TRM 257

TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 1 of 8
STADIUM DESIGN – STANDARDS AND LOADING

CONTENTS

Introduction
Standards and Codes of Practice
Loading
Movement criteria
Dynamic performance and testing
Disproportionate collapse
References
Keywords

INTRODUCTION

Sports grounds and stadiums are governed by the Safety at Sports Grounds Act 1975 and the Fire
Safety and Safety of Places of Sport Act 1987. The Safety at Sports Grounds Act requires all
stadiums with a capacity of over 10,000 to have a safety certificate. This results in constraints on
the structural design of grandstands and stadium structures, which go beyond the requirements of
the Building Regulations and other legislation applicable to buildings in general. These requirements
are generally applied to all stadium structures irrespective of the capacity of the ground.

The stadium working group within WSP Buildings has prepared this TRM is to provide guidance on
aspects of structural design that are specific to stadium structures. Where definitive guidance has
been published, it is summarised and the source document is referenced. Where no such definitive
guidance is available, the common design approaches followed within WSP are outlined and, where
possible and appropriate, best practice is recommended.

STANDARDS AND CODES OF PRACTICE

The main standards applicable to stadium design which supplement the Building Regulations, British
Standards and Codes of Practice are:

• Guide to Safety at Sports Grounds – Fourth Edition – The Stationery Office, 1997
Generally referred to as ‘The Green Guide’ this is the main point of reference for management,
design, scrutiny and certification of sports grounds. The broad scope covers stadium capacity
calculations, layout, required facilities, access and egress, calculation of viewing standards,
terrace dimensions and barrier loading. Guidance in some area has been superseded by
subsequent legislation and guidance, notably with respect to stadium dynamics and provision for
disabled supporters.

• Dynamic performance requirements for permanent grandstands subject to crowd action –

Interim guidance on assessment and design – IStructE, 2001.
Prepared by a joint working group of the IStructE and government departments, this document
was issued as a supplement to the Green Guide in the area of structural dynamics. The
requirements supersede and in many cases relax the provisions of section 4.5 of the Green
Guide. The finalized document should have been issued in 2007 but was still awaited at the time
of writing.

• Dynamic testing of grandstands and seating decks – IStructE, 2002.

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 2 of 8
STADIUM DESIGN – STANDARDS AND LOADING

Issued by the joint working group to be read in conjunction with the Interim Guidance, it provides
guidance on the requirement for and specification of testing and information on testing
techniques.

• Accessible Stadia – Football Licensing Authority and Football Stadia Improvement Trust,
2003
Although published as an advisory document “not intended to be prescriptive nor inhibit alternative
or creative solutions”, it has in many ways become the accepted interpretation of BS8300 and Part
M of the Building Regulations. It has a significant impact on the provision and arrangement of
elevated wheelchair spaces.

• Temporary demountable structures – Guidance on procurement, design and use – Third

edition 2007
This IStructE publication provides guidance on all aspects of temporary venues and temporary
installations including temporary stages and lighting support.

LOADING

Guidance on the loadings to be adopted in the design of sports stadium structures is covered by
three principle documents:-
• BS 6399 : Part 1 1996 – Dead and Imposed Loads
Part 2 1997 – Wind Loads
Part 3 1988 – Imposed Roof Loads (including snow)
• Guide to Safety at Sports Grounds ( “The Green Guide” )
• Dynamic Performance Requirements for Grandstands Subject to Crowd Action
Some aspects of design loading are specific to stadium structures and common principles apply to
many stadium designs.

Imposed loads – terraces

2
Seated terrace areas should be designed for a uniformly distributed load of 4 kN/m from BS6399
Part 1 Table 1 –C2 – areas with fixed seating.
2
Standing terraces should be designed for a uniformly distributed load of 5kN/m from BS6399 Part 1
Table 1 –C5 - areas susceptible to overcrowding.

These loadings include some allowance for dynamic effects providing that the structure is not
susceptible to dynamic excitation. Susceptibility to dynamic effects should be checked separately as
explained in the section on dynamic performance.

Imposed loads – concourses

Concourse areas are generally accessed directly from the terraces and are potentially subject to
2
overcrowding. They should be designed for a uniformly distributed load of 5kN/m , BS6399 Part 1
1996 Table 1 –C5. - areas susceptible to overcrowding.

Imposed loads – hospitality areas

Hospitality areas in grandstands can have a range of uses including restaurants, bars, offices,
classrooms, kitchens and plant rooms. They are not usually accessible to the crowd at large.
Appropriate design floor loading should be determined for these areas from BS6399 Part 1.

Crowd surge loads

To allow for the dynamic effects of crowd surge, the IStructE interim guidance states that, in addition
to wind loads, grandstands should be capable of withstanding an additional lateral load as given in

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 3 of 8
STADIUM DESIGN – STANDARDS AND LOADING

Table 1 of the document. For new construction the higher figure of 7.5% of the vertical load is
generally taken and applied to the terrace imposed loads only.
2 2
ie 7.5% x 4.0 = 0.3 kN/m (seated areas) or 7.5% x 5.0 = 0.375 kN/m (standing terrace) should be
applied horizontally in either of two orthogonal directions

Barrier loads
Barriers within a stadium should be designed to resist minimum horizontal loadings given in Table 1,
2 or 3 of the “Green Guide”. A barrier is any element of the sports ground ‘intended to prevent
people falling, and to retain, stop and guide people’. For design purposes, the load is always
considered to act at a height of 1.1m above the floor or datum level.

Horizontal loads on concourse walls

A certain degree of judgement is needed when assessing whether the walls inside a grandstand
should be classed as barriers for design purposes. It is clear that when a concourse forms part of
the escape route between the vomitory and the exit gates, the walls will need to contain the escaping
crowd and must be designed for crowd loadings from Table 1 of the Green Guide. Where the
concourse is wide compared to the vomitory and exit gates or does not form part of an escape route
there is a good case for arguing that the walls do not fall within the definition of a barrier in the Green
Guide.

Live Loads – Roof

Roof imposed loads should be based on BS6399 Part 3 1988. Significant snow drift loads can have
a major effect on long span and cantilever roofs. Where possible the roof geometry should be
designed to avoid abrupt changes in height. Where this is not possible, particular attention should
be given to local snow drift build-up, such as smaller corner stands abutting gables of larger main
side stands. Due to the geometry of long span cantilever roofs, consideration should be given to
avoid the build up of rainwater in the event of a drainage system failure by the introduction of
overflows.

Wind Loading
Wind loading is generally the most critical loading for the roof, impacting on sway stability, bracing,
movement and member design. As a result, effort expended in accurate determination will usually
be reflected in design economies.

Most sports stadium structures are less than 100m high and are generally light structures with few
internal walls. Although they are inherently mildly dynamic, Figure 3 of BS 6399 Part 2 sanctions the
static equivalent approach to the derivation of wind loads even for a large grandstand. Stands can
generally be considered as open-sided buildings for which internal pressure coefficients are given in
Table 18 of the code. For straightforward cases, this approach has been used without problem but a
degree of engineering judgement is needed. For combinations of grandstand structures or a
complete stadium, individual assessment may not be the most appropriate approach, and a full
boundary layer wind tunnel model study may be justified.

Particular care should be taken if there are significant openings in the rear wall of a grandstand.
Recent experience suggests that this could give rise to varying flow patterns with a proportion of the
flow fluctuating between through and over the stand. This may give rise to dynamic conditions.
Consideration should be given to either changing the layout of the stand or carrying out a wind tunnel
testing.

Loading Combinations
The number of people within the stadium at any one time is controlled and effectively limited to the
capacity of the seating deck. If the seating deck is full, it follows that the concourses must be empty.
It therefore seems reasonable to consider realistic combinations of 100% of terrace load with 25% of
concourse and stair loads and vice versa. This approach is not specifically sanctioned by any

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 4 of 8
STADIUM DESIGN – STANDARDS AND LOADING

design standard and the design engineer must judge whether it is appropriate for the particular
project. The approach can be particularly beneficial when assessing the potential for modifying and
extending existing structures, however it should not be considered in combination with the live load
reductions specified in BS 6399.

Other load combinations which could be considered as unrealistic are crowd imposed with horizontal
surge loads in combination with either full wind load or full snow load. Events will not be held under
extreme weather conditions. It is equally unlikely to combine full wind (downdraft) loading with full
uniform snow as the wind would redistribute into drift formations or remove from the roof.

MOVEMENT CRITERIA

Global / Settlement
The spacing of columns and magnitude of column loads generally leads to the use of piled
foundations to transfer load down to suitable bearing sub-strata. A maximum of 12.5mm total vertical
movement under 150% proof load testing is usually an appropriate settlement for steel-framed
stadium structures.

Thermal Expansion / Contraction

Stadium structures are by nature long structures. Touchline stands are generally around 115m long
and complete bowl stadiums are typically 200m overall. From the point of view of weather tightness
and robustness it is desirable to minimise the number of movement joints in the roof. Thermal
expansion can therefore be a significant consideration in stadium design.

The steelwork for stadium roofs is often exposed to direct sunshine and extremes of temperature.
The range of UK average internal temperatures given in BS5950 should probably be increased from
-5°C to +35°C to -10°C to +40°C. For a typical stand, 115m long and free to move at both ends,
erected at an ambient temperature in the range 10°C to 20°C, the maximum anticipated end
-6
movement would be 57.5m x 12 x 10 x (40 – 10)°C = 21mm. In practice, depending on the detail
of the connections, some of this movement will be accommodated by bolt tolerances. Most
structures with metal sheet can accommodate thermal movements of around 20mm so it is not
usually necessary to introduce movement joints in the roof of a single grandstand.

For a complete bowl structure, the maximum movement would be greater and, if the roof is
supported by trusses running the length of the stand, may be of the order of 40mm. If the structure
is high enough and has flexible cladding this still may not be a problem, but this must be checked
carefully. Where practicable, it is generally best to introduce movement joints to separate the
touchline stands from the end stands. Typical movement joint arrangements are shown in figure 1.

Slotted connection movement joint Double frame movement joint

Figure 1 – Alternative movement joint arrangements

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 5 of 8
STADIUM DESIGN – STANDARDS AND LOADING

For the body of the stadium the design temperature range is lower but there is not always facility for
movement in the connections. It is often appropriate to introduce movement joint based on slotted
connections in the lower levels of the stands.

The terrace is generally constructed of precast concrete units spanning 7.0 m to 7.6 m between
frames. These are usually located with pins welded to the steel frame which are grouted into holes
in the precast unit. It is generally advisable to introduce movement joints at intervals along the lower
terrace. This should be done by replacing the pin with a bolted fixing which can accommodate the
design movement.

Wind Induced Movement

For any cantilever grandstand structure, the most perceptible movement is at the front of the roof.

The British Standards do not give specific guidance on allowable movement in this situation. The
curvature associated with a limit of + cantilever length/100 is the same as for the accepted limit of
span/200 for a simply supported beam without brittle finishes. There seems to be a general
consensus that adopting length/100 as the limit for vertical movement of the roof tip under a 50 year
return period wind, in conjunction with an effective pin connection between rafter and column, will
produce very low ephemeral additional steel stresses, insufficient to cause any long term fatigue
problems.

Element Deflections
Roof structure – Vertical deflection due to roof structure self weight can be significant on large
cantilever roofs. In order to avoid misalignment problems, it is general practice to erect the rafters
incorporating adjustable pre-setting to facilitate tip edge alignment under dead load.

Purlins and Rails - There is generally very little equipment suspended from the purlins of a
grandstand roof but excessive deflections must be avoided to ensure that the water resistant
effectiveness of joints in the cladding is not compromised.

DYNAMIC PERFORMANCE AND TESTING

Grandstand structures are subject to crowd-induced loading which can result in vibration problems.
Seating decks that cantilever more than 2.5m or have long spans (greater than 6m) are particularly
susceptible. Dynamic loading is unlikely to cause structural damage; the major safety concern is
that excessive motion could cause discomfort or panic among spectators.

The current accepted guidance for dynamic assessment is ‘Dynamic performance requirements for
permanent grandstands subject to crowd action – Interim guidance on assessment and design’
published by the IStructE in 2001. Section 4.5 of the Green Guide has effectively been superseded
by the interim guidance document.

Design issues
The requirements for structures suitable for all types of events are:
- Minimum vertical natural frequency of 6 Hz
- Horizontal load of 7.5% of design vertical live load applied with design vertical live load.

The requirements for structures suitable for use at sporting events only, where the playing of
incidental music will be controlled by the stadium management are:
- Minimum vertical natural frequency of 3.5 Hz
- Horizontal load of 5% of design vertical live load applied with design vertical live load.

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 6 of 8
STADIUM DESIGN – STANDARDS AND LOADING

The higher standard is generally recommended for new structures. If use is to be limited to viewing
sports only, it is important that the client fully understands and accepts the implications.

Currently, there is no minimum horizontal natural frequency requirement. The 3 Hz minimum in the
Green Guide has been superseded by the IStructE interim guidance and design for the specified
horizontal load is considered to give adequate resistance to horizontal movement.

For a repetitive structural terrace arrangement comprising precast terrace units spanning 7.0 –
7.6 between main frames, the natural frequency can be established reasonably accurately using two
dimensional models of the steel frame. This can be done using CSC S-Frame or a similar analysis
package. If this type of analysis is used, the natural frequency of individual precast concrete
elements should be significantly higher than the minimum natural frequency, typically 9 Hz.

In cases where the vertical natural frequency of cantilever structures is dependent on the
contribution of stair cores and diaphragm action of floors, dynamic performance cannot be
adequately modelled using a series of two dimensional frames. Under these circumstances a 3
dimensional model including finite-element representation of the diaphragm elements is required.
Problems arise because the diaphragm action achieved in practice is less than that assumed in
design and as a result the natural frequency is lower than expected.

An alternative approach to dynamic design is to limit acceleration rather than design above a
minimum frequency. This approach has been used on some major stadiums but there are currently
no agreed UK design criteria for acceleration. Until this situation is clarified it is recommended that
stadium structures are designed to meet minimum natural frequency requirements.
The recommendation of the IStructE Working Group is that design is carried out under the close
direction of a Chartered Engineer with appropriate experience. A member of the Stadia Working
Group should be consulted.

Testing
Different licensing authorities take differing views on the requirement for dynamic testing. The
IStructE Working Group recommends that all new grandstand structures covered by the Interim
Guidance are tested to confirm the dynamic design. Testing is divided into two categories; type 1
testing provides basic information on minimum natural frequency while type 2 testing can establish
mode shapes and other information relating to the modes. In practice, the results of type 1 testing
are often inconclusive and type 2 modal testing is usually carried out. The University of Sheffield is
the only organization in the UK known to have the equipment and experience to carry out a full
modal testing of stadium structures. The cost of testing can vary, but for a 15,000 capacity stadium
may be around £20k.

Test requirements must be carefully specified. The testing programme should be designed to suit
the particular structure, on the basis of dynamic analysis results. The testing programme should be
designed in consultation with the specialist testing house.

Results of testing are not always clear cut. If a cantilever roof relies on the terrace structure for
support, as is often the case, global modes of vibration involving the roof may be picked up by the
testing. Often these cannot be excited significantly by movement of the crowd but this assessment
must be made by a Chartered Engineer with appropriate experience.

DISPROPORTIONATE COLLAPSE

Under the current Building Regulations grandstands that will accommodate more than 5,000
spectators are classed as Class 3 structures. For a stadium with a capacity of 15,000 or less
accommodated in four separate stands the capacity of each is likely to be less than 5,000. However,
if the corners are added at a later date it raises a question as to whether the whole stadium becomes

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 7 of 8
STADIUM DESIGN – STANDARDS AND LOADING

one stand. Under the Eurocodes, the anomaly will be removed as the requirement will apply to a
stadium rather than a grandstand. On balance, it is probably best to design all stadiums as though
they were class 3 structures.

The Building Regulations requirement for a Class 3 building is to base design on a systematic risk
assessment taking into account all normal hazards that may be reasonably be foreseen together
with any abnormal hazards. The problems with this approach are addressed in TRM 130. WSP
design policy on Class 3 structures, which is outlined in TRM 130, is based on a strategy of providing
a reasonable degree of robustness. This is the strategy that should be adopted for stadium
structures.

BS 5950 states that “the structure should behave as one three-dimensional entity” and that “the
layout of its constituent parts, foundations, steelwork, connections and other components should
constitute a robust and stable structure under normal loading to ensure that in the event of misuse or
accident, damage will not be disproportionate to the cause.”

In general, a redundancy and alternative load path approach should be adopted. Where
discontinuity of ties at movement joints cannot be avoided or where particular architectural
requirements dictate, a key element approach may be unavoidable.

Localization of Damage

Concourse, hospitality and terrace levels

The lower concourse and hospitality areas and terrace areas of the individual grandstand structures
are generally supported on a regular column arrangement and as such can be addressed by the
conventional approach common to all multi-storey buildings.

The precast concrete terrace units present a situation which is specific to stadiums and common to
most stadium structures. The units are usually located with two dowel bars or shear studs, one at
each end. In order to meet the requirement for a Class 2B building, these pins should be designed
to each resist a tying force equal to the weight of the terrace unit. This will provide good resistance
against them falling off their supports in the case of excessive movement of the structure and should
also hold the supporting beam in place in the event of loss of a supporting column.

Cantilever Roof
The lightweight cantilever form of roof construction that is generally used to ensure unrestricted
viewing for all spectators determines that a particular approach is adopted to satisfy robustness
requirements.

An iterative exercise of systematically introducing a pin to the cantilever rafter / column connection
(to simulate rafter failure) within the analysis model is commonly used. This should demonstrate that
the longitudinal ties would transfer loading to adjacent frames by catenary action. Because the roof
construction is lightweight the structure is usually sufficiently strong to prevent the “domino-effect” of
a progressive collapse, however it has to be accepted, that any such incidence would cause
excessive deflection beyond the 70m² limitation of the Building Regulations.

When it is necessary to introduce movement joints in the roof structure it is usually possible to do so
by means of slotted connections. In this case a tying capability can be provided across the
movement joint.

© WSP Group
 TRM 257
TECHNICAL REFERENCE Rev 1

MANUAL Date 07/08

Civil & Structural
Page 8 of 8
STADIUM DESIGN – STANDARDS AND LOADING

REFERENCES

Guide to Safety at Sports Grounds – Fourth Edition, Stationary Office, 1997

Dynamic performance requirements for permanent grandstands subject to crowd action – Interim
guidance on assessment and design, IStructE, 2001.

Advisory Note – Dynamic testing of grandstands and seating decks, IStructE, 2002.

Accessible Stadia – Football Licensing Authority and Football Stadia Improvement Trust, 2003

Temporary demountable structures – Guidance on procurement, design and use – Third edition
2007

BS 5950-1:2000 Structural use of steelwork in building – Part 1 Code of practice for design – Rolled
and welded sections

BS6399-1:1996 – Loading for buildings – Part 1: Code of practice for dead and imposed loads

BS6399-2:1997 – Loading for buildings – Part 2: Code of practice for wind loads

BS6399-3:1988 – Loading for buildings – Part 3: Code of practice for imposed roof loads

KEYWORDS

Stadium; design standards; loading; movement; dynamic performance; dynamic testing; robustness;
disproportionate collapse.

Authors: Mark Barwood (Buildings, Leeds)

Alan Liddell (Buildings, Newcastle)
Jeremy Wells (GTC)
Sponsor: Group Technical Centre
Revision record:
xx/08 First issue (rev 1)

© WSP Group