ED005 Design of Light Steel Sections To Eurocode 3

ED005 Technical Report: Design of Light Steel Sections to Eurocode 3
Discuss me ...
TECHNICAL REPORT
SCI DOCUMENT ED 005
Design of Light Steel Sections

to Eurocode 3
This material is copyright - all rights reserved. Use of this document is subject to the terms and conditions of the Steelbiz Licence Agreement
A G J Way MEng, CEng, MICE

M D Heywood MEng, PhD, CEng, MICE
Issued (in electronic format only) by:

The Steel Construction Institute
Silwood Park
Ascot
Created on 23 February 2012
Berkshire, SL5 7QN
01344 636525
www.steel-sci.org
Discuss me ...
SCI (The Steel Construction Institute) is the leading, independent provider

of technical expertise and disseminator of best practice to the steel
construction sector. We work in partnership with clients, members and
industry peers to help build businesses and provide competitive advantage
through the commercial application of our knowledge. We are committed to
offering and promoting sustainable and environmentally responsible
solutions.
Our service spans the following areas:
Membership Consultancy
Individual & corporate membership Development
Product development
Advice Engineering support
Members advisory service Sustainability
Assessment
Information SCI Assessment
Publications Specification
Education Websites
Events & training Engineering software
SCI Technical Reports
Technical Reports are intended for the cost effective dissemination of the latest research
results and design guidance, as and when they become available, or as specialist
documents for further discussion.
A Technical Report may serve as the basis for a future SCI publication or the updating of
an existing SCI publication.
 2012 The Steel Construction Institute
Apart from any fair dealing for the purposes of research or private study or criticism or review, as permitted under the
Copyright Designs and Patents Act, 1988, this document may not be reproduced, stored or transmitted, in any form or
by any means, without the prior permission in writing of the publishers, or in the case of reprographic reproduction only
in accordance with the terms of the licences issued by the UK Copyright Licensing Agency, or in accordance with the
terms of licences issued by the appropriate Reproduction Rights Organisation outside the UK.
Enquiries concerning reproduction outside the terms stated here should be sent to The Steel Construction Institute, at
the address given on the title page.
Although care has been taken to ensure, to the best of our knowledge, that all data and information contained herein
are accurate to the extent that they relate to either matters of fact or accepted practice or matters of opinion at the time
of publication, The Steel Construction Institute, the authors and the reviewers assume no responsibility for any errors in
or misinterpretations of such data and/or information or any loss or damage arising from or related to their use.
SCI Document Number: SCI ED 005

ED005V01.doc ii
Discuss me ...
FOREWORD
Light gauge cold-formed steel members (generally referred to as light steel sections)
are commonly used in a range of building types as secondary steelwork (e.g. purlins
and cladding rails in industrial buildings) and as the primary load-bearing elements in
light steel frames (e.g. in residential buildings). They may be used as individual
structural members (e.g. floor joists) or as part of a structural frame.
In Eurocode 3 Design of steel structures, design rules for such members in the UK are
given in several separate Parts, notably BS EN 1993-1-1, BS EN 1993-1-3 and BS EN
1993-1-5, each together its National Annex. This Report has been prepared to aid
designers in the application of the various rules in those documents, particularly in the
applications in wall panels and floors.
The report has been prepared by Andrew Way and Martin Heywood, both of SCI. The
worked examples have been independently checked by Stephen Napper of
Stephen Napper Associates Ltd, and his involvement is gratefully acknowledged.
The work leading to this Report was funded by Tata Steel Strip Products UK.
ED005V01.doc iii
ED005
Discuss
me
Technical
...
Report:
Design
ED005V01.doc
of
Light
Steel
Sections
iv
to
Eurocode
3
Discuss me ...
Contents
Page No
FOREWORD iii
SUMMARY vii
1 INTRODUCTION 1
1.1 Light steel sections 1
1.2 Eurocodes 2
1.3 Design to Eurocode 3 3
2 SECTION PROPERTIES 5
2.1 Core steel thickness 5
2.2 Mid-line theory 6
2.3 Corner radii 6
3 LOCAL BUCKLING 9
3.1 Effective width concept 9
3.2 Eurocode calculation procedure for unstiffened plane elements 10
4 DISTORTIONAL BUCKLING 12
4.1 Design of stiffened sections 12
4.2 Eurocode calculation procedure for stiffened elements 14
5 DESIGN OF COMPRESSION MEMBERS 17
5.1 Design issues 17
5.2 Eurocode calculation procedures 17
6 DESIGN OF MEMBERS IN BENDING 20

6.1 Laterally restrained members 20
6.2 Lateral-torsional buckling 22
6.3 Serviceability 23
7 FRAME DESIGN 25
7.1 Frame stability 25
7.2 Structural robustness 26
7.3 Frame anchorage 27
8 WORKED EXAMPLES 28
8.1 Gross section properties for a cold-formed lipped C section 29
8.2 Effective section properties for a lipped C section in compression 38
8.3 Effective properties for a lipped C section in major axis bending 43
8.4 Effective properties for a lipped C section in minor axis bending 48
8.5 Design of a wall stud in a light steel frame building 52
8.6 Design of a floor joist (restrained) in a light steel frame building 60
8.7 Design of a floor joist (unrestrained) in a light steel frame building 64
8.8 Design of a lattice floor truss in a light steel frame building 67
9 REFERENCES 78
ED005V01.doc v
This material is copyright - all rights reserved. Use of this document is subject to the terms and conditions of the Steelbiz Licence Agreement Discuss me ...
ED005V01.doc
vi
Discuss me ...
SUMMARY
This Technical Report provides guidance in accordance with Eurocode 3 for the design
of steel sections as used in light steel framing applications.
The guidance includes a brief introduction to Eurocodes and light steel sections,
followed by detailed design guidance. Since light steel members are especially prone
to local buckling, the design consequences of this behaviour are dealt with in depth,
notably the calculation of effective cross section properties. Design guidance for
members in compression and members in bending is also given.
Eight worked examples are provided to illustrate the application of the design rules to
practical building applications. The examples include the calculation of gross and
effective section properties, the design of sections subject to bending and compression
and serviceability design of light steel floors.
ED005V01.doc vii
This material is copyright - all rights reserved. Use of this document is subject to the terms and conditions of the Steelbiz Licence Agreement Discuss me ...
ED005V01.doc
viii
Discuss me ...
1 INTRODUCTION
Light gauge cold-formed steel members (generally referred to as light steel sections)
are commonly used in a range of building types as secondary steelwork (e.g. purlins
and cladding rails in industrial buildings) and as the primary load-bearing elements
in light steel frames (e.g. in residential buildings). They may be used as individual
structural members (e.g. floor joists) or as part of a structural frame. Light steel
members are often prefabricated off-site to form wall panels, floor cassettes or
volumetric modular units, but are equally suited to stick build applications.
1.1 Light steel sections

For the purposes of this report, the term ‘light steel’ refers to galvanized cold-formed
steel sections with a maximum gauge (thickness) of 4 mm, although gauges from
1.2 mm to 2.0 mm are the most common for light steel framing applications. For
purlins and cladding rails, the thickness generally lies in the range 1.4 mm to
3.2 mm. Light steel members are usually cold-formed from hot-dip galvanized strip
steel, which is supplied to the section manufacturers as pre-galvanized coil
conforming to BS EN 10346: 2009[1]. For light steel framing applications, the
commonly used grades of steel are S350, S390 and S450, while purlin and cladding
rail products tend to use S390 and S450.
Lipped C sections are the most common section shape for light steel framing
applications, including wall studs and floor joists. The C section shape is simple to
roll and widely manufactured, while the lip provides additional stiffness to the flange
and increases the stress at which local buckling occurs in the flange. Depths
commonly range from 70 mm to 120 mm for wall studs and from 120 mm to 250 mm
for floor joists. Purlins are typically made from Zed or sigma sections with depths
ranging from 140 mm to 300 mm.
A typical light steel frame is shown in Figure 1.1. The photograph shows a
prefabricated volumetric module, but similar framing arrangements are also used in
panellised and stick-built construction.
Figure 1.1 Load bearing light steel frame in a volumetric module
1
Discuss me ...
1.2 Eurocodes
The Eurocodes are a set of structural design standards, developed by CEN (the
European Committee for Standardization) over the last 30 years, to cover the design
of all types of structures in steel, concrete, timber, masonry and aluminium. In the
UK they are published by BSI under the designations BS EN 1990 to BS EN 1999,
each in a number of ‘Parts’. Each Part is accompanied by a National Annex that
implements the CEN document and adds certain UK-specific provisions.
There are ten separate Eurocodes:

BS EN 1990 Eurocode: Basis of structural design
BS EN 1991 Eurocode 1: Actions on structures
BS EN 1992 Eurocode 2: Design of concrete structures
BS EN 1993 Eurocode 3: Design of steel structures
BS EN 1994 Eurocode 4: Design of composite steel and concrete structures
BS EN 1995 Eurocode 5: Design of timber structures
BS EN 1996 Eurocode 6: Design of masonry structures
BS EN 1997 Eurocode 7: Geotechnical design
BS EN 1998 Eurocode 8: Design of structures for earthquake resistance
BS EN 1999 Eurocode 9: Design of Aluminium Structures
Each Eurocode is comprised of a number of Parts, which are published as separate

documents.
The main Eurocodes that may be required for the design of a light steel buildings
and elements are:
BS EN 1990 Basis of structural design
BS EN 1991 Actions on structures
BS EN 1993 Design of steel structures
BS EN 1990[2] can be considered as the ‘core’ document of the structural Eurocode

system because it establishes the principles and requirements for the safety,
serviceability and durability of structures. This document specifies the limit states
and load combinations that should be considered.
Eurocode 1 is used to determine the actions (loads) that structures must be

designed to resist.
Eurocode 3 comprises twelve parts (BS EN 1993-1-1 to BS EN 1993-1-12). When

designing light steel framed buildings, the following parts of BS EN 1993-1 will
generally be required:
BS EN 1993-1-1[3] General rules and rules for buildings

BS EN 1993-1-3[4] Supplementary rules for cold-formed members and sheeting
[5]
BS EN 1993-1-5 Plated structural elements
BS EN 1993-1-8[6] Design of joints
Further guidance on the Eurocode system of design for steel structures and the
documents involved is provided in SCI publications P362[7] and P387[8].
2
Discuss me ...
The focus of this report is the design of light steel elements for which
BS EN 1993-1-3 is the main reference. However, for the design of light steel frames
it is important to note that other parts of the Eurocodes include essential design
requirements, e.g. frame stability in BS EN 1993-1-1 and structural robustness in
BS EN 1991-1-7[9] (see Section 7).
1.3 Design to Eurocode 3

Although BS EN 1993-1-3 is only a ‘supplementary’ Part of Eurocode 3 and has
many references to the general rules in BS EN 1993-1-1 and to rules in some other
Parts, as noted above, it contains most of the rules that will be needed for the
design of light steel structural members. BS EN 1993-1-3 has been specifically
written for light steel and makes special allowances for the structural behaviour
commonly encountered with this product.
BS EN 1993-1-3 permits two alternative design routes:

 Design by calculation
 Design by testing.
As the name suggests, in the former route the structural designer follows an
analytical procedure laid down in BS EN 1993-1-3 to arrive at a calculated value of
section resistance (to compression, bending etc.). The method is fairly complex due
to the need to take account of local buckling through the use of effective widths, but
can be attempted by hand for relatively simple sections such as lipped C sections.
One disadvantage of this approach is that it tends to be conservative due to the
assumptions and simplifications on which the method is based. For this reason, this
approach is rarely used by purlin and cladding rail manufacturers, as it would place
their products at a commercial disadvantage. Furthermore, while the scope of
BS EN 1993-1-3 includes a range of section shapes, it is less well suited to some of
the more complex sections, especially those with multiple stiffeners and curved
webs, flanges or lips. However, despite the apparent limitations of this approach,
design by calculation remains the preferred option for designers of light steel frames
and floor joists and is the primary focus of this report.
Design by testing overcomes the limitations of the calculation approach by

permitting design resistances to be obtained accurately for almost any shape of
section. However, the benefits must be weighed up against the costs of undertaking
a programme of tests followed by the statistical analysis required by BS EN
1993-1-3 to convert raw test data into usable design values. Where testing is
undertaken, such as for purlins and cladding rails, the design data are usually
tabulated in the form of load span tables. These tables are published by the product
manufacturer and are used by potential specifiers to choose the most suitable
section for the given span and load.
The testing route can be sub-divided into two options:

 Design data derived directly from test data.
 Design data derived from a numerical model.
In the first option, an appropriate number of tests are undertaken on a range of test
specimens and the results are analysed statistically to obtain a characteristic

resistance for each section size within the range. The rules for statistical analysis
are laid out in BS EN 1993-1-3 (or alternatively the method in Annex D of BS EN
1990 may be used) and involve subtracting a prescribed multiple of the standard
3
Discuss me ...
deviation from the mean test result. The multiple is dependent on the number of
tests undertaken. This option is the simpler of the two, but has the disadvantage that
tests must be performed on multiple samples of every section size within the product
range. It is, therefore, not suitable for products with a wide range of dimensional
variations, such as purlins and cladding rails, which are normally sold in a wide
range of depths and gauges. It is, however, a useful method where the product
range is limited, such as purlin cleats or tie wires.
In the second option, testing is only conducted on a limited number of specimens

from within the full range. The test results are then normalised (by comparison with
the equivalent theoretical value) and used to derive a numerical model that aims to
accurately predict a safe design resistance across the full product range. This
approach avoids the need to test the full range of section sizes and even allows for
new sections to be introduced at a later date without the need for additional testing.
Furthermore, where test results are normalised against a theoretical value, it is often
possible to consider results from different section sizes as belonging to the same
‘family’ of data, thereby reducing the multiple of standard deviations to be
subtracted. The accuracy of the final resistances will depend on the complexity of
the numerical model and the number of tests undertaken. For example, a simple
model might be used to calculate a realistic bending resistance of a section, while
maintaining the theoretical assumption of a simply supported beam. By comparison,
a more complex model might take account of the stiffness in the end connections of
the beam, using data from a separate set of tests, in order to increase the beam’s
resistance.
A typical test on a light steel purlin is shown in Figure 1.2. In this instance, a point
load is being applied to a cleat connecting two lengths of purlin at the mid-point of a
simply supported span. The purpose of this test was to assess the moment-rotation
behaviour of the joint in order to model accurately the behaviour of a complete purlin
system. The data from this series of tests were combined with results from two other
types of test (gravity and uplift loading on a pair of purlins with sheeting attached) to
create a numerical model, which was then used to produce load span data for the
full range of purlin sizes.
Figure 1.2 Physical testing of purlin specimen
4
Discuss me ...
2 SECTION PROPERTIES
Before a light steel member’s resistance to bending, compression or other loading

type can be calculated, it is necessary to determine the dimensional properties of
the section under consideration. To those unfamiliar with light steel, this first step
might appear to be a trivial exercise of calculating the cross-sectional area and
second moments of area, but in reality it is a complex process that lies at the heart
of the Eurocode design procedure for light steel.
It is important to distinguish between the following types of section property:

 Gross section properties
 Effective section properties.
The term ‘effective section properties’ refers to the properties of a fictitious cross
section that has been reduced in area to take account of the impact of local buckling
on its resistance (see Section 3). Further reductions may also be necessary to allow
for distortional buckling (see Section 4). The bending and compression resistances
of light steel members are always calculated using the effective properties of the
section. The calculation of the effective section properties is dealt with in detail in
Section 4. The remainder of this section discusses the determination of gross
section properties because these are required for some aspects of light steel design.
As the name suggests, the term gross section properties refers to the whole cross
section without any reduction for local buckling. The process of calculating the gross
section properties is relatively straightforward for most common section shapes as it
involves little more than the summation of elemental areas and first and second
moments of area (for flanges, web, stiffeners etc.), the calculation of the position of
the major and minor centroidal axes and, from these values, the second moment of
area for the whole section. A similar process can be repeated for other properties as
required. There are, however, three important issues that need to be addressed
when considering light steel sections:
 Core steel thickness;
 Use of mid-line theory;
 Impact of corner radii.
2.1 Core steel thickness

The galvanized cold rolled strip steel of the type used in light steel construction is
normally delivered pre-coated. Therefore, when specifying the thickness of the steel
it is common practice to include the thickness of the coating in the specified value.
However, BS EN 1993-1-3 (§3.2.4) requires that all section properties are based on
the core thickness of the steel, excluding the coating. The standard zinc coating for
construction products is 275 g/m2 (referred to as Z275), which corresponds to a
coating thickness of 0.02 mm on each surface. Hence, the nominal (specified) steel
thickness should be reduced by 0.04 mm for design purposes.
The tolerances to which the steel is manufactured and specified should also be
considered when determining the thickness to use in design calculations in

accordance with BS EN 1993-1-3 (§3.2.4(3)). The application of this clause is
explained in detail in SCI advisory note, AD 358: Design thickness of cold formed
members and sheeting[10].
5
Discuss me ...
2.2 Mid-line theory

When calculating the section properties of light steel sections, it is standard practice
to measure all dimensions along the mid-lines of the individual elements. Initially,
corner radii are ignored (see Section 2.3), resulting in an idealised section consisting
of a series of thin rectangular elements. In calculating the lengths of the individual
elements, an allowance must be made for the intersection between adjacent
elements, to avoid double-counting the over-lapping corner regions. Using the mid-
line theory, this is simply achieved by measuring each element length between the
points of intersection of the mid-lines. This results in a reduction in the element
length below its nominal value of either t/2 or t, depending on the number of corners.
The mid-line dimensions for a lipped C section are shown in Figure 2.1.
Figure 2.1 Mid-line dimensions for a lipped C section
2.3 Corner radii

The use of mid-line theory described in Section 2.2 results in an idealised section
that is easy to analyse. However, without modification, the impact of the rounded
corners on the section properties, which could be significant, is ignored. The
problem is illustrated by Figure 2.2.
Figure 2.2 Rounded corner of a light steel section
6
Discuss me ...
According to mid-line theory, the web and flange intersect at the point X (the
intersection of the two mid-lines. The true intersection point is P, a distance gr from
X, given by:
     
g r  rm  tan   sin    (Eq.1)
 2  2 
where
t
rm  r 
2
It is apparent that any section properties derived from mid-line theory will not be
exact. The important question for designers is whether or not the error is significant.
BS EN 1993-1-3 gives some guidance in this respect (in §5.1) stating that the
influence of rounded corners on cross section resistance may be neglected provided
that both of the following conditions are satisfied:
r  5t
r  0.1b p
where bp is the width of the element measured between the midpoints of the corners
(see Figure 2.2).
Consider the following example:

A typical lipped C section has a nominal width of 65 mm, corner radii of 3.0 mm and
a nominal thickness of 1.5 mm.
Allowing for the standard 275 g/m2 zinc coating, the core thickness t = 1.46 mm.
The flange width measured between mid-lines = 65 – 1.5 = 63.5 mm.
(It is assumed that the nominal width of 65 mm includes the galvanising, so it is

appropriate to subtract the nominal thickness when calculating the mid-line
dimension.)
t
rm  r  = 3.73 mm
2
     
g r  rm  tan   sin    = 1.09 mm
 2  2 
bp  63.5  2 g r  61.32 mm
(which corresponds to a reduction in area of about 3.4%)
Checking the corner radii:

5t = 7.3 mm, r = 3.0 mm, therefore r  5t
0.1bp = 6.13 mm, r = 3.0 mm, therefore r  0.1bp
7
Discuss me ...
Therefore, in this instance, the influence of the rounded corners may be neglected
when calculating the cross section resistance.
Note: The influence of rounded corners should always be taken into account when
calculating cross section stiffness properties (BS EN 1993-1-3 (§5.1(3)).
Where the influence of rounded corners needs to be accounted for, this is achieved
by first calculating the section properties assuming sharp corners (i.e. ignoring the
corner radii) and then applying reduction factors as follows:
For area, Ag  Ag,sh 1    (Eq.2)
For second moment of area, I g  I g,sh 1  2  (Eq.3)
For warping, I w  I w,sh 1  4  (Eq.4)
In these expressions, the subscript ‘sh’ denotes the section property based on sharp
corners and δ is a reduction factor given by:
n j
j1 r j 90
  0.43 m
(Eq.5)
i1 bp,i
where:
rj is the internal radius of curved element j

n is the number of curved elements (number of corners)
øj is the angle between two plane elements
bp,i is the notional flat width of plane element i
m is the number of plane elements.
The same reduction factors may also be applied to the effective section properties
(Aeff, Iy,eff, Iz,eff and Iw,eff) provided that the notional flat widths of the plane elements
are measured to the points of intersection of their midlines.
BS EN 1993-1-3, 5.1(6) states that where r is greater than 0.04t E / fy then the
resistance of the cross section should be determined by physical testing. This
situation is unlikely to arise for any of the standard sections used in light steel
framing, but designers need to be aware of this limit when dealing with some of the
more unusual section shapes that are introduced into the light steel market from
time to time.
8
Discuss me ...
3 LOCAL BUCKLING
Light steel sections of the types used in construction are profiled in such a way that
they are extremely efficient in terms of their use of material. However, the
associated penalty from a designer’s point of view is the need to consider local
buckling and its impact on the structural resistance of the cross-section.
Designers of hot-rolled structural steelwork will be familiar with the concept of

section classification in which a section’s susceptibility to local buckling is
determined by comparing dimensional ratios (e.g. the depth to thickness ratio of the
web) to a set of specified limits. This process results in each section being assigned
a ‘class’ (which in some cases is dependent on the magnitude of any compression
in the member). BS EN 1993-1-1 describes four such classes and prescribes design
rules for each class that reflect the impact of local buckling on the resistance of the
section. The classifications range from ‘Class 1’, which is defined as being a section
capable of sustaining its full plastic moment while accommodating rotation of a
plastic hinge, to ‘Class 4’, defined as a section whose moment resistance is limited
by local buckling. (The bending resistance of a Class 4 section is less than that
corresponding to elastic bending resistance of the gross cross-section.)
The approach used in light steel design is quite different. Rather than classifying the
cross-section, there is an implied assumption that the section is class 4 (although
this term is not used in BS EN 1993-1-3). Having made this assumption, the design
process focuses on the calculation of effective section properties, following the

same procedures for design resistance as for class 4 sections to BS EN 1993-1-1.
The use of effective section properties stems from the need to simplify the complex
stress distributions associated with local buckling, in order to minimise the required
computational effort, without being over conservative in terms of the cross section
resistance.
3.1 Effective width concept

When dealing with local buckling of slender plate elements, the behaviour of the
plate may be analysed approximately using the effective width method, as illustrated
by Figure 3.1.
Figure 3.1 The effective width concept applied to a plate

In this approach, the actual stress distribution acting over element width b is
replaced by simplified equivalent stresses acting over two equal widths of beff/2. The
central portion of the plate, the region most affected by local buckling, is assumed to
have no stress and is ignored completely. The result is a simple model in which a
9
Discuss me ...
uniform stress equal to the yield strength of the steel is assumed to act over a
reduced width of plate.
The method adopted by BS EN 1993-1-3 takes the effective width concept

illustrated above and applies it to the cross section of a light steel member. The
cross section is divided into elements (flanges, web, lips etc.), with each element
being treated like the flat plate in Figure 3.1. An effective width beff is calculated for
all elements that are subjected to compressive stress (either due to axial
compression or bending). The effective area of the element Aeff is then obtained by
multiplying beff by the section thickness t. Elements not subjected to compressive
stress are not susceptible to local buckling, so the full element width b may be used
in the calculation of the effective section properties.
Having obtained the elemental effective widths and areas, the properties of the
effective cross section are determined in the usual manner by calculating the
position of the neutral axis followed by the first and second moments of area about
this axis. The resulting set of ‘effective properties’ should be used when calculating
the resistance of the cross section to bending or compression as appropriate.
Since the distribution of compressive stress across the section differs between
sections subjected to pure axial compression and those subjected to bending, it
follows that the effective section properties will differ between these two cases.
Furthermore, asymmetric sections subjected to bending may have one set of
effective properties when sagging and another when hogging. It is important to use
the relevant effective properties for the case under consideration.
For sections at the heavier end of the ‘light steel’ range, especially those with
relatively stocky webs and flanges, it might not be appropriate to reduce the cross-
sectional properties to allow for local buckling. Since BS EN 1993-1-3 does not
permit the classification of the section in the manner familiar to designers of hot-
rolled structural steel, the procedure outlined above must be followed even for
stocky sections. However, in this case, the calculation procedure will automatically
yield effective widths beff equal to the full widths b, resulting in effective section
properties equal to the gross section properties.
3.2 Eurocode calculation procedure for unstiffened

plane elements
This Section focuses on the calculation of beff for an element subjected to
compression.
The calculation of beff for plane elements without stiffeners is introduced in §5.5.2 of
BS EN 1993-1-3. However, the detail of the method, including the relevant
equations can be found in BS EN 1993-1-5.
For each element, the effective width is given by:

beff  b (Eq.6)
where:
b is the width of the element

ρ is the reduction factor to allow for local buckling.
10
Discuss me ...
The reduction factor ρ takes account of the slenderness of the element, whether it is
an internal or outstand element and the stress distribution within the element.
For an internal element, ρ is given by:
 p  0.0553  
 2
 1.0 (Eq.7)
p
For an outstand element, ρ is given by:
 p  0.188
 2
 1.0 (Eq.8)
p
where ψ is the stress ratio between the ends of the element and p is the
slenderness of the element given by:
fy bt
p   (Eq.9)
 cr 28.4 k σ
where
fy is the design strength
σcr is the elastic critical plate buckling stress
b is the appropriate width of the compression element

t is the steel core thickness (i.e. minus the coating)
kσ is the buckling factor corresponding to the stress ratio ψ and the
boundary conditions. Values of kσ should be obtained from Tables 4.1
and 4.2 of BS EN 1993-1-5 for internal and outstand elements
respectively.
  235 / f y
From the discussion in Section 3.1, it follows that there must be limits of slenderness
below which local buckling does not influence the resistance of the section. These
limits correspond to ρ = 1 in equations 7 and 8 and, according to BS EN 1993-1-5,
the limits are:
Internal compression elements:
 p  0.673
Outstand compression elements:
 p  0.748
Where p is lower than the appropriate limit, ρ should be taken as 1.0 in the
calculation of the effective width of that element. This does not necessarily mean
that the section is fully effective, since there may be other elements for which
ρ < 1.0.
11
Discuss me ...
4 DISTORTIONAL BUCKLING
In the discussion on local buckling, it was assumed that the corners of the section
remained fixed in position, so that the buckling deformation takes place within the
length of the element. This case is represented on the left hand side of Figure 4.1.
By contrast, the right hand side of Figure 4.1 shows a situation in which the right
hand corners of the flanges are not fixed in position, allowing the flanges to rotate.
This is known as distortional buckling.
Figure 4.1 Local and distortional buckling

The susceptibility of a section to distortional buckling depends on the ability of the

stiffeners to prevent displacement of the adjacent flange corners. This is dependent
on the geometry of the stiffeners relative to the flanges and, in particular, their
relative stiffness. BS EN 1993-1-3 presents a detailed procedure for the design of
stiffened sections based on a simple spring model. This approach is described in
Section 4.1 with a summary of the procedure given in Section 4.2.
4.1 Design of stiffened sections

Due to the susceptibility of light steel sections to local and distortional buckling, it is
common practice for manufacturers to roll stiffeners into the sections. It is most
common to stiffen the free edge of the flanges by the provision of a lip, or in some
cases a double lip, but intermediate flange stiffeners may also be used in order to
permit the use of thinner gauge steel. It is less common to stiffen the web in sections
used for light steel framing applications, although deeper purlins and floor joists may
benefit from this technique. Flange and web stiffeners are often used in trapezoidal
deck profiles of the types used for roofing and flooring applications due to the very
light gauge steel used in these products.
BS EN 1993-1-3 provides guidance to cater for all of the options discussed above.
In all cases, the underlying assumption is that the stiffener behaves like a
compression member with a continuous partial restraint. This is a reasonable
assumption since, whether the member is subjected to pure axial compression or
bending, one flange and its stiffener at least will be subjected to a longitudinal
compressive stress. In the design model, the stiffener is represented by a linear
spring of stiffness K, as shown in Figure 4.2.
12
Discuss me ...
Figure 4.2 Linear spring model
The spring stiffness depends on the boundary conditions and the flexural stiffness of
the adjacent elements. The spring is assumed to act at the centroid of the effective
stiffener section. Figure 4.2 shows two varieties of edge stiffener and one
intermediate stiffener. In each case, the ‘effective stiffener section’, which is
depicted as a dark solid line, comprises the stiffener itself plus an adjacent length (or
lengths) of flange.
In order that the stiffener provides sufficient stiffness and to avoid buckling of the
stiffener, BS EN 1993-1-3 gives limits for the geometry of the stiffener in relation to
the adjacent flange as follows:
 For a single or double-lipped section, the length of the lip c, measured
perpendicular to the flange width b, should lie in the range 0.2 ≤ c/b ≤ 0.6.
 For a double-lipped section, the return length of the lip d, measured parallel to the
flange (width b), should lie in the range 0.1 ≤ d/b ≤ 0.3.
For the case of the edge stiffeners of lipped C sections and Z sections, the spring
stiffness K1 for flange 1 may be obtained from the following equation:
E t3 1
K1   2 (Eq.10)
4(1   ) b1 hp  b1  0.5b1b2 hp kf
2 3
where:
b1 and b2 are the distances from the web-to-flange junction to the centroid of the
effective stiffener section for flanges 1 and 2 respectively.
hw is the depth of the web.
kf is the ratio of the effective areas of the two edge stiffeners (including
the effective portion of the flange).
All other symbols have their usual meanings.

13
Discuss me ...
The term kf can have the following values:

 For sections subjected to axial compression, such that flange 1 and flange 2 are
both in compression: kf = As2 / As1 (where As1 and As2 are the effective areas of
the edge stiffeners).
 For sections subjected to bending about the major axis so that flange 1 is in
compression and flange 2 is in tension: kf = 0
 For symmetric sections in compression: kf = 1
Having calculated the equivalent spring stiffness of the stiffener, the method
described in BS EN 1993-1-3 proceeds to determine the elastic critical stress for the
stiffener σcr followed by the relative slenderness  d and the reduction factor for the
distortional buckling resistance χd. Finally, χd is used to determine the reduced
effective area of the stiffener, which is generally represented as a reduced thickness
when determining the effective properties of the section.
A detailed algorithm for determining χd and hence the reduced thickness of the
stiffener is described below.
4.2 Eurocode calculation procedure for stiffened elements

Detailed procedures for the design of plane elements with edge or intermediate
stiffeners are presented in §5.5.3 of BS EN 1993-1-3. The procedures combine the
calculation of beff with the calculation of a reduced thickness for the stiffener. The
former takes account of local buckling within the length of the element, while the
latter makes an allowance for the impact of distortional buckling. The procedure
described below is for a flange with an edge stiffener. It is divided into three steps,
the last of which involves an optional iteration in order to refine the value of the
reduction factor χd. BS EN 1993-1-3 also presents procedures for flanges with
intermediate stiffeners, stiffened webs and trapezoidal decking profiles.
Step 1:
The procedure begins with the calculation of the effective width of the flange beff
following the method described in Section 3.2. At this point in the procedure, it is
assumed that the edge stiffener is infinitely stiff and, therefore, provides full restraint
to the free end of the flange. This corresponds to the left hand side of Figure 4.1, in
which the corners of the section are fixed in position and failure is due to local
buckling. It is also assumed that the maximum compressive stress in the flange is
equal to the design strength of the material, i.e.
 com,Ed  f yb  M0 (Eq.11)
Step 2:
In the second step, the edge stiffener is considered in isolation in order to calculate
the reduction factor χd for distortional buckling. At this point, the infinitely stiff spring
used in Step 1 is replaced by a spring of stiffness K, as illustrated in Figure 4.2. The
spring stiffness K may be calculated from Equation 10, using the initial effective
cross section of the stiffener determined in Step 1.
14
conditions
Discuss
and me ...
Once K is known, the elastic critical buckling stress σcr,s of the stiffener may be
calculated from:
2 KEI s
terms
 cr,s  (Eq.12)
As
Is and As are the effective second moment of area and the effective cross-sectional
the
area respectively of the stiffener.
The relative slenderness of the stiffener for distortional buckling d is given by:
to
f yb
d  (Eq.13)
 cr,s
subject
where fyb is the basic design strength of the steel.
The reduction factor for distortional buckling χd is dependent on the slenderness d

is
as follows:
For  d  0.65 ,  d  1.0

document
Eq.14a)
For 0.65   d  1.38 ,  d  1.47  0.723 d (Eq.14b)
For  d  1.38 ,  d  0.66  d (Eq.14c)

this
Step 3:
The value of χd may be refined iteratively by returning to Step 1 and calculating a
of
modified effective flange width beff based on a revised compressive stress σcom,Ed.
This is achieved by calculating a modified value of ρ (see Section 3.2) using a
Use
reduced p given by:
 p,red   p  d (Eq.15)
reserved.
Step 2 may then be repeated for the modified effective section to obtain a new value
of χd. Steps 1 and 2 may be repeated until the desired degree of convergence on
the value of χd has been achieved.
rights
Step 3 is entirely optional and it is perfectly acceptable to use the initial value of χd
for the calculation of the reduced area of the stiffener. Where the designer chooses
to iterate to obtain an improved value of χd, one or two iterations should suffice.
all
Once χd has been calculated to the desired degree of refinement, the reduced
effective area of the stiffener As,red may be calculated using :
2012
f yb /  M 0
-
As,red   d As (Eq.16)
 com,Ed
copyright
February
where σcom,Ed is the compressive stress at the centreline of the stiffener based on
the effective cross section.
It is usually more convenient to work in terms of a reduced thickness such that:

is
23
t red  t As,red As (Eq.17)

material
on
Created
15
This
Discuss me ...
This may be calculated directly using:

t red  t d (Eq.18)
Worked Examples 2 and 3 in Section 8 illustrate the use of this procedure for a
lipped C section under pure axial compression and in bending, respectively.
16
Discuss me ...
5 DESIGN OF COMPRESSION MEMBERS
5.1 Design issues

Light steel members are often required to carry axial compression loads, such as in
the case of studs in a load-bearing wall. As with their hot-rolled counterparts, the
failure of light steel compression members is likely to be governed by buckling rather
than yielding of the cross section, resulting in a member resistance significantly
lower than the squash load of the section. The design method for such a member is,
therefore, focused on the calculation of its buckling resistance and is similar in many
respects to the design of hot-rolled steel columns. However, the behaviour of light
steel wall studs differs in a number of respects from that of hot-rolled columns and
these differences must be accounted for in the design procedure.
Unlike columns, which act as independent members within a structural frame, light
steel wall studs are used in conjunction with plasterboard, and often some form of
sheathing board, to form a load-bearing panel. The presence of the boards will
provide a certain degree of lateral restraint in the minor axis of the studs, which may
be utilized when calculating the buckling resistance. However, any restraint must be
verified by testing, using studs of a representative slenderness range and a similar
build-up of boards to that used in practice.
While flexural buckling usually governs the behaviour of hot-rolled steel columns,
many light steel sections are also susceptible to torsional-flexural buckling. If

torsional-flexural buckling occurs at a lower magnitude of load than flexural buckling,
this mode of failure will naturally govern the resistance of the member. This is
reflected in the Eurocode design rules, in which the elastic critical buckling load
used for design is taken to be the smallest of the elastic critical buckling loads for
flexural buckling, torsional buckling and torsional-flexural buckling.
Finally, as noted in Sections 3 and 4, light steel sections are susceptible to local and
distortional buckling, both of which can have an adverse impact on the compression
resistance of a member. This should be accounted for by using the effective cross-
sectional area instead of the area of the gross cross section when calculating the
compression resistance.
5.2 Eurocode calculation procedures

The design procedures for light steel compression members are given in §6.2 of
BS EN 1993-1-3. However, due to the similarities with the design of hot-rolled
columns, designers are referred to §6.3 of BS EN 1993-1-1 for much of the detail,
including the buckling curves.
The design buckling resistance of a member subjected to axial compression is given

by:
Aeff f y
N b,Rd  (Eq.19)
 M1
where
χ is the reduction factor for flexural buckling

Aeff is the area of the effective cross section (Sections 3 and 4)
17
con
Discuss
and me ...

terms
γM1 is the partial safety factor for buckling
Since fy and γM1 are known values (γM1 = 1.0 in the UK NA) and the calculation of
Aeff has been dealt with earlier in this chapter, the procedure described below
the
focuses on the calculation of χ.
The reduction factor χ is used to quantify the reduction in resistance below the
to
squash load of the section due to buckling. It may be obtained from BS EN 1993-1-1
using the appropriate buckling curve and the value of slenderness corresponding
subject
to the critical mode of failure. BS EN 1993-1-1 offers a choice of 5 buckling curves,

but this is restricted to 3 curves (a, b and c) for light steel according to §6.2.2 of
BS EN 1993-1-3. The appropriate choice of curve for various types of cross section
is given in Table 6.3 of BS EN 1993-1-3.
is
The relationship between χ and slenderness for buckling curves a, b and c is shown
in Figure 5.1. The squash load of the section corresponds to χ = 1.0.
document
this
of
Use
reserved.
rights
all
Figure 5.1 Buckling curves a, b and c

2012
The slenderness is given by:

-
Aeff f y

copyright
(Eq.20)
February
N cr
Ncr is the elastic critical buckling load, which for flexural buckling is equal to the
Euler load and is given by:
is
23
 2 EI
N cr  2
(Eq.21)
Lcr
material
on
Created
18
This
Discuss me ...
where:
E is Young’s modulus for the material.
I is the appropriate second moment of area (for the gross cross section).
Lcr is the buckling length in the plane considered.
Alternatively, may be obtained from:
Lcr Aeff Agr

 (Eq.22)
i 1
where i is the radius of gyration and λ1 is given by
E
  (Eq.23)
fy
If the buckling lengths differ between the major and minor axes, for example where
a mid-height noggin provides restraint to the minor axis of a wall stud, values of
should be obtained for both axes (since major axis flexural buckling might govern in
this case). In cases where the critical mode of failure is either torsional buckling or
torsional-flexural buckling, should be obtained from Equation 20 using the value
of Ncr corresponding to the critical mode of failure (i.e. the elastic critical buckling
load for torsional or torsional-flexural buckling). Equations for Ncr for both modes of
failure are given in BS EN 1993-1-3.
The reduction factor χ may be obtained directly from the buckling curves printed in
BS EN 1993-1-1 or from the following equations:
1
 but ≤ 1.0 (Eq.24)
   2 2

  0.5 1     0.2   2  (Eq.25)
α is the imperfection factor corresponding to the chosen buckling curve. Values for α
are given in Table 6.1 of BS EN 1993-1-1.
The design of members subject to axial compression is demonstrated in Worked

Examples 5 and 8 in Section 8.
19
Discuss me ...
6 DESIGN OF MEMBERS IN BENDING
Several construction applications require light steel members to carry loads in

bending. Examples include floor joists, roof purlins and wall studs (when subjected
to wind loading). As with hot-rolled steel beams, it is essential that a distinction is
made between members which are laterally restrained and those that are
susceptible to lateral-torsional buckling, since the bending resistance will differ
significantly between the two cases. While many joists, purlins and beams have
some form of attachment providing restraint, designers should be aware of the
potential lack of restraint during construction and the risk of load reversal. In addition
to checking the member resistance, it may also be necessary to check serviceability
limits such as deflections and dynamic response.
6.1 Laterally restrained members

Beams and similar structural members may be considered to be laterally restrained
when their compression flange is held in position to the extent that lateral-torsional
buckling is prevented. This is often the case with light steel framing members due to
the attachment of sheathing or plasterboard (for walls) or flooring products (e.g.
wooden floor boards or concrete slabs). Purlins and cladding rails are also often
laterally restrained due to the attachment of cladding. However, as the cladding is
only attached to one side of the member, which may be subjected to sagging or
hogging bending depending on the direction of the loading, purlins and cladding rails
must usually be designed for lateral-torsional buckling for at least one load case.
The design of laterally restrained light steel beams is similar to the design of the
equivalent hot-rolled members. As such, the following issues need to be considered:
 Bending resistance
 Shear resistance
 Local web failure
 Deflections (at SLS).
As noted previously, the key difference between light steel and hot-rolled steel is the
susceptibility to local and distortional buckling, both of which are dealt with by the
use of effective section properties. However, the use of light gauge material has
other consequences, such as the increased risk of shear buckling and of crushing,
crippling or buckling of the web under local transverse forces. A typical failure of a
light steel member subjected to bending is shown in Figure 6.1. The member shown
is a Z section purlin, but similar failure modes can be observed in lipped C sections
of the type used in framing applications.
20
Discuss me ...
Figure 6.1 Bending failure of a light steel purlin
The resistance of a light steel cross section to bending is considered in §6.1.4 of

BS EN 1993-1-3, which states that the moment resistance is given by:
Weff f yb
M c,Rd  (Eq.26)
M0
Weff is the elastic modulus of the effective cross section, as discussed in Sections 3
and 4 (See Worked Examples 2 and 3 in Section 8).
Equation 26 assumes that failure is due to yielding of the compression flange.

Where yielding occurs first in the tension flange, plastic reserves in the tension zone
may be utilized, as explained in §6.1.4.2 of BS EN 1993-1-3. In this case, the
bending moment will be limited by the maximum compressive stress σcom,Ed reaching
fyb / γM0.
The resistance to shear is considered in §6.1.5 of BS EN 1993-1-3, which states

that the design shear resistance is given by:
hw
tf bv
sin 
Vb,Rd  (Eq.27)
M0
where:
fbv is the shear strength allowing for buckling
hw is the web height between the mid-lines of the flanges
 is the slope of the web relative to the flanges.
It is apparent from the use of the term Vb,Rd that this is a buckling resistance rather
than a cross section resistance. This is due to the susceptibility of some light steel
sections to shear buckling. The risk of failure due to shear buckling is dependent on
the slenderness of the web, so deep sections made from very light gauge steel are
most at risk. Shear buckling is accounted for in the design procedure by the use of
21
Discuss me ...
fbv, which is a function of the basic yield strength fyb and the relative web
slenderness w. Values of fbv may be obtained from Table 6.1 of BS EN 1993-1-3.
Local transverse forces are considered in §6.1.7 of BS EN 1993-1-3. Several

equations are presented for the local resistance of the web Rw,Rd, taking into account
the location of the applied load (e.g. whether close to the end of the member), the
number of webs in the cross section and the inclusion or absence of stiffeners.
6.2 Lateral-torsional buckling

Ideally, light steel members should be incorporated into systems that provide lateral
and torsional restraint and, therefore, prevent lateral-torsional buckling. However,
there are occasions when this is not possible and the member has to be designed
as an unrestrained beam. This situation is considered in §6.2.4 of BS EN 1993-1-3.
Alternatively, for purlins and cladding rails, the cladding may provide full lateral
restraint when the flange to which it is attached is in compression, but only partial
restraint when it is in tension. Design rules for this situation are provided in §10.1.4
of BS EN 1993-1-3.
The behaviour of an unrestrained beam loaded in bending about its major axis is
analogous to the behaviour of a column under axial load and the buckling curves
depicted in Figure 5.1 are equally applicable to lateral-torsional buckling. The
essential feature of this mode of failure is that the compression flange becomes
unstable and, as it is not restrained, attempts to buckle laterally. However, since it is
attached to the tension flange via the web of the beam, it cannot move freely and
must pull the tension flange over as it deforms. The tension flange resists, resulting
in the classical combination of lateral and torsional deformation, commonly known
as lateral-torsional buckling. This type of failure only occurs when a member is bent
about its major axis; members bent about their minor axis will always fail by minor
axis bending and never by lateral-torsional buckling.
The impact of lateral-torsional buckling on the bending resistance of a beam is

dependent on a number of factors. Principal among these are the geometry of the
cross section and the slenderness of the member. In terms of geometry, sections
that have a low minor axis flexural rigidity relative to the flexural rigidity of the major
axis are most likely to fail by lateral-torsional buckling. The ease with which a
section is able to twist and warp is also important. Light steel sections
(e.g. C sections and Z sections) tend to be poor on both counts. By contrast, square
hollow sections cannot suffer lateral-torsional buckling.
The relationship between slenderness and bending resistance can be represented

by a buckling curve of the type shown in Figure 5.1. For very stocky members, the
bending resistance will be limited by the resistance of the cross section (given by
Equation 26), but as the slenderness increases, the influence of lateral-torsional
buckling also increases, resulting in a significant reduction in the bending resistance.
As with the case of axial compression, it is convenient to quantify this reduction in
terms of a reduction factor, expressed as a proportion of the section capacity. For
unrestrained beams, the Eurocode symbol for this reduction factor is χLT.
Due to the similarities between the design of light steel unrestrained beams and the
equivalent hot-rolled steel members, designers are referred to §6.3 of BS EN
1993-1-1 for the detailed design procedures and the buckling curves. However,
important guidance is given in §6.2.4 of BS EN 1993-1-3 regarding the choice of

design method and buckling curve.
22
Steelbiz
Discuss
the me ...
The design buckling resistance of a member subjected to bending is given by:

of
 LTWeff,y f y
M b,Rd  (Eq.28)
 M1
conditions
where
χLT is the reduction factor for lateral-torsional buckling
and
Weff,y is the elastic modulus of the effective cross section (major axis)
terms
γM1 is the partial safety factor for buckling (= 1.0 according to UK NA).
BS EN 1993-1-1 gives two alternative methods for the calculation of χLT. However,
the
only the ‘General case’ given in §6.3.2.2 is permitted for light steel members. This
method resembles that used for column buckling and uses the same buckling
curves. As such, the equations resemble those given in Section 5.2 with the addition
to
of the subscript ‘LT’.

subject
The reduction factor χLT is given by:

1
 LT  but ≤ 1.0 (Eq.29)
is
2 2
 LT   LT  LT
document
and

 LT  0.5 1   LT LT  0.2  LT 2  (Eq.30)
αLT is the imperfection factor corresponding to the chosen buckling curve and is
this
given by Table 6.3 of BS EN 1993-1-1. However, since §6.2.4 of BS EN 1993-1-3

states that buckling curve b should always be used for light steel in bending, αLT will
of
always have the value 0.34.
Lateral torsional buckling design checks are demonstrated in Worked Examples 5

Use
and 7 in Section 8.
reserved.
6.3 Serviceability
In addition to checking the resistance of the member to the applied loading and
associated bending moments and shear forces, designers should also check that
rights
the member is adequate at the serviceability limit state (SLS). For normal building
applications, this involves checking the imposed load deflections against specified
limits. Occasionally, the dynamic response of light steel floors will also need to be
all
checked. Guidance for light steel floors is given in P301[12] and specialist guidance
for more general consideration of dynamic response is given in P354[11].
2012
-
For the purpose of calculating deflections, the second moment of area should be
calculated using the following equation:
copyright
February
 gr
I fic  I gr 
 gr
I  I  eff  (Eq. 31)
is
23
material
on
Created
23
This
Discuss me ...
where:
Igr is the second moment of area of the gross cross section.
σgr is the maximum compressive bending stress at SLS based on gross
cross section properties.
I(σ)eff is the second moment of area of the effective cross section calculated
for a maximum stress σ ≥ σgr.
Serviceability limit state checks are demonstrated in Worked Examples 5, 6 and 8 in

Section 8.
24
Discuss me ...
7 FRAME DESIGN
Verifying that each member within a steel frame is capable of withstanding the
applied loads is an essential part of the structural design process, but it is not
sufficient by itself. Other design considerations such as frame stability and structural
robustness must also be considered.
7.1 Frame stability

The stability of a building as a whole must be checked. While the methods used to
provide stability differ between hot-rolled and cold-formed steel, the principles
remain the same:
 A suitable load path must be provided to safely transmit horizontal forces to the
foundations.
 The system used to provide stability must be sufficiently stiff to avoid excessive
lateral deflections.
 Some account should be taken of the instability arising from frame imperfections.
 Where appropriate, account should be taken of second order effects.
The four issues outlined above are dealt with by the design rules in BS EN
1993-1-1. These rules cover hot-rolled and cold-formed steel structures and no
distinction is made between the two forms of construction.
The stability of light steel framing is usually achieved through one of the following
methods:
 Integral bracing
 X bracing
 Diaphragm action.
Integral bracing consists of C section members placed diagonally between the

vertical wall studs and within the depth of the studs. The use of a C section means
that the integral bracing members are capable of carrying tension and compression
forces. However, careful detailing and connection design are important.
X bracing consists of diagonal crossed flat straps attached to the face of the vertical
studs. Unlike integral bracing, the flats usually extend across several studs and
should be connected to every stud that they cross. Each individual bracing element
is only capable of acting in tension (hence the need for the X arrangement).
As an alternative to steel bracing members, the frame designer may choose to rely
on the racking resistance of the wall itself. In this case, stability is provided by
diaphragm action in the plane of the wall due to the attached board or cladding.
Board options include:
 Plywood
 Cement Particle Board
 OSB
 Plasterboard.
25
Discuss me ...
The racking resistance of a particular board and frame combination should be

determined by testing.
7.2 Structural robustness

The term ‘robustness’ when used in the context of building design relates to the
ability of the structure to withstand accidental actions without the spread of damage
or disproportionate collapse. In this sense, robustness is synonymous with structural
integrity.
The essential principles of robustness, which apply across all forms of construction
and materials, are summarised below:
 Robustness relates to the ability of a structure to withstand events such as
explosions, impact or the consequences of human error.
 The structure does not have to be serviceable after the event. Large
deformations and plasticity are permitted. It is expected that the structure will
need to be repaired before it can be re-occupied. In some cases, it may need to
be demolished.
The objectives are:

 To restrict the spread of localised damage and to prevent collapse of the
structure disproportionate to the original cause.
 To ensure the safety of the structure while building occupants make their escape
and the emergency services are in attendance.
Structural robustness and design to avoid disproportionate collapse is a requirement

of Building Regulations. The required level of robustness is dependant on the
building class, which is related to the size, type and use of the building. The same
building classification system is given in Approved Document A, BS 5950-5 and
BS EN 1991-1-7.
The requirement to design and construct buildings to have robustness and avoid
disproportionate collapse under accidental design situations is given in BS EN 1990.
Details of how the requirement should be met are given in BS EN 1991-1-7. The
guidance given in BS EN 1991-1-7 is material independent. The only guidance that
is specific for light steel framing is found in the UK National Annex, where it states in
§NA.3.1 that:
In the case of lightweight building structures (e.g. those whose primary

structure is timber or cold formed light gauge steel), the values for minimum
horizontal tie forces in expression A.1 and A.2 should be taken as 15 kN
and 7.5 kN, respectively.
This guidance is comparable to that given in BS 5950-5, which was based on the
guidance given in SCI publication P301[12]. Light steel multi-storey structures are
generally structurally robust because of their construction using a large number of
regularly distributed structural elements, with a high degree of connectivity. In most
applications, the provision of continuous ties between the components is
straightforward because of the multiple inter connections. However, the connection
resistances must be checked to ensure the appropriate requirements are satisfied.
26
Discuss me ...
7.3 Frame anchorage

Light steel framed buildings must be adequately held in position by a suitable
anchorage system to prevent sliding, overturning or lifting off the foundations. Two
types of anchorage are generally employed:
 Holding down bolts connecting the stud and track to the ground slab.
 Steel straps connecting wall studs to the foundations.
27
Discuss me ...
8 WORKED EXAMPLES
There are 8 worked examples which show the design of light steel sections to
BS EN 1993-1-3. The examples include the calculation of effective section
properties and member capacities. The examples are for cross-sections and
member designs that may typically be used in light steel framing.
The examples are:

8.1 Gross section properties for a cold formed lipped C section
8.2 Effective section properties for a lipped C section in compression
8.3 Effective properties for a lipped C section in major axis bending
8.4 Effective properties for a lipped C section in minor axis bending
8.5 Design of a wall stud in a light steel frame building
8.6 Design of a floor joist (restrained) in a light steel frame building
8.7 Design of a floor joist (unrestrained) in a light steel frame building
8.8 Design of a lattice floor truss in a light steel frame building

The marginal references to clauses in various Eurocode Parts are given in the form
3-1-3/ 5.1, meaning clause 5.1 of BS EN 1993-1-3.
28
Discuss me ...
Job No. BCS 320 Sheet 1 of 9 Rev
Title Example 1
Silwood Park, Ascot, Berks SL5 7QN Subject Gross properties of a lipped C section
Telephone: (01344) 636525
Fax: (01344) 636570 Made by AW Date Oct 2010
Client SCI
CALCULATION SHEET Checked by SN Date Jan 2011
8.1 Gross section properties for a cold-formed lipped

C section
This worked example presents design calculations for the gross section
properties of a cold-formed lipped C section.
Section dimensions and material properties
Section depth h = 200 mm
Flange width b1 = b2 = 65 mm
Stiffener depth c = 25 mm
Corner radius r = 3 mm
Nominal thickness tnom = 2 mm
Core thickness t = 1.96 mm
b1 b p1
c cp
r r
h hp
t t
b2
Mid-line dimensions
Web depth hp = h  t nom  200  2  198 mm

Flange width bpl = bp2  b1  t nom  65  2  63 mm
Stiffener depth cp = c  t nom 2  25  2 2  24 mm
Mean radius rm = r  t 2  3  1.96 2  3.98 mm
Corner gr = rm tan  / 2  sin  / 2   3.98tan 45  sin 45  1.17 mm
29
Discuss me ...
Example 1: Gross properties of a lipped C section Sheet 2 of 9 Rev
Flat width dimensions
Web bp,h  h p  2 g r  198  2 1.17  195.7 mm 3-1-3/ 5.1

Flange bp,b  b p  2 g r  63  2 1.17  60.66 mm
Stiffener bp,c  c p  g r  24  1.17  22.83 mm
Geometry checks
Checks on the geometry of the cross section to ensure that the dimensions are 3-1-3/ 5.2
within the scope of 3-1-3/ .
bt  65 1.96  33.16  60 OK
ct  25 1.96  12.76  50 OK
ht  200 1.96  102.0  500 OK
Check on the dimensions of the stiffener.
c b1  25 65  0.38 0.2  0.38  0.6 OK
Check to see whether the rounding of the corners may be neglected. 3-1-3/ 5.1(3)
r t  3 1.96  1.53  5 OK
r bp, h  3 195.7  0.02  0.10 OK
r bp, b  3 60.66  0.05  0.10 OK
r bp, c  3 22.83  0.13  0.10 Outside limit
Therefore, the influence of rounded corners should be considered in the 3-1-3/ 5.1(4)
calculation of section properties. The influence of rounded corners on section
properties may be taken into account by reducing the properties calculated for
an otherwise similar cross section with sharp corners, using the following
approximations.
Ag  Ag, sharp 1   
Ig  I g, sharp 1  2 
Iw  I w, sharp 1  4 
where
n 
 r j 90j
j 1
  0.43 m
 bp,i
i 1
n 
 r j 90j 4  3
90
  0.43
j 1
= 0.43 90  0.014
m
195.7  2  60.66  2  22.83
 bp,i
i 1
30
Discuss me ...
Gross section properties (sharp corners)

BS 3-1-3/ Annex C has been used to calculate the gross section properties that 3-1-3/ Annex C
may be required for structural calculations. The section is divided into flat
elements. Each element is defined by the coordinates of its ends points based on
the mid-line properties, as shown below. The end points of each element are
called nodes.
3 4
2 y 1
Node y coordinate z coordinate

j yj zj
mm mm
0 63 24
1 63 0
2 0 0
3 0 198
4 63 198
5 63 174
Start Node End Node

Part No. y z No. y z Thick- Length Area
coord coord coord coord ness
i yi-1 zi-1 yi zi ti dAi
mm mm mm mm mm mm mm2
1 0 63 24 1 63 0 1.96 24 47.0
2 1 63 0 2 0 0 1.96 63 123.5
3 2 0 0 3 0 198 1.96 198 388.1
4 3 0 198 4 63 198 1.96 63 123.5
5 4 63 198 5 63 174 1.96 24 47.0
Area
Agr = Σ dAi = 729.1 mm2
31
Discuss me ...
Centroid
The z coordinate of the centroid (zgc) defines the position of the y axis of the 3-1-3/ Annex C
gross section. The y axis is central to the section as the section is symmetrical
about the y axis. Hence, zb1 will be equal to zb2. The calculation of zgc is shown
here for completeness. The y coordinate of the centroid defines the position of
the major z axis of the gross section. The terms are defined on the figure below.
z
ygc = yb1 ylip
zb1
y y
zb2 = zgc
z
Determine the z coordinate of the centroid.

Part (zi + zi-1).dAi / 2 3-1-3/ Annex C
i mm3
1 564.0
2 0.0
3 38420
4 24450
5 8742
Total 72180
First moment of area for the y axis is the total from above,
Sy0 = 72180 mm3.
The z coordinate of the centroid is given by:
zgc = Sy0 / Agr
zgc = 72180 / 729.1
zgc = 99.0 mm
zgc = zb1 = zb2 = 99.0 mm
Determine the y coordinate of the centroid.
Part (yi + yi-1).dAi / 2
i mm3
1 2961
2 3890
3 0.0
4 3890
5 2961
Total 13700
32
Discuss me ...
First moment of area for the z axis is the total from above,
Sz0 = 13700 mm3.
The y coordinate of the centroid is given by:
ygc = Sy0 / Agr
ygc = 13700 / 729.1
ygc = 18.8 mm
ygc = yb1 = 18.8 mm
ylip = bp1 - yb1
ylip = 63.0 – 18.8
ylip = 44.2 mm
Second Moment of Area
Second moment of area about major axis.
Part (zi2 +zi-12 +zi.zi-1).dAi / 3 3-1-3/ Annex C
i mm4
1 9024
2 0.0
3 5072000
4 4842000
5 1628000
Total 11550000
The second moment of area about the y axis with respect to origin (bottom
flange) is the total from above,
Iy0 = 11550000 mm4
The second moment of area about the y axis with respect to the centroid is
given by:
Iy = Iy0 – Agr zgc2
Iy = 11550000 – 729.1  99.02
Iy = 4404000 mm4
Second moment of area about minor axis.

Part (yi2 +yi-12 +yi. yi-1).dAi / 3 3-1-3/ Annex C
i mm4
1 186500
2 163400
3 0.0
4 163400
5 186500
Total 699800
The second moment of area about the z axis with respect to origin (web) is the
total from above,
Iz0 = 699800 mm4
The second moment of area about the z axis with respect to centroid is given by:
Iz = Iz0 – Agr ygc2

Iz = 699800 – 729.1 × 18.82
Iz = 442100 mm4
33
Discuss me ...
Radius of Gyration
Radius of gyration for major axis
iy = (Iy / Agr)0.5
iy = (4404000 / 729.1)0.5
iy = 77.7 mm
Radius of gyration for minor axis
iz = (Iz / Agr)0.5
iz = (442100 / 729.1)0.5
iz = 24.6 mm
Elastic Section Modulus
Elastic section modulus for major axis
Wy = Iy / Max{ zb1 zb2}
Wy = 4404000 / 99.0
Wy = 44480 mm3
Elastic section modulus for minor axis
Wz = Iz / Max{ ygc ylip}
Wz = 442100 / 44.2
Wz = 10000 mm3
Shear Centre
The position of the shear centre is calculated with respect to the centroid. The 3-1-3/ Annex C
section is symmetrical about the y axis. Hence, the shear centre will be a point
on the major axis. The terms are defined on the figure below.
z
Shear Centre
y y
ysc
zsc
y0
Part (2.yi-1.zi-1 + 2.yi.zi + yi-1.zi + yi.zi-1) dAi / 6 3-1-3/ Annex C

i mm4
1 35530
2 0.0
3 0.0
4 770300
5 550700
Total 1357000
34
Discuss me ...
Product moment of area with respect to origin is the total from above,
Iyz0 = 1357000 mm4.
Product moment of area with respect to centroid is given by:
Iyz = Iyz0 – (Sy0.Sz0 / Agr)
Iyz = 1357000 – (72180 × 13700 / 729.1)
Iyz = 716.9 mm4
Sectorial Coordinates 3-1-3/ Annex C

Part ω0 i ωi
i yi-1.zi  yi.zi-1 ωi-1 + ω0i
0 na 0
1 1512 1512
2 0.0 1512
3 0.0 1512
4 12470 13980
5 1512 15490
Sectorial Constants
i Iyω0 Izω0 Iωω0 (ωi-1 + ωi)dAi /
2
1 2.24 ×106 2.84 ×105 3.58 ×107 3.553 ×104
2 5.88 ×106 0.00 2.82 ×108 1.867 ×105
3 0.00 5.81 ×107 8.87 ×108 5.868 ×105

4 3.82 ×107 1.89 ×108 9.02 ×109 9.566 ×105
5 4.37 ×107 1.29 ×108 1.02 ×1010 6.925 ×105
Total 9.00 ×107 3.76 ×108 2.04 ×1010 2.458 ×106
Iω = Σ(ωi-1 + ωi) dAi / 2

Iω = –2458000 mm3
Iyω = Iyω0 – Sz0 × Iω / Agr
Iyω = –9.00×107 – 13700 × (–2458000) / 729.1
Iyω = –43710000 mm4
Izω = Izω0 – Sy0 × Iω / Agr
Izω = –3.76×108 – 72180 × (–2458000) / 729.1
Izω = –132700000 mm4
Iωω = Iωω0 – Iω2 / Agr
Iωω = 2.04×1010 – (–2458947)2 / 729.1
Iωω = 1.211×1010 mm4
The y coordinate of the shear centre is given by: 3-1-3/ Annex C
ysc = (Izω Iz – Iyω Iyz) / (Iy Iz – Iyz2)
ysc = (–132700000  442100 – (–43710000)  716.9) /
(4404000  442100 – 716.92)
ysc = –30.10 mm
The distance of the shear centre from the centroid is given by:
y0 = –ysc + ygc
y0 = 30.10 + 18.8
y0 = 48.9 mm
35
Discuss me ...
The z coordinate of the shear centre is at mid height.

zsc = 99.0 mm
Warping Constant
The warping constant is given by:
Iw = Iωω + zsc × Iyω – ysc × Izω 3-1-3/ Annex C
Iw = 1.211×1010 + 99.0 × –43710000 – –30.10 × –132700000
Iw = 3.797 ×109 mm6
Torsion Constant
Part dAi.(ti)2 / 3 3-1-3/ Annex C
i mm4
1 60.2
2 158.1
3 497.0
4 158.1
5 60.2
Total 933.6
The torsion constant is given by the total from the table above.
It = Σ dAi.(ti)2 / 3
It = 933.6 mm4
Summary of Gross Properties for a section with sharp corners

Area Agr = 729.1 mm2
centroid from flange zgc = 99.0 mm
centroid from web ygc = 18.8 mm
centroid from lip ylip = 44.2 mm
Second moment of area about major axis Iy = 4404000 mm4
Second moment of area about minor axis Iz = 442100 mm4
Radius of gyration for major axis iy = 77.7 mm
Radius of gyration for minor axis iz = 24.6 mm
Elastic section modulus for major axis Wy = 44480 mm3
Elastic section modulus for major axis Wz = 10000 mm3
Shear centre from flange zsc = 99.0 mm
Shear centre y from web ysc = –30.10 mm
Shear centre from centroid y0 = 48.9 mm
Warping constant Iw = 3.797 ×109 mm6
Torsion constant It = 933.6 mm4
Gross section properties (rounded corners)
The influence of rounded corners on section properties is taken into account by
reducing the properties calculated for an otherwise similar cross section with
sharp corners, using the following approximations:
Ag  Ag, sharp 1   
 I g, sharp 1  2 
Ig
Iw  I w, sharp 1  4 
36
Discuss me ...
For properties such as the position of the neutral axes and the position of the
shear centre, the values for the section with rounded corners are taken as equal
to those calculated for the section with sharp corners.
The section properties that are reduced to take account of rounded corners are
calculated below.
Ag  Ag, sharp 1     729.1(1  0.014)  718.9 mm2
Iy  I y,sharp 1  2   4404000(1  2  0.014)  4281000 mm4
Iz  I z,sharp 1  2   442100(1  2  0.014)  429700 mm4
iy  ( I y / Ag ) 0.5  (4281000 / 718.9) 0.5  77.2 mm

iz  ( I z / Ag ) 0.5  (429700 / 718.9) 0.5  24.5 mm
Wy  I y / zgc  4281000 / 99.0  43240 mm3
Wz  I z / ylip  429700 / 44.2  9724 mm3
Iw  I w, sharp 1  4   3.797  109 (1  4  0.014)  3.584  109 mm6
Summary of Gross Properties for a section with rounded corners

centroid from flange zgc = 99.0 mm
centroid from web ygc = 18.8 mm

centroid from lip ylip = 44.2 mm
Radius of gyration for major axis iy = 77.2 mm
Radius of gyration for minor axis iz = 24.5 mm
Elastic section modulus for major axis Wy = 43240 mm3
Elastic section modulus for major axis Wz = 9724 mm3
Shear centre from flange zsc = 99.0 mm
Shear centre y from web ysc = –30.10 mm
Shear centre from centroid y0 = 48.9 mm
37
Discuss me ...
Title Example 2
Silwood Park, Ascot, Berks SL5 7QN Subject Effective properties of a lipped C section in compression
Telephone: (01344) 636525
Client SCI
8.2 Effective section properties for a lipped C section

in compression
This worked example presents design calculations for the effective section
properties of a cold-formed lipped C section in compression. The chosen section
is identical to that considered in Example 1, so some of the calculations and
checks relating to the gross cross section have been omitted.
It was shown in Example 1 that the influence of rounded corners must be
included for this cross-section. The section properties are initially calculated on
the basis that the section has sharp corners. The influence of rounded corners is
taken into account by applying reductions to the sharp corner properties.

Flange widt b1 = b2 = 65mm
Design strength fy = 350 N/mm2
Young’s modulus E = 210000 N/mm2
Poisson’s ratio v = 0.3
Partial safety factor M0 = 1.00
Mid-line dimensions

Flange width bp1 = b p 2  b1  t nom  65  2  63 mm
Effective section properties (sharp corners)
Effective properties of the flanges and lips 3-1-3/ 5.5.3.2

Step 1
Effective width of the flanges 3-1-3/ 5.5.2
For a stress ratio   1 (uniform compression), k σ  4

 = 235 f yb
38
Discuss me ...
Example 2: Effective properties of a lipped C section in compression Sheet 2 of 5 Rev
bp1 t 63 1.96 3-1-5/ 4.4

p, b    0.691
28.4  k σ 28.4  235 350  4
p, b  0.0553    0.691  0.055  3  1
  2
  0.986  1.0
p, b 0.6912
beff   bp1  0.987  63  62.2 mm
be1  be2  0.5beff  0.5  62.2  31.1 mm
Effective width of the edge stiffener 3-1-3/ 5.5.3.2(5a)
The buckling factor is given by:

if bp,c bp1  0.35 : kσ  0.5
kσ  0.5  0.83 3 bp,c bp1  0.35
2
if 0.35  bp,c bp1  0.6 :
k σ  0.5  0.833 0.38  0.35  0.582

2
bp,c bp1  24 63  0.38 so
cp t 24 1.96 3-1-5/ 4.4

p, c    0.690
28.4  kσ 28.4  235 350  0.58
p, c  0.188 0.690  0.188

  2
  1.05
p, c 0.690 2
but   1 so  1
The effective width is given by: 3-1-3/ 5.5.3.2(5a)
ceff   cp  1  24  24 mm 5.5.3.2(6)
The effective area of the edge stiffener is:
As  t be2  ceff   1.96  31.1  24   108.0 mm 2
Step 2
The elastic critical buckling stress for the edge stiffener is given by: 3-1-3/ 5.5.3.2(7)
2 K E Is
 cr,s 
As
where K is the spring stiffness per unit length and Is is the effective second
moment of area of the stiffener.
E t3 1 3-1-3/ 5.5.3.1(5)
K   2
4(1   ) b1 hp  b1  0,5 b1 b2 hp k f
2 3
be2t be2 2 31.1  1.96  31.1 2

b1  bp1   63   54.22 mm
(be2  ceff )t (31.1  24)  1.96
b2  b1  54.22 mm (for a section with equal flanges)
As2 108
kf    1.0 for a member in axial compression
As1 108
K  0.421 N mm 2
39
Discuss me ...
2 2
b t3 c 3 t  ceff 2  c  ceff 2 
Is  e2  eff  be2 t    ceff t  eff   
12 12  2be2  ceff   2  2be2  ceff   
 6101 mm 4
As the section has equal flanges, the spring stiffness K and second moment of
area Is are applicable to both edge stiffeners. Had the section been asymmetric,
it would have been necessary to repeat the process shown above for the upper
and lower edge stiffeners.
2  0.421  210000  6101
 cr,s   430.1 N mm 2
108.0
d  f y  cr,s  350 430  0.902 3-1-3/ 5.5.3.1(7)
Since 0.65  d  1.38 ,  d  1.47  0.723 d

d  1.47  0.723  0.902  0.818
Step 3
3-1-3/ permits the optional iteration to refine the value of χd. This iteration has
not been undertaken for this example, so the initial value must be used.
t red  t d  1.96  0.818  1.60 mm 3-1-3/ 5.5.3.2(12)
Effective properties of the web
For uniform compression, the stress ratio   1 and the buckling factor k σ  4 3-1-5/ 4.4
(for an internal compression element).
hp t 198 1.96
p, h    2.172
28.4  kσ 28.4  235 350  4
p, h  0.0553    2.171  0.055  3  1

  2
  0.414
p, h 2.1712
heff  hp  0.414  198  82.0 mm
he1  he2  0.5heff  0.5  82.0  41.0 mm

40
Discuss me ...
Effective properties of the whole cross section

The theoretical effective section is shown below. The effective properties of the
whole section are calculated using this theoretical effective section.
b e1 b e2
c eff
h e1 t red
y y
h e2 t red
c eff
b e1 b e2
Aeff  t 2be1  he1  he2  2be2  ceff  d 

Aeff  1.962  31.1  41.0  41.0  231.1  24.0 0.816
Aeff  459.0 mm 2
Since the section is symmetrical and subjected to pure axial compression, the
position of the centroidal axis remains unchanged from that of the gross cross
section, i.e. 99.0 mm from either flange.
Other effective properties for the section subject to axial load are given below
but the calculation of these properties is not shown.
Summary of Effective Properties for a section with sharp corners
Effective area Aeff = 459.0 mm2
Effective centroid from flange zgc = 99.0 mm
Effective centroid from web ygc = 25.0 mm
Effective centroid from lip ylip = 38.0 mm
41
Discuss me ...
Effective section properties (rounded corners)
Influence of rounded corners

The influence of rounded corners must be included for this section. For
properties such as the position of the neutral axes, the values for the section
with rounded corners are taken as equal to those calculated for the section with
sharp corners.
calculated below.
Aeff  Aeff,sharp 1     459.0(1  0.014)  452.6 mm2
Iy  I y,sharp 1  2   3780000(1  2  0.014)  3675000 mm4
Iz  I z, sharp 1  2   311800(1  2  0.014)  303100 mm4
Summary of Effective Properties for a section with rounded corners

Effective centroid from flange zgc = 99.0 mm
Effective centroid from web ygc = 25.0 mm
Effective centroid from lip ylip = 38.0 mm
42
Discuss me ...
Title Example 3
Subject
Effective properties of a lipped C section in bending
Silwood Park, Ascot, Berks SL5 7QN
Telephone: (01344) 636525
about the major axis
Fax: (01344) 636570 Made by MDH Date Oct 2010
Client SCI
8.3 Effective properties for a lipped C section in major

axis bending
This worked example presents design calculations for effective section
properties of a cold-formed lipped C section in bending about the major axis.
taken into account by apply reductions to the sharp corner properties.

Flange width b1 = b2 = 65mm
Young’s modulus E = 210000 N/mm2
Poisson’s ratio v = 0.3
Mid-line dimensions

The effective cross section for a lipped C section in bending is shown below. 3-1-3/ 5.5
Note the ineffective portions of the flange and web and the reduced thickness
t d of the stiffener and adjacent section of flange.
43
Discuss me ...
Example 3: Effective properties of a lipped C section in bending Sheet 2 of 5 Rev
The effective properties of the flange and web are determined separately as
shown below, after which the effective properties of the whole cross section may
be calculated.
Effective properties of the compression flange and lip 3-1-3/ 5.5.3.2
Step 1
Effective width of the compression flange 3-1-3/ 5.5.2

and
For a stress ratio   1 (uniform compression), kσ  4 3-1-5/ 4.4
  235 f y
bp1 t 63 1.96
p, b    0.691
28.4  k σ 28.4  235 350  4
p, b  0.0553    0.691  0.055  3  1 3-1-5/ 4.4

    0.986  1.0
p, b 2 0.6912
beff   bp1  0.986  63  62.1 mm

be1  be2  0.5beff  0.5  62.1  31.1 mm
Effective width of the edge stiffener 3-1-3/ 5.5.3.2(5a)
The buckling factor is given by:

if bp,c bp1  0.35 : kσ  0.5
kσ  0.5  0.83 3 bp,c bp1  0.35
2
if 0.35  bp,c bp1  0.6 :
bp,c bp1  24 63  0.38 so k σ  0.5  0.83 3

0.38  0.352  0.582
cp t 24 1.96 3-1-5/ 4.4

p, c    0.690
28.4  kσ 28.4  235 350  0.58
p, c  0.188 0.690  0.188 3-1-5/ 4.4

    1.05
p, c 2 0.690 2
but   1 so  1
44
Discuss me ...
The effective width is given by: 3-1-3/ 5.5.3.2(5a)

ceff   cp  1  24  24 mm 5.5.3.2(6)
The effective area of the edge stiffener:

As  t be2  ceff   1.96  31.1  24   108.0 mm 2
Step 2
The elastic critical buckling stress for the edge stiffener is given by: 3-1-3/ 5.5.3.2(7)
2 K E Is
 cr,s 
As
where K is the spring stiffness per unit length and Is is the effective second
moment of area of the stiffener.
E t3 1 3-1-3/ 5.5.3.1(5)
K   2
4(1   ) b1 hp  b1  0.5 b1 b2 hp kf
2 3
be2t be2 2 31.1  1.96  31.1 2

b1  bp1   63   54.22 mm
(be2  ceff )t (31.1  24)  1.96
kf  0 (for major axis bending)
K  0.586 N mm
2 2
b t3 c t
3  ceff
2  c ceff
2 
Is  e2  eff  be2 t    ceff t  eff  
12 12  2be2  ceff    2 2be2  ceff  
 6101 mm 4
2  0.586  210000  6101 3-1-3/ 5.5.3.2(7)

 cr,s   507.4 N mm 2
108.0
d  f y  cr,s  350 507.4  0.831 3-1-3/ 5.5.3.1(7)
Since 0.65  d  1.38 ,  d  1.47  0.723 d

 d  1.47  0.723  0.831  0.869
Step 3
3-1-3/5.5.3.2(3) permits the optional iteration to refine the value of  d . This

iteration has not been undertaken for this example, so the initial value of  d and
the associated effective properties must be used. The next and, therefore, final
step for the flange is the calculation of the reduced thickness for the stiffener.
t red  t d  1.96  0.870  1.70 mm 3-1-3/ 5.5.3.2(12)
Effective properties of the web

The position of the neutral axis with regard to the flange in compression:
hc 
  2 2
cp hp  cp 2  bp2 hp  hp 2  ceff  d 2
hc  101.1 mm
cp  bp2  hp  be1  be2  ceff  d
45
Discuss me ...
The stress ratio is given by:

hc  hp 101.1  198
    0.958
hc 101.1
Referring to EN 1993-1-5, the buckling factor for the web is given by: 3-1-5/ 4.4
kσ  7.81  6.29  9.78 2 k σ  22.81
hp t 198 1.96
p, h    0.909
28.4  k σ 28.4  235 350  22.81
p, h  0.0553    0.909  0.055  3  0.958

  2
  0.964
p, h 0.909 2
heff   hc  0.965  101.1  97.5 mm

he1  0.4heff  0.4  97.5  39.0 mm
he2  0.6heff  0.6  97.5  58.5 mm
The effective width of the web is divided into two portions as follows:
h1  he1  39.0 mm
h2  hp  hc  he2   198  101.1  58.5  155.4 mm
Aeff  t[cp  bp 2  h1  h2  be1  (be 2  ceff )  d ]

Aeff  1.96  24  63  39  155.4  31.08  31.08  24   0.870
Aeff  706.3 mm 2
The position of the neutral axis with regard to the compression flange is given
by:
zc =
    
t cp hp  cp 2  bp2 hp  h2 hp  h2 2  h1 2  ceff  d 2
2 2
 = 101.7 mm
Aeff
The position of the neutral axis with regard to the tension flange is given by:
zt =  h p  zc  198  101.7  96.3mm
h13t h2 3t bp2t
3
cp 3t be1t 3 be2 (  d t ) 3 ceff 3 (  d t )
I eff, y        
12 12 12 12 12 12 12
cp t ( z t  cp 2) 2  bp 2tz t 2  h2t ( z t  h2 2) 2  h1t ( zc  h1 2) 2 
2 2
be1t zc  be2 (  d t ) zc  ceff (  d t )( zc  ceff 2) 2
I eff, y  4235000 mm 4
I eff, y 4235000
Weff, y,c    41640 mm3
zc 101.7
I eff, y 4235000
Weff, y, t    43980 mm3

zt 96.3
46
Discuss me ...

Effective centroid from compression flange zgc = 101.7 mm
Effective centroid from tension flange zgc = 96.3 mm
Second moment of area about major axis Ieff,y = 4235000 mm4
Section modulus about major axis compression flange Weff,y,c = 41640 mm3
Section modulus about major axis tension flange Weff,y,t = 43980 mm3

The influence of rounded corners must be included for this section. For
properties such as the position of the neutral axes, the values for the section
with rounded corners are taken as equal to those calculated for the section with
sharp corners.
calculated below.
Iy  I y,sharp 1  2   4235000(1  2  0.014)  4116000 mm4
I eff, y 4116000
Weff, y,c    40480 mm3
zc 101.7
I eff, y 4116000
Weff, y, t    42740 mm3
zt 96.3

Effective centroid from compression flange zgc = 101.7 mm
Effective centroid from tension flange zgc = 96.3 mm
Second moment of area about major axis Ieff,y = 4116000 mm4
Section modulus about major axis compression flange Weff,y,c = 40480 mm3
Section modulus about major axis tension flange Weff,y,t = 42740 mm3
47
Discuss me ...
Title Example 4
Subject
Effective properties of a lipped C section in bending
Silwood Park, Ascot, Berks SL5 7QN
Telephone: (01344) 636525
about the minor axis
Fax: (01344) 636570 Made by MDH Date Oct 2010
Client SCI
8.4 Effective properties for a lipped C section in

minor axis bending
This worked example presents the design procedure for the calculation of the
effective section properties of a cold-formed lipped C section in bending about
the minor axis such that the web is in compression.
taken into account by apply reductions to the sharp corner properties.

Flange width b1 = b2 = 65mm
Mid-line dimensions

The effective properties of the flange and web are determined separately as
shown below, after which the effective properties of the whole cross section may
be calculated.
48
Discuss me ...
h e1
hc
h e2
z z
h2
Effective properties of the web in compression 3-1-3/ 5.5.3.2
Step 1
Effective width of the web in compression 3-1-3/ 5.5.2

and
For a stress ratio   1 (uniform compression), kσ  4 3-1-5/ 4.4
  235 f y
hp t 198 1.96
p, b    2.172
28.4  k σ 28.4  235 350  4
p, b  0.0553    2.172  0.055  3  1 3-1-5/ 4.4

    0.414  1.0
p, b 2 2.172 2
heff  h p  0.414  198  82.0 mm
Effective width of the part of the flange in compression

The effective area of the section considering only the web as having an
ineffective part is calculated.
Aeff = Agr – (hp – heff).t
Aeff = 729.1 – (198.0 – 82.0) × 1.96
Aeff = 501.7 mm2
The position of the effective neutral axis with regard to the web in compression
is calculated.
hc = t (2.bc.cc + 2.bc.(bc/2)) / Aeff
hc = 1.96 (2 × 63.0 × 24.0 + 2 × 63.0 (63.0/2)) / 501.7
hc = 27.32 mm
The stress ratio ψ is given by: 3-1-5/ Table 4.1
ψ = (hc – bc) / hc
ψ = (27.32 – 63.0) / 27.32
ψ = –1.31
For ψ between –1 and –3, the buckling factor is given by:
kσ = 5.98 (1 – ψ)2
kσ = 5.98 (1 – –1.31)2
kσ = 31.91
49
Discuss me ...
bp / t 63.0 / 1.96 3-1-5/ 4.4

p,b = = = 0.245
28.4 k σ 28.4  235 / 350  31.91
p, b  0.0553    0.245  0.055  3  1.31

 = 2
  2.533
p, b 0.2452
but   1 so  1
The flanges are fully effective.
The effective width of the zone in compression of the web is:
heff = ρ × hc
heff = 1.0 × 27.32 = 27.32 mm
The effective width of the flange near the web in compression is:
he1 = 0.4 heff = 10.93 mm
The effective width of the flange near the lips in tension:
h2 = bp – hc + 0.6 heff = 52.07 mm
The lips are fully effective because they are in tension.

Aeff = 501.7 mm2
The position of the neutral axis with regard to the web in compression is given
by:
yc 
 
t 2 cp bp  2bp bp / 2  = 27.3 mm
Aeff
The position of the neutral axis with regard to the lips in tension is given by:
yt  b p  yc  63  27.3  35.7 mm
Other effective properties for the section subject to minor axis bending are given
below but the calculation of these properties is not shown.
Second moment of area about minor axis Iz,eff = 325800 mm4
Effective elastic modulus about minor axis Weff,z = 9126 mm3
Effective centroid from web yc = 27.3 mm
Effective centroid from lips yt = 35.7 mm
50
Discuss me ...

The influence of rounded corners must be included for this cross-section. For
properties such as the position of the neutral axes, the values for the section with
rounded corners are taken as equal to those calculated for the section with
sharp corners.
calculated below.
Iz  I z, sharp 1  2   325800(1  2  0.014)  316700 mm4

Effective centroid from web yc = 27.3 mm
Effective centroid from lips yt = 35.7 mm
51
Discuss me ...
Title Example 5
Silwood Park, Ascot, Berks SL5 7QN Subject Design of a wall stud
Telephone: (01344) 636525
Client SCI
8.5 Design of a wall stud in a light steel frame building

This worked example presents the design of a wall stud in a light steel frame
building. Ultimate limit state and serviceability limit state design checks are
included.
Wall details
The wall stud is located in a load bearing external. Therefore, the stud is
subjected to wind loads and gravity loads. The wall stud is restrained at mid
height by a lateral restraint within the wall construction.
Dimensions
Stud height L = 2700 mm
Stud spacing s = 600 mm
Actions
Design value of axial force NEd = 4.5 kN

Characteristic wind load qk = 1.2 kN/m2
Maximum design moment due to wind My,Ed = 0.98 kNm
The wall stud is a lipped C section, manufactured from S350 steel with a Z275
coating to BS EN 10346.
Flange width b = 45 mm
Stiffener depth c = 12.0 mm
Corner radius r = 1.5 mm
Nominal thickness tn = 1.2 mm
Basic yield strength fyb = 350 N/mm2
Modulus of elasticity E = 2100000 N/mm2
Shear modulus G = 80770 N/mm2
Partial factor γM0 = 1.0
Section Properties
The calculation of section properties is not included in this example. See
Examples 1 to 4 for procedure of calculation of gross and effective section
properties.
52
Discuss me ...
Example 5: Design of a wall stud Sheet 2 of 8 Rev
Gross Properties
Radius of gyration about y axis iy = 40.2 mm
Radius of gyration about z axis iz = 16.8 mm
Position of y axis from flange yflange = 49.4 mm
Position of z axis from web zweb = 13.9 mm
Position of shear centre with respect to the z axis yo = 35.0 mm
Position of shear centre with respect to the y axis zo = 0.0 mm
Second moment of area about y axis Iy = 392000 mm4
Effective Section Properties
Effective area subject to compression Aeff = 153.6 mm2
Elastic section modulus for bending about y axis Weff,y = 6929 mm3
Elastic section modulus for bending about z axis Weff,z = 2101 mm3
Position of y axis from flange (due to compression) yflange = 49.4 mm
Position of z axis from web (due to compression) zweb = 15.5 mm
Second moment of area about y axis (due to bending) Iy,eff = 360800 mm4
Resistance of Cross-Section
Axial Compression
Design resistance of cross section for compression is given by: 3-1-3/ 6.1.3
Nc,Rd = Aeff × fyb / γM0
Nc,Rd = 153.6 × 350 ×10-3/ 1.0 = 53.8 kN
Bending
Design moment resistance for bending about y axis is given by: 3-1-3/ 6.1.4
Mcy,Rd = Weff,y × fyb / γM0
Mcy,Rd = 6929 × 350 ×10-6/ 1.0 = 2.4 kNm
Design moment resistance for bending about z axis:

Mcz,Rd = Weff,z × fyb / γM0
Mcz,Rd = 2101 × 350 ×10-6/ 1.0 = 0.7 kNm
Combined Compression and Bending
The shift in the position of the effective section neutral axis relative to gross 3-1-3/ 6.1.9
section neutral axis causes an additional moment due to axial load which is
assumed to be applied at the gross section neutral axis.
The shift in y axis due to axial compression is given by:
eNy = 49.4 – 49.4 = 0.0 mm
The shift in z axis due to axial compression is given by:
eNz = 15.5 – 13.9 = 1.6 mm
Additional y axis moment due to shift in axis:

ΔMy,Ed = eNy × Ned = 0.000 kNm
53
Discuss me ...
Additional z axis moment due to shift in axis:

ΔMz,Ed = eNz × Ned = 4.5 × 1.6 ×10-3 = 0.007 kNm
N Ed M y, Ed  M y, Ed M z, Ed  M z, Ed 3-1-3/ 6.1.9
  1
N c, Rd M cy,Rd M cz, Rd
4.5 0.98  0.0 0  0.007
  1
53.8 2.4 0.7
0.08 + 0.41 + 0.01 = 0.50 ≤ 1.0 OK
Buckling Resistance of Member
Flexural Buckling about major axis (y axis)

From 3-1-3/Table 6.3, the buckling curve for a lipped C section buckling about 3-1-3/ 6.2.2
any axis is buckling curve b.
The buckling length for buckling about the y axis is taken as equal to the
member length.
Lcr,y = 2700 mm
E 210000 3-1-1/ 6.3.1.3

1 = = = 76.95
fy 350
Non dimensional slenderness factor

Aeff 153.6
L A gr 2700 242.7
 = cr  = = 0.695
i 1 40.2 76.95
For buckling curve b, the imperfection factor α is 0.34 3-1-1/ Table 6.1
  
= 0.5 1     0.2   2
 = 0.51  0.340.695  0.2  0.695  = 0.825

2
1
 = ≤ 1.0
    2 2
1
 = = 0.787
0.825  0.825 2  0.695 2
Flexural buckling resistance

 Aeff f y
Nb,Rd =
 M1
0.787  153.6  350  10 3
Nb,Rd = = 42.3 kN
1.0
Validation
NEd / Nb,Rd = 4.5 / 42.3 = 0.11 < 1.0 OK
54
Discuss me ...
Flexural Buckling about minor axis (z axis)

The buckling length for buckling about the z axis is taken as half the member
length because there is a mid-height lateral restraint in the wall.
Lcr,z = 2700 / 2 = 1350 mm
E 210000 3-1-1/ 6.3.1.3

1 = = = 76.95
fy 350
Non dimensional slenderness factor

Aeff 153.6
Lcr Agr 1350 242.7
 =  = = 0.830
i 1 16.8 76.95
  
= 0.5 1     0.2   2
 = 0.51  0.340.830  0.2   0.830 

2
= 0.952
1
 = ≤ 1.0
   2 2
1
 = = 0.706
0.952  0.952 2  0.830 2
Flexural buckling resistance

 Aeff f y
Nb,Rd =
 M1
0.706  153.6  350  10 3
Nb,Rd = = 37.9 kN
1.0
Validation
NEd / Nb,Rd = 4.5 / 37.9 = 0.12 < 1.0 OK
Torsional Buckling
The torsional buckling lengths are:
LT,y = 2700 mm
LT,z = 1350 mm
The polar radius of gyration is calculated as below:
io2 = iy2 + iz2 + yo2 + zo2
io2 = 40.22 + 16.82 + 35.02 + 0.02 = 3123 mm2
io = 55.9 mm
55
Discuss me ...
The elastic critical force for torsional buckling of a simply supported member is 3-1-3/ 6.2.3(5)
given by:
1   2E Iw 

Ncr,T = G I 
io2  
t
LT2 
 8 
 80770  108.8    210000  1.41  10  = 54.27 kN
2
1
Ncr,T =
55.9 2  1350 2 
 
Non dimensional slenderness factor: 3-1-1/ 6.3.1.4

Aeff f y
T =
N cr
153.6  350  10 3
T = = 0.995
54.27
For buckling curve b, the imperfection factor α is 0.34 3-1-1/ 6.3.1.2
T  
= 0.5 1     0.2   2
T = 0.51  0.340.995  0.2  0.995  2

= 1.131
1
T = ≤ 1.0
    2 2
1
T = = 0.600
1.131  1.1312  0.995 2
Torsional buckling resistance

 T Aeff f y
Nb,Rd =
 M1
0.600  153.6  350  10 3
Nb,Rd = = 32.3 kN
1.0
Validation
NEd / Nb,Rd = 4.5 / 32.3 = 0.14 < 1.0 OK
Torsional-Flexural Buckling
 2E Iy 3-1-3/ 6.2.3(7)
Ncr,y =
L2cr,y
 2  210000  392000  10 3
Ncr,y = = 111.5 kN
2700 2
2 2
y   35.0 
 = 1   o   1   = 0.608
 io   55.9 
56
Discuss me ...
The elastic critical force for torsional-flexural buckling is given by: 3-1-3/ 6.2.3(7)
 2 2 
N cr,y  N cr,T  N cr,T   yo  N cr,T 
Ncr,TF = 1  1    4 
i  N
2   N cr,y  N  
  cr, y   o  cr, y

111.5  54.27 
2 2
 54.27   35.0  54.27 
Ncr,TF = 1  1    4  = 43.41 kN
2  0.608  111.5  111.5   55.9  111.5 
 
Aeff f y
TF =
N cr
153.6  350  10 3
TF = = 1.113
43.41
TF  
= 0.5 1     0.2   2
TF = 0.51  0.341.113  0.2  1.113 

2
= 1.274
1
TF = ≤ 1.0
   2 2
1
TF = = 0.528
1.274  1.274 2  1.1132
Torsional-flexural buckling resistance:

 TF Aeff f y
Nb,Rd =
 M1
0.528 153.6  350  10 3
Nb,Rd = = 28.4 kN
1.0
Validation
NEd / Nb,Rd = 4.5 / 28.4 = 0.16 < 1.0 OK
Lateral Torsional Buckling

The maximum length of member between points of minor axis lateral restraint is
half the member length.
L = 1350 mm
13]
Coefficients dependent on loading and end restraint: SN003[
C1 = 1.127
C2 = 0.454
Effective length factor for end rotation on plan k = 1.00

Effective length factor for end warping kw = 1.00
57
Discuss me ...
The distance from the point of load application to the shear centre is taken as
half the stud depth.
zg = 100 / 2 = 50 mm
The factor g is used in the calculation of Mcr, it may conservatively be taken as
14]
1.0 or may be calculated as below. SN002[
I 68550
g = 1  z g = 1 = 0.908
Iy 392000
The elastic critical moment for lateral torsional buckling is given by:
 
π 2 EI z   k  I w kL  GI t
2 2
Mcr = C1 2  
   2 
2

 C2 zg  C2 zg 

kL  g   k w  I z π EI z 
 
Mcr = 2.82 kNm
For lateral torsional buckling curve b should be used. 3-1-3/ 6.2.4
Wy,eff f y
LT =
M cr
6929  350  10 6
LT = = 0.927
2.82
LT  
= 0.5 1     0.2   2
LT = 0.51  0.340.927  0.2  0.927  2

= 1.054
1
LT = ≤ 1.0
   2 2
1
LT = = 0.644
1.054  1.054 2  0.927 2
Lateral torsional buckling resistance:

 lT Weff,y f y
M b,Rd 
 M1
0.644  6929  350  10 6
M b,Rd  = 1.56 kNm
1.0
Validation
MEd / Mb,Rd = 0.98 / 1.56 = 0.63 < 1.0 OK
58
Discuss me ...
Combined Bending and Axial Compression

The interaction of bending moment and axial force may be obtained from a 3-1-3/ 6.2.5
second order analysis or by using the formula below.
0.8 0.8
 N Ed   M 
    Ed   1.0
N  M 
 b, Rd   b, Rd 
Nb,Rd is the minimum value from flexural, torsional and torsional-flexural
buckling.
MEd includes any additional moments due to the shift in the neutral axis.
0.8 0.8
 4.5   0.98  0.0 
     0.23 + 0.69 = 0.92 < 1.0 OK
 28.4   1.56 
Serviceability Deflections
For cross section stiffness properties the influence of rounded corners should 3-1-3/ 5.1(3)
always be taken into account. For this example it is assumed that the maximum
stress at serviceability is the design yield strength divided by 1.5.
3-1-3/ 7.1(3)
Ific
 1 

= I gr    I gr  I eff 
 1.5 
 1 
Ific = 392000   392000  360800 = 371200 mm4
 1.5 
The corner factor δ is given by: 3-1-3/ 5.1(4)
n
 rj 4  1.5
j 1
 = 0.43 = 0.43 = 0.01
m
(98.8  2  43.8  2  11.4)
 bp,i
i 1
Taking account of rounded corners, the second moment of area for

serviceability is given by:
ISLS = I (1 – 2δ) = 371200 (1 – 0.02) = 363800 mm4
Total wind load:
W = L s qk = 2700 × 600 × 1.2 × 10-6 = 1.94 kN
Deflection due to wind is given by;
5 WL3 5 1.94  2700 3 10 3
wind =  = 6.5 mm
384 EI SLS 384 210000  363792
Deflection limit is taken as length divided by 360,
δlimit = 2700 / 360 = 7.5 mm > 6.5 mm OK
59
Discuss me ...
Title Example 6
Silwood Park, Ascot, Berks SL5 7QN Subject Design of a floor joist (restrained)
Telephone: (01344) 636525
Client SCI
8.6 Design of a floor joist (restrained) in a light steel

frame building
This worked example presents the calculations for a restrained floor joist in a
light steel frame building. Ultimate limit state and serviceability limit state
design checks are included.
Floor details
For this example the floor joist is assumed to be restrained along the length of
the top flange by the connection to the timber floor boarding. The unrestrained
situation during construction is considered in Example 7.
Dimensions
Floor span L = 4800 mm
Joist spacing s = 600 mm
Actions
Permanent action (characteristic) gk = 0.5 kN/m2
Variable action (characteristic) qk = 1.5 kN/m2
Factors on actions
Permanent γG = 1.35
Variable γQ = 1.50
60
Discuss me ...
Example 6: Design of a floor joist (restrained) Sheet 2 of 4 Rev
Section Properties
properties.
Gross Properties
Second moment of area about y axis Ieff,y = 4116000 mm4
Design Moment
Applied design moment is given by,
Mcy,Ed = (γG × gk + γQ × qk) s L2 / 8
Mcy,Ed = (1.35 × 0.5 + 1.50 × 1.5) 600 × 48002 × 10-9 / 8
Mcy,Ed = 5.1 kNm
Bending Moment
Design moment resistance for bending about y axis is given by, 3-1-3/ 6.1.4
Mcy,Rd = 40480 × 350 ×10-6/ 1.0 = 14.2 kNm
Validation
My,Ed / Mcy,Rd = 5.1 / 14.2 = 0.36 < 1.0 OK

For this example the floor joist is assumed to be restrained along the length of
the top flange by the connection to the timber floor boarding. Therefore, the joist
is not required to be checked for lateral torsional buckling.
The joist is unrestrained during construction; this situation is considered in
Example 7.
3-1-3/ 7.1(3)
Ific
 1 
 
= Igr    I gr  I eff
 1.5 
 1 
Ific = 4281000   4281000  4116000 = 4171000 mm4
 1.5 
61
Discuss me ...
The influence of rounded corners has been taken account in the calculation of
the Igr and Ieff values used to calculate Ific. Therefore, the second moment of area
for serviceability is given by;
ISLS = Ific = 4171000 mm4
Serviceability Criteria
For light weight steel floors there are four serviceability criteria that should be
checked to ensure acceptable performance of the floor in service. The criteria
are detailed in SCI publication P301 and in Chapter 6.10 of the NHBC
Standards.
Criterion 1
Dead load plus imposed load deflection less than span/350 or 15 mm whichever
is smaller.
Total load is given by:
W = L s (gk + qk) = 4800 × 600 (0.5 + 1.5) × 10-6 = 5.76 kN
Deflection due to load is given by:
5 WL3 5 5.76  4800 3 10 3
 =  = 9.5 mm
384 EI SLS 384 210000  4171000
Deflection limit:
δlimit = 4800 / 360 = 13.3 mm > 9.5 mm OK

Criterion 2
Imposed load deflection less than span/450.
Total load:
W = L s qk = 4800 × 600 × 1.5 × 10-6 = 4.32 kN
5 WL3 5 4.32  4800 3 10 3
 =  = 7.1 mm
384 EI SLS 384 210000  4171000
Deflection limit:
δlimit = 4800 / 450 = 10.7 mm > 7.1 mm OK
Criterion 3
Natural frequency of the floor not less than 8 Hz.
Total load for this criterion is given by:
W = L s (gk + 0.2 qk) = 4800 × 600 (0.5 + 0.2 × 1.5) × 10-6 = 2.30 kN
5 WL3 5 2.30  4800 3 10 3
 =  = 3.8 mm
384 EI SLS 384 210000  4171000
Deflection limit for 8 Hz is;
δlimit = 5.0 mm > 3.8 mm OK
62
Discuss me ...
Criterion 4
Deflection of floor system less than critical value subject to 1 kN point load.
Total load for this criteria is given by:
W = 1.0 kN
The number of effective joists depends on the floor construction. For this Table 6.1 of SCI-
example the floor construction is taken as chipboard on the floor joists spaced at P301
600 mm.
Neff = 2.35
1 WL3 1 1.0  4800 3 10 3
 =  = 1.1 mm
48 EI SLS N eff 48 210000  4171000  2.35
The deflection limit for this criterion is dependant on the span of the joist. Table 6.2 of SCI-
δlimit = 1.37 mm > 1.1 mm OK P301
63
Discuss me ...
Title Example 7
Silwood Park, Ascot, Berks SL5 7QN Subject Design of a floor joist (unrestrained)
Telephone: (01344) 636525
Client SCI
8.7 Design of a floor joist (unrestrained) in a light steel

frame building
This worked example presents the design calculations for an unrestrained floor
joist during the construction stage in a light steel frame building.
Floor details
Dimensions
Actions
Factors on actions
Variable γQ = 1.50
64
Discuss me ...
Example 7: Design of a floor joist (unrestrained) Sheet 2 of 3 Rev
Section Properties
properties.
Gross Properties
Second moment of area about z axis Iz = 429700 mm4
Warping constant Iw = 3.57 × 109 mm6
Design Moment
Applied design moment is given by,
My,Ed = (γG × gk + γQ × qk) s L2 / 8
My,Ed = (1.35 × 0.3 + 1.50 × 0.7) 600 × 48002 × 10-9 / 8
My,Ed = 2.5 kNm
Bending Moment
Design moment resistance for bending about y axis is given by, 3-1-3/ 6.1.4
Mcy,Rd = 40480 × 350 ×10-6/ 1.0 = 14.2 kNm
Validation
My,Ed / Mcy,Rd = 2.5 / 14.2 = 0.18 < 1.0 OK

The joist is unrestrained during construction. Therefore, it must be checked for
lateral torsional buckling.
The maximum length of member between points of minor axis lateral restraint is
taken as the whole length of the member.
L = 4800 mm
Coefficients dependent on loading and end restraint; SN003
C1 = 1.127
C2 = 0.454
Effective length factor for end rotation on plan k = 1.00
Effective length factor for end warping kw = 1.00
The distance from the point of load application to the shear centre is taken as
half the stud depth.
zg = 200 / 2 = 100 mm
65
Discuss me ...
Example 7: Design of a floor joist (unrestrained) Sheet 3 of 3 Rev
The factor g is used in the calculation of Mcr, it may conservatively be taken as

1.0 or may be calculated as below. SN002
I 429700
g = 1  z g = 1 = 0.948
Iy 4281000
The elastic critical moment for lateral torsional buckling is given by:
 
π 2 EI z   k  I w kL  GI t
2 2
Mcr = C1 2  
 
   C z
2
 C z

2 g
kL  g   k w  I z π EI z
2 2 g

 
Mcr = 3.14 kNm
For lateral torsional buckling curve b should be used. 3-1-3/ 6.2.4
Wy,eff f y
LT =
M cr
40480  350  10 6
LT = = 2.12
3.14
 
= 0.5 1     0.2   2
LT
LT = 0.51  0.342.12  0.2   2.12 

2
= 3.07
1
LT = ≤ 1.0
   2 2
1
LT = = 0.189
3.07  3.07 2  2.12 2
Lateral torsional buckling resistance:

 lT Weff,y f y
Mb,Rd =
 M1
0.189  40480  350  10 6
Mb,Rd = = 2.68 kNm
1.0
Validation
MEd / Mb,Rd = 2.5 / 2.68 = 0.93 < 1.0 OK
66
Discuss me ...
Title Example 8
Silwood Park, Ascot, Berks SL5 7QN Subject Design of a lattice floor truss
Telephone: (01344) 636525
Client SCI
8.8 Design of a lattice floor truss in a light steel frame

building
This worked example presents design calculations for a restrained lattice floor
truss in a light steel frame building. Ultimate limit state and serviceability limit
state design checks are included.
Floor details
Dimensions
Depth of truss DT = 200 mm
Number of bays in truss N = 12
Length of each bay Lbay = 400 mm
Angle of brace member θ = 45o
Length of brace member Lbrace = 283 mm

Actions
Factors on actions
Variable γQ = 1.50
Lattice truss dimensions and material properties

The lattice truss is fabricated from lipped C sections, manufactured from S350
steel with a Z275 coating to BS EN 10346. The same size section is used for the
top chord, bottom chord and brace members. The lattice truss is shown below.
DT
Lbay
67
Discuss me ...
Example 8: Design of a lattice floor truss Sheet 2 of 11 Rev
The section dimensions are:

In this example the nodal eccentricities of the neutral axes of the brace members
with the top and bottom chords is considered in the design. The eccentricity
values used in this example are estimated.
Nodal eccentricity at the top chord etop = 60 mm
Nodal eccentricity at the bottom chord ebottom = 20 mm
nodal eccentricity
nodal eccentricity
Section Properties
properties.
Gross Properties
Area Ag = 263.0 mm2
Position of y axis from flange yflange = 36.7 mm
Position of z axis from web zweb = 12.9 mm
Radius of gyration for y axis iy = 30.4 mm
Radius of gyration for z axis iz = 14.8 mm
Second moment of area about strong axis y-y Igry = 242500mm4
Second moment of area about weak axis z-z Igrz = 57230m4
Position of shear centre with respect to the z axis yo = 31.8 mm
Position of shear centre with respect to the y axis zo = 0.00 mm
68
Discuss me ...

Effective area subject to compression Aeff = 222.3 mm2
Position of y axis from flange (section in compression) yflange = 36.7 mm
Position of z axis from web (section in compression) zweb = 13.2 mm
Modulus for bending about z axis (lips in compression) Weff,z,lip = 2247 mm3
Modulus for bending about z axis (web in compression) Weff,z,web = 2179 mm3
Section modulus for bending about z axis (minimum) Weff,z = 2179 mm3
Second moment of area about weak axis z-z Ieff,z = 51530 mm4
Analysis of forces in lattice truss

A cross section of the lattice truss is shown below.
Centroid of top chord
Chord centroid
from w eb
Depth
Centroid of lattice truss
of truss
Chord centroid
from w eb
Centroid of bottom chord
Maximum moment is given by:

My,Ed = (γG × gk + γQ × qk) s L2 / 8
My,Ed = (1.35 × 0.6 + 1.50 × 1.5) 600 × 48002 × 10-9 / 8 = 5.3 kNm
End reaction is given by:
R,Ed = (γG × gk + γQ × qk) s L / 2
R,Ed = (1.35 × 0.6 + 1.50 × 1.5) 600 × 4800 × 10-6 / 2 = 4.4 kN
Maximum compression in top chord is taken as:
Nc,Ed = My,Ed / (DT – 2 × zweb)
Nc,Ed = 5.3 × 103 / (200 – 2 × 13.2) = 30.5 kN
Maximum tension in bottom chord is taken as:
Nt,Ed = My,Ed / (DT – 2 × zweb)
Nt,Ed = 5.3 × 103 / (200 – 2 × 13.2) = 30.5 kN
Maximum compression in brace member is given by:
Nc,Ed = R,Ed × Lbrace / DT
Nc,Ed = 4.4 × 283 / 200 = 6.2 kN
For the bay in the lattice truss adjacent to the central bay:
Compression in top chord Nc,Ed = 29.6 kN
Tension in bottom chord Nt,Ed = 29.6 kN
Moment from nodal eccentricity in the top chord is taken as:
ΔMe,top = ΔNc,Ed × etop

Me,top = (30.5 – 29.6) × 60 = 0.05 kNm
69
Discuss me ...
Moment from nodal eccentricity in the bottom chord is taken as:

Me,bottom = ΔNc,Ed × ebottom
Me,bottom = (30.5 – 29.6) × 20 = 0.02 kNm
Moment from load applied between nodes in the top chord is taken as:
Mw,top = (γG × gk + γQ × qk) s Lbay2 / 8
Mw,top = (1.35 × 0.6 + 1.50 × 1.5) 600 × 4002 × 10-9 / 8 = 0.04 kNm
Moment from load applied between nodes in the bottom chord is taken as zero.
Mw,bottom = 0.0 kNm
Total moment in top chord Mz,Ed,top = 0.05 + 0.04 = 0.09 kNm
Total moment in bottom chord Mz,Ed,bottom = 0.02 + 0.0 = 0.02 kNm
Axial Tension
The average yield strength may be used to determine tension resistance. 3-1-3/ 3.2.2
fya 
= f yb  f u  f ybk nt2
but 

f u  f yb 
Agr 2
fya = 350  420  350

7  4  1.56 2
but 
420  350
263.0 2
fya = 368.1 N/mm2

Design resistance of cross section for tension is given by: 3-1-3/ 6.1.2
Nt,Rd = Ag × fya / γM0
Nt,Rd = 263.0 × 368.1 ×10-3/ 1.0 = 96.8 kN
Validation (Bottom Chord)
Nt,Ed / Nt,Rd = 30.5 / 96.8 = 0.32 < 1.0 OK
Axial Compression
Design resistance of cross section for compression is given by: 3-1-3/ 6.1.3
Nc,Rd = Aeff × fyb / γM0
Nc,Rd = 222.3 × 350 ×10-3/ 1.0 = 77.8 kN
Validation (Top Chord)
Nc,Ed / Nc,Rd = 30.5 / 77.8 = 0.39 < 1.0 OK
Validation (Brace Member)
Nc,Ed / Nc,Rd = 6.2 / 77.8 = 0.08 < 1.0 OK
Bending Moment
Design moment resistance for bending about y axis is given by: 3-1-3/ 6.1.4
Mcy,Rd = 6297 × 350 ×10-6/ 1.0 = 2.2 kNm
Design moment resistance for bending about z axis
Mcz,Rd = Weff,z × fyb / γM0

Mcz,Rd = 2179 × 350 ×10-6/ 1.0 = 0.8 kNm
70
Discuss me ...
Validation (Top Chord)

Mz,Ed / Mz,Rd = 0.09 / 0.8 = 0.11 < 1.0 OK
Validation (Bottom Chord)
Mz,Ed / Mz,Rd = 0.02 / 0.8 = 0.03 < 1.0 OK
Combined Compression and Bending
The shift in the position of the effective section neutral axis relative to gross 3-1-3/ 6.1.9
section neutral axis causes an additional moment due to axial load which is
assumed to be applied at the gross section neutral axis.
The shift in y axis due to axial compression is given by:
eNy = 36.7 – 36.7 = 0.0 mm
The shift in z axis due to axial compression is given by:
eNz = 13.2 – 12.9 = 0.3 mm
Additional y axis moment due to shift in axis is given by:
ΔMy,Ed = eNy × Ned = 0.000 kNm
Additional z axis moment due to shift in axis is,
ΔMz,Ed = eNz × Ned = 0.3 × 30.5 ×10-3 = 0.009 kNm
Validation (Top Chord) 3-1-3/ 6.1.9
N Ed M y,Ed  M y,Ed M z,Ed  M z,Ed
  1
N c,Rd M cy,Rd M cz,Rd
30.5 0.0  0.0 0.09  0.009

  1
77.8 2.2 0.8
0.39 + 0.00 + 0.12 = 0.51 ≤ 1.0 OK
Combined Tension and Bending
Validation (Bottom Chord) 3-1-3/ 6.1.8
N Ed M y,Ed M z,Ed
  1
N t,Rd M cy,Rd M cz,Rd
30.5 0.0 0.02
  1
96.8 2.2 0.8
0.32 + 0.00 + 0.03 = 0.35 ≤ 1.0 OK
The buckling resistance of the top chord in compression is checked in this
example. The buckling resistance of the brace members in compression should
also be checked but are omitted from this example.
Flexural Buckling about major axis (y axis)
From Table 6.3 of BS 3-1-3/ the buckling curve for a lipped C section buckling 3-1-3/ 6.2.2
about any axis is buckling curve b.
The buckling length for buckling about the y axis is taken as equal to the truss
bay length because the top chord is restrained at the node points.
Lcr,y = 400 mm
71
Discuss me ...
E 210000 3-1-1/ 6.3.1.3

1 = = = 76.95
fy 350
Non dimensional slenderness factor:

Aeff 222.3
L A gr 400 263.0
 = cr    0.157
i 1 30.4 76.95
  
= 0.5 1     0.2   2
 = 0.51  0.340.157  0.2  0.157  = 0.505

2
1
 = ≤ 1.0
   2 2
1
 = = 1.02 but ≤ 1.0 = 1.0
2 2
0.505  0.505  0.157
Flexural buckling resistance:

 Aeff f y
Nb,Rd =
 M1
1.0  222.3  350  10 3
Nb,Rd = = 77.8 kN
1.0
Validation
NEd / Nb,Rd = 30.5 / 77.8 = 0.39 < 1.0 OK
Flexural Buckling about minor axis (z axis)
The buckling length for buckling about the z axis is taken as equal to the truss
bay length because the top chord is restrained at the node points.
Lcr,z = 400 mm
E 210000 3-1-1/ 6.3.1.3
1 = = = 76.95
fy 350
Non dimensional slenderness factor:

Aeff 222.3
Lcr A gr 400 263.0
   0.324
i 1 14.8 76.95
  
= 0.5 1     0.2   2
 = 0.51  0.340.324  0.2   0.324 

2
= 0.574
72
Discuss me ...
1
 = ≤ 1.0
   2 2
1
 = = 0.955
0.574  0.574 2  0.324 2
Flexural buckling resistance:
 Aeff f y
Nb,Rd =
 M1
0.955 222.3  350  10 3
Nb,Rd = = 74.3 kN
1.0
Validation
NEd / Nb,Rd = 30.5 / 74.3 = 0.41 < 1.0 OK
Torsional Buckling
The torsional buckling lengths are:
LT,y = 400 mm
LT,z = 400 mm
The polar radius of gyration is calculated as below:
i o2 = iy2 + iz2 + yo2 + zo2
2
io = 30.42 + 14.82 + 31.82 + 0.02 = 2154 mm2
io = 46.4 mm
The elastic critical force for torsional buckling of a simply supported member is 3-1-3/ 6.2.3(5)
given by:
1   2E Iw 

Ncr,T = 2 G It  
io  LT2 
1   2  210000  6.64  10 7 
Ncr,T =  80770  213.4   = 407.5 kN
46.4 2  400 2 

Non dimensional slenderness factor is given by: 3-1-1/ 6.3.1.4
Aeff f y
T =
N cr
222.3  350  10 3
T = = 0.437
407.5
T  
= 0.5 1     0.2   2
T = 0.51  0.340.437  0.2  0.437 2

= 0.636
73
Discuss me ...
1
T = ≤ 1.0
   2  2
1
T = = 0.911
0.636  0.636 2  0.437 2
Torsional buckling resistance is given by:
 T Aeff f y
Nb,Rd =
 M1
0.911 222.3  350  10 3
Nb,Rd = = 70.9 kN
1.0
Validation
NEd / Nb,Rd = 30.5 / 70.9 = 0.43 < 1.0 OK
Torsional-Flexural Buckling
 2E Iy 3-1-3/ 6.2.3(7)
Ncr,y =
L2cr,y
 2  210000  242500  10 3
Ncr,y = = 3142 kN
400 2
2 2
y   31.8 
 = 1   o  = 1    = 0.530
 io   46.4 
The elastic critical force for torsional-flexural buckling is given by: 3-1-3/ 6.2.3(7)
 2 2 
N cr,y  N cr,T  N cr,T   yo  N cr,T 
Ncr,TF = 1  1   4 
2  N cr,y  N cr,y 
    io  N cr,y 

111.5  407.5 
2 2
 407.5   31.8  407.5 
Ncr,TF = 1  1    4  = 382.5 kN
2  0.530  3142  3142   46.4  3142 
 
Non dimensional slenderness factor is given by: 3-1-1/ 6.3.1.4
Aeff f y
TF =
N cr
222.3  350  10 3
TF = = 0.451
382.5
TF  
= 0.5 1     0.2   2
TF = 0.51  0.340.451  0.2  0.451  2
= 0.644
1
TF = ≤ 1.0
   2 2
1
TF = = 0.905
0.644  0.644 2  0.4512
74
Discuss me ...
Torsional-flexural buckling resistance is given by:

 TF Aeff f y
Nb,Rd =
 M1
0.905 222.3  350  10 3
Nb,Rd = = 70.4 kN
1.0
Validation
NEd / Nb,Rd = 30.5 / 70.4 = 0.43 < 1.0 OK
Lateral torsional buckling of the top and bottom chords is not considered as
there are no major axis moments applied to these sections.
Lateral torsional buckling of the whole lattice truss is not considered as the top
chord (compression flange) is assumed to be restrained by the timber floor
system.
During construction the lattice truss will be unrestrained. In practice, the
unrestrained capacity of the lattice truss should be check for construction
loading but it is omitted from this example.
For minor axis (z axis): 3-1-3/ 7.1(3)
Ific
 1 

= I gr    I gr  I eff 
 1.5 
 1 
Ific = 57230   57230  51530 = 53430 mm4
 1.5 
The corner factor δ is given by: 3-1-3/ 5.1(4)

n
 rj
j 1
 = 0.43 m
= 0.02
 bp,i
i 1
Taking account of rounded corners, the second moment of area for

serviceability is given by:
ISLS = I (1 – 2δ) = 53430 (1 – 0.04) = 51290 mm4
The top chord is subject to axial compression therefore ISLS is used which is
based on effective properties and includes allowance for rounded corners.
The bottom chord is subject to axial tension therefore Igr,round is used which is
based on gross properties and includes allowance for rounded corners.
Igr,round = I (1 – 2δ) = 57230 (1 – 0.04) = 54940 mm4.
75
Discuss me ...
For this example the effective neutral axis of the truss subject to bending is
taken at mid-height of the truss.
The moment of inertia for the lattice truss is given by:
Ieff,truss = (ISLS + Aeff.y2) + (Igr,round + Agr.y2)
Ieff,truss = (51290 + 222.3 ×(100 – 13.2)2) + (54940 + 263.0 ×(100 – 12.9)2)
Ieff,truss = 3776000 mm4 :
Serviceability Criteria
For light weight steel floors there are four serviceability criteria that should be
checked to ensure acceptable performance of the floor in service. The criteria
are detailed in SCI publication P301 and in Chapter 6.10 of the NHBC
Standards.
Criterion 1
Dead load plus imposed load deflection less than span/350 or 15 mm whichever
is smaller.
Total load:
W = L s (gk + qk) = 4800 × 600 (0.6 + 1.5) × 10-6 = 6.05 kN:
5 WL3 5 6.05  4800 3  103
 =  = 11.0 mm
384 EI 384 210000  3776000
Deflection limit:
δlimit = 4800 / 360 = 13.3 mm > 11.0 mm OK
Criterion 2
Imposed load deflection less than span/450.
Total load:
W = L s qk = 4800 × 600 × 1.5 × 10-6 = 4.32 kN
Deflection due to load is given by;
5 WL3 5 4.32  4800 3  10 3
 =  = 7.9 mm
384 EI 384 210000  3776000
Deflection limit:
δlimit = 4800 / 450 = 10.7 mm > 7.9 mm OK
Criterion 3
Natural frequency of the floor not less than 8 Hz.
W = L s (gk + 0.2 qk) = 4800 × 600 (0.6 + 0.2 × 1.5) × 10-6 = 2.59 kN
5 WL3 5 2.59  4800 3  10 3
 =  = 4.7 mm
384 EI 384 210000  3776000
Deflection limit for 8 Hz
δlimit = 5.0 mm > 4.7 mm OK
76
Discuss me ...
Criterion 4
Deflection of floor system less than critical value subject to 1 kN point load.
W = 1.0 kN
The number of effective joists depends on the floor construction. For this Table 6.1 of SCI-
example the floor construction is taken as chipboard on the floor joists spaced P301
at 600 mm.
Neff = 2.35
1 WL3 1 1.0  4800 3  10 3
 =  = 1.24 mm
48 E I N eff 48 210000  3776000  2.35
The deflection limit for this criterion is dependant on the span of the joist. Table 6.2 of
δlimit = 1.37 mm > 1.24 mm OK SCI-P301
77
Discuss me ...
9 REFERENCES
1. BS EN 10346: 2009 Continuously hot-dip coated steel flat products. Technical

delivery conditions
BSI 2009
2. BS EN 1990: 2002 Eurocode: Basis of structural design.
BSI, 2002
3. BS EN 1993-1-1: 2005 Eurocode 3: Design of steel structures. General rules and
rules for buildings (incl. Corrigendum, February 2010)
BSI, 2010
4. BS EN 1993-1-3: 2005 Eurocode 3: Design of steel structures. Supplementary
rules for cold-formed members and sheeting (incl. Corrigendum, April 2010)
BSI, 2010
5. BS EN 1993-1-5: 2005 Eurocode 3: Design of steel structures. Plated structural
elements (incl. Corrigendum, February 2010)
BSI, 2010
6. BS EN 1993-1-8: 2005 Eurocode 3: Design of steel structures. Design of joints
(incl. Corrigendum, August 2010)
BSI, 2010
7. BRETTLE, M. E. and BROWN, D. G.
Steel Building Design: Concise Eurocodes (P362)

SCI, 2009
8. BRETTLE, M. E. and BROWN, D. G.
Steel Building Design: Worked Examples for Students (P387)
SCI, 2009
9. BS EN 1991-1-7:2006 Eurocode 1: Actions on structures. General actions.
Accidental actions (incl. Corrigendum, April 2010)
BSI, 2010
10. AD 358: Design thickness of cold formed members and sheeting
New Steel Construction, 2011 (available on www.steelbiz.org)
11. SMITH, A.L,HICKS, S.J. and DEVINE, P.J. Design of Floors for Vibration: A New
Approach (P354)
SCI, 2009
12. GRUBB, P.J, GORGOLEWSKI, M.T. and LAWSON, R.M. Building Design using
Cold Formed Steel Sections: Light Steel Framing in Residential Construction
(P301)
SCI, 2001
13 SN003, NCCI: Elastic critical moment for lateral torsional buckling
www.steel-ncci.co.uk
14 SN002 NCCI: Determination of non-dimensional slenderness of I and H sections
www.steel-ncci.co.uk
78

ED005 Design of Light Steel Sections To Eurocode 3

Uploaded by

Copyright:

Available Formats

ED005 Design of Light Steel Sections To Eurocode 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ED005 Design of Light Steel Sections To Eurocode 3

Uploaded by

Copyright:

Available Formats

ED005 Technical Report: Design of Light Steel Sections to Eurocode 3

Design of Light Steel Sections

A G J Way MEng, CEng, MICE

Issued (in electronic format only) by:

Berkshire, SL5 7QN

SCI (The Steel Construction Institute) is the leading, independent provider

Our service spans the following areas:

SCI Technical Reports

documents for further discussion.

 2012 The Steel Construction Institute

SCI Document Number: SCI ED 005

6 DESIGN OF MEMBERS IN BENDING 20

1.1 Light steel sections

Figure 1.1 Load bearing light steel frame in a volumetric module

There are ten separate Eurocodes:

Each Eurocode is comprised of a number of Parts, which are published as separate

BS EN 1990[2] can be considered as the ‘core’ document of the structural Eurocode

Eurocode 1 is used to determine the actions (loads) that structures must be

Eurocode 3 comprises twelve parts (BS EN 1993-1-1 to BS EN 1993-1-12). When

BS EN 1993-1-1[3] General rules and rules for buildings

BS EN 1993-1-8[6] Design of joints

1.3 Design to Eurocode 3

BS EN 1993-1-3 permits two alternative design routes:

Design by testing overcomes the limitations of the calculation approach by

The testing route can be sub-divided into two options:

specimens and the results are analysed statistically to obtain a characteristic

In the second option, testing is only conducted on a limited number of specimens

Figure 1.2 Physical testing of purlin specimen

Before a light steel member’s resistance to bending, compression or other loading

It is important to distinguish between the following types of section property:

2.1 Core steel thickness

considered when determining the thickness to use in design calculations in

2.2 Mid-line theory

Figure 2.1 Mid-line dimensions for a lipped C section

2.3 Corner radii

Figure 2.2 Rounded corner of a light steel section

Consider the following example:

The flange width measured between mid-lines = 65 – 1.5 = 63.5 mm.

(It is assumed that the nominal width of 65 mm includes the galvanising, so it is

(which corresponds to a reduction in area of about 3.4%)

Checking the corner radii:

5t = 7.3 mm, r = 3.0 mm, therefore r  5t

0.1bp = 6.13 mm, r = 3.0 mm, therefore r  0.1bp

For area, Ag  Ag,sh 1    (Eq.2)

For second moment of area, I g  I g,sh 1  2  (Eq.3)

For warping, I w  I w,sh 1  4  (Eq.4)

rj is the internal radius of curved element j

Designers of hot-rolled structural steelwork will be familiar with the concept of

process focuses on the calculation of effective section properties, following the

3.1 Effective width concept

Figure 3.1 The effective width concept applied to a plate

The method adopted by BS EN 1993-1-3 takes the effective width concept

3.2 Eurocode calculation procedure for unstiffened

For each element, the effective width is given by:

b is the width of the element

ρ is the reduction factor to allow for local buckling.

For an internal element, ρ is given by:

b is the appropriate width of the compression element

Outstand compression elements: