
Favourable 1.00
Variable actions
Unfavourable 1.50
Accidental actions
( I) The above apply to persistent and transient design situations.
(2) For accidental design situations {G. A = 1.0.
(3) Partial safety factor for the prestressing force (rp) is generally 1.0.
Accidental
1.00
1.00
1.00
1.00
1.00
1.00
(4) For imposed deformation, {Q for unfavourable effects is 1.2, when linear methods are used. For non-
linear methods, {Q = 1.5.
Table 2.4. 1/) values
Variable actions 'ljJo 'ljJ1 'ljJ2
Imposed loads
Dwellings 0.7 0.5 0.3
Offices 0.7 0.5 0.3
Shopping and congregation areas 0.7 0.7 0.6
Storage 1.0 0.9 0.8
Parking 0.7 0.7 0.6
Wind loads 0.5 0.2 0.0
Snow loads (for altitudes ,:S; 1000 m) 0.7 0.2 0.0
For the purposes of Eurocode 2, the three categories of variable actions in the table should be treated as
separate and independent actions.
( 4) Frequent combination.
IPkj+P) + 1/)1, lQk, 1 + L7f!2,iQk,i i> 1
This represents a combination that is likely to occur relatively frequently in service
conditions, and is used for checking cracking.
(5) Quasi-permanent combination.
LGk./+P) + L7f!2,i
Q
k,i i:2: 1
This will provide an estimate of sustained loads on the structure, and will be appropriate
for the verification of creep, settlement, etc.
H should be realized that the above combinations describe the magnitude of loads which
are likely to be present simultaneously. The actual arrangement of loads in position and
direction within the structure to create the most critical effect is a matter of structural
analysis (e.g. loading alternate or adjacent spans in continuous beams).
Values of I and 1/) applicable in the UK are given in Tables 2.3 and 2.4.
II
12
2.1-2.4 illustrate the use of the combinations noted above. in practice
simplified methods given in Section 2.5.4 below are likely to be all that is needed for the
majority of structures. Also, in practical examples the dominant loads are likely to be fairly
obvious, and therefore the designer will generally not be required to go through all the
combinations.
2.5.4. Simplified load combinations
Unlike the ENV version of EN 1992-1-1, EN 1990 does not have simplified load combinations.
For normal building structures, expressions (6.10), (6. lOa) and (6.10b) for the ultimate limit
state may also be represented in a tabular form (Table 2.5).
Material properties
2.6. I. Characteristic values
A material property is represented by a characteristic value X
k
, which in general corresponds
to a fractile (commonly 5%) in the statistical distribution of the property, i.e. it is the value
below which the chosen percentage of all test results are expected to fall.
Generally, in design, only one (lower) characteristic value will be of interest. However, in
some problems such as cracking in concrete, an upper characteristic value may be required,
i.e. the value of the property (such as the tensile strength of concrete) above which only a
chosen percentage to the values are expected to fall.
2.6.2. Design values
In order to account for the differences between the strength of test specimens of the
structural materials and their strength in situ, the strength properties will need to be reduced.
This is achieved by dividing the characteristic values by partial safety factors for materials
('M)' Thus the design value Xd =XkhM" Uncertainties in the resistance models are also
Table 2.5. Partial safety factors for load combinations in EN 1990 - ultimate limit state
Permanent load
Load combination Adverse Beneficial
load combination (6. I 0)
Permanent + imposed 1.35 1.00
Permanent + wind 1.35 1.00
Permanent + imposed + 1.35 1.00
wind
load combination (6.1 Oa)
Permanent + imposed 1.35 1.0
Permanent + wind 1.35 1.0
Permanent + imposed + 1.35 1.0
wind
load combination (6.1 Ob)
Permanent + imposed e 1.35 = 1.25 ~ 1.0 = 0.925
Permanent + wind e 1.35 = 1.25 ~ 1.0=0.925
Permanent + imposed + e 1.35 = 1.25 ~ 1.0 = 0.925
wind
(I) It is assumed that wind is not the leading action.
(2) 7 / J ~ will vary with the use of the building.
Variable load
Imposed
Adverse Beneficial Wind
1.50 0
1.50
1.50 0 1.50 X 0.5
=0.75
7 / J ~ I.S
1/)0 1.5 = 0.9
7 / J ~ 1.5 7/J
o
1.5 = 0.9
1.5
- 1.5
1.5 7/Jo 1.5 = 0.9
Prestress
1.00
1.00
1.00
1.0
1.0
1.0
1.0
1.0
1.0
Table 2.6. Partial factors for material
Combination
Fundamental
Accidental except fire situations
Accidental - fire situations
Concrete, IC
1.5
1.2
1.0
Reinforcement and
prestressing tendons, IS
1.15
1.0
1.0
* See also UK National Annexes for EN 1992-1-1 and EN 1992-1-2, and the background paper.
(I) These factors apply to the ultimate limit state. For serviceability, 1M = I.
(2) These factors apply if the quality control procedures stipulated in the code are followed. Different
values of IC may be used if justified by commensurate control procedures.
(3) These factors do not apply to fatigue verification.
(4) IS = 1.15 should be applied to the characteristic strength of 500 MPa.
covered by 1M' Although not stated in the code, 1M also accounts for local weaknesses and
inaccuracies in the assessment of resistance of the section.
The values of partial safety factors for material properties are shown in Table 2.6.
2.1. Geometric data
The structure is normally described using the nominal values for the geometrical parameters.
Variability of these is generally negligible compared with the variability associated with the
values of actions and material properties.
In special problems such as buckling and global analyses, geometrical imperfections
should be taken into account. The code specifies values for these in the relevant sections.
Traditionally, geometrical parameters are modified by factors which are additive.
Verification
(1) When considering overall stability, it should be verified that the design effects of
destabilizing actions are less than the design effects of stabilizing actions.
(2) When considering rupture or excessive deformation of a section, member or connection,
it should be verified that the design value of internal force or moment is less than the
design value of resistance.
(3) It should be ensured that the structure is not transformed into a mechanism unless
actions exceed their design values.
(4) It should be verified that the design effects of actions do not exceed a nominal value or a
function of certain design properties of materials; for example, deflection under quasi-
permanent loads should be less than span/2S0, and compression stress under a rare
combination of loads should not exceed O.6fck'
(5) In most cases, detailed calculations using various load combinations are unnecessary, as
the code stipulates simple compliance rules.
Durability
As one of the fundamental aims of design is to produce a durable structure, a number of
matters will need to be considered early in the design process. These include:
13
14
@ the use of the structure
.. the required performance criteria
.. the expected environmental conditions
.. the composition, properties and performance of the materials
.. the shape of members and the structural detailing
.. the quality of workmanship, and level of control
.. the particular protective measures
e the maintenance during the intended life.
The environmental conditions should be considered at the design stage to assess their
significance in relation to durability and to enable adequate provisions to be made for
protection of the materials.
overall
lOentltv the various load
Assume office use for this
load combinations for the of elements could be
characteristic dead
characteristic dead
characteristic live
characteristic live
characteristic wind
The fundamental load combination to be used is
i> 1
of the actions is to be when
a distinction between the favourable and unfavourable effects needs
to be made. The values for the different accordance with the UK
Table A 1 Set are as follows:
0.9
1.1
=0.7
Case I
Treat the wind load as the dominant load
Case 2
Treat the mrpm;eu load on roof as dominant load
Case 3
Treat the ImlDm,ed on floors as the dominant load
Note
1"1"'.7(-'1""",1 another set of combinations will need to be considered.
the combinations
2.2. Frame (f'"',,min!p 2.1)
2.3. Frame '-A,' .... U!e 2.1, case I)
1,1G'A +
O,5(15Q
kR
l
+
O,5(1.5W
k
} _,!z::;tzL;;::Z::Z:UI====1
2.4. Frame i-l<:;mnlp 2.1, case 2)
2.5. Frame
15
16
2.6. continuous beam
Example
'''',,,'''h1 the various load combinations for the of a continuous beam for
the ultimate limit state Assume that spans 1-2 and 2-3 are to domestic
to use.
Notation:
characteristic '''''"'',.,o<>rI
characteristic ImlPos;eo
The fundamental load combination to be used 1S
+ i> 1
The same value of may be to all spans, i.e.
clause as it will worse effects than <In''''''ItHT
The load cases to be considered are:
.. alternative spans loaded
.. spans loaded
Example 2.3
1992-1-1,
Set
For the continuous beam shown in the critical load combinations for the
for the ultimate limit state. Assume that the beam is to dead and uu,",velv,",
loads and a load at the end of the cantilever from dead loads of the external
wall.
Notation:
characteristic HHitJVO'''U
P characteristic
Fundamental combinations
used.
in Table
Example 2.4
A water tank of
H m has an
lateral loads for the ultimate limit state.
EN 1991-4 Part 4: Silos and
the be noted:
of EN 1990 should be
ofwaterh m Calculate the
should be used for dr>'rpr,rYll1",'
.. The actions to values that have a of
that will be exceeded in a reference of 1 year. In this it is assumed
that the h of water has been determined on this basis.
A ~ _
B_
Maximum positive moment in 1-2 and maximum moment in column 1
c_
Maximum positive moment in 3-4 and maximum moment in columns 3 and 4
D_
Maximum positive moment in 2-3 and maximum moment in columns 2 and 3
E_
Maximum positive moment in 4-5 and maximum moment in column 5
F_
Maximum moment at support 2
G_
H_
Maximum moment al support 3
2.7. load cases (b:amiDle
17
..
18
accidental situations.
two are shown in 2.14 and 2.15.
2.8. Continuous beam
Maximum cantilever 13M, maximum anchorage oj negative steel over 3;
also maximum column moment at 3 (see Fig. 2.12)
2.9. Continuous beam .... A ' ~ I " U " : : : 2.3, case I)
Maximum negative moment at column 2
2.10. Continuous beam case 2)
Maximum positive 1-2, column moment column 1 and
possible maximum moment column 2 (see Fig. 2.12)
2.1 I. Continuous beam case 3)
a!2 Fig.2.i1)
2.12.
;--- ~
i
~
2.13. Water tank
p density of water
L2ph
2.14. Pressure
1.0pH
p density water
2.15. Alternative - deformation
19
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
J
I
3
nalysis
l.l. Introduction
The purpose of analysis is the verification of overall stability and establishment of action
effects, i.e. the distribution of internal forces and moments. In turn, this will enable the
calculation of stresses, strains, curvature, rotation and displacements. In certain complex
structures the type of analysis used (e.g. finite-element analysis) will yield internal stresses
and strains and displacements directly.
To carry out the analysis, both the geometry and the behaviour of the structure will need to
be idealized.
Commonly, the structure is idealized by considering it as made up of elements depicted in
Fig. 3.1.
In terms of the behaviour of the structure, the following methods may be used:
e elastic analysis
elastic analysis with limited redistribution
plastic analysis
non-linear analysis.
The first two models are common for slabs and frames, and plastic analysis is popular in the
design of slabs; non-linear analysis is very rarely used in day-to-day design. The above methods,
with the exception of plastic analysis, are suitable for both serviceability and ultimate limit states.
Plastic methods can be used generally only for ultimate limit state (i.e. strength) design.
In addition to global analysis, local analyses may also be necessary, particularly when the
assumption of linear strain distribution does not apply. Examples of this include:
anchorage zones
e members with significant changes in cross-section, including the vicinity of large holes
beam-column junctions
locations adjacent to concentrated loads.
In these cases strut and tie models (a plastic method) are commonly employed to analyse
the structures.
Load cases and combination
In the analysis of the structure, the designer should consider the effects of the realistic
combinations of permanent and variable actions. Within each set of combinations (e.g. dead
and imposed loads) a number of different arrangements of loads (load cases) throughout the
structure (e.g. alternate spans loaded and adjacent spans loaded) will need consideration to
identify an envelope of action effects (e.g. bending moment and shear envelopes) to be used
in the design of sections.

22
(d)
(e)
3.1. Definition of structural elements for analysis. (a) Beam. (b)
way spanning slab (subject predominantly to ultimate design load)

I Freeedges
beam. (c) Slab. (d, e) One-
CHAPTER 3. ANALYSIS
Minimum 8/10, nol less than 50 mm
Section
(f)
Overall depth h
A-A
A
(9)
~ ~ - - - - - - - - - - - - - - - - - - - - - - - - - - ~
(h)
3.1. (Contd.) (f) Ribbed and waffle slabs (conditions to be met to allow analysis as solid slabs).
(g) Column. (h) Wall
23
DESIGNERS' GUIDE TO EN 1992- -I AND EN 1992-1-2
Clause 5.1.3
Clause 5.1.1
24
As stated in Chapter 2, EN 1990 provides the magnitude of the design loads to be used
when loads are combined. Account is taken of the probability of loads acting together, and
values are specifed accordingly.
The EN code for actions (Eurocode 1) specifies the densities of materials (to enable the
calculation of permanent actions and surcharges), and values of variable action (such as
imposed gravity, wind and snow loads). It also provides information for estimating fire loads
in buildings, to enable fire engineering calculations to be carried out.
Although EN 1992-1-1 forms part of a suite of codes including those which specify loads,
there is no reason why the Eurocode cannot be used in conjunction with other loading codes.
It has been assumed in the Eurocode system that the loads specified in Eurocode 1 are
characteristic values with only 5% of values likely to fall above them. Note that for wind loads
it is 2%. The definition of the characteristic value will affect the overall reliability.
While the general requirement is that all relevant load cases should be investigated to
arrive at the critical conditions for the design of all sections, EN 1992-1-1 permits simplified
load arrangements for the design of continuous beams and slabs. The arrangements to be
considered are:
(1) alternate spans loaded with the design variable and permanent loads (1.35G
k
+ 1.5Qk)
and other spans carrying only the design permanent load (1.35G
k
)
(2) any two adjacent spans carrying the design variable and permanent loads (1.35G
k
+ 1.5Qk)'
with an other spans carrying only the design permanent load (1.35G
k
).
Although not stated, the above arrangements are intended for braced non-sway structures.
They may also be used in the case of sway structures, but the following additional load cases
involving the total frame will also need to be considered:
(1) all spans loaded with the design permanent loads (1.35G
k
) and the frame subjected to
the design wind load (1.5W
k
), when W
k
is the characteristic wind load
(2) all spans at all floor levels loaded with (1.35G
k
+ 1.5Qk) and the frame subjected to the
design wind load of 1.05W
k
(3) in sensitive structures (sensitivity to lateral deformation), it may be necessary to
consider the effects of wind loading in conjunction with patterned imposed loading
through out the frame.
Clause 5.1.3 of EN 1992-1-1 also allows the National Annexes to specify simplification of
load arrangements, and the UK National Annex permits the following additional choices.
e For frames:
all spans loaded with the maximum design ultimate load (1.35G
k
+ 1.5Qk)
alternate spans loaded with the maximum ultimate load noted above and all other
spans loaded with 1.35G
k

e For slabs only: a single load case of maximum design load on all spans or panels provided
the following conditions are met:
in a one-way spanning slab the area of each bay exceeds 30 m
2
(here a bay means a
strip across the full width of a structure bounded on the other two sides by lines of
supports)
the ratio of the characteristic variable load to the characteristic permanent load
does not exceed 1.25
the characteristic variable load does not exceed 5 kN/m
2

The resulting support moments (except those at the supports of cantilevers) should be
reduced by 20%, and the span moments adjusted upwards accordingly. No further
redistribution should be carried out.
Clause 5.1.1(8) of EN 1992-1-1 states that in linear elements and slabs subject predominantly
to bending, the effects of shear and axial forces on deformation may be neglected, if these are
likely to be less than 10%. In practice, the designer need not actually calculate these
additional deformations to carry out this check.
CHAPTER 3. ANALYSIS
Vc1 Vc2 Ven
-:aI'--- ..,..,..+...-f ____ ::..
c
----......J- 4.He
. ~ ________ ::..b________ ~ _ 4 . ~
.'1-___
F
:".:lo;;,;o;,:"r,;;;.8___ -'_ 4.H.
3.2. Application of the effective geometrical imperfections: braced structure (number of
vertically continuous members = 2)
Deflections are generally of concern only in members with reasonably long spans.
In such members, the contribution of shear to the deflections is never significant for
members with normal (span/depth) ratios. When the spans are short, EN 1992-1-1 provides
alternative design models (e.g. truss or strut and tie) in which deflections are rarely, if ever, a
consideration.
The contribution of axial loads to deflections may be neglected if the axial stresses do not
exceed 0.08fck'
Imperfections
I. General
Perfection in buildings exists only in theory; in practice, some degree of imperfection is
unavoidable, and designs should recognize this, and ensure that buildings are sufficiently
robust to withstand the consequences of such inaccuracies. For example, load-bearing
elements may be out of plumb or the dimensional inaccuracies may cause eccentric application
of loads. Most codes allow for these by prescribing a notional check for lateral stability. The
exact approach adopted to achieve this differs between codes.
EN 1992-1-1 has a number of provisions in this regard, affecting the design of (1) the
structure as a whole, (2) some slender elements and (3) elements which transfer forces to
bracing members.
3.3.2. Global analysis
For the analysis of the structure as a whole, an arbitrary inclination of the structure eo = 1/200
is prescribed as a basic value. This is then modified for height and for the number of
members involved.
The design value will be
where
an == 2/-vl
where l is the total height of the structure in metres (0.67 :s; an :s; 1.0), and
am = -V[O.5(1 + 11m)]
where m is the number of vertically continuous elements in the storeys contributing to the
total horizontal force on the floor. This factor recognizes that the degree of imperfection is
statistically unlikely to be the same in aU the members.
As a result of the inclination, a horizontal component of the vertical loads could be
thought of at each floor level, as shown in Figs 3.2 and 3.3. These horizontal
25
Clause
26
!l.Hj as Fig. 3.2
Fig. 3.3. Application of the effective geometrical imperfections: unbraced structure (number of
vertically continuous members = 3)
3.4. Minimum tie force for perimeter columns
forces should be taken into account in the stability calculation. This is in addition to other
design horizontal actions, such as wind.
3.3.3. Design of slender elements
In the design of slender elements, which are prone to fail by buckling (e.g. slender columns),
EN 1992-1-1 requires geometrical imperfection to be added to other eccentricities. For
example, in the design of the columns, an eccentricity of 8/
0
/2 is assumed for geometrical
imperfection (where lo is the effective length of the column).
3.3.4. Members transferring forces to bracing ""I., ..... ", .. , , ~ ' "
In the design of these elements (such as a floor diagram), a force to account for the possible
imperfection should be taken into account in addition to other design actions. This
additional force is illustrated in Fig. 3.4. This force need not be taken into account in the
design of the bracing element itself.
Second-order effects
As structures subject to lateral loads deflect, the vertical loads acting on the structure
produce additional forces and moments. These are normally referred to as second-order
effects. Consider a cantilever column shown in 3.5. The deflection caused by the
horizontal load alone is .1
1
' In this deflected state, the verticalloadPwill contribute a further
bending moment and increase the lateral sway, and the final deflection will be .1
2
' This
phenomenon is also commonly referred to as the 'P L1 effect'.
EN 1992-1-1 requires second-order effects to be considered where they may significantly
affect the stability of structure as a whole or the attainment of the ultimate limit state at
critical sections.
In the Application Rules, the code further states that for normal buildings, second-order
effects may be neglected if the bending moments caused by them do not increase the first-
CHAPTER 3. ANALYSIS
p
H_
/
Fig. 3.S. Second-order effect
/
p
!
/"
/
order bending moments (i.e. bending moments calculated ignoring the effects of displacements)
by more than 10%.
Although this would suggest that the designer would first have to check the second-order
bending moment before ignoring it, the code provides some simplified criteria, to verify if
second-order analysis is required. These tests are summarized in clause 5.8.3.1 for isolated
elements and clause 5.8.3.3 for structures. These checks essentially ensure that adequate
stiffness is provided.
In most practical building structures, second-order effects are unlikely to be significant as
serviceability criterion to limit the lateral deflections will ensure that structures are not
prone to P Ll effects.
3.S. Timemdependent effects
The main effects to be considered are creep and shrinkage of concrete and relaxation of
prestressing steel. EN 1992-1-1 provides information in clause 7.2.2, when such effects need
not be considered explicitly.
3.6. Design by testing
EN 1992-1-1 itself does not provide any useable guidance on this subject. However, there is
some guidance in EN 1990. According to EN 1990, testing may be carried out in the
following circumstances:
if adequate calculation models are not available
if a large number of similar components are to be produced
to confirm by control checks assumptions made in design.
Annex D of EN 1990 provides further information, particularly regarding the statistics.
Design entirely based on testing is not common in building structures. However, this is
an accepted method in some other fields (e.g. pipes). In such work, test programmes
should be designed in such a way that an appropriate design strength can be established,
which includes proper allowance for the uncertainties covered by partial safety factors in
conventional design. It will be necessary to establish the influence of material
Clause 5.8.3.1
Clause 5.8.3.3
Clause 7.2.2
27
Gause 5.4(2)
Gause 5.4(3)
Gause 5.3.2.2
28
on behaviour and their variability so that a characteristic hence design)
response can be derived. When testing is carried out on elements which are smaller than
the prototype, size effects should be considered in the interpretation of results (e.g. shear
strength).
Testing may also be undertaken for other reasons, including:
appraisal of an existing structure
* to establish data for use in design
to verify consistency of manufacture or performance of components.
Test methods and procedures will obviously be different in each case. Before planning a test,
the precise nature of the information required from the test together with criteria for judging
the test should be specified.
Structural analysis
I. analysis with or without
General
EN 1992-1-1 provides limited guidance on analysis. Elastic analysis remains the most
popular method for frame (e.g. moment distribution and slope deflection).
Braced frames may be analysed as a whole frame or may be partitioned into subframes
(Fig. 3.6). The subframes may consist of beams at one level with monolithic attachment to
the columns. The remote ends may be assumed to be 'fixed' unless a 'pinned' end is more
reasonable in particular cases. As a further simplification, beams alone can be considered
to be continuous over supports providing no restraint to rotation. Clearly this is more
conservative.
In unbraced structures, it is generally necessary to consider the whole structure, particularly
when lateral loads are involved. A simplified analysis may be carried out, assuming points of
contraflexure at the mid-lengths of beams and columns (Fig. 3.7). However, it should be
remembered that this method will be inaccurate if (1) the feet of the column are not fixed
and/or (2) the beams and columns are not of similar stiffnesses.
Stiffness parameters
In the calculation of the stiffness of members it is normally satisfactory to assume a 'mean'
value of the modulus of elasticity for concrete and the moment of inertia based on the
uncracked gross cross-section of the member. However, when computing the effects of
deformation, shrinkage and settlement reduced stiffness corresponding to cracked cross
section should be used.
Effective spans
Calculations are performed using effective spans, which are defined below. The principle is
to identify approximately the location of the line of reaction of the support.
leff = in + aj + a
2
where leff is the effective span, In is the distance between the faces of supports, and a
j
and a
2
are the distances from the faces of the supports to the line of the effective reaction at the two
ends of the member.
..
..
"
Typical conditions are considered below.
Case 1: intermediate support over which member is continuous (Fig. 3.8) .
Case 2: monolithic end support (Fig. 3.9). a
j
is the lesser of half the width of support or
half the overall depth of number.
Case 3: discontinuous end support 3.10). a
j
is the lesser of half the width of support
or third of the overall depth of member.
CHAPTER 3. ANALYSIS
3.6.
Fuilirama
r ----- ----- ----- -----f,----- ------,
L -----j----- --I---I------f----- f------W
h
.J
h
h\ /
Stillness halved
Beam only
Column only
!"';U'1",f"nnlno of multi-storeyed braced structures for analysis
Case 4: discontinuous end support on bearings 3.11). a
j
is the distance from the face
of the to the centre line of the bearing.
Case 5: isolated cantilever. a
j
= 0, Le. the effective span is the length of the cantilever
from the face of the
29
-
~
-
,r/
., ./
Total frame
t
1
~
1
I
1
=L

L

L
Ell
'" '"
l--
Subframe
3.1. Simplified model for the analysis of unbraced structures
Centre line
Fig. l.8. Intermediate support over which member is continuous
L
8
j
min{O.Sh; O.SI}
fig. 3.9. Monolithic end support
/.J<--- 8
j
= min{O.5h; O,SI}
3.10. Discontinuous end support
30
CHAPTER 3. ANALYSIS
V//V//1
!
-J
,
I
3.1 I. Discontinuous end support on bearings
(a)
beff = bw + 0.10\, < b
(b)
fig. l.12. Effective flange width: (a) T beam; (b) L beam. b is the beam spacing and b
l
is half the
dear spacing between beams
fig. 3.1 J. Definition of 10 for the calculation of the effective flange width
Effective width of flanges
As a result of 'shear lag', the stress in the parts of a wide flange distant from the web would be
much less than that at the flange-web junction. The calculation of the variation is a complex
mathematical problem. Therefore, codes of practice allow approximations by which an
'effective' width can calculated. A uniform distribution of stress is assumed over the effective
width.
For building structures the effective widths shown in 3.12 may be used.
lo in Fig. 3.13 is the distance between points of contraflexure.
Note that:
(1) The length of the cantilever should be less than half the adjacent span.
(2) The ratio of adjacent spans (other than the cantilever) should be between 0.67 and 1.5.
In analysis, EN 1992-1-1 permits the use of a constant flange width throughout the span.
This should take the value applicable to span sections. This applies only to elastic design: with
or without redistribution and not whcn more rigorous non-linear methods are adopted.
Redistribution
The moment-curvature response of a true material will be typically as shown
at 3.14. The after Mp is reached UH,IJU\/" a large rotation capacity. Consider a
Clause 5.3.2. f
31
DESIGNERS' GUiDE TO EN 1992-1-1 AND EN 1992-1-2
32
continuous beam made of such a material and loaded as shown in Fig. 3.1S(a). When a
bending moment in a critical section (usually at a support) reaches M
p
, a plastic hinge is said
to be formed. The structure will be able to withstand further increases in loading until
sufficient plastic hinges form to turn the structure into a mechanism.
The bending moments at 'failure' will be as shown in Fig. 3.1S(c). Had the beam been
elastic, then the bending moment will be as shown in 3.1S(b). Comparing these two
figures, it can be seen that the elastic bending moment at the support has been reduced from
(3/16)puL to (l/6)PuL, or redistributed by about 11 %. It should also be noted that the span
moment has increased by about 7%.
The process illustrated is plastic analysis. Clearly, to exploit plasticity fully, the material
must possess adequate ductility (rotation capacity). Concrete has only limited capacity in
this regard. The moment redistribution procedure is an allowance for the plastic behaviour
without carrying out plastic hinge analysis. Indirectly, it also ensures that the yield of sections
under service loads and large uncontrolled deflections are avoided.
EN 1992-1-1 allows moment redistribution up to 30% in braced structures. No
redistribution is permitted in sway frames or in situations where rotation capacity cannot be
defined with confidence.
1Ir
3.14. Moment-curvature: idealization
(a)
(b)
Mp
Mp= PuU6
(e)
l.IS. (a, b) 1'\"",riir"T moment of an elastic beam at failure. (c) Bending moment of an elasto-
beam at failure
CHAPTER 3. ANALYSIS
35
Limit for class Band C reinforcement
30
, :-..'
":: "--,
,., ......
, :-..'
"::
..... '" .....
25
, "
'\;:, ",
Limit for class A reinforcement
?i " , ....
20
'"5 .... " ..
: , ....
]l 15 ' ",',
ill --- fek 0; 50 MPa "\ "',
a: - - - - - - fcJ< = 55 MPa " -'::',
10
- - - fek 60 MPa "\, ""',
.... " ....
"\ ",',
fek = 70 MPa
- - fek = 80 MPa
5
, '
, ",',
" ",',
, '
, ",'
" <"
, '
.... " .....
L-____ ____ ____ -L ____ -L ____ ____ ____ L_ ____ ____
0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0.5 0.55 0,6
xld
3.16. Redistribution of moment and limiting x/d values
When redistribution is carried out, it is essential to maintain equilibrium between the
applied loads and the resulting distribution of bending moments. Thus, where support
moments are reduced, the moments in the adjacent spans will need to be increased to
maintain equilibrium for that particular arrangement of loads.
In EN 1992-1-1 the limit to the amount of redistribution is related to the ductility
characteristics of the reinforcement used:
class B or class C reinforcement - 30%
class A reinforcement - 20%.
In the case of continuous beams or slabs, subject predominantly to flexure and in which
the ratio of the lengths of the adjacent spans lie in the range 0.5 to 2.0, the code permits
redistribution to be carried out without the rotation being explicitly checked provided the
following expressions are satisfied (see the UK National Annex).
For ick :s:; 50 MPa:
8;:::: 0.4 + [0.6 + (0.0014/c
cu
2)]x/d
For iCk > 50 MPa
8;:::: 0.4 + [0.6 + (0.0014/ccu2)]x/d
where 8 is the ratio of the redistributed moment to the moment before redistribution,x is the
depth of the neutral axis at the ultimate limit state after redistribution, and d is the effective
depth. FN values of Ccu2 refer to Table 4,1.
These are presented graphically in 3.16. Table 3.1 gives thex/d values
for different values of 8.
It should be noted that the above limitations try to ensure sufficiently ductile behaviour.
Higher-strength concrete tends to be and hence the more onerous limits on x/d. x/d
also increases with the steel percentage, and therefore indirectly affects the ductility of the
section.
33
DESIGNERS' GUIDE TO EN 1992-1-1 AND EN ! 992-1-2
Clause
34
3.7.2.
from moment EN 1992-1-1 allows the use of plastic analysis without
any direct check on rotation capacity, provided certain conditions are satisfied:
(1) The area of reinforcement at any location in any direction should not exceed a value
corresponding to xu/d == 0.25 for concrete strength fek ~ 50 MPa and xjd == 0.15 for concrete
strengthfek ~ 55 MPa.
(2) Reinforcing steel should be class B or C.
(3) The ratio of the moments over continuous edges to the moments in the span should be
between 0.5 and 2.0.
The first of the conditions can be expressed as a reinforcement percentage for a balanced
section, i.e. when both concrete and reinforcement have reached their limiting strains (Fig. 3.17):
== 0.8x(0.85fek/1.5)b
Substituting x == 0.25d:
== 0.113bdfek
Fs ==As(0.87fy)
Substituting As == (p/lOO)bd and equating and we obtain
p == 12.99fek1f
y
Table 3.2 gives the values of p for different combinations offck andfy == 500 MPa. Note that
the stress block is a function offck for concrete strength higher than 50 MPa.
Table 3.1. Limiting neutral axis depths and redistribution of moment
xld
5 fck S 50 MPa fck = 55 MPa fck = 60 MPa fck = 70 MPa fck = 80, 90 MPa
0.70 0.260 0.152 0.148 0.143 0.140
0.80 0.360 0.247 0.240 0.232 0.228
0.85 0.410 0.295 0.286 0.277 0.272
0.90 0.460 0.342 0.332 0.322 0.316
d
=---
'--------"'"-.. As
Cross-section Strain diagram I nlernai forces
3.11. Strain and stress conditions in a balanced section
CHAPTER 3. ANALYSIS
Table 3.2. Maximum values of p = I OOAJbd for plastic
reinforcement with frk = 500 MPa
fCk P
20 0.52
25 0.65
30 0.78
35 0.91
40 1.04
45 1.17
50 1.30
60 1.00
70 1.09
80 1.13
90 1.16
The most common plastic method used in practice is the yield line analysis for slabs.
For details of the method, standard textbooks should be consulted. In conjunction with
plastic analysis, an appraisal of the behaviour of the structure at serviceability should be
undertaken.
Non-linear analysis
This takes into account the non-linear deformation properties of reinforced-concrete
sections. This method is seldom used in practical design as it is complex, requiring a
computer and prior knowledge of reinforcement details throughout the structures. It could
be useful in appraising the capacity of existing structures or when a large repetition of a
particular structure is considered (e.g. precast frames).
The method involves plastic hinge analysis (see redistribution of moments, p. 31), which
includes explicit calculation of rotations at hinges. In this method, 'failure' is reached (1)
where limiting rotation occurs at a plastic hinge or (2) when sufficient hinges form to render
the structure into a mechanism, whichever occurs first. A hinge is said to have formed when
the steel starts to yield.
Consider an encastre beam, as shown in Fig. 3.18. The procedure to predict the ultimate
load capacity may be as follows:
(1) Knowing the reinforcement, section properties and concrete grade, calculate the MYk
at A and C. Myk is the bending moment which produces a stress iyk in the tension
reinforcement. Also, calculate Myd by applying 1m'
(2) Calculate the loading which will produce an elastic bending moment at A and B equal
to M
yk
'
(3) Increase the loading in stages. At each stage calculate (a) the rotation ofthe hinges at A
and B by integrating the curvatures of the beam between hinges (this will require the
beam to be divided into a number of sections. At each section the curvature should be
calculated using equation (7.18) of EN 1992-1-1); (b) the bending moment at C; and (c)
the strains in steel and concrete at A, Band C.
(4) Compare the rotation at A and B with the limiting rotation given in 5.6N in
EN 1992-1-1.
(5) Failure is reached when either the limiting rotation is reached at A and B or MYk is
reached at C.
(6) Calculate the load corresponding to 'failure' as noted above.
Note that the rotations are calculated using mean values of mechanical properties whereas
the are calculated design i.e. 1m is used.
35
36
The above procedure is set out to clarify the steps involved in a relatively simple example.
It must be clear that the analysis of a continuous beam will be fairly complex even with a
computer. Also, a number of boundary conditions will need to be imposed in the analysis.
In practice, therefore, this procedure is rarely used, and elastic analysis with moment
redistribution is preferred.
models
Strut-and-tie models utilize the lower-bound theorem of plasticity, which can be summarized
as follows: for a structure under a given system of external loads, if a stress distribution
throughout the structure can be found such that (1) all conditions of equilibrium are satisfied
and (2) the yield condition is not violated anywhere, then the structure is safe under the given
system of external loads. This approach particularly simplifies the analysis of the parts of the
structure where linear distribution of strain is not valid. Typical areas of application are
noted in Section 3.1 above.
Typical models are shown in Fig. 3.19. As can be seen, the structure is thought of
as comprising notional concrete struts and reinforcement ties. Occasionally, concrete
ties may also be considered (e.g. slabs without stirrups; anchorages without transverse
reinforcement). While there might appear to be infinite freedom to choose the orientation
of struts and ties, this is not so. Concrete has only a limited plastic deformation capacity;
therefore, the model has to be chosen with care to ensure that the deformation capacity is
not exceeded at any point before the assumed state of stress is reached in the structure.
Generally, it will be safe to model the struts and ties in such a way that they closely follow the
'stress paths' indicated by the theory of elasticity. The angles between the struts and ties
should generally be greater than 45 in order to avoid incompatibility problems. Where
several models are possible, the 'correct' model is the one with the least number of internal
members and least deformation. In this context, the deformation of the struts may be
neglected, and the model optimized by minimizing the expression
Me: Mp
Mc< Mp
(a)
(b)
3.18. Plastic of an encastre beam: (a) elastic moments when the first hinge(s) form at
A and B; (b) rotation (J at the first when the load is increased
"-"'-'-'-'r"'-'-'-'r'-"
i I I i
. I )--" I
I I / .... I
I / .... I
I (& ) I
! l-J!
I.... / I
. I .... / I .
I I )--'f' I ~
I I I I I I
L . _ . J . _ . _ . _ . L . ~ . _ . _ . _ . L . _ . ~
II ill tI II I I 1+ lilt I " I I 1+ I I I I I
t t t t
L
3.19. strut-and-tie models
Members with holes
a
1
End blocks
Beam--<:olumn junctions
Half joints
\
\
.... \
CHAPTER 3. ANALYSIS
.)
........ ~ - - - - - -
I
I
I
i
I
37
Clause 6.5.2
Clause 6.5.4
38
where is the force in tie, li is the of tie and e
mi
is the mean strain in tie i.
Having idealized the structure as struts and ties, it is then a simple matter to arrive at the
forces in them based on equilibrium with external loads. Stresses in the struts and ties should
be verified including those at nodes, where a number of members meet.
Limiting stresses are as follows:
(1) in ties,!yd
(2) in struts with no transverse tension, 0" Rd. max = fed
(3) in struts with transverse tension, O"Rd. max = O.6v'fed' where v' = 1 fek/250
(4) in nodes where no ties are anchored, O"Rd.max = 1.0v'!cd
(5) in compression-tension nodes, where ties are anchored in one direction only, O"Rd.max =:0
O.85v,!cct
(6) in compression-tension nodes, where ties are anchored in more than one direction,
O"Rd.max =:0 O.75v'!ed
(7) where a load distributed uniformly over an area AeO is dispersed to a larger area
Ac1 (which is concentric to Aco), the applied load F
Rdu
should be limited as F
Rdu
Acofed(Ae/Aeo)05 but limited to 3.OA
cO
fect.
In general, if the above limits are observed, the 'node' regions are unlikely to be critical.
Schlaich and Schafer! recommend a procedure for checking the node regions.
Bearing in mind that strut and tie models fall under plastic analysis, it is interesting to note
that the code does not impose any conditions similar to those noted in Section 3.7.2. This is
an inconsistency. It is therefore important to follow the theory of elasticity fairly closely in
choosing the model as already discussed above.
3.8. aids and simplifications
I. spanning slabs and continuous beams
The bending moment coefficients shown in Fig. 3.20 may be used for three or more equal
spans subject to equal uniformly distributed loads. This assumes supports which do not offer
rotational restraint. The coefficients for locations 3-6 may be used (without sacrificing too
much economy) even when continuity with internal columns is to be taken into account.
However, continuity with external columns has a significant effect on the bending moments
at locations 1 and 2. The following modifications may be made, depending on the ratio of the
total column wall stiffness to the slab stiffness.
It should be noted that the values of bending moments in Fig. 3.20 for variable loads are
the maximum values obtained by a linear analysis considering the cases of loading alternate
and adjacent spans. It will not, therefore, be possible to redistribute the bending moments, as
the loading case applicable will vary with each span.
A further approximation may be made if: (1) the characteristic variable load does not exceed
the characteristic permanent load; (2) loads are predominantly uniformly distributed over three
or more spans; and (3) variations in span lengths do not exceed 15% of the longest span. Under
these conditions the bending moments and shear forces may be obtained using Table 3.3.
Table 3.3 assumes that the ratio of the stiffness of the end columns to that of the slab (or
beam) does not exceed 1.0. No redistribution of moments is permitted to the bending
moments obtained using the above table.
General
For rectangular slabs with standard edge conditions and subject to distributed
loads, normally the bending moments are obtained tabulated coefficients. Such
coefficients are provided later in this section.
CHAPTER 3. ANALYSIS
Svnnm.,lrir.al about this line
2 3 4
Permanent action 0 O.072GL -O.105GL O.033GL -O.079GL 0.046GL
Variable action o O.099QL -O.116QL 0.079QL 0.1060L 0.Oa6QL
(a)
Kb 1 2 3
I
I
Permanent action -0.030GL 0.06GL
O.S
Variable action -O.03QL O.07SQL
~ I I
I .5
Permanent action -O.044GL 0.054GL
rn
co
1.0 E
.m
Variable action -0.0530L O.066QL
t.>
iE
I I
'"
0
(,)
Permanent action -0.057GL O.OSGL
2.0
Variable action -0.065QL 0.057QL
I
Kc = stiffness of external wall
Kb = stiffness of slab, spans 1-3
(b)
3.20. Bending moment coefficients: (a) continuous slab on point supports; (b) continuous slabs
with continuity at external columns
Table 3.3. Ultimate bending moment and shear forces in one-way spanning slabs (continuous
beams and flat slabs)
End support and end span
Simple Monolithic
At first Middle
At outer Near middle At outer Near middle interior of interior Interior
support of end span support of end span support spans supports
Moment 0 O.086FI -O.040FI O.075FI -O.086FI -O.063FI -O.063FI
Shear O.40F O.46F O.6F O.50F
F is the total design ultimate load (1.35G
k
+ 1.50Qk); I is the effective span.
For slabs with irregular plan shapes and slabs subject to a combination of point loads and
distributed loads, Johansen's yield line analysis and the HiHerborg strip method provide
powerful methods for strength calculations (e.g. see Johansen,2 CEB Bulletin 35
3
and Wood
and
39
40
U1n,nrU,rr slabs
Where (1) corners of slabs are free to lift and (2) no provision is made to resist forces at the
corners, the maximum moments per unit width are given by the following expressions:
Msx = bending moment in strips with span lx
= O!sxql}
Msy = bending moment in strips with span ly
- l2
- O!syq x
where lx is the shorter span of the panel, ly is the longer span of the panel and q is the design
ultimate load per unit area. Values of O!sx and O!sy are given in Table 3.4 for different ratios of
ly and lx, where ly is the longer span.
Rectangular panels with restrained edges
Where corners are (1) prevented from lifting and (2) reinforced to resist torsion, the
maximum bending moments per unit width are given by the following expressions:
Msx = f3sAI}
Msy = f3
sy
qlx2
where Msx is the maximum design moment either over supports or at midspan on strips with
span lx, Msy is the maximum design moment either over supports or at midspan on strips with
span fy, q is the design ultimate load per unit area, Ix is the shorter span and ly is the longer
span.
For different ratios of the longer to shorter spans (l/IJ, values of f3
sx
and f3s
y
are tabulated
in Table 3.5 for different edge conditions. Conditions for the use of the tabular value are:
(1) The table is based on the yield line (i.e. plastic analysis). Therefore, conditions (1 )-(3) in
Section 3.7.2 should be observed.
(2) The permanent and variable loads in adjacent panels should be approximately the same
as on the panel being designed.
(3) The span of the adjacent panels in the direction perpendicular to the line of the common
support should be approximately the same as the span of the panel being designed.
(4) Corners at the junction of simply supported edges should be reinforced over the zone
shown in Fig. 3.21.
(5) The panel should observe the detailing rules in Chapter 10 of this guide.
In the corner area shown:
(1) provide top and bottom reinforcement
(2) in each layer provide bars parallel to the slab edges
Table 3.4. Bending moment coefficients for slabs spanning in two
directions at right angles, simply supported on four sides
Vlx
a
sx
1.0 0.062
1.1 0.074
1.2 0.084
1.3 0.093
1.4 0.099
1.5 0.104
1.75 0.113
2.0 0.118
a
sy
0.062
0.061
0.059
0.055
0.051
0.046
0.037
0.029
CHAPTER 3. ANALYSIS
Table 3.5. Bending moment coefficients for rectangular panels supported on four sides with
provision for torsion at corners
Long span
Short span coefficients (3,x coefficients
Type of panel (3,y' for all
and moments Vlx: values of
considered 1.0 1.1 1.2 1.3 1.4 1.5 1.75 2.0
Vlx
Interior panels
Negative moment 0.031 0.037 0.042 0.046 0.050 0.053 0.059 0.063 0.032
at continuous edge
Positive moment 0.024 0.Q28 0.032 0.Q35 0.037 0.040 0.044 0.048 0.024
at mid-span
One short edge discontim.!ou5
Negative moment 0.039 0.044 0.048 0.052 0.055 .0.58 0.063 0.067 0.037
at continuous edge
Positive moment 0.029 0.033 0.036 0.039 0.041 0.043 0.047 0.050 0.Q28
at mid-span
One long edge discontinuous
Negative moment 0.039 0.049 0.056 0.062 0.068 0.073 0.082 0.089 0.037
at continuous edge
Positive moment 0.030 0.036 0.042 0.047 0.051 0.055 0.062 0.067 0.028
at mid-span
Two adjacent edges discontimJol.Is
Negative moment 0.047 0.056 0.063 0.069 0.074 0.078 0.087 0.093 0.045
at continuous edge
Positive moment 0.036 0.042 0.047 0.051 0.055 0.059 0.065 0.070 0.034
at mid-span
Two short edges discontinuous
Negative moment 0.046 0.050 0.054 0.057 0.060 0.062 0.067 0.070
at continuous edge
Positive moment 0.034 0.Q38 0.040 0.043 0.045 0.047 0.050 0.053 0.034
at mid-span
Two long edges discontinuol.ls
Negative moment 0.045
at continuous edge
Positive moment 0.034 0.046 0.056 0.065 0.072 0.Q78 0.091 0.100 0.034
at mid-span
Three edges discontinuous (one long edge continuous)
Negative moment 0.057 0.065 0.071 0.076 0.081 0.084 0.092 0.098
at continuous edge
Positive moment 0.043 0.048 0.053 0.057 0.060 0.063 0.069 0.074 0.044
at mid-span
Three edges discontinuous (one short edge continuous)
Negative moment 0.058
at continuous edge
Positive moment 0.042 0.054 0.063 0.071 0.078 0.084 0.096 0.105 0.044
at mid-span
Four discontim..lolls
Positive moment 0.055 0.065 0.074 0.081 0.087 0.092 0.103 0.111 0.056
at mid-span
41
42
(3) in each of the four layers the area of reinforcement should be equal to 75% of the
reinforcement required for the maximum span moment
(4) the area of reinforcement in (3) above may be halved if one edge of the slab in the corner
is continuous.
Unequal edge conditions in adjacent panels
M_l and M_2 are the support moments for panels 1 and 2, respectively, and M+l and M+2 are
the span moments for panels 1 and 2, respectively.
In some cases, the bending moments at a common support, obtained by considering the
two adjacent panels in isolation, may differ significantly (say by 10%), because of the
differing edge condition at the far supports or differing span lengths or loading.
Consider panels 1 and 2 in Fig. 3.22. As the support on grid A for panel 1 is
discontinuous and support on grid C for panel 2 is continuous, the moments for panels 1
and 2 for the support on grid B could be significantly different. In these circumstances, the
~ 1)5 l
'I
Simply supported
fig. l.21. Corner reinforcement: two-way spanning slabs
I I I
c L _____ iL ______ _
-1- _____ ...,-1-_____ _ __ ....Y.J.-
I I I
I I I
I I I
I I I
I I I Panel 2 I I Ix
I ! I I(
B L - - - - - ~ J - - - - - - - - _ - L ~ - - -
I - - - - - - ~ - - - - - - - - - - ~ ~ - - - -
I I I I I
I I I I I
I I I I I
I I I I I
I I Panel 1 I I I
II I I I I x
I I I I
A - - - - - ~ ~ - - - - - - - -
l.22. WO-W<IV spanning slabs: unequal condition in
CHAPTER 3. ANALYSIS
D 1O:.... __ --lL-_______ ----'" C
Fig. 3.23. Loads on beams supporting two-way spanning slabs
slab may be reinforced throughout for the worst case span and support moments. However,
this might be uneconomic in some cases. In such cases, the following distribution procedure
may be used:
(1) Obtain the support moments for panels 1 and 2 from Table 3.6. TreatingM_
1
andM_
2
as
fixed end moments, the moments may be distributed in proportion to the stiffnesses of
span Ix in panels 1 and 2. Thus, a revised bending moment M ~ f 3 may be obtained for
support over B.
(2) The span moments in panels 1 and 2 should be recalculated as follows:
M'+l = (M_l +M+
1
) - M ~ B
M'+2 = (M_2 + M_2 + M+
2
) M ~ B - M_2
(Note that this assumes that the final moment over C is M_2.)
(3) For curtailment of reinforcement, the point of contraflexure may be obtained by
assuming a parabolic distribution of moments in each panel.
Loads on supporting beams
Loads on supporting beams may be obtained using either Fig. 3.23 or Table 3.6.
Note that:
(1) the reactions shown apply when all edges are continuous (or discontinuous)
(2) when one edge is discontinuous, the reactions on all continuous edges should be
increased by 10% and the reaction on the discontinuous edge may be reduced by 20%
(3) when adjacent edges are discontinuous, the reactions should be adjusted for elastic
shear, considering each span separately.
slabs
Flat slab structures are defined as slabs (solid or coffered) supported on point supports.
Unlike two-way spanning slabs on line supports, flat slabs can fail by yield lines in either of
the two orthogonal directions (Fig. 3.24). For this reason, flat slabs must be capable of
carrying the total load on the panel in each direction.
Methods of analyses
Many recognized methods are available. These include:
(1) the equivalent frame method
(2) simplified coefficients
(3) yield line analysis
( 4) analysis.
43
44
In this guide, only the first two of these methods will be considered. For other approaches
specialist literature should be consulted (e.g. CIRIA Report 11 0
5
).
Equivalent frame method
Division into frames
The structure may be divided in two orthogonal directions into frames consisting of columns
and strips of slabs acting as 'beams'. The width of the slab to be used for assessing the
stiffness depends on the aspect ratio of the panels and whether the loading is vertical or
horizontal.
Vertical loading. When the aspect ratio is less than 2 the width may be taken as the distance
between the centre lines of the adjacent panels. For aspect ratios greater than 2, the width
may be taken as the distance between the centre lines of the adjacent panels when
considering bending in the direction of longer length spans of the panel and twice this value
for bending in the perpendicular direction. (See Fig. 3.25.)
The width of beams for frame analysis is as follows:
Table 3.6. Shear force coefficients for uniformly loaded rectangular panels supported on four sides
with provision for torsion at corners
f3
vx
for values of I'yil'x
Type of panel and
location 1.0 1.1 1.2 1.3 1.4 1.5 1.75 2.0
f3
vy
fOLlI'" edges continuous
Continuous edge 0.33 0.36 0.39 0.41 0.43 0.45 0.48 0.50 0.33
One short edge discontinuous
Continuous edge 0.36 0.39 0.42 0.44 0.45 0.47 0.50 0.52 0.36
Discontinuous edge 0.24
One edge discontinuous
Continuous edge 0.36 0.40 0.44 0.47 0.49 0.51 0.55 0.59 0.36
Discontinuous edge 0.24 0.27 0.29 0.31 0.32 0.34 0.36 0.38
Two edges discontinuous
Continuous edge 0.40 0.44 0.47 0.50 0.52 0.54 0.57 0.60 0.40
Discontinuous edge 0.26 0.29 0.31 0.33 0.34 0.35 0.38 0.40 0.26
Two short edges discontinuous
Continuous edge 0.26 0.30 0.33 0.36 0.38 0.40 0.44 0.47
Discontinuous edge 0.40
Two long edges discontinuous
Continuous edge 0.40
Discontinuous edge 0.26 0.30 0.33 0.36 0.38 0.40 0.44 0.47
Three edges discontinuous (one long edge continuous)
Continuous edge 0.45 0.48 0.51 0.53 0.55 0.57 0.60 0.63
Three edges discontim.lous (one short edge
Continuous edge 0.45
four discontinuous
CHAPTER 3. ANALYSIS
Top
Bottom
Top
fig. 3.24. Possible failure modes of flat slabs
I
I
I
I
I
I I
------,--------,-------
I I
I I
I I
I I
I I
I I
I I
--------1-------
I
I
I
1 i j L.
1 ~ - - - - - - - - - - - ~ - - - - - - - - - - ~ 1
fig. 3.25. Flat slabs: division into frames
When lx < ly < 2lx:
Wx = (lxl + Ix2)12
~ = (IYI + [y2)/2
When Iy > 21x:
Wx = (lxl + Ix2)/2
~ = 2 ~
Horizontal loading. Horizontal loading in the frame will be considered only in unbraced
structures. In these cases, the question of restraint to the columns, and hence the effective
length of columns, is a matter of judgement. If the stiffness of the slab framing into the
column is overestimated, the effective length of the column will reduce correspondingly. As
the stiffness at slab-column junctions is a grey area, codes of practice adopt a cautious
For the slab, half the stiffness applicable to vertical loading is used.
45
46
Stiffness properties are generally based on the gross cross-section (ignoring the
reinforcement). Additional stiffening effects of drops or solid concrete around columns in
coffered slabs may be included, but this will complicate hand calculations. It should be noted
that drops (and solid areas) should only be taken into account when the smaller dimension of
the drop (and solid areas) is at least 33% of the smaller dimension of the surrounding panels.
Analysis
The equivalent frames may be analysed using any of the standard linear elastic methods such
as moment distribution (see Section 3.7.1). Braced structures may be partitioned into
subframes consisting of the slab at one level continuous with columns above and below. The
far ends of the columns are normally taken as fixed unless this assumption is obviously wrong
(e.g. columns with small pad footings not designed to take moments).
The load combinations given in Section 3.2 may be used.
The bending moments obtained from the analysis should be distributed laterally in the
'width' of the slab in accordance with the allocation of moments between strips (see p. 47).
Simplified coefficients
In braced buildings with at least three approximately equal bays and both slabs subject
predominantly to uniformly distributed loads, the bending moments and shear forces may be
obtained using the coefficients given in Section 3.8.1 (Table 3.3).
The bending moments obtained from the above, should be distributed laterally in the 'width'
of the slab in accordance with h the allocation of moments between strips (see p. 47, Table 3.7).
Lateral distribution of moments in the width of the slab
In order to control the cracking of the slabs under service conditions, the bending moments
obtained from the analysis should be distributed taking into account the elastic behaviours of
the slab. As can be imagined, the strips of the slab on the lines of the columns will be stiffer
than those away from the columns. Thus, the strips closer to the column lines will attract
higher bending moments.
Division of panels
Flat slab panels should be divided into column and middle strips, as shown in Fig. 3.26.
3.26. Flat slabs: definition of panels
CHAPTER 3. ANALYSIS
Table 3.7. Distribution of design moments in panels of flat slabs
Negative
Positive
Apportionment between column and middle strip expressed
as a percentage of the total negative or positive design
moment
Column strip (%)
60-80
50-70
Middle strip (%)
40-20
50-30
For the case where the width of the column strip is taken as equal to that of the drop, and the middle
strip is thereby increased in width, the design moments to be resisted by the middle strip should be
increased in proportion to its increased width. The design moments to be resisted by the column strip
may be decreased by an amount such that the total positive and the total negative design moments
resisted by the column strip and middle strip together are unchanged.
As drawn, Fig. 3.26 applies to slabs without drops (or solid areas around columns in
coffered slabs). When drops (or solid areas around columns in coffered slabs) of plan
dimensions greater than [xl3 are used, the width of the column strip may be taken as the
width of the drop. The width of the middle strip should be adjusted accordingly.
Allocation of moments between strips
The bending moments obtained from the analysis should be distributed between the column
and middle strips in the proportions shown in Table 3.7.
In some instances the analysis may show that hogging moments occur in the centre of a
span (e.g. the middle span of a three-bay structure, particularly when it is shorter than the
adjacent spans). The hogging moment may be assumed to be uniformly distributed across
the slab, if the negative moment at midspan is less than 20% of the negative moment at
supports. When the above condition is not met, the moment is concentrated more in the
middle strip.
Moment transfer at edge columns
As a result of flexural and torsional cracking of the edge (and corner) columns, the effective
width through which moments can be transferred between the slabs and the columns will be
much narrower than in the case for internal columns. Empirically, this is allowed for in
design by limiting the maximum moment the slab (without edge beams) can transfer to the
columns:
M
tmax
= 0.17b
c
d
2
fck
where be is the effective width of the strip transferring the moment as defined in Fig. 3.27,
and d is the effective depth of the slab.
M
t
max should not be less than 50% of the design moment obtained from an elastic analysis
or 70% of the design moment obtained from grillage or finite element analysis. If M
t
max is less
than these limits, the structure should be redesigned.
When the bending moment at the outer support obtained from the analysis exceeds M
t
max'
then the moment at the outer support should be reduced to M
t
max' and the span moment
should be increased accordingly.
The reinforcement required in the slab to transfer the outer support moment to the
column should be placed on a width of slab C
x
+ 2r (Fig. 3.28). For slabs with a thickness less
than 300 mm, r = C
y
' and for thicker slabs, r = 1.67C
y
In the latter case, it is essential to
provide torsional links along the edge of the slab. However, U bars (as distinct from L bars)
with longitudinal anchor bars in the top and bottom may be assumed to provide the
necessary torsional reinforcement 3.29).
47
/
be= Cx + Cy
be = Cx + y/2
/I
/
/
/
3.27. Effective slab widths for moment transfer
Cx
~ '
~
A!!'.%
...
~ I
Reinforcement lor ouler support
Momenllo be transferred to column placed in this strip
3.28. Flat slabs: detailing at outer supports
U-bar
Transverse bar Boltom bar
3.29. Flat slabs: detailing at slab
48
CHAPTER 3. ANALYSIS
Bending moments in excess of M
t
max may be transferred to the column only if an edge
beam (which may be a strip of slab) is suitably designed to resist the tension.
Beams
For the design of beams, the bending moment coefficients given in Fig. 3.20 or Table 3.3 may
be used, noting the conditions to be complied with, which are discussed in Section 3.8.1.
Simplifications
EN 1992-1-1 permits the following simplifications regardless of the method of analysis used:
(1) At a support assumed to offer no restraint to rotation (e.g. over walls), a beam or a slab
which is continuous over it may be designed for a support moment which is less than the
moment theoretically calculated on the centre line of the support. The permitted
reduction in moment is then FEd, suppt/S, where supp is the design support reaction and t
is the breadth of the support. This recognizes effect of the width of support and
arbitrarily rounds off the peak in the bending moment diagram.
(2) Where a beam or a slab is cast monolithically into its supports, the critical moment may
be taken as that at the face of the supports (but see also (3) below). This provision is
quite reasonable, as failure cannot occur within the support.
(3) The design moment at the faces of rigid supports should not be less than 65% of the
support moment calculated assuming full fixity at the faces of support. This ensures a
minimum design value for the support moment, particularly in the case of wide supports.
(4) Loads on members supporting one-way-spanning continuous slabs (solid and ribbed)
and beams (including T beams) may be assessed on the assumption that supports
offer no rotational restraint. This is reasonable; but the effect of continuity should be
considered when designing the support such as columns or walls.
49
4
aterials and design data
I. Concrete
4.1.1. General
EN 1992-1-1 covers concrete of strength up to 90 MPa in normal-weight concrete and
80 MPa in lightweight concrete. It relies on EN 206-1 for the specification and production of
concrete.
4.1.2. Strength
In EN 1992-1-1 the compressive strength of concrete is denoted by concrete strength classes,
which relate to the characteristic cylinder strength ICk or the cube strength Ick, cube' The
characteristic strength is defined as that strength below which not more than 5% of all test
results are likely to fall. The relationship between the cylinder and cube strengths is given in
the code, and is reproduced in Table 4.1.
The code provides models for strength development with time, and these should be used
when concrete strength need to be calculated.
Table 4.1 also gives the mean tensile strength of concrete.
4.1.3. Elastic deformation
EN 1992-1-1 recognizes that deformation properties are crucially dependent on the composition
of concrete, and in particular on the aggregates. Therefore, it provides only indicative
information, which should be sufficient for most normal structures. However, in structures
that are likely to be sensitive to deformation, it is advisable to determine the properties by
controlled testing and appropriate specification.
Table 4.1 reproduces the mean values for the modulus of elasticity given in the code. The
values given apply to concretes with quartzite aggregates. The reduction factors for other
Table 4.1. Strength classes and associated properties
fCk, cy (MPa)
15 20 25 30 35 40 45 50 55 60 70 80 90
fCk, cube (MPa)
20 25 30 37 45 50 55 60 67 75 85 95 105
fem (MPa) 24 28 33 38 43 48 53 58 63 68 78 88 98
{ctm (MPa) 1.9 2.2 2.6 2.9 3,2 3.5 3.8 4.1 4.2 4.4 4.6 4.8 5.0
f
ctk
,0.05 (MPa) 1.3 1.5 1.8 2.0 2.2 2.5 2.7 2.9 3.0 3.1 3.2 3.4 3.5
f
ctk
,0.95 (MPa) 2.5 2.9 3.3 3.8 4.2 4.6 4.9 5.3 5.5 5.7 6.0 6.3 6.6
Ecm (GPa) 29 30 31 33 34 35 36 37 38 39 41 42 44
Ecu2 (%0) 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.1 2.9 2.7 2.6 2.6
52
aggregates are as follows: limestone aggregates, 10%; basalt aggregates, 20%; and sandstone
aggregates, 30%.
Poisson's ratio may be taken as 0.2 for uncracked concrete, and 0 for cracked concrete.
The coefficient of linear thermal expansion may be taken as 10 x 1O-
6
/K.
4.1.4. Creep and shrinkage
The creep and shrinkage of concrete depend on a number of factors, including the ambient
humidity, the dimensions of the element, the composition of the concrete and the age of
concrete at the time of loading. The long-term values of the creep coefficient cp( 00, to) are
shown in Figs 4.1 and 4.2 for inside and outside conditions, respectively. The resulting strain
in the member will be cp(oo, to)(aJEc)' When the compressive stress at time to exceeds
AS/ck(t
O
) , the creep coefficient should be multiplied by a factor shown in Table 4.2,
depending on the ratio
k" = aj/cm(t
O
)
to allow for the non-linearity of creep development.
Shrinkage has two components, namely drying shrinkage (ccd) and autogenous shrinkage
(c
c
.). Drying shrinkage is the result of expiration of moisture from the concrete to the
surrounding air. On the other hand, the background document to EN 1992-1-1 defines
autogenous shrinkage as 'the macroscopic volume reduction of cementitious materials when
cement hydrates, after initial setting. It does not include the volume change due to loss
or ingress of substances, temperature variation or the application of external force and
restraint'. Autogenous shrinkage is related inversely to the water/cement ratio, and will be
significant for concretes of high strength.
The drying shrinkage strain at age t is f3dit, ts)khccd,O' Tables 4.3, 4.4 and 4.5 give the data
for calculating shrinkage at 30, 100 and 1000 days, respectively.
Autogenous shrinkage strain at age t is given by cca(t) = f3ascca(oo). Values at 30,100 and
1000 days are given in Table 4.6.
1.5. Stress-strain relationships
A distinction should be made between the stress-strain curve used for analysis and that used
for the design of cross-sections. Figure 4.3 defines the parameters for the curve used for
analysis. For numerical values, see EN 1992-1-1.
EN 1992-1-1 offers two alternatives for the design of cross-sections. These are parabola-
rectangle (Fig. 4.4) and bi-linear (Fig. 4.5) diagrams. For numerical values of the parameters,
see EN 1992-1-1.
2
3
5
20
30
50
100
~ ~ ~
1-8 ~ I . ' "
""
"
~ \ .
"'"
~
\
,
\
\
7.0 6.0 5.0 4.0 3.0 2.0 1.0 0 100 300 500 700 900 1100 1300 1500
<ph to) ho (mm)
4.1. coefficients for concrete in indoor conditions
CHAPTER 4. MATERIALS AND DESIGN DATA
2
3
.........
"0

'"
I
'-.."
r------
5
20
30

I
I
I

I'
Ill..
'\

C30/37
C35/45
C40/50
50
i
100
\
6.0 5.0 4.0 3.0 2.0 1.0
cp(=, to)
0100 300 500 700 9001100 13001500
ho (mm)
4.2. Creep coefficients for concrete in outdoor conditions
Table 4.2. Multipliers for non-linear creep
ka
0.5
0.6
0.7
0.8
0.9
1.0
Table 4.3. Values of ccd (%0) for drying shrinkage
20/25
40/50
60/75
80/95
90/105
Relative humidity
Indoor conditions, 50%
0.55
0.44
0.35
0.29
0.26
Multiplier
1.078
1.252
1.455
1.690
1.964
2.282
Outdoor conditions, 80%
0.31
0.25
0.20
0.16
0.15
EN 1992-1-1 also permits a rectangular stress block to be used for section design. This is .
shown in Fig. 4.6. The basic compressive stress is fed = aecfekhe' The coefficient a
ee
is said to
take account of long-term effects on the compressive strength and of the unfavourable
effects from the way the load is applied. EN 1992-1-1 recommends a value of 1.0 for
this coefficient, but the UK in its National Annex proposes 0.85. This reduced value, which
was used in the ENV version of EN 1992-1-1, is considered necessary as a calibration factor
53
Table 4.4. kh to shrinkage
coefficient for size of members
ho (mm)
100
200
300
~ 500
1.00
0.85
0.75
0.70
(I) ho is the notional size of the cross-section
in miilimetres.
(2) ho = 2 X (area of cross-section)/(perimeter
of that part of the cross-section which is
exposed to drying).
Table 4 .5. Multipliers fJds(t, t
s
) for shrinkage development
t - ts days
ho (mm)
30 days 100 days
100 0.428 0.714
200 0.209 0.469
300 0.126 0.325
~ 500 0.063 0.183
1000 days
0.962
0.898
0.828
0.691
t - ts is the age of concrete in days from the end of curing when the shrinkage strain is being calculated.
Table 4.6. Autogenous shrinkage coefficients at different ages cca(oo)fJas(t) (%0)
fck (MPa)
20 25 30 35 40 45 50 60 70 80 90
cca(=)fJas
0.025 0.038 0.050 0.063 0.075 0.088 0.100 0.125 0.150 0.175 0.200
(1000 days)
cca(=)fJas
0.022 0.033 0.043 0.054 0.065 0.075 0.086 0.108 0.129 0.151 0.172
(100 days)
cca(=)fJ
as
0.017 0.025 0.033 0.042 0.050 0.056 0.067 0.083 0.100 0.116 0.133
(30 days)
O"c
"m
4.3. Stress-strain relation for structural
54
fCk -----------:;: -
I
I
o
ec2
4.4. Parabola-rectangle diagram for concrete under compression
fCk ________________ tj- ___ _
I:
I :
I
fCd -- - --- - - - _L __ --;.: ----------...;
I
o
4.5. Bi-linear stress-strain relation for concrete
4.6. Rectangular stress distribution for concrete
between the predicted strength and that obtained in experiments. See the background paper
to the UK National Annex.
6
The other parameters defining the rectangular stress block are ).. and ry. These are
functions of concrete strength, but have constant values of 0.8 and 1.0, respectively, for
fek:::; 50 MPa. Table 4.7 lists the values for different concrete grades.
The value of design tensile strength fCld = OIcJctk,o.oshe' In the UK National OI
et
= 1.00.
55
56
I.
EN 1992-1-1 deals with additional requirements for lightweight concrete in Chapter 11. The
requirements for normal-weight concrete are generally applicable to lightweight concrete
unless specifically varied. The code uses the subscript'!' to distinguish lightweight concrete
~ c k , E
lcm
, Slcu2, etc.)
Lightweight concrete is defined as concrete having a closed structure and a density not
exceeding 2200 kg/m3.
Density classes
Six density classes are defined in EN 206-1. Table 4.8 reproduces the relevant data. In each
class, a range is given for the density and the nominal density to be used in design calculations.
Many properties of lightweight concrete are related to its density p. The coefficient
'rJj = 0.4 + 0.6p/2200
is used to modify the relevant property of normal-weight concrete. In this expression, p
refers to upper limit of the density for the relevant density class.
Table 4.7. Values of A and 7) for different strengths of concrete
fek (MPa) A 7)
::; 50 0.800 1.00
60 0.775 0.95
70 0.750 0.90
80 0.725 0.85
90 0.700 0.80
Table 4.8. Density classes for lightweight concrete
Density class
1.0 1.2 1.4 1.6 1.8 2.0
Density range (kg/m
l
) 801-1000 1001-1200 1201-1400 1401-1600 1601-1800 1801-2000
Nominal design denSity
(kg/m
l
)
Plain concrete 1050 1250 1450 1650 1850 2050
Reinforced concrete 1150 1350 1550 1750 1950 2150
Table 4.9. Strength classes and associated properties for lightweight concrete
ftek. cy (MPa)
16 20 25 30 35 40 45 50 55 60 70 80
ftek. cube (MPa)
18 22 28 33 38 44 50 55 60 66 77 88
ht frem (MPa) 22 28 33 38 43 48 53 58 63 68 78 88
ftetm (MPa) fretm := fetm 7) 1
fretk.0.05 (MPa) freck. 0.05 = fetk. 0.057)1
fretk.0.95 (MPa) fretk.0.95 = feck, 0.957)1
E
1em
(GPa)
E!cm =0 Ecm7)E
"!cu2 (%0) 3.57)1 3.57)1 3.57)1 3.57)1 3.57)1 3.57)1 3.57)1 3.1711 2.9711 2.7711 2.67)1
See the text for the definition of 7)1 and 71E'
(a)
f,= kfo,2k ------
fYk
O ' ~ ~ f - - cuk
(b)
4.7. Stress-strain diagrams of typical reinforcing steel. (a) Hot-rolled steel. (b) Cold-worked steel
modifying ....... "ffi, .. i""'Mlt-c:
The mean value of the modulus of elasticity is obtained by multiplying the value for
normal weight concrete by 'TIE = (p/2200f
The creep coefficient !p( 00, to) for normal-weight concrete should also be multiplied by
'TIE to obtain the coefficient for lightweight concrete. The creep strains so derived should
be further multiplied by 1.3 for /rek ::::; LC16/20.
Shrinkage strains for lightweight concrete should be obtained by multiplying the values
of ced,O for normal-weight concrete by 'TI3 = 1.5 for/rek ::::; LC16/20 and 1.2 for/rek2 LC20/25.
Table 4.9 summarizes the properties of lightweight concrete.
Design compressive c:rt.,.i410rn
The value of the design compressive strength is
,fted = alecfrckire
The recommended value in the code for alec is 0.85. The UK National Annex proposes to
adopt the same value.
Similarly, the value of the design tensile strength is
/retd = alet/retk/rc
with a
let
= 0.85.
Reinforcing steel
EN 1992-1-1 rules are valid for ribbed bars, de-coiled rods, welded fabrics and lattice girders.
The range of essential properties is listed in the normative Annex C of the code. Although the
reinforcement standard EN 10080 will be published soon, it will confine itself to the testing
requirements of properties and not define the properties themselves.
I. Strength {fyJ
The grade of reinforcement steel denotes the specified characteristic yield stress lfyk)' It is
obtained by dividing the characteristic yield load by the nominal cross-section area of the bar.
For products without a pronounced yield stress, the 0.2% proof stress!O.2k is substituted as the
yield stress. Typical and idealized stress-strain diagrams are given in 4.7. EN 1992-1-1 is
valid for yield stress of reinforcement in the range 400-600 MPa. In the UK, the characteristic
strength lfyJ of the commonly used grade of reinforcement is likely to be 500 MPa.
In general, ductility is inversely related to yield stress. Therefore, in applications where
ductility is critical (e.g. seismic design), it is important to ensure that the actual strength
57
58
fpk _ _ _ _ ____ ~ _ : : . : . . : : . . __ : . . . : _ : . . : _ ; . ; ; ; _ ~ __ "' ___ ,
4.8. Stress-strain diagram for typical prestressing steel
Table 4.10. Ductility classes for bars, decoiled rods and wire fabrics
Ductility class A
2': 1.05
2': 2.5
Ductility class B
2': 1.08
2': 5.0
Ductility class C
2': 1.15 but < 1.35
2': 7.5
does not exceed the specified value by a large margin. Limits for the ratio of the actual
yield stress to the specified strength are specified for such applications. Currently, no UK
standard gives any guidance.
There is no technical reason why other types of reinforcement should not be used in
conjunction with EN 1992-1-1, provided suitable allowance is made for the behaviour. For
instance, the UK National Annex provides some advice on the use of plain bars with
fyk = 250 MPa. Relevant authoritative publications should be consulted when other types of
reinforcement are used.
Ductility is defined using the strain at maximum load (Cuk) and the ratio between the
maximum and the yield strengths ifJf)k' Ductility is an essential property if advantage is to
be taken of the plastic behaviour of structures. The greater the ductility, the greater the
elongation in axially loaded members, and the greater the rotation capacity in members
subjected to flexure. In members where the ultimate strength is governed by the yielding of
reinforcement (shallow members with a low percentage of steel), clearly the higher the value
of cuk the greater the ductility (Le. the plateau of the stress-strain diagram is long).
When the ultimate strength is controlled by the strain in concrete reaching the limiting
value, the length of the plastic zone influences the rotation. The longer this length, the
greater is the rotation. ft/.t;, indicates the length over which yield takes place (i.e. the length of
the plastic zone) (Fig. 4.8).
EN 1992-1-1 defines three ductility classes (A, Band C ( in terms of Cuk and (ft/fy))' These
are shown in Table 4.10.
EN 1992-1-1 states that its requirements will be satisfied if the prestressing steel ,",V'UiJ""'''
with EN 10138. The definition of the relevant parameters is shown in 4.8.
5
esign of sections for bending
and axial rce
This chapter is concerned with the design for the ultimate limit state of sections subject to
pure flexure, such as beams or slabs, sections subject to combined bending and axial load,
such as columns, and prestressed sections. The material in this chapter is covered in
EN 1992-1-1 in the following clauses:
Design stress-strain curves for reinforcement
Design stress-strain curve for prestressing steel
Design stress-strain curves for concrete
Basic assumptions for section design
Minimum reinforcement
Limitations imposed by redistribution
I. Bask assumptions
Clause 3.2.3
Clause 3.3.3
Clause 3.1. 7
Clause 6.1
Clauses 7.3.2 and 9.2.1.1
Clause 5.5
The basic assumptions about section behaviour are very similar to those adopted by many, if
not most, modern codes of practice. The formulation used in EN 1992-1-1 is taken from the
CEB Model Code for Concrete Structures.
7
The assumptions in EN 1992-1-1 differ in detail to
those in the UK code, BS 8110 in ways which, as will be seen, make calculations rather
more complex in some cases, but the practical outcome is not significantly different. The
assumptions define the stress-strain responses to be assumed for steel and concrete, and the
assumptions to be made about the strains at the ultimate limit state. It is these assumptions
about the strain that define failure.
5.1.1. curves
The information required to obtain the design stress-strain curves for concrete, ordinary
reinforcement and prestressing steel are to be found in clauses 3.1. 7, 3.2.3 and 3.3.3,
respectively, of EN 1992-1-1.
For reinforcement and prestressing steel the code specifies the use of bi-linear stress-
strain curves. These are given in clauses 3.2.3 and 3.3.3 for reinforcing and prestressing
steels, respectively. In each case, it is possible to choose between two possible bi-linear
diagrams for the design of sections: one with a horizontal top branch and one with an
inclined top branch. These curves are shown in Fig. 5.1. Where the horizontal top branch is
used, no limit on the tensile strain is imposed; however, the characteristics of the inclined top
branch depend on the ductility class of the reinforcement. These are given in Table 3.3 in
EN 1992-1-1 for and are repeated in Table 5.1 for convenience.
Clause 3.1.7
Clause 3.2.3
Clause 3.3.3
Clause 3.2.3
Clause 3.3.3
DESIGNERS' GUIDE TO EN 1992- -I AND EN 1992-1-2
Clause 3.2.3
60
The ultimate strain in design when the inclined upper branch is used is 0.9c
u
k.
This is assumed to correspond to a maximum stress of (j/f)dykh
s
. Clearly, use of
the curves with inclined top branches will give some economic advantage over the use
of the horizontal top branch. Potentially, this advantage could reach an 8% saving in
reinforcement, but a saving approaching this value will only rarely be achievable, and will be
at the expense of considerably more complex calculations. The fun saving will only be
available where it is certain at the design stage that a high-ductility steel will be used and
where the neutral axis depth at the ultimate limit state will be around 0.25. Above this the
steel stress will be below ftkhs' while for lower neutral axis depths the effect of limiting the
tension strain will lead to a lower average stress in the compression zone, and hence a deeper
neutral axis than if the strain was unlimited.
In order to keep the design equations and design aids reasonably simple, the horizontal
upper branch will be used in the derivation of all equations, charts and tables in this chapter.
For concrete, three possibilities are described in clause 3.2.3 of EN 1992-1-1. The preferred
idealization is the parabolic-rectangular diagram, but a bi-linear diagram and a rectangular
diagram are also permitted. These three diagrams are compared in Fig. 5.2 with the limiting
strains tabulated in Table 5.2.
Characteristic curve
----------.\ - ~ -=-- ~ -..::. -:--:= -.; i-
fYk" ------- r
I
Design curves
Es 200 kN/mm2
(a)
Strain
Characteristic curve
fpk ------------- - . \ - ~ - = - - - : : . : - - __
fpO.1k "
-------- r
I
fpo.1,bs - -- -- -- I - = = = = - - ~ - - . . - - - - - - : . . .
Ep (value varies)
Strain
(b)
5.1. (a) Stress-strain rii",cr .. :>,m for .. "i"t,.."r,-i .. ,,, steel. (b) and characteristic stress-strain
curves for steel
CHAPTER 5. DESIGN OF SECTIONS FOR BENDING AND AXIAL FORCE
Table 5.1. Parameters defining the ductility classes of reinforcement
Ductility class
A
B
C
Strain
Ultimate tensile strain, Euk
0.D25
0.050
0.075
Parabolic-rectangular Bi-linear
Fig. S.2. Idealized stress-strain distributions
Table S.2. Parameters defining design stress-strain curves for concrete
Grade
Ecu Ed Ed
n
::;C50 0.0035 0.0020 0.00175 2.0
C55 0.0031 0.0022 0.00180 1.75
C60 0.0029 0.0023 0.00190 1.60
GO 0.0027 0.0024 0.00200 1.45
CSO 0.0026 0.0025 0.00220 1.40
C90 0.0026 0.0026 0.00230 1.40
1.05
1.08
1.15 ::; (t;lfy)k::; 1.35
Rectangular
A 7)
0.800 1.000
0.788 0.975
0.775 0.950
0.750 0.900
0.725 0.850
0.700 0.800
Extension of the code to include concrete grades up to 90 N/mm2 has required modification
of the idealized stress-strain curves for concrete strengths greater than 50 N/mm2. This
arises because research has shown that higher concrete strengths exhibit more brittle
behaviour. Above a characteristic strength of 50 N/mm2, the parabolic-rectangular diagram
is not, in fact, parabolic-rectangular even though the code describes it as such. The formula
for the curved part of the response is
O"c = fet[l - (1 - E/E
c2
tl
n varies from 2 to 1.4, depending on the concrete strength, and is clearly only a parabola
whenn =2.
It will be noted that the maximum design stress is given by the characteristic concrete
strength multiplied by a coefficient nee and divided by the partial safety factor for the
concrete, i.e.
fed = ncelc)"!c
The coefficient nee is described in clause 3.1. 6 in EN 1992-1-1 as 'taking account of long term
effects on the compressive strength and of unfavourable effects resulting from the way the load
is applied'. A value of 1.0 is suggested, based on the that there should be
no long-term effects at ultimate not already included in the data from which
the stress-strain curves were derived. It arguable that the reason for the
Clause 3.1.6
61
62
introduction of this factor is more related to the idealization of the shape of the stress-strain
diagram as used for flexure than to the nature of the loading (i.e. it is needed to allow for the
way the load is applied). Figure 5.3 tries to illustrate this. It will be seen for the area
under the actual curve and the idealized curve to be the same, the maximum stress level for
the idealized curve must be below the maximum stress of the 'true' diagram. Furthermore,
study of the data available on the behaviour of compression zones at failure suggests that the
use of 1.0 is unconservative. For this reason, the UK N ationaJ Annex recommends a value for
ace of 0.85, as is proposed in the CEB Model Codes.
Derivation of stress block parameters
Table 5.3 compares the three permitted idealizations in terms of the average stress over a
rectangular compression zone and the distance from the compression face of the section to
the centre of compression as a fraction of the neutral axis depth (f3). In this table, a has been
assumed to be 0.85 and Ie has been taken as 1.5. As well as providing a convenient
comparison of the idealizations, Table 5.3 will be found to provide information in a useful
form for design calculations.
fek - - - - - - - - - - - - - - - - . ~ - - - ~
True curve
/'
/' /' Idealized curve
/
/
I
'I
'I
5.3. Comparison of true and idealized stress-strain curves
Table 5.3. Comparison of stress block parameters
Parabolic-rectangular Bi-linear rectangular Rectangular
Average stress Centroid Average stress Centroid Average stress Centroid
Grade (N/mm2) factor, {3 (N/mm2) factor, {3 (N/mm2) factor, {3
12 5.51 00416 5.10 0.389 5.44 0.40
16 7.34 0.416 6.80 0.389 7.25 0.40
20 9.18 00416 8.50 0.389 9.07 0.40
25 11047 0.416 10.63 0.389 11.33 0.40
30 13.76 0.416 12.75 0.389 13.60 0.40
35 16.06 0.416 14.88 0.389 15.87 0.40
40 18.35 0.416 17.00 0.389 18.13 0.40
45 20.64 0.416 19.13 0.389 20.40 0.40
50 22.94 0.416 21.24 0.389 22.67 0.40
55 23.19 0.393 22.10 0.374 23.93 0.39
60 23.58 0.377 22.88 0.363 25.03 0.39
70 24.86 0.360 24.55 0.349 26.78 0.38
80 27.1 0.355 26.51 0.342 27.94 0.36
90 29.75 0.353 28.44 0.337 28.56 0.35
o
~ - ~ ~ ......................... - - - - - - . - - - - - - - - , - - - - - - - , - - - ~ .. ~
1
h
Strain diagram for
section with neutral
axis within section
(1 - cC2/cC")h
or (1 ccjc,,)h
i
j
~ Strain diagram for
section in axial
compression
. ~ ~
Tension
Strain diagram for section
with zero strain at the least
compressed face
S.4. Ultimate strain distributions
Compression
Strain diagram for section
fully in compression but
with some bending
It will be seen from Table 5.3 that the results obtained from the three idealizations will be
so similar as to be indistinguishable for all normal purposes for concrete strengths up to
50 N/mm
2
.
relating to at the ultimate limit state
The basic assumptions here are that plane sections remain plane and that the strain in
reinforcement is the same as the strain in the concrete at the same level. These assumptions
are universally accepted for the design of members containing bonded ordinary
reinforcement. For prestressed sections, allowance has to be made for the strain in the steel
prior to it being bonded to the concrete. After bonding, the change in strain in the steel is
assumed to be the same as the change in strain in the concrete. The assumption is, of course,
invalid for beams with unbonded tendons. These lie outside the scope of EN 1992-1-1.
However, the assumptions are not strictly true. The deformations within a section are very
and, locally, plane sections do not remain plane. Nor, due to local bond slip, are the
strains in the concrete exactly the same as those in the steel. Nevertheless, on average, the
assumptions are correct, and are certainly sufficiently true for practical purposes for the
design of normal members. One area where they are not adequate is in short members. For
this reason, deep beams are not designed using the provisions of this chapter, and nor are
members such as corbels.
It is universal to define failure of concrete in compression by means of a limiting compressive
strain. The formulation of the limit varies from code to code, for example the American
Concrete Institute code, ACI 318, uses a limit of 0.003, while the UK code BS 8110 uses
0.0035. For concrete strengths not exceeding 50 N/mm2, the Eurocode adopts values taken
from the CEB Model Code. These comprise a limit of 0.0035 for flexure and for combined
bending and axial load where the neutral axis remains within the section, and a limit of
between 0.0035 and 0.002 for sections loaded so that the whole section is in compression.
This is illustrated in Fig. 5.4. The values for the ultimate compressive strains for situations
where the neutral axis lies within the section and for axial compression vary for higher-
strength concretes, with the ultimate strain strength for cases where
the neutral axis is within the section, while the value for axial compression increases with
63
64
The logic behind the reduction in the strain limit for axial compression is in axial
compression, failure will occur at the strain to the attainment of the maximum
compressive stress. This is 0.002 for concrete strengths not exceeding 50 N/mm2. In flexure,
considerably higher strains can be reached before the maximum capacity of the section is
reached, and the value of 0.0035 has been obtained empirically. A means is needed to
interpolate between the value of 0.0035 for flexure and 0.002 for axial load, and 5.4
provides this.
5.1.1. reinforced beams and slabs
The conditions in a singly reinforced rectangular section at the ultimate limit state are
assumed to be as shown in Fig. 5.5.
From Fig. 5.5, it will be seen that the following equations can be derived from consideration
of equilibrium of axial forces and moments:
x/d = pfyd/fav
M = favbx(d - (3x) or M/bd
2
= fav(l - (3x/d)x/d
(D5.l)
(D5.2)
whereAs is the area of tension steel'!av is the average stress over the compression zone,!yd is
the design yield strength of the steel, (3 is the ratio of the distance of the centre of
compression from the compression face to the neutral axis depth, and p is the reinforcement
ratio (Ajbd).
For the parabolic-rectangular stress-strain curve for a concrete strength not greater than
50 N/mm2 and taking the partial safety for concrete as 1.5 and 0; as 0.85, we can write
fay = 0.459fck
fyd = f
y
k/1.5
(3 = 0.416
Values for higher strength concretes can be obtained from Table 5.3.
Substituting for x/d in equation (D5.2) from equation (D5.1) gives
pfYd/fav = 1/2(3 - -V(1/4(32 + Kaj(3)
where Kav = Mctfbd
2
fav or, substituting the values for (3 and fay given above,
As = f
ck
bd(0.633 - -V(0.4 -1.46K)/fYk
in whichK =M/bd
2
fck.
(D5.3)
It may also sometimes be useful to be able to obtain the neutral axis
arm. These are given by
and the lever
r-
b
-1
TI__:____.------?--------------------------,------L----,
j
As Zl
A/
Yd
--- -.---e--e-- ------------------ ------------------ ----------------------+-----'----1----------------- ------------------
L-____ --J _______________________________________________________________________________ _
Strains Forces
5.5. Conditions in a reinforced section at the ultimate limit state
1.00
0.95
"'"
tl
0.90
J!!
c
0
0.85
~
.Q
c
1ii
'6
0.80 <J)
cr: Reinforcement
0.75
0.16 0.18
I
I
I
I
I
I
. I
1---,-------
0.2 0.22 0.24 0.26
Limiting values of Md/bd2fcd
0.28
fig. 5.6. Limiting values of Mibd2fcd as a function of the amount of redistribution
x/d = 1.895fy0Afe/Yd)
z/d = 1 - 0.416x/d
~
VI
0.3
(D5.4)
(D5.5)
It is now necessary to consider the limits to the use of singly reinforced sections. Clause 5.5
in EN 1992-1-1 gives limits to the neutral axis depth at the ultimate limit state as a function of
the amount of redistribution carried out in the analysis. The limit is repeated here for
convenience:
x/d:::; (8 - 0.4)/(0.6 + 0.0014/E
eu
) (D5.6)
In applying this formula, some degree of interpretation of the code is necessary since the
reader is referred to Table 3.1 of EN 1992-1-1 for values of the ultimate strain. Table 3.1 gives
several values, and it has been assumed that the ultimate strain is Eeul in this table.x/d should
not be greater than 0.45 for concrete grades C50 or below, and not greater then 0.35 for
grades C55 or higher.
8 in equation (D5.6) is the ratio of the redistributed moment to the moment before
redistribution. 8 is limited as a function of the type of reinforcement used as follows: for Class
B and Class C steel, 8 ~ 0.70 and for Class A steel, 8 ~ 0.8.
The neutral axis limits set out above can now be substituted into equation (D5.2), to give
limiting values of the parameterized moment, M
d
/bd
2
fed' This gives the limits shown in
5.6. The sloping parts of the relationships for a given concrete strength are the parts of the
relationship controlled by the equation given above while the vertical parts are governed by
the neutral axis limits of 0.45d or 0.35d as appropriate for the particular concrete strength.
The expressions for the calculation of the areas of reinforcement can be presented as
design charts. This is done in Fig. 5.7, which gives the mechanical reinforcement ratio,
Alyd/bdfed' in terms of the non-dimensional moment parameter, M/bd
2
fed' It should be noted
that M/bd
2
fed should not be taken as greater than the limiting value given by 5.6. The
scale on the upper edge of Fig. 5.7 gives values ofx/d at ultimate. This is of use where flanged
sections are being considered. If a greater moment capacity is required, then a doubly
reinforced section will be required. This is considered below. It can easily be shown that
there is no need to distinguish between different concrete grades, as the line for a 90 N/mm2
concrete is indistinguishable from that for a 50 N/mm2 concrete. This chart can now be used
for the design of singly reinforced rectangular sections or flanged sections where the neutral
axis remains within the The use of the chart is illustrated below by a simple C)I.(UHIJ!C'.
Clause 5.5
65
66
for the ultimate limit state.
Assume that the effective is 450
the concrete and the reinforcement are, ",p,,"p.r>tn1e>h
then is 435 . This enables be calculated as
0.139
the 111""11<,,,",,a1 ratio can be read off as
of reinforcement of
0.151 x 300 x 450 x 663
A check should also be carried out to ensure
the amount of carried out. Since
and the maximum allowahle value
reinforced section
reinforced sections
is
an area
If the design moment is greater than the limiting value obtained from Fig. 5.6, it is necessary
to add compression reinforcement to maintain the neutral axis depth at the limiting value.
The equations
(D5.7)
As + As, lim (D5.8)
will achieve this. M
lim
is the moment corresponding to the limiting value of M/bd
2
fcd obtained
from Fig. 5.6. As, lim is the value of As obtained when M = M
lim
This can be obtained from Fig. 5.7.
It should be noted that the equations derived in this section implicitly assume firstly that
the area displaced by the compression reinforcement is ignored and secondly that the
compression reinforcement is working at its design yield strength. This second assumption is
only true if
fsd/Es < Ecu(X - d')/x (D5.9)
xld
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
0.28
i
i I .

V I I
0.24

--- . /
V
-
0.20
"
"tl
0.16
:!?
:;:;;"
0.12
/,V
I

/'"
f---
v...- I
c---
V
I
I /i-""
,
I
--
I

/'
,

0.08
0.04
__ I A I

c---

!
i

I

0.00
/'" I !
0.00 0.04 0.08 0.12 0.16 0.20 0.24 0.28 0.32 0.36
5.7. chart for reinforced sections
CHAPTER 5. DESIGN OF SECTIONS FOR BENDiNG AND AXIAL FORCE
Figure 5.7 in conjunction with equations (DS.7) and (DS.8) can be used for the design of
doubly reinforced sections, as illustrated in 5.2.
with compression rll'!!inf,arfE':lI'!rnp,nt
The amounts of tension and C01TIoreSSlOfl reinforcement are to be calculated for a section
300 mm wide and 450 mm overall ultimate moment of 275 kN m.
concrete is
for the ultimate limit state.
a
tension reinforcement as
:::: 220 + 1688 = 1908
Design of flanged sections (T 01" l beams)
it is
can now be used to obtain the
rt'lrrp<:n{,n{i<: to
the area of
Provided that the neutral axis at the ultimate limit state remains within the flange of a T or L
beam, the section may be treated as a rectangular section, and the equations derived above
may be used. From experience, it will only very infrequently be necessary to do other than
consider a flanged beam as rectangular. If the neutral axis does need to be deeper than the
flange depth, then it will be sufficiently accurate to assume that the whole of the outstanding
parts of the flanges are carrying a concrete stress of fed' This leads to the equilibrium
equations given below:
AJyd = fedhf(b - b
r
) + f a v b ~
M = favbrx(d - f3x) + fedhf (b br)(d - hf/2)
(DS.lO)
(DS.11)
where b is the overall breadth of the section, b
r
is the breadth of the rib and h
f
is the depth of
the flange. An alternative way of looking at these equations is to see that the outstanding part
of the flanges carry a moment equal to
Mnange = fCdhib br)(d - hf/2)
which requires a reinforcement area of
Af = fedhf(b - bJlfyd
The remaining part of the beam can be considered as a rectangular beam of width b
r
and
effective depth d supporting a moment of M - Mflangc' Thc reinforcement required for this
may be calculated using 5.6 and 5.7. The total area of tension reinforcement is thenAf
that required to resist M - Mnangc.
the moment resistance of mOl"e section shapes
It is not possible to derive equations for all possible situations, and it is useful to have a more
general approach available. A suitable iterative for the calculation of the moment
of resistance of any section is as follows:
67
68
(1) Guess a value for x.
(2) Calculate the compressive force in the compression zone using any of the allowable
idealizations of the compressive stress-strain curve. If the parabolic-rectangular diagram
is used, a numerical technique may be most convenient. For simple hand calculations,
the rectangular idealization is much the easiest to apply.
(3) From the neutral axis depth and the ultimate strain, calculate the strain in each layer of
reinforcement and hence the stress and the force in each layer.
(4) Check whether the total compressive force is equal to the total tension force. If this is so,
then the neutral axis depth is correct, and the moments of the internal forces may be
taken to obtain the moment of resistance. If the tension and compression forces are not
equal, adjust the value of the neutral axis depth and repeat the calculation from step (2).
S.2.S. Design IOf rectangular clOlumn sectilOns
The basic equations for equilibrium of a rectangular section subjected to combined bending
and axial load are as given below for situations where the neutral axis remains within the
section:
N Rd = fav
bx
+ fsAs
M
Rd
= favbx(h/2 - ;3x) + - d;)
(D5.12)
(D5.13)
In equation (D5.13), moments have been taken about the centroid of the concrete section.
The summation signs indicate a summation over all the levels of reinforcement within the
section. In carrying out the summation, tensile stresses must be taken as negative. d; is the
distance from the compressive face of the section to the ith layer of reinforcement.
Assuming that the values for the partial safety factors on the steel and concrete are taken
as 1.15 and 1.5, respectively, and taking ace as 0.85, we can substitute 0.459fck forfav and 0.416
for ;3. The resulting equations are rigorous for situations where the neutral axis remains
within the section. Where the whole section is in compression, more complex expressions are
required to deal with (1) the portion of the parabolic curve cut off by the bottom of the
section and (2) the reduction in the ultimate strain at the compressive face according to
5.4. Additionally, more complex equations are required for cases where the concrete
strength exceeds 50 N/mm
2
. The resulting equations are unwieldy, and it is not appropriate
to present them here. A simpler approach is to use design charts. Figures 5.8a-l provide
a series of parameterized charts for the design of symmetrically reinforced rectangular
columns derived using the parabolic-rectangular diagram. Three sets of charts are included
here. The first set (Figs 5.8a-d) can be used for sections where either the reinforcement is
concentrated at the corners, as shown in the diagrams on the charts, or for sections where the
steel is distributed along the sides parallel to the axis of bending. The second set of
charts (Figs 5.8e-h) are for sections where the reinforcement is distributed in the most
disadvantageous way along the sides of the section which are perpendicular to the axis of
bending. The final set of charts (Figs 5.8i-l) are for sections where the reinforcement is
distributed evenly around the section perimeter. The use of these charts is illustrated by
Example 5.3.
Rigorously, a further complete set of charts is required for each concrete grade above
50 N/mm2, as the parameters defining the concrete stress-strain curve change. What has
been done here is to provide a further complete set of charts for 90 N/mm2 concrete (Figs
5.9a-l). Interpolation may be used between the charts for 50 and 90 N/mm2 concrete for
intermediate strengths.
Charts can also be produced for circular column sections, and a set of parameterized
charts are given in Figs 5.lOa-d. The charts are drawn assuming that the section contains six
reinforcing bars, which is the minimum that can reasonably be used in a circular section. It is
found that there is no single of the reinforcement relative to the axis of bending
which will always give the minimum strength. The charts are therefore drawn to give a lower
bound to the interaction for various of the bars.

13"
::2 1.0

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(a)
M/
bh2f
cd
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(b)
M/
bh2f
cd

I
1.8 d'l]e Ih
1 . 6
1.4 d, e CD -.L
1. 2 I d,l h = 0.15

::2 1.0 ..--.-_j

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(e)
M/
bh2f
cd
5.8. charts for r<>rr",nOl
columns (ck ::;; 50)
69
DESIGNERS' GUIDE TO EN 992-1 I AND EN 1992-1-2
2,0
1,8 d,
1,6
1.4
d,
1.2
U
~ "
.c:
1,0
:'2
:z:.U
0,8
0,6
0.4
0,2
0,0

0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
(d)
Mi
bh2f
cd
2,0
1,8
1,6
1.4
1,2
U
,-0
.c:
1.0
:'2
:z:.U
0,8
0,6
0.4
0,2
0,0

0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
(e)
Mi
bh2f
cd
2,0
1,8
A/yJbhf
cd
~ - t
d ~
~ b ~
1,'0-- -,-- 1
.. e I
1.6

h
1.4
~
<II
1,2
= 0,10
u
,,-0
.c:
1,0
:'2
:z:.U
0,8
0,6
0.4
0,2
0,0

0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
(I)
Mi
bh2f
cd
5.8.
70
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(g)
Mi
bh2f
cd
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(h)
M/
bh2f
cd
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(i)
M/
bh2f
cd
5.8.
71
72

d,:r-:l
i
I I h
d, 0lIl> I 4&
U 1. 2 __ -"""'-J.-"'....-"k----".,-+',,-<--__+---.--+_______i I d,l h = 0.10
1.0 0; 50
;;:.u

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(j)
Mcl
bh2f
cd

U 1.2 ..... ---+----+__---j I d,lh = 0.15
1.0 .. --.-t----+----'-1 ___ 0; 50
;;:.U
O. 8 -1'--+----+_---1
I
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(k)
Mi
bh2f
cd

I
1.8 +----+----1----+-.-+.---1 d
, <II 1& lID I
1.6 .... I_-.-__+---+_______i I. h
<lie 4&
U 1.2 d,l h 0.20
1.0 -k-_ ...... __ __+-_.-.+.-.-+_-..,..-f-"'Ck'-;0;-5-0--.. ,_--I
;;:.u

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(I)
Mi
bh2f
cd
5.8.
(a)
u
,,-0

0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
MJbh
2
f
cd
; 1 ,0
<.u
(b)
0, 8
0,0
0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
MJbh
2
f
cd
2,0
1.8 d
,
n- Ih

1.4 d
,
II e (9
I' d/h=0,15
__ "8 f
ck
::::: 90
; 1 ,0
<.U
(e)
5.9.
0,05 0,1 0,15 0,2 0,25 0,3 0,35 0.4 0.45 0,5 0,55 0,6
Mo/
bh2f
cd
charts for """t"'j"",n,,,, columns (fck =
73
74

(d)
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd

(e)
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd

1 . 8
1.6 __ "", ... ", ... -+ ..
U
, .. .0
: 1.0 .. - .. -i-----+--'''''--''''''''-''-'''''r----I
2:,u

(f)

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd
5.9.
(g)
(h)

0.8
0.6
0.4
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd

i A/yibhfcd .-:ll+-- b --I
1.8 d
,
De ID I
1.6 h
1.4 d
,
GIl 1& II .-L
1.2 I d/h 0.05
,,-0 fCk = 90
1.0
(i)
5.9.

0.6
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Mi
bh2f
cd
75
o 0.05 0.1 O. i5 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(j)
MJbh
2
f
ed
2.0
1.8
1.6
1.4
1.2

.c:
1.0
:'2

0.8
0.6
0.4
0.2
0.0

---t 4- b--l

d
,
Tr-Ol> It -,

.. h
d,--.:L GIl e "

r 0.15

fc/{ = 90

i\\\
,\
'\
1\ \ \\
\ \ \ ",\\1
1/
'))
I) \ 1\ \ \ \\ 1\

//
// // '//
')
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(k)
MJbh
2
f
ed
2.0
1.8
1.6
1.4
1.2

.c:
1.0
:'2

0.8
0.6
0.4
0.2
0.0

b--l
d'::[J
T

h
d, 411

r =0.20

fek = 90

0[\\\\
\\
'\
\
,\ \ \
i\\\
/
()
I !
1\ \ \
\ \ \\

v//
//
))A )
o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
(I)
MJbh
2
f
ed
.5.9. (Contd)
76
1.8
AJyi
h2f
ed
1.6
7.0
1)
1.4
08
h d ~ ,
1.2
~ /
..........
~ - g 1.0 d/h = 0.9
::<:: fek:S; 50
?U 0.8
0.6
0.4
0.2
0.0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(a)
Mi
bh3f
ed
1.8
1.6
1)
1.4 h d ~ ,
1.2
~ .. / .... --
~ 1 l 1.0 d/h= 0.8
~
fek:S; 50
?U 0.8
0.6
0.4
0.2
0.0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(b)
Mi
bh3f
ed
1.8
AJyi
h2f
ed
1)
1.6
1.4
h d " ,.
1.2 ~ ... /
.... --
~ " 8 1.0
d/h= 0.7
::<::
fek:S; 50
?U 0.8
0.6
0.4
0.2
0.0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
(c)
Mi
bh3f
ed
5.10. charts for circular columns
77
78

1.6 _ ---I-----r
1.4
1.2
0.4
0.2
o 0.05 0.1 0.15
(d)
5.10. (Contd)
5.3: rectangular ..... ,1 .......... se<:troin
0.2 0.25
dlh = 0.6
f
ek
,; 50
0.3 0.35
'''','';Ul1'.<''<1.1 column section 500 mm and 300 mm wide is to be to carry a
ultimate axial load of 1875 kN and a ultimate moment of 280 kN m. The
characteristic strem'!th of the reinforcement is 500
The -drmenSlonal moment and
= 17
::::435
x x 500 x
x x
load
0.735
0.22
and that of the concrete 30
are calculated as follows:
h'JIULULUF, that the centroid of the steel in each face is 50 mm from the face and that the
reinforcement is concentrated at the corners of the 5.8b can be used to
establish that the mechanical reinforcement ratio is OA. This a total steel area of
OAO x 17 x 500 x =2345
Design for biaxial niPnl'lll ... O'
EN 1992-1-1 does not directly give a method for designing biaxially bent columns other than
working from first principles. This is not easy to do without design charts or a suitable
computer program. In principle, the development of a suitable computer program is relatively
simple, and the necessary assumptions are illustrated in Fig. 5.11. To develop the most
rigorous program, it is convenient to divide the compression zone into strips parallel with the
neutral axis and then calculate the stress in each strip using the parabolic-rectangular
diagram. The force and the moment about the x and y axes of each at the ultimate limit
state can then be calculated and summed to find the moment and axial force developed at the
ultimate limit state by the concrete in compression. Such a program can be used to generate
design charts for biaxial bending, if desired.
In this section, a number of simplified methods will be considered which will allow the
design of rectangular sections to be carried out relatively
Three methods of dealing with biaxially bent here.
The methods are in order of ,U'-"'-'VWHJl"-
5.1 i. Biaxial bending of a rectangular section
Table 5.4. Values of the exponent a in equation (05.15)
NdNRd
a
S; 0.1 1.00
0.7 1.5
1.0 2.0
The first is an approach given in clause 5.8. 9(4) of EN 1992-1-1 for the design of biaxially
bent sections in slender columns. This is based on the observation that the form of the Mx -My
interaction diagram can conveniently be represented by a super-ellipse, and is believed to
have been first developed by Bresler. A super-ellipse has an equation of the form
Jt1+ya=k (DS.14)
If a = 2, this equation becomes a circle, while if a 1 it describes a straight line. At loads
approaching the squash load, the Mx-My interaction diagram approaches a circle, while in
the region of the balance point it is close to a straight line. Clause 5.8.9(4) adopts the
equation below as a means of describing the complete interaction surface:
(DS.lS)
A convenient parameter for defining the proximity to the squash load is the ratio N/Nuz'
and EN 1992-1-1 assumes the relationship between this parameter and the exponent a given
in Table 5.4. Intermediate values may be interpolated.
In Table 5.4, N
Rd
is the squash load of the column, and may be calculated from
N
Rd
The the approach from the of view is that it cannot be used as
a direct design method since N Rd can only be established once the reinforcement area has
been found. It therefore has to be used An initial estimate is made of N/N uz' the
Clause 5.B.9(4)
Clause 5.B. 9(4)
79
80
section is designed, a corrected value of N/N uz can then be and the process
repeated until a correct solution is obtained.
It is not clear why the method described above is included only in the code for the design of
slender columns, as it was originally devised for general use, and is considered to be
adequately accurate for column sections where the reinforcement is concentrated near the
corners. If the reinforcement is more uniformly distributed around the perimeter of the
section, then the method is conservative.
A much simpler, but considerably more approximate, method has been adopted in
BS 8110. This is a slightly modified version of a method given in eEB Bulletin d'Information
141.8 Design is carried out for an increased uniaxial moment, which takes account of the
biaxial effects. The required uniaxial moment is obtained from whichever is appropriate of
the two relationships set out below:
(D5.I6)
if Mjh' < M/b' then M; = My + (3b'Mx/h' (D5.17)
In the above relationships, and My are the design moments about the x and y axes,
respectively, while M ~ and are the effective uniaxial moments for which the section is
actually designed. b' and h' are effective depths as indicated in Fig. 5.12. The factor (3 is defined
in BS 8110 as a function of N/bhfcu' In terms offck' it can be obtained from the relationship
(3 = 1 - N/bhfck (0.3 < (3 < 1.0) (D5.I8)
This approach has the great advantage of being very simple. It is, however, an approximate
approach, and it effectively defines the interaction diagram illustrated in Fig. 5.13. A
possible way to use this simplified method is to use it to obtain a first approximation to the
required area of reinforcement and then check the result using equation (D5.15).
y
L. b ~
.+ .
I
. + .
.<:::
X
:l:::
I
.+.
b'
y
S.12. Notation used in biaxial column design method
fig. 5.13. Comparison of simplified and rigorous biaxial methods
bent column section
rectarlgular column section of 500 mm x 400 mm is to be to
sustain a axial load of 2000 kN combined with moments of 100 kN m about
the minor axis and 234 kN m about the axis. The centres of the may
be assumed to be 50 mm from the shorter faces of the section and 60 mm from the longer
faces. This = 0.125 and d'/b 0.2. The concrete is C50/60, and the
reinforcement is 460 . The reinforcement assumed to be concentrated
dose to the corners of the ,,""UUJtl.
The area of reinforcement will be "ct,o,.,UlctlC'U
with the more exact has been carried out using
a total area ofreinforcement 3216 . This value
cOluparc;(\with the results obtained the to
their accuracy.
The method from BS will be tried first.
The first is to assess about which axis the effective uniaxial moment should be
"nn,hpfi h'::::: 400 - 50::::: 350 and b
l
::::: 300 60::::: 240. these Mih' = L.:J'+I_',HJ
0.67 and 0.417. > is carried outfor an effective
uniaxial moment about the
The coefficient ;3 is as
1 =1 =0.6666
The effective uniaxial moment is thus
x x
:::::
kNm
The
necessary to
0.2.
charts can now be used to obtain a steel area. It be seen that it be
The paI'ametenze
33L2x
between the charts for bars in the corners with d'/h 0.1 and
axial force and moment are
300 50
50x
=0.588
0.244
81
82
method may
earlier
reinforcement area it cou Id
in EN 1992-1-L This will now be done.
resistance for uniaxial to use the chart for
For 5.
This moment of
0.215 x 400 x
The factor a must now he found.
of the tahle in
For the section to be
+
the results:
+ + =
This very close to 1 that the reinforcement area
Prestrain
1
o
l
5.14. Effective steel stress-strain curve for
the methods
reinforcement
of nr .. sections
4794 kN
EN 1992-1-1
results

1nO __ '_._ of nr,p<;:1i"rll"cc,,,,,ri C.,.."'I"IIl'U"C
The design of prestressed sections introduces no major new problems except that allowance
must be made for the in the prestressing steel. This is effectively done by fixing a
false origin for the stress-strain curve for the prestressing steel at a point corresponding to
zero stress and a strain equal to the prestrain. This is illustrated in Fig. 5.14.
83
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
6
Shear, punching shear and
torsion
I. Shear - general
Shear is dealt with in clause 6.2 of EN 1992-1-1, torsion in clause 6.3 and punching shear in
clause 6.4.
Probably more research has been carried out into shear behaviour than into any other
mode of failure of reinforced or prestressed concrete. Despite this, there remain considerable
areas of uncertainty and disagreement. Furthermore, unlike flexural behaviour, there is no
generally accepted overall model describing shear behaviour. Despite these uncertainties,
design for shear can be carried out with confidence for an normal members because the
design methods given in codes have had the advantage of being tested against, and being
adjusted to fit, a very large body of experimental data. The first part of this chapter will give
an outline of shear behaviour as it is currently understood with an indication of how the
provisions in the code reflect this.
Background to the code provisions
Four aspects of the strength of sections subjected to shear loading will be considered here.
These are:
strength of members without shear reinforcement
strength of members with shear reinforcement
maximum shear that can be carried by a member
behaviour close to supports.
Each of these will be considered in turn.
6.2.1. Members without reinforcement
This area is probably of limited importance for beams where some shear reinforcement will
always be provided but is of major importance for slabs where it is often very inconvenient to
provide shear reinforcement. It is of particular importance for design for punching (see
Section 6.5). This has been the most researched area of shear performance, but, as yet, no
generally accepted theory describing the ultimate behaviour of a member without shear
reinforcement has been developed. The formulae given in codes should therefore be
considered to be basically empirical. Because of the amount of testing that has been carried
the effect of the major variables can be clearly established, and the resulting formulae
can be considered as reliable for normal types of element.
Clause 6.2
Clause 6.3
Clause 6.4
86
The principal variables governing the shear strength of members without shear reinforcement
are the concrete strength, the reinforcement ratio and the depth of the member.
It is fairly self-evident that concrete strength will have an influence on shear strength,
though the nature of the failure suggests that it is likely to be the tensile strength rather than
the compressive strength which is important. In fact, experimental evidence suggests that the
influence of concrete strength is rather less than proportional to the tensile strength.
EN 1992-1-1 gives the tensile strength as varying in proportion to the compressive strength to
the power of 2/3. The code indicates the shear strength of sections unrein forced in shear to
vary in proportion to the compressive strength to the power of 1/3. In this it agrees with UK
practice as incorporated in BS 8110.
Shear strength increases with increasing reinforcement ratio, but the rate of increase
reduces as reinforcement ratio increases. This behaviour may be modelled in various ways.
EN 1992-1-1, like BS 8110, employs a cube root relationship between shear strength and
steel percentage up to a maximum value of p of 0.02.
The absolute section depth is found to have a significant influence on shear strength over
and above that expected from normal geometrical scaling (i.e. there is a size effect). This
effect is sufficiently large that it is worth taking into account in the design of shallow
members such as slabs. Most recent codes of practice therefore have a term in their
equations to allow for this and which gives a higher shear strength for shallow members.
Taking these various factors into account, EN 1992-1-1 gives the following equation for
the strength of sections without shear reinforcement:
VRd,c = + 0.150"cp]
V
Rd
, c should not be taken as less than
+ 0.150"cp)
where
k = 1 + -V(200/d) ::; 2.0
d is in millimetres,
Pl =As/bwd ::; 0.02
(D6.1)
(D6.2)
and O"cp is the average longitudinal stress (note compression is positive). The tensile axial
force is simply taken into account by the introduction of a negative value for O"cp'
In equation (D6.1), the term 0.18hc and the constant 0.15 are nationally determined
parameters. The UK has, however, accepted the recommended values, and these are as
given above. For convenience, they have been included explicitly in the equation.
The minimum requirement for V
Rd
, c takes account of the fact that a member without any
reinforcement still has some shear strength.
6.2.2. Strength of sections with shear reinforcement
There is a generally accepted model for the prediction of the effects of shear reinforcement.
This is the 'truss model', illustrated in Fig. 6.1 for the commonest case where vertical
links are used. In this model, the top and bottom compression and tension members are,
respectively, the concrete in the compression zone and the tension steel. The members
connecting the top and bottom members are represented by steel tension members and
virtual concrete 'struts'. The truss in Fig. 6.1 can be analysed to give the forces in the various
members as indicated below:
Fl = N/2 + V(a - viz - 0.5 cot e)
= N/2 - V(a - viz + 0.5 cot e)
F2 = V/sin e
v
(D6.3)
(D6.4)
(D6.5)
(D6.6)
CHAPTER 6. SHEAR, PUNCHING SHEAR AND TORSION
5
z
2
4
N
e
3
v
fig. 6.1. Idealized truss
y
c
C _Ce_nt;_e __
x
I I I I I I I I I I
Fig. 6.2. Idealization of a beam with shear reinforcement as a 'smeared' truss
Two factors can be observed from this analysis. Firstly, the forces in the vertical tension
member 4 and the compression strut are both independent of the axial force, N. Secondly,
the forces in the compression and tension chords both differ from that calculated from the
moment alone by O.SV cot e. (Note that Va
v
= M, and the forces in the upper and lower
chords are given by bending theory as M/z. )
In developing the truss under discussion into a design approach for a reinforced or
prestressed concrete beam, it is necessary to consider it as a 'smeared' truss, as illustrated in
Fig. 6.2. In this idealized truss, the vertical stirrups are represented by a uniform vertical
tensile stress of Pwfyk or a vertical force of PwfYkbw per unit length of the beam. This assumes
that, at failure, the stirrups yield. The virtual compressive strut is replaced by a uniform
uniaxial compressive stress of (Yc acting parallel to the line of action of the strut over the
concrete between the centroid of the tension reinforcement and the centre of compression.
A complete analysis of this smeared truss may be carried out by considering the equilibrium
of two sections:
Vertkal equlibrium across a section parallel to the virtual compression strut. In this
section, aU the shear is supported by the shear reinforcement (section A-A in Fig. 6.2).
This gives
(D6.7)
87
88
e Vertical equilibrium across a section perpendicular to the strut (section B-B in Fig. 6.2).
Across this section, some of the shear is carried by the strut and some by the stirrups. The
total strut force is
Fe = o-ebwz/cos e (D6.8)
The horizontal projection of the section is given by z tan e, and hence equilibrium is
given by
V = pJ'yIiJwz tan e + Fe cos e
Since, from equation (D6.7),
PJ'YkbwZ = V tan e
we can rewrite this as
V = V tan 2 e + O-ebwz tan e
This can be rearranged to give
V = o-ebwz/( cot e + tan e) (D6.9)
These equations can be converted to design values by simply replacing the characteristic
material properties by the appropriate design values. O-e is replaced by (YcwVJed' The parameter
Vi is an efficiency factor which allows for the actual distribution of the stress within the strut
and the effects of cracking. (Yew is a coefficient taking account of any applied compression
force. Both these factors are Nationally Defined Parameters, and the recommended value
for Vi is v, given by
V = 0.6(1 - fek/250) (D6.10)
A higher value is permitted where the design stress in the shear reinforcement is less than
0.8f
y
k
(Yew takes a value of 1 for non-prestressed structures or structures subjected to tension. For
cases where an axial compression is applied, equations (6. 11. aN), (6.1l.bN) and (6. 11. eN) in
EN 1992-1-1 provide values.
Equation (D6.9) then gives the maximum shear that can be carried by a section before
failure by crushing of the notional compression struts.
Equations for inclined shear reinforcement can be derived in a similar way by replacing
the vertical tension by a uniform tension inclined at an angle to the horizontal.
The design forms of equations (D6.7) and (D6.9) appear in EN 1992-1-1 as equations (6.8)
and (6.9).
There are two areas where methods in various codes differ. The first is in the choice of a
value for the truss angle, e. The second is whether the truss should carry all the shear force or
whether part of the shear can be considered to be carried by the concrete.
Internationally, by far the most common approach is to assume that cot e is 1 (i.e. the
truss is assumed to form at 45 to the axis of the beam). When this assumption is made, it
is found that the shear capacity of a beam with shear reinforcement is underestimated, and
it is found that a generally consistent prediction of strength is achieved if the shear strength
is assumed to be equal to the shear capacity of the shear reinforcement calculated according
to equation (D6.9) plus the capacity of the concrete given by equation (D6.1) (or its
equivalent in whatever code is being considered). The shear response is thus assumed to be
plastic. Both BS 8110 and the ACI code follow this approach.
EN 1992-1-1 adopts a second approach. In this approach, all the shear is assumed to be
carried by the shear reinforcement, but the truss angle, e, can take any value between
cor
l
0.4 and coc
l
2.5. This variable strut angle approach is considered to be the more
rigorous ofthe two methods, and is also slightly more economical in some cases. However, as
written in EN 1992-1-1, it is open to a misunderstanding. The code implies that the designer
may select any strut angle between the specified limits. This concept of free choice does not,
~ Location of shear failure imposed by nature of loading
Fig. 6.3. Shear failure close to a support
however, reflect the behaviour of a beam. Beams will fail in a manner corresponding to a
strut angle of roughly cot -1 2.5 unless constrained by the detailing or the geometry of the system
to fail at some steeper angle. A steeper-angled failure could be induced either by the way in
which the tension steel was curtailed or where the load is so close to the support that only a
steeper failure can occur or where the shear strength is limited by the crushing strength of
the strut. This last consideration will be considered further after the crushing strength of the
strut has been discussed. Figure 6.3 illustrates this aspect of behaviour.
6.2.3. Maximum shear strength of a section
The previous section has been concerned with the behaviour of the tension members in
the truss. The capacity of these can be increased to whatever level is required simply
by increasing the amount of reinforcement provided. The capacity of the compression
members, the virtual diagonal struts, cannot, however, be so easily varied. It is the strength of
these struts that provides an absolute upper limit to the shear that can be supported by the
beam. The shear supported by these members has been derived above as
V = o-cbwz/(cot e + tan e)
At the ultimate load, the strut will crush, and the average stress in the strut, O-e' can be
expected to be proportional to the compressive strength of the concrete. This is achieved by
defining the limiting value of o-c as ifed' where v is an empirically obtained efficiency factor
which can be considered to take account of the actual distribution of stress across the section
at ultimate. Thus:
(D6.1l)
where V
Rd
, max is the maximum shear capacity of the section. EN 1992-1-1 states that z may
generally be assumed to be O.9d.
These equations are appropriate for vertical shear reinforcement.
It will be noted from equation (D6.7) that the smaller the angle e, the greater is the shear
capacity based on the shear reinforcement. However, the shear capacity based on the
crushing strength of the strut, given by equation (D6.11), decreases with decreasing values of
e below 45. Hence the maximum capacity corresponds to the situation where the capacity
based on the shear reinforcement just equals the capacity based on the strength of the strut.
This implies that the actual conditions at failure may be established by using equation
(D6.1l) to estimate the value of e for which V = V
Rd
max' and then using this value of e to
obtain the required amount of shear reinforcement. It should be noted that V
Rd
, max reaches a
maximum value when cot e = 1, and hence values of e greater than 45 will only occur if other
factors constrain the failure to occur at such an angle. The result of combining the strength
defined by the crushing strength of the strut and the limitations applied to the strut angle is
illustrated in Fig. 6.4.
89
90
Shear capacity enhancement near supports
There is very extensive test work that shows that much higher shear strengths than are given
by equation (D6.1) can be obtained in short members such as corbels or in beams where the
load is applied close to the support. The reason for this is simply that, for any sections closer
to the support than the critical section defined by the 'natural' strut angle of coc
1
2.5 (Fig.
6.5), a substantial proportion of the load will be carried through to the support directly by the
strut and not via the normal actions of shear and bending. The closer the load is to the
support, the greater the proportion of the load which will be transmitted to the support in
this way. Experimental evidence suggests that a convenient method of treating this problem
is to assume that the shear carried by this mechanism is a function of d/x, where x is the
distance from the face of the support to the face of the load (see Fig. 6.5). Figure 6.6 shows
experimental results for failures close to supports. EN 1992-1-1 deals with this problem by
increasing the design shear resistance of the concrete section, V
Rd
, c' according to the
following modified equation:
(D6.12)
EN 1992-1-1 does not mention the shear reinforcement that would be required in circumstances
where this was required at sections closer than 2d to the face of a support. Clearly, in such
cases the truss angle e is constrained so that cot e at the section considered is not greater than
x/O.9d.
A further point that should be noted is that, clearly, this enhancement can only be applied
where the load is applied to the top face of the beam and the support is at the bottom,
o
Maximum capacity with cot e = 1.0
~ Shear capacity governed by V Rd, max
(cot e varies from 2.5 to 1.0)
~ Shear capacity governed by cot e 2.5
..--- Minimum shear reinforcement (EN 1992-1-1, equation (9.4))
fig. 6.4. Relationship between the design shear force and the amount of shear reinforcement
,--
--------
,.-
/ ~ Trajectory of compressive force
I
1
6.5. Direct transfer of the load to the support
,.
'"
10

. .-
.,.
01

~ I
8 .1.
"
8v
,r
-+
1

' ..

Experiment
...
-- EC21ine
,
I
6
i

4

I
~
Ii

2
. *

1
,
OL-______ L-______ ~ ______ ~ ______ ~
o 2
avid
3 4
Fig. 6.6. Experimental results for shear strength near supports
and where the longitudinal reinforcement is properly anchored beyond the centreline of
the support.
6.2.5. Summary
This section has tried to give a brief picture of the phenomenon of shear failure and the
factors which influence it. This should help in understanding the provisions of EN 1992-1-1
and in applying them with confidence. It should be understood, however, that this picture is a
simplified one: shear is still not fully understood, and remains a very complex phenomenon.
6.3. Summary of the provisions in clause 6.2 of EN 1992-1-1
Tables 6.1-6.3 will enable the basic quantities used in the shear calculation to be read off for
most normal situations which will be met with for reinforced concrete.
Table 6.1 can be used to obtain basic values of V
Rd
ibwd for different values of longitudinal
reinforcement ratio and concrete grade. These values have been calculated ignoring the
factor k. To obtain the correct value for V
Rd
, c / b ~ , the values will need to be multiplied by the
91
k factor given in Table 6.2. Table 6.3 then gives the minimum value of V
Rd
, c1bvfl. For
members an axial force (i.e. a column or a prestressed beam), the value of VRd,c
obtained from Tables 6.1 and 6.3 are increased by adding a shear equal to O.15NI(Acbwd),
where N is the axial force andAc is the cross-sectional area of the element.
Where shear reinforcement in the form of vertical stirrups is to be provided, Fig. 6.7 may
be used to calculate the required amount. To use the chart, first estimate the value of
VEdlbvfl. The design relationship is then defined by two curves: (1) a straight line defined by
Table 6.1. Values of (O.18/rc)(1 OOpi/'::k) 1/3
longitudinal Concrete compressive strength, fck (N/mm2)
reinforcement
(%) 25 30 35 40 45 50 55 60 70 80 90
0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.27 0.29 0.30 0.31
0.4 0.26 0.27 0.29 0.30 0.31 0.33 0.34 0.35 0.36 0.38 0.40
0.6 0.30 0.31 0.33 0.35 0.36 0.37 0.38 0.40 0.42 0.44 0.45
0.8 0.33 0.35 0.36 0.38 0.40 0.41 0.42 0.44 0.46 0.48 0.50
1.0 0.35 0.37 0.39 0.41 0.43 0.44 0.46 0.47 0.49 0.52 0.54
1.2 0.37 0.40 0.42 0.44 0.45 0.47 0.48 0.50 0.53 0.55 0.57
1.4 0.39 0.42 0.44 0.46 0.48 0.49 0.51 0.53 0.55 0.58 0.60
1.6 0.41 0.44 0.46 0.48 0.50 0.52 0.53 0.55 0.58 0.60 0.63
1.8 0.43 0.45 0.48 0.50 0.52 0.54 0.56 0.57 0.60 0.63 0.65
2.0 0.44 0.47 0,49 0.52 0.54 0.56 0.57 0.59 0.62 0.65 0.68
Table 6.2. Values of depth factor, k
Effective depth (mm) k
100 2.00
200 2.00
300 1.82
400 1.71
500 1 ~ 3
600 1.58
700 1.53
800 1.50
900 1,47
1000 1.45
Table 6.3. Values ofv
min
Concrete compressive strength, fck (N/mm2)
Effective
depth (mm) 25 30 35 40 45 50 55 60 70 80 90
100 0.67 0.73 0.79 0.85 0.90 0.95 0.99 1.04 1.12 1.20 1.27
200 0.35 0.38 0.41 0,44 0.47 0,49 0.52 0.54 0.59 0.63 0.66
300 0.27 0.30 0.32 0.34 0.36 0.38 0.40 0.42 0.45 0.48 0.51
400 0.24 0.26 0.28 0.30 0.32 0.33 0.35 0.37 0.40 0,42 0.45
500 0.22 0.24 0.26 0.28 0.29 0.31 0.33 0.34 0.37 0.39 0.42
600 0.21 0.23 0.25 0.26 0.28 0.30 0.31 0.32 0.35 0.37 0.40
700 0.20 0.22 0.24 0.26 0.27 0.29 0.30 0.31 0.34 0.36 0.38
800 0.20 0.22 0.23 0.25 0.26 0.28 0.29 0.30 0.33 0.35 0.37
900 0.19 0.21 0.23 0.24 0.26 0.27 0.29 0.30 0.32 0.35 0.37
1000 0.19 0.21 0.23 0.24 0.26 0.27 0.28 0.30 0.32 0.34 0.36
CHAPTER 6. PUNCHING SHEAR AND TORSION
10.00
9.00
8.00
7.00
6.00
l:l
~
5.00
::::."'
4.00
3.00
2.00
1.00
0.00
0.00 2.00
cot 8 = 2.5 2.2 2.0 1.8 1.6
/
/
/
/
/
'--.,.k--,-/-:, ~ 1 3 5
/ ,// 30
y/ f
ck
= 25
4.00 6.00 8.00
Fig. 6.7. Design chart for the selection of shear reinforcement
/
/
55
1.4
10.00
1.2
1.0
12.00
the maximum permissible value of cot e (cot e 2.5) and (2) a curve defined by the grade of
concrete. The broken lines on the chart give the values of cot e. These may be useful when
detailing the longitudinal reinforcement, as the curtailment depends to some degree on the
value of cot e. The right-hand broken line, defining the end of the curves, corresponds to
cot e == 1 and gives the maximum amount of vertical shear reinforcement that may be used.
Minimum areas of shear reinforcement are not marked on Fig. 6.7, as this would overcomplicate
the chart; however, values can be obtained from equation (9.5) in EN 1992-1-1:
(AswfYWk/bws)min == 0.08-v fck (D6.13)
In selecting the critical section for shear design, it should be noted that clause 6.2.2(3) in
EN 1992-1-1 indicates that the shear resistance need not be checked closer to the support
than the point where a line drawn at 45 from the inner face of the support intersects the
elastic centroidal axis of the member. This takes account of the increased shear resistance
near the supports. For concentrated loads, this effect is accounted for by the shear force
reduction factor given earlier.
It should be noted again that all the equations derived and presented in this chapter deal
only with the case of vertical shear reinforcement. No new principles are involved in
extending the derivations to the case of inclined shear reinforcement, it simply complicates
the equations and makes the derivation less clear. Furthermore, from the practical point of
view, the use of inclined shear reinforcement is becoming increasingly rare due to the
considerably increased expense of the reinforcement fixing and difficulties in maintaining
tolerances. Should inclined shear reinforcement be needed, the necessary equations are an
presented in EN 1992-1-1 itself.
The best way to illustrate the design procedure is by means of examples.
The shear resistance of a section without shear reinforcement is given by
V
Rd
c == [(0.18hJk(lOOpJck)1/3 + O.150'cp]b,.d ~ (v min + 0.15CT
cp
)b,.d
<::Ht,nf'.rtp,tl T beam illustrated in 6.8 is to be for The beam
and reinforcement with a characteristic ",uvu,,,-cu
for flexure has resulted in a for 2200
Clause 6.2:2(3)
93
94
1*-----750 -----.I
r--------'l III
-L 315 l
Reinforcement: fyf< = 500 N/mm2
Concrete: fek = 30 N/mm2
: =: __ -_--_-_-_-_-_-_-__ -_-_-_-_J_,_4.1f_

Dimensions in mm
Fig. 6.8. Shear design example
longitudinal reinforcement. This is provided by 2 x No. 32 mm and 2 x No. 20 mm bars,
giving an area of 2236 mm
2
The effective depth of the section is 325 mm, and the depth
from the top of the flange to the elastic centroid of the section is 125 mm.
The shear force at the supports is given by
(1.35 x 40 + 1.5 x 20) x 5/2 ::::: 210 kN
However, the design shear may be taken as that at a distance from the face of the support
equal to the height from the bottom of the beam to the elastic centroid of the section from
the face of the support. Assuming that the distance from the support centreline to the face
of the support is 75 mm, the critical design shear is that at a distance of
75 + 400 - 125 = 350 mm
from the support centreline. This shear can be calculated as
::::: 210 x (5000/2 - 350)/(5000/2) = 180.6 kN
V
Rd
, c is now calculated. The reinforcement ratio is given by
Asibvfl ::::: 2236/150/325 = 4.59%
However, the maximum value that may be used in assessing V
Rd
, cis 2%. From Table 6.1,
the basic value of V
Rd
Jbvfl is 0.47 N/mm2, and from Table 6.2, the correction factor for
depth is 1.78. Hence, '
V
Rd
, c::::: 0.44 x 1.78 x 150 x 325/1000:::: 40.8 kN
This is much less than V
Ed
, and hence designed shear reinforcement is needed:
V Ed1bvfl ::::: 180.6 x 1000/150/325::::: 3.69
Figure 6.7 gives the required value ofAswfYWk1b,.p as 2.5. Hence,
Asv'!s :::: 2.5 x 150/500 =: 0.75
The maximum spacing of links according to equation (9.6) in EN 1992-1-1 is
0.75d:::: 0.75 x 325 = 244 mm
The most economical solution which will not violate the maximum stirrup spacing rules
would thus seem to be to use 10 mm diameter stirrups at 200 mm centres.
CHAPTER 6. PUNCHING SHEAR AND TORSiON
The minimum
0.08 x -V30 == 0.438
giving (Asw/s )min as
0.438 x 150/500 = 0.131
value is
Assuming that a bar diameter less than 8 mm will not be used, and recalling that the
maximum bar spacing is 244 mm suggests that minimum links have to be 8 mm links at
225 mm centres, giving A,Js = 0.447. The shear capacity of this choice of minimum
stirrups can be calculated from equation (D6.8) (noting that, in this case, cot Bwill be 2.5)
as
V
Rd
s
== 0.447 x 0.9 x 325 x 500 x 2.5/1.15/1000 kN = 142 kN
This indicates that minimum stirrups will be adequate over the central 3.4 m of the span,
with the heavier designed shear reinforcement near the ends.
6.4. Torsion
6.4.1. Introduction
In normal building structures, torsion generally arises as a secondary effect, and specific
calculations are not necessary. Torsional cracking is generally adequately controlled by
reinforcement provided to resist shear. Even when torsion occurs, it rarely controls the basic
sizing of members, and torsion design is often a check calculation after the members have
been designed for flexure. Also in some of these cases, the loading that causes the maximum
torsional moments may not be the same that induces the maximum flexural effects. In some
cases, reinforcement provided for flexure and other forces may prove adequate to resist
torsion.
It is important to recognize two basic types of torsion:
Equilibrium torsion, which is essential for the basic stability of the element or structure,
e.g. a canopy cantilevering off an edge beam (Fig. 6.9a). Equilibrium torsion is clearly a
fundamental design effect, which cannot be ignored. Structures in which torsion design
is carried out to include beams curved on plan, helical staircases and box girders.
Compatibility torsion, which arises in monolithic construction when compatibility of
deformation of the connected parts have to be maintained, e.g. edge beams in a normal
framed building with slabs and beams framed in on one side only. In these cases
the torsional restraint can be released without causing a collapse of the structure;
however, at serviceability, cracking may occur in the absence of sufficient reinforcement
(Fig. 6.9b).
6.4.2. Evaluation of torsional moments
When torsion has to be taken into account in design of framed structures, the torsional
moments may be obtained on the basis of an elastic analysis using torsional rigidity (GC)
calculated as follows:
G = O.42Ecm
C = 0.5 x St Venant torsional stiffness for a plain concrete section
(Note: 0.5 has been justified by test results.)
For a rectangular section, the St Venant torsional stiffness is k
1
b
3
d, where kl is obtained
from Table 6.4, b is the overall breadth d) and d is the overall depth.
Non-rectangular sections should be divided into a series of rectangles, and their torsional
stiffness should be summed up. The division should be carried out such as to maximize the
stiffness.
95
96
6.4.3. Verification
In EN 1992-1-1, the torsional resistance is calculated on the basis of a thin-walled closed
cross-section. Solid sections are replaced by equivalent thin-waned sections. The wall
thickness is taken as A/u, where A is the total area of cross-section within the outer
circumference including inner hollow areas and u is the outer circumference of the cross-
section. The effective thickness is to be not less than twice the distance between the outer
edge and the centreline of the longitudinal reinforcement. In hollow sections; the effective

ITT" I I Il'T1i ''''''IIIYTTWTrIlI
rTT
UiTnIlT1'"1IIT'IIII"""IInTl W""'III
m1
1 W
MT
!l
TTTl
III4
... ::.:.::.:: ..
I
(al
(b)
Fig. 6.9. Examples of (a) primary or equilibrium torsion and (b) torsion in statically indeterminate
structures
Table 6.4. Values of kl for a rectangular section
d/b k,
I 0.14
1.5 0.20
2 0.23
3 0.26
5 0.26
>5 0.33
Effective cross-section
Coverc
Centre line
Circumference U
* - ' ~ ~ - - - Circumference Uk
Fig. 6.10. Notation for torsion
thickness is not to be taken as greater than the actual wall thickness. Figure 6.10 defines the
relevant parameters.
The procedure for verification may be summarized as follows:
(1) Choose an angle e for the struts such that the following relationship is satisfied:
TEd = TRd,max = 2vfcrf'1lef sin e cos e
in which TEd is the applied torsion and v = 1 - (fck/250); Ak is the area enclosed by the
centre lines of the connecting walls including the inner hollow areas, cot e should lie in
the range 1.0 and 2.5. (Note that the above expression has to be multiplied by a factor OOew
if axial stresses exist. See EN 1992-1-1 for details.)
(2) The transverse reinforcement required should be calculated using the following expression:
Asw/s = T
Ed
/2AJyw,d cot e
where tyw, d is the design strength of the transverse reinforcement.
(3) The longitudinal reinforcement required to resist torsion is given by
ASl = TEdU
k
cot e/2AJ
y
d
in which tyd is the design strength of the longitudinal reinforcement and Uk is the
perimeter of areaA
k

(4) In a member subject to torsion and shear, the following inequality should be satisfied:
TEd/T
Rd
, max + VEd/V
Rd
, max::; 1.0
in which V
Ed
is the applied shear force and V
Rd
max is the maximum shear resistance of the
cross-section in accordance with equation (D'6.11).
(5) Only minimum reinforcement is required when
TEd/TRd,e + VEctfVRd,c::; 1.0
in which T Rd, e = 2Aledet, d and V
Rd
, e is as defined in equation (D6.1).
97
Clause 6.4
98
6.5. Punching shear
6.5.1. General
Punching shear is a local shear failure around a concentrated load on a slab. The most
common situations where punching shear has to be considered is the region immediately
surrounding a column in a flat slab or where a column is supported on a pad footing or
foundation raft. Design for punching shear is covered in clause 6.4 of EN 1992-1-1.
Punching shear failures may be considered to be shear failures rotated around the loaded
area so as to give a failure surface which has the form of a truncated cone. This is illustrated
in Fig. 6.11. The 'critical section' for shear failure in a beam is transformed into a 'basic
control perimeter' when punching shear is considered. This conversion of the problem from
a basically two-dimensional one to a three-dimensional problem does not change the basic
phenomenon as deseribed in Section 6.2, though there are a number of practical points
which need further consideration. The first point which will be considered is the definition of
the control perimeter.
6.5.2. Bask control perimeter
The starting point for design for punching shear is the definition of the critical perimeter.
This has greater importance than the selection of the critical section in a beam because, as
perimeters closer to the loaded area are considered, the length of the perimeter rapidly gets
shorter, and hence the shear force per unit length of the perimeter rapidly increases.
Observation of failures shows that the outer perimeter of a punching failure takes the
general form sketched in Fig. 6.12a. For this reason, EN 1992-1-1 proposes the idealized
form shown in Fig. 6.lOb. Some national codes (e.g. the UK code, BS 8110) employ a
rectangular perimeter. This is less realistic but slightly easier to use. Another advantage of
the rectangular perimeter is that, when reinforcement is needed, this must be provided
within the perimeter. It is much easier to do this with a rectangular perimeter since
reinforcement is commonly provided on a rectangular grid. In fact, as will be seen below,
6.1 I. Schematic illustration of a punching shear failure
kd
(a) (b)
Fig. 6.12. Rationalisation of failure perimeter
since the design equations for punching shear are basically empirical, the precise choice of
the shape of the perimeter in a particular code is not too important.
Having chosen a basic form for the perimeter, it is next necessary to consider the distance
from the loaded area at which it should be located. Since the treatment of shear is empirical,
there is no absolutely defined answer to this question. The same strength could be obtained
by using a high value for the shear strength eombined with a short perimeter close to a
column or by using a lower shear strength combined with a longer perimeter further from the
column. Different eodes have adopted different solutions to this question; the UK code
takes a perimeter l.5d from the face of the column or loaded area while the US ACI code
uses a perimeter O.5d from the face of the column but with a much higher stress. In drafting
EN 1992-1-1, it was decided that the shear strength should be given by the same formula for
punching shear as is used for shear in beams. Having decided on this and on the shape of the
perimeter, the distance of the perimeter from the loaded area can be established from test
data. This led to a value of 2d. The interpretation of this to define the actual perimeter is
dealt with very thoroughly in clauses 6.4.2(1)-6.4.2(6) of EN 1992-1-1, and needs no further
discussion here.
A question which does require some further clarification is the definition of the control
perimeter in cases where there is a 'drop' or 'column head' around the column. This is
covered in EN 1992-1-1 in clauses 6.4.2(8) and 6.4.2(9)] and Figs 6.17 and 6.18, but is not easy
to follow. Although the code does not explicitly say so, it appears that, where there is a drop,
a circular control perimeter is assumed with a radius r
cont
The provisions in clauses 6.4.2(8)
and 6.4.2(9) are concerned with defining this radius. The flow chart in Fig. 6.13 is intended to
clarify the interpretation of these provisions.
6.5.3. Design shear force
It is usual to assume in design that the distribution of shear force around the critical
perimeter is uniform. In fact, this is untrue, particularly in the case of a slab-column
connection where there is a moment transfer between the slab and the column. In such cases,
a rigorous analysis would show that the distribution of shear varied markedly around the
perimeter and was accompanied by torsional moments. Extensive experimental work shows
that punching shear strength can be significantly reduced where substantial moment transfer
occurs. This implies that punching shear is not an entirely plastic phenomenon, and the shear
at the ultimate limit state cannot be fully redistributed around the perimeter. A way of
dealing with this in design is to increase the design shear force by a factor which is a function
of the geometry of the critical perimeter and the moment transferred. The provisions in
EN 1992-1-1 introduce a multiplier, (3, to increase the average shear stress around the
perimeter such that
(D6.14)
where V
Ed
is the design ultimate shear stress on the perimeter considered, V
Ed
is the design
Clauses
6.4.2( 1)-
6.4.2(6)
Clause 6.4.2(8)
Clause 6.4.2(9)
99
100
ultimate shear force on the perimeter considered, u
i
is the length of the control perimeter
considered, d is the mean effective depth of the slab at the control perimeter, and f3 is the
shear multiplier, taking account of moment transfer.
The principle behind the definition of f3 is that the effect of transferring a moment
between the slab and a column can be modelled by considering a distribution of shear around
the control perimeter considered such that it provides a moment equal to the moment
transferred. The distribution of shear assumed is sketched in Fig. 6.14a, which is redrawn
after Fig. 6.19 in EN 1992-1-1. It will be seen that the magnitude of the distributed shear
force is a function of the moment transferred, the distance of the perimeter from the loaded
area and the shape of the loaded area. It will be seen that
f3V
Ed
lu
i
= VEdlu
i
+ Llv
hence
(D6.15)
In principle, Llv can be calculated from simple statics, since the moment transferred
between the slab and the column, LlM
Ed
, must be equal to the moment produced by the shear
Llv distributed around the perimeter, u
i
Inspection of Fig. 6.12b shows that this may be
written as
LlM
Ed
= 4 Llv[c/2 x c/4 + c
2
(c
j
/2 + 2d) + 2d1l/2(c/2 + 2d x 2111)]
= Llv(c/12 + c
j
c
2
+ 4c
2
d + 16d
2
+ 211dc j) = Wj Llv
Hence,
d
lJrf
H H
Control perimeter
/
/
I
\
,
\
\
\
~ / - - - - - - - - - - -
/
I ..
,-
,
\
~ I
--------:1 \
, '
: 1, 1
, '
-- ----...! /
/
/
/
I
Is
head circular?
No
r
cont
lesser of:
(1) 2d + 0.56--11, /
2
(2) 2d + 0.69/,
Is
there a column
head?
>-----No
Yes
Check two sections:
rcont 101 = IH + 2d + d2
r
cont
. e,t = 2( d + I
H
) + d2
Control
perimeter as
clauses
6.4.2( 1 )-6.4.2(6)
Fig. 6.1 J. Interpretation of clauses 6.4.2(8)-6.4.2(11) in EN 1992-1-1 for establishing the location
of the basic control perimeter where column heads are used
rD
c, + 4d --- ----
1 ............ .
(c)
fig. 6.14. Assumed shear distribution to resist moment transfer
Table 6.5. Values for the aspect ratio factor, k
C/C
2
:0; 0.5
1.0
2.0
;::: 3.0
Substituting for .6.v into the equation for (3 gives
(3= 1 + .6.MEdU/WjVEd
k
0.45
0.60
0.70
0.80
It is found that a further correction factor is required to adjust the value of (3 for the
aspect ratio of the column section. The full equation in the code is thus
(D6.16)
The coefficient k is given as a function of the aspect ratio of the loaded area in Table 6.1 in
EN 1992-1-1 (repeated here for convenience as Table 6.5). In the table, C
j
is the dimension of
the cross section perpendicular to the axis of bending, while c
2
is the dimension parallel to
the axis of bending.
It should be noticed that .6.M
Ed
/V
Ed
is actually the eccentricity of the concentrated load
relative to the centroid of the loaded area. In some equations, EN 1992-1-1 uses 'e' for this
rather than .6.M
Ed
/V
Ed
It should also be noted that EN 1992-1-1 uses the term M
Ed
for the
moment transferred between the slab and the column. This seems a possible cause of
confusion, since the moment transferred is not the moment in the slab but rather the
difference in moment between one side of the column and the other. For this reason, .6.M
Ed
has been used here for the moment transferred.
The derivation of the equation for (3 for a rectangular internal column has been derived in
order to show the basic logic of the system. There are many other possible configurations of
column, and EN 1992-1-1 occupies three pages setting out the appropriate values. For
convenience, these are set out in Table 6.6.
For structures where the lateral stability does not depend on the frame action of the slabs
and columns, EN 1992-1-1 permits the use of simplified values for (3. These values are
(3 = 1.15 for interior columns, (3 1.4 for edge columns and (3 = 1.5 for corner columns. These
values are Nationally Determined Parameters.
101
DESIGNERS' GUIDE TO EN 1992-1-1 AND EN 1992-1,,2
102
Table 6.6. Values of the shear enhancement factor for various types of column
Case
Internal rectangular column. uniaxial bending
Internal rectangular column. biaxial bending
Rectangular edge column; axis of bending
parallel to the slab edge. eccentricity is
towards the interior
Rectangular edge column; bending about both
axes. Eccentricity perpendicular to the slab
edge is towards the exterior
Rectangular edge column; bending about both
axes. Eccentricity perpendicular to the slab
edge is towards the interior
Rectangular corner column. eccentricity is
towards the interior
Rectangular corner column, eccentricity is
towards the exterior
Interior circular column
Circular edge or corner columns
of
Value for (3
(3 I + k(L3.MdVEd)(U/WI)
WI = c/!2 + C
I
C
2
+ 4c
2
d + 16d
2
+ 2Tidc
I
Values of k from Table 6.4
I + 1.8-V{[L3.M
Edy
l(c
z
+ 4d)f + [L3.MEd)(C
y
+ 4d)]2}
L3.M
Edy
and L3.M
Edz
are respectively the moments
transferred in the y and z directions while c
y
and C
z
are respectively the section dimensions in the y
and z directions
(3 = u / u ~ (i.e. shear is assumed uniformly
distributed over perimeter u ~ as defined in Fig.
6.200 of EN 1992-1-1)
(3 = I + k(L3.M
Ed
/VEd)(U/W
I
)
WI calculated by taking moments about the
centroid of the control perimeter. u
l
(3 = U/U'I + k(L3.M
Ed
. par/VEd)(u/W
I
)
WI = c/14 + C
I
C
2
+ 4c
l
d + 8d
2
+ Tidc
2
L3.M
Ed
. par is the moment transfer about an axis
perpendicular to the slab edge. k is determined
from Table 6.4 with c/c
2
put equal to 0.5(c/c
2
)
c
i
is the section dimension perpendicular to the
slab edge; c
2
is the dimension parallel to the slab
edge
(3 = u / u ~ (i.e. punching force is considered
uniformly distributed along perimeter u ~ in Fig.
6.20b of EN 1992-1-1)
,8 = I + k(L3.M
Ed
/VEd)(U/W
1
)
WI = c
l
2
/2 + C
I
C
2
+ 4c
2
d + 16d
2
+ 2Tidc
I
(3 = I + O.6Ti(L3.M
Ed
/VEd)/(D + 4d)
D is the diameter of the column
No information given
Provided that V
Ed
::; V
Rd
, c' no shear reinforcement is required. V
Rd
, c is given by
V
Rd
c = (0. 18hJk(100pdck)1!3 + O.la
cp
V
Rd
, c should not be taken as less than
O.035k3/2fck1!2 + O.la
cp
In equation (D6.17),
k = 1 + -Y(2001d) ::; 2.0
where d is in millimetres,
PI = -Y(PlyPlZ) ::; 0.02
(D6.17)
(D6.18)
where Ply and Pl
z
are, the reinforcement ratios in the y and z directions
calculated for the reinforcement within a width
each side. a
cp
is the average stress
These expresions differ from those for beams unreinforced in shcar by the necessity
to obtain average values for PI and (Tcp and the reduction of the coefficient applied to the axial
stress (Tcp from 0.15 to 0.1.
6.S.5. shear
The punching strength of a slab with shear reinforcement is given for vertieal shear
reinforcement by
(D6.19)
where is the area of one perimeter of shear reinforcement around the column, d is the
average effective depth of the flexural reinforcement (in millimetres),
fywd,ef = 250 + 0.25d -:;.fywd N/mm2
(d in millimetres) and sr is the radial spacing of perimeters of shear reinforcement (in
millimetres)
It will be noted that thc approach to the design of shear reinforcement is in principle
different to that used in beams where the concrete is assumed not to contribute to the shear
strength of a section reinforced in shear. a contribution from the concrete is assumed
but, unlike the gencral international consensus, the concrete contribution is reduced from
the full strength of the section without shear reinforcement while cot e is effectively assumed
to be 1.5 rather than the more common 1.0. The formula has to be seen to be basically
empirical, and one cannot avoid an impression that all the parameters have been changed
slightly from previous formulae just for thc sake of being different.
The concepts of a perimeter of shear reinforcement and the spacing of the perimeters
are awkward, and will have to be interpreted for each particular case. A perimeter of
reinforcement is presumably reinforcement provided on a perimeter parallel to the control
pcrimeter u
j
but lying inside it. Unfortunately, it is not clear that shear reinforcement
is actually provided in this way; indeed, Fig. 6.22 of EN 1992-1-1 makes this difficulty
clear. It is also unfortunate that the rules in clauses 6.4.5(4) and 9.4.4.3(1) appear to be
uninterpretable. The following is an attempt to develop a set of rules which can reasonably
be interpreted in practice.
Having found that the shear reinforcement is required on the control perimeter at a
distance of 2d from the face of the loaded area (first control perimeter), the control
perimeter beyond which shear reinforcement is no longer required is found from
(D6.20)
having calculated Uout ef' the distance of this perimeter from the loaded area may easily be
calculated. While f a i l ~ r e will not take place on this perimeter, it can occur on any perimeter
within this (even, in theory, 1 mm inside it). The failure is predicted to occur over a distance,
measured radially inwards from the perimeter, equal to 2d. Shear reinforcement for this
failure must, therefore, be within this zone between the perimeter and a parallel perimeter
2d inside it. Reinforcement close to either boundary of this zone is unlikely to be effective,
and therefore a is needed on either edge. On the inner edge, Fig. 9.12a of EN 1992-1-
1 indicates that reinforcement should not be closer than 0.3d to the loaded area, which
constitutes the inner boundary of the failure zone for the first control perimeter. The
maximum radial spacing of shear reinforcement is O.75d, and at least two sets of reinforcement
are required. This would suggest, for the first control that reinforcement could
reasonably be provided at distances of O.5d and 1.25d from the loaded area (column face),
leaving a gap of O. 75d between the outermost reinforcement and the first control perimeter.
It is suggested that this procedure is generalized into the fonowing rules.
Considering any possible perimeter, un' lying between u
out
, ef and the first control perimeter,
shear reinforcement should be arranged so that a total area of shear reinforcement equal
to tot lies within a zone between a perimeter situated O.3d and a perimeter situated
1.7d inside un' The radial of the reinforcement should not exceed 0.75d, and the
Clause 6.4.5(4)
Clause 9.4.4.3(1)
103
104
First control Outermost control
J I
perimeter, u, ~ perimeter, uoat,ett ~
S:O,75d
l
This zone should contain at least Asw, tot
fig. 6.15. Arrangement of punching reinforcement
, ,
, ,
,
, ,
~ - - - - - ~ ~
- -,- ---,..----,-
, ,
-'
,
circumferential spacing should not exceed 1.5d for the reinforcement for the first control
perimeter or 2d for subsequent perimeters. Asw, tot is given by
Asw, tot :2 und(v
Ed
O.75v
Rd
, c)/fywd, eff
This procedure is illustrated in Fig. 6.15.
Example 6.2: lightly loaded slab-column connection
Check the punching shear strength of the slab around an internal column supporting a
225 mm thick flat slab having 6 m spans in both directions. The column is 300 x 400 mm,
and the design shear force established from analysis of the slab, which has 6.5 m spans in
both directions and supports a design ultimate load of 9 kN/m2, is 400 kN. Design of
the slab for flexure gave an average value for the reinforcement ratio as 0.0077. The
characteristic concrete strength is 30 N/mm2. Bending is about the major axis of the
column only, and the moment transferred between the slab and the columns is 45 kN m.
Assuming 20 mm cover and 12 mm diameter bars give the average effective depth as
225 - 20 - 12 = 193 mm
Table 4.8 of EN 1992-1-1 gives the basic design shear strength as 0.34 N/mm2 for 30 N/mm2
concrete. The critical perimeter is
2(300 + 400) + 193 x 4 x 3.1416 = 3825 mm
To find the design effective shear stress it is now necessary to calculate (3. For an
internal rectangular column, Table 6.6 gives
Wj = c//2 + c
1
C
Z
+ 4czd + 16d
2
+ 2Tidc
1
For the slab and column dimensions considered, this gives
W
j
= 400
2
/2 + 400 x 300 + 4 x 300 x 193 + 16 x 193
2
+ 211 x 193 x 400 = 1512.6 x lW
c1ic
z
== 400/300 == 1.333
From Table 6.5, k = 0.63. This gives
(3 = 1 + 0.63 x 45 x 10
6
x 3825/(400 x 1000 x 1512.6 x 1000) = 1.17
The effective design shear stress is
CHAPTER 6. SHEAR, PUNCHiNG SHEAR AND TORSION
400 x 1000 x x 0.633
The depth
(D6.17) gives
k, for a slab with an effective depth less than 200 mm is 2.0. Equation
0.18/1.5 x 2.0 x (100 x 0.0077 x 30)1/3 0.684
This exceeds 0.633 N/mm2, hence no shear reinforcement is needed.
Example 6.3: heavily loaded slab-column connection requiring shear
reinforcement
The slab-column connection considered in Example 6.2 will now be designed for an
increased design shear force of 600 kN. All other factors will be assumed to be as in
Example 6.2.
In calculating vF.d' Wi and C
1
/c2 will remain as in Example 6.2, but ,6 will change because
the eccentricity of the load will reduce. ,6 can now be calculated as
('J::;::: 1 + 0.63 x 45 x 10
6
x 3825/(600 x 1000 x 1512.6 x 1000) = 1.12
vEd is now given by
V
Ed
= 600 x 1000 x 1.12/(3825 x 193) = 0.91 N/mm2
As before, = 0.684, thus punching shear reinforcement is required.
It is convenient at this stage to calculate U"uter to establish the extent of puncbing
reinforcement:
Uouter = 1.12 x 600 x 1000/(0.684 x 193) == 5090 mm
This can easily be calculated to be a perimeter at a distance
(5090 - 1400)/2'li" == 587 mm
or 3.04d from the column face.
The area of reinforcement required across the failure zone can be calculated for tbe
first control perimeter and for U"uter' At any point in between these, the required area can
be found by linear interpolation. For tbe first control perimeter, the total area of
reinforcement that must be provided within tbe perimeter is given by
As, tDt = 3825 x 193 x (0.91 - 0.75 x 0.684)/435 =::: 674 mm
2
At tbe outer perimeter, tbe total area is given by
As, tot == 5090 x 193 x (0.91 - 0.75 x 0.684)/435 = 896 mm
2
Figure 6.16 sbows tbe required steel area for perimeters within this area.
Assuming tbat the first perimeter of reinforcement is provided at 125 mm from the
column face and then at 150 mm centres as far as necessary, the lengtb of eacb perimeter
can be calculated, and, bence, from the maximum spacing rules, the minimum number of
bars whicb sbould be provided, For example, tbe lengtb of a perimeter 125 mm from tbe
column face is 2185 mm. Tbe maximum permissible spacing around tbe perimeter is
1.5d = 290 mm. To meet tbis requirement, 2185/290 = 8 bars are required. Trial and error
suggests that 8 mm diameter bars, whicb are tbe smallest bars generally available, sbould
be used on aU perimeters. Table 6.7 gives tbe steel areas supplied on each perimeter in
this case.
The areas in Table 6.7 together with the assumption that reinforcement is only effective
for a particular failure zone when it is at least 0.3d from tbe inner edge of the failure zone
permits the total reinforcement provided for any outer perimeter of a failure zone to be
calculated. This is plotted on Fig. 6.16. It will be seen that adequate reinforcement is
provided at all perimeters and that all the detailing rules are obeyed. It will also be
seen that the amount of reinforcement provided is defined by the minimum
105
circumferential and radial bar "fJ''''U''F,
to be
in many places, there are, unavoidably, three
failure zone in many situations.
of reinforcement within the
106
Table 6.7. Calculations for each perimeter of reinforcement in Example 6.3
Perimeter of reinforcement
Distance from column face (mm)
(Distance from column face)/d
Perimeter (mm)
Maximum spacing
No. of bars
Bar diameter
Area on perimeter
1800
r;-- 1600
125
0.65d
2185
290
8
8
402
E
.s 1400
Area of reinforcement provided
'"
~ 1200
~ 1000
E
<l>
e 800
0
C
600 '@
<a
400 '0
f-
200
Area of reinforcement required
7
0
0 100 200 300 400 500
Distance from column face (mm)
Fig. 6.16. Provision of shear reinforcement in Example 6.3
2
250
I.3d
2970
290
10
8
603
U{)uter
600 700
3
375
1.9d
3756
290
13
8
553
800
CHAPTER 7
Slender columns and beams
7.1. Scope
This chapter covers the material given in clause 5.8 of EN 1992-1-1. Also covered is the
material in Annex D, which provides supplementary information on global second-order
effects.
7.2. Background to design of columns for slenderness effects
When an element is subjected to an axial load combined with a moment, it will deflect. This
deflection will increase the moment at any section in the element by an amount equal to the
axial force multiplied by the deflection at that point. This extra moment will cause the
resistance of the element to be reduced below that calculated ignoring the deflections. In
many, if not most, practical situations, the effect of deflections is so small that it can be
ignored. Indeed, in many practical cases the effect of deflections on the design of a member
is not merely insignificant but actually nil. How this comes about will be explained later. A
slender member is one where the influence of deflections on the ultimate strength is
significant and must be taken into account in the design.
Slenderness effects in struts have been the subject of theoretical and experimental study
from the earliest days of the science of structural mechanics. Euler (1707-1783) seems to
have been the first person to study this problem mathematically, and the Euler equation for
the buckling strength of a strut must be well known to any engineer. The Euler equation
deals with the failure by instability of a strut made of infinitely strong, elastic material.
Failure occurs not due to failure of the material but because the deflection under the
critical buckling load becomes infinite. The problem that the designer faces in the design of
reinforced concrete members is very rarely one of classical buckling; failure of the material
occurs at a lower load than would be predicted if the deflections were ignored, due to the
moments being somewhat higher than predicted. The other critical difference from classical
buckling theory is that reinforced concrete is definitely not elastic at the ultimate limit
state.
In order to understand what the design rules given in EN 1992-1-1 are aiming to achieve, it
is necessary to have an idea of the forms of deflection which may occur. There are two modes
of deflection which are normally possible: sidesway of the whole structure, and deflection df
a single column without sidesway. These two possibilities are shown schematically in Fig. 7.1.
It will be seen that the two modes give very different forms of bending moment diagram in
the columns. In the sidesway mode it is important to note that all the columns within the
storey are subjected to the same deflection.
Code rules dealing with slenderness effects commonly include three basic steps:
Clause 5.8
108
(1) establish which modes of deflection are likely
(2) establish whether slenderness effects are likely to be significant (i.e. whether the structure
or member considered is slender)
(3) if the slenderness effects are likely to be significant, design to make allowance for them.
This is the approach adopted by EN 1992-1-1, and the basic principles will be introduced
here. Detailed guidance with use of the provisions will be given later.
In order to establish the deflection modes which should be considered, EN 1992-1-1 uses
the following classification system. Structures are classified as:
..

sway or non-sway and
braced or unbraced .
A sway structure is one where sidesway (global second-order effects) is likely to be
significant (as in Fig. 7.1a), while in non-sway structures, sidesway is unlikely to be significant.
'Significance' is defined in this context as a situation where lateral displacement of the
ends of the columns increase the critical bending moments by more than 10% above the
moment calculated ignoring the displacements. Basically, whether a structure is sway or non-
sway will depend upon its stiffness against lateral deformation. In most cases a structure
can be classified by inspection. Section 7.3.6 of this guide deals with this question in more
detail.
A braced structure is one which contains bracing elements. These are vertical elements
(usually walls) which are so stiff relative to the other vertical elements that they may be
assumed to attract all horizontal forces. In general, where the bracing elements in a braced
structure consist of shear walls or cores, braced structures may be assumed to be non-sway
without further checking.
..
Having classified the structure, the following can be said:
Non-sway structures: only the deformation of individual columns of the type illustrated
in Fig. 7.1b need be considered as, by definition, sidesway will be insignificant.
Sway structures:
Braced structures. As mentioned above, braced structures may normally be assumed
to be non-sway unless the bracing elements are relatively flexible. If they are, then
the bracing elements or structure should be analysed for the effects of sidesway but
( (
(a)
Bending moment
in columns
due to deflection
)
(b)
Bending moment
in column
due to deflection
Fig. 7. I. Modes of deflection of columns in a structure: (a) sidesway of whole structure; (b) deflection
of a single column
CHAPTER 7. SLENDER COLUMNS AND BEAMS
Notional point of
contraflexure ___ , _ ~ _
: ------ -----r-
/
':'\T I
Column
height
POints of: Effective
00""""'1 LI
(a) (b)
Sidesway +-
I
I
Effective
length
Fig. 1.2. The concept of effective length: (a) isolated column; (b) column with sidesway
the braced elements within the structure may be assumed to deform, as illustrated in
Fig. 7.1b, and sidesway may be ignored.
Unbraced structures. These structures should first be designed to take account of
sidesway of the whole structure as illustrated in Fig. 7.1a, and then each column in
turn should be analysed for the possibility of its deforming, as illustrated in Fig. 7.1b.
The second element in the code provisions is establishing whether slenderness effects are
likely to be significant or whether they can be ignored. EN 1992-1-1 defines significance as
being when slenderness effects increase the first-order moments by more than 10%. It is
useful to have this stated as a general principle, but, from the practical point of view, it is
unhelpful since it can only be established that the effect is less than 10% by doing a full
calculation while the objective is to establish when the calculation is unnecessary. Some
simplified rules which may be deemed to satisfy this requirement are therefore needed. As
with most other codes, EN 1992-1-1 formulates these rules in terms of two parameters:
..
..
the effective length of the column
the slenderness ratio .
Effective length is a concept used almost universally in the treatment of second-order
effects. The effective length can best be defined as the length of a pinned ended strut having
the same cross-section as the column considered and which would have the same buckling
strength. The concept is illustrated in Fig. 7.2. The effective length will be seen to depend
upon the type of deflection considered and the end fixity of the column. A non-sway column
will have an effective length which is between 0.5 and 1.0 times the actual clear height of the
column while a sway column will have an effective length greater than the clear height. Only
in very rare cases will the effective height exceed twice the overall height. Formulae can be
derived which will give the effective height as a function of the stiffnesses of the column and
the members framing into each end of the column. The following equations are given in
EN 1992-1-1 to establish the effective length:
(1) For isolated columns in non-sway frames (braced columns):
109
110
Table 1.1. Approximate values for effective length factors (effective length = factor x dear height)
End condition at bottom
End condition at top 2 3
Braced or non-sway columns
I 0.75 0.80 0.90
2 0.80 0.85 0.95
3 0.90 0.95 1.00
Sway columns
I 1.2 1.3 1.6
2 1.3 1.5 1.8
3 1.6 1.8
4 2.2
to = 0.51 (1 + k
j
)(1 + k2 )
0.45 + kl 0.45 + k2
(D7.1)
(2) For columns subjected to sway:
fo = 1 x the greater of a or b below:
a = 1+10 k
j
k
2
kj +k2
(D7.2)
b=(l+ 1:
1
kJ(1+ 1:
2
kJ
(D7.3)
where 10 is the effective length of the column, I is the clear height of the column between
restraints, k j is the ratio of the column stiffness to the stiffness of the upper joint and k2 is
the ratio of the column stiffness to the stiffness of the lower joint
These equations are time-consuming to use, and Table 7.1 (from BS 8110) provides a simple
and conservative means of estimating the effective length for columns in regular structures.
The values in the table are difficult to justify rigorously, and the interpretation of the clauses
in EN 1992-1-1 are, in any case, unclear. The lack of clarity results from the definition of k
j
and k
2
. k, which may be either kl or k2' is defined by
k (8/M) (El/I) (D7.4)
where 8 is the rotation of the joint at the end of the column considered for bending moment
M applied to the restraining members by thc buckling member or members. Elil is the
bending stiffness of the compression member or members considered to be buckling.
If there is a compression member above the joint at the upper end of the column
considered or below the joint at the lower end of the column considered then it is necessary
to consider whether the column is likely to contribute to the deflection or to restrain it.
If the column is likely to contribute to the deflection, then El/l should be replaced by
(El/I)a + (EIII)b, where a and b indicate the columns above and below the joint. on the
other hand, the column is likely to act to restrain the deflection then, logically, its stiffness
should be included in the calculation of 8/M, though this is not stated in the code. In general,
it is believed that the possibility of the columns above and below a joint buckling at the same
time is improbable and, though the column will deflect somewhat, its deflection will be much
smaller than that of the failing column. The values in Table 7.1 are roughly consistent with
assuming that the outer columns neither drive nor restrain the deflection, and this seems to
give a practically reasonable value for effective length.
..
The definitions of the four end conditions in Table 7.1 are as follows:
Condition 1. The end of the column is connected monolithically to beams on either
side which are at least as deep as the overall dimension of the column in the plane
considered. Where the column is connected to a foundation, this should be of a form
specifically designed to carry moment.
Condition 2. The end of the column is connected monolithically to beams or slabs on
either side which are shallower than the overall dimension of the column in the plane
considered.
Condition 3. The end of the column is connected to members which, while not specifically
designed to provide restraint to rotation of the end of the column, will, nevertheless,
provide some nominal restraint.
Condition 4. The end of the column is unrestrained against both lateral movement and
rotation (e.g. the free end of a cantilever column in an unbraced frame).
The other factor used to define the slenderness of a column is the slenderness ratio. This is
defined as the ratio of the effective length to the radius of gyration. The radius of gyration of
a section is defined by
i =I/A
where i is the radius of gyration, I is the second moment of area of the cross-section andA is
the area of the cross-section. For a rectangular section, 1= bh
3
/12 and A = bh, hence the
radius of gyration = 0.2887h, where h is the dimension of the section perpendicular to the
axis of bending considered. The radius of gyration of a circular section is equal to its actual
radius.
Many codes define the slenderness ratio as the ratio of the effective length to the section
depth. It is not easy to argue that one method is intrinsically better than another for
reinforced concrete since, as discussed earlier, the problem being dealt with is not one of
classical buckling and nor is reinforced concrete an elastic material. Use of the radius of
gyration does, however, have the advantage of providing a reasonable way of dealing with
non-rectangular sections.
EN 1992-1-1 gives limits for the slenderness ratio above which it will be necessary to take
account of slenderness effects for non-sway columns. If the slenderness ratio is less than A
lim
,
then second-order effects may be ignored.
Alim = (D7.5)
where
(D7.6)
and Band C are Nationally Defined Parameters, and their recommended values are as
follows.
A = 1/(1 + O.2<Pef) (D7.7)
where <Pef is the effective creep coefficient. Details of how to assess <Pef are given in clause
5.8.4 of EN 1992-1-1, but, in the absence of information, A may be taken as 0.7.
(D7.8)
Since the objective of the calculation is usually to find the value of As, this is unlikely to be
known, though a lower-bound estimate may be made by findingA/yd/AJcd for the first-order
moments using the appropriate design chart from Chapter 5. Alternatively, a value of 1.1
may be assumed.
(D7.9)
where MOl and M02 are the end moments of the column chosen so that IMo21 :2: IMOlI. In the
Clause 5.8.4
III
Clause 5.8.7.2
112
expression for C, M02 and MOl should be given their correct relative algebraic signs, thus, for
the normal case of a column in double curvature, C will always be greater than 1.7.
On this basis, for a column in double curvature, a conservative minimum value for "'lim
would be 27/-Yn.
No simple limits are given in EN 1992-1-1 for sway frames, but it seems reasonable to
assume that any structure that is classifiable as a sway frame will need to be designed for the
effects of deformations.
7.J. Design for slenderness effects
7.3.1. Basic approaches
Having concluded that a column or structure requires to be designed for slenderness effects,
the remaining, and biggest, problem is how this should be done. The rest of this chapter will
be concerned with this problem.
EN 1992-1-1 provides four basic approaches:
(1) A general method based on non-linear analysis of the structure including geometric
non-linearity. This approach will require a suitable computer package, and will not be
considered further here. For a relatively rigorous analysis of this type, it is necessary to
know the reinforcement details before the analysis is attempted, and so it can only be
used as a check or iteratively.
(2) A second-order analysis based on the nominal stiffnesses of the column or structure.
This can be done using less sophisticated software than the first approach, but still
requires knowledge of the reinforcement areas. It can therefore also only be used
iteratively or as a check. Establishment of the nominal stiffness is set out in clause 5.B. 7.2
of EN 1992-1-1. Since this is required for the moment magnification method discussed
below, this will be considered further when the moment magnification method is
discussed.
(3) The moment magnification factor method. In this method, the design moment including
second-order effects is obtained by multiplying the first-order moment by a factor. This
method will be discussed in more detail below.
(4) The nominal curvature method. In this method, an estimate is made of the ultimate
deflection, and, from this, an estimate of the second-order moment which is added to the
first-order moment. This is in principle the method currently used in BS 8110, and
should be familiar to UK designers. The method will be described in more detail below.
7.3.2. First-order moments
Both the moment magnification method and the nominal curvature method require an
estimate of the first-order moment at around mid-height of the column, since this is the point
where the second-order moment will be a maximum. Both methods use the same method to
establish the first-order moment, which is as follows.
Classic analyses of buckling commonly considers the deformation of a pinned ended strut,
but this is not the normal configuration of a column in a building. A normal column built
monolithically into a structure at its top and bottom will deform, and be subjected to
moments like those shown schematically in Fig. 7.3. It will be seen that the section of
the column between the points of contraflexure in the final state of the column may be
considered to be a pinned ended strut equivalent to that for which the analysis was carried
out. The distance between the points of contraflexure is the effective length of the column.
The estimation of this length has been covered earlier. The maximum moment due to
deflection will be seen to occur at mid-height of the effective column. This will normally be
somewhere close to mid-height of the real column. Clearly, the total moment to which the
critical section is subjected IS made up of the maximum moment due to the deflection plus
the first-order moment at this height plus any allowance for accidental effects. A reasonable
estimate of the first-order moment near mid-height of the column is given by
N
!
No sidesway
Deflected shape First-order
moments
Bending moment
due to deflection
Fig. 1.3. Moments and deformations of a braced isolated column
Total moment diagram
(D7.1O)
where Mmax and M min are, respectively, the numerically greater and lesser end moments. This
relationship, written in terms of eceentricities, is included in EN 1992-1-1 (clause 5.8.8.2)
together with a further precautionary limitation that the moment should not be less than
OAM
max

All methods require that the possibility of the structure being accidentally constructed
out-of-plumb should be considered. This is allowed for by introducing an accidental
eccentricity of the axial load in the ease of an isolated column or by considering an accidental
inclination from the vertical for a sway frame. Clause 5.2 defines the accidental inclination as
an angle 0i in radians:
(D7.11)
where a
h
is a reduetion factor for height 2/3 :s; 2/YL :s; 1), am is a reduction factor for the
number of members (= ,f[O.S(l + 11m)]), L is the length of the member or height of the
building and m is the number of vertical members contributing to the total effect.
For an isolated member such as a braced column, L is the actual length of the column, and
m = 1. In this circumstance the effect can be considered to be given by an additional
eccentricity at mid-height equal to OJrJ2. This may always be conservatively assumed to be
equal to [01400, resulting in a moment of
Mi NEd 10 1400
From Fig. 5.1a2 in EN 1992-1-1, it is clear that this accidental moment is a maximum at mid-
height of the column. It follows that the accidental moment should not be added to Mmax
andlor M min but should be added to MOe. Thus,
M
OEd
= MOe + Mi
7.3.3. Moment magnifier method
At first glance, this method seems deceptively simple. More detailed investigation shows that'
it is actually quite complex to apply. To aid application, a flow chart has been provided (Fig.
7.4) this has been slightly simplified from the full description as it has been drawn
specifically for columns in double curvature. That is, columns where the moment at one end
is either zero or has the opposite sign to that at the other end. The notation used in 7.4 is
that used in the code, but, for convenience, is repeated below:
Clause 5.8.8.2
Clause 5.2
113
114
(1)
(2)
(3)
(4)
(5)
Is
fPh to) 0;2
and MOEiNEd;o, h
and /0; 75?
NO
Is r;o, 0.01?
Fig. 7.4. Moment magnifier method flow chart
Ae area of the concrete section
As total area of the reinforcement
YES
E nominal modulus of elasticity of the column
Ecd design modulus of elasticity of the concrete
modulus of elasticity of the reinforcement
I nominal second moment of the area of the section about the axis of bending
considered
Ie second moment of the area of the concrete cross-section
Is second moment of the area of reinforcement about the centroid of the concrete
section
kl concrete strength factor
k2 axial force and slenderness factor
kc creep effect factor
ks reinforcement contribution factor
10
MOl
M02
M
OEd
M
oEqp
M
Ed
NB
NEd
n
<;J( 00, to)
<;Jeff
A
effective length of the column
numerically smaller end moment from first-order analysis
numerically larger end moment from first-order analysis
design first-order momcnt at about mid-height of the column
design first-order moment under the quasi-permanent load combination (also at
about mid-height of the column)
design ultimate moment
buckling load bascd on the nominal stiffness 7T
2
EI/lo2)
design ultimate axial force
relative axial force (=NEd/AJcd)
final creep coefficient of concrete loaded at time to
effective creep coefficient
slenderness ratio
p reinforcement ratio AjAJ
The various steps in the flow chart are numbered 1 to 8, and notes are given on each stage
below:
(1) The creep coefficient can be obtained from clause 3.1.4 in EN 1992-1-l.
(2) This provides a simplification for some situations where creep is unlikely to have any
significant effect on the behaviour.
(3) The assumption is made that creep dominantly occurs under the quasi-permanent load
and that the effect on creep of the additional load above the quasi-permanent load is
relatively short-term. It may be noted that, in internal columns in frames with uniform
spans, the quasi-permanent load is unlikely to produce significant moment, and hence
M
oEqp
will be close to zero, resulting in <;Jeff close to zero.
(4) This permits a simplification in the calculation of the effective stiffness where the
reinforcement percentage exceeds 1 %. A reasonable approach may be to assume that
the reinforcement percentage will exceed 1% and use the simplified route generally,
with a check at the end of the design procedure that this is actually the case.
(5) k) and k2 indirectly take account of cracking as well as other factors.
(6) EI is the effective stiffness of the column section.
(7) This formula is identical to that in BS 8110. It aims to give an estimate of the moment
around mid-height of the column where the additional moment due to deflection will be
a maximum, IMo21 ;::: IMOlI. In evaluating the formula, the moments should be given
their correct algebraic sign.
(8) The factor 8 in the denominator of the expression is correct for columns in double
curvature. Different values are given in clause 5.S. 7.3 of EN 1992-1-1 for other shapes of
bending moment diagram. The final requirement that M
Ed
should not be less than the
numerically larger first-order moment is not included in the code, but it would seem
nonsense if the design ultimate moment for a slender column was smaller than the
moment that would be used if the column was short. Another way of considering this is
that if the column was, say, stiffer than had been assumed and the deflection did not
occur, then it would be the moment at the end which would be critical and not the
moment near mid-height.
The difficulty with this method of establishing the moments in a slender column is that,
though considerable efforts have gone into calibrating the method to give similar results as
other methods, it provides no clear picture of the actual behaviour. In this respect, the
nominal curvature method is considerably superior.
7.3.4. Nominal curvature method
This method is set out in EN 1992-1-1 in clause 5.S.S.
If a rigorous non-linear analysis is carried out for a slender column, it will result in a load-
deflection curve of a form similar to one of the curves in Fig. 7.5, depending on the
Clause 3.1.4
Clause 5.8.7.3
Clause 5.8.8
115
116
slenderness ratio. The object of a rigorous analysis is to calculate such curves, and hence find
the maximum load capacity. The ideal simplified method would aim to establish this ultimate
load, or something close to it, in a single calculation without the necessity to calculate the full
load-deflection response. What the method in EN 1992-1-1 aims to do is to predict the
deflection at which failure of the concrete will initiate (i.e. when the maximum compressive
strain = IOu). It will be seen from Fig. 7.5 that, if this point can be established, then it will either
correspond to the actual ultimate load or to a conservative estimate of the ultimate load.
Such a method will therefore, in principle, give a lower bound estimate of the strength. The
calculation of the ultimate deflection is carried out as follows for a pinned ended strut:
(1) Estimate the ultimate curvature ofthe section. For a balanced section, the strain at the
compression face of the section is 0.0035, while the strain in the reinforcement is given by
fyd/Es' From this, the curvature can be written as
(D7.12)
For axial loads above the balance point, the curvature will be less than this, as can be seen
from Fig. 7.6. To allow for this, the balanced curvature is multiplied by a factor kr' where
kr = (nu - n )/(nu - n
ba1
) :S; 1
nu = 1 + AJyd/AJCd
(D7.13)
and nbal is the design load capacity of a balanced section. For a symmetrically reinforced
rectangular section, this may be taken as 0.4.
It will be seen that, for a balanced section, kr = 1, and that kr reduces to zero as the load
approaches the axial load capacity of the section. This approach provides an approximate
estimate of the ultimate curvature. A better method might be to define kr as cjx, though this
may not be so practically convenient.
Figures 7.7 and 7.8 can be used to obtain values of kr for symmetrically reinforced
rectangular sections.
The curvature at the critical section may also be increased by creep. This is allowed for by
multiplying the curvature by a factor k"" where
k", = 1 + (0.35 + fck/200 - A/150)IPef (D7.14)
Crushing of concrete
--.... --Ultimate load
=---_-;;;:::---Ultimale load
Crushing of concrete
Deformation
1.5. Load-deformation curves for columns of different slenderness
0.002
0""0 0.0027
/
Nud
0.002
0.0035
~ . 0 0 2
Moment 0.002
Fig. 7.6. Variation in curvature in a column at different levels of axial load
(2) Calculate the ultimate deflection. This may be done by assuming that the column
deflects in the form of a sine curve. Thus, if it is assumed that the curvature is proportional to
the moment, which is equal to the deflection times the axial load, we can write
a = H(lIr) sinCrrx/lo) dx
= lo\lIr) sine 'ITX/lo)!T(z
at mid-height, x/lo = 0.5 and, hence, approximately,
a = O.l(lIr)lo Z
The ultimate deflection at mid-height of the column can thus be calculated. On the one
hand, the estimate is likely to be conservative since the assumption of a sinusoidal variation
of curvature over the height of the column will be conservative due to the non-linear
behaviour of reinforced concrete. On the other hand, some of this conservatism will be
counterbalanced by the coefficient kr giving an unconservative estimate of the curvature in
some cases.
In EN 1992-1-1, the deflection is, for convenience, given as an eccentricity, and, thus,
e
Z
= O.l(l/r)lo Z (D7.15)
where e
z
is the second-order eccentricity.
(3) Define the design moments in the structure. This is covered in Section 7.3.2. The
design moment is now given by
M
Ed
= M
OEd
+ M
z
where M
z
= NEde
Z
'
It will be seen from Fig. 7.3 that as well as the moment near mid -heigh t being modified by
the deflections, the end moments are also affected. The numerically larger end moment will
actually be reduced while the numerically smaller end moment will be increased. Thus,
rigorously, there are three possible conditions which should be considered in establishing the
critical conditions for section design:
117
118
____ ______ ______ _________ 0T.4 _________

0.4
+---------10.4
0.0'-------'-------'-------'------'------.... 0.0
0.00 0.1 0.2 0.3 0.4 0.5
Mlbtffck
Fig. 7.7. k2 values for section with d'/h = 0.2
0.00 0.1 0.2 0.3 0.4 0.5
1.4 .---,,-,---,------,-------,------y-------, 1.4
0.2
0.0'------..l------'--------.!.-----.l-.-----...l0.0
0.00 0.1 0.2 0.3 0.4 0.5
Mfbtffck
Fig. 7.8. k2 values for section with d'/h = 0.1
(1) The numerically larger end moment assuming that no deflection occurs. In most columns
of intermediate slenderness, this is the critical design condition and,though the deflection
influences the moments at mid-height of the column, it does not influence the design. In
EN 1992-1-1 the range of columns where this can be relied upon to be the case are those
which have slenderness ratios lying below '\rit. Nevertheless, columns may frequently
occur where the design eccentricity given by equation (5.31) in EN 1992-1-1 is less than
the larger end eccentricity. Clearly, the larger end eccentricity must be taken as the
design value in this case.
(2) The moment at mid-height, allowing for the ultimate deflection. This is the condition
covered explicitly in EN 1992-1-l.
(3) The numerically smaller end moment increased by the effects of deflections. The
increase will be less than that corresponding to the second-order eccentricity at mid-
height. The UK code, BS 8110, assumes that the second-order moment at this end of the
column is equal to half the second-order moment at mid-height. EN 1992-1-1 does not
require a check at this point, and, in fact, it is rarely critical.
Example 7.1: uniaxially bent rectangular column
Calculate the design ultimate moment for which the column shown in Fig. 7.9 should be
designed. The column is in a braced frame. The stiffness of tbe beams and columns
framing into the column considered may all be assumed to have the same value of LIE! as
the column. It may be assumed that the column is loaded at 30 days, giving an effective
creep coefficient IPeff::;:: 0.87.
(1) Calculate tbe effective beigbt:
From equation (5.15) of EN 1992-1-1, Kl =K
z
::;:: 0.75, and hence the effective length is
0.81 x 7.0::;:: 5.67 m.
(2) Calculate tbe sleuderness ratio:
slenderness ratio::;:: effective height/radius of gyration.
(a)
5670/(0.2887 x 300)::;:: 65.5
Section
40kNm
Design moments
Design axiat force
NEd = 1500 kN
fcl<=30N/mm
2
~ k '" 460 N/mm2
~ Dimensions in mm
(b)
Fig. 7.9. Details of the column in Example 7.1
119
120
Calculate the slenderness limits:
in EN 1992-1-1
:::: 20ABC/-Yn
A :::: 1/(1 + O.21peff) 1/(1 + 0.2 x 0.87) 0.85
Since the reinforcement is not known, B will have to be assumed as suggested in the code
as 1.1.
r m -40/80 :::: -0.5
giving C :::: 2 . 2 ~
n = 1500 x 1000/(300
2
x 20) = 0.83
and hence
:::: 20ABC/-Yn =0 20 x 0.85 x 1.1 x 2.2/-Y0.83 = 45.15
The column is thus slender, and, since the slenderness exceeds A
lim
, specific measures to
deal with slenderness are necessary.
(4) Calculate Mj)Ed:
Moe is the greater of
0.6 x 80 - 0.4 x 50 = 28
or
0.4 x80= 32
and hence MOe:::: 32. The accidental eccentricity is
e
j
:::: l0l400 = 5670/400:::: 14.2
and hence the accidental moment is
1500 x 14.2/1000 kN m = 21.3 kN m
The total first-order design moment is thus
32 + 21.3 :::: 53.3 kN m
The design will now be carried out using the nominal curvature method.
(5) Calculate the second-order eccentricity, e
2
:
The ultimate eccentricity at about mid-height is given as
0llo2KrK",(CYd/0.45d)
From equation (5.37) in EN 1992-1-1 it will be found that K", = 1.06. For 500 N/mm
2
reinforcement,
:::: 500/(1.15 x 200000):::: 0.00217
Since, from Fig. 7.9, it can be seen that
d:::: 300 - 50:::: 250 mm
the eccentricity is thus given by
0.00217 x 5670
2
xKr x 1.06/(0.45 x 250 x 10):::: 65.8K
r
mm
This gives the second-order moment as
65.8 x 1500 x KJI000 kN m:::: 98.7K
r
kN m
A value for Kr can be found from Fig. 7.7 by iteration in the fonowing way:
:::: 1500 x x300x 0.56
A first estimate of the moment may be obtained by assuming Kr 1. This
M/bh'1ck = (53.5 + 98.7) x 1()6/(300' x 30) == 0.19
From Fig. 7.7, this gives Kr = 0.73. This leads to a modified total moment of
Mlbh'tk:::: (53.5 + 0.73 x 98.7) x 10
6
/(300
3
x 30):::: 0.155
From Fig. 7.7, this leads to a second estimate of Kr of 0.69 and a modified value of
M/bh
2
fck = 0.15. This is close enough to the previous result to suggest that this is a
reasonably accurate assessment of K
r
Figure 5.9 may now be used to obtain a value for
pfYk1fck of 0.51, and hence a steel area of 2754 mm
2

The calculation may now be repeated using the moment magnification method.
(6) Calculate EI and the buckling load, N B:
EI == K
t
K
2
E
e
ij(1 + YJeff) + E"lg
Kl = .v(JCk/20) = .v(30/20) = 1.22
K2 = nA/170 = 0.56 x 65.5/170 0.216
Ecd E
cn
/1.2 == 32000/1.2 == 26700 N/mm2
Ie bh
3
112 == 300
4
/12 == 675 x 10
6
mm
4
U sing the steel area calculated by the nominal curvature method, I, may be calculated as
1 x 2754 x lOOZ/2 = 27.5 x 1()6 mm
4
The buckling load may now be calculated:
N
B
:::: 1\2EI/l02
== 1\2[1.22 x 0.216 x 26700 x 675 x 10
6
/(1 + 0.87) +
200000 x 27.5 X 10
6
]/(5670
2
x 1000)
2468kN
(7) Calculate tbe desigu moment:
The design moment is now given by
M
Ed
= MoEd[l + j3/(NJN
Ed
-1)]
where j3 = 1\2/8. Thus,
M
Ed
53.3[1 + (1\2/8)/(2468/1500 - 1)] = 155 kN m
This compares with 126 kN m obtained using the nominal curvature method.
7.3.5. Other factors
The sections above have been concerned exclusively with cases where the second-order
effects act about the same axis as the first-order moments and where only uniaxial moments
act. While this will frequently be the case, it will by no means always be so, and clause Clause S.B.2.P(4)
5.8.2.P(4) of EN 1992-1-1 states that: 'the structural behaviour shall be considered in the
direction in which deformations can occur, and biaxial bending shall be taken into account when
necessary'. This clause is difficult to interpret, and much less clear than the
requirements of either BS 8110 or the pre-standard ENV 1991-1.
In principle, it seems possible that the accidental eccentricity, e
j
and the second-order
eccentricity, e
2
, could occur about either axis of the column, independently of the first-order
moments. This certainly must be true of the accidental eccentricity. It is also obviously true
that where a column with one dimension much greater than the other was designed for first-
121
122
order bending about the major axis, then it is highly likely that the second-order eccentricity
will occur perpendicularly to the minor axis. How, then, should one decide upon a reasonable
design situation to choose for the general case? The following proposal is not given in any
code, but seems a reasonable way to reduce the problem to consideration of a single load case.
The column will deflect under the action of the first-order moments and the accidental
moment. 1t is proposed that the second-order moment will occur in whichever direction the
deflection due to the first-order moment as a proportion of the effective length of the
column (a/l
o
) is greatest. 1t is assumed, though this is not stated in the code, that the
accidental moment and the second-order moments will only occur in one direction and not in
both directions at once.
An approximate analysis shows that the ratio a/io is proportional to MA/bh
2
, where h is the
dimension of the cross-section perpendicular to the axis of bending and b is the dimension
parallel to the axis of bending. It now remains to establish what situation gives the maximum
value for this parameter and about which axis of bending. In the following, the notation
shown in Fig. 7.10 is used. It is unimportant which is the minor and which is the major axis of
the section. We will now define Tlz and Tl
y
as follows:
TJz == (M
ocz
+ NEd ejy ) \/h
Tly == (Mocy + NEdeizy/b
The subscriptsy andz refer to moments, eccentricities and slenderness for bending about the
y and z axes, respectively.
It is assumed that if rJz ;::: TJ
y
then the second-order deflection will occur about the z axis,
otherwise it will occur about the y axis. On this basis, the design moments about the z and y
axes may be written as follows. If TJz ;::: TJ
y
then, for the nominal curvature method,
M
Edz
== Mocz + Ne jy + M2z
MEdy ==Mocy
Or, for the moment multiplier method,
M
Edz
== (M
oez
+ Ne
iy
)[1 + f3/(NB/NED -1)]
z axis
Centroid of
load
Fig. 7.10. Notation for biaxial bending
yaxis
If TJ
y
> TJz then, for the nominal curvature method,
M
Edz
= MOez
M
Edy
MOey + Neiz + M
2y
Or, for the moment multiplier method,
M
Edy
= (MOey + Neiz) [1 + p/(NB/NED -1)]
Note that, if the first-order bending moment only acts about one axis, then either MOez or Mocy
maybe zero.
The conditions set out above may lead to a necessity to design the column section for
biaxial bending. Where this is necessary, the provisions of clause 5.8.9.4 may be used. These
are discussed in Section 5.2.6.
EN 1992-1-1 does, however, define situations where a simplified approach whereby
checks on the basis of uniaxial bending may be carried out for the two axes independently.
These are cases where the first-order bending is close to being uniaxial at the critical section
for buckling. This condition is defined in clauses 5.8.9(2) and 5.8.9(3) as being when
(1) \/>"z::; 2 and \/\::; 2
(2) either (ey/h)/(e/b)::; 0.2 or (e/h)/(eyfb)::; 0.2.
7.3.6. Global second-order effects
The body of EN 1992-1-1 only deals in detail with the design of isolated columns for second-
order effects. This is the commonest situation where an individual column may deflect to
failure while the remainder of the structure does not deflect significantly. It is thus assumed
that there is no significant sidesway at the time that the column fails. Deflections of the
whole structure will, however, occur which will impose sidesway on all vertical members and,
if this sidesway is significant, it will need to be considered in the design. Clause 5.8 does give
rules for deciding whether the possible sidesway may be ignored, but gives no methods
for dealing with the problem if sidesway is significant. Methods are, however, given in
informative Annex D. Being an informative annex, Annex D may be adopted as normative by
individual countries, or they may introduce their own methods within the National Annex.
The rules by which it can be decided that global second-order effects may be ignored are
given in clauses 5.8.2(6) and 5.8.3.3. Clause 5.8.2(6) states the general principle that second-
order effects may be ignored if they are less than 10% of the corresponding first-order
effects. This is a useful principle to have stated, but is practically unhelpfuL Clause 5.8.3.3
provides an operational rule which states that global second-order effects may be ignored if
Fv, Ed::; kjnj(n
s
+ 1.6) x 'LECdIciL
2
where F v Ed is the total vertical load on both braced and bracing elements, ns is the number of
storeys, L is the total height of the structure above the level of moment restraint, Ie is the
second moment of area calculated on the basis of the concrete section of the members
considered to support lateral loads, and E C is the design modulus of elasticity of the concrete.
In general, it is likely that it can be established by inspection whether or not global second-
order effects need consideration.
The method given in Annex D for designing for global second-order effects will not be
considered in detail here as the description in the annex seems clear. Basically, the method
calculates a magnified horizontal load for which the structure should be designed.
7.3.7. Walls
Reinforced concrete walls may be treated by the either the moment multiplier method or the
nominal curvature method in exactly the same way as columns except that it is only necessary
to consider the possibility of deformation about the minor axis (i.e. only deformation
perpendicular to the plane of the wall need be considered). The approach for the design of
Clause 5.8.9.4
Clause 5.8.9(2)
Clause 5.B. 9(3)
Clause 5.B
Clause 5.8.2(6)
Clause 5.8.3.3
123
Clause 5.9
124
walls assumes that the first-order analysis for the vertical and lateral loads will give a
distribution of axial load along the wall. Consideration of the interaction of the floors and
beams framing into the wans will lead to transverse first-order moments. Sections of wall will
then be designed for the maximum axial load per unit length of wall within the section
considered combined with any transverse moments and the appropriate accidental and
second-order eccentricities.
7.3.0. lateral buckling of slender beams
Where a beam is narrow compared with its span or depth, it may fail by lateral buckling of the
form illustrated schematically in Fig. 7.11. Beams where this is likely to be a problem are
relatively rare, and hence a simple, conservative check is normally sufficient. EN 1992-1-1
includes such a check in clause 5.9. This states that the safety will be adequate without further
check provided that:
(1) For persistent situations,
h/b ~ 2.5
(2) For transient situations,
lo/b ~ 70/(h/b) 1/3 h/b ~ 3.5
Mid-span section
Fig. 7.1 I. Buckling of a slender beam
b
Fig. 7.12. Cross-section of a slender beam with a parapet
where lOt is the distance between torsional restraints, h is the overall depth of the beam in the
central part of lOt and b is the width of the compression zone or flange.
No indication is given in EN 1992-1-1 as to how a further check should be formulated,
should this be necessary. The CEB-FIP 1990 Model Code does, however, give a more
detailed method of analysis in its clause 6.7.3.3.4. This method has similarities with the
nominal curvature method in that it postulates an ultimate deflected shape and then ensures
that the critical section can withstand the resulting internal actions. Reference should be
made to the CEB-FIP 1990 Model Code for details of the method.
A problem which occasionally occurs in practice with slender beams is where, for example,
an edge beam is designed with a thin parapet cast monolithically as sketched in Fig. 7.12.
Such a beam would normally be designed ignoring the effect of the upstand parapet;
however, rigorous interpretation of rules such as those in EN 1992-1-1 would imply that such
a member cannot be used because of the slenderness of the parapet. It must, in such
circumstances be satisfactory to state that if the beam is adequately safe without the parapet,
then the addition of the parapet cannot make it less safe.
125
CHAPTER 8
Serviceability limit states
8. I. General
EN 1992-1-1 deals in some detail with three common serviceability limit states. These are:

Limitation of stresses
Control of cracking
Control of deflections
Design for any limit state requires the definition of four quantities. These are:
Clause 7.2
Clause 7.3
Clause 7.4
(1) definition of the appropriate loading and methods of analysis so that the design load
effects can be established
(2) definition of the design material properties to be assumed in the verification
(3) definition of criteria defining the limit of satisfactory performance
(4) definition of suitable methods by which performance may be predicted.
Of these, only (3) and (4) will be found in Section 7 of EN 1992-1-1; (1) will be found in
EN 1990 (Section 6 and clause 2.5.3.1) and (2) in Section 3. All these matters will be
summarized in this chapter. It should be noted that in most cases it will not be necessary to
carry out explicit calculations for the serviceability limit states, as simple, 'deemed to satisfy'
procedures are given in the code for dealing with all three of the limit states covered. This
approach is acceptable because serviceability is intrinsically less critical than the ultimate
limit states, and major calculation effort is not justified. For example, if the structure is
wrongly designed and the strength is even 1 % below the imposed loads, the structure
collapses. By comparison, if the crack width turns out to be 0.33 mm instead of 0.3 mm,
nothing more serious than some grumbling from the owner and a cosmetic repair is likely to
result.
8.1.1. Assessment of design action effects
EN 1990 (clause 6.5.3) defines three combinations of actions which may need to be considered
when designing for a serviceability limit state. These are:
Characteristic combination:
0k,j + (P) + Qk, 1 + L 1/Ja, iQk,i
(D8.1)
Frequent combination:
Ok,} + (P) + QI, 1+ L iQk,i
(D8.2)
Quasi-permanent combination:
0k,j + (P) + L 1/J2,iQk,i
(D8.3)
Clause 5.10.9
Clause 3.2.7(4)
Clause 3.3.6(2)
Clause 3.3.6(3)
Clause 3.1.3
Clause 3. t .4
128
It will be understood from these formulae that the partial factor on the loads is always 1.0 for
serviceability limit states. For buildings, the following simplifications which may be used for
the characteristic or frequent combinations:
Where there is only one variable action:
G
k
,} + (P) + Q
k
Where there are two or more variable actions:
Ok,} + (P) + 0.9I,Qk,i
Analysis may be elastic (without redistribution). This analysis may normally be based on the
stiffness of the uncracked section; however, if it is suspected that cracking may have a significant
unfavourable effect on the performance, then a more realistic analysis taking account of the
cracking should be used. This possibility may be ignored for normal building structures.
The determination of the effects of prestress is covered in clause 5.10.9.
8.1.2. Material properties
Clause 6.5.4 of EN 1990 states that the partial safety factor applied to material properties
should generally be 1.0.
The properties of the materials which are normally significant in serviceability calculations
arc the modulus of elasticity of the reinforcement, the modulus of elasticity, the creep
coefficient, the shrinkage strain and the tensile strength of the concrete. All this information
is given in Section 3, but is summarized here for convenience.
Modulus of elasticity of reinforcement
For ordinary reinforcement this may be taken as 200 kN/mm2 (clause 3.2. 7(4. For prestressing
steel, 205 kN/mm2 may be assumed for wires and bars, and 195 kN/mm2 for strand (clauses
3.3.6(2) and 3.3.6(3.
Modulus of elasticity of concrete
Values of the modulus of elasticity are given in clause 3.1.3 and Table 3.1. The clastic
modulus of concrete varies with other factors than just the strength (e.g. the aggregate type),
and, if an accurate prediction of serviceability conditions is required, it will be necessary to
establish the value of Ecru by tests on the type of concrete actually being used.
Creep coefficient
The creep coefficient, which is defined as the ratio of the creep deformation to the
instantaneous elastic deformation, depends upon many factors, the most significant of which
are the age at loading, the time under load, the relative humidity, the section geometry, the
concrete strength and the type of cement. Other factors such as the aggregate type may also
be significant. As with the modulus of elasticity, the only way of determining the creep
performance of concrete with any real reliability is to obtain data from creep tests on the
concrete actually being used. This information is rarely available, and more approximate
means have to be adopted. The equations given in EN 1992-1-1 should be viewed in this light.
Full equations for the prediction of the creep coefficient are given in Annex B. These have
been used to produce design graphs (Fig. 3.1 in clause 3.1.4).
Free shrinkage strain
The free shrinkage strain depends upon the same basic variables as creep, and equations are
given in Annex B for its prediction. Simplified values are given in clause 3.1.4.
Tensile strength of concrete
Table 3.1 gives the mean tensile strength and the upper and lower characteristic tensile
strengths as a function of the characteristic compressive strength of the concrete. Equations
are also given in Table 3.1.
CHAPTER 8. SERVICEABILITY LIMIT STATES
8.2. limitation of stresses under serviceability conditions
8.2.1. General
The question of the possible limitation of stresses under serviceability loads was the subject
of a great deal of discussion during the drafting of EN 1992-1-1. The basic reasons for
differences of opinion arose from the different way in which design methods had developed
in different countries and a natural desire in each country that the new and untried Eurocode
should not force design too far from existing experience in that country. Limitations on the
compressive stress in the concrete were particularly contentious. In many countries, section
design for both prestressed and reinforced concrete has been carried out on the basis of
ultimate strength methods, and no explicit check has been required on the concrete stresses
under service loads. These countries were determined that no such checks should be
introduced since these could be shown to have a significant effect on the economy of some
types of member, notably columns. Other countries have not so far adopted ultimate load
methods, and still base their design on an elastic analysis of sections combined with a stress
limit. While these countries were prepared to accept ultimate strength methods for section
design, they also wanted to retain elastic design as an additional requirement to ensure that
member sizes, particularly columns, did not become much smaller or more lightly reinforced
than those arising from their current design methods. The check on the compressive stress in
the concrete at the serviceability limit state provided them with this safeguard.
In order to develop a rational draft for this section of the code which would have some
hope of general acceptance, an attempt was made to establish what rational basis existed for
a limit on stresses under serviceability conditions. Two limits need to be considered:
compressive stress limits and tensile stress limits. These will be considered in turn.
Limits to compressive stress
Two reasons are usually put forward for limiting compressive stresses in the concrete. These
are (1) to avoid the formation of microcracks in the concrete which might reduce durability
and (2) to avoid excessive creep. There is some logic in both reasons. It is commonly
accepted that microcracking will start to develop in concrete when the compressive stress
exceeds about 70% of the compressive strength. However, none of the established
publications on durability suggest any relationship betweeen stress level, micro cracking and
durability problems (e.g. see CEB Bulletin d'Information 185
9
). EN 1992-1-1 suggests that, in
aggressive environments (exposure classes XD, XF and XS), it may be reasonable to limit
the stress to 0.6fck if other measures, such as an increase to the cover to any compressive
reinforcement or the provision of confinement to the compressed concrete by transverse
reinforcement.
The design methods used for checking serviceability assume that creep will be proportional
to stress. This assumption is increasingly invalid as the stress increases above about 50% of
the compressive strength. It could therefore be logical to impose a limit on compressive
stress under the quasi-permanent combination of loads where a higher level of creep than
might be predicted could seriously impair performance. This problem has been dealt with in
clause 3.1.4, where a correction equation has been introduced which enables the creep to be
calculated for levels of compressive stress above 0.45fck.
It will be seen that there is no case for a blanket limitation to compressive stresses but that,
in particular cases, it may be prudent to apply such limits if the effects of non-linear creep are
not explicitly calculated. EN 1992-1-1 (clause 7.2) suggests that non-linear creep should be
taken into account if the compressive stress exceeds 0.45fck under the quasi-permanent load.
In UK practice, compressive stresses have been checked in the design of prestressed i
members where high compressive stresses are likely to be present permanently, but no
check has been required in the design of reinforced sections over the last 50 years. There
has been no evidence of problems with performance that could be related to high
compressive stress levels in service. It therefore seems reasonable to assume that there is no
general requirement to check the compressive stresses in reinforced-concrete members
Clause 3. 1.4
Clause 7.2
129
Clause 7.2(5)
Clause 7.2(5)
130
under service conditions, although there may be special circumstances where this might
be prudent.
Limitation of tensile stresses
Limits may be defined for the tensile stresses in the concrete or in the reinforcement. In the
case of reinforcement, it seems reasonable to ensure that inelastic deformations of the steel
are avoided under service loads. Such dcformations would, firstly, invalidate any calculations
of cracking or deflections which assume that the reinforcement behaves elastically and,
secondly, could result in excessively large cracks. This could be particularly critical in areas
subjected to frequent variations in loading. The limits suggested in clause 7.2(5) are aimed at
ensuring that inelastic deformations do not occur, and include some margin of safety against
this occurring.
Prestressing steels are much more sensitive to corrosion damage than are normal
reinforcing bars. This is partly due to differences in metallurgy but mostly due to their
smaller diameter and the high level of stress at which they normally operate. A prestressing
tendon may frequently operate for its whole life at a stress of up to 70% of its tensile strength
over its whole length. A reinforcing bar is unlikely to be stressed above about 50% of its yield
strength for any significant period, and then only in a very limited area close to sections of
maximum moment. It is therefore necessary to take much more stringent precautions
against corrosion with prestressed members than with reinforced members. In any parts of a
member which could be exposed to salt, it is necessary to ensure that no direct paths exist
which could give direct access for the salt to the surface of the tendon. Cracks could provide
such a path, and so cracks which could penetrate to the tendons should be avoided in salty
environments. This objective can conveniently be achieved by ensuring that no tension
develops in the concrete surrounding the tendons. Clause 7.2(5) provides an element of extra
safety by requiring that, in environments where this precaution is necessary, the tendon is
surrounded by at least 25 mm of uncracked concrete.
8.2.2. Procedure for stress ched(s
In general, stress checks may be avoided for reinforced concrete building structures if the
advice given in the National Annex is deemed relevant.
If stresses are to be checked, the calculation is done on the basis of the following
assumptions:
(1) Plane sections remain plane.
(2) Reinforcement and concrete in tension are assumed to be elastic.
(3) Concrete is assumed to be elastic up to its tensile strength,fctm' If the stress has exceeded
Ictm' the section is assumed to be cracked, and the concrete in tension is ignored.
(4) EN 1992-1-1 does not give details of how creep under varying loads should be handled.
However, a reasonable approach seems to be to assume that creep may generally be taken
into account by assuming that the ratio of the elastic modulus for steel to that for concrete
(modular ratio) is 15. A lower value, based on the actual elastic modulus of the concrete
may be used where less than 50% of the stresses arise from quasi-permanent loads. A
more accurate assessment of the effects of creep may, of course, be used if desired.
The code states that shrinkage and temperature stresses should be taken into account
where these are likely to be significant. This will not usually be necessary for normal
reinforced and prestressed members in buildings; however, appropriate methods for doing
this will be considered on p. 148.
In general, stress checks can be carried out using standard elastic formulae. Thus, for
beams subjected only to flexure, stresses may be calculated from
stress = My/I (D8.4)
where M is the applied moment, y is the distance from the neutral axis to the point
considered and I is the second moment of area of the section (this should be based on either
a cracked or uncracked section, as appropriate).
CHAPTER 8. SERViCEABILITY LIMIT STATES
Table 8.1. Neutral axis depths and moments of inertia for flanged beams
(a) hf/d = 0.2
bJb
1.0 0.5 0.4 0.3 0.2
a
ep x/d I/bd
3
x/d I/bd
3
x/d I/bd
3
x/d l/bd
3
x/d l/bd
3
0.020 0.181 0.015
0.030 0.217 0.022 0.217 0.022 0.217 0.022 0.217 0.022 0.217 0.022
0.040 0.246 0.028 0.248 0.028 0.248 0.028 0.249 0.028 0.249 0.028
0.050 0.270 0.033 0.274 0.033 0.275 0.033 0.276 0.033 0.278 0.033
0.060 0.292 0.038 0.298 0.038 0.300 0.038 0.302 0.038 0.304 0.038
0.070 0.311 0.043 0.320 0.043 0.322 0.043 0.325 0.043 0.327 0.043
0.080 0.328 0.048 0.340 0.047 0.343 0.047 0.346 0.047 0.349 0.047
0.090 0.344 0.052 0.358 0.052 0.361 0.052 0.365 0.051 0.369 0.051
0.100 0.358 0.057 0.375 0.056 0.379 0.056 0.383 0.055 0.388 0.055
0.110 0.372 0.061 0.390 0.060 0.395 0.059 0.400 0.059 0.406 0.059
0.120 0.384 0.064 0.405 0.063 0.410 0.063 0.416 0.063 0.422 0.062
0.130 0.396 0.068 0.418 0.067 0.424 0.066 0.430 0.066 0.437 0.065
0.140 0.407 0.072 0.431 0.070 0.437 0.070 0.444 0.069 0.452 0.069
0.150 0.418 0.075 0.443 0.073 0.450 0.073 0.457 0.072 0.466 0.071
0.160 0.428 0.078 0.455 0.076 0.462 0.076 0.470 oms 0.478 0.074
0.170 0.437 0.082 0.466 0.079 0.473 0.078 0.481 0.078 0.491 0.077
0.180 0.446 0.085 0.476 0.082 0.484 0.081 0.493 0.080 0.502 0.079
0.190 0.455 0.088 0.486 0.085 0.494 0.084 0.503 0.083 0.513 0.082
0.200 0.463 0.091 0.495 0.087 0.504 0.086 0.513 0.085 0.524 0.084
0.210 0.471 0.094 0.504 0.090 0.513 0.089 0.523 0.088 0.534 0.086
0.220 0.479 0.096 0.513 0.092 0.522 0.091 0.532 0.090 0.543 0.089
0.230 0.486 0.099 0.521 0.094 0.531 0.093 0.541 0.092 0.552 0.091
0.240 0.493 0.102 0.529 0.097 0.539 0.095 0.549 0.094 0.561 0.093
(b) hf/d = 0.3
bJb
0.5 0.4 0.3 0.2
P
x/d l/bd
3
x/d l/bd
3
x/d I/bd
3
x/d l/bd
3
x/d I/bd
l
0.020 0.181 0.015
0.030 0.217 0.022
0.040 0.246 0.028
0.050 0.270 0.033
0.060 0.292 0.038
0.070 0.311 0.043 0.311 0.043 0.311 0.043 0.311 0.043 0.311 0.043
0.080 0.328 0.048 0.328 0.048 0.329 0.048 0.329 0.048 0.329 0.048
0.090 0.344 0.052 0.345 0.052 0.345 0.052 0.345 0.052 0.346 0.052
0.100 0.358 0.057 0.360 0.056 0.361 0.056 0.361 0.056 0.362 0.056
0.110 0.372 0.061 0.375 0.060 0.375 0.060 0.376 0.060 0.377 0.060
0.120 0.384 0.064 0.388 0.064 0.389 0.064 0.390 0.064 0.391 0.064
0.130 0.396 0.068 0.401 0.068 0.402 0.068 0.403 0.068 0.404 0.068
0.140 0.407 0.072 0.413 0.071 0.414 0.071 0.416 0.071 0.417 0.071
0.150 0.418 0.075 0.425 0.075 0.426 0.075 0.428 0.075 0.430 0.075
0.160 0.428 0.078 0.436 0.078 0.437 0.078 0.439 0.078 0.441 0.078
0.170 0.437 0.082 0.446 0.081 0.448 0.081 0.450 0.081 0.452 0.081
0.180 0.446 0.085 0.456 0.084 0.458 0.084 0.461 0.084 0.463 0.084
131
Table 8.1. (Contd)
br/b
1.0 0.5 0.4 0.3 0.2
P
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
0.190 0.455 0.088 0.466 0.087 0.468 0.087 0.471 0.087 0.473 0.087
0.200 0.463 0.091 0.475 0.090 0.477 0.090 0.480 0.090 0.483 0.089
0.210 0.471 0.094 0.483 0.093 0.486 0.092 0.489 0.092 0.493 0.092
0.220 0.479 0.096 0.492 0.095 0.495 0.095 0.498 0.095 0.502 0.095
0.230 0.486 0.099 0.500 0.098 0.503 0.098 0.507 0.097 0.511 0.097
0.240 0.493 0.102 0.508 0.100 0.511 0.100 0.515 0.100 0.519 0.099
(c) hfJd = 0.4
bJb
1.0 0.5 0.4 0.3 0.2
P
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
x/d I/bd
l
0.020 0.181 0.015
0.030 0.217 0.022
0.040 0.246 0.028
0.050 0.270 0.033
0.060 0.292 0.038
0.070 0.311 0.043
0.080 0.328 0.048
0.090 0.344 0.052
0.100 0.358 0.057
0.110 0.372 0.061
0.120 0.384 0.064
0.130 0.396 0.068
0.140 0.407 0.072 0.407 0.072 0.407 0.072 0.407 0.072 0.407 0.072
0.150 0.418 0.075 0.418 0.075 0.418 0.075 0.418 0.075 0.418 0.075
0.160 0.428 0.078 0.428 0.078 0.428 0.078 0.428 0.078 0.428 0.078
0.170 0.437 0.082 0.438 0.082 0.438 0.082 0.438 0.082 0.438 0.082
0.180 0.446 0.085 0.447 0.085 0.447 0.085 0.448 0.085 0.448 0.085
0.190 0.455 0.088 0.456 0.088 0.457 0.088 0.457 0.088 0.457 0.088
0.200 0.463 0.091 0.465 0.091 0.465 0.091 0.466 0.091 0.466 0.091
0.210 0.471 0.094 0.473 0.094 0.474 0.094 0.474 0.094 0.474 0.093
0.220 0.479 0.096 0.481 0.096 0.482 0.096 0.482 0.096 0.483 0.096
0.230 0.486 0.099 0.489 0.099 0.490 0.099 0.490 0.099 0.491 0.099
0.240 0.493 0.102 0.496 0.101 0.497 0.101 0.498 0.101 0.498 0.101
Similarly, for uncracked reinforced members subject to moments and axial load or
prestressed beams, the stresses may be calculated from the standard relationship
stress = My/! + N/Ac (D8.S)
where N is the applied axial force and Ac is the area of the total concrete section.
Tables 8.1 and 8.2 give values for the neutral axis depth and second moments of areas of
cracked and uncracked rectangular and flanged sections. For cracked sections, only singly
reinforced members are considered. Figures 8.1 and 8.2 give values for the cracked neutral
axis depths and second moments of area of doubly reinforced rectangular sections; these
have been taken from Rowe.
1O
132
Table 8.2. Properties of uncracked sections
hf/h
1.0
bJb x/h
0.100
0.150
0.200
0.250
0.300 0.500
0.350
0.400
0.450
0.500
."
.Q
0.4 0.3
l/bh
3
x/h I/bh
3
x/h l/bh
3
0.265 0.020 0.245 0.019
0.292 0.026 0.280 0.026
0.315 0.032 0.309 0.032
0.336 0.037 0.334 0.037
0.083 0.355 0.042 0.356 0.042
0.372 0.046 0.375 0.046
0.388 0.050 0.391 0.050
0.401 0.054 0.406 0.054
0.414 0.057 0.419 0.057
0.4.-------,------r---.,.--.,..-....--.......
0.3 t---------j-----
~ 0.21---------j-----I--+-+-+-/-------I
~
O ~ ~ - - - - - - ~ ~ - - - - - - - - - - ~ - - - - - - - - ~
0.2 0.4 0.6
xJd
Fig. 8. i. Neutral axis depths calculated on the basis of a cracked section
0.2
x/h l/bh
3
0.243 0.019
0.288 0.026
0.322 0.031
0.350 0.036
0.373 0.041
0.392 0.045
0.408 0.049
0.421 0.052
0.433 0.055
The calculation of stresses in a cracked member subject to axial forces and moments is less
simple since the second moment of area of the cracked section is a function of the axial force.
In such cases it is generally easier to write out the equations of equilibrium and solve these
iteratively. An approximate calculation of the steel stress in such sections can, however, be
made by the use of the design chart in Fig. 8.3. This is used as follows:
(1) Calculate the average axial stress in the section by dividing the longitudinal force by the
area of the concrete section. This is then divided by the reinforcement ratio (Ajbvfl) and
multiplied by the factor bwlb, to give the longitudinal force factor, NbwlAcb.
133
134
0.04 0.08
IdbcJ3
0.12 0.16
Fig. 8.2. Second moments of the area of rectangular sections calculated on the basis of a cracked section
f/5p (N/mm2)
100 200 300 400 500
6 0 0 r - - - - - - y - - - - - - - r - - - - - - ~ - - - - ~ - - - - - - _ r - - - - ~
500
.Q
?
Ci:
400
750
0
S
NE
1000
~
))l300
~
1500
t5
m
c7l
200
100 200 400 500
Fig. 8.3. Design chart for the approximate calculation of the stress in reinforcement in a
prestressed beam
(2) Calculate the 'notional tensile stress'., it' at the tension face of the section. This is
calculated on the basis of an uncracked section, even though the section is cracked. This
is then used to calculate the factor 1/5.
(3) The above two factors can then be used in Fig. 8.3 to give the steel stress in any
unprestressed reinforcement in a prestressed beam or the change in stress in the
prestressing tendons from the condition where the stress in the concrete immediately
surrounding the steel is zero.
Example 8.1
Calculate the stresses in a T beam with the dimensions given below when subjected to a
moment of 100 kNm:
Overall breadth: 600 mm
Overall depth: 500 mm
Effective depth: 450 mm
Flange depth: 150 mm
Rib breadth: 200 mm
Steel area: 2455 mm
2
The concrete grade is C30/37.
From Table 3.1, the mean tensile strength of the concrete is 2.8 N/mm2. From Table 8.2
the neutral axis depth and second moment of area of the uncracked section can be found
to be
Xl = 0.367 x 500 = 184 mm
II = 0.045 x 600 x 500
3
== 3375 X 10
6
mm
4
The tensile stress at the bottom of the section assuming an uncracked section is thus given by
.ft = 100 x (500 - 184)/3375 = 9.36 N/mm"'
This is greater than the tensile strength of the concrete, so the section should be assumed
to be cracked. The neutral axis depth and second moment of area of the cracked section
can be obtained by interpolation from Tables 8.1b and 8.le. The reinforcement ratio
multiplied by the modular ratio is given by
p = 2455/600/450 x 15 == 0.136
From the tables,
Xu = 0.41 x 450 = 185 mm
In == 0.07 x 600 x 450
3
= 3827 X 10
6
mm
4
The stresses can now be calculated:
compressive stress in the concrete == 100 x 185/3827 == 4.83 N/mm2
tensile stress in steel = 100 x (450 - 185) x 15/3827 = 104 N/mm2
Both these stresses are wen below the limits given in the code.
Stress checks for reinforced concrete columns
The basic approach for checking the stresses in columns is the same as for beams and,'for
uncracked members, the formulae given in Section 8.2.2 above for combined axial force and
bending may be used. For cracked sections, the algebra becomes rather more involved, and
the design charts given in Figs 8,4a-x may be used. The first set of charts (Figs 8.4a-l) give the
reinforcement areas necessary to meet the limit of O.6ick while the second set (Figs 8.4m-x)
give the reinforcement areas necessary to meet the limit of 0,45fck.
135
136
"7
..!<
.<::
.Q
~
(al
(b)
0.0 0.1 0.2 0.3 0.4 0.5
1.4 1.4
Ratio d'fh = 0.05
...,r-
1.2
1.0

~
k-L
b
L
0.8
'I '1
0.8
0.6
0.4 0.4
0.2
0.0
0.1 0.2
M/(bh
2
fckl
0.4 0.5
0.0 0.1 0.2 0.3 0.4 0.5
1.4 ,.;;..------..-------..,..-------__ ..------....,------....., 1.4
Ratio d'th = 0.10
r--..------,

..,Ie...-!--. ___ . - - , ~
L b l
0.8
"-""""-.+-----"--+-----"""""--""'1
0
.2
~ ________ ~ ~ L _ ~ ____ ~ __ ~ L _ ~ __ ~ __ ~ ~ ~ ~ __________ ~ __________ ~ O . O
0.1 0.2 0.3 0.4 0.5
Fig. 8.4. Design charts for ensuring that the compressive stress in reinforced concrete columns
does not exceed O.6f
ck
0.3 0.4
0.5
1
.
4
Ratio d'th = 0.20
I'-
1.2

1.2
1.0 1.0

~
L
b
L
0.8
'I
.,
0.8
"2
-.l!
~
~
0.6 0.6
0.4
0.4
0.2 0.2
0.0
0.1 0.2
MI(btffck)
0.3 0.4 0.5
(d)
Fig. 8.4. (Contd)
137
0.0 0.1 0.2 0.3 0.4 0.5
1.4 1.4
Ratio d'lh = 0.15
1.2
.. ..
1.2
i
I
.. .. I
I
1.0 ~ ~ t--------- 1.0
':55 .. ..
$
"-
"-l
b
L
0.8
1 '1 0.8
-:;:
-.!<
J::
.Q
~
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.1 0.2 0.3 0.4 0.5
(e)
M/(btffck)
0.0 0.1 0.2 0.3 0.4 0.5
1
.4
1.4
Ratio d'lh = 0.10
,f-
.. ..
1.2
1.2
.. ..
1.0 1.0
.. ..
~
-'<-L
b
L
0.6
1 '1
0.8
x
-.
:
~
0.6 0.6
0.4 0.4
0.2 0.2
0.0
0.1 0.2 0.3 0.4 0.5
(f)
MI(bh
2
fck}
8.4. (Contd)
138
1.4
0.0 0.1 0.2 0.3 0.4 0.5
1.4
Ratio d'lh = 0.15
'--
1.2 1.2

1.0

~
L
b
L
O.S
'l 'I
O.S
"" ~
:
~
0.6
0.4
0.2
0.0
0.1
(g)
0.2
MI(btffck)
0.3 0.4 0.5
0.0 0.1 0.2 0.3 0.4 0.5
1.4 1.4
Ratio d'th = 0.20
.,1'-

1.2

1.0
g
L
b
l
O.S
'I
.,
O.S
"" ~
:
~
0.6
0.4 0.4
0.2 0.2
0.0
0.2
MI(btffck)
0.3 0.4 0.5
(h)
Fig. 8.4. (Contd)
139
0.0 0.1 0.2 0.3 0.4 0.5
1.4 1.4
Ratio d'ih = 0.05
f-
1.2 1.2

1.0

~
~ l
b
L
0.8
'I 'I
0.8
'" ..!<
:5
~
0.6 0.6
0.4 0.4
0.2 0.2
0.0
0.1 0.2
MI(btffCk)
0.3 0.4 0.5
(i)
0.0
1.4
0.1 0.2 0.3 0.4 0.5
1.4
Ratio d'ih = 0.10
I'-
1.2

1.2

1.0 1.0

S
L
b
L
0.8 1
'1
0.8
-:;;:
..!!
:5
~
0.6 0.6
0.4 0.4
0.2 0.2
0.0
0.1 0.2
MI(btffck)
0.3 0.4 0.5
0)
8.4. (Contd)
140
0.0 0.1 0.2
1.4
0.3 0.4
0.5
1
.4
Ratio d'ih = 0.15
.,f-

1.2 1.2

1.0 1.0
g:
l
b
L
'I
.,
0.8
"" ..;e
:8
~
0.6
0.4 0.4
0.2
0.0
0.1
(k)
0.2
MI(btffck)
0.3 0.4 0.5
0.0 0.1 0.2 0.3 0.4 0.5
1.4 1.4
Ratio d'lh = 0.20
1.2

1.2

1.0
1.0

~
L
b
}
'I
0.8
"" ..;e
:8
~
0.6
0.4
0.2
0.0
0.1 0.2
MI(btffck)
0.3 0.4 0.5
(I)
fig. 8.4. (Contd)
141
0.0 0.1 0.2 0.3
1.0 1.0
Ratio d'lh = 0.05
..,f--

0.8

0.8

0.6 l
b
l 0.6
.,
1
-:;;-
""
J:::
..Q

0.4 0.4
0.0 ro:;;,. ______ ....lo... __ __ ...:lb.. __ _ _....Jc.... __ ::O"'_.....;:lI.__....:lI ___ L_ _ ___' 0.0
0.0 0.1 0.2 0.3
(m)
MI(btffck)
0.0 0.1 0.2 0.3
1.0 1.0
Ratio d'lh = 0.10

0.8 0.8

...,"--
L
b L
0.6 0.6
'I ...
:]"
:fS

0.4 0.4
(n)
Fig. 8.4. (Contd)
142
0.1 0.2 0.3
1.0
Ratio d'lh = 0.15

0.8 0.8

L
b
L
0.6
1
'1 0.6

..!<
13

0.4 0.4
0.2 ... .... .. .. - .. .. .. -.--_+--_I 0.2
__ __ __
0.2 0.3
(0)
0.1 0.2 0.3
1.0
Ratio d'lh = 0.20
'--

0.8 0.8
g
0.6
,'--L
b
L
0.6
'1

..!<
.<:::
.Q

0.4 0.4
8.4. (Contd)
143
0.1 0.2 0.3
1.0
Ratio d'lh = 0.05
...,,<-

0.8

0.8

~
0.6
<-L
b
L
0.6
"
Ij
'" ..!!
:
~
0.4 0.4
~ - - - - - - - - ~ ~ - - ~ - - ~ - - ~ - - ~ - - ~ - - ~ - - ~ - - - - - - - - - - - - - - - - - - L - - - ~ O . O
(q)
M/(btffck)
0.3
0.1 0.2 0.3
1.0
Ratio d'lh = 0.10

0.8
:g:
<-L
b
L
0.6
'1 'I
-:;;-
..!!
is
~
0.4
(r)
8.4. (Contd)
144
0.1 0.2 0.3
1.0
Ratio d'lh = 0.15

0.8 0.8
.c::

g:
..,LL
0.6
b
l
0.6
1 'I
~
.,J!
.c::
.Q
~
0.4 0.4
~ ______ ~ __ ~ - L ~ ~ L - ~ ~ - L __________ ~ ~ ________________ ~ ____ ~ O . O
(s)
0.3 0.1
M/(btffck)
0.2
0.1 0.2 0.3
1.0
Ratio d'lh = 0.20
<-

0.8
g
l
b
'I
L
~
0.6
~
2i
~
0.4
~ - - - - - - ~ ~ ~ ~ ~ - L ~ ~ - - - - - - - - - - - - - - - - ~ - - - - - - - - - - - - - - - - - - - - ~ - - ~ O . O
0.1 0.2 0.3
(t)
M/(btffck)
Fig. 8.4. (Contd)
145
0.1 0.2 0.3
1.0
Ratio d'lh = 0.15
<-

0.8

0.8

0.6
L
b
L
0.6
1 1
'" ..f2
.c:
.Q

0.4
(u)
MI(btffck)
0.2 0.3
0.1 0.2 0.3
1.0
Ratio d'ih 0.20
--

0.8

0.8
g:
?'-L
b
}
0.6
1
0.6
'" -.E
.c:
.Q

0.4 0.4
0.2
0.1 0.2 0.3
(v)
fig. S.4. (Contd)
146
0.1 0.2 0.3
1.0
Ratio d'ih = 0.05
..,f-

0.8

0.8

~
l
b
L
0.6 'I "1 0.6
-"
..!2
-<::
.C)
~
0.4 0.4
~ __________ ~ ____ ~ L - ~ __ ~ ~ __ ~ __ ~ __ ~ __ ~ ______________ ~ __ --d0.0
0.1 0.3
(w)
0.1 0.2 0.3
1.0
Ratio d'lh = 0.10
'"-

0.8

0.8
g
'-L
b
J 0.6
"
0.6
"
'"
-<::
.C)
~
0.4 0.4
0.1
(x)
MI(btftck)
0.2 0.3
8.4. (Contd)
147
Clause 7.3
148
Stresses due to temperature changes
A problem which occasionally has to be dealt with in deeper members is the estimation of the
stresses induced in a member due to non-uniform temperature distributions. For members
which are uncracked and unrestrained, well-established methods of analysis exist. Under the
influence of non-uniform and non-linear temperature changes, the member will change in
overall length and will also bend. The overall length change will be proportional to the
average temperature change. For any general section subject to any general distribution of
temperature, the average shortening will be given by
h
c
av
= a/A f biy dy
(DS.6)
o
where A is the total area of the section, h is the overall depth of the section, by is the breadth
of the section at depthy, ty is the temperature change at depthy and a is the coefficient of
thermal expansion.
The curvature can be calculated by taking moments about the section centroid. This gives
(DS.7)
where Y is the depth to the centroid of the section and I is the second moment of area of the
section.
The actual strain at level y is now given by
(DS.S)
The stress at any level can now be obtained from the difference between the actual strain
and the free temperature movement at that leveL Thus,
(DS.9)
While these equations may look very complicated, they can, in fact, be solved quite easily
by numerical methods. A very simple approach is to use a spreadsheet program.
The above method has been developed assuming that the member considered is unrestrained.
The effects of restraint can be introduced into the analysis by introducing springs at the
member ends which will resist axial and rotational movement. An iterative calculation is
then required to develop a situation giving equilibrium between the spring forces and the
forces in the member.
A similar approach can be adopted to deal with cracked sections subjected to temperature
change. The method becomes highly iterative, and is beyond the scope of this guide. In
general, cracking, by reducing the stiffness of the member, will lead to a major relaxation in
the stresses induced. It will therefore generally be satisfactory to ignore temperature-
induced stresses in cracked members.
8.l. Control of cracking
8.3.1. General
This section covers the material in clause 7.3 of EN 1992-1-1.
Crack width limits
There are many reasons for wishing to limit the widths of cracks to a relatively low value.
Among the most commonly cited reasons are:
to avoid possible corrosion damage to the reinforcement due to deleterious substances
penetrating to the reinforcement down the cracks
to avoid, or limit, leakage through cracks - this is commonly a critical design consideration
in water retaining structures
to avoid an unsightly appearance.
AU of the above reasons have been researched to some degree, but no clear definition of
permissible crack width has emerged from any of these studies. The following provides a very
brief summary of the results from these studies.

Cracking and corrosion. This is the most extensively researched area. Summaries of the
findings have been published by a number of authors (e.g. Beebyll and Schiessl and
Raupach12). The development of corrosion is a two-phase process. In fresh concrete the
reinforcement is protected from corrosion by the alkaline nature of the concrete.
This protection can be destroyed by two mechanisms: carbonation of the concrete to
the surface of the reinforcement or ingress of chlorides. Cracks will lead to a local
acceleration of both processes by permitting more rapid ingress of either carbon dioxide
or chlorides to the surface of the reinforcement. Once the protection provided by the
concrete has been destroyed, corrosion can start if the environmental conditions are
right. The period from construction up till the initiation of corrosion is usually referred
to as the 'initiation phase', while the period after the initiation of corrosion is termed the
'active phase'. The length of the initiation phase is likely to be influenced by crack width.
However, this period is likely to be short at a crack, and some corrosion can usually be
found on the bar surface where a crack reaches a bar after as little as 2 years even with
very small cracks. It is found, however, that this initial corrosion does not develop in
cases where the cracks are small or where the bars intersect the cracks. Quite possibly
the corrosion products block the cracks and inhibit further corrosion. A more serious
situation exists where a crack runs along the line of a bar. There is limited evidence to
suggest that, in this case, sustained corrosion may develop in salty environments where
the crack width exceeds about 0.3 mm.
Though less research has been done on the relation between cracking and corrosion in
prestressed concrete members, it is generally believed that the risks posed by cracks are
greater, and therefore more stringent criteria should be imposed. For this reason,
EN 1992-1-1 does not permit cracks to penetrate to the prestressing tendons where the
member is exposed to aggressive environments.
Leakage. Only very limited research has been carried out so far into this problem, and
this has not led to any agreed basis for crack width limits. Practical experience has
suggested that cracks of less than 0.2 mm width which pass right through a section will
leak somewhat initially but will quickly seal themselves. This problem is not specifically
considered in EN 1992-1-1 as liquid-retaining structures are covered in EN 1992-3.
Appearance. Limited studies suggest that noticeable cracks in structural members cause
concern to the occupants of structures, and it is therefore advisable to keep cracks below a
width that will not generally be noticed by a casual observer. On a smoothly finished
concrete surface, it appears that cracks are unlikely to lead to complaint if the maximum
width is kept below 0.4 mm. Clearly, larger widths may be used on rougher forms of
surface or where the cracking cannot be seen. This is mentioned in the note to Table 7.1 N.
The general considerations set out above led to the recommended limits on crack width
set out in EN 1992-1-1 in Table 7.1N. It should be noted, however, that the information on
which the criteria were based is far from being unambiguous. The actual values used in
particular countries can be defined in their National Annexes, but the values given in the
table have been accepted for buildings in the UK.
Causes of cracking
There are many possible causes of cracking and only a few of these lead to cracks that can be
controlled by measures taken during the design. The following are the more common causes.
Plastic shrinkage or plastic settlement. These are phenomena which occur within the
first few hours after casting while the concrete is still in a plastic state. The likelihood of
cracks being caused by these phenomena depends upon the bleeding rate of the mix and
the evaporation rate. The resulting cracks may be large: up to 2 mm is not uncommon.
149
150
.. Corrosion. Rust occupies a greater volume than the metal from which it is formed. Its
formation therefore causes internal pressures to build up around the bar surface, which
will lead to the formation of cracks running along the line of the corroding bars and,
eventually, spalling of the concrete cover.
Expansive chemical reactions within the concrete. Expansive reactions occurring at the
concrete surface tend to lead to scaling of the concrete rather than cracking; however,
some reactions, such as the alkali-silica reaction, occur within the body of the concrete
and can lead to large surface cracks.
Restrained deformations, such as shrinkage or temperature movements.
Loading.
Of this list, only the last two causes can be treated by the designer. They are probably the two
least serious causes of cracking.
8.3.2. Minimum areas of reinforcement
General principles
An essential condition for the validity of formulae for crack width calculation is that the
reinforcement remains elastic. If the reinforcement yields, then the deformation becomes
concentrated at the crack where the yielding is occurring, which is not the condition
considered in the derivation of the formulae. Particularly critical is the condition immediately
after formation of the first crack. Should formation of the first crack lead to yield, then only
one single crack will form, and all the deformation will be concentrated at this crack. The
principles involved here can be most easily demonstrated by considering a member subjected
to pure tension.
The force necessary to cause the member to crack is given by
Nr=Acfet
where N
r
is the cracking load, Ae is the area of concrete and fet is the tensile strength of the
concrete.
The strength of the steel isAJy. Thus, for the steel not to yield on first cracking, and hence
for spread cracking to develop,
AJy>AJet
or
As >AJC/fy
This provides the minimum reinforcement area required for controlled cracking. It can be
shown that where the cracking is caused exclusively by loading, this limitation is unimportant
since no cracks will form under service conditions. However, if tensile stresses may
be generated by the restraint of imposed deformations such as shrinkage or thermal
movements, then it is essential to ensure that at least the minimum reinforcement area is
provided. This result is true for flexure as well as for pure tension. Since it is only in rare cases
that it can be said with confidence that there is no possibility of stresses arising from
restraint, it is advisable to apply the rules for the minimum steel area in the great majority of
practical cases.
While the principles illustrated above for pure tension apply generally, the actual equations
will differ for different types of member. To avoid excessive complexity in the rules, a
factor kc is introduced into the above relationship to adjust for different forms of stress
distribution.
A further factor, k, is included to allow for the influence of internal self-equilibriating
stresses. These arise in cases where deformations of the concrete in the member itself are
restrained. The most common forms of such restrained deformation are deformations
resulting from either shrinkage or temperature change. These effects do not occur uniformly
throughout the section, occurring more rapidly near the member surface. As a result, the
o
o o
Shrinkage greatest
at surfaces
(a)
Shrinkage stress in member restrained at ends
o 0
o 0
Uniform stress due
to end restraints
+
(b)
Self-equilibrating stress
due to non-linear
shrinkage
Higher surface stress
Total stress
Fig. 8.5. Development of (a) shrinkage and (b) shrinkage-induced stresses in a reinforced-concrete
member
deformation of the surface concrete will be restrained by the interior concrete, and higher
tensions will be developed near the surface. This is illustrated in Fig. 8.5. The higher tensile
stresses at the surface will lead to cracking occurring at a lower load than would be predicted
on the basis of a linear distribution of stress across the section. Since the cracking load will be
smaller than predicted, a smaller amount of reinforcement will be necessary to ensure that
controlled cracking occurs. The function of the factor k is thus to reduce the minimum steel
area in cases where such non-linear stress distributions can occur.
Finally, it is necessary to consider what value should be chosen for the tensile strength of
the concrete, tet. It will be seen from the equations above that the minimum area of
reinforcement is proportional to the tensile strength of the concrete. It thus seems logical to
take as the tensile strength of the concrete an estimate of its likely maximum value. Such a
value is not easy to establish, and, furthermore, would lead to levels of reinforcement that
would be considered impractically high. For this reason, a more pragmatic approach has
been adopted, and the provisions of the code aim to give a minimum area which is not too
different from current practice.
The above considerations lead to equation (7.1) given in clause 7.3.2 for minimum
reinforcement:
Clause 7.3.2
151
152
As, mino-s + Alj 2:..o-p ~ ki-fct, effAjo-s
where
kc is a coefficient taking account of the form of the loading.
(D8.10)
k is a coefficient taking account of the possible presence of non-linear stress
distributions.
fct, eff is the mean value of the tensile strength of the concrete effective at the time when
the cracks first form. EN 1992-1-1 suggests that this should be taken as fctm, given in
Table 3.1, though a lower value may be used where it is clear that the cracking will
occur before 28 days.
Act is the area of concrete in tension immediately before the formation of the first
crack.
o-s is the steel stress, which can generally be taken as the characteristic yield strength
of the steel, though, as will be seen, there are cases where it may be more
convenient to use other, lower values.
Ap is the area of prestressing tendons within 150 mm of the ordinary bonded
reinforcement.
2:..0-p is the increase in stress in the prestressing tendons above that corresponding to
zero stress in the concrete at the same level.
~ J is a coefficient taking account of the different bond properties of the prestressing
tendons and the ordinary reinforcement.
Interpretation of the formula
EN 1992-1-1 provides a considerable amount of guidance on how to arrive at values for the
coefficients k and kc and the other parameters in the equation. This is summarized below:
Values for kc
kc takes a value of 1.0 for pure tension, 0.4 for pure flexure in rectangular reinforced concrete
sections or the webs of reinforced concrete box or flanged sections. The value of kc can be
reduced if there is a compressive force applied to the section, either by prestress or other
means. Similarly, kc will be increased where there is an applied tension.
Where, under the characteristic combination of loads, the stress in the section remains
more compressive thatfct eff and kc may be taken as zero, and no minimum reinforcement is
required. In other circumstances, values for kc may be calculated for rectangular sections and
the webs of flanged or box sections from
kc = 0.4{1- o-c/[kj(hlh')fct eff]} ::; 1.0
where
(D8.11)
o-c is the average stress in the concrete due to external forces (i.e. if the external force
is N, o-c = NIAJ
h' = h (the overall depth of the section in metres), where h < 1 m or 1.0 m where
h ~ 1.0 m.
k
j
is a coefficient taking account of the nature of the applied external force. kJ = 1.5 if
the applied force is compressive or 2h '/3h if it is tensile.
For a flange in tension of a flanged beam or box section, kc may be calculated from
kc = 0.9F
cr
IActict, eff ~ 0.5
where
(D8.12)
is the absolute value of the tensile force in the flange immediately prior to the
formation of the first crack. The cracking load is calculated on the basis of fct, eff'
Act is the area of the flange considered.
CHAPTERS. SERVICEABILITY LIMIT STATES
Values for k
The basic problem here is to establish whether the possible imposed deformations are
imposed from outside the element in question or whether they arise from shortening of the
member itself. In cases where the imposed deformation is external to the element such as the
settlement of a foundation, the question of non-linear distributions of strain across a section
does not arise, and k must take a value of 1.0. Where the deformations are generated within
the member considered, such as those caused by shrinkage or change in temperature of the
member, lower values of k will be appropriate. In this case, k should be taken as 1.0 for
members less than 300 mm deep or 0.65 for members of depth greater than 800 mm.
Intermediate values may be interpolated.
Mixtures of prestressing tendons and ordinary reinforcement
The coefficient ~ I takes account of the different surface characteristics of prestressing
tendons and ordinary reinforcement and also differences in diameter which will affect the
relative bond performance. ~ I is given by
(D8.13)
where ~ is the ratio of the bond strength of the prestressing tendons to that of the ordinary
reinforcement, CPs is the diameter of the ordinary reinforcement and CPp is the diameter of the
prestressing tendons. If there is no ordinary reinforcement, then ~ 1 = ; j ~ .
It will be seen that the estimation of the minimum reinforcement area has become a fairly
complex procedure; possibly unnecessarily so in view of the extreme uncertainty about many
aspects of the analysis. Nevertheless, it should be remembered that the areas of reinforcement
provided in large areas of most building structures (particularly slabs) will be governed by
the minimum reinforcement rules, and so these calculations will have a significant influence
on the overall economy of the structure. Furthermore, there is at least some discernible logic
behind the calculation compared with the arbitrary rules for minimum reinforcement given
in previous design codes. What is possibly more surprising and disappointing is that these
arbitrary rules are still maintained in Sections 8 and 9 of EN 1992-1-1.
S.l.3. Principles of the cracking phenomena
For simplicity, the development of cracking will be described in terms of a reinforced
concrete member subjected to pure axial tension.
Cracking caused by loading
If a continuously increasing tension is applied to a tension member, the first crack will form
when the tensile strength of the weakest section in the member is exceeded. The formation
of this crack leads to a local redistribution of stresses within the section. At the crack, all
the tensile force will be transferred to the reinforcement, and the stress in the concrete
immediately adjacent to the crack must clearly be zero. With increasing distance from the
crack, force is transferred by bond from the reinforcement to the concrete until, at some
distance, So, from the crack, the stress distribution within the section remains unchanged
from what it was before the crack formed. This local redistribution of forces in the region of
the crack is accompanied by an extension of the member. This extension, plus a minor
shortening of the concrete which has been relieved of the tensile stress it was supporting, is
accommodated in the crack. The crack thus opens up to a finite width immediately on its
formation. The formation of the crack and the resulting extension of the member also
reduces the stiffness of the member. As further load is applied, a second crack will form' at
the next weakest section, though it will not form within So of the first crack since the stresses
within this region will have been reduced by the formation of the first crack. Further
increases in loading will lead to the formation of further cracks until, eventually, there is no
remaining area of the member surface which is not within So of a previously formed crack.
The formation of each crack will lead to a reduction in the member stiffness. After all the
153
154
cracks have formed, further loading will result in a widening of the existing cracks but no new
cracks. Stresses in the concrete will be relieved by limited bond slip near the crack faces and
by the formation of internal cracks. This process leads to further reductions in stiffness, but,
clearly, the stiffness cannot reduce to below that of the bare reinforcement. Figure 8.6
illustrates the behaviour described above. The stepped nature of the load-deformation
response is not normally discernible in tests, and the response is usually shown as a smooth
curve.
Cracking due to imposed deformations
If the member is subjected to a steadily increasing strain rather than a steadily increasing
load, the response illustrated in Fig. 8.6 is radically changed. When a crack forms, the
reduction of stiffness resulting from the formation of the crack leads to a reduction in the
Cracking
load
Innial
width
Initial width
Deformation
(a)
Deformation
(b)
fig. 8.6. (a) Load-deformation response of a member subjected to a steadily increasing load.
(b) Crack width-deformation response in a load-controlled test
'C
~
...J
1
1/
I
Formation of cracks
/'
/ /'
/ /'
/ /'
Deformation
(a)
Average width for ;"l
load-controlled lest "'" -::::-./
\. ".'" V
\. "'" .
~ Apparent average width
in slrain-controlied lest
Deformation
(b)
Fig. 8.1. (a) Load-deformation response for a member subjected to steadily increasing deformation.
(b) Crack width-deformation response for a member subjected to steadily increasing deformation
tensile force supported by the member. This, in turn, leads to a reduction in the tensile
stresses at all points along the member. Only when the strain has increased sufficiently to
develop the tensile stress equal to the tensile strength of the next weakest point on the basis
of the reduced stiffness will the next crack form. This results in the 'saw-toothed' load-
deformation and load-crack width relationships shown in Fig. S.7.
8.3.4. Derivation of crack prediction formulae
The development of formulae for the prediction of crack widths given in clause 7.3.4 of
EN 1992-1-1 follows from the description of the cracking phenomenon given above. If it is
assumed that all the extension occurring when a crack forms is accommodated in that crack,
then, when all the cracks have formed, the crack width will be given by the following
relationship, which is simply a statement of compatibility:
(DS.14)
where w is crack width, Srrn is the average crack spacing and C
m
is the average strain.
The average strain can be more rigorously stated to be equal to the strain in the reinforcement,
taking account of tension stiffening, C
sm
' less the average strain in the concrete at the surface,
Clause 7.3.4
155
156
Cern' Since, in design, it is a maximum width of crack which is required rather than the average,
the final formula given in EN 1992-1-1 is
(D8.15)
Since no crack can form within So of an existing crack, this defines the minimum spacing of
the cracks. The maximum spacing is 2S
o
, since if a spacing existed wider than this, a further
crack could form. It follows that the average crack spacing will lie between So and 2S
o
' It is
frequently assumed to be 1.5S
o
.
It is in the calculation of Srm that the most significant differences arise between the
formulae in national codes. The distance So, and hence Srm' depends on the rate at which
stress can be transferred from the reinforcement, which is carrying all the force at a crack, to
the concrete. This transfer is effected by bond stresses on the bar surface. If the bond stress is
assumed to be constant along the length So and that the stress will just reach the tensile
strength of the concrete at a distance So from a crack, then
Tf\So = A Jet
where T is the bond stress,A
e
is the area of concrete,!et is the tensile strength of concrete and
is the bar diameter.
Taking P = 1I"2/4A
c
and substituting for Ac gives
So = !
et
1l"!4p
From this,
Srm = 0.25k!p
where k is a constant depending on the bond characteristics of the reinforcement. This is the
oldest form of relationship for the prediction of crack spacings. More recent studies have
shown that the cover also has a significant influence, and that a better agreement with test
results is obtained from an equation of the form
Srm = kc + 0.25k]! p
where c is the cover. This formula has been derived for members subject to pure tension. In
order to be able to apply it to bending, it is necessary to introduce a further coefficient, k
2
,
and to define an effective reinforcement ratio, Peff' These modifications take account of the
different form of the stress distribution within the tension zone and the fact that only part of
the section is in tension. k2 and Peff can be derived empirically from tests. The resulting
formula is
(D8.16)
Here, k) is a coefficient taking account of the bond properties of the reinforcement. A value
of 0.8 is taken for high bond bars and 1.6 for smooth bars. k2 is a coefficient depending on the
form of the stress distribution. A value of 0.5 is taken for bending and 1 for pure tension.
Intermediate values can be obtained from
(D8.17)
where C) and Cz are, respectively, the greater and lesser tensile strains at the faces of the
member. Peff is the effective reinforcement ratio, where As is the area of tension
reinforcement contained within the effective area of concrete in tension, Ac, eff' This is the
area of concrete in tension surrounding the reinforcement of a depth equal to 2.5 times the
distance from the tension face of the member to the centroid of the tension reinforcement. A
figure in the code gives definitions for other, less typical cases.
In design, it is not the average crack width which is required but a value which is unlikely to
be exceeded. EN 1992-1-1 uses the characteristic crack width, which is defined as a width
with a 5% probability of exceedance. It is found experimentally that a reasonable estimate of
the characteristic width is obtained if the maximum crack spacing is assumed to be 1.7 times
the average value. In EN 1992-1-1, therefore, the maximum spacing is used, and this is
assumed to be given by
(D8.18)
The other parameter in the crack width equation is the average strain, G
sm
- Gem' This is
obtained from equation (7.9) in EN 1992-1-1, and is repeated below for convenience:
G
sm
- Gem = O"s IEs - kJet, efi 1 + ();eP
p
, eff)IEsp
p
, eff:2: 0.60"s IEs (D8.19)
where o"s is the stress in the tension reinforcement calculated assuming a cracked section, ();e
is the modular ratio (E/E
c
)' and k
t
is a factor depending on the duration of the load (0.6 for
short-term loads and 0.4 for long-term loads).
There remains a situation where the formula described above can lead to a significant
overestimate of the likely cracking. The reason for this may be understood by considering the
element shown in Fig. 8.5. This is an unreinforced element subjected to an axial load applied
at an eccentricity sufficiently large for part of the section to be in tension. If the load is
sufficient, the section will crack. The formation of this crack will not result in failure but
merely a redistribution of stresses locally to the crack. Some distance away from this crack
the stresses will remain unaffected by the crack. It is found that the crack affects the stress
distribution within a distance roughly equal to the height of the crack on either side of the
crack. Thus, by the same arguments used earlier, the spacing of the cracks can eventually be
expected to be between the crack height and twice the crack height. This leads to the
following relationship:
Srm=h-x (D8.20)
where h is the overall depth of the section andx is the depth of the neutral axis. This formula
applies not only to members subject to axial compression but also to any situation where the
cracks, when they form, do not pass right through the section. The effect of bonded
reinforcement in the section is almost always to give a calculated crack spacing much smaller
than given by the above equation. The equation does, however, give a maximum, or limiting,
value for the spacing, and there are a number of practical situations where it can be used with
advantage. One particular case is that of a prestressed beam without any bonded normal
reinforcement. The bond of prestressing tendons or wires is often much inferior to that on
normal high bond reinforcement. A safe estimate of the crack spacing, and hence the crack
width, can be made by treating the prestress as an external load, calculating the depth of the
tension zone under the loading considered and applying the above formula. Clause 4.4.2.4(8) Clause 4.4.2.4(8)
permits this procedure.
8.l.S. Checking cracking without direct calculation
Flexural or tension cracking can normally be controlled by application of simplified detailing
rules (clause 7.3.3). These take the form of tables of maximum bar diameters or maximum Clause 7.3.3
bar spacings. The tables (Tables 7.2 and 7.3 in EN 1992-1-1) were derived from parameter
studies carried out using the crack width calculation formulae described above. It should be
noted that, for load-induced cracking, either Table 7.2 or Table 7.3 may be used; it is not
necessary to satisfy both tables. For cracking due to restrained imposed deformations, Table
7.2 should be used. For reinforced concrete, the tables can be presented graphically, as in
Fig. 8.8. For load-induced cracking, the steel stress is the stress under the quasi-permanent
combination of loading. This may be calculated approximately from the relation:
(D8.2,l)
where Js
qp
is the steel stress under the quasi-permanent load, Nqp is the quasi-permanent load
and Nud is the design ultimate load.
For prestressed concrete beams containing bonded untensioned reinforcement, it is
possible to make an estimate of the stress in the untensioned steel from Figure 8.9. This is
done by calculating a fictitious value for Jet, eff equal to the tensile stress in the concrete at the
157
158
Stress in reinforcement under quasi-permanent load: N/mm2
1oo __________ ______
300 r-
100 200 300
Stress in reinforcement under quasi-permanent load: N/mm2
Fig. 8.S. Maximum bar sizes and spacings for crack control
-1.0 o
25
20
12
c
o

"c

0.4
0.3
-1.0
O
A2
face
Al
Most tensile face
Compression
o
NI(bwhfct,eff)
Tension
Fig. 8.9. Design chart for assessing minimum areas of reinforcement
0.4
1.0
beam soffit calculated on the assumption that the concrete is uncracked. A value of N/bhfet, eff
can now be calculated and a value for PUs/fet, elf' Substituting for P andfet, eff will give a value for
Us' An example will illustrate the procedure.
Example 8.2
Carry out a check for cracking for a prestressed T beam having a rib breadth of 250 mm,
an overall depth of 600 mm, a flange breadth of 1000 mm and a flange depth of 180 mm.
The prestress N after allowance for losses is 700 kN at an eccentricity of 280 mm and the
moment under the frequent combination of loads is 330 kN m.
It can be found from standard tables that the section modulus for stresses at the bottom
of the section is 0.222bh
2
:::: 20 X 10
6
, and the area of the section is 285 x 1(fl mm
2
The
stress at the soffit of the section is thus given by
fet eff = 700/285 + 700 x 0.28/20 - 330/20 :::: -4.09
Hence
N/AJct eff = 700/285/4.09 = 0.6
From Fig. 8Ac,
Aso-sfbrhfct,eff = 0.08
Hence,
Asu
s
:::: 0.08 x 250 x 600 x 4.09 :::: 49080
Assume we choose 2 x No. 12 mm bars. This gives As = 226 mm
2
, and hence
is = 217 N/mm2. For this stress, Table 4. Z 1 in EN 1992-1-1 gives a maximum bar diameter
of between 12 and 16 mm, so the design is satisfactory. Table 4.12 gives a maximum bar
spacing of about 130 mm, which would be difficult to achieve with only two bars.
However, since Table 4.11 has been satisfied, this does not matter.
It is, of course, possible to calculate the stress in the reinforcement rigorously on the
basis of a cracked section. This would have given a significantly lower stress if proper
allowance had been made for the prestressing tendons in the section, so the proposed
method is on the safe side. Using the rigorously calculated stress and the formulae given
for the calculation of crack width results in a width below 0.2 mm.
EN 1992-1-1 gives no way of using the tables where there is no untensioned reinforcement
in the section. By interpreting the rules given in clause 4.4.2.4(4), it seems a reasonable
approximation to calculate a steel stress as described above. This will be a change in stress
in the tendons beyond the condition where the concrete at the level of the tendons is at
zero stress. To allow for the inferior bond of the tendons, the bar diameters given by Table
4. Z Z or the bar spacings from Table 4.12 should be divided by 2.5.
8.3.6. Checking cracking by direct calculation
The formula for calculating the design crack width has been discussed in Section 8.3.3, and
no further discussion is considered necessary.
159
CHAPTER 9
Durability
9.1. General
The Construction Products Directive defines certain essential requirements relating to
fitness for purpose, mechanical resistance and stability and various other factors. These
requirements 'must, subject to normal maintenance, be satisfied for an economically reasonable
working life'. A structure having adequate durability is one which satisfies this last stipulation.
This chapter is concerned with the rules given in EN 1992-1-1 and associated documents to
ensure that adequate durability is achieved.
Design for durability is not covered fully by EN 1992-1-1, and reference needs to be made
to a number of other standards. These are:

BS EN 206-1 (Concrete, Part 1: Specification, Peiformance, Production and Conformity).
BS 8500-1 (Concrete - Complementary British Standard to BS EN 206-1, Part 1:2002,
Method of Specifying and Guidance for the Specifier).
BS 8500-2 (Concrete - Complementary British Standard to BS EN 206-1, Part 2:2002,
Specification for Constituent Materials and Concrete).
ISO 9690 (Production and Control of Concrete. Classification of Chemically Aggressive
Environmental Conditions Affecting Concrete).
In fact, BS EN 206 is not, in itself, a very helpful document for the designer or specifier and
all the information needed for specification of concrete in the UK is included in the
complementary British standard, BS 8500.
In EN 1992-1-1, durability is mainly covered in Section 4 ('Durability and cover to
reinforcement'), but parts of clauses 7.2 (on stresses) and 7.3 (on cracking) are also relevant.
A good source of general information on durability is CEB Design Guide on Durable
Concrete Structures. 13
9.1.1. Historical perspective
Up until about the mid-1970s, durability was not seen as a serious issue for concrete. While
all codes gave minimum covers for protection of the reinforcement, it was rarely felt
necessary to do more than this. For example, inspection of the contents list of the CEB-FIP
1978 Model Code,? which formed the original base document for the drafting of EN 1992-1-
1, will reveal no explicit mention of durability. During the 1970s, however, durability
problems arose in very many countries. This has resulted in a complete change in attitude to
the design and construction of concrete structures. A few of the more notable of these
problems are:
Very serious deterioration of bridge decks in the USA due to corrosion of the
reinforcement contaminated by de-icing salts.
Clause 7.2
Clause 7.3
162
e Major deterioration problems in the Middle East due to chloride-induced corrosion in a
particularly aggressive environment.
Major problems in the UK in the 19708 due to reductions in strength of high-alumina
cement with time. More recently, this has also become a major problem in Spain.
Severe cracking in structures in many countries resulting from the alkali-silica reaction.
While the number of affected structures may not have been great, the problem generated
a large amount of publicity.
Major problems resulting from corrosion initiated by de-icing salts on bridges in almost
all countries where these are used in significant quantities in winter.
As a result of these problems, durability and, in particular, corrosion of reinforcement was
probably the major area of research in the structural field during the 1980s. Inevitably, the
drafters of codes of practice also found it essential to treat the subject much more fully. The
Eurocodes have followed this trend.
9.1.2. Common mechanisms leading to the deterioration of concrete
structures
This section lists the main deterioration mechanisms which may need to be considered in
design, gives a brief description of the phenomena involved and the possible methods for
dealing with them.
Corrosion of reinforcement or prestressing tendons
In normal circumstances, the highly alkaline nature of concrete protects steel embedded
within it. Except under the circumstances discussed below, the pH value of the pore solution
in concrete is in the region of 12-14. Steel will not generally corrode in uncontaminated
concrete until the pH drops below 10. The protection is afforded by the formation of a very
thin, coherent layer of iron oxide over the surface of the bar under alkaline conditions. Steel
protected in this way is described as being in a passive state. Two mechanisms can lead to the
destruction of this passive state. These are:
(1) Carbonation of the concrete. This is a reaction between carbon dioxide in the atmosphere
and the alkalis in the cement matrix. This process starts at the surface, and, with time,
penetrates slowly into the concrete. The rate of penetration of carbonation into the
concrete depends upon the environment and the quality of the concrete. The rate is
fastest where the relative humidity is in the range 50-60%. It is slower at higher
humidities, being effectively zero at 100%. Good-quality concrete carbonates more
slowly than poor quality material. The speed of the phenomenon depends on the rate at
which carbon dioxide can diffuse into the concrete. This will decrease with a decrease in
the water/cement ratio and hence with an increase in strength. The effect of carbonation
is to reduce the alkalinity of the concrete to a level where the natural protection
is lost, and corrosion may then occur if the concrete immediately surrounding the
reinforcement is carbonated.
(2) The presence of chlorides in the concrete. Chlorides have the capacity to destroy the
passivity of steel even where the alkalinity remains high. This usually occurs locally,
giving rise to 'pitting corrosion'. Chlorides may get into the concrete from various
sources, but the commonest are seawater in marine environments, de-icing chemicals on
roads and additives such as calcium chloride, which was used extensively in the past as an
accelerator. The rate at which chlorides penetrate into concrete depends upon the rate
of application of chlorides to the concrete surface and, as with carbonation, on the
quality of the concrete.
Once the passivity of the steel has been destroyed, corrosion can occur if there is
(1) sufficient moisture and (2) sufficient oxygen. It is found that these two requirements
can act against each other since, if the concrete is wet, oxygen cannot penetrate and, if it is
CHAPTER 9. DURABILITY
dry so that there is plentiful oxygen, there is insufficient moisture for the reaction to
progress. As a result, the greatest risk of corrosion is in members subjected to wetting and
drying.
The normal way to design against corrosion is to ensure that there is an adequate cover to
the reinforcement and that the concrete in the cover region is of a high quality and is well
cured. In particularly aggressive environments, however, there are other, more expensive
measures which may be taken. Possibilities are:
Use reinforcement coated with epoxy or similar. Over recent years, this approach seems
to have been somewhat discredited due to some high-profile problem cases.
Use stainless steel reinforcement.
Apply surface coatings to the concrete to inhibit the ingress of chlorides or carbon
dioxide. Such coatings would have to be meticulously maintained to be successful for
long periods of time.
Apply cathodic protection to the structure.
One major factor in the avoidance of corrosion problems is the form of the structure. Areas
of exposed concrete on which water can stand or can drain across are particularly at risk.
Frost attack
If saturated concrete is subjected to frequent freezing and thawing, the expansive effects of
ice formation will disrupt the concrete. The usual manifestations offrost damage are surface
spalling or the formation of systems of closely spaced surface cracks. These cracks can be
seen as the precursors of spalling. Concrete which is not close to being saturated is not at risk
from frost as the expansion that occurs on freezing can be accommodated in the non-water-
filled pores. Frost damage can be avoided by the following methods:
Protecting the concrete from saturation.
Using an air-entrained concrete mix. The small bubbles within the matrix can provide
pressure relief.
Using high-strength concrete. Concrete with a strength of 45 N/mm
2
or more is generally
immune to frost damage.
Alkali-aggregate reactions
There are two basic forms of reaction which occasionally occur and can damage concrete:
the alkali-silica reaction and the alkali-carbonate reaction. The alkali-silica reaction is the
more common. It is a reaction between the alkalis in the cement and certain forms of silica
which results in the formation of a hygroscopic silica geL This gel takes up water and
expands, causing cracking. Identification of the alkali-silica reaction as the cause of damage
is not straightforward. The first stage is to establish that silica gel has formed within the
matrix. This can only be done by taking thin sections of the concrete and examining them
under a microscope to establish the existence of gel-filled microcracks. Establishing the
presence of gel is not sufficient proof in itself that the alkali-silica reaction is the cause of the
observed damage as limited amounts of gel can frequently be present in concrete without any
deleterious consequences. There needs to be evidence of large quantities of gel. Externally,
the effects of these reactions are the formation of cracks, which may be large (several
millimetres wide is not uncommon). In relatively unstressed and unreinforced concrete,
these cracks can form a random 'map' pattern. In other cases they will tend to form parallel
to the direction of compressive stress or reinforcement. Though large, the cracks are usually
not deep, only extending 50-70 mm into the section. Their effect on structural performance
is not as great as might be imagined from looking at the cracks. A reduction in the
compressive and tensile strength of the concrete occurs, but this is commonly not more than
about 20-30%.
The alkali-silica reaction can be avoided by three methods:
163
164
(1) Using aggregates which experience has shown to perform satisfactorily. No fully adequate
test has yet been devised for assessing the potential reactivity of aggregates, so experience
remains the only reliable method.
(2) Using cement with a low alkali content. The level which will avoid problems is likely to
vary from country to country, depending on the geology. In the UK, a limit of 3 kg/m
3
of
sodium oxide equivalent is considered adequate.
(3) Inhibiting the ingress of water.
Attack from sulphates
In the presence of water, sulphate ions can react with the tricalcium aluminate component of
the matrix. This reaction causes expansion, leading to cracking and eventual disintegration
of the concrete. The commonest source of sulphates is in the earth surrounding foundations
but other sources are sometimes significant. Seawater contains significant amounts of
sulphate but the presence of chlorides renders it harmless.
If it is established that there is a risk of sulphate attack, the possible measures to avoid it
are:
the use of sulphate-resisting cement (Portland cement having a low tricalcium aluminate
content)
the use of blended cements incorporating ground granulated blastfurnace slag
for medium levels of risk, the use of additions such as fly ash or pozzolans are effective
in particularly aggressive situations, it may be necessary to provide an impervious
coating to the concrete.
Acid attack
Acids attack the calcium compounds in concrete, converting them to soluble salts, which can
then leach away. The effect of acids is therefore to eat away, or render weak and permeable,
the surface of the concrete. Very substantial amounts of acid are required to do serious
damage to concrete. Acid rain, for example, will do no more than etch the surface of
concrete over any reasonable design life. If the concrete is likely to be exposed to major
amounts of acid, for example from some industrial process, the only way to avoid damage is
to provide an impermeable coating to the concrete.
Leaching by soft water
This process is in effect a mild version of acid attack. Calcium compounds (e.g. calcium
carbonate and calcium hydroxide) are weakly soluble in soft water, and can be leached out if
the concrete is constantly exposed to running soft water. The process is very slow, but, in
time, can lead to exposed concrete surfaces taking on an 'exposed aggregate' appearance.
Abrasion
Abrasion of concrete surfaces may occur due to trafficking of the concrete or due to sand or
gravel suspended in turbulent water.
Resistance to abrasion can be obtained by using higher-strength concrete and abrasion-
resistant aggregates. Resistance is also markedly improved by good curing of surfaces likely
to be exposed to abrasive action.
9.1.3. Relative importance of deterioration mechanisms
Clearly, the relative importance of the various mechanisms will vary from country to country
and even region to region; and no generally applicable ordering of the mechanisms can be
made. However, there seems no doubt that the commonest and most serious form of
degredation worldwide is corrosion of reinforcement. It can also be stated that, of the two
CHAPTER 9. DURABILITY
initiating mechanisms for corrosion, carbonation and chlorides, chlorides have led to the
greater amount of damage by far.
9.2. Design for durability
There are two basic steps in designing for durability. These are:
(1) Establish the aggressivity of the environment to which the member is exposed. This is
analogous to establishing the design loading where the ultimate or serviceability limit
states are being considered.
(2) Select materials and design the structure to be able to resist the environment for a
reasonable lifetime.
The aggressivity of the environment should, in principle, be defined separately for each
degredation mechanism, since the factors acting to promote one form of degredation are not
necessarily the same as those promoting another. EN 1992-1-1 does this in Table 4.1. This
classifies environments into six basic classes. These are:
(1) no risk of corrosion or attack
(2) corrosion induced by carbonation
(3) corrosion induced by chlorides
(4) corrosion induced by chlorides from sea water
(5) freeze-thaw attack
(6) chemical attack.
Each of these basic classifications, which cover the common forms of attack, are subdivided
into subclasses which define the severity of the risk of attack. For example, Class 3,
'corrosion induced by chlorides', is divided into three subclasses of increasing severity: XD1,
'moderate humidity'; XD2, 'wet, rarely dry'; and XD3, 'cyclic wet and dry'. Each of the
environment classes are defined by examples.
Durability is, however, difficult to define on a uniform basis for all countries which
might use EN 1992-1-1, and therefore the durability provisions are all Nationally Defined
Parameters. The Nationally Defined Parameters for durability in the UK are set out in
the UK National Annex, and are reproduced here in Tables 9.1 and 9.2 (overleaf) for
convenience. Table 9.1 gives provisions for a 50 year design life while Table 9.2 gives values
for a 100 year design life.
Measures relating to the selection of materials and design of a suitable concrete mix are
most conveniently found in BS 8500, while measures relating to the design of the structure,
such as cover to the reinforcement, are included in EN 1992-1-1. The design of a durable
structure thus requires attention to the rules in both documents
For the selection of the quality of concrete subject to freeze-thaw and concrete in
aggressive ground conditions, reference should be made to Annex A of BS 8500-1.
165
0'
0'
Table 9.1. Recommendations for normal-weight concrete quality for selected exposure classes and cover to reinforcement for a 50 year intended working life and
20 mm maximum aggregate size
Equivalent designated concrete: strength class, maximum water/cement ratio, minimum cement or combination content (kg/m
3
)
Cement/
Exposure Exposure combination
Nominal cover to reinforcement including prestressing steel:
class conditions types
.
15 + D.c 20 + D.c 25 + D.c 30 + D.c 35 + D.c 40 + D.c 45 + D.c 50 + D.c
Carbonation-induced corrosion
XC2 Wet, rarely All C25/30,
dry 0.65,
260
or RC30
XC3 Moderate All except C40/50 C32/40 C28/35 C25/30
humidity IVB 0.45, 0.55, 0.60, 0.65,
XC4 CycliC wet 340 300 280 260
and dry or RC50 or RC40 or RC35 or RC30
Chloride-induced corrosion excluding chlorides from seawater
XDI Moderate All C40/50, C32/40 C28/35
humidity 0.45, 0.55, 0.60,
360 320 300
XD2 Wet, rarely I, IIA, IIB-S, C40/50 C32/40 C28/35
dry SRPC 0.40, 0.50, 0.55,
380 340 320
liB-V, lilA C35/45 C28/35 C25/30
0.40, 0.50, 0.55,
380 340 320
IIIB,IVB C32/40 C25/30 C20/25
0.40, 0.50, 0.55,
380 340 320
XD3 Cyclic wet I,IIA,IIB-S, C45/55 C40/50 C35/45
and dry SRPC 0.35, 0.40, 0.45,
380 380 360
liB-V, iliA C35/45 C32/40 C28/35
0.40, 0.45, 0.50,
380 360 340
o
m
VI
G)
Z
m
~
G)
C
o
m
-l
o
m
Z
..s;)
..s;)
IV
,

z
o
m
Z
..s;)
..s;)
IV
,
,
IV
IIIB,IVB C32/40 C28/35 C25/30
0.40, 0.45, 0.50,
380 360 340
Seawater-induced corrosion
XSI Airborne I, IIA, IIB-S, C50/60 C40/50 C35/45
salts but SRPC 0.35, 0.45, 0.50,
no direct 380 360 340
contact
liB-V, IliA C45/55 C35/45 C32/40
0.35, 0.45, 0.50,
380 360 340
IIIB,IVB C35/45 C28/35 C25/30
0.40, 0.50, 0.55,
380 340 320
XS2 Wet, rarely I, IIA, IIB-S, C40/50 C32/40 C28/35
dry SRPC 0.40, 0.50, 0.55,
380 340 320
liB-V, lilA C35/45 C28/35 C25/30
0.40, 0.50, 0.55,
380 340 320
IIIB,IVB C32/40 C25/30 C20/25
0.40, 0.50, 0.55,
380 340 320
XS3 Tidal, splash I, IIA, IIB-S, C45/55 C40/50
and spray SRPC 0.35, 0.40,
zones 380 380
liB-V, iliA C35/45 C32/40 C28/35
0.40, 0.45, 0.50,
380 360 340
IIIB,IVB C32/40 C28/35 C25/30
()
0.40, 0.45, 0.50, I
380 360 340
):>
-0
-I
'See BS 8500-2:2002, Table I.
m
;;0
indicates that the quality of concrete given in the cell to the left should not be reduced. :-0
0
C
~
OJ
r
C1'
I ~ .....
C/'
00
Table 9.2. Recommendations for normal-weight concrete quality for selected exposure classes and cover to reinforcement for a 100 year intended working life and
20 mm maximum aggregate size
Equivalent designated concrete: strength class, maximum water/cement ratio, minimum cement or combination content (kg/m
3
)
Cement/
Exposure Exposure combination
Nominal cover to reinforcement including prestressing steel:
class conditions types
.
15 + fl.c 20 + fl.c 25 + fl.c 30 + fl.c 35 + fl.c 40 + fl.c 45 + fl.c 50 + fl.c
Carbonation-induced corrosion
XCI Dry of All C20/25,
permanently 0.70,
wet 240
or RC25
XC2 Wet, rarely All C25/30,
dry 0.65,
260
or RC30
XC3 Moderate All except C40/50, C35/45, C32/40, C28/35,
humidity IVB 0.45, 0.50, 0.55, 0.60,
XC4
Cyclic wet
340 320 300 280
and dry
or RC50 or RC45 or RC40 or RC35
'See BS 8500-2:2002, Table I.
indicates that the quality of concrete given in the cell to the left should not be reduced.
o
m
VI
C)
Z
m
~
G)
C
o
m
-I
o
m
Z
..0
..0
IV
,

z
o
m
Z
..0
..0
IV
,
,
IV
CHAPTER 10
Detailing
I 0.1. General
The main guidance on detailing is contained in Sections 8 and 9 of EN 1992-1-1. Cover
requirements are in Section 4, under durability.
The main aspects of detailing are concerned with anchoring bars carrying stress and
lapping of reinforcement. Most of the provisions regarding anchorage and laps were reputedly
deduced from tests carried out in many countries, principally in Austria, France, Germany,
Sweden and the USA. It is understood that the global safety factor I'm I'f against the 5%
fractile of test results is 2.1. As I'm = 1.5 for concrete, this gives a value of I'f = 1.4 for loads,
which is clearly considered acceptable in EN 1992-1-1.
Anchorage and lap requirements are checked at the ultimate limit stage. The general
detailing provisions are deemed to ensure satisfactory behaviour of the structure at serviceability
conditions.
The code covers (1) high-bond bars, (2) welded mesh and (3) different types of structural
members. Bars may be used in bundles with certain limitations.
In the main, the detailing requirements are governed by bond-related phenomena, which
are significantly influenced by:
..

the surface characteristics of the bars (plain, ribbed)
the shape of the bars (straight, with hooks or bends)
the presence of welded transverse bars
the confinement offered by concrete (mainly controlled by the size of the concrete cover
in relation to the bar diameter)
the confinement offered by non-welded transverse reinforcement (such as links)
the confinement offered by transverse pressure.
The rules governing detailing make allowance for the above. Particular emphasis is placed
on the need for adequate concrete cover and transverse reinforcement to cater for tensile
stresses in concrete in regions of high bond stresses.
Bond stresses for high-bond bars are a function of the tensile strength of concrete (fclk)'
The guidance for detailing different types of members includes requirements for minimum
areas of reinforcement. This is stipulated in order to (1) prevent a brittle failure, (2) prevent
wide cracks and (3) resist stresses arising from temperature effects, shrinkage and other
restrained actions.
10.2. Discussion of the general requirements
In this section the main features of the detailing requirements are arranged in a practical
order and discussed.
Clause 8.3(3)
170
10.2.1. Covel" to bar reinforcement
See Chapter 9 of this guide for the minimum cover to reinforcement to meet durability
requirements. See Chapter 12 for cover requirements for fire resistance.
10.2.2. Spacing of bars
The basic principle is that the reinforcement bars in a member should be arranged in such a
way that concrete can be placed and compacted satisfactorily so that adequate bond will
develop between the bars and concrete.
Figure 10.1 defines the spacing Smin between bars in a layer and that between layers of
reinforcement. Srnin should be taken as diameter of the bar or diameter of the aggregate plus
5 or 20 mm, whichever is the largest.
10.2.3. Mandrel diameters for bars
EN 1992-1-1 prescribes separate mandrel diameters for (1) avoiding damage to the
reinforcement as a result of bending the bars and (2) for avoiding damage to the concrete
when a hook or a bend carries stress.
For avoiding damage to the bars themselves, the diameter of the mandrel is 4 for bars of
16 mm diameter or less and 7 for larger bars.
When a bent bar carries stress, compressive stresses are generated inside the bend. For a
given force the stress inside the bend is inversely related to the diameter of the bend (i.e. the
stress decreases with an increase in the diameter of the bend). EN 1992-1-1 gives a formula
relating the diameter of the mandrel, diameter of the bar, the force in the bar and the spacing
of the bars. Table 10.1 sets out the same information in a readily usable form for different bar
spacing and concrete grades.
10.2.4. Basic anchorage length
In EN 1992-1-1 the parameter lb, reqd (basic anchorage length) is used in the calculation of
design anchorage and lap lengths:
lb, reqd = ( <PI 4) ( (j sd /fbd)
where is the diameter of the bar, (jsd is the stress in the bar and fbd is the design value of
ultimate bond stress. The axial force in the bar is assumed to be distributed uniformly over
this length through bond stress between the reinforcement and concrete.
The ultimate bond stress basically depends upon the tensile strength of concrete and the
location of the bar within the concrete. The latter is referred to as 'bond condition'. EN 1992-
1-1 defines 'good' and 'poor' bond conditions, and these are shown in Fig. 10.2. There is also
test evidence to show that the ultimate bond stress has some dependence on the size of the
bar. EN 1992-1-1 de-rates the bond stress for large-diameter bars. In the UK National Annex
it is defined as bars larger than 40 mm. The above factors are reflected in the expression for
the ultimate stress:
fbd = 2.
25
rIJTJlfctk,O.oshJ
where TJl is 1.0 for 'good' bond conditions and 0.7 for 'poor' bond conditions and TJ2 = 1.0 for
bar :::; 32 mm and (132 - )/100 for bar > 32 mm. In the UK this expression should strictly
be (140 - )/100.
Table 10.2 shows the ultimate bond stress for different grades of concrete together with
values of basic anchorage lengths assuming full design stress in the bars.
10.3. Anchorage of longitudinal bars
When reinforcement is designed to carry stress, it needs to be anchored into adjacent parts
such that (1) the required stress will be able to develop and (2) the force in the bar is safely
transmitted to the surrounding concrete without causing longitudinal cracks or spalling.
CHAPTER 10. DETAILING
Fig. 10.1. Spacing of bars
10 0 o l ~
~
w
but sufficient for
insertion 01 vibrator
Table 10.1. Mandrel diameters of curved bars as mUltiples of bar diameter
Concrete grade (MPa)
b
20 25 30 35 40 45
2 25.6 20.5 17.1 14.6 12.8 11.4
3 21.3 17.0 14.2 12.1 10.6 9.5
4 19.2 15.4 12.8 11.0 9.6 8.6
5 18.0 14.4 12.0 10.2 9.0 8.0
10 15.4 12.3 10.3 8.8 7.7 6.8
50 ;:: 55
10.2 9.3
8.5 7.7
7.7 7.0
7.0 6.5
6.1 5.6
(I) 0b is half the centre-to-centre distance between bars. For a bar adjacent to the face of a member, 0b
is the cover plus half the diameter of the bar.
(2) The table assumes that the stress in the bar is fykhs. i.e. 500/1.15 MPa. The table values may be
multiplied by As, reqd lAs, prov;ded'
(a)
(b)
I
~ ~
250mm
(e)
' = m m I ~ I
(d)
fig. 10.2. Bond conditions: (a) 45 < a < 90 for all h values; (b) h < 250 mm; (c) 250 mm < h < 600 mm;
(d) h > 600 mm (hatched zones have poor bond conditions)
171
172
Generally, transverse reinforcement will be required in the anchorage zone to resist the
secondary forces induced locally.
Normal methods of anchorage are shown in Fig. 10.3.
Table 10.2. Bond stress and basic anchorage length for different concrete grades
Concrete grade (MPa)
20 25 30 35 40 45 50 55 2:: 60
fbd (MPa) 2.25 2.70 3.00 3.30 3.75 4.05 4.35 4.50 4.65
('good' bond
conditions)
lb. rqd
48 40 36 33 29 27 25 24 23
(,good'bond
conditions)
lb. rqd
69 58 52 47 41 38 36 35 33
('poor'bond
conditions)
(I) For 'poor' bond conditions the table values should be multiplied by 0.7.
(2) The table values apply to bar diameters S; 40 mm. For larger bars the values should be multiplied by
(140- )/IOO.
(3) The basic anchorage length lb. rqd has been calculated assuming that the stress in the bar is fykhs' i.e.
500/1.15. The values of lb. rqd may be multiplied by As. reqd lAs. prOVided'
1
(a)
<I>
L
Ib net
l
1 'I
(b)
90' S; CI. <150'
---
(e)
---
}
'bnet
L
~ Not less than 5<1>
.r---t (d)
<1>, not less than 0, 6<1>
1---
1
L
,
L
(e)
fig. 10.3. Methods of anchorage: (a) straight bar; (b) hook; (c) bend; (d) loop; (e) welded
transverse bar
--
10.3.1. Design anchorage length
The basic anchorage length can be modified to allow for such effects as the shape of the bars,
the size of the concrete cover, confinement offered by transverse reinforcement or transverse
pressure. In EN 1992-1-1, the design anchorage length is obtained by multiplying the basic
anchorage length by a number of factors:
lbd = 0:10:20:30:40:s lb. reqd > max{0.3I
b
reqd; 10<,1:>; 100 mm}
for anchorage in tension and
lbd = 0:10:20:30:40:Slb,reqd > max{0.6I
b
,reqd; 10<,1:>; 100 mm}
for anchorage in compression; the product 0:20:30:S should be ;::: 0.7. Values of the different
multipliers together with the conditions that should be met are given in the code and are not
reproduced here. However, the applicability of these in practice are discussed below.

..

The following applies to bars in tension:
0:
1
allows for the shape of the bars. For straight bars, 0:
1
= 1.0. For curved bars, 0:
1
may be
taken as 0.7, provided that the lesser of the side cover to the bars or half the clear spacing
between bars is ;::: 3<,1:>. It will be difficult to meet this condition in all but a few practical
cases.
0:
2
allows for the effect of the size of concrete cover. For straight bars in tension, some
reduction will be possible when the parameter Cd is between <,I:> and 3<,1:>, where Cd is the
least of the side or bottom cover or half the clear spacing between the bars. In the case of
curved bars this benefit does not accrue until Cd is larger than 3<,1:>. Again, advantage can
be taken of this reduction only in a limited number of practical cases.
0:
3
allows for the effect of confinement offered by transverse reinforcement, which is not
welded to main bars. The transverse reinforcement should be placed between the
concrete surface and the bar that is being anchored. The reduction will be enhanced if
the transverse reinforcement is in the form of links. Even so, the reduction that can be
achieved in beams will only be 0.925 when LAst = 1.OA
S
'
0:
4
= 0.7 can be used in all cases where the transverse reinforcement is welded to the
main bars, provided the diameter of the transverse bar is at least 0.6<,1:> and it is located at
least 5<,1:> inside lb, reqd from the free end of the bar.
O:s accounts for the effect of any pressure p (in megapascals) transverse to the potential
plane of splitting, and is taken as (1 - 0.04p).
In compression anchorage, 0:
1
= 0:
2
= 0:
3
= 0:
4
= 1.0; O:s does not apply to anchorage in
compression.
In summary, the conditions that need to be satisfied to take advantage of the reduction
factors are such that they will only apply in a limited number of practical cases in building
structures.
10.3.2. Transverse reinforcement at anchorage
In clause 8.4.], EN 1992-1-1 states that transverse reinforcement should be provided if
necessary. There is no other direct guidance. In the section dealing with design anchorage
length there are references to transverse reinforcement. This could be interpreted to mean
that there is no specific requirement for transverse reinforcement. This is unlike the
requirements in the ENV, which still appear to be prudent, and are repeated here:
In beams, transverse reinforcement should be provided for anchorage in tension,' if
there is no transverse compression due to support reaction (e.g. indirect supports); and
for all compression anchorage.
The minimum total area of the transverse reinforcement (the area of the legs parallel to
the plane of the longitudinal reinforcement) is 25% of the area of one anchored bar.
(Fig. lOA).
Clause 8.4.1
173
174
The transverse reinforcement should be distributed evenly along the anchorage length,
and at least one bar should be placed in the region of the hook, bend or loop of any
curved bars.
e For bars in compression, transverse reinforcement should surround the bars, being
concentrated at the end of the anchorage and extend beyond it to a distance of at least
four times the diameter of the anchored bar. This is a precaution against bursting forces
that are likely to arise due to the 'pin effect' at the end of the bar.
T--,
I
I
I As
II
I
I I
(b)
Fig. 10.4. Lateral reinforcement: (a) beam; (b) slab (Ast is the area of one bar of the transverse
reinforcement; As is the area of one anchored bar)
(a)
Fig. 10.5. Anchorage of links
8$ not less than 70 mm
(b)
Not less than 2$
Not less than 20 mm
Not less than 50 mm
(e) (d)
,.. 'i'
~ ~ ~ * = ~ ~
Not less than
Fs
- = = = ~ F = = = = = =
*If > 4q" the lap length shall be increased by the amount by which the clear gap exceeds 4q,
Fig. 10.6. Spaces between lapped bars
10.4. Anchorage of links
These are shown in Fig. 10.5.
10.5. Laps for bars
10.5.1. General
As the length of reinforcement bars is generally likely to be less than the length of the
structure, lapping (or splicing) of bars will be necessary in most structural elements. At laps,
forces are transmitted from one bar to another. This can be achieved through the concrete
surrounding the lapping bars or by welding of the bars or by mechanical couplers.
Lapped bars should comply with the arrangement shown in Fig. 10.6.
10.5.2. Design lap length
Design lap length is calculated by modifying the basic anchorage length (see Section 10.2.4
above) by applying multipliers:
1
o
= a
1
a
2
a
3
a
S
aib,rqd:?: max{0.3aib,rqd; 15; 200 mm}
where aI' a
2
, a
3
and as are as discussed for anchorage and a
6
allows for the effect of the
amount of reinforcement that is lapped at one section:
a
6
= (p/25)OS
with values in the range 1.0-1.5 (for Pl > 50%). The amount of reinforcement lapped at one
section includes any laps occurring within 0.65/
0
on either side of the section.
10.5.3. Transverse reinforcement
The requirements are shown in Fig. 10.7.
10.6. Additional requirements for large diameter bars
Clause 8.8 applies to bars with a diameter larger than 32 mm, but in the UK to bars> 40 mm.
Such bars should be anchored either as straight bars with links provided as confining
reinforcement or using mechanical devices.
Large-diameter bars should not be lapped unless the minimum dimension of the member
is 1 m, or the stress in the bar does not exceed 80% of the design strength.
In the absence of transverse compression, transverse reinforcement should be provided in
anchorage zones. This is in addition to any shear reinforcement provided. The area of
transverse reinforcement is shown in Fig. 10.8.
Unless cracking is verified by calculations, surface reinforcement should be provided such
that its area is O.OlA
et
, ext in the direction parallel to the large diameter bars and 0.01 Act, ext in
the direction perpendicular to it. Act, ext is the area of concrete between the links and the outer
Clause B.B
175
176
surface of the member below the neutral axis. Surface reinforcement is not practical and
should be avoided.
I 0.7. Requirements for weld mesh
The mandrel sizes for bends in mesh are shown in Fig. 10.9.
The methods of lapping are shown in Fig. 10.10.
Intermeshed laps should comply with the requirements for bars discussed in Section 10.5
above. In the calculation of lo, the value of 0;3 should be taken as 1.0.
Layered laps of main bars in a mesh should be generally situated in zones, where the
calculated stress in the main reinforcement does not exceed 80% of its design strength.
The permissible percentage of the main reinforcement that can be lapped in one section is
as follows:
for intermeshed laps of fabrics using high bond bars: 100%
.. for layered laps withAs/S :s; 1200 mm2/m: 100%
.. for layered laps withAs/S > 1200 mm2/m: 60%.
Ast/2 Ast/2

,y2 ,y2
.5..- c:::::::::::. 8t$1 "w f=====::J4<P
b"d
fig. 10.7. Placing oftransverse reinforcement
fig. 10.0. Additional transverse reinforcement
or " ..... __ _
5
o Anchored bar
Continuous bar
or
d?: 3: 5
d < 3 or welding within the curved zone: 20
Note: the mandrel size for welding within the curved zone may be reduced to 5
where the welding is carried out in accordance with prEN ISO 17660, Annex B
fig. 10.9. Minimum diameter of the mandrel: (a) welds outside bends; (b) welds inside bends
-
Fig. 10.10. Layering of wires: (a) main reinforcement; (b) transverse reinforcement
Fig. 10.1 I. Minimum area of longitudinal reinforcement
The permissible percentage of secondary reinforcement that can be lapped in one section
is 100%.
I 0.8. Bundled bars
Unless stated otherwise, the rules given above for individual bars also apply to bundled bars.
The bundle is replaced by a notional bar of diameter n = -Yn
b
, in which nb is the number
of bars with diameter . nb is limited to 4 or below for vertical bars in compression and for
bars in a lapped joint; and 3 or below in all other cases.
In good bond conditions, two touching bars placed one over the other need not be
considered as a bundle.
Individual bars in bundles with n 32 mm should be anchored in tension such that they
are staggered by 1.3/
h
reqd in the longitudinal direction. For the calculation of lb, reqd' the
diameter of the individual bar may be used.
In compression anchorages of bundles with n 32 mm at least four links with bars of
12 mm diameter or more should be provided.
In lapped joints of bundles comprising two bars and n 32 mm or of lapped joints in
bundles comprising three bars, laps of the individual bars in the bundle should be staggered
in the longitudinal direction by at least 1.3/
0
,
10.9. Detailing requirements for particular member types
10.9.1. Beams
Longitudinal reinforcement
(1) Minimum areaAst,min:
A
st
, min = but not less than O.0013b
t
d, where fyk is the characteristic yield
stress of reinforcement (Fig. 10.11).
177
178
At supports in monolithic construction where simple supports are assumed in the
design, reinforcement As!, sup required to cope with partial fixity is 0.15 xA
st
span
(Figs 10.12 and 10.13).
(2) Maximum area A
st
, max or A
sc
, max:
A
st
, max or A
sc
, max = O. 04Ac where Ac is the cross-section area of concrete.
(3) Distribution of support reinforcement at internal supports of continuous flanged beams.
Total support reinforcement (Ast) should be distributed over the width of the effective
flange. Part of it may be concentrated over the width of the web (Fig. 10.14).
(4) Longitudinal compression reinforcement (diameter ) should be contained by link
reinforcement, the maximum spacing of which should not exceed 15.
Shear reinforcement
(1) General:
Shear reinforcement should form an angle of 90-45 with the mid-plane of the
beam.
Shear reinforcement may consist of a combination of:
links enclosing the longitudinal tensile reinforcement and the compression
zone
bent-up bars
shear assemblies of cages, ladders, etc. which do not enclose the longitudinal
reinforcement, but which are properly anchored in the compression and
tension zones (Fig. 10.15).
For combinations of links and shear assemblies, all shear reinforcement should be
effectively anchored. Lap joints on the leg near the surface of the web are only
permitted for high-bond bars. At least 50% of the necessary shear reinforcement
should be in the form of links.
Fig. 10.12. Longitudinal reinforcement at supports in monolithic construction
D
e As.:
.. Ast
Fig. 10.13. Maximum area of longitudinal reinforcement
~ . "s _______
u
' ~ u UQ'
u u
----;
beff1 bw
beff2
II) t11l
Fig. 10.14. Placing of tension reinforcement in flanged cross-section
-
Stirrup cage as
shear assembly
-
1-+--- Stirrups
(a)
"-
'.-
(b)
,
-
~
Ladd ers as
assembly shear
Fig. 10.15. Combination of links and shear assemblies: (a) stirrup cage as a shear assembly;
(b) ladders as a shear assembly
(2) Minimum areaA
sw
:
The minimum area should be calculated from
Pw.min =Asw/sbw sin 0: = 0.08fckO.
5
/fYk
where P
w
, min is the minimum shear reinforcement ratio, Asw is the area of shear
reinforcement with a longitudinal spacing sand 0: is the angle between the shear
reinforcement and the longitudinal steel.
(3) Maximum spacing of shear reinforcement Smax:
Maximum longitudinal spacing of links = 0.75d(1 + cot 0:).
Maximum longitudinal spacing of bent-up bars = 0.6d(1 + cot 0:).
(4) The transverse spacing of a series of shear links should not exceed 0.75d nor 600 mm.
Curtailment of longitudinal reinforcement
Any curtailed reinforcement should be provided with an anchorage length lb, net but not less
than d from the point where it is no longer needed. This should be determined taking into
account the tension caused by the bending moment and that implied in the truss analogy
used for shear design. This can be done by shifting the point of the theoretical cut-off based
on the bending moment by a
l
(see below for definition) in the direction of decreasing
moment. This procedure is also referred to as the 'shift rule'.
a
j
= z(cot e - cot 0:)/2
where e is the angle of the concrete struts to the longitudinal axis and 0: is the angle of
the shear reinforcement to the longitudinal axis. Normally, z can be taken as 0.9d. For
reinforcement in the flange, placed outside the web, a
j
should be increased by the distance of
the bar from the web.
Anchorage at supports
(1) End support:
When there is little or no fixity at an end support, at least one-quarter of the span
reinforcement should be carried through to the support. The code recommends that
the bottom reinforcement should be anchored to resist force of V,da/d + N
sd
' where
Vsd is the shear force at the end, a
l
is as defined above for the shift rule, and Nsd is the
axial force, if any, in the member.
(2) Intermediate supports (general requirements):
At intermediate supports, ;:: 25% of the midspan bottom reinforcement should be
carried to the support.
The minimum anchorage of bottom reinforcement beyond the face of the support is
1O for straight bars; or the diameter of the mandrel for bars of 16 mm diameter or
more and with hooks or bends; or twice the diameter of the mandrel in other cases.
However, this does not mean that the support must be greater than 204> wide, as the
bars from each side of the support can be lapped. It is, however, recommended that
continuous reinforcement is provided to resist accidental forces (Fig. 10.16).
179
J
180
Skin reinforcement
Skin reinforcement to control cracking should normally be provided in beams over 1.0 m in
depth where the reinforcement is concentrated in a small portion of the depth. This
reinforcement should be evenly distributed between the level of the tension steel and the
neutral axis and be located within the links.
Surface reinforcement
Surface reinforcement may be required to resist spalling of the cover, e.g. arising from fire or
where bundled bars or bars greater than 32 mm are used. This reinforcement should consist
of small-diameter, high-bond bars or wire mesh placed in the tension zone outside the links.
The area of surface reinforcement parallel to the beam tension reinforcement should not
be less than O.OlAct,ex!' where Act, ext is the area of concrete in tension external to the links.
The longitudinal bars of the surface reinforcement may be taken into account as longitudinal
bending reinforcement, and the transverse bars as shear reinforcement, provided that
they meet the arrangement and anchorage requirements of these types of reinforcement
(Fig. 10.17).
~ t--i f
10d
t::=====
L. Ib, net I,
t 1
Fig. 10.16. Continuous reinforcement
Surface
reinforcement
1
10.11. Arrangement and anchorage requirements
Ib, net L
1
I
I '
I
!
Fig. 10.18. Span reinforcement
Fig. 10.19. End supports with partial fixity
10.9.2. Slabs
(1) Minimum areaAst,min:
Ast, min = 0.26ictmbtd/iYk
Aot
but not less than O.0013b
t
d, where iyk is the characteristic yield stress of reinforcement.
(2) Maximum area A
st
, max:
A
st
, max;::: O.OMc
where Ac is the cross-section area of concrete.
(3) Maximum spacing Smax:
Generally, Smax = 3h :::; 400 mm for main reinforcement and 3.5h:::; 450 mm for
secondary reinforcement.
Local to concentrated loads, Smax = 2h :::; 250 mm for main reinforcement and
3h :::; 400 mm for secondary reinforcement.
(4) Reinforcement near supports:
In simply supported slabs a minimum of 50% of the reinforcement in the span
should be anchored at supports (Fig. 10.18).
Where partial fixity is likely to exist despite the assumption of simple support in
design, 25% of the reinforcement required to resist the maximum span moment
should be provided at the top of end supports.
At the end supports, the reinforcement should extend from the face of the support,
at least 0.2 times the adjacent span (Fig. 10.19).
At intermediate supports, the reinforcement should be continuous across the
support.
Transverse reinforcement
The minimum area As is 20% of the longitudinal reinforcement.
Corner and edge reinforcement
Suitable reinforcement is required where slab corners are restrained against lifting. Normally,
U bars extending 0.2l into the span should be provided at aU edges (Fig. 10.20).
181
182
Shear reinforcement
(1) Minimum slab depth. h
min
= 200 mm where shear reinforcement is to be provided.
(2) General:
The requirements given in Section to.9.1 for beams apply generally to slabs, with the
following modifications:
Form of shear reinforcement: shear reinforcement may consist entirely of
bent-up bars or shear assemblies where VEd:O; 0.33VRdmax'
Maximum longitudinal spacing:
smax = 0.75d(1 + cot a)
for links and 1.0d for bent up bars.
Maximum transverse spacing of shear reinforcement is 1.5d.
10.9.3. Columns
(1) Minimum diameter: 12 mm.
(2) Minimum area As. min:
As, min = 0.1N
Ed
liyd or O.
0024
e
whichever is greater, where NEd is the design axial force, iyd is the yield strength of
reinforcement andAc is the area of concrete.
(3) Maximum area As, max:
As, max = O.
04A
c
outside laps and
As, max = O.OBAc
at laps.
(4) Minimum number of bars (Fig. 10.21).
(1) General:
All transverse reinforcement must be adequately anchored.
Every longitudinal bar (or group of bars) placed in a corner should be held by
transverse reinforcement.
No longitudinal bar in a compression zone should be further from a restrained bar
than 150mm.
(2) Minimum diameter:
The diameter of the transverse bar should not be less than 6 mm or 0.25 times the
diameter of the largest bar being restrained.
(3) Spacing:
Generally, the maximum spacing Smax should be the least of the following: 20 times the
diameter of the longitudinal bar; or the lesser dimension of the column; or 400 mm.
Longitudinal
U-bar reinforcement
10.10. Reinforcement at free edges
0
"0 0
. .: .. : : :
Fig. 10.21. Minimum number of bars
Polygonal:
one bar per
corner
Circular:
four bars
minimum
For a distance equal to the larger dimension of the column, above and below slabs or
beams the spacing noted above should be reduced by a factor of 0.6.
The above reduced spacing is also required at laps of longitudinal bars of diameter
greater than 14 mm. A minimum of three transverse bars should be evenly positioned
over the lap length.
10.9.4. Walls
Vertical reinforcement
(1) Minimum area Asv, min
Asv. min = O,OOMe
(2) Maximum areaA
sv
max:
The code permits this to be doubled if the designer can show that the integrity of
concrete is not affected and the full strength can be achieved at the ultimate limit state.
The distance between adjacent bars should not exceed three times the thickness of
the wall or 400 mm, whichever is less.
(4) When minimum reinforcement controls the design, 50% of the minimum reinforcement
should be placed on each face.
Horizontal reinforcement
(1) Placement:
Horizontal reinforcement should be placed between the vertical reinforcement and
the face of the wall.
(2) Minimum area Ash, min:
ASh, min = 0.25A
sv
but not less than O.OOlAe.
Smax = 400 mm
(4) Minimum diameter:
The code does not specify a value, but it will be prudent to use a minimum of p.25
times the diameter of the vertical reinforcement.
Where the area of vertical reinforcement exceeds O.OMe transverse reinforcement in the
form of links should be provided in accordance with the requirements for columns.
183
184
10.9.5. Corbels
General
Where a
e
S zoo a simple strut-and-tie model may be used (Fig. 1O.22a).
Anchorage of the primary tie reinforcement
Unless a length [bnet is available, the primary horizontal tie As should be anchored on both
sides beyond the bearing area using U bars or a welded cross bar.
Provision of finks
When a
c
< O.Sh
e
, closed horizontal or inclined links should be provided (Fig. 10.22b). The
area of the link should be at least 0.25 times the area of the primary tie reinforcement.
When a
e
;::: 0.5he and the load applied on the corbel exceeds V
Rdc
' vertical closed links
should be provided in addition to the horizontal links (Fig. 1O.22c). The area of the vertical
link should be at least 0.5F
wd
/jyd' where FWd is the force in the main tie reinforcement.
~ d , m a x
FEd ~
(a)
As,maln A
LAs,lnk ~ As, main
(b) As,lnk '2:: k1As,main (c)
[KJ Anchorage devices or loops []] Links
fig. 10.22. Corbels: (a) strut-and-tie model; (b) reinforcement for G
e
< O.She; (c) reinforcement for
Ge;:O: O.She
I] Slab edge
Note: y can be > c
y
Note: zean be > C
z
and yean be> c
y
(a) (b)
Fig. 10.23. Effective width, be' of a flat slab: (a) edge column; (b) corner column
10.9.6. Nibs
The code does not give any specific guidance for designing or detailing such members, which
normally project from the faces of beams or walls. Nibs less than 300 mm thick may be
designed as short cantilevers. The line of action of load should be taken as the outer edge of
the loaded area. The bending moment should be calculated on the line of the nearest vertical
reinforcement in the member from which the nib projects (i.e. the leg of a link in a beam or
vertical reinforcement in a wall). Nibs should be sized so that shear reinforcement is not
required. Reduction of flexural shear permitted in clause 6.2.2(6) could be taken into Clause 6.2.2(6)
account in assessing the reinforcement. The tension reinforcement should be adequately
anchored by forming loops in the vertical or horizontal plane. Where nibs hang from the
bottom of other members, sufficient tension reinforcement should be provided to transfer
the loads from the nib to the member.
10.9.7. Reinforcement in flat slabs
Concentration of reinforcement over the columns will generally be required to meet the
serviceability requirements. In the absence of rigorous calculations, top reinforcement with
an area of 0.5At' should be placed in a width equal to the sum of 0.125 times the panel width
on either side of the column. At represents the area of reinforcement required to resist the
full negative moments in a panel.
At least two bars forming the bottom reinforcement in the slab should pass through the
internal columns in each orthogonal direction.
Reinforcement required to transfer bending moments from slab to columns, at right
angles to an edge, should be placed within an effective width as shown in Fig. 10.23.
EN 1992-1-1 recognizes the use of both proprietary shear reinforcement and conventional
link reinforcement. In the case of the former, the design and detailing should comply with
the relevant European technical approval.
Where punching shear reinforcement is required, it should be provided between the
loaded area and 1.5d within the control perimeter at which shear reinforcement is no longer
required. Such reinforcement should be provided in at least two perimeters of spacing not
exceeding 0.75d. The spacing of the perimeters in which links are provided should not exceed
0.75d. The spacing of the legs oflinks around a perimeter should not exceed 1.5d (Fig. 10.24).
Where shear reinforcement is required, the area of one leg of link reinforcement should
comply with
Asw, min(1.5 sin a + cos a)/srSt ;::;: 0.08jck.5/jYk
where a is the angle between the shear reinforcement and the longitudinal steel, and Sf and
St are the spacings of the shear reinforcement in the radial and tangential directions,
respectively.
185
I
186
(a)
fAl Outer control perimeter requiring
~ shear reinforcement
~ First control perimeter not requiring
shear reinforcement
O:O.25d
f L r - B ~
",2d
(b)
Fig. 10.24. Punching shear reinforcement: (a) spacing of links; (b) spacing of bent-up bars
t
Bent-up bars passing through the loaded area and within O.25d on either side of it may be
used as punching shear reinforcement.
CHAPTER II
Prestressed concrete
I I. I. General
In EN 1992-1-1, prestressed concrete is treated as part of a much wider material group
known as reinforced concrete, which covers normal reinforced concrete through partially
prestressed concrete to fully prestressed concrete. The principles and methods given in
EN 1992-1-1 are, in general, applicable to the full range of reinforced concrete construction.
This chapter covers the aspects of design and detailing that are particular to prestressed
concrete; those that apply equally to reinforced concrete are covered in other chapters of
this guide.
EN 1992-1-1 covers the design of prestressed concrete members in which the prestress is
applied by tendons; the tendons may be embedded in the concrete and may be pretensioned
and bonded or post-tensioned and bonded or unbonded. The tendons may also be external
to the structure with points of contact occurring at deviators and anchorages. It should be
noted that second-order moments can be induced in prestressed members with external
tendons if their points of contact with the concrete are sufficiently far apart.
I 1.2. Summary of main clauses
The main clauses relating to the design of prestressed concrete in EN 1992-1-1 are summarized
in Table 11.1.
I 1.3. Durability
Steel reinforcement and prestressing tendons are protected against corrosion by complying
with EN 1992-1-1 requirements on stress levels, crack widths and concrete cover. EN 1992-1-1
specifies a 'minimum' cover, based on the greater of the requirements for bond or durability,
to which a construction tolerance of 10 mm must be added.
For post-tensioned members, the minimum cover for bond requirements should not be
less than the duct diameter (for circular ducts), or the greater of the smaller dimension or
half the larger dimension (for rectangular ducts). In no case does it need to be greater than
80 mm. Although EN 1992-1-1 specifies requirements for the minimum cover perpendicular
to the plane of curvature for curved reinforcing bars, it does not specify similar requirements
for curved prestressing tendons. This is an unfortunate omission, and the designer is
recommended to refer to other codes, e.g. BS 8110, where specific requirements are stated
for the prevention of bursting of the cover perpendicular to the plane of curvature. The
requirements of BS 8110 are given in Table 11.2.
For pretensioned members, the minimum cover for bond requirements should be not less
than 1.5 the tendon diameter, or 2.5 times the diameter when indented wires are used. The
Clause 2.3.1.4
Clause
4.4.1.3( I)P
Clause
4.4.1.2(3)
Clause
4.4.1.2(3)
Clause
4.4.1.2(5)
188
minimum cover is not related to the aggregate size, but this will probably not be a problem in
practice when the construction tolerance is taken into account.
The minimum cover to prestressing tendons and concrete strengths for durability
requirements are given in Tables 4.3N, 4.5N and E.1N. The UK National Annex replaces
these tables with two tables specifying minimum covers and concrete strengths, maximum
water/cement ratios, minimum cement content, cement/combination types and equivalent
designated concrete for different exposure classes and for intended working lives of 50 and
Table I 1.1. Clauses in EN 1992-1-1 for the design of prestressed concrete members
Requirement
Durability
Environmental conditions
Concrete cover
Design data
Concrete
Prestressing steel
Mechanical properties
Jacking force
Partial factors for prestress
Bending and longitudinal force
Values of prestress
Shear
General
Reduction in web width
Shear capacity of concrete
Maximum design shear stress
Shear reinforcement
Enhancement near supports
Variable depth members
Inclined tendons
Torsion
Values of prestress
Stress levels
Cracking
Deformation
Prestress losses
Relaxation
Elastic deformation
Shrinkage
Creep
Draw-in
Duct friction
Anchorage zones
Pretensioned members
Post-tensioned members
Detailing
Spacing of tendons/ducts
Anchorages and couplers
Deviators
Minimum area of tendons
Tendon profile
Minimun shear reinforcement
Spacing of shear reinforcement
Clause/annex
4.2
4.4.1
3.1
3.3
3.3.2-3.3.6
5.10.2,5.10.3
2.4.2.2,5.10.9
6.1
2.4.2.2,5.10.3,5.10.8
6.2.1
6.2.3(6)
6.2.2
6.2.1 (6),6.2.2(6),6.2.3(3),6.2.3(4)
6.2.3
6.2.2(6),6.2.3(8)
6.2.1 (2)
6.2.1 (3)
6.3
5./0.3,5.10.9
7./,7.2
7.3
7.4
3.3.2,5.10.6, /0.5.2, Annex 0
5.10.4,5.10.5.1
3.1.4(6),5.10.6, Annex B
3.1.4(2)-3.1.4(5), 5.10.6, Annex B
5.10.4,5.10.5.3
5.10.4,5.10.5.2
6.5, B.10.2
6.5, 6.7, B.I 0.3
B.IO.I
B.I 0.2-8.1 0.4
B.10.5
5.10.1 (6), 7.3.2, 9.2.1.1, 9.3.1.1
9.2./.3
9.2.2(5), 9.3.2
9.2.2(6)-9.2.2(8), 9.3.2
Table I 1.2. Minimum cover to curved ducts perpendicular to the plane of curvature
Minimum cover (mm)
Duct internal diameter (mm):
Radius of
19 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
curvature
Tendon force (kN):
of duct (m) 296 387 960 1337 1920 2640 3360 4320 5183 6019 7200 8640 9424 10338 11248 13200
2 50 55 155 220 320 445 Radii not
4 50 70 100 145 205 265 350 420 normally
6 50 65 90 125 165 220 265 310 375 460 used
8 55 75 95 115 150 185 220 270 330 360 395
10 50 65 85 100 120 140 165 205 250 275 300 330
12 60 75 90 110 125 145 165 200 215 240 260 315
14 55 70 85 100 115 130 150 170 185 200 215 260
16 55 65 80 95 110 125 140 160 175 190 205 225
18 50 65 75 90 105 115 135 150 165 180 190 215
20 60 70 85 100 110 125 145 155 170 180 205
22 55 70 80 95 105 120 140 150 160 175 195
24 55 65 80 90 100 115 130 145 155 165 185
26 50 65 75 85 100 110 125 135 150 160 180
28 60 75 85 95 105 120 130 145 155 170
30 60 70 80 90 105 120 130 140 150 165
32 55 70 80 90 100 115 125 135 145 160
34 55 65 75 85 100 110 120 130 140 155
36 55 65 75 85 95 100 115 125 140 150
I ~
38 50 60 70 80 90 105 115 125 135 150
40 50 50 50 50 50 50 50 60 70 80 90 100 110 120 130 145
-I
m
(I) The tendon force shown is the maximum normally available for the given size of duct (taken as 80% of the characteristic strength of the tendon).
I,.,
(2) Where tendon profilers or spacers are provided in the ducts and these are of a type which will concentrate the radial force, the values given in the table will need to be
increased.
I ~
(3) The cover for a given combination of duct internal diameter and radius of curvature shown in the table may be reduced in proportion to the square root of the tendon
force when this is less than the value tabulated, subject to a minimum value of 50 mm.
,.,
m
Vl
Vl
m
0
n
0
z
n
,.,
m
00
liTi -.0
Clause 3.1.3
Clause 3. 1.2
Clause 3.3.2
Clause 3.3.6
Clause
5.10.2.1 (I )
190
100 years (see Chapter 9). The tables are limited to exposure classes XC, XD and XS for a
working life of 50 years and exposure class XC for a working life of 100 years because there is
insufficient data to make recommendations for members exposed to other conditions. For
selecting the quality of concrete subject to freeze-thaw and concrete in aggressive ground
conditions, the designer is referred to Annex A of BS 8500-1.
The UK tables allow a greater range of concrete strengths and covers and take advantage
of the improved durability of blended cements to reduce minimum concrete strengths and
increase water/cement ratios for the same cover. As far as the minimum cover to ensure
durability is concerned, the UK Annex does not distinguish between normal reinforcement
and prestressing tendons. EN 1992-1-1, on the other hand, suggests that greater covers
should be provided to prestressing steel. Presumably this is because it is generally believed
that the risks posed to prestressing tendons from the breakdown of the protection provided
by the concrete are greater. A simplified composite version of the tables in EN 1992-1-1 is
presented as Table 11.3.
I 1.4. Design data
I 1.4.1. Concrete
EN 1992-1-1 does not specify minimum concrete strengths for prestressed concrete members.
The design data to be assumed for concrete are the same for prestressed members as for
normal reinforced members, and are given in Table 11.4.
I 1.4.2. Prestressing steel
Prestressing steel should comply with EN 10138.
EN 1992-1-1 adopts a bilinear stress-strain diagram for prestressing steel with either a
horizontal or sloping top leg, as shown in Fig. 11.1.
The maximum initial stress applied to the tendon, i.e. the jacking stress, is expressed as a
percentage of the ultimate strength fpk or the 0.1% proof stress fpOlk of the tendon. In
Table I 1.3. Minimum cover for durability requirements of prestressing tendons according to
EN 1992-1-1
Minimum cover (mm)
Exposure Working Minimum concrete strength:
class life (years) CI2!15 C20/25 C25/30 C30/37 C35/45 C40/50 C45/55
XO 50 10 10 10 10 10 10 10
100 20 20 20 15 15 15 15
XCI 50 25 25 20 20 20 20
100 35 35 30 30 30 30
XC2 50 35 35 30 30 30
100 45 45 40 40 40
XC3 50 35 30 30 30
100 45 40 40 40
XC4 50 40 40 35 35
100 50 50 45 45
XDI,XSI 50 45 45 40 40
100 55 55 50 50
XD2 50 50 50 45 45
100 60 60 55 55
XS2 50 50 50 45
100 60 60 55
XD3, XS3 50 55 55 50
100 65 65 60
For values to be used in the UK, see Chapter 9 the National Annex.
CHAPTER I I. PRESTRESSED CONCRETE
Table I 1.4. Concrete design data
Concrete
fek fck, cube fetm fctkO.05 fctkO.95 Eem
class (N/mm
2
) (N/mm2) (N/mm2) (N/mm2) (N/mm2) (kN/mm2)
CI2!15 12 15 1.6 1.1 2.0 27
C16/20 16 20 1.9 1.3 2.5 29
C20/25 20 25 2.2 1.5 2.9 30
C25/30 25 30 2.6 1.8 3.3 31
C30/37 30 37 2.9 2.0 3.8 33
C35/45 35 45 3.2 2.2 4.2 34
C40/50 40 50 3.5 2.5 4.6 35
C45/55 45 55 3.8 2.7 4.9 36
C50/60 50 60 4.1 2.9 5.3 37
fpk - - - - - ---- --- - - --- - --- - - - -- - - - . : = ~ . ,
Idealized ~ _------ I
--------- J
fpO,1k - - - - - T ___ - - - I fpk/rs
I :
Design
,,/Ep
cud = 0,9c
u
k Cuk
Fig. 11.1. Idealized and design stress-strain diagram for prestressing steel
Table I 1.5. Jacking forces and initial prestress for different tendons
.
Jacking Initial
Strand Diameter Area
fpk f pO. 1k
force prestress
type (mm) (mm
2
) (N/mm2) (N/mm2) (kN) (kN)
YI770S7 15.2 140.0 1770 1520 192 181
16.0 150.0 1770 1520 205 194
YI860S7 12.5 93.0 1860 1600 134 126
13.0 100.0 1860 1600 144 136
15.2 140.0 1860 1600 202 190
16.0 150.0 1860 1600 216 204
YI860S7G 12.7 112.0 1860 1610 162 153
YI820S7G 15.2 165.0 1820 1560 232 219
YI700S7G 18.0 223.0 1700 1470 294 278
'Based on data from BS EN 10138-3
practice'/pO.lk "" 0.85fpk' and the requirement expressed in terms of the 0.1 % proof stress will
be the more critical. Similar requirements apply to the stress in the tendon after transfer.
Values of jacking loads for various strand types are given in Table 11.5, where the requirement
based on the 0.1 % proof stress governs.
EN 1992-1-1 allows overstressing of tendons during tensioning to a stress of 0.95fpo.lk'
provided that the force in the jack can be measured to an accuracy of 5% of the final value
of the prestressing force.
Clause 5.10.3(2)
Clause
5.10.2.1 (2)
191
Clause 2.4.2.2
Clause 5.10.9
Clause
2.4.2.2(1)
Clause
2.4.2.2(2)
192
I 1.4.3. Partial factors
The partial factors to be applied to the prestressing force in the design of prestressed
concrete members are summarized in Table 11.6.
I 1.5. Design of sections for flexure and axial load
I 1.5.1. Ultimate limit state
EN 1992-1-1 specifies two partial safety factors for prestress when considering overall
analysis and the design of sections: 'fp, fav = 1.0 and 'f
p
. unfav = 1.3. However, the UK National
Annex specifies modified values of 'fp, fay = 0.9 and 'fp, unfav = 1.1. In most cases, prestressing is
a favourable effect, and 'f
p
, fay should be used.
In the design of sections, the initial strain in the tendons, Epo, multiplied by the partial
factor, 'fp' is taken into account together with the increase in strain at the ultimate limit state,
fl.E
p
The increase in strain at the ultimate limit state is determined as follows:
For bonded tendons, the increase in strain in the tendon is equal to the increase in strain
in the concrete above zero at the level of the tendon (Fig. 11.2).
Table I 1.6. Partial factors to be used for prestress
Partial factors
Serviceability limit state Ultimate limit state
Effect Maximum Minimum Maximum Minimum
Flexure
Pretensioned or unbonded
Post-tensioned and bonded
Shear
Anchorage zones
EN 1992-1-1
UK
EN 1992-1-1
UK
EN 1992-1-1
UK
1.05
1.0
1.1
1.0
0.95 1.3 1.0
1.0 1.1 0.9
0.9 1.3 1.0
1.0 1.1 0.9
1.3 1.0
1.1 0.9
Anchorage of pretensioned tendons
Anchorage zones
See Section I I .8. I
1.2
C
eu
------
CpO the initial strain in the tendon at a section (Le. allowing only for losses due 10
friction and draw-in) + L\c
po
L\cpO = 0, for prelensioned members
the strain in the concrete due to streSSing the tendon (or to stressing the first
tendon when a number of tendons are stressed successively), for post-tensioned
members
Fp the force in the prestressing tendons at the ultimate limit state
EpAp(L\cp + Y
P
, ,,,cpo)
Fig. I 1.2. Strain in bonded prestressing tendons at the ultimate limit state
Table 11.7. Criteria for controlling cracking
Exposure class
XO,XCI
XC2, XC3, XC4
XD I , XD2, XD3, XS I, XS2, XS3
Unbonded tendons: Bonded tendons
quasi-permanent
loads
0.3'
0.3
0.3
Frequent loads
0.2
0.2
Decompression
Quasi-permanent loads
Decompression
'For XO, XC I exposure classes, crack width has no influence on durability and this limit state is set to
guarantee acceptable appearance. In the absence of appearance conditions this limit may be relaxed.
For unbonded and external tendons, the increase in strain has to be calculated based
on the deformation of the concrete member. Alternatively, an increase in stress of
100 N/mm2 may be assumed. When the deformation is calculated using mean values of
the member properties, the increase in stress should be multiplied by a partial factor,
1't.p. 1't.p should be taken as 0.8 or 1.2, depending on whether the increase in stress is a
beneficial or adverse effect. Alternatively, a linear analysis using uncracked section
properties can be used with 1't.p = 1.0.
11.5.2. Serviceability limit state
EN 1992-1-1 distinguishes between three serviceability load combinations, i.e. characteristic,
frequent and quasi-permanent. All are relevant to the design of prestressed concrete
members, but their relative magnitude depends on the loads applied and the type of
structure being designed.
EN 1992-1-1 also adopts an upper or lower characteristic value (rsup and r
inf
) for the
prestressing force (whichever is more critical) at the serviceability limit state. The upper and
lower characteristic values are taken as 1.05 and 0.95 times the mean value of the prestress
respectively for pre tensioned or unbonded tendons, and as 1.1 and 0.9 for post-tensioned
bonded tendons. However, the UK National Annex adopts a single value of 1.0, which
simplifies the analysis. In the design equations presented later in this section, rsup and rinf are
explicitly identified so that the equations can be easily applied in other countries.
Limitation on stress
To avoid the occurrence of longitudinal cracks in regions of high compression, EN 1992-1-1
limits the compressive stress under the characteristic load combination to 0.6fck in areas
exposed to environments of exposure class XD, XF or XS unless other measures, such as an
increase in cover to reinforcement in the compressive zone or confinement by transverse
reinforcement, are taken. In order to ensure that creep deformation remains linear, the
concrete stress under the quasi-permanent load combination is limited to 0.45fck. Otherwise,
non-linear creep should be considered (see clause 3.1.4). Compressive stresses at transfer
should generally be limited to O.6fck. However, for pretensioned members, this limit can be
increased to O.7fck if tests or experience demonstrate that longitudinal cracking is prevented.
To avoid stresses in the tendons under serviceability conditions that could lead to
unacceptable cracking and deformation, EN 1992-1-1 limits the mean stress in the prestressing
tendons to 0.75fpk' after allowance for losses. This is unlikely to be a critical criterion in
most cases.
In calculating stresses at the serviceability limit state, sections should be assumed to be
uncracked if the maximum tensile stress does not exceed the mean concrete tensile strength,fctm.'
Cracking
EN 1992-1-1 specifies two criteria for the control of cracking: decompression and limiting
crack widths to a specified value (Table 11.7). The decompression limit requires all parts
of the tendon or duct to lie at least 25 mm within concrete in compression. Members
Clause 5.10.8(2)
Clause 5.10.8(3)
Clause 5.10. 9( I)
Clause 7.2(2)
Clause 7.2(3)
Clause 3.1.4(4)
Clause
5.10.2.2(5)
Clause 7.2(5)
Clause 7. f (2)
Clause 7.3.1 (5)
Clause 7.3.1 (6)
193
Clause 7.3.3
Clause 7.3.3(2)
Clause 7.3.4
194
prestressed with a combination of bonded and unbonded tendons should meet the requirements
for bonded tendons. Cracking checks for prestressed concrete are carried out under the
frequent or quasi-permanent load combinations using the upper or lower characteristic
prestressing force.
In order to control cracking, EN 1992-1-1 tabulates bar size and spacing criteria (see
Tables 7.2N and 7.3N) to ensure that the crack widths are limited to the appropriate values, as
long as a minimum area of reinforcement or prestressing tendons is provided. Alternatively,
formulae are provided to allow the design crack width to be calculated.
The approach adopted by EN 1992-1-1 for the design of partially prestressed concrete
may be more rigorous than the approach adopted by some other codes (e.g. BS 8110), but it
is more complex and difficult for the designer to apply. Tables 7.2N and 7.3N can be applied
to pretensioned members, where the crack width is controlled by the tendons, by using a steel
stress equal to the increase in stress in the tendons due to the applied load, t.O"p. They can
also be used for post-tensioned members where the crack width is controlled by ordinary
reinforcement by allowing for the effect of the prestress force when calculating the stress in
the normal reinforcement.
Alternatively, the designer must use the crack width formulae. The procedure is as
follows:
(1) Calculate the stress in the reinforcement, O"s, assuming a cracked section and the design
value of the prestress force (rin? m=>' For pretensioned members, o"s can be replaced by
the increase in tendon stress, .0.0" p'
(2) Determine C
sm
- Cern' allowing for tension stiffening using equation (7.9).
(3) Determine sr, max using equation (7.11) or (7.14).
(4) Calculate the design crack width W
k
using equation (7.8).
(5) Compare w
k
with the required value, and adjust the normal reinforcement or prestressing
force and/or its eccentricity.
(6) Repeat steps (1)-(5) until the required value OfWk is obtained.
Design equations
In order to assist the designer, the stress and crack width criteria discussed in the previous
sections can be expressed mathematically as a series of inequalities. The following
nomenclature is used (sagging moments are positive, and e is taken as positive when it is
below the neutral axis):
A
e
lek
fetm
MDL
Mp
M
Qp
Mk
P
mD
P
m
=
r
inf
, rsup
Zb
Zb(25)
Z(
area of concrete section
eccentricity of tendon
28 day cylinder strength
mean tensile strength of concrete
moment acting at transfer
moment under frequent loads
moment from quasi-permanent loads
moment under characteristic loading
prestress force at transfer (t = 0)
final prestress force (t = 00)
partial safety factors (= 1.0 in the UK)
section modulus of bottom fibre
section modulus 25 mm below the tendon
section modulus of top fibre
Lower chord, compression at transfer
(
Ie) MDL f-'
T,upPmo --- +--;:::: -Je(m
A Zt
for
Otherwise the calculation should be based on a cracked section.
Upper chord, tension at transfer
If the tensile stress at the upper chord exceeds fetm' check that the crack width does not
exceed the appropriate value from Table 11.7. In this calculation, rsup should be used in order
to maximize the tensile stress, i.e. the prestressing force should be taken as P = r s u ~ mO'
Lower chord, tension at the serviceability limit state
Check decompression:
1 linf(lIA +
< ------'--'--
P
m
= - (MF or M
QP
)/Zb(25)
for
Otherwise the designer should use a cracked section and adjust P and e until the neutral axis
is 25 mm below the tendon.
Check the crack width: the crack width should be determined using rinf in order to
maximize tension, i.e. P = rin? m=' The bending moment should be MF for members with
bonded tendons, or MQp for members with unbonded tendons.
Check the tendon stress: calculate the tendon stress using a cracked section. The bending
moment should be M
k
Both rsup and rinf should be considered, although the higher value is
more likely to be critical.
Upper chord, compression at the serviceability limit state
Check that
1 linf (l/A - e/Z
t
)
- > -'=-'-------'-'-
P
m
= - 0.6fck -Mk/Zt
1 > linf (lIA - e/Z
t
)
P
m
= - 0.45fck -MQp/Zt
for
for
Otherwise, the calculation must be carried out on a cracked section.
Deformation
The span/depth ratios given in clause 7.4.2 apply only to reinforced concrete. The designer
must use the guidance given in clause 7.4.3 when it is necessary to determine the deflections
of prestressed concrete members.
I I .6. Design of sections for shear and torsion
11.6.1. Shear
In the design of sections for shear, EN 1992-1-1 follows the same procedure as for reinforced
concrete, treating the prestressing force as an applied axial compression. The design shear
force at a section should be calculated taking into account the external loads and the effect of
any inclined tendons.
Where the prestressing force increases the shear capacity, it should be multiplied by the
partial safety factor Ip, fay = 1.0 (0.9 in the UK); in other cases it should be multiplied by
Ip, unfav = 1.3 (1.1 in the UK). Because the prestressing force can occur on both the load and
the resistance side of the equation, Ip, fay and Ip, unfav should be used consistently and the worst
case adopted.
In members where the design shear force is less than the shear capacity of the concrete
section alone, no design shear reinforcement is required, although the minimum shear
reinforcement in accordance with clause 9.2.2 should be provided unless the member is a slab
or a lintel with a span::; 2 m.
One of the factors governing the shear capacity of the concrete section alone is the area of
longitudinal reinforcement in the tensile zone. EN 1992-1-1 does not state whether this
Clause 7.4.2
Clause 7.4.3
Clause 6.2.1 (3)
Clause 2.4.2.2(1)
Clause 2.4.2.2(2)
Clause 6.2.1 (3)
Clause 9.2.2
Clause 6.2.2(1)
195
should be the total area of both the normal reinforcement and the prestressing tendons or
whether the areas should be adjusted to reflect the different design strengths. It is suggested
that the sum of the areas be used for bonded tendons, as this term is included to account for
the change in level of the neutral axis, which is governed more by reinforcement area than by
its strength.
Clause 6.2.2( I) The presence of prestress increases the shear capacity of the concrete section by 15% of
the stress due to prestressing.
Clause 6.2.2(2) For the special case of single span prestressed members without shear reinforcement, in
Clause 6.2.3(6)
zones where the maximum flexural tensile stress at the ultimate limit state is less than
fetd = fetkO,oshe' the shear capacity should be checked by limiting the maximum principal tensile
stress in the cross-section to fetd' For pretensioned members, the effect of the prestress force
is reduced in the transmission zone. The code draws the designer's attention to the fact that,
in a member of varying width, the maximum principal tensile stress may not occur at the
neutral axis. The section should be checked in several positions, and the lowest value for the
shear capacity should be taken.
EN 1992-1-1 requires the width of the web used in determining the shear capacity by
limiting the principal tensile stress to be reduced as follows:
..
..
Non-grouted ducts, grouted plastic ducts and unbonded tendons: 1.2 x the sum of the
duct diameters.
Grouted metal ducts with a diameter> 1/8 of the web width: 0.5 x the sum of the duct
diameters.
.. Grouted metal ducts with a diameter::;; 1/8 of the web width: O.
EN 1992-1-1 explains that the value of 1.2 is to avoid the splitting of the concrete
compressive struts due to transverse tension, and suggests that it can be reduced to 1.0 if
Clause 6.2.2( I) adequate transverse reinforcement is provided. It is not clear from the code that this
reduction should also be applied when calculating the shear capacity of the concrete alone by
the general method in clause 6.2.2(1), but it is recommended that this should be the case
where the ducts are located in the compression zone.
Clause 6.2.2(5) The designer's attention is specifically drawn to the fact that, in regions cracked in flexure,
the tensile force in the longitudinal reinforcement is increased by the presence of shear
above that required for bending. Provision for this effect is made by shifting the bending
moment diagram by a distance equal to the effective depth of the section so that the moment
at any section is always increased.
Where shear reinforcement is required, it must be designed in accordance with the
Clause 6.2.3(3) variable strut inclination method. The presence of prestress does not affect the area of shear
reinforcement required, but it does increase the maximum shear capacity of the section,
providing the mean prestress is less than 0.6fed' This will be the case in most prestressed
Clause 6.2.3(6) concrete members. In determining the maximum shear capacity of the section, the width of
the concrete section taken into account should be reduced to allow for the presence of ducts,
as described above.
Clause 6.2.3(7) Again, the designer's attention is drawn to the fact that the tensile force in the longitudinal
reinforcement is increased by the presence of shear above that required for bending.
Clause 9.2.1.3(2) Provision for this effect is most easily made by shifting the bending moment diagram so that
the moment at a given section is always increased. This is discussed in Chapter 10. As the
angle of the struts with the horizontal is reduced, the area of shear reinforcement reduces,
but there will be a corresponding increase in the longitudinal reinforcement requirement.
Clause 6.2.1 (2) The shear capacity of a member with shear reinforcement should be modified by taking
into account the effects of the vertical components of the design force in inclined
compression and tension chords. The effect of the inclined chords will be either positive or
Clause 6.2. 1(6) negative depending on the direction of inclination and the sign of the corresponding bending
moment. The maximum shear capacity of a section reinforced in shear should be greater
than the vector sum of the design shear force and the vertical components of the design
forces in any inclined chords.
196
Serviceability
If shear reinforcement is detailed in accordance with the requirements discussed in Chapter
10, then it is not necessary to carry out any shear checks at the serviceability limit state.
11.6.2. Torsion
Torsional resistance must be calculated when the static equilibrium of the structure depends Clause 6.3.1 ( I )P
on its torsional resistance. EN 1992-1-1 states that the design should be carried out at both
the serviceability and ultimate limit states; but design rules are only given for the ultimate
limit state. The reinforcement is designed using the variable strut inclination method; the Clause 6.3.2(2)
inclination of the struts must be the same as that used in designing for any coexistent shear
forces. It can be assumed that serviceability requirements are met if the detailing rules
discussed in Chapter 10 are followed.
When torsion arises from consideration of compatibility only, EN 1992-1-1 requires the
member to be designed to avoid excessive cracking. In practice, this means that stirrups and Clause 6.3.1 (2)
longitudinal reinforcement should satisfy specified detailing rules.
11.7. Prestress losses
11.1.1. General
The designer's attention is drawn to the need to allow for losses of prestress when calculating
the design forces in tendons at the various stages considered in the design. The causes of
these losses are listed in EN 1992-1-1. Because ofthe uncertainty in estimating such losses, it
is suggested that values are taken from European technical approval documents where these
are available. In the absence of such data, EN 1992-1-1 suggests values for the various
parameters that can be used for design.
I 1.1.2. Friction in jack and anchorages
Loss of prestressing force due to friction in the jack and anchorages is not mentioned in
EN 1992-1-1. It is not generally a problem for the designer, because the design is normally
based on the tendon force on the concrete member side of the jack. However, it does need to
be allowed for in calibration of the actual jack used for stressing.
11.1.3. Dud friction
Clause 5.10.3
EN 1992-1-1 specifies the standard formula for calculating the force lost in overcoming duct Clause 5. (0.S.2
friction when the tendon is stressed. This separates the loss into two parts:
(1) that caused by the tendon and duct following the specified profile
(2) that caused by slight variations in the actual line of the duct, which may cause additional
points of contact between the tendon and the duct wall.
The losses are calculated in terms of the coefficient of friction fJ and an unintentional
angular displacement per metre length k. Values of fJ for a range of internal tendons which Clause 5.10.5.2(2)
fill about 50% of the duct and for a range of external unbonded tendons in different ducts are
given in EN 1992-1-1, which also suggests that k should be taken between 0.005 and Clause 5.10.5.2(3)
0.01 rad/m. The loss of prestress due to unintentional angular displacements may be ignored Clause 5. f 0.5.2(4)
for external tendons. Rather than using these values, it is generally better to refer to Clause 5.10.5.2(1)
European technical approval documents or test results produced by the manufacturers of
prestressing components; EN 1992-1-1 allows the designer to adopt this approach.
11.7.4. Elastic: deformation
The method specified for the calculation of losses due to the immediate elastic deformation
of the concrete when the prestressing force is applied follows normal procedures.
Clause 5.10.4
Clause 5.10.5. I
197
I 1.7.5. Anchorage draw-in 01" slip
Clause 5.10.5.3 EN 1992-1-1 mentions this cause of loss of prestressing force, but does not specify the values
to be used. Appropriate values can be obtained from European technical approval documents
and from the anchorage manufacturers, and should be checked on site, particularly if the
member is short, when the loss due to this cause can be critical.
Clause 5.10.6
Clause 3.3.2(5)
Clause 3.3.2(6)
198
The loss due to anchorage draw-in, !::"P is given by
M= E p ~ s D +mL
L
where
~ E p A p s D
L = --;;;-:::; length of tendon
Ep is the modulus of elasticity of tendons
Aps is the area of prestressing tendons
D is the anchorage draw-in
m is the rate of change of prestressing force due to friction, which may be taken as
PoJL(a+k)
JL is the coefficient of friction
a is the angular deviation per unit length
k is the unintentional angular deviation per unit length
I 1.7.6. Time-dependent losses
Time-dependent losses are those due to relaxation of the prestressing tendons and shrinkage
and creep of the concrete. EN 1992-1-1 considers the loss due to each effect, as if that effect
were acting alone, and then combines them taking 80% of the relaxation loss to account for
the effect of creep and shrinkage reducing the stress in the tendon. The formula given in
EN 1992-1-1 further reduces the combined loss to allow for the reduction in compressive
stress in the concrete as the prestress force reduces with time.
Relaxation of steel
The starting point is the relaxation loss at 1000 h from an initial stress of 70% of the actual
tensile strength of the tendon. This is best obtained from test certificates, or from the
appropriate material standard. EN 1992-1-1 suggests values for this parameter in the
absence of any other source of data. These values are numerically the same as those
commonly taken as being the 1000 h relaxation loss from an initial stress of 70% of the
characteristic strength of the tendon.
Table 11.8. Relaxation losses as a percentage of initial stress, at 20C
Initial Loss after (h)
stress
Tendon type CT
pi
5 100 500 1000 5000 10000 50000 100000 500000
Class I (wire 0.6f
p
k 0.3 0.5 1.2 2.0 2.4 3.9 4.8 7.8 9.6 15.5
or strand) O.7f
pk
1.0 1.4 2.8 4.0 4.7 6.7 7.9 11.3 13.2 19.0
0.8f
p
k 3.3 4.1 6.5 8.3 9.2 11.7 13.0 16.5 18.3 23.3
Class 2 (wire 0.6f
pk
0.0 0.1 0.2 0.3 0.4 0.6 0.8 1.3 1.5 2.5
or strand) 0.7f
p
k 0.2 0.3 0.6 0.8 1.0 1.4 1.6 2.3 2.7 3.9
0.8f
p
k 0.8 1.1 1.7 2.2 2.4 3.0 3.4 4.3 4.8 6.1
Class 3 (bars) 0.6f
p
k 0.1 0.2 0.5 0.8 1.0 1.6 1.9 3.1 3.8 6.2
0.7f
p
k 0.5 0.7 1.3 1.8 2.1 3.1 3.6 5.2 6.0 8.7
0.8f
p
k 1.7 2.2 3.4 4.3 4.8 6.1 6.7 8.6 9.5 12.1
Table 11.9. Factor to allow for non-linear creep
a Jfcm(to) Factor
::; 0.45 1.00
0.50 1.08
0.55 1.16
0.60 1.25
0.65 1.35
0.70 1.46
0.75 1.57
The 1000 h relaxation loss is multiplied by a factor which depends on the relaxation class
of the tendon, the ratio of the initial stress in the tendon to its characteristic tensile strength
and duration. A duration of 500000 h can be used to determine the final (long-term)
relaxation loss. Relaxation losses are given in Table 11.S.
The values for the relaxation losses in Table l1.S vary from those that the designer might
have expected from previous experience. In particular, the values for the loss at 1000 hare
less than those which would normally have been previously assumed. This can be explained
by the fact that the losses in the table are for initial stresses expressed as a function of the
characteristic tendon strength. Actual tendon strengths can be up to 14% higher than the
characteristic strength, so that an initial stress of 70% of the actual tendon strength could be
numerically equal to an initial stress of 70 x 1.14", SO% of the characteristic strength. For
this value of initial stress it can be seen that the 1000 h loss is approximately equal to the
suggested values, and the formulae in EN 1992-1-1 can be used with confidence.
Shrinkage
EN 1992-1-1 suggests typical values within the main text for the final shrinkage strain of
concrete (Tables 3.2 and 3.3), with more detailed information in Annex B.
Creep
EN 1992-1-1 suggests typical values within the main text for the final specific creep strain (or
creep coefficient), ( 00, to), for concrete (Fig. 3.1), with more detailed information in Annex
B. The designer should note that the creep coefficient is related to the tangent modulus of
the concrete, which may be taken as LOSE em' The creep deformation under a constant
compressive stress, (Jc, is given by
( ) ( )
(Jc
E oct = oct
cc '0 , 0 1.0SE
cm
When the compressive stress at the time of loading exceeds O.4SfCk(t
O
)' then the creep
deformations can develop non-linearly. The non-linear creep coefficient, i 00, to)' is
determined hy multiplying the linear creep coefficient, ( 00, to), by a factor obtained from
Table 11.9.
I I .8. Anchorage zones
I 1.8.1. Pretensioned members
Guidance is provided in EN 1992-1-1 for estimating the transmission lengths in pre tensioned
members. A distinction is drawn between the transmission length over which the
prestressing force is fully transmitted to the concrete lpl; the dispersion length over which
the concrete stresses gradually disperse to a distribution compatible with plane sections
remaining plane ldisp; and the anchorage length over which the tendon force at the ultimate
limit state is fully transmitted to the concrete lbpd' The designer will find that this distinction
Clause 3.3.2(7)
Clause 3.3.2(8)
Clause 3.1.4(6)
Clause 3.1.4(2)
Clause 3.1.4(3)
Clause 3.1.4(4)
Clause 8. 10.2
Clause 8.10.2. f
199
Clause 8.10.2.2 helps to clarify the distribution of forces within the anchorage zone. Typical transmission
Clause 8.10.2.3 and anchorage lengths for good bond conditions are given in Table 11.10.
Clause 8./ 0.2.3( I) The anchorage of pretensioned tendons at the ultimate limit state need only be checked if
Clause 8. f 0.3
Clause 2.4.2.2(3)
Clause 8.10.3(4)
Clause 8.10.3(3)
Clause 8.10.3(4)
Clause 8.10.3(5)
Clause 6.5.3(3)
200
the concrete tensile stress in the anchorage zone exceeds fetO,Osk' The tendon force to be used
should be calculated assuming a cracked cross-section and include the effect of shear.
Often pretensioned tendons are well distributed across the cross-section of the member so
that forces induced by transfer of the prestressing force from the tendon positions to a linear
distribution are low. In slabs these forces can often be resisted by the concrete alone without
the need for ordinary reinforcement, whilst in beams it is generally satisfactory to reduce the
stirrup spacing to approximately 75 mm throughout the transmission length. Where this is
not the case, reinforcement to maintain overall equilibrium and carry any spalling forces
should be provided as described in Section 11.8.2.
I 1.8.1. Post-tensioned members
The design of anchorage zones, or end blocks, in post-tensioned members is a very important
area of design, but one that is rarely well-treated in codes of practice. EN 1992-1-1 is no
exception, although it does draw attention to this important area and outlines the principles
of the design.
The designer's attention is drawn to the need to consider bearing stresses behind the
anchorage, overall equilibrium and transverse tensile forces, i.e. bursting and spalling forces.
The design is carried out at the ultimate limit state, and is based on '1p (= 1.2) times
the prestressing force. No check of crack widths is required when reinforcement stresses
are limited to 300 MPa. Bearing stresses under anchorage plates should be checked in
accordance with the relevant European technical approval. In practice, this means that the
designer should check with the supplier the minimum concrete strength and reinforcement
required adjacent to the anchorage. It is suggested that overall equilibrium be checked using
a strut-and-tie method, and guidance is given on this and on the angle of dispersion of the
prestressing force into the concrete section. Formulae are also given for determining the
bursting forces behind an anchorage. This is of benefit to the designer. However, no
guidance is given on how to design against spalling forces.
The designer should give particular attention to anchorage zones that have a cross-section
of different shape from that of the general cross-section of the member.
For more information on the design and detailing of anchorage zones in prestressed
concrete, the designer is referred to the excellent documents produced by CIRIA
14
and the
Institution of Structural Engineers.
ls
Table 11.10. Transmission lengths, I
pt
and anchorage lengths, i
bpd
' for good bond conditions:
number of diameters
Concrete grade
Tendon type 25/30 30/37 35/45 40/50 45/55 50/60
7 -wire strands
f
pt
60-90 50-80 45-70 40-65 40-60 35-55
f
bpd
130 115 100 95 85 80
Indented wire
f
pt
90-135 80-120 75-110 65-100 60-90 55-85
i
bpd
170 150 135 125 115 110
Ultimate stress in tendon = 1304 MPa.
Stress in tendon immediately after release = 1250 MPa.
Stress in tendon after all losses = 875 MPa.
Table I 1.1 I. Minimum dear spacing requirements for prestressing tendons and ducts
Pretensioned
Post-tensioned
Vertically
Horizontally
Vertically
Horizontally
? d
g
?d
g
+5mm
? d
g
? d
g
+5 mm
is the diameter of a tendon or duct, as appropriate; d
g
is the aggregate size.
11.9. Detailing
? 2
? 2? 20 mm
? ? 40 mm
? ? 50 mm
The detailing provisions given in Sections 8 and 9 of EN 1992-1-1 apply to both prestressed
and normally reinforced concrete. In this section, only provisions that specifically relate to
prestressed concrete are discussed. A more general commentary on the detailing requirements
of EN 1992-1-1 is given in Chapter 10.
11.9.1. Spacing oftendons and ducts
EN 1992-1-1 specifies the minimum allowable clear spacing between adjacent tendons or
ducts, which is generally the greatest of twice the tendon diameter, the duct diameter or the
aggregate size plus 5 mm. Absolute minimum clear spacings are also specified, but these are
unlikely to be critical when account is taken of the need to be able to compact the concrete
around the tendons. For post-tensioned members, EN 1992-1-1 bases its requirements on
the outside dimensions of the duct rather than on the internal dimensions, the latter being
more normal UK practice. The requirements are summarized in Table 11.11.
EN 1992-1-1 does not specify any additional requirements for the minimum clear spacing
in the plane of curvature between curved tendons. This is a serious omission, as very high
local forces can be developed, particularly when tendons in outer ducts are stressed before
the inner ducts are grouted. The designer is recommended to use the requirements given in
other codes, such as BS 8110, which are reproduced in Table 11.12.
11.9.2. Anchorages and couplers
EN 1992-1-1 requires that not more than 50% of the tendons be coupled at anyone cross-
section unless it can be shown that a higher percentage will not cause more risk to the safety
of the structure. This is contrary to the generally accepted UK practice of coupling all the
tendons at one cross-section. The reason for the EN 1992-1-1 requirement is that when a
coupled tendon is stressed, the force between the previously stressed concrete and the
anchorage decreases, and the local deformation of this concrete is reduced. Increased
compressive forces are induced adjacent to the anchorage, and balancing tensile forces
between adjacent anchorages. Uncoupled tendons across the joint reduce this effect, as well
as assisting in carrying the induced tensile forces. Alternatively, it is suggested that such
forces could be carried by properly detailed normal reinforcement.
11.9.3. Minimum area of tendons
EN 1992-1-1 specifies minimum areas of reinforcement to avoid brittle failure and wide
cracks occurring and also to resist forces arising from restrained actions. In prestressed
members, minimum reinforcement need not be provided to meet this requirement where
one of the following conditions is satisfied:
(1) the maximum tensile stress in the concrete at serviceability under the characteristic load
combination is less thanfelm or
(2) the prestress and ordinary reinforcement is sufficient to limit crack widths to the
required values
and brittle failure is avoided by:
Clause 8. 10. f .2
Clause 8. 10. 1.3
Clause 8. to. f. 1
Clause 8.10.4(5)
Clause 9.1
Clause 7.3.2(4)
Clause 5.10.1 (6)
201
IY
I ~
0
IY
Vl
Ci
Z
m
C2
G)
c
0
Table I 1.2. Minimum distance between centre-lines of ducts in the plane of curvature
1
m
-i
Minimum distance (mm)
0
m
Z
Duct internal diameter (mm):
-.0
19
-.0
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
IY
Radius of
,
-
Tendon force (kN):
,
curvature -
of duct (m) 296 387 960 1337 1920 2640 3360 4320 5183 6019 7200 8640 9424 10338 11248 13200

Z
2 110 140 350 485 700 960
0
Radii not m
4 55 70 175 245 350 480 610 785 940 normally
Z
6 38 60 120 165 235 320 410 525 630 730 870 1045 used -.0
8 90 125 175 240 305 395 470 545 785 855 940
-.0
655 IY
,
10 80 100 140 195 245 315 375 440 525 630 685 750 815
,
12 160 205 265 315 365 435 525 570 625 680 800
IY
14 140 175 225 270 315 375 450 490 535 585 785
16 160 195 235 275 330 395 430 470 510 600
18 180 210 245 290 350 380 420 455 535
20 200 220 265 315 345 375 410 480
22 240 285 310 340 370 435
24 265 285 315 340 400
26 260 280 300 320 370
28 345
30 340
32
34
36
38
40 38 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340
(I) The tendon force shown is the maximum normally available for the given size of duct (taken as 80% of the characteristic strength of the tendon).
(2) Values less than 2 x duct internal diameters are not included.
(3) Where tendon profilers or spacers are provided in the ducts and these are of a type which will concentrate the radial force. the values given in the table will need to be
increased. If necessary, reinforcement should be provided between ducts.
(4) The distance for a given combination of duct internal diameter and radius of curvature shown in the table may be reduced in proportion to the tendon force when this is
less than the value tabulated. subject to metting the minimum requirements of Table 11.1.
(1) the member containing pretensioned tendons
(2) access being provided to enable non-destructive testing and monitoring of thc condition
of the tendons to be carried out easily
(3) satisfactory evidence being provided on the reliability of the tendons or
(4) proportioning the section to ensure that, under the frequent combination of loads,
cracking occurs before the ultimate bending capacity is reached.
For members with unbonded or external tendons, EN 1992-1-1 requires that the ultimate Clause 9.2.1.1(4)
bending capacity exceeds the cracking moment by at least 15%. This seems an appropriate
value to adopt also for members with bonded tendons.
11.9.4. Tendon profiles
The design for shear in EN 1992-1-1 leads to a requirement for longitudinal reinforcement in
addition to that required for bending. This reinforcement is generally allowed for by a
horizontal displacement of the bending moment envelope (the 'shift' rulc). When the
flexural design of a prestressed concrete member is governed by the ultimate limit state,
which could be the case when the design criterion is a limiting crack width, the longitudinal
reinforcement will need to be increased and/or the tendon profile adjusted in order to resist
the tensile forces arising from the effect of inclined cracks in the webs.
Clause 9.2.1.3
203
I
I
CHAPTER 12
Structural fire design
12. I. Aims of design
Structural fire design is covered in EN 1991-1-2. Under fire conditions, a structure will be
required to fulfil load bearing and/or separating function. The code provides guidance for
satisfying the above requirements and this chapter of the manual discusses the salient
features. Building regulations normally specify the fire rating for the building. It can also be
calculated from first principles using EN 1991-1-2 (Actions on Structures Exposed to Fire).
The structure is required to comply with criteria R, I and E as appropriate:
Criterion R is assumed to be satisfied if the structure maintains adequate mechanical
resistance for the required duration of fire exposure.
Criterion I is the insulation criterion, which also applies to members fulfilling separating
function. It is deemed to be satisfied if the average rise in temperature on the non-
exposed surface does not exceed 140 K and the maximum rise does not exceed 180 K.
Criterion E applies to elements required to fulfil separating function, and will be
satisfied if the integrity of the structure is maintained for the required duration such that
the spread of flames, hot gases and excessive heat is avoided.
12.2. Design procedure
The designer has a wide choice of methods. The method for verifying the fire resistance in a
project should be chosen on the basis of acceptability to regulatory authorities. The design
can be based on thermal actions given by either nominal fire exposure or those calculated
using the physical parameters of the building. The former is the most common method, and
the latter should only be undertaken by designers with adequate appreciation of all aspects
of fire engineering.
In both procedures it is further permitted to base the design on (1) member analysis
(2) analysis of parts of the structure or (3) analysis of the whole structure.
In member analysis - which includes tabulated data - only the effects of thermal gradients
across the cross-section is considered; the effects of axial or in plane thermal expansions are
neglected. Also, the boundary conditions are assumed to remain unchanged throughout the
fire exposure.
In the analysis of a part of the structure, a section of the structure is isolated such that
stationary boundary conditions could be prescribed in the analysis. Within the part that
is isolated, the relevant failure mode in fire exposure, temperature-dependent material
properties and the effects of thermal expansions and deformations should be taken into
account in the analysis.
206
In the analysis of the whole structure, none of the approximations noted above are used,
and the global analysis is undertaken accounting for the temperature-dependent material
properties, member stiffness and the effects of thermal expansions and deformations.
The code gives only limited guidance for methods 2 and 3 above.
12.3. Actions and partial factors
12.3.1. Actions
Where calculations are undertaken, the load combination to be used in fire conditions is
given in EN 1990 (Basis of Structural Design). It is
Gk +Ad + 'ifJ1, lQk, 1 + 'ifJ2,iQk,i
whereA
d
is the design value of the accidental action (in this case actions caused by fire), Qk,l
is the leading variable action and Qk i represents other accompanying actions, if applicable.
For member analysis, where the ~ f f e c t s of thermal deformation are neglected and where
there is only one variable action, the action effect under fire conditions may be taken as 7]fi
times the value of action effect used in normal temperature design. The value of 7]fi depends
on the ratio of variable and permanent loads and the use of the building. EN 1992-1-2
provides a figure to arrive at the value. A value of 0.7 may be used as a safe approximation in
all cases.
12.3.2. Material factors
EN 1992-1-2 recommends a value of 1.0 for both "ie and "is' The UK National Annex has also
decided to adopt these values. The reasoning is as follows.
In probabilistic terms
"iM,fi = exp(0.7VR -1.645V
f
)
where
V
R
= (V
m
2
+ VG
2
+ V/)05
Vm is the coefficient of variation for model uncertainty, VG is the coefficient of variation for
geometry of element and Vr is the coefficient of variation for material property. Values for V
f
and VG were taken as the same as for normal design at room temperature, and V
m
, fi was taken
as 2V
m
, room temperature' This resulted in "ie = 0.84 and "is = 1.0. It was decided to use the value of
1.0 for all material properties.
12.4. Member analysis using tabular data
12.4.1. Scope
Section 5 of EN 1992-1-2 sets out minimum dimensions and axis distances (cover plus half
the diameter of the reinforcement bar) for different types of structural element for fire
exposure periods of 30,60,90, 120, 180 and 240 min. Some detailing requirements are also
given. The member types covered are columns, load-bearing and non-load-bearing walls,
beams (simply supported and continuous) and slabs (simply supported and continuous
boundary conditions, one- and two-way spanning slabs, solid and ribbed slabs, and flat slabs).
The table values apply to concrete made with siliceous aggregates and with a density in the
range 2000-2600 kg/m3. If calcareous or lightweight aggregate is used, the table values for
minimum dimensions for slabs and beams may be reduced by 10%.
When using tabulated data, no further checks are required for shear, torsion, anchorage
details and explosive spalling of concrete (spaHing related to the moisture content of the
concrete). However, the code requires surface reinforcement when the axis distance in a
member exceeds 70 mm to prevent falling off of the cover concrete.
CHAPTER 12. STRUCTURAL FIRE DESIGN
Rather than repeating the data tabulated in the code here, this guide discusses the basis of
the tables and some of the limitations.
12.4.2. Basis for the tabulated data
The tables have been developed on an empirical basis, general experience in Europe and
theoretical evaluation of tests. In particular, the data for method A for columns are based on
recent tests in Belgium and Canada. The data for the more common types of structural
elements are derived making approximate and conservative assumptions.
When the thickness of walls and slabs complies with the minimum dimensions, the
integrity and insulation criteria (see Section 12.4.1 above for definition) for the separating
function may be assumed to be satisfied.
For the load-bearing function, the following assumptions apply:
Ed f)R
d
fi S 1.0, where Ed fi is the design effect of actions under fire conditions andR
d
fi is
design resistance of the element under fire conditions. '
Unless otherwise noted, it is assumed that Tlfi = Ed fJE
d
= 0.7, where Ed is the design
effect of actions in normal temperature design. '
The critical temperature of reinforcement (Bcr) (i.e. the temperature of reinforcement at
which the failure of the member is expected to occur under fire conditions for a given
level of steel stress) is 500C. This assumes that the stress level = as, fJfyk = 0.6, Tlfi = 0.7
and "Is = 1.15.
For prestressing tendons, Bcr = 400C, and for prestressing wires and strands Ber = 350C.
The stress level for both is taken as 0.55.
The thickness of slabs in the tables is to ensure adequacy for the separating function, and
may include the thickness of any non-combustible floor finishes in addition to the thickness
of the structural floor.
The axis distance in the tables should be compared with (1) the axis distance to a single
layer of reinforcement of the same diameter; or (2) the weighted axis distance of bars in the
case of reinforcement in multiple layers or multiple sizes.
12.4.3. Discussion of some features
Columns
Two alternatives are provided. Both methods apply to braced structures. No guidance is
given for unbraced frames.
Method A is based on tests, and has been derived by curve fitting. For different fire
resistance periods and load levels, it provides the minimum dimension of columns and the
axis distance to the main reinforcement. The following limitations to the validity are imposed
to reflect the limits of the test data:
..
The effective height under fire conditions is 10, Ii S 3 m. Under fire conditions, Zo, fi may be
taken as 0.5/ for intermediate floors and between O.Sl and 0.71 for columns in top storeys
of buildings. In the above, I is the distance between the centres of lateral restraints. The
effective lengths might seem small at first sight, but it should be appreciated that the
effective length depends on the stiffness (El) of members meeting at a joint. Due to the
action of fire, the value of E will be considerably lower in the members in the storey in
which the fire occurs compared with those in the floors above and below.
The code limits the first-order eccentricity to that used in the tests. Under fire conditions
e = M
OEd
fJN
oEd
, fi should not exceed e
max
The code recommends a value of O.ISh (or b)
but a range of between 0.15h (or b) and 0.40h (or b). The value of the
eccentricity e under fire conditions may be taken to be the same as in normal
temperature design.
Data are given for different load levels fLfi =NEd,fi/NRd' where NEd,fi is the load in a fire
situation and N
Rd
is the design resistance under normal temperature conditions, calculated
using EN 1992-1-1.
207
208
The code also gives a formula for arnvmg at the fire resistance period when the
parameters are different to those used for the table. This includes load level, ace' axis
distance, number of bars and effective length of the column. The value calculated should be
greater than the actual fire period required.
The alternative method, B, is a subset of Annex C, which has been derived by buckling
analysis as set out in Annex B3. It compares reasonably well with the results of tests although
it is more conservative compared with method A. This method gives data similar to that in
method A, but another variable, the mechanical reinforcement ratio (w), has been brought
into the table. The limits of validity are as follows:
The first-order eccentricity e under fire conditions should not exceed 0.025b or 100 mm.
The slenderness ratio Afi = (lo, fJi) should not exceed 30. The effective length of columns
may be taken as that defined for method A.
The load level n = N
OEd
, f/0.7(AJcd + As!yd)' NOEd,fi may be taken as rlrlVoEd' and 'fJfi may be
taken conservatively as 0.7.
Annex C comprises nine tables grouped as follows: Tables Cl, C2 and C3 apply to w = 0.1
and e = 0.025b, 0.25b and 0.5b, respectively; Tables C4, C5 and C6 deal with w = 0.5b and
e = 0.025b, 0.25b and 0.5b, respectively; and Tables C7, C8 and C9 are for w = 1.0 and
e = 0.025b, 0.25b and 0.5b, respectively. In each table, the minimum size of columns and the
axis distance to the main bars are given for different fire periods, load levels and slenderness
ratios.
Beams
In continuous beams it should be noted that there are requirements for top reinforcement
over supports and beam width over the first intermediate support. In order to use the data
for continuous beams, the above requirements should be complied with, and redistribution
of bending moments for normal temperature design should not exceed 15%. Otherwise the
beams should be treated as simply supported for fire design.
Slabs
The data for two-way spanning slabs are applicable only to slabs supported on all four edges.
If this is not met, the slab should be treated as one-way spanning.
The thickness of the slab for continuous slabs is the same as for simply supported slabs.
But some reduction in the axis distance will apply to continuous slabs if the redistribution of
bending moments for normal-temperature design does not exceed 15% and top reinforcement
over intermediate supports complies with the requirements mentioned for beams.
There is also a requirement to provide top reinforcement of at least 0.5% Ac over
intermediate supports if any of the following apply: (1) use of cold worked reinforcement;
(2) the end supports in a continuous slab do not have fixity or are not detailed in accordance
with EN 1992-1-1; or (3) the load effects cannot be distributed laterally.
In flat slabs, redistribution of bending moment for normal-temperature design should not
exceed 15%. For fire ratings of 90 min and above, top reinforcement should be provided
over the full span in the column strips. The area of reinforcement should be at least 20% of
the total top reinforcement required over intermediate supports.
12.5. Simple calculation methods
Three calculation methods, which will have considerable application in practice, are noted in
Annexes B andE.
12.5.1. 500C isotherm method
This method requires knowledge of the temperature distribution across the cross-section of
the member. EN 1992-1-2 provides such profiles for slabs, beams and columns inAnnexA for
CHAPTER 12. STRUCTURAL FIRE DESIGN
standard time-temperature relationships. The method may be applied to sections, the
minimum width of the cross-section of which complies with the values in Table BI.
The thickness of the zone damaged by heat is first identified. This is taken as the average
depth of the 500C isotherm in the compression zone of the cross-section. This damaged
concrete is discounted and assumed not to contribute to the load-bearing capacity. The
remaining cross-section is assumed to retain the initial values for strength and modulus of
elasticity.
Although some reinforcement bars may lie outside the reduced cross-section, they are still
taken into account in the calculations. By applying the reduction factors appropriate to their
temperature, the strength of all reinforcement is reduced.
With the above assumptions, the problem reduces to the calculation of the resistance of
the cross-section using the methods used for normal-temperature design.
12.5.2. Zone method
This method is more laborious, but more accurate, than the 500C isotherm method.
The first step in the calculation is the estimation of the thickness a
z
of fire damaged zone.
This is carried out for an equivalent wall exposed to fire on both sides. The thickness of the
equivalent wall is the actual thickness in the case of members subject to fire on opposite faces
(e.g. walls, webs of beams and rectangular columns). The thickness is twice the actual
thickness in the case of members subject to fire on one face only (e.g. slabs). The procedure
then is as follows:
..
Divide the half thickness of the equivalent wall into a number (minimum of three) of
parallel zones.
Using Fig. B5a or figures in Section 4 of EN 1992-1-2 in conjunction with the temperature
profile inAnnexA, estimate the reduction factor (kc((J)) at the centre of each zone and
on the centreline of the equivalent wall (kc(BM)' The code provides formulae for the
calculation of the mean reduction factor for half the width of the equivalent wall and the
thickness of the damaged section az.
Proceed with the calculation of the resistance with the reduced cross-section as in the
500C isotherm method.
Although not mentioned in the code, it is the authors' opinion that as an alternative to the
above method, the following procedure can be used:
Discount 25 mm from the faces of the members exposed to fire.
using the temperature profiles in Annex A, estimate the temperature at the level of
reinforcement and in a number of layers of concrete.
It should now be possible to calculate the resistance of the cross-sections from first
principles.
12.5.3. Annex E. method
This is a useful annex for the calculation of the resistance of beams and slabs subject to fire.
Guidance is given for both simply supported and continuous members.
209
References
1. Schlaich, J. and Schafer, K. (1991) Design and detailing of structural concrete using
strut-and-tie models. Structural Engineer, 69,113-125
2. Johansen, K. W. (1962) Yield Line Theory. Cement and Concrete Association, London
3. CEB Commission 'Dalles et Planchers' (1962) The application of yield line theory for
the calculation of the flexural strength of slabs and flat slab floors. CEB Bulletin 35. FIB,
Lausanne.
4. Wood, R. H. and Armer, G. S. T (1970) The Strip Method for Designing Slabs. BRE,
Watford, Current Paper 39170.
5. Whittle, R. T. (1994) Design of Rein forced Concrete Flat Slabs toBS 8110. CIRlA Report
110, revised edn. CIRIA, London.
6. British Standards Institute (2005) Background Paper for the UK National Annex to
BS EN 1992-1-1. BSI, London, BS PD 6687.
7. Comite Euro-International du Beton (1978) CEB/FIP Model Code for Concrete Structures.
FIB, Lausanne.
8. Comite Euro-International du Beton (1982) CEB/FIP Manual on Bending and
Compression. CEB Bulletin d'Information 141. FIB, Lausanne.
9. Co mite Euro-International du Beton (1988) Manual for the Design of Durable Structures.
CEB Bulletin d'Information 185. FIB, Lausanne.
10. Rowe, R. E. (1987) Handbook to British Standard BS 8110:1985: Structural Use of
Concrete. Palladian, London.
11. Beeby, AW. (1978) Cracking and Corrosion. Concrete in the Oceans Report No. 2.
CIRIA Underwater Engineering Group, London.
12. Schiessl, P. and Raupach, M. (1997) Laboratory studies and calculations on the influence
of chloride-induced corrosion of steel in concrete. ACI Materials Journal, 94, 56-62.
13. Comite Euro-International du Beton (1989) CEB Design Guide on Durable Concrete
Structures, 2nd edn. CEB Bulletin d'Information 182. FIB, Lausanne.
16. Clarke, J. L. (1976)A Guide to the Design of Anchor Blocks for Post-tensioned Prestressed
Concrete. Guide No.1. CIRIA, London.
17. Institution of Structural Engineers (2006) Manual for the Design of Concrete Building
Structures to Eurocode 2. Institution of Structural Engineers, London.
Index
abrasion 164
accidental actions 6
accidental eccentricity 121-2
acid attack 164
actions 6-12
characteristic values 6-7
classification 6, 7
design values 7-12
actual tensile strength 198, 199
alkali-aggregate reactions 163-4
alkali-carbonate reaction 163
alkali-silica reaction 163-4
anchorage draw-in or slip 198
anchorage, friction in 197
anchorage of links 174,. 175
anchorage of longitudinal bars 170-4
basic anchorage length 170, 172
design anchorage length 173
transverse reinforcement at anchorage 173-4
anchorage of prestressed concrete 201
anchorage zones 199-200
post-tensioned members 200
pre tensioned members 199-200
Annex E method 209
Application Rules 1-2, 26-7
aspect ratio 101
autogenous shrinkage 52
axial compression, strain limit for 63, 64
basic anchorage length 170, 172
beam
anchorage at supports 179, 180
bending moment coefficients for 49
continuous, design of 38, 39
continuous reinforcement 179, 180
curtailment of longitudinal reinforcement 179
definition 22
design 49
detailing requirements for 177-80
end support 179
intermediate supports 179
longitudinal reinforcement 177-8
shear reinforcement 178-9
singly reinforced 64-6
skin reinforcement 180
structural fire design 208
surface reinforcement 180
bend anchorage 172
bending moment coefficients
for beams 49
for continuous slabs 38, 39
for rectangular panels 41
for two-way spanning slabs 40
biaxial bending
design for 78-82
notation 122
of a rectangular section 78, 79
biaxial column design method, notation in 80
biaxial methods, simplified vs rigorous 80, 81
biaxillary bent column section 81-2
bilinear stress-strain relation 52, 55, 60
bond conditions 170, 171
braced structures 108-9
analysis 28, 29
bracing elements, members transferring forces
to 26
buckling load 121
bundled bars 177
bursting and spalling forces 200
calcium carbonate erosion 164
calcium hydroxide erosion 164
carbonation of concrete 162
CEB Model Codes 62
characteristic tensile strength 198, 199
chloride-induced corrosion 162
chlorides in concrete 162
coefficient of linear thermal expansion 52
cold-worked steel
stress-strain diagram 57
columns
detailing requirements 182-3
longitudinal reinforcement 182
structural fire design 207-8
214
transverse reinforcement 182-3
compatibility torsion 95
continuous beams, design of 38, 39
corbels, detailing requirements 184
anchorage of primary tie reinforcement 184
provision of links 184
corrosion
active phase 149
as cause of cracking 150
design against 163
initiation phase 149
of reinforcement, causes of 161-3
couplers of prestressed concrete 201201
cover to bar reinforcement 170
crack prediction formulae 155-7
crack width limits 148-9
cracking and corrosion 149
leakage 149
appearance 149
cracking
caused by loading 153-4
causes of 149-50
control of 148-59
corrosion as cause of 150
criteria for controlling 193
due to imposed deformations 154-5
phenomena, principles of 153-5
prestressed concrete 193-4
cracking checking
by direct calculation 159
without direct calculation 157-9
creep
of concrete 52, 53
of prestressed concrete 199
creep coefficient 128
cube strength 51
cylinder strength 51
deep beam, definition 22
deflections, spans and 25
deformation 51-2,195
de-icing salts, contamination by 161, 162
design aids 38-49
design anchorage length 173
design by testing 27-8
design chart
for approximate calculation of stress in
prestressed beam 134
for assessing minimum areas of reinforcement
158
for circular columns 68, 77-8
for cracked reinforced concrete columns 136-7
for rectangular columns 68, 69-76
for singly reinforced sections 65, 66
design lap length 175
design moment 121
design tensile strength 55
detailing 169-86
of prestressed concrete 201-3
deterioration of concrete structures 162-4
abrasion 164
acid attack 164
alkali-aggregate reactions 163-4
carbonation of concrete 162
chlorides in concrete 162
corrosion of reinforcement 162-3
design for 165
frost attack 163
leaching by soft water 164
prestressing tendons 162-3
relative importance of deterioration
mechanisms 164-5
sulphates, attack from 164
discontinuous end support 28, 30
on bearings 29,31
doubly reinforced rectangular sections 66-7
drying shrinkage 52
duct friction 197
duct spacing of prestressed concrete 201
ductility classes of reinforcement 61
durability 13-19,161-8
definition 161
historical perspectives 161-2
of prestressed concrete 187-90
effective length/height 109-10, 119
effective reinforcement ratio 156
effective spans 28-9
elastic analysis 21
with limited redistribution 21
with or without redistribution 28-34
elastic deformation of prestressed concrete 197
EN 19902-3,5-6,8-10,12,24
testing 27
EN 199710
encastre beam 35, 36
end conditions 110, 111
epoxy reinforcement 163
EQU8
equilibrium torsion 95
equivalence 2
equivalent frame method 43, 44-6
analysis 46
division into frames 44, 45
horizontal loading 45-6
vertical loading 44-5
Euler equation 107
expansive chemical reactions as cause of
concrete cracking 150
finite-element analysis 21, 47
500C isotherm method 208-9
flange, effective width 31
flanged sections (T or L beams), design of 67
flat slabs 43-9
failure of 45
methods of analysis 43-9
free shrinkage strain 128
frost attack 163
fundamental requirements 5-6
GE09
geometric data 13
grillage analysis 43, 47
high-alumina cement, reduction in strength 162
Hillerborg strip method 39
hogging moment 47
hook anchorage 172
hot-rolled steel
imperfections 25-6
design of slender elements 26
global analysis 25-6
members transferring forces to bracing
elements 26
indirect action 6
intermediate support 28, 30
isolated cantilever 29
jack, friction in 197
jacking stress 190-1
Johansen's yield line analysis 39
L beams
design of 67
effective flange width 31
ladders as shear assembly 179
laps for bars 175
design lap length 175
spaces between lapped bars 175
transverse reinforcement 175, 176
large diameter bars, additional requirements for
175-6
layout 1-2
leaching by soft water 164
lightweight concrete
creep coefficient 57
definition 56
density classes 56
design compressive strength 57
design tensile strength 57
general 56
modifying coefficients 57
modulus of elasticity 57
shrinkage 57
strength classes and associated properties 56, 57
limit states 6
limiting compressive strains 63-83
design for biaxial bending 78-82
design of flanged sections (T or L beams) 67
design of prestressed sections 83
design of rectangular column sections 68-78
doubly reinforced rectangular sections 66-7
moment of resistance of more complex
section shapes 67-8
singly reinforced beams and slabs 64-6
limits
to compressive stress 129-30
to tensile stresses 130
linear thermal expansion, coefficient of 52
load cases and combinations 21-5
load-crack width relationship 154, 155
load-deformation response
for slender columns 115, 116
cracking and 154, 155
load-induced cracking 150, 157
longitudinal reinforcement ratio
concrete grade 91, 92
loop anchorage 172
mandrel diameter for bars 170, 171, 176
material properties 12-13
characteristic values 12
design values 12-13
mechanical reinforcement ratio 65, 208
modulus of elasticity
of concrete 128
of reinforcement 128
moment-curvature response 31-3
moment magnifier method 112, 113-15
moment multiplier method 122, 123
moment of inertia for flanged beams 131-2
moment of resistance of more complex section
shapes 67-8
moment, redistribution of 31-3
monolithic end support 28, 30
Nationally Determined Parameters (NDPs) 2, 9,
88,101,111-12,165
neutral axis depths for flanged beams 1 3 1 ~ 2
nibs, detailing requirements 185
node regions, analysis of 38
nominal curvature method 112, 115-21, 122, 123
calculate the ultimate deflection 117
design moments 117-19
ultimate curvature of the section 116
non-dimensional moment parameter 65
non-linear analysis 21, 35-6
non-sway structures 108
notation 5
notional tensile stress 135
one-way spanning slab
analysis 22
design of 38, 39
P L1 effect 26, 27
parabolic-rectangular diagram 60, 61, 64
for concrete under compression 52, 55
partial factors
for prestress 192
iNDEX
215
216
partial safety factors
for actions 11
for load combinations 12
for material properties 12, 13
pH of pore solution, corrosion of reinforcement
and 162
pin effect 174
pitting corrosion 162
plastic analysis 21, 32, 34-5
plastic hinge analysis 35
plastic settlement, cracking and 149
plastic shrinkage, cracking and 149
Poisson's ratio 52
Portland cement 164
prestress losses 197-9
anchorage zones 199-200
cracking 193-4
deformation 195
design data 190-2
design equations 194-5
lower chord, compression at the
serviceability limit state 195
lower chord, compression at transfer 194
lower chord, tension at the serviceability
limit state 195
upper chord, tension at transfer 195
design of sections for flexure and axial load
192-5
detailing 201-3
durability 187-90
limitation on stress 193
main clauses 187, 188
partial factors 192
prestress losses 197-9
serviceability limit state 193-5
shear 195-7
torsion 197
ultimate limit state 192-3
prestressed sections, design of 83
prestressing steel 58, 190-1
corrosion damage 130
prestressing tendons 162-3
Principles 1-2
punching shear 98-106
basic control perimeter 98-9
design shear force 99-101
enhancement factor 102
resistance of slabs without shear
reinforcement 102-3
reinforcement 103-6, 186
radius of gyration 111
rectangular column sections, design of 68-78
rectangular columns,
design charts for 68, 69-76
rectangular diagram 60
rectangular stress distribution for concrete 53, 55
reinforced concrete columns, stress checks for
135-47
reinforcement in flat slabs 185-6
reinforcement, minimum areas of 150-3
mixtures of prestressing tendons and ordinary
reinforcement 153
values for k 153
values for kc 152
reinforcing bars, corrosion damage 130
reinforcing steel
ductility 57-8
essential properties 57
strength 57-8
related documents 2
relaxation of steel 198-9
ribbed slabs, analysis 23
saw-toothed load-deformation 155
second-order eccentricity 120-2
second-order effects 26-7
serviceability limit states
assessment of design action effects 127-8
characteristic combination 10
control of cracking 148-59
definition 6
frequent combination 11
limitation of stresses under 129-48
material properties 128
procedure for stress checks 130--48
quasi-permanent combination 11-12
verification 13
shear 85-95, 195-7
general 85
maximum shear strength of a section 89
members without shear reinforcement 85-6
serviceability 197
shear capacity enhancement near supports 90-1
strength of sections with shear reinforcement
86-9
ultimate limit state 195-6
shear resistance 93
shift rule 179, 203
shrinkage
of concrete 52, 53, 54
of concrete cracking 150, 151
of prestressed concrete 199
simplifications 49
simplified coefficients 43, 46-8
allocation of moments between strips 47
division of panels 46-7
lateral distribution of moments in the width of
the slab 46
moment transfer at edge columns 47-9
simplified load combinations 12
singly reinforced beams 64-6
singly reinforced slabs 64-6
slab-column connection
lightly loaded 104-5
heavily loaded 105-6
slabs
analysis 22
anchorage at supports 179
corner and edge reinforcement 18]
detailing requirements 181-2
flat, reinforcement in 185-6
longitudinal reinforcement 181
shear reinforcement 182
singly reinforced 64-6
skin reinforcement 180
span reinforcement 181
surface reinforcement 180
transverse reinforcement 181
slender elements, design of 26
slenderness
code rules 107-8
column design 107-12
slenderness effects, design for 112-25
basic approaches 112
design moments 117-19
first-order moments 112-13
global second-order effects 123
lateral buckling of slender beams 124-5
moment magnifier method 112, 113-15
nominal curvature method 112, 115-21
ultimate curvature of the section 116
ultimate deflection calculation 117
walls 123-4
slenderness ratio 109, 111, 119,208
smeared truss 87
spacing
of bars 170, 171
between lapped bars 175
specification and production 51
squash load of a column 79
St Venant torsional stiffness 95
stainless steel reinforcement 163
stiffness parameters 28
stirrup cage as shear assembly 179
STR 8-9
straight bar anchorage 172
strength of concrete 51
stress block parameters 62-3
stress checks, procedure for 130-48
stress-strain curves/relationships 59-63
bi-linear 59
for concrete 52-5, 59, 61
hot-rolled steel 57
idealized 60, 61
for ordinary reinforcement 59,60
for prestressing steel 59, 60, 82, 83
structural analysis 28-38
structural fire design
actions 206
aims of design 205
beams 208
calculation methods 208-9
columns 207-8
design procedure 205-6
material factors 206
member analysis using tabular data 206-8
partial factors 206
scope 206-7
slabs 208
strut-and-tie models 21, 36-8, 184, 200
strut angle 88-9, 90
sulphates, attack from 164
super-ellipse 79
surface coatings 163
sway frames 112
sway structure 108
Tbeams
design of 67
effective flange width 31
T section, shear design 93-5
temperature changes
as cause of concrete cracking 150, 151
stresses due to 148
tendon profiles in prestressed concrete 203
tendon spacing of prestressed concrete 201
tendons, minimum area of, in prestressed
concrete 201-3
tensile strength of concrete 128
terminology 2-3
time-dependent effects 27
time-dependent losses of prestressed concrete
198-9
torsion 95-7,197
evaluation of torsional moments 95
notation 97
verification 96-7
transverse reinforcement
at anchorage 173-4
for lapped bars 175, 176
truss angle 88
truss model 86, 87
two-way spanning slabs, design of 38-43
loads on supporting beams 43, 44
rectangular panels with restrained edges 40-2
simply supported slabs 40
unequal edge conditions in adjacent panels 42-4
ultimate limit state
accidental design situation 10
definition 6
equilibrium 9
persistent and transient situations 8-9
prestressed concrete 192-3, 195-6
shear 195-6
strain at 63
strength 9-10
verification 13
ultimate strain distributions 63
unbraced structures 108, l09
analysis 28, 30
INDEX
217
218
uniaxially bent rectangular column
design ultimate moment 119-21
verification 13
waffle slabs, analysis 23
walls
detailing requirements 183
horizontal reinforcement 183
reinforced concrete 123-4
transverse reinforcement 183
vertical reinforcement 183
water, soft, leaching by 164
weld mesh, requirements for 176-7
welded transverse bar anchorage 172
welds inside bends 176
welds outside bends 176
yield line analysis 35, 43
zone method 209

DESIGNERS'
GUIDES TO THE
EUROCODES
. . . ~ : i J ' ~
an ice initiative

Designers' Guide To en 1992-1-1 and en 1992-1-2

Uploaded by

Copyright:

Available Formats

Designers' Guide To en 1992-1-1 and en 1992-1-2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Designers' Guide To en 1992-1-1 and en 1992-1-2

Uploaded by

Copyright:

Available Formats

What are the Eurocodes?

What topics are covered in the Eurocodes?

DESIGNERS'

You might also like