PD6688 2

PD 6688-2:2011
BSI Standards Publication
PUBLISHED DOCUMENT
Background to the National

Annex to BS EN 1991-2
Licensed copy:Arup, 02/03/2020, Uncontrolled Copy, © BSI
Traffic loads on bridges
This publication is not to be regarded as a British Standard.
NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAW
raising standards worldwide™

PD 6688-2:2011 PUBLISHED DOCUMENT
Publishing and copyright information

The BSI copyright notice displayed in this document indicates when the
document was last issued.
© BSI 2011
ISBN 978 0 580 58680 4
ICS 91.010.30; 93.040
The following BSI reference relates to the work on this standard:

Committee reference B/525/10
Publication history
First published March 2011
Amendments issued since publication
Date Text affected

PUBLISHED DOCUMENT PD 6688-2:2011
Contents
Foreword ii
1 Scope 1
2 Normative references 1
3 BS EN 1991-2:2003, Eurocode 1: Actions on structures – Part 2:
Traffic loads on bridges 1
Annexes
Annex A (informative) Vibration serviceability recommendations for
foot and cycle track bridges 11
Bibliography 19
List of figures
Figure 1 – Comparison of pedestrian load intensity (excluding crowd
loading) variation with loaded length for road (and rail) bridges
supporting footways or cycle tracks and for footbridges 8
Figure 2 – Reduction factor, λ, to allow for the unsynchronized
combination of pedestrian actions within groups and crowds 10
Figure A.1 – Synthetic time history for a random bridge/pedestrian
model 13
Figure A.2 – Reduction of pedestrian speed and flow rates with
density 14
Figure A.3 – Configuration factor K 15
Figure A.4 – Dynamic response factor Ψ 16
List of tables
Table 1 – Loading classes 6
Table A.1 – Typical values of logarithmic decrement of decay of
vibration (δ ) between successive peaks 11
Summary of pages
This document comprises a front cover, an inside front cover,
pages i to ii, pages 1 to 20, an inside back cover and a back cover.
© BSI 2011 • i
Foreword
Publishing information
This Published Document is published by BSI and came into effect on
31 March 2011. It was prepared by Subcommittee B/525/10, Bridges,
under the authority of Technical Committee B/525, Building and civil
engineering. A list of organizations represented on this committee
can be obtained on request to its secretary.
Relationship with other publications

This Published Document is a background paper that gives
non-contradictory complimentary information for use in the UK
with the Eurocode for actions on structures, BS EN 1991-2 and its
UK National Annex.
BS 5400-2 has been withdrawn and superseded by BS EN 1990,
BS EN 1991-2 and BS EN 1991-1-7. References to BS 5400-2 in this
Published Document are informative only and are included to provide
readers with a comparison between the withdrawn British Standard
and the current Eurocodes.
The following standards BS EN 1991-1-3, BS EN 1991-1-4, BS EN 1991-1-5
and BS EN 1991-1-6 should be consulted when designing bridges.
Presentational conventions
The word “should” is used to express recommendations of this
Published Document. The word “may” is used in the text to express
permissibility, e.g. as an alternative to the primary recommendation
of the clause. The word “can” is used to express possibility, e.g. a

consequence of an action or an event.
Notes and commentaries are provided throughout the text of this
standard. Notes give references and additional information that are
important but do not form part of the recommendations. Commentaries
give background information.
Contractual and legal considerations

This publication does not purport to include all the necessary provisions
of a contract. Users are responsible for its correct application.
Compliance with a Published Document cannot confer immunity from
legal obligations.
ii • © BSI 2011
1 Scope
This Published Document gives non-contradictory, complementary
information on traffic loads on bridges for use in the UK in conjunction
with BS EN 1991-2 and its UK National Annex.
2 Normative references
The following referenced documents are indispensable for the
application of this document. For dated references, only the edition
cited applies. For undated references, the latest edition of the
referenced document (including any amendments) applies.
BS EN 1990:2002, Eurocode – Basis of structural design
BS EN 1991-2:2003, Eurocode 1: Actions on structures – Part 2: Traffic
loads on bridges
NA to BS EN 1990:2002+A1:2005, UK National Annex for Eurocode –
Basis of structural design
NA to BS EN 1991-2:2003, UK National Annex to Eurocode 1: Actions
on structures – Part 2: Traffic loads on bridges
3 BS EN 1991-2:2003, Eurocode 1: Actions on

structures – Part 2: Traffic loads on bridges
3.1 Models for loaded lengths greater than 200 m

[BS EN 1991-2:2003, 4.1 (1), Note 2]

A study has been carried out to compare type HA loading that
was given in BS 5400-2 (BD 37/01 1) [1]) and Load Model 1 (LM1) in
BS EN 1991-2 with the parameters given in the UK National Annex.
The LM1 effects are within the acceptable range for spans up to
1 500 m. It is concluded that, assuming type HA loading in BS 5400-2
was suitable for bridges with a span up to 1 500 m, the application of
LM1 is appropriate for bridges up to 1 500 m loaded length.
3.2 Adjustment factors α for LM1

[BS EN 1991-2:2003, 4.3.2 (3), Notes 1 and 2]
The values given in the NA to BS EN 1991-2:2003, Table NA.1 have
been derived based on calibration against HA serviceability limit state
(SLS) loading from BS 5400-2 and verified using reliability analysis
(see Atkins’ Background to the UK National Annexes to BS EN 1990
and BS EN 1991-2 [2]). This loading should be used together with the
partial factors given in the NA to BS EN 1990:2002+A1.
3.3 Adjustment factor β for LM2

[BS EN 1991-2:2003, 4.3.3 (2)]
The NA to BS EN 1991-2, NA.2.14 provides for a wheel load of 200 kN
compared with the 100 kN load that was given in BS 5400-2:2006,
6.2.5. However, the wheel contact area has been adjusted as in the
1)
The technical content of BS 5400-2:2006 was extracted from BD 37/01.
© BSI 2011 • 1
NA to BS EN 1991-2, NA.2.15, which results in similar contact pressures

to those previously given in BS 5400-2.
3.4 Wheel contact surface for LM2

[BS EN 1991-2:2003, 4.3.3 (4)]
The wheel contact area given in the NA to BS EN 1991-2:2003, NA.2.15
results in a characteristic pressure of 1,25 N/mm2 compared with
1,1 N/mm2 that was given in BS 5400-2:2006. The corresponding
design pressures are 1,69 N/mm2 and 1,82 N/mm2. This also makes the
wheel contact areas the same for LM1 and LM2.
3.5 Upper limit of the braking force on road bridges

[BS EN 1991-2:2003, 4.4.1 (2)]
This value of 900 kN has been retained for harmonization although
the value was 750 kN in BS 5400-2, 6.10.1.
NOTE The value of α q1, used to calculate the braking force is given as 1,0
in the note to the NA to BS EN 1991-2:2003, Table NA.1, which differs from
the value used for vertical loading. This is because the α q1 values needed
to calibrate the braking force and vertical loads are different.
3.6 Horizontal forces associated with LM3

[BS EN 1991-2:2003, 4.4.1 (3)]
3.6.1 Longitudinal braking force

[NA to BS EN 1991-2:2003, 2.18.1]

Where the loaded length is less than the length of the SV or SOV
vehicle (see Figures NA.1 to NA.3), only the weight of the axles
likely to occupy the loaded length when the vehicle is in its most
unfavourable position should be considered in calculating ω . Where a
structure is designed for SV196, the maximum braking force resulting
from either the SV100 or SV196 vehicle should be considered. SV80
and SV100 vehicles produce approximately the same magnitude
of braking force because the increase in the basic axle load ω is
counteracted by the decrease in the deceleration factor δ .
3.6.2 Centrifugal force

[NA to BS EN 1991-2:2003, 2.18.2]
Where the loaded length is less than the length of the SV or SOV
vehicle, only the weight of the axles likely to occupy the loaded
length when the vehicle is in its most unfavourable position should be
considered in calculating W. All three SV vehicle models produce the
same level of centrifugal force for loaded lengths over 14 m, and it is
therefore sufficient to consider only one vehicle in this case. However,
for loaded lengths under 14 m, differences might arise due to the
number of axles present and the axle weights. The centrifugal force
should therefore be checked for different vehicles.
The horizontal forces should be combined with the vertical loads for
SV or SOV vehicles as given in Table NA.3 for Group 6.
NOTE The basis for this can be found in the Background to the UK
National Annexes to BS EN 1990 and BS EN 1991-2 [2].
2 • © BSI 2011
3.7 Horizontal force transmitted by expansion joints or

applied to structural members
[BS EN 1991-2:2003, 4.4.1 (6)]
The horizontal force transmitted by expansion joints given in
BS EN 1991-2:2003 of 180 kN/axle should be used as there was no
equivalent value given in BS 5400-2. It is not dissimilar to the value
of 180 kN/m run of the joint given in BD 33/94 [3] given that the UDL
from each horizontal wheel force of 90 kN applied over a 400 mm
width is 225 kN/m.
3.8 Lateral forces on road bridge decks

[BS EN 1991-2:2003, 4.4.2 (4)]
The transverse force due to skew braking or skidding, Qtrk, should be
considered to act simultaneously with Qlk.
BS EN 1991-2:2003 recommends a value of 25% of Qlk which is
approximately 90 kN while BS 5400-2:2006, 6.11.1 gave a constant
force for skidding of 300 kN. The latter is considered to be excessive.
3.9 Fatigue load models

[BS EN 1991-2:2003, 4.6]
Background to the fatigue load models can be found in the Derivation
of the UK National Annex to Clause 4.6: Fatigue Load Models [4].
Dynamic additional amplification factor due to

3.10
expansion joints
[BS EN 1991-2:2003, 4.6.1 (6)]
No modification is proposed as the recommended values are reasonably
close to those that were given in BS 5400-10 2).
3.11 Fatigue LM4

[BS EN 1991-2:2003, 4.6.1 (2) Note 2(e) and 4.6.5 (1)
Note 2]
Fatigue LM4 consists of sets of standard lorries that together produce
effects equivalent to those of real traffic. The NA to BS EN 1991-2:2003,
Table NA.5 defines the properties of standard lorries for use in fatigue
design on all routes.
The characterization of road traffic for fatigue design purposes
depends on the traffic lane configuration, traffic flow rates,
proportions of heavy goods vehicles, types of heavy goods vehicles
and vehicle usage. Traffic flow rates are route-dependent and the
NA to BS EN 1991-2:2003, Table NA.4 should be used to determine
the traffic flows. However, road provision is made approximately
in relation to demand, and it is recommended that the highway
traffic flow rates that are presented in the NA to BS EN 1991-2:2003,
Table NA.4 be adopted for design as stated in the National Annex.
2)
Now withdrawn and superseded by BS EN 1993-1-3.
© BSI 2011 • 3
3.12 Fatigue LM5 (based on recorded traffic data)

[BS EN 1991-2:2003, 4.6.6 (1)]
Fatigue LM5 consists of the direct application of recorded traffic
flows at specific sites. The traffic flow data for the development
of such a model can be supplemented, if appropriate, by statistical
extrapolation.
The following guidelines should be used in the development of a
site-specific fatigue load model.
• Limited data indicates that the nature of use and, therefore, the
weight distributions of each type of heavy goods vehicle appear to
depend more on the country (and thus the legislative and vehicle
control frameworks) than to sites within individual countries.
• Certain sites are particularly subject to asymmetric flows and
vehicle populations might contain unusually large fractions of
either fully laden or empty vehicles. Such sites might include
access routes to mines or quarries, to large industrial locations
and to port facilities.
• It is most important to accurately model the numbers of the
heavier goods vehicle types. Estimates of the “total goods vehicle
flow” can therefore be very misleading, owing to the variable
means of monitoring and classifying twin axle lorries. Care
should be taken to accurately estimate the number of lorries with
three or more axles, and to correctly match the numbers of such
vehicles to those in general traffic models where these are used.
3.13 Effects of collision forces on vehicle restraint systems

[BS EN 1991-2:2003, 4.7.3.3]
3.13.1 Global effects, 4.7.3.3 (1)

The magnitude of forces transferred to a structure during vehicle
collision against a vehicle restraint system depends on a complex
interaction between the speed of impact, mass and stiffness of the
vehicle, the strength and stiffness of the restraint system, the stiffness
of the connection between the vehicle restraint system and the
structure and the stiffness of the structure. For design purposes simple
rules are required that take account of the key parameters.
In BS 5400-2 the forces were based on the containment level and
mass of the structure. In BS EN 1991-2 the forces are based on the
stiffness of the connection. However, it was felt that designers could
have difficulty in determining whether the stiffness of a connection is
weak or strong. The approach recommended in the National Annex
takes into account only the containment level and the stiffness of the
parapet, which can be more readily determined by the designers. The
magnitude of forces given in the NA to BS EN 1991-2:2003, Table NA.6
is broadly in line with those that were specified in BS 5400-2:2006, 6.7.
The transverse forces in the NA to BS EN 1991-2:2003, Table NA.6
correspond to the values given in BS EN 1991-2, Table 4.9 (n) while
the longitudinal and vertical forces were based on BS 5400-2:2006, 6.7
with partial factors set to unity. It is possible that BS EN 1991-2 does
not take into account the longitudinal and vertical forces as they were
assumed not to have a significant influence on the design.
4 • © BSI 2011
Note that BS 5400-2, 6.7.2 did not take into account global forces for
all parapets other than the very high containment parapets. Therefore
an amendment to the NA to BS EN 1991-2:2003, NA.2.30 has been
prepared. The proposed amendment is that the last paragraph of the
NA to BS EN 1991-2:2003, NA.2.30.1 should be amended as follows:
For classes A and B the transverse forces in Table NA.6 should be
applied 100 mm below the top of the vehicle restraint system or 1,0 m
above the level of carriageway or footway, whichever is lower, and on
a line 0,5 m long. For classes C and D, the forces in Table NA.6 should
be applied uniformly over a length of 3 m at the top of the traffic face
of the vehicle restraint system and in a position along the line of the
vehicle restraint system that produces the maximum effects on the
part of the structure under construction.
It is recommended that the appropriate class of forces that should be
used is agreed on a project specific basis.
The position of the application of the vertical force has been based on
BS 5400-2:2006, 6.7.2.1. The use of the vertical force, which models a
wheel climbing the parapet, makes sense only at the top of the parapet.
The co-existing loading due to normal traffic given in BS EN 1991-2,
4.7.3.3 (1), Note 3, (0,75α Q1Q1k) is considerably lower than the full
type HA loading + the accidental vehicle load that was specified in
BS 5400-2:2006, 6.7.2.2. The latter is obviously more conservative
for use in an accidental situation and, therefore, the requirement in
BS EN 1991-2 might appear to be less conservative. For this reason a
reduction factor of 0,75 on the lane 1 loading for LM1 (0,75α Q1Q1k),
which is the frequent value, in addition to the accidental vehicle
loading, has been adopted in the NA to BS EN 1991-2:2003, NA.2.30.2.

This is considered to give an adequate representation of the design
situation.
3.13.2 Local effects, 4.7.3.3 (2)

The approach given in the National Annex in essence follows
BS EN 1991-2, supplemented with additional guidance that was
contained in BS 5400-2. The underlying principle of the two standards
is the same.
The magnitude of the collision forces calculated using the National
Annex will not be the same as those previously found in BS 5400-2.
The National Annex uses characteristic resistance multiplied by 1,25
[BS EN 1991-2:2003, 4.7.3.3 (2)], while BS 5400-2 used design resistance
(which was lower than the characteristic) multiplied by a partial factor
of 1,4 or 1,5 (BS 5400-2:2006, 6.7.1.4).
1,25 times the characteristic resistance gives the approximate mean
strength for a concrete parapet and is slightly higher than the mean for
a metal parapet post. This is likely to be adequate since the supporting
member will be designed with a partial factor on its resistance.
No other loading is considered in BS EN 1991-2 to act simultaneously
with the horizontal collision forces for local effects, while BS 5400-2
considered the coexistence of the vertical loads attributable to the
accidental vehicle loading. The latter approach has been adopted in
the National Annex as the two sets of loads are bound to coexist during
vehicle collision with a parapet (see the NA to BS EN 1991-2, NA.2.30.2).
The accidental vehicle loading given in BS EN 1991-2 is considerably
lighter than that given in BD 21/01 [5], the assessment standard for
© BSI 2011 • 5
UK highway structures, and the former design standard BS 5400-2.

Designers should consider designing bridges for higher accidental
wheel per vehicle loads for individual projects where the likelihood of
such loading is considered significant. The following paragraph should
be considered in addition to the NA to BS EN 1991-2:2003, NA.2.30.2
where a higher accidental wheel per vehicle load is considered to be
appropriate.
Members supporting central reserves, outer verges and footways, which
are not protected from vehicular traffic by an effective barrier, should
be assessed for an enhanced accidental vehicle loading in addition
to the requirements in BS EN 1991-2. For the purposes of considering
this accidental vehicle loading, a new effective carriageway should
be defined, extending between bridge parapets or between other
effective barriers. LM1, comprising a single tandem system and 3 m
width of UDL, should be applied in whatever lateral or longitudinal
position produces the most adverse effect on the element considered.
Where the application of any wheel or wheels has a relieving effect,
it or they should be ignored. Smaller widths of UDL should also be
considered where this is more adverse than the full 3 m width. Lane
factors for Lane 1 should be applied to the loading. This additional load
case should be considered in the accidental combination of actions and
should not be considered in conjunction with other variable actions.
3.14 Actions on pedestrian parapets

[BS EN 1991-2:2003, 4.8 (1), Note 2]
The choice of the parapet loading class depends on the particular
situation of the structure and is normally project specific.

In situations where crowd loading is expected, e.g. access to a sports
stadium, it would be normal to specify a parapet with a load class of
at least Class H (2,8 kN/m) in accordance with Table 1.
Table 1 Loading classes
Class Characteristic value of horizontal uniformly

distributed load
kN/m
A 0,4
B 0,8
C 1,0
D 1,2
E 1,6
F 2,0
G 2,4
H 2,8
J 3,0
6 • © BSI 2011
3.15 Model for vertical loads on abutments and wing

walls adjacent to bridges
[BS EN 1991-2:2003, 4.9.1]
3.15.1 Loading from normal traffic

[NA to BS EN 1991-2:2003, NA.2.34.2]
The vehicle configuration given in the NA to BS EN 1991-2:2003,
Figure NA.6 corresponds to the 32 tonne rigid vehicle among the set
of vehicles given in BD 21/01, Annex D [5].
This vehicle is the most compact and has the highest load density
(weight/length is 3,8 tonnes/m) among the set of vehicles. This vehicle
produces the most severe bending and shearing effects for loaded
lengths up to 15 m.
The dynamic amplification factor and the overload factors have been
derived by calibrating the average ultimate limit state (ULS) load
effect from the model vehicle against the HA loading effect, found in
BS 5400-2, for bending and shearing in a simply supported beam for
loaded lengths of between 10 m and 15 m.
3.15.2 Loading from special vehicles

[NA to BS EN 1991-2:2003, NA.2.34.3]
The 0,75 factor on the loading from associated normal traffic
represents its frequency value. The SV-TT vehicle (military tank
transporter) used for the assessment of structures in accordance with
BD 86/07 [6], is not currently considered for design loading. It has
been assumed that the SV-TT with its 4-wheel 250 kN rear axles is
less critical as military vehicles do not overload to the same extent as
commercial abnormal vehicles and also because of their 3,5 m width,
they displace associated loading from the adjacent lane.
3.16 Load models for inspection gangways

[BS EN 1991-2:2003, 5.2.3 (2)]
The load model for inspection gangways within road bridges has
been taken as in BS EN 1991-2 since an equivalent model was not
given in BS 5400-2.
3.17 Uniformly distributed load for road and rail

bridges supporting footways or cycle tracks and
for footbridges
[NA to BS EN 1991-2:2003, NA.2.36]
The recommended values for footbridge loading (for crowd loading
and reduced pedestrian loading) are contained in BS EN 1991-2:2003,
5.3.2.1 (2). It is anticipated that, in an amendment to BS EN 1991-2, the
uniformly distributed load qfk given in the NA to BS EN 1991-2:2003,
NA.2.36 is to be adopted for BS EN 1991-2:2003, 5.3.2.1 (2) (footbridges)
as well as 5.3.2.1 (1) (footways on road bridges). The same loading
(NA.2.36) is also to be used for footways on rail bridges.
© BSI 2011 • 7
The basis for NA.2.36 can be found in the Background to the UK

National Annexes to BS EN 1990 and BS EN 1991-2 [2]. The characteristic
value given in the NA to BS EN 1991-2:2003, NA.2.36 is calibrated
against the nominal loading that was given in BS 5400-2:2006, 6.5.1.
However, this value is to be used with an SLS partial factor of 1,0 and
a ULS partial factor of 1,35 as given in the NA to BS EN 1990:2002,
NA.2.3.9 and NA.2.3.7 respectively. The pedestrian loadings specified
(but excluding crowd loading) by the different standards are compared
in Figure 1.
Figure 1 Comparison of pedestrian load intensity (excluding crowd loading) variation with loaded length
for road (and rail) bridges supporting footways or cycle tracks and for footbridges
x - Loaded length (m)

y - Load intensity (kN/m2)
Key
1 BS EN 1991-2 nominal A) 4 BS 5400-2:2006 ULS
2 BS 5400-2:2006 nominal 5 NA to BS EN 1991-2 nominal C)
B)
3 BS EN 1991-2 ULS 6 NA to BS EN 1991-2 ULS
A)
For footbridges only [see BS EN 1991-2:2003, 5.3.2.1 (2)]. For road or railway bridges supporting footways or cycle
tracks [see BS EN 1991-2:2003, 5.3.2.1 (1)], the recommended nominal pedestrian load intensity is a constant of
5 kN/m2.
B)
For footbridges only [see BS EN 1991-2:2003, 5.3.2.1 (2)]. For road or railway bridges supporting footways or cycle
tracks [see BS EN 1991-2:2003, 5.3.2.1 (1)], the recommended pedestrian load intensity at ULS is a constant of
6,75 kN/m2.
C)
For road or railway bridges supporting footways or cycle tracks (see NA to BS EN 1991-2:2003, NA.2.36).
8 • © BSI 2011
The model in the NA to BS EN 1991-2 provides a good fit to the

nominal loading that was given in BS 5400-2:2006, 6.5.1. A pedestrian
load intensity of 5,0 kN/m2 is considered to represent a very dense
crowd. Values greater than 7,0 kN/m2 would be inconceivable, except
where there is the possibility that people are stationary on the bridge,
although the dynamic increment of loading can reduce with the
extent of crowding.
3.18 Concentrated load

[BS EN 1991-2:2003, 5.3.2.2 (1)]
The BS EN 1991-2:2003 value has been adopted as there was no
equivalent load model in BS 5400-2.
3.19 Horizontal force on footbridges

[BS EN 1991-2:2003, 5.4 (2)]
The BS EN 1991-2:2003 value has been adopted since an equivalent
load model was not available in BS 5400-2.
3.20 Dynamic models for pedestrian loads on footbridges

[BS EN 1991-2:2003, 5.7 (3)]
The values of reduction factor given by the NA to BS EN 1991-2:2003,
Figure NA.9 are considered to be erroneous in some cases. A replacement
figure has been prepared to address this (see Figure 2).
It is recommended that values of the reduction factor should be agreed
on a project specific basis.

The background to the vibration serviceability requirements for
pedestrian footbridges is given in Annex A and further information
can be found in Design Methodology for Pedestrian Induced
Footbridge Vibration [7].
3.21 LM3: Models of special vehicles

[BS EN 1991-2:2003, Annex A]
The SV vehicle models have been developed based on the previous
work carried out for the development of BD 86/07 [6]. The SV196
vehicle is slightly more onerous than the assessment SV-Train vehicle
but this avoids the need for considering different vehicle types for the
design of a structure. Further information on the development of SV
models is given in the Atkins’ Background to the UK National Annexes
to BS EN 1990 and BS EN 1991-2 [2].
In common with the SV vehicle models in BD 86/07 [6], the SV196
model has the following limitations.
• The case of two or more STGO vehicles occurring simultaneously
within a lane over a bridge is not accounted for since it is not
permitted under the STGO Regulations.
• The simultaneous occurrence of two or more abnormal vehicles
in adjacent lanes over a bridge is not accounted for since it is not
permitted under the STGO Regulations.
• The dynamic amplification factors have been determined based
on work undertaken for the development of BD 86/07 [6].
© BSI 2011 • 9
Figure 2 Reduction factor, λ, to allow for the unsynchronized combination of pedestrian actions within
groups and crowds
x - Structural damping – logarithmic decrement, δ

y - Reduction factor on effective number of pedestrians, γ
Key
1 Effective span, Seff (m)
Pedestrian groups
Crowd loading
10 • © BSI 2011
Annex A (informative) Vibration serviceability recommendations

for foot and cycle track bridges
A.1 General
For superstructures, where the fundamental natural frequencies of
vibration exceed 8 Hz for the unloaded bridge in a vertical direction
and 1,5 Hz for the loaded bridge in a horizontal direction, the
necessary vibration serviceability can be deemed to be satisfied.
In general, a linear elastic model is appropriate for calculation of
the structure’s response to dynamic actions. Nevertheless, attention
is drawn to the fact that the constructed footbridge might have
different natural frequencies from those calculated (e.g. due to the
interaction between structural and non-structural parts) and therefore
might respond differently from predictions.
Typical values for the logarithmic decrement of decay of vibration (δ )
(structural damping) to be used in design are provided in Table A.1.
Structural damping is extremely difficult to predict and bridge fittings,
such as parapets, can significantly alter these values. Therefore, it is
suggested that the intrinsic damping be measured before providing
additional damping to a constructed bridge. The value of structural
damping might also depend on the mode and load case under
consideration and this needs to be taken into account during design.
Table A.1 Typical values of logarithmic decrement of decay of vibration (δ )

between successive peaks
Material of construction Logarithmic decrement of decay

of vibration (δ )
Steel 0,03
Steel and concrete composite 0,04
Concrete 0,05
Timber 0,06 to 0,12
Aluminium alloy 0,02
Glass or fibre reinforced plastic 0,04 to 0,08
At the time of publication, research suggests that in crowded conditions

the mass of pedestrians might also act to add vertical damping to the
bridge and so further reduce the response to dynamic actions from
pedestrians. If this effect is proven to be a reliable source of energy
dissipation, the extra damping provided may be added to the estimate
of structural damping and, in all other respects, the response to dynamic
actions from pedestrians calculation will be unaltered. However, under
no circumstances should such additional damping be used in the
avoidance of unstable lateral responses.
The pedestrian actions described in the NA to BS EN 1991-2, NA 2.44
correspond to an average pedestrian weight of 700 N. In general,
the dynamic action imposed by the excitation of a bridge structure as
pedestrians move over it, is proportional to the average weight. For
bridges that are to be designed for locations where the national average
weight is known to be different to that of the UK, consideration might
be given to adjusting the applied actions accordingly.
© BSI 2011 • 11
A.2 Background to “Dynamic actions representing the

passage of single pedestrians and pedestrian groups”
[NA to BS EN 1991-2:2003, NA.2.44.4]
The dynamic actions represent a group of N pedestrians where the
following applies.
• Pedestrians in a group make a single crossing of the bridge
together.
• One pedestrian in the group is assumed to walk with a pace
frequency that is exactly matched to the frequency of the mode
being investigated (the reduced likelihood of this occurring in
practice is dealt with subsequently by the factor k(fv) obtained
from the NA to BS EN 1991-2:2003, Figure NA.8).
• All other pedestrians in the group (N-1) are assumed to walk
with phase and pace rates that are randomly chosen from the
pedestrian population model.
• In order to allow for the increased likelihood that pedestrians in
small groups might walk with frequencies that are similar to each
other, the standard deviation of the distribution of frequencies
used for the group pedestrian model has been chosen to be 0,1
[as opposed to the value of 0,3 that was used in the development
of the crowd model and the factor k(fv)].
In this way the group model aims to provide an assessment of the
effect of groups of pedestrians who are walking more purposefully
than those represented by the crowd model.
A.3 Background to “Steady state modelling of pedestrians

in crowded conditions”
[NA to BS EN 1991-2:2003, NA.2.44.5]
In order to describe the range of possible bridge responses to dynamic
actions from pedestrians, which might be caused by crowd loading, it
is normal to describe the responses in a probabilistic manner. A useful
measure of the amplitude of the response is the root mean square (RMS)
of the population of possible responses. As this amplitude is frequently
exceeded by a large amount, the RMS is not suitable for comparison
with comfort criteria.
NOTE In the extreme there is a real, but very small risk that everyone
within a crowd might be fully synchronized for the duration of the loading.
Mathematical techniques are readily available to obtain maximum
design values for given return periods. For a bridge response frequency
of approximately 2 Hz, it can be shown that 4,0 times the standard
deviation (σ) needs to be used to provide a maximum design value with
an average return period of 15 minutes. At face value this appears to be
a suitable maximum design value. However, if real response signals are
examined, it is evident that the use of 4,0 times the standard deviation
would be unnecessarily severe.
Figure A.1 illustrates a synthetic time history produced from a random
bridge response population model centred on a bridge response
frequency of 2 Hz. Horizontal lines mark the boundaries of ±4,0, ±2,5 and
±1,0 times standard deviations for this typical 7½ minute time history.
12 • © BSI 2011
Figure A.1 Synthetic time history for a random bridge/pedestrian model
x - Time (seconds)
y - Bridge response frequency (Hz)
Key
σ Standard deviation
It has been decided that 2,5 times standard deviation is an appropriate

maximum design value for use for bridge responses to crowd loading,
and the loading provided by the NA to BS EN 1991-2, NA.2.44.4
incorporates this assumption. This permits short durations of response
that are above the target comfort criteria but the average level of
comfort is significantly better than the RMS value.
Other design guides (such as the Sétra Technical Guide for the
Assessment of vibrational behaviour of footbridges under pedestrian
loading [8]) use maximum design accelerations that are equivalent
to the 4,0 standard deviations illustrated in Figure A.1. In certain
circumstances, subject to agreement on a project specific basis, a
more stringent limit to the maximum bridge response may be applied
by increasing the design maximum loading from 2,5 to 4,0 times
standard deviations.
In very crowded conditions, the mean velocity of pedestrians tends
to reduce as the density increases (see Figure A.2) and the population
statistics change. In the calculation of vertical bridge responses to
the NA to BS EN 1991-2, NA.2.44.4, the maximum crowd density of
1 persons/m2 is used to take account of this effect.
There is some evidence to suggest that additional damping might
be available in very crowded conditions. In addition, there is rather
more evidence to suggest that simple probabilistic models of crowd
behaviour, such as that used in the NA to BS EN 1991-2, NA.2.44.4,
tend to produce calculated bridge responses that are somewhat larger
than field measurements of comparable cases. However, at the time of
publication, such measured responses cannot be relied upon for design.
© BSI 2011 • 13
Figure A.2 Reduction of pedestrian speed and flow rates with density
a)
b)
a) x - Density (people/m2)
y - Speed
b) x - Flow rate
y - Density (people/m2)
A.4 A simplified method for deriving the maximum

vertical acceleration due to single pedestrians and
pedestrian groups
This method is valid only for a single span or two or three-span
continuous, symmetric superstructures of constant cross-section and
supported on bearings that can be idealized as simple supports.
The maximum vertical acceleration α (m/s2) can be taken as:
 F′ 2 
α = γ imax  KΨ
 Mi 
where:
F’ is the amplitude of the moving dynamic load (N) given in
the NA to BS EN 1991-2, NA.2.44.4 (1). That is:
( )
F ′ = F0 k fv ( )
1+ γ N − 1 ;
Mi is the generalized mass of the mode of interest i (kg);
γ imax is the maximum vertical component of mode shape I;
K is the configuration factor (see Figure A.3);
Ψ is the dynamic response factor (see Figure A.4).
14 • © BSI 2011
Figure A.3 Configuration factor K
Key
1 Bridge configuration
l length of main span (m)
l1 length of side spans (m)
NOTE For three-span continuous bridges, intermediate values of K may be obtained by linear interpolation.
In all other situations, the maximum vertical acceleration for single

pedestrians and pedestrian groups is to be determined from a time
history calculation of the bridge response, which explicitly models
fluctuating loads passing across the span using the parameters
described in the NA to BS EN 1991-2, NA.2.44.4.
© BSI 2011 • 15
Figure A.4 Dynamic response factor Ψ

x - Main span length (m)

y - Dynamic response factor Ψ
Key
Walking
Jogging
NOTE 1 Main span length, l, is indicated in Figure A.3.

NOTE 2 Values of δ , the logarithmic decrement of decay of vibration for different types of construction are
given in Table A.1.
16 • © BSI 2011
A.5 A method for deriving the maximum vertical

acceleration in crowded conditions
This method is generally applicable.
The maximum vertical acceleration α (m/s2) can be taken as:
 S 
 ∫ ( )
w ′b γ i x dx
γ imax 
1
α = 0
 Mi  2ξ
where:
w’ is the amplitude of the vertical dynamic load (N/m2) given in
the NA to BS EN 1991-2, NA.2.44.3(b)(1). That is:
F 
( )
w ′ = 1.8  0  k fv
 A
γ N λ;
b is the width of the walking surface of the bridge (m);

Mi is the generalized mass of the mode of interest i (kg);
γ i ( x ) is the vertical component of mode shape i at position x (m)
along the span S (m);
NOTE In this equation the absolute, positive value of the
mode amplitude is used in the integration.
γ imax is the maximum vertical component of mode shape I;
ξ is the structural damping when expressed as a damping
δ
ratio: ξ = ;
2π
δ is the logarithmic decrement of decay of vibration between

successive peaks.
For uniform width spans, where the mode shape is approximately
sinusoidal, the first equation in A.5 can be further simplified to:
 0.634w ′bs 2  1
α = γ imax 
 Mi  2ξ
A.6 Damage from forced vibration

Consideration is to be given to the possibility of permanent damage to
a superstructure by a group of pedestrians deliberately causing resonant
oscillations of the superstructure. As a general precaution, therefore,
the bearings need to be of robust construction with adequate provision
to resist upward or lateral movement.
For prestressed concrete construction, resonant oscillation might result
in a reversal of up to 10% of the static live load bending moment.
Providing that sufficient unstressed reinforcement is available to prevent
gross cracking, no further consideration needs to be given to this effect.
© BSI 2011 • 17
A.7 Background to “The avoidance of unstable lateral

bridge responses due to crowd loading”
[NA to BS EN 1991-2:2003, NA.2.44.7]
The stability condition described in the NA to BS EN 1991-2, NA.2.44.4
is significantly different from the comfort criteria described in respect
of vertical vibrations. While the level of comfort experienced by bridge
users is very subjective and varies from person to person and site to
site, the lateral stability condition has a measurable and clearly defined
limit that is not to be exceeded if large uncontrolled lateral motions
are to be avoided.
The process provided in the NA to BS EN 1991-2, NA.2.44.7 is based
on a uniform deck mass. Lower critical densities can arise where the
deck mass per unit length is irregular; a conservative approach would
be to use the lowest mass per unit length. This method might also
be conservative when other parts of the structure, such as arch ribs
or suspension cables, add significantly to the modal mass. In these
circumstances, specialist advice needs to be sought.
At the time of publication, reliable test measurements are only
available for footbridge lateral frequencies in the range of 0,5 to
1,1 Hz. The extensions to the stability curve beyond this region are
based upon a theoretical model of bridge response only and are to be
used with caution.
18 • © BSI 2011
Bibliography
Standards publications
For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including
any amendments) applies.
BS 5400-2:2006 3), Steel, concrete and composite bridges – Part 2:
Specification for loads
BS 5400-10:1980 4), Steel, concrete and composite bridges – Part 10:
Code of practice for fatigue
BS EN 1991-1-3, Eurocode 1 – Actions on structures – Part: 1-3: General
actions – Snow loads
actions – Wind actions
actions – Thermal actions
actions – Actions during execution
actions – Accidental actions
Other publications
[1] GREAT BRITAIN. BD 37/01 Loads for highway bridges, Design
Manual for Roads and Bridges, Vol 1, Section 3, Part 14. London:
The Highways Agency, 2001.
[2] ATKINS HIGHWAYS AND TRANSPORTATION. Background to the
UK National Annexes to EN1990: Basis of Structural Design –
Annex A2: Application for Bridges and EN1991-2: Traffic
Loads on Bridges. The Highways Agency, 2004. http://www.
standardsforhighways.co.uk/tech_info/files/Background%20
to%20UK%20NA_EN1990_EN1991-2.pdf
[3] GREAT BRITAIN. BD 33/94, Expansion joints for use in highway
bridge decks, Design Manual for Roads and Bridges, Vol 2,
Section 3, Part 6. London: The Highways Agency, 1994. http://
www.standardsforhighways.co.uk/dmrb/vol2/section3/bd3394.pdf
[4] FLINT AND NEILL PARTNERSHIP. Derivation of the UK National
Annex to Clause 4.6: Fatigue Load Models: Eurocode 1: Actions
on structures – Part 2: Traffic loads on bridges. The Highways
Agency, 2004. http://www.standardsforhighways.co.uk/tech_info/
files/738-20-Rp01-100204-R1a.pdf
[5] GREAT BRITAIN. BD 21/01 The assessment of highway bridges
and structures, Design Manual for Roads and Bridges, Vol 3,
Section 4, Part 3. London: The Highways Agency, 2001. http://
www.standardsforhighways.co.uk/dmrb/vol3/section4/bd2101.pdf
3)
Now withdrawn and superseded by BS EN 1990:2002+A1 and BS EN 1991-1-7.
4)
Now withdrawn and superseded by BS EN 1993-1-9.
© BSI 2011 • 19
[6] GREAT BRITAIN. BD 86/07 The assessment of highway bridges and

structures for the effects of special types general order (STGO)
and special order (SO) vehicles, Design Manual for Roads and
Bridges, Vol 3, Section 4, Part 19. London: The Highways Agency,
2007. http://www.standardsforhighways.co.uk/dmrb/vol3/section4/
bd8607.pdf
[7] BARKER, C. and MACKENZIE, D. Design Methodology for
Pedestrian Induced Footbridge Vibration, Proceedings of
Footbridge 2008, Porto, 2008.
[8] FRANCE. Footbridges – Assessment of vibrational behaviour
of footbridges under pedestrian loading. Paris: Ministère des
Transports, 2006. http://www.setra.equipement.gouv.fr/IMG/pdf/
US_0644A_Footbridges.pdf
20 • © BSI 2011
This page deliberately left blank

British Standards Institution (BSI)
BSI is the independent national body responsible for preparing British Standards
and other standards-related publications, information and services.
It presents the UK view on standards in Europe and at the international level.
It is incorporated by Royal Charter.
Revisions Information on standards

British Standards are updated by amendment or revision. Users of British Stan- BSI provides a wide range of information on national, European
dards should make sure that they possess the latest amendments or editions. and international standards through its Knowledge Centre.
It is the constant aim of BSI to improve the quality of our products and serv- Tel: +44 (0)20 8996 7004 Fax: +44 (0)20 8996 7005
ices. We would be grateful if anyone finding an inaccuracy or ambiguity while Email: [email protected]
using this British Standard would inform the Secretary of the technical com-
mittee responsible, the identity of which can be found on the inside front Various BSI electronic information services are also available which
cover. give details on all its products and services.
Tel: +44 (0)20 8996 9001 Fax: +44 (0)20 8996 7001 Tel: +44 (0)20 8996 7111 Fax: +44 (0)20 8996 7048
Email: [email protected]
BSI offers Members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards. BSI Subscribing Members are kept up to date with standards
Tel: +44 (0)20 8996 7669 Fax: +44 (0)20 8996 7001 developments and receive substantial discounts on the purchase price
Email: [email protected] of standards. For details of these and other benefits contact Membership Ad-
ministration.
Buying standards Tel: +44 (0)20 8996 7002 Fax: +44 (0)20 8996 7001
You may buy PDF and hard copy versions of standards directly using a
credit card from the BSI Shop on the website www.bsigroup.com/shop. Information regarding online access to British Standards via British
In addition all orders for BSI, international and foreign standards publications Standards Online can be found at www.bsigroup.com/BSOL
can be addressed to BSI Customer Services.
Further information about BSI is available on the BSI website at www.bsi-
Tel: +44 (0)20 8996 9001 Fax: +44 (0)20 8996 7001 group.com/standards
In response to orders for international standards, it is BSI policy to Copyright

supply the BSI implementation of those that have been published
Copyright subsists in all BSI publications. BSI also holds the copyright,
as British Standards, unless otherwise requested.
in the UK, of the publications of the international standardization bodies. Ex-
cept as permitted under the Copyright, Designs and Patents Act 1988 no ex-
tract may be reproduced, stored in a retrieval system or transmitted in any
form or by any means – electronic, photocopying, recording or otherwise –
without prior written permission from BSI. This does not preclude the free
use, in the course of implementing the standard of necessary details such as
symbols, and size, type or grade designations. If these details are to be used
for any other purpose than implementation then the prior written permission
of BSI must be obtained. Details and advice can be obtained from the Copy-
right & Licensing Manager.
Tel: +44 (0)20 8996 7070
BSI Group Headquarters

389 Chiswick High Road London W4 4AL UK
Tel +44 (0)20 8996 9001
Fax +44 (0)20 8996 7001
www.bsigroup.com/standards
raising standards worldwide™

PD6688 2

Uploaded by

Copyright:

Available Formats

PD6688 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PD6688 2

Uploaded by

Copyright:

Available Formats

PD 6688-2:2011

BSI Standards Publication

Background to the National

Traffic loads on bridges

This publication is not to be regarded as a British Standard.

NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAW

raising standards worldwide™

Publishing and copyright information

ISBN 978 0 580 58680 4

ICS 91.010.30; 93.040

The following BSI reference relates to the work on this standard:

Amendments issued since publication

Date Text affected

Relationship with other publications

of the clause. The word “can” is used to express possibility, e.g. a

Contractual and legal considerations

3 BS EN 1991-2:2003, Eurocode 1: Actions on

3.1 Models for loaded lengths greater than 200 m

[BS EN 1991-2:2003, 4.1 (1), Note 2]

3.2 Adjustment factors α for LM1

3.3 Adjustment factor β for LM2

NA to BS EN 1991-2, NA.2.15, which results in similar contact pressures

3.4 Wheel contact surface for LM2

3.5 Upper limit of the braking force on road bridges

3.6 Horizontal forces associated with LM3

3.6.1 Longitudinal braking force

[NA to BS EN 1991-2:2003, 2.18.1]

3.6.2 Centrifugal force

3.7 Horizontal force transmitted by expansion joints or

3.8 Lateral forces on road bridge decks

3.9 Fatigue load models

Dynamic additional amplification factor due to

3.11 Fatigue LM4

3.12 Fatigue LM5 (based on recorded traffic data)

3.13 Effects of collision forces on vehicle restraint systems

3.13.1 Global effects, 4.7.3.3 (1)

loading, has been adopted in the NA to BS EN 1991-2:2003, NA.2.30.2.

3.13.2 Local effects, 4.7.3.3 (2)

UK highway structures, and the former design standard BS 5400-2.

3.14 Actions on pedestrian parapets

situation of the structure and is normally project specific.

Table 1 Loading classes

Class Characteristic value of horizontal uniformly

3.15 Model for vertical loads on abutments and wing

3.15.1 Loading from normal traffic

3.15.2 Loading from special vehicles

3.16 Load models for inspection gangways

3.17 Uniformly distributed load for road and rail

The basis for NA.2.36 can be found in the Background to the UK

x - Loaded length (m)

The model in the NA to BS EN 1991-2 provides a good fit to the

3.18 Concentrated load

3.19 Horizontal force on footbridges

3.20 Dynamic models for pedestrian loads on footbridges

on a project specific basis.

3.21 LM3: Models of special vehicles

x - Structural damping – logarithmic decrement, δ

Annex A (informative) Vibration serviceability recommendations

Table A.1 Typical values of logarithmic decrement of decay of vibration (δ )