Stability of Buildings Part 3 Shear Walls

Stability of buildings
Part 3:
Shear walls
March 2015
Author
A Gardner MEng(Hons) MA(Cantab) CEng MIStructE (The Institution of Structural Engineers)
Consultees
P Perry BSc(Hons) CEng MIStructE MICE MHKIE (CH2M Hill) Chairman of the Reviewing Panel
E Bennett MEng (Arup)
O Brooker BEng CEng MIStructE MICE MCS (Modulus)
Dr A S Fraser BEng(Hons) PhD CEng MIStructE MICE (Arup)
J Guneratne BSc(Hons) CEng MIStructE (CH2M Hill)
R Marshall BEng(Hons) MIPENZ (Buro Happold)
Acknowledgements
The use of Arup internal guidance documents in developing this Guide is gratefully
acknowledged.
Photographs and digital imagery have been supplied by courtesy of and are published
with the permission of the following organisations and individuals:
Figures 2.4, 3.1, 6.7, 6.8, 7.2: Arup
Figure 7.11a: The Structural Timber Association
Figures 7.11b, 7.14: Milner Associates
Figure 8.2b: ‘Reinforced concrete frame with brick masonry infill walls, India’ by
A. Charleson (GEM Nexus website [or http://www.nexus.globalquakemodel.org]) is
licensed under CC BY 3.0 (http://creativecommons.org/licenses/by/3.0/)
Box 3.2: J.K. Nakata, United States Geological Survey
Box 3.3: Halcrow Atkins Joint Venture
Box 6.3: SKM anthony hunts
Boxes 7.1, 7.5, 7.6b: Arup
Box 7.3: Wellcome Images
Box 7.4: P Buffett
Box 7.6a: Frank P Palmer
All other photographs and all hand illustrations: A Gardner
Published by The Institution of Structural Engineers,

47–58 Bastwick Street, London EC1V 3PS, United Kingdom
Telephone: þ44(0)20 7235 4535 Fax: þ44(0)20 7235 4294
Email: mail@istructe.org Website: www.istructe.org
First published 2015

ISBN 978-1-906335-27-4
# 2015 The Institution of Structural Engineers
The Institution of Structural Engineers and those individuals who contributed to this Guide have
endeavored to ensure the accuracy of its contents. However, the guidance and recommendations
given in the Guide should always be reviewed by those using the Guide in the light of the facts of
their particular case and specialist advice obtained as necessary. No liability for negligence or
otherwise in relation to this Guide and its contents is accepted by the Institution, the author, the
consultees, their servants or agents. Any person using this Guide should pay particular
attention to the provisions of this Condition.
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any
form or by any means without prior permission of The Institution of Structural Engineers, who may
be contacted at 47–58 Bastwick Street, London EC1V 3PS, United Kingdom.
Glossary
The following definitions are provided to explain how the listed terms are used specifically in this Guide. They
may differ to definitions found in other documents.
Term Definition
Action or load An influencing effect, normally external to the structure, that causes movement, deformation and/or internal
stresses. The two terms are largely interchangeable; ‘action’ is favoured by the Eurocodes while ‘load’ is more
common throughout the design codes of English-language countries. Both terms are used herein.
Braced Stabilised in sway by other connected elements or systems such that the subject element does not experience a
significant sway moment. Structural elements or systems that are not braced are ‘unbraced’. The term braced is
not to be confused with ‘bracing’ which is a type of structural component.
Centroid The point on a cross-section, defined by the intersection of the neutral axes, through which a pure axial load will
result in a uniform stress.
Coupled Two or more elements with joints that resist longitudinal shear and make the flexural stiffness of the system
greater than the sum of the parts.
Deflection head A detail that allows differential vertical movement at the top of a non-loadbearing wall/panel. The detail may or
may not provide in- and/or out-of-plane shear resistance.
Effective height The length of an ideal pin ended wall that would have the same buckling load as the actual wall to which it relates.
Force A type of action, causing both stresses and strains within a resisting static structure.
Horizontal An element, frame or assembly orientated in a horizontal or near horizontal plane that transfers lateral actions
stability system through the structure (generally connecting the façade to the vertical stability systems).
Hybrid system A structure that contains more than one type of vertical stability system. This is usually with a combination of
shear walls, framed bracing or portalisation.
In-plane Characteristics (e.g. stiffness) or effects (e.g. actions) considered in a vertical plane that are orthogonal to a
wall’s or wall system’s major axis. For a planar wall, these are the characteristics or effects in the plane of the
wall. Out-of-plane characteristics or effects are those orthogonal to a system’s minor axis.
Insulated Also known as ‘structural insulated panel systems’ (SIPS), a pre-cast composite wall element combining layers of
sandwich materials to deliver structural and insulation (acoustic and/or thermal) performance. The structural component is
panels typically either reinforced concrete or timber.
Lateral load- The total structural system that acts to resist lateral loads comprising both horizontal and vertical stability
resisting system systems together with façade elements (windposts, cladding rails, etc.) and the substructure.
Length The longer plan dimension of a wall. For planar walls, this is orthogonal to the major axis.
Loadbearing A wall that resists vertical actions. Walls that resist in-plane lateral forces but not vertical actions are defined as
‘non-loadbearing’ herein. Meanwhile partitions and façade elements that only resist out-of-plane local pressures
are defined as ‘non-structural’.
Load path The complete route by which any applied or induced stress is transmitted through a structure to the foundations
via a system of interconnected elements.
Modular ratio The ratio of moduli of elasticity for two composite materials. In the context of reinforced concrete, it is the ratio
of the steel modulus divided by the concrete modulus.
Non-structural An element that can be removed without detrimental impact on the retained structure.
Precast Used generically for any construction process that is not completed in situ. This includes processes that are
more accurately described as pre-formed or pre-assembled.
Secant modulus The average gradient (stiffness) plotted as a straight line between two defined points on a stress–strain curve. It
of elasticity differs to the tangent stiffness which is the gradient of a tangent to the stress–strain curve at a given point on
the curve.
Shear centre The longitudinal axis through which a transverse shear will cause a linear displacement without twist, and about
which a torque will cause pure rotation.
Slender An element that is prone to buckling at a load less than the material strength would imply. Structural elements
that are not slender are ‘stocky’.
Soft storey Those storeys that have undergone ill-conceived structural alterations leading to a structural system that is
particularly highly utilised and vulnerable to failure.
U-value The thermal transmittance of a material, it is a measure of the power loss per metre squared per degree Kelvin
temperature gradient (standard units W/m2/K). It is the inverse of the thermal resistivity.
Vertical stability An element, frame or assembly orientated in a vertical or near vertical plane that transfers lateral actions through
system the structure (generally down towards the ground). These systems form part of the lateral load-resisting structure.
iv The Institution of Structural Engineers Stability of buildings Part 3

Notation
The following notation is used for hand drawn figures:
An applied force
An applied bending moment
A movement/displacement
An out-of-plane surface pressure
Centre of stiffness
Lateral restraint
Solid wall section
Structural lintel beams across wall openings
Components of stiffness
The following notation is used in equations. Further notation is defined in the body text and within figures
where used.
A Cross-section area
E Young’s modulus of elasticity
G Shear modulus
I Section second moment of area
J Section torsion constant
L Wall height
Le Effective wall height
b Wall cross-section length on plan
k Stiffness
t Wall cross-section thickness
e Strain
s Normal stress
t Shear stress
The Institution of Structural Engineers Stability of buildings Part 3 v

Contents
Glossary iv 7 Non-monolithic shear wall construction 40

7.1 Introduction 40
Notation v 7.2 Precast construction 40
7.3 Precast reinforced concrete wall construction 42
Foreword vi 7.4 Hybrid precast in situ reinforced concrete wall
construction 43
Part 3: Shear walls 7.5 Timber and light gauge steel ‘platform’ frame
construction 45
1 Introduction 1 7.6 Mass timber 47
7.7 Loadbearing masonry 49
2 Design overview 4 7.8 Steel plate diaphragm walls in steel framed
2.1 Form and configuration 4 buildings 51
2.2 Resistance and force transfer 4 7.9 References 52
2.3 Failure mechanisms 5
2.4 Materials 6 8 Shear infill panels 53
2.5 Monolithic and jointed construction 6 8.1 Introduction 53
2.6 Coupled behaviour 6 8.2 Common characteristics of infill systems 53
2.7 Buckling and buckling restraint 6 8.3 Masonry infill panels 54
2.8 Slenderness and effective heights 7 8.4 References 56
2.9 Limit state philosophy and initial wall sizing 9
2.10 References 10
3 Requirements of walls 11
3.1 Introduction 11
3.2 Wall locations 11
3.3 Non-structural partitions and non-loadbearing
panels 11
3.4 Cores 12
3.5 Vertical access and transportation 12
3.6 Service risers and distribution 14
3.7 Insulation and compartmentalisation 15
3.8 References 15
4 Elastic theory of thin-walled sections 16

4.1 Introduction 16
4.2 Complementary shear 16
4.3 Torsion 16
4.4 Warp and warp restraint 16
4.5 Lintel beams in sections subject to torsion 18
4.6 Centroid and shear centre 19
5 Modelling and analysis 21

5.1 Introduction 21
5.2 Modelling simplifications 21
5.3 Modelling vertical stability structures 22
5.3.1 1-dimensional element models 22
5.3.2 Grillage models 26
5.3.3 2-dimensional finite element models 28
5.4 Modelling horizontal stability systems 30
5.5 Manually apportioning actions between vertical stability
systems 31
5.6 Modelling boundary conditions 31
5.7 Elastic and plastic analysis 32
5.8 References 32
6 Monolithic reinforced concrete shear wall

construction 33
6.1 Introduction 33
6.2 Modelling the stiffness of concrete 33
6.3 Ultimate and serviceability limit state design of reinforced
concrete sections 34
6.4 Concrete classes 35
6.5 Minimum wall thickness 35
6.6 Reinforcement and embedments 35
6.7 Construction 36
6.8 References 39
The Institution of Structural Engineers Stability of buildings Part 3 iii

Part 3: Shear walls
1 Introduction
This Guide is the third part in a four-part series on plane horizontal forces may be termed shear walls,
lateral load-resisting ‘stability’ systems for buildings. cross walls, spine walls, raking panels or vertical
Its focus is the use, analysis and design of shear diaphragms. These terms are largely interchangeable,
walls. though each has origin in a different material trade.
To avoid ambiguity, the term ‘shear wall’ is adopted
Walls have provided stability to buildings generally throughout this Guide. A further term ‘shear
intentionally or otherwise for many centuries. They infill panel’ is used to describe the subset of walls
are almost exclusively vertical and are inherently that resist only in-plane lateral loads (i.e. not vertical
stiff in-plane. However, perhaps more significant to loads). Meanwhile, the term ‘wall system’ is used
their widespread application, walls are essential herein to discuss any arrangement of coupled shear
components of most buildings to divide space walls.
and isolate environmental conditions. This remains
as true today as it has done throughout history This Guide discusses walls that are planar, flanged or
and it is rare that the structural design of walls is arranged as a core. The torsional behaviour of these
devoid of non-structural criteria. Continuing to wall systems is given specific attention – with
evolve are acoustic, thermal and fire criteria Chapter 4 dedicated to fundamental theory.
together with material technologies and Chapter 5 discusses modelling techniques focusing
construction techniques. mainly on computer analysis methods.
Relatively recent changes in building form have seen This Guide describes common characteristics of
an ascendency in open-plan, flexible shear walls before considering various
accommodation, paired with ever taller and larger constructions in turn. These are split into three
buildings. Through this evolution, framed structures categories: in situ monolithic shear walls
have come to the fore and walls have become (Chapter 6), non-monolithic shear walls (Chapter 7)
engineered systems that fulfil specific criteria. Modern and non-vertical loadbearing (non-loadbearing)
construction distinguishes between non-structural shear infill panels (Chapter 8).
‘partitions’, ‘cladding’ and ‘façade’ elements that are
isolated in-plane from the structure on the one hand, Relevance and aims
and ‘structural walls’ that transfer in-plane forces on This Guide is an introduction written primarily for
the other. Within the subset of walls classed as design engineers, particularly those approaching a
structural, walls can resist vertical forces, in-plane professional review. Together with Part 1 of the
horizontal forces, out-of-plane horizontal forces or a Institution’s sister publication Stability of buildings,
combination thereof (Figure 1.1). Walls that resist in- Parts 1 and 2: General philosophy and framed
bracing, it introduces many generic principles –
theoretical and applied – that are associated with
shear wall design. However, it is not the intention that
this Guide provides detailed ‘how to’ instructions on
design, and reading it will not automatically make a
Vertical actions designer proficient to work autonomously. Rather, this
Guide aims to supplement supervised learning by
increasing background knowledge on the topic. To
this end, we hope the Guide helps promote a
In-plane thoughtful attitude towards design that is based on a
horizontal breadth of knowledge.
actions
Finally it is important to note that this Guide is not
limited to a discussion of ‘best practice’ systems.
Chapters 7 and 8 intentionally include a variety of
systems not all of which are in favour, at least in the
UK. These systems exist in buildings standing today
and many of today’s engineers will be required to
carry out designs for retrospective alterations. This is
the motivation for including any guidance on
Out-of-plane controversial systems.
horizontal
actions
Further reading: stability
The following text is a recommended source of further guidance:
– Institution of Structural Engineers. Stability of buildings.
Parts 1 and 2: General philosophy and framed bracing.
London: IStructE, 2014
Figure 1.1 Actions on structural walls
The Institution of Structural Engineers Stability of buildings Part 3 1

Introduction
Designers’ checklist
Actions – applied:
Are minimum and maximum gravity load cases considered?
Are wind, soil, ground surcharge and hydrostatic lateral forces considered?
Are accidental and extreme actions including impact, fire and earthquakes considered?
Actions – induced:
Will actions result from the restraint of arches, domes, catenaries, nets?
Will actions result from initial imperfections?
Will actions result from inelastic strains?
Will actions result from the restraint of post-tensioning and other elastic strains?
Second-order PD effects:
Is the structure sway sensitive/do PD effects need to be considered?
Combinations of actions:
Are all critical combinations for all elements/failure mechanisms evaluated?
Accommodating movement:
Are viable movements understood and quantified?
Are any movement joints necessary and/or incorporated?
Are these accurately portrayed in the analysis?
Are significant movements resisted by the structure?
Are corresponding forces (actions and reactions) allowed for throughout the load path?
Does the design take due account of force redistribution resulting from creep or ground movement?
Are all parts of the structure adequately served by load paths to ensure stability, noting load paths and movement joints are irreconcilable?
How many independent structures exist; is each one stable?
Load paths:
How do forces acting on the façade transfer to the horizontal stability systems? Where the façade spans onto beams, are they
restrained or bending in their minor axis?
How do forces acting on the horizontal stability systems transfer to the vertical stability structures?
How stiff are both the horizontal stability structures and the connections from the horizontal to vertical stability structures?
How do forces transfer through the vertical stability structures?
How are forces transferred from the superstructure into the substructure?
How are forces transferred from the substructure into the soil?
Are the interfaces of the above six line items adequate?
Are there any aspects of the structure, small or large, that do not follow the normal pattern? Do these have suitable load paths of
resistance?
Are all eccentricities accounted for in the analysis?
Braced or unbraced:
Is the structure braced, unbraced, or a hybrid?
Are effective heights correctly determined, taking account of relative stiffnesses and joint rotations where necessary?
Design – stability, strength, service and robustness:
Is the structure in static equilibrium: rotational and linear?
Are all elements and connections adequate to transfer the design actions?
Are deflections, rotations and the natural frequency each within permissible bounds?
Is the structure deemed robust in the event of failure to any of the stability structures? Does the design safeguard against
progressive collapse?
Construction:
Is the disposition of the stability system, and are all design assumptions communicated to the contractor?
Is the disposition of the stability system, and are all design assumptions communicated to the contractor?
Are all parties clear and in agreement on their responsibility?
Is the transfer of information understood by and compatible to all parties, e.g. are actions characteristic or factored values?
Where existing structures are involved, is the stability of these understood before demolition works start?
Are new and existing parts to be connected or isolated from one another?
Alterations and maintenance:
Will new structure provide support to, or act on existing structures?
Are ‘as built’ records available for the existing structure?
Are these accurate to the structure and inclusive of any previous modifications?
Can elements within the completed structure be maintained?
Figure 1.2 Designers’ checklist for design of lateral load resisting stability systems
2 The Institution of Structural Engineers Stability of buildings Part 3

Introduction
Limitations
This Guide does not cover general considerations of
lateral load-resisting ‘stability’ systems or the broader
topics of actions, movements and load paths. These
are discussed in Part 1 of this series. It is
recommended that designers familiarise themselves
with this before embarking on a stability design.
Figure 1.2 reproduces the designers’ checklist from
Part 1 for quick reference.
Note that the checklist given in Fig. 1.2 is intended to

serve as a prompt for designers. It only concerns the
design of lateral load-resisting systems and is
generic, to be considered with the project context in
mind.
In line with Part 1 and Part 2 of this series,

earthquake design is largely omitted from this Guide.
So too are the advanced topics of non-linear and
dynamic analysis. Each of these is introduced in
Part 1 but is considered too advanced to be catered
for in this introductory text.

2 Design overview
2.1 Form and configuration
There are three basic types of structure that include

shear walls (Figure 2.1):
– Cellular structures with vertical loadbearing shear
walls.
– Framed structures with vertical loadbearing shear
walls.
– Framed structures with non-loadbearing shear infill
panels.
(a)
Each contains wall elements that are generally much
stiffer in one principal axis than they are in the other.
However, how these elements behave is largely
governed by whether they are isolated (and acting as
a uniform planar section) or coupled (and acting as a
compound section). Examples of each are shown in
Figure 2.2.
Within a building, isolated or coupled walls can be

arranged in any configuration that provides:
– Adequate resistance in two orthogonal horizontal
directions
(b)
– Adequate torsional resistance
– Adequate robustness
Note that, in this context, ‘resistance’ should be

interpreted to mean both stiffness and strength.
Fig. 2.1 shows the three structural types idyllically in

isolation but they can equally contribute to a hybrid
system. Two examples are shown in Figure 2.3: one
with the different systems acting in a single axis and
the other with the different systems acting orthogonal
to one another.
(c)
Figure 2.1 Structural forms: (a) cellular, (b) framed with

loadbearing walls, (c) framed with infill walls
2.2 Resistance and force transfer
to vertical and horizontal actions acting in their plane.
Shear walls act as vertical stability systems within In this way they transfer forces vertically to their
load paths that extend from a building’s extremities to founding structure. This transfer is via an internal
its foundations. They are generally regarded as stress distribution that results from compatible axial,
primary structure and provide relatively stiff resistance shear, torsional and flexural strains.
Tubular
Flanged ‘core’
Coupling across
openings
Coupling at
intersections


























(a) (b)
Figure 2.2 Wall systems: (a) isolated, (b) coupled

Design overview 2.3
components, local strengthening with additional

elements/reinforcement and/or local restrictions on
the placement and size of wall or slab penetrations.
In multi-walled statically indeterminate structures, the

relative stiffness of each wall together with its
boundary conditions will influence the distribution of
forces between walls. The wall-floor connections can
be considered as one set of boundary conditions
and the founding structure another. None will be of
Shear infinite stiffness and collectively there can be
walls significant stiffness variation within a structure. The
Framed bracing rotational stiffness of the founding structure and
translational stiffness of the floor systems tend to
have greatest impact.
Note that any movement joints form boundary

conditions with zero stiffness in the axis of the free
movement. This is critical when devising the stability
strategy and is discussed in more detail in Part 1 of
this series2.1.
Shear 2.3 Failure mechanisms

walls
Framed bracing / Shear walls

The ultimate limit state of a shear wall may be
governed by one of four mechanisms shown in
Figure 2.3 Example hybrid stability systems Figure 2.5. These mechanisms are additional to the
equilibrium failure modes that include ‘rigid-body’
The overall height to length aspect ratio of a wall, overturning and sliding.
together with the material characteristics, will often
The critical mechanism will be dependent on all
govern whether shear or flexural stresses dominate
aspects of the wall: its geometry, loading, material
the global behaviour. Shear tends to govern in
properties, restraint, and construction. While failure
short, stocky walls and it is usually acceptable to
could occur anywhere, ill-conceived joints are
dismiss in-plane flexural deformation for walls with a
particularly vulnerable and can create paths of
height to length (L/b) ratio of less than 2.
weakness in non-monolithic structures. This is,
Conversely, taller more slender walls are dominated
however, generally to be avoided as it can lead to
by flexure. Joint slip in precast assemblies and local poor robustness.
deformations around voids in coupled walls will,
however, influence the global behaviour and can lead Note that the failure mechanisms listed are those
to complex stress distributions in which plane resulting from vertical (axial) and in-plane horizontal
sections do not remain plane. actions only. Walls with significant out-of-plane
horizontal actions may also fail in minor axis bending
Actions may apply directly to a wall (e.g. gravity or shear.
acting on the wall’s mass) but a large component of
the total action tends to transfer from the connecting
structure (usually via beams and/or the slab at each
floor – Figure 2.4). This can cause stress 2.4 Materials
concentrations at the wall-floor connections that
extend into both the wall and the surrounding floor.
Common options to overcome insufficiencies in these Reinforced concrete, masonry and timber are the
connection regions include increasing the wall/slab most common materials for shear wall construction
thickness, increasing the size of any connection worldwide.
(a) (b) (c) (d)
Figure 2.4 Wall-floor beam connections showing steel beams Figure 2.5 Structural failure mechanisms: (a) tension and/or compression, (b) horizontal
and reinforced concrete core wall shear, (c) vertical shear, (d) buckling

2.5 Design overview
Masonry and reinforced concrete are widely available, 2.5 Monolithic and jointed construction
have high compressive strength and are non-
combustible. They also both allow large, heavy
panels to be constructed in situ with manageable Structural shear walls fall into two categories: in situ
constituents. and monolithic or precast with joints. Only in situ
reinforced concrete leads to structures that can be
Timber, used mostly as an engineered product, is considered wholly monolithic. All other materials tend
lightweight and sustainable when sourced from to result in structures made up of parts and joints.
managed plantations. It is however combustible and
the least stiff of the materials listed. It also suffers The implications of each are discussed in more detail
from significant shear deformation and can suffer in Chapters 6 and 7.
dimensional instability.
Steel diaphragms are used in substitution for

reinforced concrete in certain regions including North 2.6 Coupled behaviour
America and North East Asia. These are suitable for
reasonably tall structures. Meanwhile modular steel-
concrete hybrid systems (e.g. Tata Corefast2.2) are Many wall systems, monolithic or jointed, contain
available and can be favourable where the speed of coupled panels (Fig. 2.2). The panels that make up
construction is critical2.3. these systems may be in a single plane framing
openings with coupling beams (lintels) above and
Plasterboard is a recognised option for sheathed below the opening; otherwise they may be non-
panels in lightly loaded modest structures (e.g. planar with multiple wall planes joining to form a
English Building Regulations Class 1 structures2.4) flanged or channel section in plan.
although wood-based sheathing (e.g. plywood) is
widely considered more robust. Other materials Two neighbouring wall panels can be considered
including glass and composites have been used for coupled when the interface transfers longitudinal
structural walls when warranted. These are however (vertical) shear to resist the deformation mode shown
highly specialist, non-standard, and are outside the in Fig. 2.5(c). This stress arises whenever a section
scope of this Guide. experiences a flexural or restrained warping stress
and its magnitude is dependent on the stiffness of
the coupling element.
Depending on this stiffness, the performance of a

coupled section will fall between that of:
– An ideal uniform element of similar gross plan
Plan: cross-section.
– The combined performance of the independent
(non-coupled) component parts.
Elevation: It is important to appreciate that coupling enhances

the overall flexural stiffness disproportionately to the
shear stiffness. For buildings of equal overall stiffness,
one containing coupled walls will deform more in
shear (and less in flexure) than one with isolated
walls. This change to the deflection bias can heighten
Stress:
the relative sensitivity of the coupled wall structure to
second order PD effects2.1.
(a) (b) (c) (d)
In a similar light, coupling may influence the torsional
Figure 2.6 Stress distributions for planar walls: (a) solid, (b) and (c) coupled, (d) discrete stiffness disproportionately to the flexural stiffness
(see Section 4.5).
The coupling effect is illustrated in Figure 2.6 for a

series of planar walls, and Figure 2.7 for a non-planar
arrangement. Both figures plot the longitudinal stress
Plan: resulting from a major axis bending moment. In
reality, longitudinal shear deformations together with
joint slip (where joints are present) can impact on this
stress distribution.
Elevation:
2.7 Buckling and buckling restraint
Shear walls and panels subject to in-plane shear and/

Stress:
or axial compression are vulnerable to buckling failure
if slender. Three modes can act:
(a) (b) (c) – Euler major axis (in-plane) buckling due to axial
compression
Figure 2.7 Stress distributions for flanged walls: (a) perfectly coupled, (b) partially – Euler minor axis (out-of-plane) buckling due to axial
coupled, (c) discrete compression

Design overview 2.8
Vz
Nx Nx Pinned Rigid or semi-rigid
δy δy
δz
Rigid Pinned
x or rigid
x x
y y y Shear walls Moment frame

z z z
Figure 2.8 Buckling modes for walls subject to in-plane shear and axial force Figure 2.9 Comparison of behaviour of shear wall and
moment frame structures
– Lateral torsional (out-of-plane) buckling due to an adopted by BS EN 19962.5 and BS EN 19922.6

in-plane (major axis) bending moment respectively.
These are shown for a planar wall in Figure 2.8 but More recently, the slenderness ratio has also been
may apply to walls of more complicated cross- defined as a function of the resistance without buckling
section geometry. divided by the elastic critical bucking load. This is the
definition used, for example, by BS EN 19932.7.
Both lateral torsional buckling and Euler buckling can
apply simultaneously, impacting on the axial The slenderness ratio relates to a ‘slenderness limit’
compression and major axis bending moment that is the cut-off between elements being classed
capacities of a wall. However, only one Euler mode ‘slender’ (and thus vulnerable to buckling) or ‘stocky’.
will govern; whether this is in the major or minor axis Both the ratio and the limit must always be taken
is dependent on the effective unrestrained heights from the same code.
and flexural stiffnesses of the wall in each of these
axes. Slenderness of unrestrained wall system
The slenderness of an unrestrained wall system
Individual shear walls can often be regarded as concerns global buckling and is a function of the
statically determinate ‘unbraced’ cantilevers in their
major axes. Floor systems will usually provide an in-
plane link between walls but do not normally provide x
a significant coupling effect. This differentiates the z
majority of shear wall structures from moment frames
(Figure 2.9). y
(Ley = Lez = 2L)
The minor axis may similarly be unbraced, in which

case minor axis Euler buckling will always govern
L
over major axis. A single central core is a common

scenario when this may apply (Figure 2.10).
Alternatively, an individual wall may be restrained at

discrete points through its height. Restraint can be to
each or either of the major and minor axes of a
system. However, in practice, it is almost exclusively
Figure 2.10 Euler buckling effective heights of an
applicable to the minor axis of a wall system with low
unrestrained system
minor axis stiffness (‘low’ stiffness being measured
relative to the stiffness of the restraining systems
(Figure 2.11).
x z
2.8 Slenderness and effective heights
y
(Ley = 2L)
Slenderness is a measurement used to combat the

Lez
L
buckling in members subject to axial compression. A

similar approach is used for walls as is used for
columns.
The ‘slenderness ratio’ of a wall is traditionally

defined as a function of the effective height divided
by either the effective thickness or the radius of Figure 2.11 Euler buckling effective heights of a system
gyration of the wall section. These definitions are restrained in the minor axis

2.8 Design overview
overall height of the wall L. The effective height Le is

that of a strut or cantilever that is free at its tip
(Fig. 2.10).
Slenderness of restrained wall systems and

component parts
The slenderness of a restrained wall system, or of a
restrained component within a system, primarily (a) (b) (c)
concerns the spacing of the restraints. Restraint can
be provided by floor diaphragms linking between a
number of vertical stability systems (Fig. 2.11).
Vertical return walls within non-planar coupled wall
systems may also provide restraint to individual
panels. Thus a single wall panel can be:
– Restrained along one edge (Fig. 2.12(a))
– Restrained along two opposite edges (Fig. 2.12(b))
– Restrained along two adjacent edges (Fig. 2.12(c))
(d) (e) (f)
– Restrained along three edges (Figs 2.12(d) and
2.12(e)) Notes
– Restrained along four edges (Fig. 2.12(f )) a Condition (a) is unrestrained and equivalent to the
condition shown in Fig. 2.10.
These restraint conditions influence the effective b Condition (b) is equivalent to the conditions shown in
height, usually by means of a factor applied to the Fig. 2.11.
actual wall dimension and/or restraint spacing. Each
of the Institution’s manuals on concrete2.8 (Section
Figure 2.12 Wall panel restraint configurations
5.6.2.1) and masonry2.9 (Section 5.3.2) define
simplified factors for walls restrained along two
opposite edges (i.e. one-way spanning braced walls). to act as vertical edge restraints; further simplified
The latter also defines factors for braced masonry guidance for small buildings of traditional (masonry)
walls restrained on three and four edges. Meanwhile, construction is contained in the English Building
Clause 12.6.5.1 of BS EN 1992 Part 1-12.6 lists Regulations2.4.
simplified factors for lightly reinforced concrete walls
restrained along three and four edges. In accordance Figure 2.13 illustrates the requirements for concrete
with Clause 5.8.3.2(7), these factors may be used for walls as set out in BS EN 19922.6. Where these, or
normally reinforced concrete. similar, rules are not achieved by a return wall (i.e. the
return wall is too small), the return can be counted as
BS EN 1992 Part 1-12.6 Clause 12.6.5.1(3) and the part of the panel’s geometry contributing towards the
Institution’s manual for masonry2.9 each give effective thickness or radius of gyration. To this end,
guidance on the disposition of return walls necessary the arrangement shown in Fig. 2.13 may be regarded
L
br >
5
b
Restraint from slab
diaphragms top and bottom
hr > 0.5h
Return wall
Subject wall L
where:
b is the length of the wall
L is the height of the wall
h is the thickness of the wall
br is the length of the return (supporting) wall Short stub wall inadequate as a restraint but may count towards
hr is the thickness of the return wall the subject wall’s effective thickness or radius of gyration
Figure 2.13 Illustration to support BS EN 1992 Part 1-1 Table 12.1 and Clause 12.6.5.1(3)

Design overview 2.9
as two panels, each locally restrained on three sides

(Figure 2.14).
Carefully located and sized return walls can have a

considerable enhancing effect on the slenderness of
individual panels. To work most efficiently, return walls
should be of sufficient length to provide edge support
to the adjoining panel, but not be so long that they
become critically slender elements in themselves.
When a wall is classed as slender (with the

slenderness ratio above the slenderness limit) the
design must account for second order effects. This
may be conducted via a non-linear second order
analysis. Alternatively many codes allow simplified
methods. The most common is to design for a
moment that must be considered in addition to the
actions resulting from first order linear elastic analysis.
This additional moment makes allowance for the
tendency to buckle by assuming the axial force acts
eccentric to the section. BS EN 1992 Part 1-12.6, by
example and specific to reinforced concrete, includes Figure 2.14 Restraint conditions for the wall shown in Fig. 2.13 (exploded view)
two alternative methods to account for this additional
moment in Clauses 5.8.7.3 and 5.8.8.2.
This is a significant difference, not to be overlooked.
Further reading: slenderness and effective heights Taking a monolithic slab spanning a narrow doorway
The following text is a recommended source of further between walls as an example (Figure 2.15): in the
guidance on slenderness and effective heights of reinforced absence of a lintel a slab will contribute very little
concrete shear walls: coupling stiffness between walls. It is generally
– The Institution of Structural Engineers (2014) Technical inadequate to provide meaningful coupling effect and
Guidance Note 10 (Level 2) ‘Design of reinforced concrete can therefore be ignored at the ultimate limit state.
walls’, The Structural Engineer, 92(4), pp36-40 However, in service, the degree of deformation
caused by the connection must be assessed for
adequacy to ensure the slab does not become
unduly strained.
2.9 Limit state philosophy and initial wall Serviceability performance criteria can often govern
sizing shear wall design, whether it be movement,
acceleration, cracking, another characteristic, or a
combination thereof that dominates. Both deflection
Analysis at the ultimate limit state (ULS) may consider and dynamic behaviour are related to stiffness which
only those components of the structure that are is in turn dependent on geometry and material elastic
essential to the principal load path. Meanwhile, moduli. Height to length (L/b) ratios are a useful
analysis at the serviceability limit state (SLS) should starting point for sizing walls and/or choosing viable
consider all components of the structure that may systems. Table 2.1 lists guide values for a number of
undergo elastic deformation. materials and constructions.
Wall Pinned strut
Slabs
Real structure SLS model ULS model
Figure 2.15 Ultimate and serviceability models of a monolithic wall and slab system

2.10 Design overview
Table 2.1 Typical height to length ratios for common

materials
Material and construction Typical L/b ratios
Reinforced concrete – in situ 7 to 10
Reinforced concrete – precast 5 to 8
Masonry – unreinforced 1.5 to 2.5
Timber – cross laminated 3 to 4
Timber – sheathed panels 1 to 2
Steel plate diaphragms 7 to 10
where:
L is the overall height of the wall.
b is the overall length of the wall on plan.
Notes
a These are guide ratios only; they are not rules and design
solutions may fall outside the ranges given.
b Ratios are derived for wall sections without dominant
openings (i.e. where the second moment of area, I, is
proportional to L4 and vertical shear deformation is not
exceptional).
c The performance of any specific structure at any nominated
ratio will depend on the loading which is usually a function
of the tributary load area and/or building mass.
2.10 References
2.1 Institution of Structural Engineers. Stability of buildings.

Parts 1 and 2: General considerations and framed
bracing. London: IStructE, 2014
2.2 Corus. Bi-Steel design and construction guide.

Scunthorpe: British Steel, 1999
2.3 Gough, V. ‘Fast steel cores’. New Steel Construction,

14(5), May 2006
2.4 HM Government. The Building Regulations 2010.

Structure: Approved Document A (2004 edition
incorporating 2010 and 2013 amendments – for use
in England ). London: NBS, 2013
2.5 BS EN 1996-3: 2006: Eurocode 6 – Design of

masonry structures – Part 3: Simplified calculation
methods for unreinforced masonry structures. London:
BSI, 2006
2.6 BS EN 1992-1-1: 2004: Eurocode 2: Design of

concrete structures – Part 1-1: General rules and rules
for buildings. London: BSI, 2004
2.7 BS EN 1993-1-1: 2005: Eurocode 3: Design of steel

structures. Part 1-1: General rules and rules for
buildings. London: BSI, 2005
2.8 Institution of Structural Engineers. Manual for the

design of concrete building structures to Eurocode 2.

design of plain masonry in building structures to
Eurocode 6. London: IStructE, 2008

3 Requirements of walls
3.1 Introduction Requirements for, and the likelihood of, future layout
changes should be considered when planning
structural walls. While cellular construction is often
Together with floor slabs, walls are often integral efficient as designed, the widespread structure can
components in the form, environment and function of be prohibitive to future changes of use. This can lead
a building. In this way they differ from columns, to low whole-life efficiency if the inflexibility of the
beams and foundations that usually satisfy structural layout becomes the cause of premature
functions only. This chapter introduces a number of decommissioning or major structural remodelling
non-structural performance requirements. It focuses works. Hence it can be advantageous to design for
on the direct and/or indirect effects these have on the fewer structural walls especially where the walls are
structural design. only lightly stressed. Additional partitions can then be
added and removed as non-structural items.
Note that any remodelling of the structural walls will

3.2 Wall locations usually affect the horizontal stability systems and the
foundations in addition to walls that form part of the
remodelled load paths.
The placement of structural walls will always be linked
to the architecture and building function. Some
buildings present more opportunities for walls than
others; cellular single-occupancy dormitory buildings 3.3 Non-structural partitions and
are one extreme while open-plan offices, arenas and non-loadbearing panels
industrial buildings are at the other (Box 3.1).
Structural engineers should not take for granted that Partitions that are temporary are not suitable for use as
structural walls can be included in the façade structural stability elements unless the design
especially where large doorways are required. Natural specifically allows for the partitions to be removed. Non-
ventilation and/or daylighting requirements can dictate structural partitions should be considered as nothing
strict criteria for the geometry of windows, while the more than superimposed mass and a surface attracting
installation and maintenance strategy for significant lateral actions that transfer to the adjoining structure.
plant or operational equipment may affect the
permanence of parts of the façade. Walls that are discontinuous from one storey to the
next may be used as non-loadbearing shear infill
Similarly structural engineers should consider all panels within a frame (Chapter 8). They are, however,
special storeys that might break the continuity of the generally not suitable as vertical load carrying shear
vertical structure from one storey to the next. By walls unless they land upon suitable transfer structures
example ground floors, basement levels and any that transfer both vertical and horizontal loads to
internal plant floors will often have very different structural elements continuing below.
functional requirements to typical floors. These
requirements may prevent walls being continuous Any non-structural partitions or façade elements
from one floor to the next and heighten the risk of should be installed with connections that can
localised instability (Box 3.2). accommodate in-plane differential movement. This is
Box 3.1 Traditional cellular construction

Many cellular buildings including schools, hotels, prisons and student accommodation adopt a regular wall arrangement with
transverse ‘cross walls’ between rooms and longitudinal ‘spine walls’ flanking a central corridor. Spine walls are generally punctured
by doorways and service routes. Meanwhile the façade is often non-structural to permit unobstructed glazing, a proprietary façade
system, and/or doorways.
Spine wall
Cross wall
Non-structural façade
Doors and service
voids to corridor

3.4 Requirements of walls
Box 3.2 Local storey instability and alterations

A single storey may be the cause of an instability failure where it provides particularly little resistance, measured relative to the
action. This can occur in new structures but often comes about through ill-conceived structural alterations to existing buildings
(e.g. the removal of walls to create open-plan accommodation). Where alterations have a detrimental effect on the resistance, the
resulting storey is often dubbed a ‘soft storey’.
Despite the name, a soft storey is not necessarily the weakest storey in a structure. Rather, it is the part of the structure most
vulnerable to failure. This could be one of the strongest storeys (e.g. the ground floor storey which is subject to maximum shear
force).
Further information on structural alterations can be found in Part 1 of this series3.1.
Global sway failure Local member failure
to prevent unintentional load paths that could Whether a building has centralised or dispersed
otherwise lead to premature failure of one or more stairwells is often a function of emergency egress
panels. Removed from the load path, these non- criteria. These criteria, together with the functional
structural elements must not be relied upon to brief, inform the initial concept developed by the
provide any in-plane resistance or stiffness. However, architect. However, while the building layout is fluid,
they must be self-stable in all unrestrained axes and structural engineers can contribute structural drivers
of adequate strength and stiffness to withstand direct into the mix, collaborating with the wider design team
actions (e.g. wind pressure) without loss of integrity. to guide the architect through the early space
planning.
Important structural drivers may include:

3.4 Cores – symmetry and/or eccentricities
– linear stiffness
– torsional stiffness
Most multi-storey buildings contain at least one core – movement joints
housing communal services including stairs, lifts, – foundation requirements and avoidance of transfers
toilets and service risers. These often have walls
continuous up the height of the building that are most The importance of these is discussed in Part 13.1.
likely suitable to be used as primary components of a
building’s structure (Figure 3.1). When a building’s form can accommodate or favours
a non-structural core, the core becomes nothing more
than a series of slab penetrations, superimposed
actions and non-structural partitions (supported either
at intervals through the superstructure or cantilevering
from the foundation). Non-structural cores can be
advantageous when the structural design is
progressing ahead of other disciplines, or when the
building services are significant and likely to change
markedly during the life of the building (e.g. in
industrial buildings). Otherwise, precast non-structural
lift (elevator) shafts that are independent of the
structure are increasingly common in low-rise steel
framed structures (Figure 3.2).
3.5 Vertical access and transportation
Counter to the scenario shown in Fig. 3.2, it is

convenient in many structural systems to surround
stair and lift shafts with coupled structural shear walls
in the form of a flanged or tubular core (Fig. 2.2). This
minimises the need for walls or bracing elsewhere
throughout the structure, while it also increases the
Figure 3.1 Structural core walls under construction for a efficiency by which fire compartmentalisation can be
400m tall tower achieved (if required).

Requirements of walls 3.5
Where the walls are working hard (typically in taller

buildings) there may be no alternative but to use all
the available enclosure as structural elements. In this
case, care is needed to make the wall spanning the
doorways thick enough to accommodate the stress
concentrations around the openings.
Irrespective of whether the wall containing the

doorways is structural or not, it is recommended
that doorways are not located immediately
adjacent to a structural return wall (Figure 3.3). It is
better to have an adequate offset from the flanking
wall to:
– provide support to the lintel beam
– provide rotational stiffness at the lintel to wall
junction
– decrease the minor axis slenderness of the return
wall
It should be noted that the decision on the placement

of doors is typically at the behest of the architect and
not the structural engineer. The engineer should make
recommendations to the architect at the earliest
appropriate time.
Note Lifts
The steel frame is independently stabilised, in this instance Specific to lifts, overruns are often necessary at both
with framed bracing. the top and bottom and often define the vertical
extents of the walls. These may have an impact on
Figure 3.2 Non-structural precast lift shaft providing support
the design actions, the global slenderness, and the
to lift guide rails only
interfaces with other structural components (the roof,
substructure or transfers).
Both lift and stair shafts require access openings, Lift systems will often require smooth, accurately
usually regular in size and location at every storey. formed internal walls, constructed to tight tolerances
The result is often a ladder effect, with planar sections and without projections (e.g. surface-mounted
of wall punctuated in a regular arrangement, as connections). BS 5655-63.2, by example,
shown in Figs 2.6b, 2.6c and 3.1. recommends that shafts should be sized with a
50mm tolerance zone provided to all four sides for
Penetrations the size of doorways and stacked one lifts up to 20 storeys, with an additional 1mm per
above another will often have significant impact on storey up to a maximum of 100mm for lifts over
the behaviour of a wall, bringing into play the coupling 70 storeys.
effects introduced in Section 2.6. Where possible
there may be advantages in making the penetrated Although critical, internal shaft dimensions often
walls non-structural and discounting them from the remain in abeyance through much of the design
analysis. Among other benefits, this allows the development until the preferred lift manufacturer is
structural design to proceed in advance of precise contracted (often past the point at which the
locations and sizes of the doors being known. It also architectural space planning becomes fixed). The
allows for delayed construction and the future structural engineer should bear this in mind, stating
removal of the non-structural element, possibly any assumptions on preliminary drawings, and
needed for the installation and replacement of a lift possibly allowing tolerance when initially sizing the
car. A drawback however is the loss of torsional walls to ensure adequate thickness in the event that
rigidity that comes with an open section (Chapter 4). the internal shaft dimensions increase.
Unrestrained Wall end restrained Short wall return to

wall end by return provide restraint
Lintel
Doorway continuous
with lintel into wall
Door with
Wall lintel Non-structural
thickness surround to
often doorway
governed
by lintel
Poor Improved Max. flexibility
Figure 3.3 Positioning large penetrations in walls

3.6 Requirements of walls
Box 3.3 Stair core buildability at Bond Street Station, London

The images below show the construction and permanent conditions of the over-site development commissioned as part of the
Bond Street Underground Station upgrade in Central London.
Temporary Permanent
(construction) stage stage
Here, stair flights and sections of floor slabs were omitted throughout the duration of the superstructure construction. This was to
allow for a temporary crane hall within the stair core footprint that provided construction access to works on the station tunnels below.
The omission of the stairs meant that the planar shear wall, which is situated on the perimeter of the building adjacent the site
boundary, was unrestrained in the temporary condition. Temporary bracing could not be provided as it would have impeded the
crane hall. Instead the wall had to be designed for the temporary unrestrained condition. Temporary loads included all the self-
weight of the complete superstructure plus concentrated crane loads from construction equipment supported off the wall.
Stairs 3.6 Service risers and distribution

More so with stairs than lifts it is important that the
structural design engineer understands the sequence
of construction, the dependency of the walls on the Service risers are shown on architect’s drawings as
stairs as means of restraint, and the robustness box-outs, cupboards or small rooms, typically with
requirements of any joints3.3. Each can influence the door or hatch access from each floor. Traditionally
fixings embedded in, or post-fixed to, the walls and they were quite small. However this has changed as
the nature, alignment and magnitude of the reaction buildings have become more heavily serviced and
forces. safe installation and maintenance provisions have
become mandatory.
Short-term buildability logistics should also be
thought through. Where stairs are precast or Risers are often clustered around stair and lift cores:
prefabricated they may be craned in from above, or – to minimise impact on a building’s plan layout
need to be manoeuvred through adequately sized – to be close to the main circulation routes
openings in the walls. Other in situ stairs may require – to be accessible from communal areas
formwork or temporary work platforms. Finally,
where the stair flight provides permanent lateral While functionally ideal, this is not without challenges
restraint to the walls, temporary bracing may be to the structural engineer.
needed prior to the stair being completed.
Most risers require large voids in the floor
Without temporary bracing, walls may need to be diaphragm to fulfil their function and allow vertical
designed for a governing temporary condition service distribution through the height of the
(Box 3.3). building. Voids in the walls (additional to any doors)
are also required to allow horizontal distribution of
the services out to the floors (mostly into false floor
and ceiling voids). These wall voids often limit the
Non-structural opportunity to use riser walls as structural elements.
partitions to risers However, the voids in the floor diaphragms which
Penetrations
to partitions are close but external to a structural core can be
Structural core equally troublesome, impacting on the load path
between the wider floor diaphragm and the
structural walls (Figure 3.4).
Horizontal service distribution routes to each storey

generally follow circulation corridors above ceilings and
below floors. They invariably track into the cores,
passing through, under or over critical lintel beams.
Any services passing across structural wall lines
should be thoroughly coordinated by the different
Penetrations in engineering disciplines. This is best conducted by first
slab diaphragm setting clear parameters early in the concept stage to
define the zones that can and cannot be assumed by
Figure 3.4 Penetrations to service risers the different engineering disciplines. Zones overcome

Requirements of walls 3.7
the need to know exact penetration sizes. However, it achieved by meeting minimum cover and concrete
is recommended that structural engineers do class criteria; for steel it may be via an insulating
understand the nature of absolute penetration barrier layer (e.g. intumescent paint or a solid box-
dimensions to better identify and collaboratively resolve out); meanwhile for timber, a sacrificial char layer or
conflicts. Insulation, isolation and minimum separation an insulating barrier are common.
distances3.4, together with practical installation
tolerances3.5, can add significantly to the dimensions Finally a need for moisture resistance may be defined
of ducts, pipes and cable trays typically cited on the either by the serviceability requirements of the
service engineer’s drawings. accommodation or by any degrading effects of
moisture on the material of the construction (e.g.
Further reading: vertical transportation and service rusting of steel, leaching of calcium from mortar,
integration rotting of timber, etc.).
The following texts are recommended sources of further Further reading: moisture and water resistant wall
guidance on service integration: construction
– McKenna, P.D. and Lawson, R.M. Design of steel framed
buildings for service integration: interfaces. SCI Publication The following texts are recommended sources of further
166. Ascot: SCI, 1997 guidance on moisture and water resistant wall construction:
– Co-Construct. Services integration with concrete buildings – Institution of Structural Engineers. Design and construction
– Guidance for a defect-free interface. IEP 3/2004. of deep basements including cut-and-cover structures.
Bracknell: BSRIA, 2004 London: IStructE, 2004
– Co-Construct. Services coordination with structural beams. – Mott MacDonald Special Services Division. Water-resisting
IEP 2/2003. Bracknell: BSRIA, 2003 basement construction – a guide – safeguarding new and
existing basements against water and dampness. CIRIA
Report 139. London: CIRIA, 1995
3.7 Insulation and compartmentalisation

3.8 References
Requirements for thermal insulation, acoustic
insulation, moisture resistance and fire
compartmentalisation often impose performance 3.1 Institution of Structural Engineers. Stability of buildings.
criteria on walls. Parts 1 and 2: General philosophy and framed bracing.
Each of these is typically dependent on the complete
wall construction (including non-structural parts such 3.2 BS 5655-6: 2011: Lifts and service lifts – Part 6:
as insulation and protective coatings) and can benefit Code of practice for the selection, installation and
from cavity or sandwich wall systems. The peripheral location of new lifts. London: BSI, 2011
detailing can also be influential: voids, edge details
and anything that detrimentally ‘bridges’ the 3.3 Billington, C. ‘Achieving robustness of precast concrete
favourable constituents of the wall will often govern. stairs using proprietary cast-in inserts’. The Structural
Engineer, 92(2), February 2014, pp34-36
Thermal insulation is usually defined by a system’s
U-value. Design values are normally established by a 3.4 Ministry of Defence. Space requirements for plant
mechanical or building environment engineer and access, operation and maintenance. Defence Works
should satisfy legislated requirements (e.g. those Functional Standard Design & Maintenance Guide 08.
criteria set by the English Building Regulations Part L3.6). London: HMSO, 1996
Acoustic insulation is similarly defined by a sound 3.5 BS 8313: 1997: Code of practice for accommodation
reduction index (SRI) which is a logarithmic function of building services in ducts. London: BSI, 1997
of the construction’s transmission coefficient (the ratio
of incident to transmitted sound energy)3.7. The 3.6 HM Government. The Building Regulations 2010.
design value is normally established by an architect, Approved Document L1A: Conservation of fuel and
possibly in consultation with an acoustics specialist, power in new dwellings; Approved Document L2A:
and may be based on occupancy-specific guidance Conservation of fuel and power in new buildings other
such as that published in Building Bulletin 933.8. than dwellings [2013 edition for use in England].
London: NBS, 2014; Approved Document L1B:
Meanwhile, fire performance is defined by a time- Conservation of fuel and power in existing dwellings;
based rating that concerns each of: Approved Document L2B: Conservation of fuel and
– the structural adequacy (i.e. the ability to fulfil a power in existing buildings other than dwellings (2010
structural function) edition incorporating further 2010 and 2011
– the integrity of the system (i.e. the ability to amendments). London: NBS, 2011
compartmentalise a space without breach)
– the insulation of the system (i.e. the ability to 3.7 BS 8233: 2014: Guidance on sound insulation and
contain heat) noise reduction for buildings. London: BSI, 2014
The fire rating is usually established by the architect, 3.8 Department for Education and Skills. Acoustic design for
often in consultation with a fire engineer, the client’s schools: a design guide. Building Bulletin 93. London:
insurance manager, mechanical engineers (involved in The Stationery Office, 2004 [Section 1 updated by
the design of a sprinkler system), and the local fire Education Funding Agency. Acoustic performance
authority. It will often have a direct impact on the standards for the priority schools building programme.
structural design, the detail of which varies by 2012. Available at: https://www.education.gov.uk/
material. Fire rating for concrete components may be publications/standard/publicationDetail/Page1/BB93]

4 Elastic theory of thin-walled sections
4.1 Introduction Longitudinal shear due to flexure is one example of

complementary shear. Box 4.2 gives the general
equation for an ideal 1-dimensional element; all
Often subject to both linear and torsion actions, shear elements that maintain a plane section when subject
wall systems experience various stress components. to flexure must be able to transfer a stress derived in
How these interact and cause compatible accordance with Equation 4.1 through any
deformations (strains) is critical to the behaviour and longitudinal plane.
must be captured in the analysis.
A similar longitudinal shear stress acts in all walls
A wall system can be considered analytically as a subject to flexure. However the magnitude of the
collection of 2-dimensional elements. However, most stress deviates from that given in Equation 4.1 if
wall systems can also be approximated as thin- plane sections deform. This deviation is critical to the
walled 1-dimensional elements that are regular along analysis of coupled wall systems. Anywhere a lintel
their vertical (longitudinal) axes. In this way classical beam or joint is less stiff than the equivalent solid
beam theories can be applied. section, the shear stress across the weakened
section will decrease while the shear stresses in the
monolithic sections will increase (Figs 2.6 and 2.7).
Overall this ensures shear equilibrium but equates to
4.2 Complementary shear a loss of gross section stiffness.
As thin-walled systems, it is acceptable to neglect

through-thickness stresses in most walls. However, 4.3 Torsion
complementary stresses must be considered in-
plane; anywhere a shear is applied (from either a
linear or torsional action), an equal shear stress must Box 4.3 shows two mechanisms by which torsion is
occur on all four through-thickness faces of a resisted within elastic sections. The first is with a
infinitesimally small cut block. This is to maintain linear stress profile that acts through the thickness
equilibrium (Box 4.1). of all parts of the cross-section (Box 4.3a). This is
referred to as St Venant’s resistance and it acts
irrespective of whether the section is open or
Box 4.1 2-dimensional complementary shear of a closed. The second mechanism has a uniform
thin-walled structure stress profile and acts only within closed sections
Consider an infinitesimally small block subject to uniform (Box 4.3b). The second mechanism generates a
shear stress t1: resistance which is proportional to the square of the
area enclosed by the closed section and tends to
t dominate over St Venant’s resistance where
applicable.
4
4.4 Warp and warp restraint

1
a 3 Complementary longitudinal shear stresses exist with
both resistance mechanisms introduced in Section
4.3. Where unrestrained, these stress profiles cause
warp: a minimum-energy longitudinal deformation
that distorts plane sections.
2
Open sections that are subject to torsion and are free
b to warp exhibit uniform deformation along their
length. However, when warp is restrained with one or
For vertical equilibrium: more sections held plane, the deformation cannot be
t1at ¼ t3at ! t1 ¼ t3 uniform and axial stresses develop proportional to the
restraint stiffness.
For moment equilibrium:
b b a a
t1 at þ t3 at ¼ t2 bt þ t4 bt ! t1 þ t3 ¼ t2 þ t4
2 2 2 2
For horizontal equilibrium:

t2bt ¼ t4bt ! t2 ¼ t4
Hence:
t1 ¼ t2 ¼ t3 ¼ t4

Elastic theory of thin-walled sections 4.4
Box 4.2 Transverse and longitudinal complementary shears resulting from flexure
Vz
Cross-section
area A
Neutral axis of gross
z
Vz cross-section
Longitudinal
complementary
shear
Myy Vz Shear stress
Vertical normal
stress
For elements that deform such that plane sections remain plane, the longitudinal shear stress tx can be derived by evaluating
Equation 4.1:
Vz Az
t2 ¼ . . .Eqn 4.1
Iyy t
where:
Vz is the applied transverse shear force
Az is the first moment of area of the cut section measured about the neutral axis of the gross section
Iyy is the second moment of area of the gross section
t is the thickness of the cut section
Both Az and Iyy should be measured about an axis orientated perpendicular to the shear force.
Box 4.3 Shear stresses resulting from torsion for (a) an open section and (b) a closed section
Enclosed area Ω
to centre of walls
Cross-section:
(a) Open section (b) Closed section
Through-thickness
stress:
St. Venant’s St. Venant’s + Uniform

stress stress stress
St. Venant’s stress
The shear stress at the extreme fibres tmax and the torsion constant J are given by Equations 4.2 and 4.3 respectively:
3T
tmax ¼ . . .Eqn 4.2
bt 2
1Xn
J¼ bt 3 . . .Eqn 4.3
3

4.5 Elastic theory of thin-walled sections
Box 4.3 Continued

Note that Equations 4.2 and 4.3 apply where b t and for small rotations where the global flexural stresses due to rotational
deformations (Figure 4.1) are small.
Uniform stress of a closed section

The uniform component of shear stress t and the corresponding torsion constant J are given by Equations 4.4 and 4.5
respectively:
T
t¼ . . .Eqn 4.4
2T V
4V2
J ¼ð . . .Eqn 4.5
1
ds
t
where:
T is the applied torque
t is the section thickness
b is the cross-section length of the elements that make up the section
V
Ð is the area enclosed by the centrelines of the walls forming a closed section
ds is the line integral of elements within a section
Thus warp restraint is a third mechanism by which considered in the walls and also as actions on the
torsion can be resisted. The resulting stress profile is supporting structure (Figure 4.3).
additional to those shown in Box 4.3 and is
commonly known as ‘warping stress’. It is associated Longitudinal stresses due to restrained warp are
with the local in-plane bending resistance of the most pronounced in open sections and can be of
individual wall panels as the section twists similar order to stresses resulting from flexure. Closed
(Figure 4.1). sections have significantly larger torsional resistance
(Box 4.3) rendering the effect of restrained warp
Importantly, warp restraint increases the rotational usually negligible, while circular sections are the only
stiffness of a section but in doing this it adds to the sections where restrained warp causes zero
longitudinal stresses in the wall. Stresses tend to additional stress (as circular sections do not warp).
peak close to the restraint and diminish moving
towards unrestrained ends (Figure 4.2). Shear walls
extending to foundations are generally considered
fully restrained at the base and free to warp at the 4.5 Lintel beams in sections subject to
top. Other shear walls (e.g. those terminating on torsion
transfer beams, or infill panels situated within a
framed system) may have less rigid warp restraint as
a function of the supporting element’s stiffness. In When an open channel twists, the section naturally
either instance, the longitudinal stresses should be wants to warp causing the wall ends to displace
longitudinally in opposite directions. This longitudinal
unrestrained movement is vertical in walls
(Figure 4.4).
Lintel beams locally close the section and partially

restrain this movement forcing compatibility of
Flanges bend stresses and strains across the opening. In doing
as section twists this, they resist warp and can add significant torsional
stiffness. The lintel beam experiences significant in-
plane (vertical) shear and bending; stresses that
induce a complementary horizontal shear. This
circulates through the closed section causing an
increased torsion constant locally in the section with
the lintel (Box 4.3). In turn, this enhanced stiffness
reduces the tendency for rotation and warp,
increasing the overall torsional stiffness of the whole
system.
The vertical stresses in walls resulting from torsion

will vary significantly with the stiffness of the lintel
beam. Even relatively shallow lintel beams can
result in a considerable increase in overall torsional
stiffness, resulting in an equally considerable drop
in vertical warping stress in the walls and acting on
the restraining foundations or transfer structures.
Figure 4.1 In-plane bending of individual wall panels due to However, engineers must be mindful that any
twist weaknesses in the lintel (e.g. those resulting from

Elastic theory of thin-walled sections 4.6
Original Free Approaches constant shear

shape
Height
Approaches constant
rate of rotation
Section deforms
through full height
with constant shear
and rate of rotation
iont
Rota
r
Shea
Restrained
Shear rotation
Free to warp Restrained warp
Figure 4.2 Free and restrained warp of an open channel section
service penetrations) can have a significant – The centroid is significant for axial (vertical) forces;
detrimental effect. This sensitivity emphasises the an axial force through the centroid results in pure
importance of developing a coordinated scheme axial strain
that makes adequate provision for service – The shear centre is significant for transverse
distribution voids and doorways alike (horizontal) forces; a horizontal force through the
(Chapter 3). shear centre results in pure shear (without torsion)
It should be noted that torsion and warp relate to the

vertical axis that runs through the shear centre
4.6 Centroid and shear centre equivalent to how flexural deformations relate to a
section’s neutral axes.
Two centres govern the behaviour of a section: the Both centres lie on axes of symmetry and are only
centroid and the shear centre: coincident when a section is doubly-symmetrical. The
Original
shape
Wall
Restrained warp
stresses across
interface
Open section Closed section

Restraining structure
(substructure or transfer)
Figure 4.4 Warp deformations across openings with and
Figure 4.3 Warp restraint reactions (exploded view) without lintel beams

4.6 Elastic theory of thin-walled sections
shear centre can be derived by considering the

Box 4.4 Shear centre of a regular channel
equilibrium of complementary shear stresses in a
section
section subject to a transverse shear. A simplified
statically-determinant example for an open channel The following is a simplified derivation for the shear centre of
section is shown in Box 4.4. In practice the shear a regular channel section with constant thickness t and equal
centre of a flanged or tubular wall system will be flange lengths B. It simplifies the minor axis inertia of the
dependent on the stiffness of the coupling within the channel to that of the two flanges only. It is accurate when
section. The shear centre may also vary through the the minor axis inertia of the web, measured about the
height of the building if the wall section changes. section’s neutral axis, is negligible.
F
Further reading: elastic theory
The following texts are recommended sources of further A B
guidance on elastic theory: QF
– Millais, M. Building structures: from concepts to design.
2nd ed. Abingdon: Spon Press, 2005 t
– ACI 445.1R-12: Report on torsion in structural concrete. D/2
Farmington Hills, MI: ACI, 2012
D/2 Qw
QF
Shear in flanges:
ð
FB
from t dA ! QF ¼
2D
Shear in web:
from vertical equilibrium Qw ¼ F
Moments about shear centre:
FB B
D FA ¼ 0 ! A ¼
2D 2

5 Modelling and analysis
5.1 Introduction 5.2 Modelling simplifications
A model is a tool for analysis being a representation of The analysis of shear walls is a key part of a project’s
a structure, its physical behaviour and the forces and design development, from concept right through to
environmental conditions to which it is subject. All detailed production information and checking. The
models are a simplification in one way or another: rigour with which a structure is modelled should
whether they only represent part of the structure, part reflect the design stage, certainty and risk. Figure 5.1
of the behaviour, or part of the exposure conditions. In presents possible model options; these options make
most cases, the simplest possible model to achieve reference to element dimensions defined in Box 5.2.
the goal should be used. This will usually be the
easiest model to interrogate, update and (perhaps
most importantly) check.
The method of analysis and modelling should be Single 1-D element

actively planned by the lead structural engineer. approximation
Decisions should be recorded to assist both
computing power
Ease of interpretation
colleagues and checkers. This is best done by way of
Increasing
a documented modelling plan contained within the
Simplifications
calculations and model files. 1-D element grillage

approximation
In planning a model, it is important the engineer
considers all parameters that could be critical
(Box 5.1). Knowing how sections behave when loaded
2-D finite element
and how behaviours are represented by various
approximation
model techniques is fundamental; Chapter 4 provides
important background reading relevant to the discussion Limit on
that follows on 1-dimensional element models. model accuracy
Real-world
Box 5.1 Modelling parameters structure testing
The following is a list of common modelling parameters and
associated considerations that can influence the accuracy of Figure 5.1 Modelling options and complexity
a model:
– Applied actions: type, location, relationship to original and
deformed geometry, variability
Box 5.2 Modelling dimensions
– Boundary conditions: location, stiffness/fixity, strength,
variability Wall sections are typically modelled using 1- or 2-dimensional
– Imperfections of the structure: material moduli, densities, elements. These element dimensions have no relation to the
setting out errors, out-of-plumb errors, locked in assembly/ dimensions of the model as a whole:
casting stresses – 1-dimensional elements connect two node points defining
– Construction sequence: staged loading, temporary conditions a line. They usually exhibit constant properties along their
– Material behaviour: elastic, plastic, brittle, ductile length and closely resemble linear elements within a frame
– Non-linear behaviour: material non-linearity (including structure
shrinkage, creep and cracking), action non-linearity – 2-dimensional elements connect three, four, six or eight
(including PD effects), reaction non-linearity, element points arranged as either a planar triangle or quadrilateral.
buckling, global buckling, plastic hinge failure They exhibit properties in two orthogonal axes and can
– Model compatibility: are the parameters compatible across closely resemble both walls and floor slabs.
all models being used to analyse a structure?
There is very little dispute that computer analysis can be

far more powerful than more traditional methods.
However, while the following sections refer mostly to
computer methods, there is rarely a project that cannot 1-D element
start with simple hand calculations as part of the
conceptual design. These should not be
underestimated. Not only do they provide focus leading
to coherent concepts but they also provide information
against which more detailed computer outputs can be
appraised. Parts of the discussion herein can be
applied to hand calculations just as it can to computer
methods.
Note that differences between computer output and

results obtained via hand calculations must always be 2-D element
examined and justified.

5.3 Modelling and analysis
Real structure (a) (b) (c)
Figure 5.2 1-dimensional element models of a structure: (a) grillage model, (b) single element per wall system model, (c) single element amalgamated model
It is worth noting that it is often advantageous in Each simplification requires that the properties of the
more complex structures to use more than one of the more complicated model are known. Thus the
modelling options through the different stages of the simplifications are rarely useful for design
design development and/or in the checking. Testing development but are valued techniques for checking.
different models against one another is often far more
efficient than directly checking a single model for For design development, it is most likely the models
errors and can unearth fundamental shortcomings are used in the reverse of the order shown in Fig. 5.2.
that would not otherwise be apparent. The single element amalgamation shown in Fig. 5.2(c)
is useful as a concept, if not formally modelled, to
Engineers must also not confuse nor automatically gain a handle on the gross requirements of the
associate modelling complexity and/or simplifications stability system in the early concept development
with model accuracy. It is possible to have a highly stage. The representation shown in Fig. 5.2(b) then
accurate simple model, just as it is possible to have becomes more useful once the stability systems have
an inaccurate complicated model. The accuracy of a been located. It allows the structural engineer to
model ultimately comes down to the cumulative quickly experiment with the properties of the
impact of the simplifications, not the simplifications in individual systems and evaluate the influence of minor
themselves. Modelling elements and joints as adjustments. Finally, the grillage model shown in
perfectly elastic is a common simplification; uniform Fig. 5.2(a) is the most detailed 1-dimensional element
actions and perfectly rigid boundary conditions are model and is best suited to element verification within
others. the detailed design. Techniques to model an accurate
grillage are discussed in more detail in the next
section.
5.3 Modelling vertical stability structures
5.3.1 1-dimensional element models
1-dimensional element modelling originates from

classical beam theory and was favoured with early
software. Today these models are valued for hand
calculation methods and are well suited to fast
conceptual work. They remain a fall-back when
2-dimensional finite element (FE) modelling is not
available and provide a means of checking more
complicated results. A significant advantage is that
the output is readily intelligible and can be used for
element design with very little post-processing.
Individual wall systems can be represented with

either a grillage or single element representation
(Figures 5.2(a) and 5.2(b)). Models can be simplified
further with a single element to represent the
amalgamation of all stability systems across a
building (Fig. 5.2(c)).
With reference to Fig. 5.2, methods exist to

determine equivalent properties for simplified models:
– Box 5.3 outlines a method for approximating the
equivalent single-element properties of a coupled
wall (i.e. changing from Fig. 5.2(a) to 5.2(b)).
– Box 5.4 outlines how properties of elements shown
in Fig. 5.2(b) can then be combined to determine
the amalgamated properties for the single element
shown in Fig. 5.2(c).

Modelling and analysis 5.3
Box 5.3 Single element representation of a pierced wall

It should be noted that the following representation is appropriate for modelling global behaviour. It fails to capture the local stress
concentrations that would be apparent in and around the lintel beams and also the effect of any localised buckling in the wall ends
adjacent to the openings.
Torsional stiffness
Un-stressed
Equivalent
shear diaphragm
Lintel beam
Deformed
δw
Thickness tw
δb
h Diaphragm of shear
stiffness GAw
Beam of flexural V
stiffness EI b and V
shear stiffness GA b
Lb Lb
True structure Representation
Thickness of the equivalent diaphragm:
VL b 3 VL
db ¼ þ b
12EI b GA b
VL b
dw ¼
GA w
A w ¼ ht w
2
1 L G 1
! ¼h b þ . . .Eqn 5.1
tw 12I b E Ab
|fflfflffl{zfflfflffl} |fflfflfflffl{zfflfflfflffl}
flexural shear
deformation deformation
of beam of beam
where:
Ab is the shear area of the lintel beam
E, G are the material moduli of the lintel beam
Ib is the second moment of inertia of the lintel beam
Lb is the lintel beam length
h is the height of the equivalent diaphragm
tw is the thickness of the equivalent diaphragm

Box 5.3 Continued

Torsion constant J:
1X 3 V2
J¼ bt þ . . .Eqn 5.2
3
|fflfflfflffl{zfflfflfflffl} L=t w
|fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflffl
ffl}
for real t w treal walls
walls only [ L /t w dominates
where:
b is the length of the perimeter walls excluding length Lb
V is the area enclosed by the perimeter walls, measured to the centre of the walls
Flexural stiffness
A conservative approximation of the flexural stiffness in both the major and minor axes can be approximated by considering the net
wall section taken from a cut through the wall at a level that bisects the largest wall penetrations. This will result in a section of
least stiffness.
Iyy
Izz
It should be noted that the centroid will be influenced by the presence of the penetrations.
Shear stiffness
for solid ‘webs’: for beams: combined:
Vw L V bL b V bL b3 VL
d¼ d¼ þ d¼
Un-stressed GA w GA b 12EI b GA
(neglecting flexural component)
where V ¼ Vw þ V b ,
Solid
‘web’
Lb L 3
L þ ðGA w Þ þ b
ðGA b Þ 12EI b
! ðGAÞ ¼ ... Eqn 5.3
Lb L 3
Deformed þ b
ðGA b Þ 12EI b
where:
Lintel beam
GA is the gross equivalent shear stiffness
GA w is the shear stiffness of the solid wall
L GAb is the shear stiffness of the lintel beam
EI b is the flexural stiffness of the lintel beam
L is the overall depth of the wall system
δ
δ
V
Lb
Assume all deformation of the
coupled wall occurs in the lintel beam
Box 5.4 Notes for amalgamating multiple linked structures into a single element
The following characteristics refer to the amalgamated element. Flexural stiffness I is quoted as a simplification to the combined
effect of flexure and shear. This can be changed throughout to a more general stiffness k.
Lower case subscripts i,uu and i,vv are used to denote the stiffnesses of the i’th element in its principal major and minor axes
respectively, and i,xx i,xy and i,yy are used for the stiffnesses of the i’th element about a common global coordinate system.
Upper case subscripts are used for the single amalgamated system: XX, XY and YY define stiffnesses in the global coordinate
system while UU and VV define stiffnesses in the amalgamated system’s principal axes.

Box 5.4 Continued

Cross-section area A This model should not be used for axial loads and the area of the section is largely
irrelevant.
Principal second moment of The principal inertias I i,uu, I i,vv of the individual contributing components must first be converted
area stiffness properties I UU, I VV to I i,xx, I i,yy, and I i,xy about a common global axis set. These can then be summed in turn to
determine I XX, I YY and I XY for the amalgamated element. Finally, I UU and I VV can be calculated.
v y
y u
x
y
Single element stiffness:

IXX, IYY, IXY
Notes
a I i,uu and I i,vv of the components must not be summed where they do not share a common
axis set.
b I i,uu, and I i,vv can be related to I i,xx, I i,yy and I i,xy, and I XX, I YY and I XY to I UU and I VV using
Mohr’s circle.
Torsional stiffness J The amalgamated torsional stiffness is the sum of: P

– The local torsional stiffness of the contributing components J i
– The product of the translational stiffness of the contributing components and the eccentricity
of this stiffness from the global centre of stiffness
y xi
xs xs
yi
ys ys
Single element
torsional stiffness J
x
Arbitrary origin P P
J¼ [J i ] þ [I i,xx(yi ys )2 þ I i,yy(x i xs )2 2I i,xy(x i x s )( y i y s )] . . .Eqn 5.4
Position The single element should be positioned at the equivalent shear centre of the systems (xs, ys),
otherwise known as the global centre of stiffness.
The coordinates can be determined relative to a reference origin by evaluating a torsion about
the origin that results in a unit linear displacement in each of the x and y axes:
P
Tx ¼ [I i,xy x i I i,xx y i ] . . .Eqn 5.5
P
Ty ¼ [I i,yy x i I i,xy y i ] . . .Eqn 5.6
x s ¼ (I XX Ty I XY T x ) / (I XX I YY I XY 2 ) . . .Eqn 5.7
y s ¼ (I YY T x þ I XY Ty ) / (I XX I YY I XY 2 ) . . .Eqn 5.8
Limitations of 1-dimensional elements

Further reading: Critical to the performance of the model shown in
Fig. 5.2b is the software’s representations of the
An understanding of Mohr’s circle is required for the stiffness
restrained warping stiffness and the shear centre.
components. For further guidance refer to:
– Hulse, R. and Cain, J.A. Structural mechanics: worked
Few software packages apply a restrained warping
examples. Basingstoke: Palgrave Macmillan, 2009
stiffness to 1-dimensional elements. This is critical for
– Parry, R.H.G. Mohr circles, stress paths, and geotechnics.
open sections where restrained warping stiffness is
2nd ed. London: Spon, 2004
often several orders of magnitude greater than pure

St Venant’s torsion stiffness. Failure to acknowledge 5.3.2 Grillage models

the restrained warping stiffness in these sections can
render a model overly flexible. More importantly, it The grillage method shown in Fig. 5.2a uses vertical
may dramatically underestimate the element stresses and horizontal elements to represent components
and reactions on supporting structures. It is for these within a wall system. Figure 5.6 shows a close up of
reasons that Fig. 5.2b shows a two-column grillage a closed core section.
for the channel section. Each column of elements
represents half the gross cross-section of the wall (in A vertical tower of elements represents each solid
this case, each is an ‘L’ section) and collectively the planar wall section. These should be centrally located
flexural stiffness of these elements, together with and of dimensions and orientation true to the solid
the shear resistance of the horizontal elements and section of wall they represent. Horizontal elements
the two support nodes, go some way to replicate the trace the path of the wall system’s cross-section,
warp restraint. This approach is recommended for all connecting the vertical elements and joining at the
non-planar open sections when warping stiffness is corners. These elements reproduce the shear
not accounted for in the software. stiffness between wall sections and ensure geometric
compatibility is maintained during deformation. Not
With regard to the shear centre, many software being wholly representative of the continuum of the
packages incorrectly assume it is coincident with wall, their properties must be modified somewhat
the centroid irrespective of cross-section (as set out in Box 5.5).
geometry. This assumption is again critical to open
sections when represented by single elements. Lintel elements are an exception and should not
Loads in the real world are invariably applied to adopt the modified properties of other horizontal
the physical structure (not the academic shear elements. These should remain true to the real
centre) and failure to acknowledge an offset to the geometrical properties of the structure.
shear centre can dramatically reduce the torsion
recorded. Modelling the core so that it is centred The overall length of lintel beams is also critical.
at the shear centre can improve the accuracy of Modelling lintel beams to the centrelines of the walls
the model provided the lateral forces are applied will underestimate the system stiffness, while
correctly through the true core alignment (i.e. at an assuming full fixity at the face of the opening will be
offset to the core elements as modelled an overestimate. Rather, the position of full fixity
(Figure 5.3)). should depend on the construction. In monolithic
construction (i.e. structures without joints) full fixity
With this approach, and where the core may be assumed at 0.6h past the face of the
cross-section changes with height, the modelled opening, where h is the effective depth of the
core elements may need to move on plan to coincide coupling beam5.1 (Figure 5.7). This is reasonably
correctly with the changing shear centre. Two layers accurate in most cases, although it becomes
of rigid links, spaced with rigid stub columns, can be conservative where the beam is relatively deep
used to best maintain continuity in such (Box 5.6). At no point should the fixity be beyond the
circumstances (Figure 5.4). It should be noted that centre of the adjacent vertical wall panels.
this detail is only needed where continuity of each of
the wall’s rotational and translational deformation is It is typical to model vertical elements spanning
needed across the step. This is the case for changes between nodes located at floor levels, allowing the
of cross-section within the height of a wall. However, horizontal elements, including lintel beams, to be at
in the scenario where a wall is simply supported on an storey intervals. This is reasonably accurate in all but
offset transfer system, a single layer of rigid links is short stocky structures (Section 2.2), or open flanged
sufficient to ensure compatible translational sections subject to large torsion. In both instances,
deformations only (Figure 5.5). shear deformation dominates and can be better
Deformed shape True position

of structure Offset modelled
Correct element
shear
centre
Load
through
centroid
Rigid link element

centroid to load
a a

Rotation + translation Translation only Approximate rotation + translation
Figure 5.3 Correcting for shear centre errors in 1-dimensional element models

Wall section 2
Floor elements
or rigid links
Section 2
~1/10th Flexurally rigid

storey stub column
height
Auxiliary rigid links

between nodes
Section 1
Wall section 1
Viewed end on
Shear centre
axes
Structure Model
Figure 5.4 Linking offset 1-dimensional element where rotational continuity is needed
Wall
Wall
Transfer beam
Rigid links
Transfer beam
into page
Structure Model Viewed end on
Figure 5.5 Linking offset 1-dimensional element where a pinned joint is appropriate
Box 5.5 Horizontal element property modifiers

The following apply to horizontal elements other than lintel
beams in a grillage model:
(1) The material property and section thickness should be as
per the vertical elements
(2) The section depths should be equal to the vertical Nodes at all
spacing of the elements intersections
(3) The major axis (in-plane) flexural stiffness should be
increased to minimise flexural deformation
(4) The torsional stiffness should be decreased so as not to
relieve vertical elements of out-of-plane bending actions Horizontal
(5) The mass should be set to zero so as not to double count ‘modified’
gravity loads and/or affect the natural frequency elements
(shown red)
Note that the torsional stiffness of the vertical elements
should also be decreased so as not to relieve the horizontal
elements of out-of-plane bending actions.
Initial modification factors of the order of 103 are
recommended for each of (3) and (4) and the torsional
stiffness of the vertical elements as listed. The sensitivity of Vertical elements
the model with respect to these factors should, however, be Elements with with properties of
reviewed in the software being used. Extremely large or small properties of lintel wall sections
factors, as well as a 0% factor, can cause numerical errors in
the analysis engine depending on the algorithm and
numerical rounding.
Figure 5.6 1-dimensional elements used in a grillage configuration

Box 5.7 Modelling clarity, complexity and

b1/2 b2/2 verification
It is not uncommon for automated software to lead to
unnecessarily complex models developed without clear intent
or focus on specific output. Engineers should be wary of this
Modified as overly complex models can be a catalyst for oversights
Lintel element horizontal and misinterpretation of results; they can also be a cause of
element inefficient working practice.
Models should be:

h – planned based on clear objectives
Corner node
to return – developed in a manner that is fit for purpose and within
wall 0.6h the limitations of the software
b2/2 < 0.6h – verified
∴lintel extends
Vertical wall
to wall node It is important for engineers to recognise that software does
element
not relinquish any responsibility – and ignorant dependency
on software is an example of working without due diligence.
Hand calculations or simple models can almost always be
t /2 b1 b2 t /2 undertaken to demonstrate that a complex model is giving
reasonable results. These checks should always be carried
out.
Figure 5.7 Modelling lintels with 1-dimensional grillage elements
captured in the model with additional mid-storey (Box 5.7), not least because FE models can produce
horizontal elements. an overwhelming amount of information which the
engineer must decipher.
5.3.3 2-dimensional finite element models
Element type and meshing
2-dimensional finite element (FE) modelling is Choice of element type and their conditioning is
potentially the most detailed technique, being the critical within FE modelling. Shear wall elements can
truest representation of the structure (Figure 5.8). It is be modelled as plane-stress elements when in-plane
also increasingly one of the most convenient stresses dominate, otherwise shell elements need to
techniques, with software developers continually be used when both in- and out-of-plane stresses
improving and adding new automated links between need to be considered. Plate elements which ignore
analysis, design and documentation software in-plane stresses are not appropriate and plane strain
packages. However, this convenience tends to play elements are inappropriate in all but 2-dimensional
down the engineer’s responsibility and workload cross-section models (the likes of which may be used
to determine the necessary area of reinforcement in a
concrete section, but not to determine the design
actions on the section).
Box 5.6 Deep beam lintels
Grillage models are not suitable for structures containing Note it is essential that engineers understand the
lintels classified as ‘deep beams’ (where the span is less default element settings adopted by the specific
than three times the depth)5.2. The behaviour of these beams software they will be using.
is dominated by shear deformation and is better modelled
within a 2-dimensional finite element model. By adopting plane-stress elements the design
engineer dictates that a model will resist all forces by
in-plane stiffness only (i.e. walls bending in their minor
axes have zero stiffness and attract no force). This
approach is valid where redistribution is justifiable and
followed through with adequately ductile detailing of
joints and/or reinforcement. Otherwise, shell elements
are needed to incorporate out-of-plane stresses.
These stresses can be manually redistributed or
designed for but cannot be ignored.
1-dimensional auxiliary elements can be used in

conjunction with 2-dimensional elements. These may
be to enhance the accuracy of a model by replicating
a property of the structure (usually a stiffness) that is
not adequately replicated by the 2-dimensional
elements. Two common instances where these
auxiliary 1-dimensional elements are used are
introduced in Box 5.8.
It is typical to use quad 2-dimensional elements for

planar walls. The ideal shape of these is square and
the further the elements deviate from this (either in the
aspect ratio or interior angles) the less accurate they
Figure 5.8 2-dimensional element model become. As a guide, the length of the smallest side

Box 5.8 1-dimensional auxiliary elements in 2-dimensional models

Auxiliary elements are needed with plane-stress 2-dimensional elements to provide stiff anchorage to out-of-plane actions or to
in-plane rotations. They can be used to ensure continuity between wall elements and floor beams, as shown below:
1-D element to
lintel beam
1-D element
to floor Auxiliary
elements
Auxiliary elements
Auxiliary elements can also be used to recreate a torsion constant J of an open section. Plane-stress elements cannot represent
this stiffness and, while small relative to the warping stiffness, the stiffness can attract stresses that may otherwise be
redistributed. A single vertical column of 1-dimensional elements can be used, with each element being assigned a constant J
representative of the system:
Single column of 1-D auxiliary elements

with J = Jwall, Ix = Iy = 0
It should be noted that 2-dimensional FE models consider restrained warping stiffness intrinsically and do not require any special
consideration in this regard.
should not be less than half the length of the longest

side.
The size of elements should be such that the stress

does not vary significantly across the side length.
Significant stress discontinuities across the
boundaries of neighbouring elements are tell-tales Min. four elements
that the density of elements should be refined. across wall
Elements are typically refined locally for in-plane
stress concentrations at abrupt changes in the
geometry (Figure 5.9). Trapezoidal or irregular
elements are unavoidable in such instances and it is
essential that all nodes of all elements meet with
nodes of adjoining elements where present. Trapezoidal
elements
It is worth noting that Fig. 5.9 shows a regular array
of elements (collectively making a ‘mesh’) in a format
that is convenient to generate manually. Many
software programmes provide auto-meshing Min. six × four
functions and these tend to produce a far less elements to lintel
regular, ‘organic’ mesh pattern. Either is suitable
provided the sizes of the elements (the ‘mesh
density’) are reflective of the stress concentrations.
Lintels are governed by flexure and should be divided

to portray the stress flow. Four elements through the
depth and six across the span are recommended as
the minimum. However, good element aspect ratios
should be attained and this may lead to significantly Figure 5.9 2-dimensional element mesh

more elements either across the span or through the – The mesh element size is considered where a force
depth (depending on the geometry of the beam). per metre
P length of wall is needed, e.g.
NX ¼ (Nx,i / bi)
Output
where:
Raw output from a 2-dimensional FE model consists
Nx,i is the force on element i
of a stress field for each element which is often
bi is the side length of element i
integrated by the software over the area of the
NX is the force per unit length
individual elements. This integration returns stress
– Moments must include allP components acting on
resultants (forces and moments), shown in
the elements, e.g. MXX ¼ (Mxx,i þ |Mxy,i|) / bi
Figure 5.10 for a shell element. However, further
post-processing is needed to obtain values where:
representative of the wall as a whole or part that are Mxx,i and Mxy,i are the components shown in
suitable for design. Increasingly software packages Fig. 5.10
automate this but engineers should verify that this is MXX is the moment per unit length
being completed appropriately.
Ny M yx 5.4 Modelling horizontal stability systems

N yx M yy
Horizontal stability systems that link vertical stability

N yz M xx systems may be modelled using elements true to the
N xy structure or by using rigid constraints; the latter is
N xz shown in Figure 5.11 for the models previously
Nx shown in Figs 5.2 and 5.8.
M xy
These elements or constraints will distribute forces
internally and allow actions to be applied to a global
Figure 5.10 Stress field for 2-dimensional shell element model independently of either the stiffness or layout
of the resisting structure. Fig. 5.11 shows forces
When post-processing mesh results, care must be applied to nodes located at the centre of area for
taken to ensure: each storey as would be used for uniform wind load.
Rigid constraints
Figure 5.11 Modelling horizontal stability systems using rigid constraints

Further nodes could be positioned and loaded at an

Box 5.9 Apportioning actions between systems linked by a rigid
eccentricity (for non-uniform actions) or at the centre
diaphragm
of mass (for equivalent horizontal and static
earthquake actions). Py
y
Local shear
Rigid constraints are the simplest way of connecting centre of wall
e
discrete stability walls to the point or points of k y,n
loading. The decision to use them should be based
on two questions: k x,n
– How stiff is the horizontal structure and is an yi
infinitely rigid approximation appropriate x
considering each of: any penetrations, narrow
sections and the stiffness of the slab to wall k θ,n
connections?
– What actions are being modelled? Rigid constraints
must not be used when applying stresses resulting
from internal strains (e.g. thermal strains or post- xi
tensioning), or when using a model to evaluate
forces in the horizontal system. The apportioned action Py,n acting on a system is given by:

Particular care should be taken not to over-constrain k y;i ek y;i x i
any elements of the structure. Rigid constraints Py;n ¼ Py P +P . . .Eqn 5.9
k y;i ðk y;i x i 2 þ k x;i y i 2 þ ku;i Þ
should only link nodes of the core that are directly
connected to the diaphragm or horizontal bracing where:
and should generally only provide constraint in Py is the applied (global) action
horizontal planes. Neither should they run parallel to e is the eccentricity of the applied action measured from the global shear centre
(or bridge) elements that are being investigated, nor ky,i is the stiffness of the i’th system parallel to the action
connect to fixed ‘boundary’ nodes. xi is the distance to the shear centre of the i’th system measured from the global
centre of stiffness and orthogonal to the action
Where the floorplate is instead modelled using kx,i is the stiffness of the i’th system perpendicular to the action
elements representative of the structure, it is typical yi is the distance to the shear centre of the i’th system measured from the global
to use 2-dimensional elements for diaphragm slabs. centre of stiffness and parallel to the action
Elements can be of plane-stress type when modelled ku,i is the local torsional stiffness of the i’th system
solely for the purpose of assessing the diaphragm
action. The mesh density should be reflective of the Note that the global centre of stiffness is analogous to the shear centre of the individual
stress concentrations, especially locally around the wall elements. It is the point through which a transverse load will cause pure shear
shear walls. (without torsion). Its position can be found using Equations 5.5 to 5.8 (Box 5.4).
Otherwise, where the model will be used for load

takedown or slab design, or where the slab
contributes to the stiffness of the vertical structure and should be commensurate with the general
(e.g. by contributing to frame action), shell elements accuracy of the analysis. Flexibility at the boundary
must be used. The mesh should be suitably dense in nodes should be considered in each of the six
all spans and around vertical supports to allow the degrees of freedom (Figure 5.12). Any variability (e.g.
flexural and shear behaviour to be determined. This of the founding soil stiffness) or uncertainty (e.g.
typically requires not less than six elements across where founding on an existing structure of unknown
each span. design) should be captured in an enveloping
sensitivity study.
5.5 Manually apportioning actions

between vertical stability systems
z
Box 5.9 details how loads can be manually

apportioned between vertical stability systems linked
by a stiff diaphragm.
y
Although quite simple in theory, the calculation can
quickly become unwieldy in practice. The stiffness of
the individual wall systems ki is dependent on both
shear and flexural stiffness and can vary through the
height of the building. This can have a knock-on
effect on the global centre of stiffness, and collectively
will cause the loads to redistribute.
5.6 Modelling boundary conditions

x
The rigour with which the boundary conditions are
modelled can have significant impact on the results Figure 5.12 Degrees of freedom

5.7 Elastic and plastic analysis Note that plastic analysis as discussed in this section
is different to plastic section design. The latter is
widely used to justify resistance of elements at the
Analysis can be elastic, rigid-plastic or elastic-plastic. ultimate limit state and is applicable in conjunction
with elastic, rigid-plastic, or elastic-plastic analysis. It
Elastic analysis is normally adopted for the derivation is important that designers distinguish analysis from
of forces throughout shear wall systems. It is element design in this way.
appropriate for all materials that exhibit homogeneous
behaviour at both the serviceability and ultimate limit
states (Figure 5.13(a)), and is applicable irrespective
of the structural form and/or the failure mechanism. 5.8 References
Rigid-plastic analysis is rarely appropriate for global
shear wall analysis. It relies on stable and predictable 5.1 Irwin, A.W. Design of shear wall buildings. CIRIA
post-elastic deformation with neither fracture nor Report 102. London: CIRIA, 1984
buckling causing premature failure.
Elastic-plastic analysis is, however, common at the concrete structures – Part 1-1: General rules and rules
ultimate limit state for walls with lintel beams. These for buildings. London: BSI, 2004
relatively shallow beams can often be designed with
sufficient ductility and stability to form predictable
plastic hinges. Hinges can either be assumed by the
designer (and modelled as springs of constant
rotational stiffness) or found by non-linear iterative
elastic-plastic analysis. Both techniques should lead
to a hinge pattern similar to that shown in
Fig. 5.13(b).
Alternatively, the in-plane stiffness of lintels can often

be omitted outright at the ultimate limit state where
the lintels are not necessary for overall stability
(Fig. 5.13(c)). This is an elastic-plastic approach,
equivalent to assuming plastic hinges that are of
negligible rotational stiffness. It has the advantage of
being able to simplify the model to one that can be
analysed elastically.
Plastic hinge Strut (of zero flexural stiffness)
Modelled
Deformed
(a) (b) (c)

Elastic ULS with plastic hinges ULS with lintels omitted
Figure 5.13 Modelling the elastic and plastic behaviour of shallow lintels

6 Monolithic reinforced concrete shear wall construction
6.1 Introduction Modelling variable parameters

In many instances the engineer may find it necessary
to carry out analysis across a range of material
This chapter looks at characteristics of in situ moduli and with upper and lower bound criteria for
reinforced concrete walls. These tend to be the section properties (see the example family of
monolithic and loadbearing. They are usually models shown in Table 6.1).
designed as compression elements under the
combined action of in-plane bending and axial forces. Cracked properties can be applied uniformly to all
elements within a model, banded through the
building height, or applied on an element-by-
element basis following an initial or iterative
6.2 Modelling the stiffness of concrete appraisal of the tensile stresses. A uniform
application is by far the simplest but is likely to
underestimate the stiffness of the structure. An
Stiffness is critical to serviceability deflection limits but element-by-element application is more accurate,
also to the load distribution within a structure. It is though may prove impractical without a semi or
dependent on the material elastic modulus, tension fully automated iteration script to determine the
stiffening, creep and the effective section. extent to which each section is in tension. It may
also lead to a false sense of accuracy, potentially
Modulus of elasticity, tension stiffening and creep not taking account of the true load conditions or
The modulus of elasticity E for concrete is a variable locked in stresses. A banded application is a
usually defined at 28 days in line with the 28 day middle road approach. Here, the degree of
compressive strength (Box 6.1). These variables are cracking is applied incrementally through the
co-dependent; both are characteristics of the height of the structure and is based on the
concrete mix and both are age (‘maturity’) approximate extent of tension at representative
dependent. Figure 6.1 plots their development for a sections to each band.
C35/45 Class R concrete6.2. This graph has been
derived in accordance with Clause 3.1 of Cracked section properties are typically modelled by
BS EN 1992 Part 1-16.3 and plots the secant reducing the second moment of area I of the solid
modulus. It shows that both the strength and wall. This can be carried out in most software
stiffness are asymptotic over time. packages with either a property modification factor or
by overriding the default value (the default having
Box 6.1 Specification of concrete strength and been derived from the gross geometry of the
stiffness element). Either method has no impact on other
properties. However, note that it would be wrong to
The 28 day strength has traditionally been and remains the change the explicit section geometry (which would
principal parameter for design and specification. There is, influence the area and axial stiffness).
however, an increasing trend to measure and limit variation of
stiffness where sway, creep, differential shortening and/or PD Meanwhile it is common practice to apply the effects
effects are significant. Monitoring of stiffness is often of tension stiffening and creep by virtue of a modified
specified up to 56 days, at which point it is approximately modulus of elasticity E. This is usually applied
95% of the asymptotic value when Portland cement is used uniformly to the whole structure and can be via a
(this may vary where cement replacements such as fly ash modification factor or override to a default material
are specified within the mix6.1). property, or by adopting a user-defined material
property.
In addition to this time-based development of
stiffness, load duration and/or load cycles influence
tension stiffening and creep (also known as
‘relaxation’). BS EN 1992 Part 1-1 Clause 7.4.3 gives
factors of 1.0 for short-term loads and 0.5 for 55 40
Compressive strength f cm(t ) (N/mm2 )
Modulus of elasticity E cm(t ) (×103 N/mm2 )
sustained or cyclic loads in recognition of this. The

Concrete Society gives guidance on when to use the 50 38
long- and short-term factors6.4. It concludes that the 45 36
long-term factor should be used for all loads
40 34
anticipated to last more than approximately 20 days.
35 32
Effective section 30 30
The effective section is a function of the area and
placement of reinforcement and the extent to which 25 f cm(t ) 28
the concrete is cracked. Cracking is a brittle 20 E cm(t ) 26
mechanism of limited predictability. It is dependent on
15 24
the most severe historical load condition experienced 28 56 100
1 10 1000
by a section (not always related to the load condition
being investigated). However, only cracks in the Age of concrete (days)
tension zone need to be considered for any given
load case. Figure 6.1 Development of stiffness and strength with concrete age

6.3 Monolithic reinforced concrete shear wall construction
Table 6.1 Example modelling matrix for concrete

Material moduli Effective section
Lower bound crackingb Upper bound crackingb
I walls 90% I gross I walls 50% I gross a
I lintel 50% I gross I lintel 30% I gross
Short-term A
3 A
3 For short-duration static and low frequency dynamic
Ecm(t ) (Model 4) (Model 2)c actions
Long-term A
3 A
3 For long-duration and permanent static actions
0.5Ecm(t ) (Model 3) (Model 1) c
Notes
a Value taken from ACI 318M-116.5.
b Percentages given for the effective lower and upper bound cracked section are guide values recommended for an initial
assessment. They should be reviewed against the stresses determined within the analysis and revised as necessary.
c The upper bound cracking models (Models 1 and 2) should be used to determine PD effects at the ultimate limit state.
coincident compression or tension with flexure and

Further reading: design properties of concrete
shear.
The following texts are recommended sources of further
guidance on the design properties of concrete: Adequate reinforcement must be provided to resist
– Reynolds, C.E., Steedman, J.C. and Threlfall, A.J. the combination of normal (predominantly vertical)
Reynolds’s reinforced concrete designer’s handbook. and shear stresses to all sections. Special attention
11th ed. Abingdon: Taylor & Francis, 2008 should be given to stress concentrations including
– Neville, A.M. Properties of concrete. 5th ed. Harlow: those at re-entrant corners around openings and at
Pearson, 2011 slab junctions. The strut and tie design method is
– Bamforth, P.B. et al. Properties of concrete for use in recommended for regions in which cross-sections
Eurocode 2: how to optimise the engineering properties of undergo non-planar deformation.
concrete in design to Eurocode 2. CCIP-029. Camberley:
The Concrete Centre, 2008 Most codes allow an idealised plastic stress block
model for section design at the ultimate limit state
(Figure 6.2). This is a simple, proven technique for
determining lower bound plastic strength. However, it
is important engineers appreciate that it assumes
6.3 Ultimate and serviceability limit state large strains equivalent to the stresses. This is
design of reinforced concrete reasonable in many scenarios but can be critical
sections when either second order PD effects or load sharing
is significant.
Reinforced concrete shear walls can be designed The Institution’s Manual for the design of concrete
using methods set out in codes of practice for building structures to Eurocode 2 Section 5.6.4.16.6
provides a simplified rigid-perfectly plastic calculation
for the vertical reinforcement in a wall not subject to
significant minor axis bending.
ε σ
An elastic stress method is an alternative approach at
the ultimate limit state and is necessary for the
serviceability limit state designs. It can be
Perfectly elastic
advantageous for the ultimate strength design where:
– A wall is particularly slender with heightened risk of
buckling instability
– A wall has particularly complicated geometry with
ε σ an irregular arrangement of reinforcement.
Increasing curvature
An elastic calculation requires the modular ratio for

Elastic – the concrete and reinforcement. This will vary
perfectly plastic depending on the modulus of elasticity of the
concrete. The modulus of elasticity of steel is
constant irrespective of load, but the design value
does vary by region. It typically lies in the range of
ε σ
200 103 to 210 103 N/mm2.
Elastic stresses must remain below the elastic limits

Rigid – of the materials, i.e. the yield stress for steel
perfectly plastic reinforcement and the compressive strength for
concrete. These limits are critical at the ultimate limit
state.
Reinforcement tensile stress is commonly used to

Figure 6.2 Elastic, elastic-perfectly plastic and rigid-perfectly plastic stress block models define elastic design parameters for the serviceability
(shown for flexure) limit state. The limiting stress is a variable for a given

Monolithic reinforced concrete shear wall construction 6.4
steel grade, dependent on each of: the allowable at laps. Overly thin and congested sections will lead
crack width, the cover, the bar diameters and the bar to challenges when placing and compacting the
spacing. A larger quantity of closer spaced, smaller concrete. The consequence of this can be special
diameter bars permits higher stresses but may have mix requirements, slow progress on-site,
implications on construction6.6. honeycombing and poor surface finish quality
(Box 6.2).
Box 6.2 Constraints on construction

6.4 Concrete classes
Constraints imposed on the construction by a design decision
should be clearly described in the contract documents and/or
Readily available concrete classes (otherwise referred on the drawings. Thin walls, especially those requiring a good
to as ‘concrete grades’) vary by geographic region. finish quality, can restrict pour heights and limit the maximum
As a general rule, higher classes are more aggregate size. These are both measures necessary to avoid
economically available in urban areas where there is segregation and honeycombing but often incur a cost
steady demand justifying large scale production. premium.
However, underpinning the availability and cost of
high strength mixes is the availability of raw materials. The Institution’s publication Standard method of
The strength of aggregate is fundamental to the detailing structural concrete6.7 recommends that
strength of the concrete and is a function of the in situ walls are no thinner than 150mm. Walls of this
geology of the quarry from which it is sourced. thickness should be detailed and constructed with
Likewise the availability of cement replacement lifts not exceeding 1.8m to ensure dense placement.
materials such as ground granulated blast furnace Reinforcement is almost certainly limited to a single
slag, fly ash and silica fume is influential. The highest layer each way positioned centrally in the wall. Not
class concretes often need fractions of these only is this inefficient, but it is also difficult to secure in
components as part-replacement to Portland cement place and may not be adequate to control surface
to achieve enhanced densities with lower void ratios. cracking.
Whether raw materials can be sourced locally will
influence the delivery cost. 180mm is a more practical minimum thickness6.6,
allowing storey-height walls to be cast in single
For most moderate to high-rise buildings, it is pours. These walls can generally accommodate two
common practice to use a mix that is one or two layers of vertical and horizontal reinforcement (one
classes stronger for the vertical elements than is used fixed near each face) and span clear heights of the
for the floor slabs. In this common scenario, care is order of 3m.
needed to ensure that either the higher strength
concrete is continuous through the floor-wall
intersection or that the lower strength concrete and
local reinforcement is adequate. Which of these 6.6 Reinforcement and embedments
approaches is adopted is often dependent on the
construction: continuing the wall through the floor-
slab is convenient for slip- and jump-forms but less Both minimum and maximum amounts of
so for traditional shuttered lifts (Section 6.6). reinforcement in each of the vertical and horizontal
axes are typically defined by codes of practice. The
Most codes of practice list common strength classes, definitions are usually given as ratios of the gross
and some go further to list standard mix designs. concrete area for a section taken orthogonal to the
Engineers should recognise how these ranges relate reinforcement.
to the applicability of the code. Extreme care should
always be taken where a design uses materials that Minimum reinforcement ratios are dependent on the
are outside the scope of the code. design requirements (stress resultants, crack control,
durability and fire resistance) and the construction
Further reading: concreting and concrete mixes method. Meanwhile, maximum reinforcement ratios
are governed primarily by buildability with limits set to
The following text is a recommended source of further prevent over-congestion and poor concreting. Both
guidance on concreting and concrete mixes: minimum and maximum ratios apply throughout
– The Concrete Society. Concrete practice: guidance on the elements but maximum ratios tend to only be critical
practical aspects of concreting. Good Concrete Guide 8. locally at junctions with starter bars or where bars are
Camberley: The Concrete Society, 2008 lapped.
Over-reinforcing the tension zone of shear walls (so

that they would fail in flexure by concrete crushing) is
6.5 Minimum wall thickness rare; most shear walls in building structures are
without significant out-of-plane actions and are
designed with equal reinforcement to each face. This
Global strength and stiffness, local stress is both to accommodate load reversal and to simplify
concentrations, fire resistance, thermal and/or construction.
acoustic insulation, the need for chase-outs or
embedded fixings and durability can all influence the Wall reinforcement should be detailed in accordance
minimum wall thickness. with codified rules and guidance given in the
Institution’s manual Standard method of detailing
Once these points have been considered, it is worth structural concrete6.7. Salient points include:
sketching out, to scale, the critical sections with the – Vertical bars should always be in a layer that allows
most congested reinforcement. These will usually be the horizontal bars to be tied once the vertical bars
within lintels, at junctions to floor beams or slabs, and are in place (Figure 6.3). This can have significant

Cage built before

forms are added
Both forms installed only when

(a) Symmetrical placement reinforcement cage is complete
Form installed last
Cross links to
Cage built off vertical bars
formed face
Form installed first

(b) Placement away from a formed surface
Note that (a) is standard for shear walls but (b) is more efficient where a dominant Note
unidirectional lateral force is acting (e.g. soil pressure on retaining walls). Horizontal bars omitted for clarity.
Figure 6.3 Wall reinforcement layering (sections shown on plan) Figure 6.5 Cross links in heavily reinforced walls
impact on the effective depth (critical in walls that construction sequence. Laps in the wall
are slender, and/or that are subject to significant reinforcement should generally avoid these areas to
out-of-plane actions) prevent unnecessary congestion.
– Cross links may be required at wall returns
(Figure 6.4) and across heavily reinforced sections Finally, it is worth noting that the Institution’s detailing
(Figure 6.5) to overcome bursting forces manual6.7 stipulates a maximum bar spacing of
– Diagonal bars should be positioned across 400mm, in line with BS EN 1992 Part 1-1 Clause
re-entrant corners (i.e. corners around openings – 9.6.36.3. However it is considered best practice to
Figure 6.6). These are usually adequate if designed limit this to 300mm. This limit is enforced by
for a tensile force equal to twice the horizontal BS EN 1992 Part 1-1 Table 7.3 which lists maximum
shear force in the vertical component of the wall, serviceability bar stresses for bar spacings up to
but should not be less than two 16mm diameter 300mm only.
bars across each corner of the opening6.6
– Laps should be detailed to avoid localised
congestion (e.g. with staggering)
6.7 Construction
Additional local reinforcement (starter bars) and/or
embedments (Box 6.3) are usually essential where
floor beams frame into walls. These can be junctions The construction method is usually a contractor’s
of high local stress and design intricacy. Early proposal with little impact on the completed
coordination across disciplines should try to avoid structure. However, the choice of method can have
service penetrations in these zones. Subsequently, significant impact on the appropriateness of a
close attention during the detailed design needs to be design. The following paragraphs describe common
given to the forces being transferred, to the methods of construction of in situ reinforced
tolerances of the individual systems and to the concrete walls.
Cross links to corner

Diagonal bar across
re-entrant corner
U-bars at wall edges

lapping to main vertical
and horizontal bars
Figure 6.6 Diagonal reinforcement across re-entrant corners

Figure 6.4 Cross links at wall returns (section shown on plan) including penetrations (shown in elevation)

Box 6.3 Embedment plates

Embedment plates are commonly used for the connection between a steel structural frame and a reinforced concrete shear wall.
They typically transfer both vertical and horizontal actions from the frame to the wall.
The photographs (below) from 1 Grafton Street6.8 show the embedment plates before and after concreting. In reinforced concrete
construction, an embedment plate is an assembly made up of a flat steel plate with shear studs and/or reinforcement welded to the
rear that projects into the wall. A fin plate that projects out from the wall and is necessary for the beam connection (Fig. 2.4) is
site-welded once the concrete is cast and formwork removed. Site welding allows the fin plate to be positioned accurately; omitting
the fin plate during the initial installation also means that the cast-in assembly can be set flush with, or nominally in from, the face
of the wall without impacting on the formwork.
Embedment
plates
Embedment plates are generally oversized to allow for tolerance in the concreting. Their design should assume the fin plate is
welded in the most onerous conceivable position relative to the shear studs, that reinforcement is welded to the rear of the
embedment plate, and that the bolted connection to the beam is at the maximum eccentricity from the wall.
Embedment plates should always be shown on reinforcement drawings to be fixed with the wall reinforcement.
Traditional shuttered lifts Box 6.4 Kickers

Traditional shuttered lifts may be used when other Kickers are small upstands cast ahead of the wall (typically
methods are not justified. This may be because the 75mm high and cast with the floor slab) that provide a useful
total number of walls in the building is small, or surface to secure formwork against. In this way they can
because they exhibit little or no repetition. Otherwise improve the quality of the column aiding both the setting out
the technique may be used to achieve a premium accuracy and the seal at the base of the form.
finish quality or texture.
Wall above
The method of construction has the walls formed one (Pour 3)
storey at a time, often in parallel with the columns. It
is slower than other methods listed but can Pour joints
Kicker (75mm typical)
accommodate the greatest variation between panels.
Construction accuracy is influenced by:

– The degree to which the formwork is erected out of
plumb
– The setting out and alignment of one wall
immediately above the one below Slab (Pour 2
– The spacing of the forms (dictating the wall with kicker)
thickness)
– The stiffness of the forms (to withstand the Wall below
pressures during and after concrete pouring) (Pour 1)
– The spacing and alignment of reinforcement
However, forming kickers is in itself not without challenge.
Acceptable limits and/or criteria should be set out Scrutiny is needed during concrete placement as the kickers
for each of these points via a project or industry/ can be prone to poorly placed/compacted concrete, debris in
national specification, e.g. the UK National structural the construction joints and/or weak concrete slurry. Kickers
concrete specification6.9. The design may also can also be unsightly and are often unpopular with both
include features such as ‘kickers’ (Box 6.4), while clients and architects where columns will remain fair faced in
rationalised reinforcement spacing and/or cover areas without a raised floor.
zones is often favoured to decrease the risk of errors Kickerless construction can overcome these quality issues but
on-site. can exacerbate issues including grout leakage from the base
of the column formwork. It is recommended that the two
Slipform techniques are considered by the design team and contractor
Slipforming is a method of concrete placement jointly.
whereby a moving form is used to create a It should be noted that the inclusion or omission of kickers
continuous wall extrusion. When used for a well- can impact the reinforcement bar scheduling and
suited structure, it is very efficient achieving unannounced site changes must be avoided.
unparalleled speed.

Jump form
Jump form, or climbing form, is a method of
construction whereby the walls are cast in discrete lifts.
It is a stop-start process with day joints formed at

each lift level. At its optimum, storeys can be cast on
a 24 hour cycle. This requires standard concrete with
28 day cylinder strength not less than 40N/mm2 to
achieve strengths of 15N/mm2 at the striking time
(typically after just 14 hours)6.10.
Like slipforming, jump forming is only efficient where

there is significant repetition in the structure through a
number of floors. It is most applicable in medium-
and high-rise buildings of ten or more storeys.
However, unlike slipforming, jump form can

incorporate discrete features other than simple
extrusions. This makes it better suited to lift cores in
Figure 6.7 Slipformed cores under construction shown with and without the climbing
particular where anchors for the lift guide rails must
platform
be positioned with high accuracy.
Jump form is also far better suited to day-time only

Flanged and core wall systems are generally suited to work hours and is less sensitive to unplanned events
slipforming where they are of regular section through causing a halt on the programme.
multiple storeys (Figure 6.7). Floor connection details
(embedments and/or starter bars), doorways, A key disadvantage of jump form relative to slip form
blockouts, steps in thickness, tapers and dense is the inclusion of lots of day joints; each can be a
reinforcement can add complication and lead to cause of defects during construction by virtue of
reduced efficiency. In general, slipforming is chosen poorly placed/compacted concrete and debris at the
where speed is the key driver and compromises to joint. These joints can also cause obvious banding in
achieve vertical regularity can be incorporated in the the concrete (Figure 6.8).
design.
Tunnel form
An efficient slipform will progress at a rate in Tunnel form construction uses a formwork system to
excess of 6m per 24 hours with non-stop working cast slabs and walls as a single pour operation. It is
day and night. The technique is less well suited at economical for cellular structures with repetition of
sites with night time restrictions on work, where the structure both horizontally and vertically (Box 3.1).
weather is likely to be prohibitive or material supply Hotel buildings, with many identical cellular rooms,
is unreliable. are ideally suited.
Accuracy and minimum wall thicknesses are Construction progresses vertically and horizontally
comparable to traditional formwork methods, simultaneously, with an inclined work front stepping
however the construction can impact on the area and back at each level up the building. As with jump form,
disposition of the reinforcement. ‘tunnels’ can be poured on a minimum 24 hour cycle,
before the wall forms are struck and relocated to the
Aesthetically, slipforming gives a rough finish with next bay.
vertical streaks caused by abrasion of the form on
the walls. Horizontal banding can also be apparent,
resulting from minor variations in the concrete
supply. Thus, it is generally not suitable for fair faced
walls.
A slipformed wall will always progress in advance of

the surrounding structure and the design must
consider the temporary condition. Although the
dead weight of the slipform structure and the
platform is relatively small, where pull-out bar boxes
are used at floor slab levels the effective thickness
of the wall can be reduced locally by up to 80mm in
advance of the slab being poured. This can be
critical while the concrete is green and the walls
unrestrained. Where walls are inadequate without
the restraint of the slabs, temporary bracing (fixing
to cast-in plates) may be necessary to brace
between panels. The installation of this bracing can
be a key consideration and must be factored into
the slipform operation.
Even with temporary bracing, the early strength gain

of the concrete can dictate the maximum speed of
the form and the mix design can have a critical
impact on the programme. Figure 6.8 A jump formed core under construction

The method requires significant free space to swing

out large units of formwork and should have 28 day
cylinder strength not less than 40N/mm2 for the same
reason given for jump form.
Further reading: in situ reinforced concrete construction

The following texts are recommended sources of further
guidance on in situ reinforced concrete construction
techniques:
– The Concrete Society. Slipforming of vertical structures.
Good Concrete Guide 6. Camberley: The Concrete Society,
2008
– The Concrete Centre. High performance buildings: using
tunnel form concrete construction. TCC/04/02. Camberley:
The Concrete Centre, 2004
6.8 References
6.1 The Concrete Society. Concrete practice: guidance on

the practical aspects of concreting. Good Concrete
Guide 8. Camberley: The Concrete Society, 2008
6.2 Narayanan, R.S. and Goodchild, C.H. Concise Eurocode

2: for the design of in-situ concrete framed buildings
to BS EN 1992-1-1:2004 and its UK National Annex:
2005. CCIP-005. Camberley: The Concrete Centre,
2006

concrete structures – Part 1-1: General rules and rules
for buildings. London: BSI, 2004
6.4 The Concrete Society. Influence of tension stiffening on

deflection of reinforced concrete structures: report of a
Concrete Society working party. Technical Report 59.
Camberley: The Concrete Society, 2004
6.5 ACI. ACI 318M-11: Building code requirements for

structural concrete (ACI 318M-11) and commentary.
Farmington Hills, MI: ACI, 2011

design of concrete building structures to Eurocode 2.
6.7 Institution of Structural Engineers and The Concrete

Society. Standard method of detailing structural
concrete: a manual for best practice. 3rd ed. London:
IStructE, 2006
6.8 Perry, P. ‘Development over London Underground

tunnels: No. 1 Grafton Street’, Proceedings of the
Institution of Civil Engineers – Structures & Buildings,
2014 [Online]. Available at: http://dx.doi.org/10.1680/
stbu.11.00025 [Accessed: 12 January 2015]
6.9 Construct. National structural concrete specification for

building construction. Fourth edition complying with
BS EN 13670:2009. Camberley: The Concrete Centre,
2010. Available at: http://www.construct.org.uk/media/
National_Structural_Concrete_Specification_for_
Building_Construction.pdf [Accessed: 12 January
2015]
6.10 The Concrete Centre. Concrete framed buildings: a

guide to design and construction. TCC/03/024.
Camberley: Concrete Centre, 2006

7 Non-monolithic shear wall construction
7.1 Introduction The pros and cons mean that off-site fabrication
generally favours buildings with highly repetitive
functional requirements including bedroom blocks for
Non-monolithic walls are often of precast hotels, universities and prisons. In each of these
construction, arranged as storey-height units with examples, there is a clear advantage to the building
horizontal joints coinciding with floor slab owner/occupier if the rooms are standardised, while
connections. Common material systems include: the repetition means that large savings are possible
– Precast reinforced concrete (including tilt-up from relatively minor design refinement.
construction)
– Hybrid precast in situ reinforced concrete Aesthetically, off-site fabrication can achieve
– Timber and light gauge steel platform frame unparallelled favourable results in terms of surface
construction finish quality. It is however very hard to mask an
– Cross laminated timber (CLT) assembled structure where the components are on
display. Hence joint patterns can be as important as
Section 7.2 discusses general characteristics of these overall form to the appearance of a structure
systems, with Sections 7.3–7.6 focusing on the (Figure 7.2). This should be brought to the attention
additional characteristics of the systems listed in the of both architect and client.
previous paragraph, in turn. Each of these sections
has a bias towards the joints which are key elements The economic viability of these systems varies
in the design. significantly internationally. Local production facilities
and market competitiveness, raw material availability,
Introductions to both loadbearing masonry and skilled labour and international financial exchange
steel plate diaphragm walls are also included in rates can have an impact. Meanwhile transport, site
Sections 7.7 and 7.8. Both systems differ access and craneage restrictions can influence
significantly to those listed here as well as from one whether an off-site fabrication option is viable at a
another. specific site. In the absence of site specific
restrictions, maximum panel sizes are usually in the
Discussion of proprietary modular ‘system builds’ order of 12m 3.5m and not more than 20 tonnes.
(e.g. Tata CorefastTM 7.1) is omitted. These systems However, to establish site specific bounds on panel
are generally developed via a manufacturing sizes and weights, it is usually essential that the
approach to research, design, testing and product designers consider the construction sequence and
iteration, and are often procured complete with crane locations early in the design development. This
specialist in-house design services. may require early contractor consultation.
Further reading: guidance on craneage
7.2 Precast construction The following texts are recommended as a source of further
guidance on craneage:
– BS 7121-1: 2006: Code of practice for safe use of
Off-site fabrication cranes – Part 1: General. London: BSI, 2006
Advantages of off-site fabrication over in situ – Skinner, H. et al. Tower crane stability. CIRIA C654.
construction can include less site labour, faster London: CIRIA, 2006
construction and better quality control. Longer lead in
times, greater need for standardisation and additional Characteristics of jointed systems
contracted party interfaces in the design and/or Joints are a key part to any off-site fabricated system.
construction are three common disadvantages. A While large areas of the panels can behave in a
more extensive, but still not exhaustive, list of pros similar manner to monolithic walls, joints will often
and cons is given in Figure 7.1. have a significant impact on the design. These usually
Pros Cons
– Faster site construction – Less inherent robustness
– Less wet trades on-site – Poor joint details can be points of structural weakness and
– Less skilled labour on-site can have a detrimental effect on fire resistance, thermal and
– Safer working conditions: less work at height, better safety acoustic insulation and waterproofing
controls – Panels require built-in erection tolerances and long-term joint
– Better control of component quality locking mechanisms
– Greater opportunity for architectural finishes (e.g. cast surface – Less opportunity for flexural continuity leading to structural
patterns and/or colour pigments in concrete panels) performance inefficiencies
– Greater component precision, allowing refined designs – Erection logistical challenges: transportation, delivery, storage
(e.g. reduced permissible concrete cover to reinforcement) craneage and temporary propping
– Longer lead-in period
Figure 7.1 Typical pros and cons of off-site prefabrication, stated relative to in situ construction

Non-monolithic shear wall construction 7.2
Figure 7.2 Buildings with exposed prefabricated elements
introduce components that may have vastly different (including robustness forces). The joints tend to be
properties to those of the panels; they can also cause located at or close to slab levels (the exact position
high stress concentrations where the joints are varies by material construction). This ensures the
discrete. These features will often add to the potential necessary out-of-plane restraint close to the joint but
failure mechanisms and can impact on the overall also allows the construction to proceed a storey at a
robustness of a structure (Box 7.1). time with a high degree of repetition.
Most precast wall systems contain both vertical and Vertical joints primarily transfer vertical shear between
horizontal joints between panels. Neither of these panels to resist the deformation shown in Fig. 2.5c.
joints tends to provide significant minor axis bending Their placement is largely dependent on the specific
resistance and panels will normally be designed to wall geometry together with limitations on fabrication,
span one-way between points of lateral support. This transportation and craneage. Openings should be
may be horizontally between return walls but is more considered when deciding on their layout. Whether
often vertically between floor slabs. an opening can be housed within a single panel will
often depend on the panel and opening dimensions
Where panels span vertically, horizontal joints (i.e. and whether the panel can sustain the temporary
those between panels stacked one above another) stresses during lifting.
transfer horizontal shear, compression and tension
Actions through both lintels and vertical joints can be
Box 7.1 Robustness reduced significantly in the completed structure by
staggering openings and joints from floor to floor
(Figure 7.3). While efficient, this is rarely practical and
tends not to be favoured by the architect and/or
services engineer.
Joints will often contain steel connection

components. These may be reinforcing bars with
couplers or grout ducts in concrete systems; nails,
screws and wall ties in timber platform construction;
or bolts with or without gusset plates in CLT
construction. Both the connector and its anchorages
into the panel must be sufficient to transfer the design
and minimum robustness forces.
Vertical joint

The direct failure of a single loadbearing precast wall panel Openings 
‘framed’ by 
caused by a domestic gas explosion led to the partial
collapse of Ronan Point, a 22-storey apartment building in 
wall and 
London. This event publically highlighted the need for
lintel panels 
robustness criteria7.2, 7.3. 
Although the failure did not result in a loss of global stability, 
the scale of the failure – which extended through the four  ‘Deep beam’
floors above the explosion and to all 17 floors below – was Openings  spanning
deemed disproportionate to the cause. Failure was concluded contained  opening below

to have been ‘progressive’, with the failure of each wall panel within 
leading to the overloading and/or loss of restraint to its single panels 
neighbour. 

The failure prompted a review of UK Building Regulations, the
findings of which established design philosophies for
Horizontal joint
robustness which are used universally adopted in modern
codes of practice.
Figure 7.3 Joints around openings

7.3 Non-monolithic shear wall construction
Joint details must provide sufficient tolerance to allow erection is usually fast, it is preceded by a significant
erection to proceed. Where this initial tolerance lead-in period during which the material order, factory
invalidates a load path that is later relied upon, a work and site delivery take place. Approvals, reviews
locking mechanism is needed. This might be and any necessary dialogue between the consulting
mechanical with an insert or locking nut, or chemical engineer and the subcontractor/supplier must also
with grout or resin. How it is fixed or installed must occur and be programmed for in this period.
be considered when planning the joint and can
impact the setting out, access requirements and load Finalised design information must be issued ‘For
capacity of the connection. In many instances the Construction’ and provided to the subcontractor in
dimensions and requirements of a joint can govern advance of the lead-in period. It must include all details
the thickness of a panel. for the end product: reinforcement, cast-ins, cut-outs,
connectors and fixings, etc. To produce this level of
Joint slip detail, the design of the building services must be well
Joint slip is movement within a joint that is not progressed with all penetrations known; the architect
representative of the wall panels. It can result from must have finalised all setting out information; and the
either tolerance between components or from elastic subcontractor must have erection, fixing and/or in situ
or plastic deformation within the highly stressed pour sequences developed sufficiently to locate
region of the joint. It is most pronounced where: cast-ins for temporary props.
– The joint’s stiffness is significantly less than the
panel stiffness The level of detail provided by the consulting engineer
– Discrete joints cause significant stress can vary between projects, with detailed design
concentrations in either the connector or the panel responsibility often split between the consultant and
– Multiple joints have a cumulative effect on the the subcontractor/supplier. Without exception, the
global stiffness of a system (e.g. where the jointed consulting engineer must, however, maintain
panels are narrow relative to the overall dimension responsibility for the global behaviour of the
of a wall) completed structure. As a minimum, they must
– Joints are positioned in areas of high stress (e.g. develop a credible scheme that presents feasible
around openings) outline designs and performance requirements for the
panels and principal joints upon which the design can
Slip resulting within a joint should be considered in be detailed. Whether the design consultant then goes
the overall stability model where it is significant to the on to detail the panels and the joints will be
global behaviour of the system (Box 7.2). Slip can be dependent on the supply chain and contractual
critical where it causes a change in global stiffness arrangements.
that has a significant impact on:
– The distribution of forces between walls in a multi-
walled structure
– The magnitude of PD effects 7.3 Precast reinforced concrete wall
– The serviceability sway deflection construction
Design responsibility and programme
Off-site fabricated elements follow a procurement Precast reinforced concrete wall construction has
model similar to structural steelwork. While site been common since the middle of the last century.
Box 7.2 Modelling joint slip with 1- and 2-dimensional elements

Joint slip may be difficult to predict accurately in analysis and a sensitivity study is often useful to determine the significance of
joint flexibility. At preliminary stages, slip can be modelled approximately by varying the stiffnesses of the panels above and below
the theoretical design value, or by imposing non-elastic deformations between nodes of the model. Some element and connection
forces will be amplified by increased stiffness, while others are amplified by decreased stiffness. Where the joint slip proves to have
significant impact on the stiffness (say by altering the stress distribution by more than 10%), physical testing may be required to
determine refined bounds on the design parameters. These parameters are often non-linear and should make allowance for any
time-based softening, friction loss and creep.
Joint slip can be modelled in either a grillage or 2-dimensional FE model. In a grillage model, it is best achieved by joining elements
with translational springs in the axis of the joint. In 2-dimensional FE models, a similar outcome is best achieved by having a
continuous or broken line of ‘soft’ elements of reduced in-plane shear stiffness coinciding with the joint plane.
Stub element with high flexural

stiffness/low axial stiffness
Low stiffness
elements
Vertical joints
Panelised wall 1-D element model 2-D element model

. Vertical joint
. Min m
Min m 75
m Steel plate
0 m
15
Oversized
washers
Pier
Elevation
Anchor bolts
Spandrel into cast-in
panel ferrules
Plan
Figure 7.4 Minimum wall thicknesses Figure 7.6 Vertical joint (shown between planar panels;
corner details similar)
Out of favour in the UK following the Ronan Point sleeves are grouted enabling the dowel to transfer
failure (Box 7.1), it has seen a steady revival fuelled by shear and tension. Compression is transferred away
the advantages of off-site production. These drivers, from the dowels via a combination of any local shims
together with innovations across the global industry, and the gross contact area of grout. Shims will
have made the technique popular for low- to usually be pre-loaded by the self-weight of the
medium-rise structures. structure before the grout is installed and this load is
unlikely to be redistributed in the permanent state.
Form The position, contact area and material of shims may
Precast structural walls are usually upward of 150mm therefore be critical and should be defined. Ensuring
thick, with either one or two layers of reinforcement. shims have adequate edge distance so that they are
Thinner shear-resistant spandrel panels as little as away from the cover zone is essential.
75mm are possible but only in conjunction with
integral piers (Figure 7.4). Vertical joints tend to use discrete steel plates,
channels or angles bolted to adjacent panels across
Panel design the joint (Figure 7.6). Bolts can fix into cast-in
Away from the joints, individual panels act as threaded anchors (sometimes known as ‘ferrules’),
monolithic reinforced concrete elements and can be usually positioned to +5mm. Alternatively they can
designed largely in accordance with generic guidance pass through a sleeve in the wall (to be grouted once
for reinforced concrete. the bolt is installed) and fixed on the far wall face. As
these pieces of steelwork are exposed, fire proofing
The internal stress field in the permanent state will be and corrosion protection are often needed and must
dependent on both the direct loads acting on the be specified by the designer.
panel and the nature of the actions transferred via
any joints. Further reading: precast reinforced concrete wall
construction
Temporary (handling) load cases may need to be
considered in addition to those for the permanent The following texts are recommended sources of further
state. These must take account of the lifting points guidance on precast reinforced concrete wall construction:
which, where necessary, should be specified by the – Elliott, K.S. Precast concrete structures. Oxford:
designer. Butterworth-Heinemann, 2002
– Southcott, M.F. and Tovey, A.K. Tilt-up concrete buildings:
Joints design and construction guide: a comprehensive guide to
It is normal practice to use vertical dowel bars the benefits, economics and practicalities of tilt-up design
housed within cast-in corrugated sleeves for and construction in the UK. Crowthorne: BCA, 1998
horizontal joints (Figure 7.5). Once assembled, the
Precast wall
7.4 Hybrid precast in situ reinforced
Bleed hole concrete wall construction
Centrally placed
Grout feed corrugated grout
sleeve Hybrid reinforced concrete walls, sometimes referred
25mm
to as ‘twin wall’ construction, comprise of precast

cassettes stitched together on-site with loose
reinforcement and filled with in situ concrete to form
solid walls. They are intended as a replacement to
Dowel bar traditional in situ reinforced concrete walls, offering
Slab poured anchored into improved finish quality, reduced formwork and faster
once wall founding structure construction (Box 7.3).
installed
Their use is increasingly common on low- to medium-
Figure 7.5 Horizontal dowelled joint (shown to a foundation) rise developments and, like precast reinforced

the central void and connects the skins of a single

Box 7.3 Hybrid precast in situ reinforced concrete
cassette. It acts to maintain the form of the cassette
wall construction to the Francis Crick
in the temporary state and to resist bursting
Institute, London
pressures during placement of the in situ concrete. It
The Francis Crick Institute is approximately 150 70m on also acts as shear reinforcement.
plan with four independently stabilised blocks. Each block is
stabilised by shear walls, the majority of which are of hybrid Total wall thicknesses are usually upwards of 200mm
construction. These were constructed with the columns one but thicknesses less than 250mm are difficult to
floor at a time7.4. achieve7.5.
This photograph taken during construction shows a hybrid
Joints
wall alongside a traditionally formed in situ wall. This side-by-
Stitch bars transfer shear and tension between
side comparison illustrates the props necessary to the hybrid
cassettes. They are fitted in situ once the cassettes
walls but also the lack of formwork.
are positioned, often as a pre-formed cage for easy
handling. These are positioned in the void between
precast skins and have reduced effective depth
relative to bars cast in the precast cassettes. This has
no impact on pure tension but means that joints tend
to have less flexural capacity than the parent panels.
Stitch bars will usually have restricted lap lengths

owing to the geometrical conflict imposed by the cast
in lattice reinforcement (Figure 7.8). This incomplete
development has a detrimental impact on a joint’s
strength and also its ductility. It affects the pure
tension, flexure and shear capacities.
In-plane panel shear

Shear between the precast and in situ faces of a
single panel must be considered as this is the load
path transferring forces from the main bars within the
concrete wall construction, they are most suited to precast skin to the stitch bars across the in situ joints.
buildings with repetitive wall arrangements consisting
of planar elements that are consistent from storey to Shear must transfer between concrete cast at
storey. different times, and guidance on this is given in
BS EN 1992 Part 1-1 Clause 6.2.57.6. In most
Form situations, the shear stress will be small owing to the
The cassettes are made up of two skins of precast very large area over which the shear acts. However,
concrete, each typically 50–80mm thick, with cast in the stress becomes more significant as the shear
main vertical and horizontal reinforcement force increases; typically when the main
(Figure 7.7). Lattice reinforcement, usually welded to reinforcement bar size and wall thickness are large.
the main bars to form ‘lattice girders’, is cast in This tends to set an effective upper thickness limit on
during the prefabrication. This reinforcement spans hybrid walls at about 400mm.
Both the concrete interface and the lattice

Wall reinforcement within reinforcement can contribute to the overall shear
Lattice reinforcement precast shims resistance. Lattice reinforcement must be anchored
across cassette into the precast skins to make an effective
25mm tolerance gap
contribution. This anchorage is often via welds to the
Loose bars packed level with shims
main bars and supplementary requirements for
to joint Full welded steel reinforcement must be adhered to; for
lap designs to Eurocodes, see BS EN 100807.7.
Timber chocks In situ slab
The roughness of the internal surfaces to the precast
skins has the most significant influence on the
concrete surface shear interlock. These surfaces are
normally unformed during casting and can be of
varying roughness.
Precast projects into
slab cover zone Full Further considerations
lap Challenging details include slabs at different levels,
double height walls with horizontal joints that are
unrestrained, and panels supporting down-stand
beams. Although it is usual to have panels
terminating above and below floor slabs (Fig. 7.7),
pull out reinforcement, couplers or ferrules can be
cast in to accommodate slabs butting into the side of
In situ core
a panel. These can be fixed to the form when casting
the precast skin to achieve good accuracy (+5mm
typically).
Precast cassette
Where a panel is to support a down-stand beam, the
Figure 7.7 Hybrid wall cross-section showing a typical junction to an in situ slab effects of end moments, local crushing, wall buckling

Cage lowered
from above
Prefabricated
reinforcement cage
Lattice reinforcement
to the cassettes
Prefabricated cage
Lattice
reinforcement
Restricted lap length
Figure 7.8 Stitch detail at vertical joints
and instability should be considered. It is not

uncommon for such joints to dictate the width of the
in situ portion of the wall and hence the overall wall
thickness. Where this has significant impact, it may
be advantageous to have an in situ column between In situ
panels, possibly protruding from the plane of the wall
(Figure 7.9). Precast
Panels should be sized to allow erection on shims,

leaving a gap of approximately 25mm at the base
(Fig. 7.7). This allows tolerance on the set out, but
also provides the principal means of checking
whether the in situ concrete reaches the base of the
pour. At the top, internal panels should either project
a small amount (typically the slab cover depth) into
the slab soffit where the slab is in situ, otherwise they
should terminate short of precast floor units allowing
room for shims and grouting. Meanwhile a shadow In situ column
gap can be incorporated where the slab uses precast between wall cassettes
biscuit units as permanent formwork (Figure 7.10).
Figure 7.9 Down-stand beam connections
At perimeter walls, it is typical to extend the outer
precast skin to the slab finish level, using it as
formwork to the slab (Fig. 7.10).
In specifying a hybrid wall system, the detailed design

engineer must specify a rise limit rate to be adopted
by the contractor during concreting. This limit will be
dependent on capacity of each of the lattice
reinforcement and precast skins to resist the pressure
of the wet concrete and can be as low as 1m/hr7.5.
Where the wall is narrow, pokering the in situ
concrete can be near on impossible and the in situ
concrete should be specified as a workable or External skin acts as
self-compacting mix with plasticisers and/or formwork to the slab
maximum 10mm aggregate.
Precast biscuit Shadow gap

slab Precast wall
7.5 Timber and light gauge steel
cassette
‘platform’ frame construction
Note
‘Platform’ frame construction is suitable for low-rise Reinforcement within the precast wall cassettes and slab biscuit is omitted for clarity.
cellular buildings, practical up to a maximum of six
storeys7.8. The walls are hollow, comprising an Figure 7.10 Perimeter wall detail with shadow gap to biscuit slab

Figure 7.11 Platform frames undergoing construction
internal stud frame with face-fixed sheathing board provide buckling restraint to the sheathing (resulting in
(Figure 7.11). Studs can be either timber or light tension in the fixings) as well as transfer in-plane
gauge steel, while the sheathing is usually either a shear (Figure 7.12).
wood-based product (typically plywood or orientated
strand board) or plasterboard. Shear buckling can be avoided by limiting the
diaphragm slenderness, measured as the distance
Walls are typically installed a storey at a time, between stud fixings divided by the thickness of the
followed at each turn by the floor structure which sheathing. BS EN 1995 Part 1-1 Clause 9.2.4.3.2(7)7.9
creates a platform off which the next level is fixed. states that shear buckling of wood-based products
The floor structures are usually of similar lightweight can be disregarded when the slenderness ratio is less
timber construction to the walls, detailed to act as than or equal to 100. This generally defines limits on
horizontal diaphragms. Being lightweight, wind uplift the spacing of the studs (varying with the sheathing
and overturning are often critical and both the thickness). However, 400 or 600mm centres are
components and their connections will often need to typical in the UK7.10. Both spacings are suited to
resist anchorage tension. Robustness tie forces must standard 2.9 1.2m sheathing board and catered for
also be considered. with standard cavity insulation.
To act as shear walls, the sheathing must act as Where a sheathed wall panel is subject to
vertical diaphragms connected via the stud frame to significant out-of-plane forces (e.g. a wind pressure
the applied horizontal actions. Fasteners connecting on a façade), minor axis flexure may also influence
the sheathings to the studs must be adequate to the stud spacing or section. The sheathing may or
Figure 7.12 Shear forces in a sample of fixings caused by in- and out-of-plane actions

Plane sections Discontinuous strains

remain plane between sections
ε ε
Composite Non-composite
Figure 7.13 Composite and non-composite flexural behaviour
may not be designed to act compositely with the

Further reading: platform frame construction
stud frame (Figure 7.13). Where it is to act
compositely, it must be continuous between The following texts are recommended sources of further
supports (typically the floors) or fully lapped, and guidance on platform frame construction:
must have adequate fixings to transfer longitudinal – Institution of Structural Engineers and TRADA. Manual for
(flexural-borne) shear stresses to the stud. This the design of timber building structures to Eurocode 5.
construction is commonly referred to as ‘stressed London: IStructE, 2007
skin’ construction. – Grubb, P.J. et al. Building design using cold formed steel
sections: light steel framing in residential construction. SCI
While the stud frame will support the sheathing, it is Publication 301. Ascot: SCI, 2001
normal that the sheathing provides essential local
restraint to the stud frame (both lateral torsional and
in-plane Euler buckling restraint). In this way, the
sheathing and stud frame are generally co-dependent 7.6 Mass timber
(with neither being sufficient to withstand loads
independently).
Cross laminated timber (CLT) and glued laminated
Both wood-based products and plasterboard have (glulam) timber are both solid engineered products
unfavourable characteristics that limit their adequacy available in dimensions far exceeding those of sawn
as primary structural elements: timber7.12. As examples of ‘mass timber’, both are
– Wood-based products are combustible and their well suited to solid wall construction and can achieve
thin nature means they have limited fire resistance. far greater capacity than is possible from sheathed
Where critical, they generally require additional fire stud systems (Figure 7.14).
protection, usually in the form of plasterboard.
– Plasterboard loses integrity when exposed to
moisture. Moisture-resistant products are available,
however these only contain water repellent
additives that delay the loss of performance; they
do not completely eradicate it.
– Plasterboard is widely recognised as a non-
structural material that is often removed unwittingly
when buildings are remodelled.
On balance, wood-based products (with plasterboard

fireproofing) are recommended for structural
sheathing in preference to plasterboard.
Note that although plasterboard may provide fire

protection to wood-based sheathing, a wall
containing both materials should only rely on the
resistance provided by the wood-based product.
This, and further design recommendations for
sheathed partitions, is contained in the British
document PD 6693 Part 1, cited in the UK National
Annex to BS EN 1995 Part 1-1. This document7.11
gives non-contradictory complementary information
to BS EN 1995 Part 1-17.9. Figure 7.14 CLT walls under construction

Form
Box 7.4 Mass timber construction to The Forté,
The bending and shear stiffness of timber are
Melbourne
comparatively low. Hence deflection is often critical –
Completed in 2012, ‘The Forté’ (pictured) is Australia’s first with shear lag, in particular, of greater impact in
CLT building7.13. It is a ten storey residential block, of cellular timber systems than in those of steel or concrete.
layout with 128mm thick CLT shear walls. All CLT panels Where a concrete wall may easily have a height to
(485 tonnes in total, with the largest measuring 16.5 3m) length aspect ratio L/b ¼ 8, a timber wall should
were imported to Melbourne from a production centre in target 3 or 4 to efficiently achieve the necessary
Europe. At the time, this was favourable due to a strong stiffness. This may ultimately limit a building’s height
Australian Dollar7.14. on any given site, and puts greater pressure on the
plan layout to achieve an efficient wall arrangement.
Another noteworthy all-timber residential building is that at
24 Murray Grove, London7.15. This project was widely
Wall thicknesses are available up to 400mm as a
publicised for completing the construction of eight of the nine
single section7.19 but nothing close to this has been
floors in 27 days by a team of four people7.16.
used on projects to date. Acoustic and fire
performance can each increase the thickness beyond
that which is needed for strength or stiffness.
Plasterboard can be used as an effective means of
fire protection avoiding a sacrificial char thickness.
Meanwhile, dual wall systems with two panels
sandwiching a narrow air gap or acoustically
absorbent spacer can be favourable when acoustic
requirements govern.
Penetrations can be incorporated into timber walls by

either taking cut-outs from a single solid panel, or by
joining two independent panels with a lintel (usually a
glulam beam). Where a cut-out is made, the
significant stress concentrations at the corners of the
opening will often govern the entire wall panel design.
This can be overcome by bolting additional sections
across the highly stressed regions (Figure 7.15). The
performance of such bolstering is highly dependent
on the strength and stiffness of the fixings. The effect
of eccentricity should also be considered where
elements lap out-of-plane.
Material properties
Material properties can vary somewhat from native
sawn timber and may be taken from manufacturer
Recent investment and developments in CLT and data with reference to codes of practice or
glulam have enabled low- and medium-rise timber standardised test measures. Both CLT and glulam
structures that would traditionally have been can contain a mix of stress grades, with higher
constructed using masonry, steel or concrete grades used at the extremities responding to the
(Box 7.4). One driving force for this has been the flexural stress distribution.
sustainability agenda, discussed in the Institution’s
publication Building for a sustainable future: an Timber is not a linear elastic material, nor is it
engineer’s guide7.17. Other drivers include the isotropic; both of these characteristics should be
favourable characteristics of timber as a workable appreciated when setting up and justifying the
material, and the abundant/discounted supply of accuracy of an analysis model. In the case of CLT, the
timber in some geographic regions7.18. behaviour depends on the orientation and build-up of
Face-fixed strengthening
to the lintel Lintel beam
Opening cut out

of single panel
Discrete wall panels
Figure 7.15 Openings and lintel strengthening

the laminates. Most CLT has a symmetrical profile

with outer laminates in the same orientation.
Joints
Horizontal joints in solid timber construction are best Floor
located immediately above the floor bearing, with Joint in
walls butt connected. This avoids axial loads passing wall immediately
through the cross-grain of floor elements. Slabs may above floor slab
be supported off steel angle bearers face-fixed to the
wall panels (Figure 7.16). Fire protection to
Moment from
surfaces including the
eccentric floor bearing
Vertical panel-to-panel joints are normally lapped steel connection pieces
taken by wall below Steel bearing angle
where the panels are co-planar, or butt jointed where
with fixings for robustness
panels are orthogonal. In both instances it is common
to have two lines of screws positioned as shown in Wall
Figure 7.17. Both details must have screws adhering
to minimum edge distances and spacing as defined
in codes of practice. Additionally, where using the Figure 7.16 Floor to wall joint
butt connection, screws must anchor into the vertical
laminates only, not the horizontal end grain.
in a range of sizes, usually larger than bricks and a

multiple of a standard brick module including mortar
(i.e. 225 75mm). They are also available in a range
of densities that correlate to the strength.
Brick and block sizes vary internationally and sizes

should always be used that are appropriate to the
region. This should be something the project architect
is mindful of, but the sizes will likely influence
structural details (e.g. the setting out of supporting
beams and/or strip foundations).
The weight and ease of handling should also be

considered when choosing a block size. CIRIA guide
C662 recommends blocks weighing over 20kg
Figure 7.17 Vertical joints between co-planar and orthogonal should be avoided where manual handling is
panels intended7.20.
Mortar is used in all but ‘dry stone wall’ construction

to provide nominal tensile and shear adhesion
7.7 Loadbearing masonry between adjacent units. Most modern construction
uses cement-based mortar, although lime-based
mortar remains necessary for the repair and/or
Loadbearing masonry is a traditional approach to wall adaption of many historic buildings where it was used
construction that remains popular in many regions. in the original construction. Lime-based mortar can
Clay bricks, concrete blocks and natural stone may generally accommodate a greater degree of
be used as constituent ‘units’. movement than cement mortar but both are
ultimately brittle.
The size of the units is critical in differentiating
masonry from the precast systems discussed in Form
previous sections. A wall may be made up of a single or multiple
‘skins’, each skin being a solid arrangement of the
Qualities of masonry include its reasonable brick or block units. The packing arrangement
compressive strength, durability, fire resistance, ability within a skin is often referred to as the ‘bond’. Most
to insulate, aesthetic quality, and ease of handling skins are a single brick or block unit thick and use a
during erection. A disadvantage is the labour- stretcher bond (Figure 7.18). This leads to a macro
intensive in situ construction process which is a wet
trade susceptible to weather conditions. Further
disadvantages of traditional unreinforced masonry
include its brittle nature and limited tensile capacity.
These latter characteristics tend to make
unreinforced masonry highly sensitive to movement
and also prone to sudden failure. These
characteristics can, however, be lessened with the
use of reinforcement.
Masonry units and mortar

Manufactured clay bricks and concrete blocks are
most widely used and are much cheaper than natural
stone. Both are available in standard sizes. In the UK,
standard bricks are 215 65 102.5mm and are
used with a 10mm mortar joint. Blocks are available Figure 7.18 Masonry stretcher bond

behaviour that tends to be non-isotropic with on the floor bearing detail; guidance is given in
differing horizontal (bed joint) and vertical Section 5.3.6 of the Institution’s Manual for the
characteristics. design of plain masonry in building structures to
Eurocode 67.21.
Skins can be standalone or form part of a cavity wall
system in which two skins are spaced apart and tied Provided the ultimate limit state is satisfied,
at regular intervals. Cavity wall construction can be serviceability checks of individual panels are seldom
favourable for insulation and acoustic isolation. Wall needed and are largely omitted from design guidance.
ties should always be used in cavity construction to A common check is to limit overall inter-storey sway.
improve robustness. They also act to enhance the The Institution’s Manual for the design of building
buckling capacity of a wall, increasing the axial load structures to Eurocode 17.22 recommends that both
carrying capacity above the sum of the component total sway and inter-storey drift should not exceed
skins. height/500.
Reinforced masonry uses steel bars or wire placed Movement

horizontally and/or vertically in one of the Bricks and blocks both exhibit dimensional instability
configurations shown in Figure 7.19. Bed-joint that can continue over a number of years. This is
reinforcement is generally restricted to bars or wire often best accommodated with the inclusion of
(up to 6mm diameter). Meanwhile, reinforcement vertical movement joints. The spacing of joints is
within the cores of hollow units or the cavity of a dependent on the nature of the unit and mortar but
cavity wall can be standard reinforcement as used for spacing of 12–15m for clay bricks and 6–8m for
reinforced concrete. The limiting bar size is generally concrete blocks is typical.
governed by the compressive strength of the
masonry unit. It should be noted that any It should be noted that fired clay bricks tend to
reinforcement within hollow cores or a cavity must be expand while concrete blocks tend to shrink; both by
grouted to be effective. as much as 0.5mm/m7.21. Cavity walls with an
external brick/internal block skin configuration should
Design have joints staggered to minimise the stresses
Axial compression in loadbearing masonry walls developing in the cavity ties.
enhances both the shear and bending capacities by
providing a pre-compression. This makes up for the Damp proof course
lack of tensile strength. However, it also makes walls External masonry walls often require a damp proof
vulnerable to buckling and out-of-plane failure is often course or membrane positioned in a bedjoint close to
critical. ground level. Such membranes impede the passage
of moisture up into the wall but can have variable
As such, out-of-plane actions at the ultimate limit structural characteristics. In particular, they can cause
state tend to dominate the design of shear walls. This a slip plane that has a detrimental impact on the
is true even where the out-of-plane actions are shear capacity of a wall. Little guidance is given in
several orders of magnitude less than the in-plane BS EN 1996 Part 1-17.23 on the friction, interlock
actions. and/or adhesion that can be assumed, and specific
product data should generally be sought from the
Minor axis slenderness is generally used as a manufacturer.
governing parameter in the design of unreinforced
walls (Section 2.8). This is a function of the wall’s Robustness
effective thickness, height and/or horizontal span. Walls should be tied to adjacent structural elements
An axial load eccentricity should be considered in order to avoid collapse. Guidance is given in the
additional to all lateral actions to determine the Institution’s Practical guide to structural robustness
critical design case. This eccentricity is dependent and disproportionate collapse in buildings7.2.
Reinforcement Cavity
wall ties
Bed-joint reinforcement Vertical and/or horizontal Cavity wall reinforcement

reinforcement in hollow units
Figure 7.19 Reinforced masonry

Further reading: introduction to masonry Box 7.6 Composite steel plate diaphragm walls to
China World Trade Centre, Beijing
The following guides provide a more extensive introduction to
masonry wall construction and are additional to the sources Composite steel-reinforced concrete walls extend through the
referenced within the main text: basement to the 16th floor of this 74 storey (330m) tower
– ‘Introduction to masonry. Technical Guidance Note’. The providing greater robustness, ductility and stiffness than is
Structural Engineer, 91(6), June 2013, pp24-26 achievable with either traditional reinforced concrete or steel
– Curtin, W.G. et al. Structural masonry designers’ manual. plate shear walls. Each wall has a web plate, welded to
3rd ed. Oxford: Blackwell, 2006 boundary elements (steel beams and columns). These steel
– Roberts, J.J. and Brooker, O. How to design masonry assemblies, complete with shear studs attached to the plate,
structures using Eurocode 6 Part 2. TCC/03/36. Revision 2. were encased within in situ reinforced concrete forming a
London: The Concrete Centre, 2013. Available at: http:// ‘composite special plate wall’7.26.
www.eurocode6.org/Published%20support%20material.htm
[Accessed: 13 January 2015]
7.8 Steel plate diaphragm walls in steel

framed buildings
Steel plate diaphragms are suited to medium- and

high-rise structures where they have been proven to
match or outperform more conventional braced
frames and reinforced concrete shear walls.
The plates, together with the surrounding frame, are

often lighter than equivalent reinforced concrete walls
and this can lead to reduced foundation loads. They
are also relatively thin (typically 5–12mm) and can
minimise the loss of lettable floor area. These are
advantages that led to the system being used for In very tall buildings, each of the following must be
Jinta Tower, Tianjin7.24. carefully considered: stiffness of the plate, axial
shortening and stiffness of the frame elements,
To their detriment, the steel plates require almost fatigue due to cyclic oscillation, and buckling at
continuous fixing to the surrounding frame elements. service loads. The flexural stiffness of the bounding
This generally necessitates either a large number of frame elements must be such that in-plane forces
bolts or site welds. along the elements do not cause significant
deflection so as to impact the load path through the
Despite some use in the UK (Box 7.5), steel plate plate. Minimum stiffnesses for each of the columns
diaphragms have been most widely used in China, and beams are defined in both the American and
Japan, and North America, in both new Canadian codes.
developments and as seismic retrofits to existing
structures. The American standard AISC 341-107.26 Finally, steel plates can be cast within reinforced
and Canadian standard CSA S16-147.27 are the most concrete walls to produce composite sections that
widely used sources of guidance; both give specific outperform traditional reinforced walls. This system
criteria for the design of these systems where subject was used within the China World Trade Centre,
to dominant seismic loads. Beijing7.28 (Box 7.6).
Box 7.5 Steel plate diaphragm walls to Embankment Place, London

This over-site development positioned above the platforms to Charing Cross Station has much of the superstructure hung to ensure
a column free space at platform level. This concentrates a large percentage of the equivalent horizontal force (referred to as
‘notional horizontal force’ at the time of the design) at the top of the cantilevering stability cores.
To cater for this force and provide sufficient stiffness so that PD effects are not overwhelming, a welded steel plate shear wall is
incorporated through the upper storeys of the superstructure7.25.

7.9 References 7.18 Fast, P. Unusual hybrid timber-steel structures

[webinar]. 19 May 2010. Available at: https://istructe.
adobeconnect.com/_a848983388/p83578900
7.1 Corus. Bi-Steel design and construction guide. [Accessed: 13 January 2015]
Scunthorpe: British Steel, 1999
7.19 Sutton, A. et al. Cross-laminated timber: an
7.2 Institution of Structural Engineers. Practical guide to introduction to low-impact building materials. BRE
structural robustness and disproportionate collapse in Information Paper IP 17/11. Watford: IHS BRE Press,
buildings. London: IStructE, 2010 2011
7.3 Elliott, K.S. and Jolly, C.K. Multi-storey precast 7.20 Ove Arup & Partners and Gilbertson, A. CDM2007 –
concrete framed structures. 2nd ed. Chichester: Construction work sector guidance for designers. CIRIA
Wiley-Blackwell, 2013 C662. 3rd ed. London: CIRIA, 2007
7.4 Partridge, R. et al. ‘The Francis Crick Institute’. The 7.21 Institution of Structural Engineers. Manual for the
Structural Engineer, 92(3), March 2014, pp10-19 design of plain masonry in building structures to
7.5 Whittle, R. and Taylor, H. Design of hybrid concrete
buildings: a guide to the design of buildings combining 7.22 Institution of Structural Engineers and Department for
in-situ and precast concrete. CCIP-030. Camberley: Communities and Local Government. Manual for the
The Concrete Centre, 2009 design of building structures to Eurocode 1 and basis
of structural design. London: IStructE, 2010
concrete structures – Part 1-1: General rules and rules 7.23 BS EN 1996-1-1: 2005 þ A1: 2012: Eurocode 6 –
for buildings. London: BSI, 2004 Design of masonry structures – Part 1-1: General
rules for reinforced and unreinforced masonry
7.7 BS EN 10080: 2005: Steel for the reinforcement of structures. London: BSI, 2012
concrete – Weldable reinforcing steel – General.
London: BSI, 2005 7.24 Sarkisian, M. et al. ‘World’s tallest steel shear walled
building’. CTBUH Journal, 1, 2011, pp28-33. Available
7.8 UK Timber Frame Association. ‘Timber Engineering at: http://ctbuh.org/LinkClick.aspx?fileticket=e2KgCTBu8
Notebook series. No. 3: Timber frame structures – yw%3D&tabid=3096&language=en-US/&_sm_au_=
platform frame construction (Part 1)’. The Structural iVVR0qQLjsLrMLsN [Accessed: 13 January 2015]
Engineer, 91(5), May 2013, pp26-32
7.25 Barrie, M. and Weston, G. ‘Embankment Place:
7.9 BS EN 1995-1-1: 2004 þ A1: 2008: Eurocode 5: building over Charing Cross Station’. The Structural
Design of timber structures – Part 1-1: General – Engineer, 70(23/24), 8 December 1992, pp405-411
Common rules and rules for buildings. London: BSI, 2009
7.26 AISC 341-10: Seismic provisions for structural steel
7.10 Institution of Structural Engineers and TRADA. Manual buildings. Chicago, Il: AISC, 2010. Available at:
for the design of timber building structures to www.aisc.org/2010sp [Accessed: 13 January 2015]
7.27 CSA S16-14: Design of steel structures. Toronto,
7.11 PD 6693-1:2012: Recommendations for the design of Ontario: Canadian Standards Association, 2014
timber structures to Eurocode 5: Design of timber
structures – Part 1: General – Common rules and 7.28 China World Tower. Available at: http://www.
rules for buildings. London: BSI, 2012 skyscrapercenter.com/beijing/china-world-tower/379/
[Accessed: 13 January 2015]
7.12 UK Timber Frame Association. ‘Timber Engineering
Notebook series. No. 2: Engineered wood products and
an introduction to timber structural systems’. The
Structural Engineer, 91(4), April 2013, pp42-48
7.13 Powney, S. Melbourne marvel. Timber & sustainable

building. 2012. Available at: http://www.timber-
building.com/features/melbourne-marvel/ [Accessed:
13 January 2015]
7.14 XE Currency Charts (AUD/EUR). Available at: http://

www.xe.com/currencycharts/?from ¼ AUD&to ¼ EUR&
view ¼ 10Y [Accessed: 13 January 2015]
7.15 MGB Architecture and Design. The Case for tall wood
buildings. 2013. Available at: www.nzwood.co.nz/
canterbury-rebuild/news-post-tag-test/ [Accessed:
13 January 2015]
7.16 KLH UK. Murray Grove – Timelapse sequence. 2009.

Available at: http://www.klhuk.com/news/stadthaus-
timelapse.aspx [Accessed: 13 January 2015]
7.17 Institution of Structural Engineers. Building for a

sustainable future: an engineer’s guide. London:
IStructE, 2014

8 Shear infill panels
8.1 Introduction including:

– The panel aspect ratio
– The disposition of openings
This chapter gives a brief introduction to systems that – The robustness of the system
provide stability without contributing to the vertical
load resistance. These systems are typically made up It should be noted that ‘non-elastic strains’
of discrete elements and used almost exclusively in include creep and shrinkage, and dimensional
framed buildings. variation caused by thermal loads and/or moisture
content.
The focus is on common considerations that are not
specific to any particular panel or frame material. Load paths
However, some specific commentary on masonry infill To provide resistance, an infill panel must form part of
panels is included which supplements the guidance a load path that locks together elements of a frame.
given in Section 7.7. Connections must be stiff, strong and without risk of
premature brittle failure (Box 8.2).
8.2 Common characteristics of infill

Box 8.2 Non-structural partitions and cladding
systems
When detailing non-structural partitions and cladding,
engineers must appreciate that it is stiffness and not strength
Both the frame and the infill can be of any suitable that defines the distribution of forces within an elastic
material (timber, masonry, concrete, steel, composites system. To avoid failure, any components that lock two points
or glass). Consequently there can be huge variation in must have either strength or ductility that is adequate for the
the specifics of the design. However, common points anticipated strains.
to be considered include:
These properties are seldom achievable in partitions and
– The interfaces and load paths between the frame
cladding. Indeed, the majority of modern buildings of three or
and the infill elements
more storeys adopt non-structural cladding and partitions
– Any eccentricity between the centrelines of the
installed, with connections that allow movement deliberately
frame and that of the infill
to prevent unintentional load paths forming. This approach
– The strain compatibility of the different materials/
allows these elements to be of low strength and does not
components (including both elastic and non-elastic
unduly penalise brittle materials such as glass, terracotta or
strains, considering the orientation of non-isotropic
stone.
materials where applicable)
– The chemical compatibility of the different materials Removed from the load paths, engineers must not assume
(e.g. bi-metallic corrosion) considering the materials that any resistance to the global system is contributed by
of the frame, the infill and the fixings these elements.
– The capacity and stiffness of the fixings and their
anchorage into the frame/infill
– The tolerance and sensitivity to workmanship and/ Mechanical fixings between the panel and the frame
or initial dimensional deviations are always recommended and are often essential to
– The temporary stability of the frame in advance of achieve robustness requirements for new or altered
the infill installation structures. However, a traditional means of
– The long-term durability and design life of the infill, achieving a load path was to create an infill panel
and requirement for maintenance that was tightly packed between columns without
– The long-term safety of the structure and the fixings. This approach provides no robustness and
heightened risk of ill-planned alterations (Box 8.1). is particularly vulnerable in extreme loading
events8.2.
These are in addition to the more standard attributes
With or without fixings, panels will usually bear on the
Box 8.1 Safety of structural shear infill panels slab or beam below and include a deflection head to
accommodate vertical differential movement of the
The use of elements that are traditionally considered floor structure above (the latter to ensure the infill
‘secondary’ or ‘non-structural’ for primary stability systems is does not attract vertical load). Thus, a gap at the top
not without serious risk. Where used, the designer must of an existing wall is not in itself an indicator that a
make clear the importance of these elements and ensure that wall panel is non-structural.
this information is adequately recorded and available to guide
future refurbishment and remodelling. Geometrical compatibility of the infill and frame
In the UK, such notices should be included on construction materials, together with relative stiffnesses and
drawings and in the building safety file under the terms of tolerance, are critical. The thermal and chemical
the CDM Regulations8.1. characteristics of both the frame and the infill can be
It is recommended that structural walls are labelled as such important in this regard. For porous or curing
on the architect’s drawings. This is in addition to the wall’s materials (including clay bricks, concrete and timber)
disposition being specified on the structural engineer’s the designer must appreciate the time-based
drawings. differential movement that can occur, and the impact
this may have on the performance.

8.3 Shear infill panels
and/or replacement strategy. This should

accompany the design data and be explicitly
communicated to, and accepted by, the client and
building manager.
In determining the design life, the risk of

degradation should be assessed using a
structured risk assessment approach. This should
consider all potential triggers and/or accelerators
specific to the project. Human behaviour may
need to be considered here and accidental,
malicious and uninformed/negligent actions can
heighten certain risks. Such an assessment is not
unique to infill panels. However, these risks can
become pronounced as the material can be easily
Figure 8.1 Indicative frame reactions showing the impact of damaged without heavy tools or significant force.
infill panels on a regular column grid Thus the risks are generally higher with
lightweight infill systems. Other triggers can
include an isolated event such as a flood.
Meanwhile environmental conditions
Failure mechanisms (temperature, humidity, chloride exposure, etc.)
All failure mechanisms shown in Fig. 2.5 apply locally can act as accelerators.
to the individual panels and their interfaces with the
frame. Forces will tend to transfer through a panel via Construction
diagonal compression struts and/or tension ties. The Structural infill panel systems can have significant
compatibility of these should be considered, as any impact on the construction programme and
strain in one diagonal will tend to cause an opposite temporary bracing to the frame is often essential
strain in the other. ahead of the panels being installed. Although the
design of the temporary works may not be the
Diagonal compression in the panel cannot be responsibility of the design engineer, the designer
transferred to the frame by direct bearing where there must make it clear to the contractor that the frame
is a deflection head. Instead, shear fixings are generally will be compromised or inadequate before the
needed to maintain vertical equilibrium. These fixings, panels are installed. This may not be immediately
together with the surrounding panel and frame, should clear to the contractor who may interpret the frame
each be checked for combined shear with to be a moment frame. As a rule, the designer must
compression. Usually the same fixings will also need to always state which elements are providing stability,
accommodate shear with tension under load reversal. and the dependencies of the different systems.
They may also need to advise the contractor of
Beyond the fixings, the frame elements that surround temporary loads and/or detail permanent works
the panels will be subject to forces resulting from the with connections for temporary elements. A design
stiffness of the panels. These forces are similar in risk assessment can formalise and assist this
nature to those seen when using eccentric bracing process, helping the designer to identify and
and it is essential they are not overlooked. Ultimately manage construction risks.
they must be tacked down through the frame and
subsequently the foundations (Figure 8.1).
Durability and maintenance 8.3 Masonry infill panels

Structural infill panels should be designed for the
same design life and probabilities of failure as the
elements that make up the structural frame. Masonry infill panels are common in low- and
medium-rise framed buildings of the last century
Where this is impractical it may be acceptable to (Figure 8.2). As a consequence they are a common
devise a clear and implementable maintenance feature of buildings undergoing remodelling.
Figure 8.2 Masonry infill panels within reinforced concrete frames (masonry panels to car park are creatively decorated to be
disguised in building’s elevation)

Shear infill panels 8.3
Forces on the frame can cause bending

Separation of and shear in the frame elements
frame from wall






 Panels with openings:
 require load path into
 the beams and generally

Figure 8.3 Infill panels under load  cannot have a deflection head





When a masonry infill is loaded in shear, separation 
of the infill from the frame tends to occur to all but 
the compression corner regions (Figure 8.3). The
infill subsequently acts as a diagonal compression Solid panel with vertical shear
strut preventing sway deformation of the frame. connections to the columns can
Stafford-Smith and Riddington8.3 define the have a movement joint (a deflection
effective area of the strut Astrut as that given in head) to the beam above
Equation 8.1:
Astrut ¼ 0.1Ld t . . .Eqn 8.1
where: Reactions transfer down

Ld is the diagonal length of panel through the frame
t is the thickness of the panel
Figure 8.4 Masonry panel struts
The direction of the strut in a given panel will depend
on the direction of the shear, which is usually
reversible under lateral loads. Hence all four corners
and each of the two diagonals should be detailed to
provide an adequate load path. Doors and other
openings can be accommodated provided that
adequate compression struts can still form. However,
these openings will often change the strut angles,
decreasing the strut area and increasing the stress
within the panel. The revised strut pattern may also
change the forces acting on the frame elements
(Figure 8.4).
Dominant failure modes for infill panels include shear

along the bedding planes, or crushing at the
compression corners. Out-of-plane buckling is
typically avoided by adhering to maximum
slenderness ratios (Section 7.7).
Note that masonry infill panels are not subject to the

pre-compression that gives loadbearing masonry
much of its flexural and shear strength.
Figure 8.5 Confined masonry construction
It should also be noted that masonry infill panels that
do not carry vertical load differ in construction and
behaviour from confined masonry walls that do.
These systems can be similar in appearance The wall is
(Figures 8.2 and 8.5). The notable difference is the constructed
order of construction: confined masonry has the after the frame Frame (cast
frame constructed after the wall is built while an infill in situ )
is built to fit a pre-formed frame (Figure 8.6). completed after
the wall panel
Confined masonry has been omitted from this Guide; is built
for more information on confined masonry walls refer
to the EERI Seismic design guide for low-rise
confined masonry buildings8.4.
Infill panel Confined masonry
Figure 8.6 Infill panels and confined masonry

8.4 Shear infill panels
Further reading: masonry infill panels

The following text is recommended as a source of further
guidance on masonry infill panels:
– Elliott, K.S. and Jolly, C.K. Multi-storey precast concrete
framed structures. 2nd ed. Chichester: Wiley-Blackwell,
2013
It should be noted that this book provides guidance on both
masonry (brickwork) and precast concrete shear infill panels;
the latter is not discussed herein.
8.4 References
8.1 HSC. Managing health and safety in construction:

Construction (Design and Management) Regulations
2007: Approved code of practice. L144. Norwich: HSE
Books, 2007
8.2 Taucer, F. et al. The 2007 August 15 magnitude 7.9

earthquake near the coast of Central Peru: EEFIT field
mission, 5-12 September 2007/Final report. Available
at: http://www.istructe.org/webtest/files/e3/e39c4dd5-
0f03-446b-8ad2-bf9cbf796ee4.pdf [Accessed:
13 January 2015]
8.3 Stafford-Smith, B. and Riddington, J.R. ‘The Design of

masonry infilled steel frames for bracing structures’.
The Structural Engineer, 56B(1), March 1978, pp1-7
8.4 Meli, R. et al. Seismic design guide for low-rise

confined masonry buildings. Oakland, CA: EERI, 2011.
Available at: http://www.world-housing.net/wp-content/
uploads/2011/08/ConfinedMasonryDesignGuide82011.
pdf [Accessed: 13 January 2015]

Stability of Buildings Part 3 Shear Walls

Uploaded by

Copyright:

Available Formats

Stability of Buildings Part 3 Shear Walls

Uploaded by

Document Information

Original Description:

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

Stability of Buildings Part 3 Shear Walls

Uploaded by

Copyright:

Available Formats

What are shear walls and how do they provide stability?

What are shear walls and how do they provide stability?

What are the failure modes of masonry infill panels?

What are the failure modes of masonry infill panels?

Stability of buildings

All other photographs and all hand illustrations: A Gardner

Published by The Institution of Structural Engineers,

First published 2015

# 2015 The Institution of Structural Engineers

iv The Institution of Structural Engineers Stability of buildings Part 3

The following notation is used for hand drawn figures:

An applied bending moment

An out-of-plane surface pressure

Structural lintel beams across wall openings

The Institution of Structural Engineers Stability of buildings Part 3 v

Glossary iv 7 Non-monolithic shear wall construction 40

4 Elastic theory of thin-walled sections 16

5 Modelling and analysis 21

6 Monolithic reinforced concrete shear wall

The Institution of Structural Engineers Stability of buildings Part 3 iii

The Institution of Structural Engineers Stability of buildings Part 3 1

2 The Institution of Structural Engineers Stability of buildings Part 3

Note that the checklist given in Fig. 1.2 is intended to

In line with Part 1 and Part 2 of this series,

The Institution of Structural Engineers Stability of buildings Part 3 3

2.1 Form and configuration

There are three basic types of structure that include

Within a building, isolated or coupled walls can be

Note that, in this context, ‘resistance’ should be

Fig. 2.1 shows the three structural types idyllically in

Figure 2.1 Structural forms: (a) cellular, (b) framed with

Figure 2.2 Wall systems: (a) isolated, (b) coupled

4 The Institution of Structural Engineers Stability of buildings Part 3

components, local strengthening with additional

In multi-walled statically indeterminate structures, the

Note that any movement joints form boundary

Shear 2.3 Failure mechanisms

Framed bracing / Shear walls

(a) (b) (c) (d)

The Institution of Structural Engineers Stability of buildings Part 3 5

Steel diaphragms are used in substitution for

Depending on this stiffness, the performance of a

Elevation: It is important to appreciate that coupling enhances

The coupling effect is illustrated in Figure 2.6 for a

2.7 Buckling and buckling restraint

Shear walls and panels subject to in-plane shear and/

6 The Institution of Structural Engineers Stability of buildings Part 3

y y y Shear walls Moment frame

– Lateral torsional (out-of-plane) buckling due to an adopted by BS EN 19962.5 and BS EN 19922.6

The minor axis may similarly be unbraced, in which

over major axis. A single central core is a common

Alternatively, an individual wall may be restrained at

Slenderness is a measurement used to combat the

buckling in members subject to axial compression. A

The ‘slenderness ratio’ of a wall is traditionally

The Institution of Structural Engineers Stability of buildings Part 3 7

overall height of the wall L. The effective height Le is

Slenderness of restrained wall systems and

8 The Institution of Structural Engineers Stability of buildings Part 3