PD 6705-2:2020

BSI Standards Publication

Structural use of steel and aluminium

Part 2: Execution of steel bridges conforming to

BS EN 1090-2 – Guide
PD 6705-2:2020 PUBLISHED DOCUMENT

Publishing and copyright information

The BSI copyright notice displayed in this document indicates when the document was last issued.

© The British Standards Institution 2020

Published by BSI Standards Limited 2020

ISBN 978 0 539 12007 3

ICS 91.080.10

The following BSI references relate to the work on this document:

Committee reference B/525/10
Draft for comment 20/30410225 DC

Amendments/corrigenda issued since publication

Date Text affected

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Contents Page

Foreword iii
1 Scope 1
2 Normative references 1
3 Terms and definitions 3
4 General 3
5 Basis for the recommended selections of choice permitted by BS EN 1090-2:2018 3
5.1 Objectives 3
5.2 Reliability level 4
5.3 Types of choice in BS EN 1090-2:2018 4
5.4 Basis for selection of execution class 5
5.5 Quantified service category 5
5.6 Options 6
5.7 Additional information 6
6 Specifications and documentation (see BS EN 1090-2:2018, Clause 4) 7
6.1 Execution class (see BS EN 1090-2:2018, 4.1.2) 7
6.2 Quality documentation (see BS EN 1090-2:2018, 4.2.1) 7
6.3 Quality plan (see BS EN 1090-2:2018, 4.2.2 and Annex C) 7
7 Constituent steel products (see BS EN 1090-2:2018, Clause 5) 7
7.1 Identification, inspection documents and traceability (see BS EN 1090-2:2018, 5.2) 7
7.2 Structural steel products (see BS EN 1090-2:2018, 5.3) 7
7.3 Steel castings (see BS EN 1090-2:2018, 5.4) 9
7.4 Welding consumables for weathering steels (see BS EN 1090-2:2018, Table 6 and 7.5.10) 10
7.5 Bolt assemblies (see BS EN 1090‑2:2018, 5.6) 10
8 Preparation and assembly (see BS EN 1090-2: 2008+A1:2011, Clause 6) 11
8.1 Identification of components (see BS EN 1090-2:2018, 6.2) 11
8.2 Marking methods (see BS EN 1090-2:2018, 6.2) 11
8.3 Quality of thermally cut surfaces (see BS EN 1090-2:2018, 6.4.3) 11
8.4 Hardness of free edge surfaces (see BS EN 1090-2:2018, 6.4.4) 11
8.5 Flame straightening (see BS EN 1090-2:2018, 6.5.3.1 and 6.5.3.2) 12
8.6 Execution of holing (see BS EN 1090-2:2018, 6.6.3) 13
8.7 Cut outs (see BS EN 1090-2:2018, 6.7) 13
8.8 Assembly – Connections for temporary assembly components (see BS EN 1090‑2:2018, 6.9) 13
8.9 Trial assembly (see BS EN 1090-2:2018, 6.10) 13
9 Welding (see BS EN 1090-2:2018, Clause 7) 14
9.1 Quality management (see BS EN 1090-2:2018, 7.1, 7.2.2 and 7.3) 14
9.2 Qualification of welding procedures (see BS EN 1090-2:2018, 7.4.1) 15
9.3 Welding co-ordination (see BS EN 1090-2:2018, 7.4.3) 16
9.4 Preparation and execution of welding (see BS EN 1090-2:2018, 7.5) 16
10 Use of bolting assemblies (see BS EN 1090-2:2018, Clause 8) 17
10.1 Welding of bolted components (see BS EN 1090-2:2018, 8.2.1) 17
10.2 Tightening of preloaded bolting assemblies (see BS EN 1090-2:2018, 8.5) 17
Table 1 — Torque values for the part turn method: step one 19
Table 2 — Rotation values for the part turn method: step two 19
11 Erection (see BS EN 1090-2:2018, Clause 9) 19
11.1 Reference temperature for setting out (see BS EN 1090-2:2018, 9.4.1) 19
11.2 Use of levelling nuts on foundation bolts (see BS EN 1090-2:2018, 9.5.4) 19
11.3 Restoration of damage of site (see BS EN 1090-2:2018, 9.6.3) 19
11.4 Use of shims for fit-up and alignment (see BS EN 1090-2:2018, 9.6.5.3 and 11.2.3.5) 20

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED i

PD 6705-2:2020 PUBLISHED DOCUMENT

12 Surface treatment (see BS EN 1090-2:2018, Clause 10) 20

12.1 Sealing of enclosed spaces (see BS EN 1090-2:2018, 10.6) 20
12.2 Repairs after cutting and welding (see BS EN 1090-2:2018, 10.9) 20
13 Geometrical tolerances (see BS EN 1090-2:2018, Clause 11) 20
13.1 Special tolerances (see BS EN 1090-2:2018, 11.1 and Annex B) 20
13.2 Functional tolerance class (see BS EN 1090-2:2018, 11.3.2 and Annex B) 21
13.3 Alternative criteria (see BS EN 1090-2:2018, 11.3.3 and Annex B) 21
13.4 Geometrical tolerances applicable to bridges (see BS EN 1090-2:2018, Annex B) 21
14 Inspection, testing and correction (see BS EN 1090-2:2018, 7.6 and Clause 12) 22
14.1 Constituent products/components (see BS EN 1090-2:2018, 5.4, 12.2.1 and 12.2.2) 22
Table 3 — Testing and acceptance levels for steel castings 23
14.2 Geometrical dimensions (see BS EN 1090-2:2018, 12.3) 23
14.3 Welding (see BS EN 1090-2:2018, 7.6 and 12.4) 23
14.4 Mechanical fastening (see BS EN 1090-2:2018, 12.5) 25
14.5 Erection 26
Table 4 — Minimum extent of supplementary NDT of shop welds in steel grades up to and
including S355 and QSC F56 27
Table 5 — Adjustments in proportions of supplementary NDT for conditions other than those
covered by Table 4 and Table 6 28
Table 6 — Minimum extent of supplementary NDT of shop welds in steel grades up to and
including S355 and QSCs F71 to F140 29
Table 7 — Weld acceptance criteria for visual inspection 30
Table 8 — Weld acceptance criteria for visual inspection for QSCs F71 to F140 where limits differ
from those for F56 in Table 7 33
Table 9 — Weld acceptance criteria for magnetic particle and penetrant testing 34
Table 10 — Weld acceptance criteria for ultrasonic testing with limited optional
radiographic testing 35
Table 11 — Production tests on run off coupon plates 37
Annex A (informative) Background to the development of European and International execution
standards and their relationship to previous British execution standards 38
Annex B (informative) Method of determining QSC and guidance for use in drafting specifications 41
Table B.1 — Definition of QSC levels 43
Table B.2 — General guide to where highest QSCs might be found in common types of bridge 46
Figure B.1 — Method of designation of QSCs for specific zones on drawings 47
Annex C (informative) Explanatory notes on selected recommendations 47
Figure C.1 — Definition of plain edge and stress raising zones for machine plasma cut
edge surfaces 50
Figure C.2 — Examples of effect of lack-of-fit on clamping force and forces in components in pre-
loaded joints 53
Annex D (informative) Full references of Parts of normative standards not provided in
BS EN 1090‑2:2018, Clause 2 58
Bibliography 63

Summary of pages
This document comprises a front cover, and inside front cover, pages i to iv, pages 1 to 64, an inside back cover and
a back cover.

ii © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Foreword
Publishing information
This part of PD 6705 is published by BSI Standards Limited, under licence from The British Standards
Institution, and came into effect on 30 June 2020. It was prepared by Subcommittee B/525/10,
Bridges, under the authority of Technical Committee B/525, Building and civil engineering structures.
A list of organizations represented on these committees can be obtained on request to their secretary.

Supersession
PD 6705‑2:2020 supersedes PD 6705‑2:2010+A1:2013, which is withdrawn.

Relationship with other publications

This Published Document gives guidance on the use of BS EN 1090‑2:2018 for the execution of steel
bridges in the UK.
BS EN 1993 requires compliance with BS EN 1090 as the key reference standard for fabrication and
erection which is necessary for the design assumptions in BS EN 1993 to be valid.

Information about this document

The guidance given in this Published Document consists of non-contradictory complementary
information (NCCI) to enable the user to apply BS EN 1090‑2:2018 in a safe and economical manner,
with particular reference to the following:
• selection of options where they are defined in BS EN 1090‑2:2018;
• selection of service categories in terms of quantified performance requirements, for use in
choosing execution requirements, where applicable; and
• additional information where permitted in BS EN 1090‑2:2018.
This revision takes account of changes to BS EN 1090‑2 made in the 2018 revision and
comments arising from its application in the drafting and use of specifications based on
BS EN 1090‑2:2008+A1:2011.
This publication can be withdrawn, revised, partially superseded or superseded. Information
regarding the status of this publication can be found in the Standards Catalogue on the BSI website at
bsigroup.com/standards, or by contacting the Customer Services team.
Where websites and webpages have been cited, they are provided for ease of reference and are
correct at the time of publication. The location of a webpage or website, or its contents, cannot
be guaranteed.

Use of this document

As a guide, this Publish Document takes the form of guidance and recommendations. It should not be
quoted as if it were a specification or a code of practice.
This publication is not to be regarded as a British Standard.

Presentational conventions
The provisions in this Published Document are presented in roman (i.e. upright) type. Its
recommendations are expressed in sentences in which the principal auxiliary verb is “should”.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED iii

PD 6705-2:2020 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.
Commentary, explanation and general informative material is presented in smaller italic type, and does
not constitute a normative element.

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.

iv © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

1 Scope
This part of PD 6705 gives guidance on the use of BS EN 1090‑2:2018 for the execution of all types of
steel bridges designed to BS EN 1993.
NOTE As BS EN 1090‑2:2018 contains many clauses which have multiple options or requires additional
information, guidance is given in this Published Document to ensure that technically sound choices are made.

This part of PD 6705 specifies appropriate controls on management systems, procedure approval,
personnel qualification, process selection, quality of materials and workmanship, inspection, testing
and recording.
This part of PD 6705 is applicable to the same scope of application as BS EN 1090‑2:2018 with the
following exceptions:
a) resistance welding; and
b) matters not related to structural integrity, e.g. visual appearance.
The recommendations given in this Published Document are only applicable when the design
requirements and recommendations in the following documents have been adopted, where relevant:
• BS EN 1990, Eurocode – Basis of structural design;
• BS EN 1991, Eurocode 1 – Actions on structures;
• BS EN 1993‑2, Eurocode 3 – Design of steel structures – Part 2: Steel bridges;
• BS EN 1994‑2, Eurocode 4 – Design of composite steel and concrete structures – Part 2: General
rules and rules for bridges; and
• any UK National Annexes and Published Documents referenced normatively or informatively in
the above standards.
This Published Document is prepared for personnel involved in the regulation, design, procurement,
fabrication, erection and certification of steel bridges when BS EN 1090‑2:2018 is used as the basis
for specifying the execution.

2 Normative references
The following referenced documents are indispensable for the application of this document. For
dated references, only the edition cited applies1). For undated references, the latest edition of the
referenced document (including any amendments) applies.
BS EN 1090‑2:2018, Execution of steel structures and aluminium structures – Part 2: Technical
requirements for steel structures
BS EN 1593, Non-destructive testing – Leak testing – Bubble emission techniques
BS EN 1990, Eurocode 0 – Basis of structural design
BS EN 1991, Eurocode 1 – Actions on structures
BS EN 1993, Eurocode 3 – Design of steel structures
BS EN 1993‑1‑9, Eurocode 3 – Design of steel structures – Part 1-9: Fatigue
BS EN 1993‑2, Eurocode 3 – Design of steel structures – Part 2: Steel bridges
BS EN 1994‑2, Eurocode 4 – Design of composite steel and concrete structures – Part 2: General rules
and rules for bridges

1)
Documents that are referred to solely in an informative manner are listed in the Bibliography.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 1

PD 6705-2:2020 PUBLISHED DOCUMENT

BS EN 10025‑1:2004, Hot rolled products of structural steels – Part 1: General technical

delivery conditions
BS EN 10025‑5:2004, Hot rolled products of structural steels – Part 5: Technical delivery conditions for
structural steels with improved atmospheric corrosion resistance
BS EN 10025‑6, Hot rolled products of structural steels – Part 6: Technical delivery conditions for flat
products of high yield strength structural steels in the quenched and tempered condition
BS EN 10088-1, Stainless Steels – Part 1: List of stainless steels
BS EN 10160, Ultrasonic testing of steel flat product of thickness equal or greater than 6 mm
(reflection method)
BS EN 10163‑2, Delivery requirements for surface condition of hot-rolled steel plates, wide flats and
sections – Part 2: Plate and wide flats
BS EN 10163‑3, Delivery requirements for surface condition of hot-rolled steel plates, wide flats and
sections – Part 3: Sections
BS EN 14399, High-strength structural bolting assemblies for preloading
BS EN 14399‑3, High-strength structural bolting assemblies for preloading – Part 3: System HR –
Hexagon bolt and nut assemblies
BS EN ISO 3834‑2, Quality requirements for fusion welding of metallic materials −
Part 2: Comprehensive quality requirements
BS EN ISO 5817, Welding – Fusion-welded joints in steel, nickel, titanium and their alloys (beam welding
excluded) – Quality levels for imperfections
BS EN ISO 6520‑1, Welding and allied processes – Classification of geometric imperfections in metallic
materials – Part 1: Fusion welding
BS EN ISO 14555:2017, Welding − Arc stud welding of metallic materials
BS EN ISO 15613, Specification and qualification of welding procedures for metallic materials –
Qualification based on pre-production welding test
BS EN ISO 15614‑1, Specification and qualification of welding procedures for metallic materials –
Welding procedure test – Part 1: Arc and gas welding of steels and arc welding of nickel and nickel alloys
BS EN ISO 23277, Non-destructive testing of welds – Penetrant testing – Acceptance levels
BS EN ISO 23278, Non-destructive testing of welds – Magnetic particle testing – Acceptance levels
PD 6695‑1‑9, Recommendations for the design of structures to BS EN 1993-1-9
PD 6695‑1‑10, Recommendations for the design of structures to BS EN 1993-1-10
PD 6702‑1, Structural use of aluminium – Part 1: Recommendations for the design of aluminium
structures to BS EN 1999
PD 6705‑3, Structural use of steel and aluminium – Part 3: Recommendations for the execution of
aluminium structures to BS EN 1090-3
ISO 10721‑1, Steel structures – Part 1: Materials and design
ISO 10721‑2, Steel structures – Part 2: Fabrication and erection

2 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

3 Terms and definitions

For the purposes of this part of PD 6705, the terms and definitions given in BS EN 1090‑2:2018 and
the following applies.

3.1 quantified service category (QSC)

category that characterizes a detail, component or structure (or part thereof), in terms of the
circumstances of its use within specified limits of static and cyclic stressing
NOTE See Annex B.

4 General
BS EN 1090‑2 superseded BS 5400‑6 in March 2010 and represented a substantial change in practice
for specifying the execution of steelwork applied to bridges in the UK.
Annex A sets out the background to the main developments in the preparation of BS EN 1090‑2, and
also identifies the main differences in the scope and technical bases of BS EN 1090‑2 as compared
to BS 5400‑6. This is to assist the user in adapting to the changes in practice necessary when using
BS EN 1090‑2:2018.
BS EN 1090‑2:2018 enables persons responsible for preparing the execution specification for steel
bridges to incorporate appropriate requirements relating to those clauses in order to comply with
BS EN 1993‑2. This applies whether the specification is intended to cover steel bridges in general,
particular types of steel bridge, parts of a steel bridge, or a bridge for an individual project.
The recommendations given in this Published Document represent the minimum requirements
needed to ensure that the design conforms to BS EN 1993‑2. Not all recommendations given in
this Published Document are relevant if the scope of the execution specification is limited. The
recommendations do not preclude the addition of further requirements in an execution specification,
provided they do not lower the level of a recommended minimum requirement.
The purpose of this Published Document is to ensure that steel bridges are executed with the same
level of assurance of reliability as that provided in BS 5400‑6.

5 Basis for the recommended selections of choice permitted by BS EN 1090-2:2018

5.1 Objectives
The objective of this Published Document is to ensure that, when using BS EN 1090‑2:2018 for
specifying the execution of steelwork in bridges, the design resistance values assumed in BS EN 1993
are realized in the final bridge construction to the level of reliability in accordance with BS EN 1990,
BS EN 1991 and BS EN 1993.
BS EN 1090‑2:2018 contains multiple options and requirements for further information. Many of
these involve decisions on specialist technical matters which users might not feel qualified to make.
Use of this Published Document should ensure that technically sound choices are made, resulting in
bridges being executed economically whilst maintaining the level of reliability defined above.
This is achieved by specifying appropriate controls on:
a) management systems;
b) procedure approval;
c) personnel qualification;

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 3

PD 6705-2:2020 PUBLISHED DOCUMENT

d) process selection;
e) quality of materials and workmanship;
f) inspection;
g) testing; and
h) recording.
Annex B sets out a method of determining QSC and guidance for its use in drafting specifications.
Annex C gives explanatory notes which detail the background to selected recommendations for
additional specification material, selection of options and choice of execution class, where the issues
involved are of a specialist nature. Annex D gives a list of references of individual Parts of normative
standards which are referenced in BS EN 1090‑2:2018, Clause 2, but where the Part references
are not given.

5.2 Reliability level

The partial factors for actions and material properties chosen by the UK and specified in the NA to
BS EN 1990, NA to BS EN 1991 and NA to BS EN 1993‑2 are based on achieving the same notional
safety margins used for general structural applications.
NOTE In BS EN 1990:2002+A1:2005 this generally aligns with reliability class (RC) 2 which is consistent
with the basis of the safety factors used for most British Standard limit state structural design codes.
BS EN 1990:2002+A1:2005, Annex B recommends that RC2 may be associated with consequence class (CC) 2.

The partial factor ϒM is necessary to allow for inherent deviations in chemical, mechanical and
geometrical properties of the members and joints in the final structure, including execution-induced
stresses and notch-like imperfections. These inherent deviations are set at maximum limits which are
not considered to be economical to reduce further and are defined by the relevant specified quality
levels. The recommendations in this Published Document control these inherent deviations so that
the values of ϒM for use in the UK are valid for all steel bridge applications.
A further factor in achieving reliability is the elimination of gross deviations due to human error.
These are not common occurrences but, if they do occur, can result in serious loss of reliability,
which cannot be economically mitigated against by increasing the values of ϒM. Mitigation against
such deviations is the prime function of quality control measures which provide assurance that the
required quality levels have been met.
In cases where a higher level of reliability is deemed to be necessary (e.g. class RC3 in
BS EN 1990:2002+A1:2005, Annex B), it might be appropriate to increase the scope of testing and
inspection to levels above that recommended in this Published Document.

5.3 Types of choice in BS EN 1090-2:2018

There are three types of choice available to the user of BS EN 1090‑2:2018.
a) Execution class: BS EN 1090‑2:2018, A.3 lists 27 technical requirements where choices can
be made by selection of execution class. Twenty one of these are related to assurance matters,
including quality documentation, identification, inspection documents, traceability, welding
management, qualification of weld procedures, welders and welding coordinators and inspection
of welding and bolting. The remaining six relate to the physical quality required for thermal

4 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

cutting, temporary attachments, butt weld terminations, continuity of weld backing and weld
acceptance criteria.
b) Options: BS EN 1090‑2:2018, A.2 lists nearly 130 requirements which can be subject to variation
by the specifier.
c) Additional information: BS EN 1090‑2:2018, A.1 lists over 80 requirements which, if relevant to
the structure, require further information to be provided by the specifier.

5.4 Basis for selection of execution class

BS EN 1090‑2:2008+A1:2011, Annex B gave informative guidelines on selection of execution class in
Table B.3 in terms of consequences class (CC), service category (SC) and production category (PC).
This information has been removed in BS EN 1090‑2:2018.
BS EN 1090‑2:2018, 4.1.2, Note refers to BS EN 1993‑1‑1:2005+A1:2014, Annex C for the basis for
the selection of execution classes. This normative annex re-confirms that execution class can be
applied either to the works as a whole, an individual component or a detail of a component. It also
contains an optional Table C.1, which is not dissimilar in principle to BS EN 1090‑2:2008+A1:2011,
Table B.3. In Table C.1, however, the recommended choice of execution class is now dependent on
reliability class (RC) or CC. The dependence of execution class on SC is now given directly in terms of
the type of applied loading (static, quasi-static, fatigue or seismic), but without reference to specific
structural applications. Dependence on type of component (referred to as “production category” in
BS EN 1090‑2:2008+A1:2011, Table B.2) is still only applicable to EXC1.
The NA+A1:2014 to BS EN 1993‑1‑1:2005+A1:2014, NA.2.27.3, Table NA.4 gives general, but
not specific, guidelines for selection of execution class in terms of which design standards apply
(BS EN 1993‑1‑9, BS EN 1993‑1‑12 and BS EN 1998 in particular). NA.2.27.3, Note 1 advises that
certain execution classes might need to be enhanced to ensure full compatibility with the design
resistance values in BS EN 1993 and reference is made to the guidance in this Published Document
for that purpose.
The main unresolved issue resulting in the UK’s original negative vote on FprEN 1090‑2:2008,
which was the technical basis of the guidelines and their perceived incompatibility with BS EN 1990,
remains. This particularly applies to the quality related items, as opposed to the assurance related
items [see 5.3a)], which in BS EN 1090‑2:2018 are still linked directly to the reliability level instead
of quantitatively defined service categories. The UK was able to implement BS EN 1090‑2:2018 due to
the flexibility provided in BS EN 1993‑1‑1, Annex C.
In BS EN 1090‑2:2018, where there is a choice of execution class in a particular clause, this Published
Document considers each case separately. In certain clauses the requirements can be the same for
more than one execution class. In these cases, the recommendation is given by the execution class
number(s). For example, if the preferred option is the same for execution class 2, 3 and 4, the choice
is indicated as EXC2/3/4.
The recommendation for selecting a particular execution class (or classes) does not imply any change
in reliability level (see 5.2).
NOTE Where the execution class has been used to differentiate a quality requirement, there might be more than
one recommended class, the distinction being dependent on QSC (see 5.5).

5.5 Quantified service category

The fatigue design rules specified in BS EN 1993‑1‑9 permit high levels of cyclic stressing for certain
categories of detail. These levels are acceptable provided a high quality of execution can be assured,
free from all but very small notches. This demands a high degree of control of execution, particularly
for welding, and inspection procedures for detection and assessment. Such stress levels are usually

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 5

PD 6705-2:2020 PUBLISHED DOCUMENT

only found in limited parts of certain bridges. However, when they do occur, extra controls should be
applied to achieve the correct level of reliability.
BS 5400 (withdrawn), ISO 10721, BS 8118 (withdrawn), PD 6702‑1 and PD 6705‑3 provide simple
means of identifying the higher risk areas. Annex B provides means of calculating and designating
the various levels of fatigue stressing and static stressing. These are compatible with the design
rules in BS EN 1993‑1‑9 and BS EN 1993‑1‑10 including their National Annexes, PD 6695‑1‑9 and
PD 6695‑1‑10. For this purpose, six levels of QSC are defined. They are designated F36, F56, F71, F90,
F112 and F140 in increasing severity of quality requirement.
QSC has effect on the recommended requirements in the following clauses:
• 7.2.3 Surface conditions;
• 7.2.4.1 Internal discontinuity quality class;
• 8.2 Marking methods;
• 8.4 Hardness of free edges;
• 8.6.1 Round holes for fasteners or pins;
• 8.6.2 Slotted holes;
• 8.7.1 Minimum radii of cut outs;
• 8.7.2 Punched cut outs;
• 9.1 Weld quality management;
• 9.2.2 Methods of weld procedure qualification;
• 9.4.3 Temporary welded attachments;
• 9.4.5 Permanent steel backing;
• 9.4.7 Stray arcing and spatter;
• 14.1.2 Inspection and testing of steel castings;
• 14.3.1 Project specific scope of weld inspection;
• 14.3.3 Project specific weld acceptance criteria; and
• 14.3.6 Production weld tests.
Annex B gives guidance on how QSC can be used in the drafting of an execution specification.

5.6 Options
Guidance on selection of any BS EN 1090‑2:2018, A.2 options which could affect the structural
reliability level in a bridge are given in Clause 6 to Clause 14, where relevant. If no guidance is given,
the stated requirement should be assumed to be acceptable for UK bridges.

5.7 Additional information

Guidance on any BS EN 1090‑2:2018, A.1 items of required additional information considered
necessary to control the structural reliability level of a bridge are given in Clause 6 to Clause 14,
where relevant.

6 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

6 Specifications and documentation (see BS EN 1090-2:2018, Clause 4)

6.1 Execution class (see BS EN 1090-2:2018, 4.1.2)
Execution class is used in Clause 6 to Clause 14 of this Published Document for the purposes of
recommending the appropriate requirement where options based on this parameter are given in
BS EN 1090‑2:2018 (see 5.4). In some cases, the applicable recommended requirement might depend
on the QSC (see 5.5).

6.2 Quality documentation (see BS EN 1090-2:2018, 4.2.1)

Quality documentation should conform to the requirements for EXC2/3/4.

6.3 Quality plan (see BS EN 1090-2:2018, 4.2.2 and Annex C)

A quality plan for execution of the works should be provided and include the items recommended in
BS EN 1090‑2:2018, Annex C, if applicable.

7 Constituent steel products (see BS EN 1090-2:2018, Clause 5)

7.1 Identification, inspection documents and traceability (see BS EN 1090-2:2018, 5.2)
7.1.1 Stages of traceability (see BS EN 1090-2:2018, 5.2)
Stages of traceability should conform to the requirements for EXC3/4.

7.1.2 Traceability of individual products (see BS EN 1090-2:2018, 5.2)

Specialist proprietary and/or bespoke products incorporating non-standardized features,
components or methods of manufacture should be individually traceable.
NOTE For example, castings, forgings, cables and their terminations, energy absorbing devices, mechanical
components such as movement joints and major bearings etc.

Small identical bespoke products such as special fasteners should be traceable to their individual
manufacturing lot.

7.1.3 Marking of different grades and/or qualities (see BS EN 1090-2:2018, 5.2)

The identification of different grades and/or qualities of each individual constituent product should
conform to the requirements for EXC2/3/4.

7.2 Structural steel products (see BS EN 1090-2:2018, 5.3)

7.2.1 Options (see BS EN 1090-2:2018, 5.3.1)

7.2.1.1 Standards for steel products

Wrought steel product forms identified in BS EN 1090‑2:2018, Table 2 to Table 4 should be specified
in accordance with the product standards listed in these tables.

7.2.1.2 Options in product standards

The only options usually needed are those covered by 7.2.1.3 to 7.2.1.5, and 7.2.2 to 7.2.5. In
particular, caution should be exercised in the use of optional variations to chemical composition. The
reason for specifying such an option should be given and it should be verified that no adverse effect
on mechanical properties, weldability or other property could result.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 7

PD 6705-2:2020 PUBLISHED DOCUMENT

7.2.1.3 Options for impact properties

The following options for impact property verification should apply:
a) Option 3 in BS EN 10025‑1:2004, 7.3.2.2: the lowest test temperature should be adopted; and
b) Option 3 in BS EN 10025‑5:2004, 7.3.2.2 and 8.4.2: the impact properties of steel grade S355WP
should be verified.

7.2.1.4 Options for suitability for hot dip galvanising

The following options for steel that is to be hot dip galvanized should apply:
a) Option 5 in BS EN 10025‑1:2004, 7.4.3: the product should be suitable for hot dip zinc coating;
b) Option 1.4 in BS EN 10210‑1:2006, 6.7.2: the product should be suitable for hot dip
galvanizing; and
c) Option 1.4 in BS EN 10219‑1:2006, 6.8.2: the product should be suitable for hot dip galvanizing.

7.2.1.5 Options for weld repair of hollow sections

The following options for weld repair of hollow sections in zones of QSC F112 and F140 should apply:
a) Option 1.5 in BS EN 10210‑1:2006, 6.8.4: repair of the body by welding should not be
permitted; and
b) Option 1.5 in BS EN 10219‑1:2006, 6.9.4: repair of the body by welding should not be permitted.

7.2.2 Thickness tolerances (see BS EN 1090-2:2018, 5.3.2)

For thickness tolerance, thickness tolerance class A should be used.
NOTE See Annex C, C.1.

7.2.3 Surface conditions (see BS EN 1090-2:2018, 5.3.3)

7.2.3.1 Plates and wide flats

For plates and wide flats, the limits on surface discontinuities and ground areas should conform to
BS EN 10163‑2, class A or class B. Repair by welding should not be carried out (subclass 3) on steels
conforming to BS EN 10025‑6 or where QSC F112 or F140 applies. Where repairs by welding are
required, they should conform to subclass 2 with the following additional requirements:
a) reports on weld repairs should be submitted;
b) UT and MT should be used for inspection of weld repairs; and
c) weld repairs on components with a QSC of F71 or F90 should be re-checked on delivery.
NOTE See Annex C, C.2.

7.2.3.2 Sections
For sections, excluding hollow sections which are covered by 7.2.3.3, the limits on surface
discontinuities and ground areas should conform to BS EN 10163‑3, class C or class D. Repair by
welding should not be carried out (subclass 3) for steels to BS EN 10025‑6 or where QSC F112 or
F140 applies. Imperfections such as cracks, shells and seams should be repaired if a permissible
repair method is available; otherwise the affected product should be treated as non-conforming.

8 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Where repairs by welding are required, they should conform to subclass 2 with the following
additional requirements:
a) reports on weld repairs should be submitted;
b) UT and MT should be used for inspection of weld repairs; and
c) weld repairs on components with a QSC of F71 or F90 should be re-checked on delivery.
NOTE See Annex C, C.2.

7.2.3.3 Hollow sections

For hot finished and cold formed hollow sections, surface defects in the body of the section should
be repaired in accordance with BS EN 10210‑1 and BS EN 10219‑1 respectively, subject to the
restrictions in 7.2.1.5. Where repairs by welding are required, the following additional requirements
should apply:
a) reports on weld repairs should be submitted;
b) UT and MT should be used for inspection of weld repairs; and
c) weld repairs on components with a QSC of F71 or F90 should be re-checked on delivery.
NOTE See Annex C, C.2.

7.2.4 Additional properties (see BS EN 1090-2:2018, 5.3.4)

7.2.4.1 Cross plates in welded cruciform and welded tee joints – transmitting primary stresses
through the plate thickness
For QSC F56, the internal discontinuity quality class S1 of BS EN 10160 should apply.
For QSC F71 and above, the internal discontinuity quality class S2 of BS EN 10160 should apply.
NOTE See Annex C, C.3.

7.2.4.2 Flange or web plates close to bearing diaphragms and/or single-sided bearing stiffeners
Flange or web plates close to bearing diaphragms and/or single-sided bearing stiffeners, where
attached by welding, should conform to BS EN 10160 internal discontinuity quality class S1 in the
areas defined in BS EN 1090‑2:2018, 5.3.4.
NOTE See Annex C, C.3.

7.2.5 Improved deformation properties (Z testing) perpendicular to the surface

(see BS EN 1090-2:2018, 5.3.4)
The constructor should decide what precautions are needed to avoid lamellar tearing in tee,
cruciform and corner joints. This should be done whether or not improved deformation properties to
BS EN 10164 have been specified for the through plates.
NOTE See Annex C, C.4.

7.3 Steel castings (see BS EN 1090-2:2018, 5.4)

The specification of grades and qualities and selection of options permitted by the product standard
is a specialist matter requiring expert knowledge and practical experience of casting technology.
Only general guidance and principles are given in this Published Document.
In the case of castings used for proprietary products, such as bearings, cable terminations etc., the
casting designer’s calculations and the manufacturer’s quality and testing requirements should be
reviewed prior to acceptance by the bridge designer.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 9

PD 6705-2:2020 PUBLISHED DOCUMENT

In the case of castings designed by the bridge designer, expert advice should be sought at the
design concept stage so that the quality and testing requirements are consistent with the design
requirements. This should include:
a) selection of material designation in accordance with BS EN 10340;
b) review of geometry with regard to castability and inspectability;
c) selection of appropriate quality control options from BS EN 1559‑2, Annex A;
d) weldability (when applicable); and
e) methods and scope of testing and inspection.
In selection of casting designation for satisfying the ULS requirements in BS EN 1993, the
mechanical properties specified for the various grades of wrought products should be the basis of
selection for castings for the same stressing conditions. This should include the proof stress (see
BS EN 10340:2007, 7.2.2.2), the minimum tensile strength and the charpy impact properties (see
BS EN 1993‑1‑10 and its National Annex). In the latter case, the lowest test temperature should be
used (see BS EN 10340:2007, 7.2.2.3).
The chemical composition limits for the selected casting designation in BS EN 10340:2007, Table 1
should be taken into account if castings are to be welded. These should be taken into account when
developing and qualifying weld procedures for cast steels (e.g. carbon equivalent values, sulphur
limits, etc.).
Recommendations for the scope of testing and NDT acceptance levels are given in 14.1.2.
NOTE See Annex C, C.5.

7.4 Welding consumables for weathering steels (see BS EN 1090-2:2018,

Table 6 and 7.5.10)
Options 1, 2 and 3 in BS EN 1090‑2:2018, Table 6 are suitable for the purposes of weather resistance
and weldability.
C-Mn consumables may be used, but only for:
a) single run fillet welds not exceeding 8 mm leg length using processes 135 and 121;
b) butt welds using one run only per side; and
c) the body of multi-pass fillet or butt welds except for the outer 3 mm thick surface zones of
the cap, root and ends (excluding any overfill). The surface zones should be deposited using
consumables conforming to options 1, 2 or 3.
NOTE See Annex C, C.6.

7.5 Bolt assemblies (see BS EN 1090‑2:2018, 5.6)

7.5.1 Preloaded bolting assemblies (see BS EN 1090‑2:2018, 5.6.4)
The following products should be used for preloaded bolting assemblies using all tightening methods,
except the HRC method:
a) bolting assemblies conforming to BS EN 14399‑3 (HR type); and
b) nuts fully lubricated by the manufacturer after the nuts have been tapped (and, if relevant, after
any coating passivation process has been completed). Lubricant coatings which are permanently
liquid and could risk contamination of faying surfaces should not be used (lubricants such as dip
spin applied dry film friction control fluid would be suitable, for example).

10 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Bolts of property class 10.9 should not be used for assemblies tightened by the part turn method
(see 10.2.6).
Electroplated property class 10.9 bolting assemblies should not be used.
Cleaning of property class 10.9 bolting assemblies prior to hot dip galvanizing should be by blast
cleaning and not by pickling.
Nuts of property class 10 may be used with bolts of property class 8.8.
NOTE See Annex C, C.7.

7.5.2 Foundation bolts (see BS EN 1090-2:2018, 5.6.7)

Reinforcing steels should not be used for foundation bolts subject to applied tension forces.

7.5.3 Locking devices (see BS EN 1090-2:2018, 5.6.8 and 8.2.1)

Prevailing torque type nuts should not be used as a substitute for nuts conforming to BS EN 14399
when used with bolts conforming to BS EN 14399 in preloaded joints.
NOTE Preloaded assemblies installed and tightened in accordance with BS EN 1090‑2:2018 are resistant
to loosening.

7.5.4 Solid rivets for hot riveting (see BS EN 1090-2:2018, 5.6.10)

Solid rivets for hot riveting should conform to BS 4620.

8 Preparation and assembly (see BS EN 1090-2: 2008+A1:2011, Clause 6)

8.1 Identification of components (see BS EN 1090-2:2018, 6.2)
A suitable identification system should identify each finished component to the documentation
supplied for its constituent product(s).
NOTE Examples of types of documentation are given in BS EN 1090‑2:2018, 12.2.1, Note 1.

8.2 Marking methods (see BS EN 1090-2:2018, 6.2)

Hard stamps, punched or drilled marks may only be used for zones with a QSC of F56 and below.
Soft or low stress stamps may be used for zones with a QSC of F90 and below.
No type of marking process involving deformation of or removal of metal from the surface should be
used for zones with a QSC of F112 and above.
Markings should be visible when steelwork is assembled and erected.

8.3 Quality of thermally cut surfaces (see BS EN 1090-2:2018, 6.4.3)

The quality of cut surfaces should conform to the requirements for EXC3/4.

8.4 Hardness of free edge surfaces (see BS EN 1090-2:2018, 6.4.4)

Free edge surfaces with hardnesses exceeding 380 (HV10) values should be ground, linished or
machined until sufficient material has been removed such that the 380 (HV10) limit is not exceeded,
subject to the following exceptions.
The hardness limit of 380 (HV10) is not applicable to machine plasma cut edges where all the
following conditions are met:
a) the steel grade does not exceed S420;
b) plain edge surfaces are not subject to a QSC exceeding F71 [subject to c)];

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 11

PD 6705-2:2020 PUBLISHED DOCUMENT

c) edge surfaces within 25 mm of stress raising features, such as re-entrant corners, openings, weld
terminations on the edge not subject to a QSC exceeding F36; and
d) edge surfaces are not subject to subsequent cold forming (unless fully fused over the full length
and width by subsequent welding).
NOTE See Annex C, C.8.

8.5 Flame straightening (see BS EN 1090-2:2018, 6.5.3.1 and 6.5.3.2)

Documented procedures for correction of distortion using flame straightening should be developed
for all steel grades and for the following typical modes of distortion:
a) an out-of-plane bow of a flat element;
b) an in-plane bow of a flat element or a bow about the minor axis of an open section; and
c) a bow about either axis of a hollow section or about the major axis of an open section.
The method of heating should be by oxy-fuel gas (acetylene recommended) using equipment with
purpose made heating nozzles. The fuel gas mixture, nozzle size and arrangement, flame distance,
heat zones and sequence of heating should be specified.
The surface temperature of the heated zones should be measured and controlled using a contact
pyrometer or temperature sensitive crayons.
Maximum surface temperatures, according to the following material groups and delivery conditions,
should not be greater than:
1) 625 °C for steels conforming to BS EN 10025‑2, BS EN 10025‑3, BS EN 10025‑5 and BS EN 10210,
but excluding those supplied in the “+M” delivery condition [see 2)]; or
2) 550 °C for steels conforming to BS EN 10025‑4, BS EN 10025‑6 and BS EN 10219, including any
other steels supplied in the “+M” delivery condition.
The temperatures in 1) and 2) are irrespective of the temperatures reached during any flame
straightening procedure trials.
The cooling phase down to 350 °C should be unassisted. Below this temperature, cooling may be
assisted by an air blast. Quenching by water should not be used except where permitted for stainless
steels [see BS EN 1090‑2:2018, 6.5.3.2c)].
Passive restraints may be used during the heating cycle. Active application of force during the heating
cycle, for example by jacking, should not be used.
For steel grades greater than S355, procedures should be qualified by applying the method to a test
piece of appropriate form and known rolling direction, thickness and material grade.
Following completion of the qualification test and cooling to ambient temperature, a set of
longitudinal tensile, impact and hardness tests should be taken from 2 mm below the material
surface of the heated zone where the maximum specified temperature had been sustained for the
longest period. The results, as tested in accordance with BS EN 10025‑1, should demonstrate that the
specified mechanical properties of the product have been maintained.
A conforming procedure qualification test should qualify procedures within the following limits:
i) all material grades and subgrades up to and including those tested;
ii) all element thicknesses up to and including that tested; and
iii) the number of full thermal cycles per heat zone up to and including the number applied
in the test.

12 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Operators should be assessed and approved by the welding coordinator to ensure satisfactory
understanding and capability of implementing the documented instruction for correction.
NOTE See Annex C, C.9.

8.6 Execution of holing (see BS EN 1090-2:2018, 6.6.3)

8.6.1 Round holes for fasteners or pins
Round holes for fasteners or pins should be drilled, or be punched at least 2 mm undersize in
diameter and reamed after punching, where any of the following conditions apply:
a) the material thickness is greater than 3 mm;
b) the QSC is F71 or higher;
c) the joint is designed to be slip resistant; or
d) the joint is designed to be preloaded.
NOTE See Annex C, C.10.

8.6.2 Slotted holes

The restrictions on forming slotted holes should be as for round holes as set out in 8.6.1, except that:
a) machining (e.g. reaming, milling) should be used to complete slotted holes formed by drilling
or punching;
b) hand thermal cutting of slotted holes should not be permitted in any material thickness; and
c) slotted holes should not be permitted in zones with a QSC above F112.

8.7 Cut outs (see BS EN 1090-2:2018, 6.7)

8.7.1 Minimum radii
The minimum radius for re-entrant corners and notches should be 5 mm for QSC F56 and 10 mm for
QSC F71 and above, subject to a smaller radius not being specified.

8.7.2 Punched cut outs

Punched cut outs should not be permitted where a QSC of F56 or above is required.

8.8 Assembly – Connections for temporary assembly components

(see BS EN 1090‑2:2018, 6.9)
Connections for temporary components which can affect the works in service, including the
reinstatement or sealing of holes and the removal or otherwise of welded attachments, should be
agreed with the designer before execution starts (see 9.4.3).

8.9 Trial assembly (see BS EN 1090-2:2018, 6.10)

A full or staged trial assembly should be undertaken, where it is necessary to prove fit-up between
components and critical dimensions especially where tolerances on overall geometry or joint fit-up
are small. This applies particularly to bolted joints with butting faces.
A full or staged trial assembly of the superstructure should be undertaken where:
a) erection operations are to be undertaken within restricted possession times;
b) the vertical camber is critical to the correct dead weight stress distribution in continuous main
girders, particularly in cable stayed or suspension bridges; and/or
c) fit-up tolerances are small in latticed structures which contain heavy members and rigid joints.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 13

PD 6705-2:2020 PUBLISHED DOCUMENT

The correction of hole alignment by reaming of bolted splices, selection of pack thicknesses,
correction of weld preparation fit-up and correct positioning of temporary alignment cleats can be
carried out during trial assembly.
The following factors should be taken into account:
1) support of the assembly parts should minimize self-weight stresses, where the unstressed
camber profile is being checked;
2) correct alignment of assembly parts should be re-established in staged trial assembly;
3) differential temperature distributions throughout the steelwork should be controlled or allowed
for if exposed to direct sunlight when surveys are made;
4) unique marking for identification and orientation of individual members should be used;
5) all shop welding should be complete;
6) all shop bolting should be complete; and
7) lack of fit between components should not be corrected by bolt tensioning.
These measures ensure that long range locked-in forces, beyond those allowed for in the design,
are controlled within acceptable limits, e.g. incorrect dead load bending moment distribution in
continuous girders.

9 Welding (see BS EN 1090-2:2018, Clause 7)

9.1 Quality management (see BS EN 1090-2:2018, 7.1, 7.2.2 and 7.3)
The quality requirements given in BS EN ISO 3834 should conform to the requirements for EXC3/4.
The exception to this is where quality requirements for EXC2 might be suitable for simple bridges of
limited span using steel grades of S355 or below, provided a QSC of no more than F56 is required.
The quality requirements given in BS EN ISO 3834 for EXC1 should not be applied to bridges.
The constructor should ensure that the range of bridge types for which their welding management
systems have been approved are clearly defined on a schedule. The schedule should support
any certification in terms of type and size of construction, material grades, product forms and
thicknesses, execution processes and QSC.
NOTE See Annex C, C.11.

When applying the conditions for welding cold formed zones in accordance with
BS EN 1993‑1‑8:2005, 4.14, Table 4.2, the “predominantly static loading” condition should be
deemed to apply to a QSC of F56 and the “fatigue predominates” condition to QSCs of F71 and above.
Attention is drawn to the change in BS EN 1090‑2:2018, 7.3, which now states that
any welding process listed in BS EN ISO 4063 may be used for structural steelwork. In
BS EN 1090‑2:2008+A1:2011, 7.3, welding processes were restricted to those for which standards
for weld procedure qualification were listed in Table 12 and Table 13 of that standard. There are
at least 20 other basic types of welding process listed in BS EN 4063 which are capable of welding
steel and for which the standards listed in BS EN 1090‑2:2018 do not adequately cover key quality
management activities. These include activities such as weld procedure specification, weld and
welder procedure qualification, welding process control, testing, inspection and acceptance criteria. If
such welding processes are to be used, the proposed welding facilities and personnel should undergo
a review by a specialist in the process concerned prior to the selection of that process for the work.

14 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

9.2 Qualification of welding procedures (see BS EN 1090-2:2018, 7.4.1)

9.2.1 Tack welds (see BS EN 1090-2:2018, 7.4.1.1 and 7.5.7)
Tack welding procedures should be qualified as required by EXC2/3/4 in BS EN 1090‑2:2018, 7.5.7.
The welding procedure specification (WPS) should include the special deposition conditions and the
tests described below.
The tolerances on cross section and length of the tack weld and any subsequent profiling
requirements, such as feathering of ends, should be specified in the WPS.
To qualify the procedure for a tack weld where it is to be incorporated in the joint, the maximum
permitted cross section of tack weld should be used in the qualification test. The positions of the ends
of the tack weld should be marked on the test piece.
Macroscopic examination and hardness testing should be carried out within the tack weld length.
This test should not be required if the original test for qualifying the complete weld includes the tack
weld tested as described in this paragraph.
NOTE See Annex C, C.12.

9.2.2 Methods of qualification (see BS EN 1090-2:2018, 7.4.1.2 and 7.4.1.3)

Methods of qualification should be in accordance with those permitted for EXC3/4 in
BS EN 1090‑2:2018, Table 12. Qualification by means of a standard welding procedure should not be
permitted. For method of qualification of tack welds, see 9.2.1.
The welding coordinator should take account of the fact that some of the imperfection acceptance
limits in BS EN 15614‑1 are less restrictive than are required for factory production control (FPC)
given in BS EN 1090‑2:2018, 7.6, particularly for the higher EXCs. The limits given in BS EN 15614‑1
might also be less restrictive than are required for fitness for purpose (FFP) given in 14.3.3,
particularly for higher QSCs. In the latter case, the relevant WPS should be reviewed, together with
the associated WPQR test record (where available), in order to establish whether any additional
controls need to be added to the WPS to ensure that the FFP limits have an adequate probability
of being met in production; for example, restrictions on tolerances for joint preparation/fit-up and
welding parameters, and limitations on welding position. In cases of doubt that the FFP limits might
not be met, a pre-production test of the joint concerned should be carried out, followed by a non-
destructive and/or destructive examination, as appropriate, to provide the necessary assurance.
NOTE See Annex C, C.13.

Alternatives to the cruciform tensile test to BS EN ISO 9018 for fillet welds on steel grades ≥ S460
should not be permitted.
Welding procedure qualification for welding process numbers 783, 784 and 786 should be
carried out by a welding procedure test in accordance with BS EN ISO 14555:2017, 10.2. For
process 783, the examination and testing of test pieces should be carried out in accordance
with BS EN ISO 14555:2017, Table 1 for comprehensive quality requirements according to
BS EN ISO 3834‑2.
Fillet, partial penetration and full penetration tee joint welds should be qualified by a supplementary
in-line butt weld procedure qualification record to demonstrate mechanical properties in accordance
with BS EN ISO 15614‑1:2017, Table 2 footnote f.

9.2.3 Welded joints in reinforcing steel

Qualification testing for welded joints involving reinforcing steel should be carried out in accordance
with BS EN ISO 17660‑1, irrespective of whether a joint is nominally classified as “load” or “non-
load” bearing.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 15

PD 6705-2:2020 PUBLISHED DOCUMENT

NOTE See NA to BS EN 17660‑1:2006, 3.2 and NA to BS EN ISO 17660‑2:2006, 3.2.

9.3 Welding co-ordination (see BS EN 1090-2:2018, 7.4.3)

Welding co-ordination should conform to the requirements for EXC2/3/4. This Published Document
gives no specific guidance on selection of execution class from BS EN 1090‑2:2018, Table 14 and
Table 15. This is because the level of technical knowledge of welding co-ordination personnel is
decided by the certification body responsible for determining the conditions of certification of the
constructor. This is in accordance with BS EN ISO 3834 (see 9.1).
NOTE See Annex C, C.14.

9.4 Preparation and execution of welding (see BS EN 1090-2:2018, 7.5)

9.4.1 Prefabrication primers: weld procedure tests (see BS EN 1090-2:2018, 7.5.1.1)
Where prefabrication primers are to be left on fusion faces or heat affected zones, the weld
procedures should be tested in accordance with the requirements for EXC2/3/4.

9.4.2 Welded joints in hollow sections (see BS EN 1090-2:2018, 7.5.4, 7.5.8.1, 7.5.11
and Annex E)
The guidance on preparation, assembly and welding given in BS EN 1090‑2:2018, Annex E should
be applied. With regards to the assembly of hollow sections, attention should be paid to the
following points.
a) Adequate access for penetration to the root in the case of full penetration butt welds. Where
access is restricted, a pre-production welding test conforming to BS EN ISO 15613 should be
conducted using the tolerances on preparation and fit-up that give the most restricted access, to
demonstrate that the required penetration can be achieved.
b) Compensation for loss of throat and leg length in the case of fillet welded joints with root gaps
(see also BS EN 1090‑2:2018, 7.5.8.1).

9.4.3 Temporary attachments: use and removal (see BS EN 1090-2:2018, 7.5.6)

In accordance with EXC3/4, restrictions on the use of temporary attachments should conform
to a) to c).
a) Temporary welded attachments should not be used in zones of QSCs F112 and above.
b) Except where excluded in c), removal of temporary welded attachments by thermal cutting,
gouging and chipping may be used provided that the finished cut surface of the remainder of
the attachment is at least 3 mm from the permanent member surface prior to grinding smooth.
Remnants of temporary materials proud of the original member surface should be removed by
grinding parallel to the member axis and the ground surface checked for freedom from cracking
using magnetic particle testing.
NOTE See Annex C, C.15.

c) Gouging and chipping should not be used on steel grades ≥ S460 or where a QSC of F71 or
above applies.

9.4.4 Butt weld run-on/run-off pieces (see BS EN 1090-2:2018, 7.5.9.1)

Run-on and run-off pieces should be used on all butt welds and should conform to the requirements
for EXC2/3/4, where access allows.

16 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

9.4.5 Permanent steel backing (see BS EN 1090-2:2018, 7.5.9.2)

Permanent steel backing should only be used if specified by the designer or if approved by the
designer in advance of execution. The continuity of permanent steel backing should conform to the
requirements for EXC3/4. The execution of the continuity weld should depend upon the QSC related
to stresses parallel to the axis of the backing component, as follows:
a) F36: no additional recommendations;
b) F56 to F90: make continuous by full penetration butt weld at any stage, followed by surface crack
detection; and
c) F112 to F140: make continuous by full penetration butt weld, followed by flush grinding and
surface crack detection before assembly.
NOTE See Annex C, C.16.

9.4.6 Stud welding (see BS EN 1090-2:2018, 7.5.12)

With reference to BS EN ISO 14555, the comprehensive quality requirements in accordance with
BS EN ISO 3834‑2 should be adopted.

9.4.7 Stray arcing and spatter (see BS EN 1090-2:2018, 7.5.16)

Stray arc sites should be checked for cracking by PT or MT on all steel grades for QSC F90 and above.
Weld spatter should be removed from all surfaces where protective coating is to be applied and from
all other surfaces where QSC F90 and above applies.

10 Use of bolting assemblies (see BS EN 1090-2:2018, Clause 8)

10.1 Welding of bolted components (see BS EN 1090-2:2018, 8.2.1)
Welding of nuts, bolts or washers should not be used on bolting assemblies above property
class 4.6. Use of welding on property class 4.6 bolting assemblies should be subject to the following
restrictions:
a) the QSC of the parent metal is not greater than F36;
b) the welded component is not coated; and
c) neither the diameter of the bolt nor the thickness of the parent material is more than 12 mm.
Written approval from the designer should be obtained prior to the welding of property class 4.6
nuts, bolts and washers. The welding should be subject to the recommendations in Clause 9, 14.3 and
BS EN 1090‑2:2018, Clause 7.
NOTE See Annex C, C.17.

Weld procedure qualification should be carried out in accordance with BS EN ISO 15613.

10.2 Tightening of preloaded bolting assemblies (see BS EN 1090-2:2018, 8.5)

10.2.1 Joint fit-up (see BS EN 1090-2:2018, 8.5.1)

10.2.1.1 Slip resistant connections

At the snug-tight stage, the exposed surfaces of the components being connected should be
aligned within 1 mm where they emerge from beneath the cover plate(s) at the joint plane. If this
tolerance is exceeded, the cover plates should be removed and corrective steel packings installed.
The tolerance on thickness difference between the two components (including any packing plates)
in BS EN 1090‑2:2018, 8.1 should also be met. To achieve adequate contact between the friction

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 17

PD 6705-2:2020 PUBLISHED DOCUMENT

surfaces, the out-of-plane bending stiffness of cover plates should be limited. This can be achieved
by using two plates of half the thickness each when the thickness of the original cover plate exceeds
the bolt size.
NOTE See Annex C, C.18.

10.2.1.2 Connections subject to applied tension

Where a connection is designed to transfer applied tensile forces through a flanged and bolted end
plate, as a minimum, the area of the mating surfaces in direct line with the tension member cross
section should be in contact when all bolts are snug-tight. Any remaining gaps in this zone should
be filled by steel shims prior to the application of the remaining preload. If necessary, bolts can be
partially slackened to enable shims to be inserted. The shims should not be loose at the final snug-
tight stage. In the event of a tapering gap, a purpose machined tapered steel shim may be used.
As an alternative, flat (parallel) steel shims of not less than 0,1 mm and of not more than 0,2 mm
thick should be inserted to refusal, in steps until the gap in the contact zone is filled, followed by
retightening to the snug-tight stage. If sealing of the joint is required, the remaining gaps should be
filled with a corrosion protection medium compatible with the corrosion protection to be applied to
the joint after final tightening of the bolt assemblies.
NOTE See Annex C, C.18.

10.2.2 Minimum preloading force (see BS EN 1090-2:2018, 8.5.1)

The values of the nominal minimum preloading force Fp,C should be in accordance with
BS EN 1090‑2:2018, Table 18 for all tightening methods.
NOTE See Annex C, C.19.

10.2.3 Torque method (see BS EN 1090-2:2018, 8.5.3)

The torque method should not be used for the tightening of preloaded bolts, except where joints are
required to be disassembled in service and fastener assemblies reused.
NOTE See Annex C, C.20.

10.2.4 Combined method (see BS EN 1090-2:2018, 8.5.4)

For property classes 8.8 and 10.9 bolting assemblies conforming to BS EN 14399‑3 (see 7.5.1),
the further rotation values in BS EN 1090‑2:2018, Table 21 should be as follows for the same
thickness ranges:
a) replace 60° by 90° (1/4 turn);
b) replace 90° by 120° (1/3 turn); and
c) replace 120° by 180° (1/2 turn).
The k value should be checked daily with the torque wrenches in accordance with
BS EN 1090‑2:2018, Annex H and the torque for the first tightening step adjusted accordingly. Any
batches that fail to conform to this test should be discarded.
NOTE See Annex C, C.21.

The simplified value of 0,75 Mr,1 = 0,094 d Fp,C should not be used.

10.2.5 HRC bolt tensioning method (see BS EN 1090-2:2018, 8.5.5)

The preload should be checked on sample assemblies from each assembly lot within seven days prior
to use, in accordance with BS EN 1090‑2:2018, Annex H. Any assembly lots showing values of preload
below the minimum specified value should be discarded.

18 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

The pre-tightening step should be repeated until the shear wrench outer socket has stopped turning
on all assemblies.

10.2.6 Other tightening methods (see BS EN 1090-2:2018, 8.5.1)

The part turn method may be used for tightening preloaded bolts if all the following conditions apply:
a) the bolts are property class 8.8 and assemblies conform to 7.5.1;
b) the required nominal minimum preloading force is not in excess of the values specified in
BS EN 1090‑2:2018, Table 18; and
c) the joint fit-up conforms to 10.2.1.
The part turn method should be carried out as for the combined method in BS EN 1090‑2:2018, 8.5.4,
except that:
d) K0 bolt assemblies may be used;
e) in the first tightening step, the nut should be tightened to the specified torque, as given in
Table 1; and
f) in the second tightening step, the nut should be tightened to the specified turn as
given in Table 2.
Table 1 — Torque values for the part turn method: step one

Bolt diameter 12 14 16 18 20 22 24 27 30 36
d, mm
Torque value 40 60 80 110 160 210 270 340 460 810
M, Nm

Table 2 — Rotation values for the part turn method: step two

Bolt diameters Grip lengthA) Rotation

mm mm °
12 to 22 t ≤ 115 180 (½ turn)
115 < t ≤ 275 270 (¾ turn)
24 to 36 t ≤ 160 180 (½ turn)
160 < t ≤ 350 270 (¾ turn)
A)
Equal to nominal thickness, t, as defined in BS EN 1090‑2:2018, Table 21.

NOTE See Annex C, C.22.

11 Erection (see BS EN 1090-2:2018, Clause 9)

11.1 Reference temperature for setting out (see BS EN 1090-2:2018, 9.4.1)
The reference temperature should be agreed with the designer in writing before execution
commences and recorded in the quality plan.

11.2 Use of levelling nuts on foundation bolts (see BS EN 1090-2:2018, 9.5.4)

Levelling nuts should be slackened off before final tensioning of foundation bolts where a post-
tensioning force is specified, to allow the post-tensioning force to be transferred to the foundations.

11.3 Restoration of damage of site (see BS EN 1090-2:2018, 9.6.3)

Procedures for restoration of damage should be documented, as required by EXC2/3/4.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 19

PD 6705-2:2020 PUBLISHED DOCUMENT

11.4 Use of shims for fit-up and alignment (see BS EN 1090-2:2018, 9.6.5.3 and 11.2.3.5)
The use of shims should be avoided where possible by suitable controls on preparation, assembly and
weld distortion and, if necessary, by machining.
If welding is required for securing purposes, it should be carried out in accordance with
BS EN 1090‑2:2018, Clause 7. The recommendations in Clause 9 and 14.3 should apply in
this context.
Where steel shims are needed to correct the fit between members, particularly in preloaded tension
joints (see 10.2.1), and where gaps are variable across the interface, flat (parallel) steel shims may be
used as an alternative to a purpose machined tapered steel shim. The gap should be filled by insertion
of shims of not less than 0,1 mm and not more than 0,2 mm thick from the widest gap position until
refusal. The excess material should be cut off on completion.

12 Surface treatment (see BS EN 1090-2:2018, Clause 10)

12.1 Sealing of enclosed spaces (see BS EN 1090-2:2018, 10.6)
All internal spaces which have been identified as being hermetically sealed should have all joints,
whether welded, mechanically fastened or bonded, leak tested prior to application of the external
protective coating (see BS EN 1779). The bubble emission technique in accordance with BS EN 1593
should be used.
Where spaces are to be fully enclosed by sealing welds to prevent ingress of moisture, weld
imperfections involving surface breaking voids otherwise permitted under the welding specification
should be sealed using an appropriate weld repair.

12.2 Repairs after cutting and welding (see BS EN 1090-2:2018, 10.9)

Protective treatment on edges and adjacent surfaces which have been damaged by cutting or after
welding should be restored in accordance with the original specification.
Coatings on pre-coated constituent products that have been damaged by welding should be restored
in accordance with the original specification or, if not practicable, by an alternative treatment
approved by the designer.

13 Geometrical tolerances (see BS EN 1090-2:2018, Clause 11)

13.1 Special tolerances (see BS EN 1090-2:2018, 11.1 and Annex B)
Where full contact end bearing is specified (see BS EN 1090‑2:2018, B.19), the maximum gap
between the two surfaces should be limited to 0,5 mm. In the case of fitted web stiffeners, the gap
should also be limited to a maximum of 0,25 mm over 60% of the fitted area.

20 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

13.2 Functional tolerance class (see BS EN 1090-2:2018, 11.3.2 and Annex B)

Functional tolerance class 1 should be adopted, except in the following cases from
BS EN 1090‑2:2018 where class 2 should be adopted:
a) Table B.1, No 1 to No 6, at bearing and bearing stiffener locations; and
NOTE Certain bearing designs might require machined surfaces.

b) Table B.6, No 3 and No 4.

13.3 Alternative criteria (see BS EN 1090-2:2018, 11.3.3 and Annex B)

The alternative criteria specified in BS EN 1090‑2:2018, 11.3.3 should only be used if there are
reasons why those in BS EN 1090‑2:2018, 11.3.2 and the criteria recommended in this Published
Document are not appropriate.

13.4 Geometrical tolerances applicable to bridges (see BS EN 1090-2:2018, Annex B)

The following tables in BS EN 1090‑2:2018, Annex B contain manufacturing tolerance requirements
which are potentially applicable to bridges:
• B.1 Welded profiles;
• B.2 Press braked profiles;
• B.3 Flanges of welded profiles;
• B.4 Flanges of welded box sections;
• B.5 Web stiffeners and cruciform joints of profiles or box sections;
• B.6 Components;
• B.7 Stiffened plating;
• B.8 Fastener holes, notches and cut edges;
• B.9 Crane beams;
• B.10 Column splices and baseplates;
• B.11 Cylindrical and conical shells;
• B.12 Lattice components; and
• B.13 Bridge decks.
The following tables in BS EN 1090‑2:2018, Annex B contain erection tolerance requirements which
are potentially applicable to bridges:
• B.19 Full contact end bearing;
• B.21 Bridge decks;
• B.22 Crane runways;
• B.23 Concrete foundations and supports; and
• B.25 Beams subject to bending and components subjected to compression.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 21

PD 6705-2:2020 PUBLISHED DOCUMENT

14 Inspection, testing and correction (see BS EN 1090-2:2018, 7.6 and Clause 12)
14.1 Constituent products/components (see BS EN 1090-2:2018, 5.4, 12.2.1 and 12.2.2)
14.1.1 Specific testing
Specific testing of proprietary products not covered by the European or International standards
referenced in BS EN 1090‑2:2018 should be specified.
NOTE For example, this might include special fasteners, tension connectors or mechanical components required
for articulation, movement or control of displacement.

Prototype testing should be included to provide evidence of structural performance to prove the
design and establish quality criteria, and production tests to provide evidence that the quality
requirements are being met in all supplied items.

14.1.2 Steel castings

The recommended testing requirements are given in Table 3. If the cyclic stressing levels vary
significantly throughout the casting, the zones of QSCs may be indicated as shown in Figure B.1.
NOTE See Annex C, C.5.

22 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Table 3 — Testing and acceptance levels for steel castings

BS EN 1090‑2:2018, Testing requirement QSC

5.4 Item F56 F71 F90 F112 F140
a) Visual inspection % of lot 100 100 100 100 100
b) Destructive tests Test piece type Product sample
c) Non-destructive tests 6) MT % of lot 20 100 100 100 100
7) UT % of lot 20 50 100 100 100
7) RT % of lot 20 20 50 100 100
Acceptance criteria MT 2 2 1 NP A)
NPA)
severity level SM
MT 3 2 1 NPA) NPA)
severity level LM/AM
UT 2 1 1 NPA) NPA)
severity level
RT 3 2 1 NPA) NPA)
severity level B)
A)
NP = no indications permitted.
B)
In accordance with BS EN 12681.

14.2 Geometrical dimensions (see BS EN 1090-2:2018, 12.3)

14.2.1 Measurement location and frequency
The inspection plan should identify all dimensions which could be at risk of non-conformity. Checks
should be made during execution and action taken to correct any non-conformity at the earliest
opportunity. All such dimensions should be checked before the steelwork is dispatched to site.
Components in the neighbourhood of site connections should be re-checked once the connections
are complete.

14.2.2 Repair of dents in hollow sections

Repair by means of welded cover plates should not be undertaken without the approval of
the designer.

14.3 Welding (see BS EN 1090-2:2018, 7.6 and 12.4)

14.3.1 Scope of project specific inspection (see BS EN 1090-2:2018, 12.4.2.4)
The scope of specific inspection for each joint on a bridge should be 100% visual and in accordance
with the extent and methods of supplementary non-destructive testing (NDT) given in Table 4,
Table 5 and Table 6 of this Published Document.
NOTE 1 Table 4 to Table 11 follow 14.5.

Table 4 specifies the minimum proportions of welded joints subjected to supplementary NDT for
shop welds in steel grades up to and including S355 and QSC F56, which represent the most common
conditions. The percentages in Table 4 should be applied to the number of joints tested up to weld
lengths of 1 m in any joint. For joints with weld lengths exceeding 1 m, the percentage should be
applied to the proportion of weld in every joint.
Where other conditions apply, the minimum recommended proportions of welded joints should be
increased (or decreased) by one or more levels in accordance with Table 5. This should be done by
applying the next highest (or lowest) proportion using the following sequence of seven increasing
levels of proportion: 0%, 2%, 5%, 10%, 20%, 50%, 100%.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 23

PD 6705-2:2020 PUBLISHED DOCUMENT

NOTE 2 For example, if the proportion in Table 4 is 20%, an increase of one level changes the recommendation to
50% and a decrease of one level to 10%.

If adjustments are made according to more than one condition in Table 5, the net number of levels
should be used, after adding and/or subtracting the number of levels given in each note.
NOTE 3 For example, if the proportion in Table 4 is 100% and the adjustments according to Table 5 are −1, −1
and +1 level respectively, the net adjustment would be −1 level, which would result in a proportion of 50%.

For QSC F71 and above for shop welds in steel grades up to and including S355, the minimum
recommended proportions of welded joints which should be subjected to supplementary NDT are
specified in Table 6. For other conditions, the proportion tested in Table 6 should be adjusted in
accordance with Table 5 as described for Table 4.
BS EN 1090‑2:2018, Annex L should not be used for determining the extent of supplementary testing
in place of Table 4, Table 5 and Table 6.
NOTE 4 See Annex C, C.23.

14.3.2 Extent of FPC inspection (see BS EN 1090-2:2018, 12.4.2.2 and 12.4.2.3)

The selection of EXC in BS EN 1090‑2:2018, Table 24 for routine FPC inspection should be
determined by the manufacturer according to the nature of the work in normal production. It should
not be less than that required for EXC3 or for the project specific inspection required for an individual
contract, whichever is the greater.
NOTE 1 Any FPC inspections undertaken on a specific bridge which conforms to Table 4, providing that the
acceptance criteria in 14.3.3 are fully met, may be regarded and documented as part of the specific inspection
recommended in 14.3.1.

NOTE 2 See Annex C, C.23.

14.3.3 Acceptance criteria for project specific inspection (see BS EN 1090-2:2018, 7.6)
The criteria for final acceptance without the need for repair should be in accordance with the tables
listed in a), b) and c), according to the method of inspection used:
a) visual inspection: Table 7 and Table 8;
b) magnetic particle and penetrant testing: Table 9; and
c) ultrasonic testing: Table 10.
The specific imperfection limits given in Table 7 to Table 10 provide the necessary criteria for
ensuring that the design requirements in BS EN 1993‑2, including BS EN 1993‑1‑9, should be met.
The method of specification and the acceptance criteria given in BS EN 1090‑2:2018, 7.6.2 should not
be used in place of those given in Table 7 to Table 10 for specific inspection of production welds.
NOTE See Annex C, C.24.

14.3.4 Acceptance criteria for FPC (see BS EN 1090-2:2018, 7.6)

The selection of the EXC for routine FPC should be determined by the manufacturer, according to
the nature of the work in normal production and should follow the recommendations for selection
of EXC in 9.1.
NOTE 1 The capability of proving that some of the criteria are met can depend on the method of testing.

The criteria in BS EN 1090‑2:2018, 7.6.1 might not necessarily exceed the criteria in 14.3.3. The
criteria given in 14.3.3 should apply irrespective of those used for FPC.
NOTE 2 See Annex C, C.24.

24 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

14.3.5 Inspection and testing of welded shear studs for composite steel and concrete structures
(see BS EN 1090-2:2018, 12.4.3)
Production tests should be carried out in accordance with BS EN ISO 14555 before the beginning
of welding operations and repeated after every 5 000 welds for each combination of stud diameter,
parent material and type of equipment that is used.
Simplified production tests should be carried out as required by BS EN ISO 14555, and at a minimum
rate of one test (comprising three studs) after every 100 welds on each piece of steelwork, with a
minimum of one test (comprising three studs) per piece. The direction of bend test should be such
that the bent stud should not interfere with other elements of the works.
Production surveillance in accordance with BS EN ISO 14555 should include the testing of every stud
on a production piece of steelwork by striking the side of the head of the stud with a 2 kg hammer.
The weld should be deemed to be acceptable if there is a clear ring tone due to the striking. Any
stud deemed to be unacceptable as a result of this test should be tested as part of the simplified
production testing given in this subclause (14.3.5).
All stud shear connectors should be aligned to within 4° of the normal to the plate to which they are
welded and should have the correct length after welding. The length of each stud after welding (LAW)
should be within +1,5 mm and −2 mm of the manufacturer’s specified nominal LAW.

14.3.6 Production tests (see BS EN 1090-2:2018, 12.4.4)

Production tests conforming to the requirements for EXC3/4 should be specified.
Specific production testing should be conducted on run-off coupon plates in accordance with
Table 11, and this should apply to all parts of the structure. Where the use of run-off coupon plates
is not practicable, e.g. in the case of circumferential butt welds in tubular members, the production
tests should be conducted at the same time on separate coupon plates. The coupon material should
be from the same batch with the same weld preparation and orientation relative to the rolling
direction(s) as in the production joint. The same welding equipment and welder/welding operator
should be used.

14.4 Mechanical fastening (see BS EN 1090-2:2018, 12.5)

14.4.1 Preloaded bolted connections: inspection before, during and after tightening
(see BS EN 1090-2:2018, 12.5.2.3, 12.5.2.4 and 12.5.2.6)
In BS EN 1090‑2:2018, 12.5.2.3, the tightening procedure should be checked before tightening as
required by EXC2/3/4.
In BS EN 1090‑2:2018, 12.5.2.4, inspection during and after tightening should conform to the
requirements for EXC2/3/4.
In BS EN 1090‑2:2018, 12.5.2.4c), the number of bolt assemblies inspected overall in a structure
should conform to the requirements for EXC3/4.
The part turn method should conform to the requirements of BS EN 1090‑2:2018, 12.5.2.4c)2)i).
In BS EN 1090‑2:2018, 12.5.2.4d), the sampling plan should conform to the requirements for EXC2/3.
Bolting assemblies should be checked for overtightening.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 25

PD 6705-2:2020 PUBLISHED DOCUMENT

For the part turn method, the inspection and testing requirements of BS EN 1090‑2:2018, 12.5.2.6
should apply except as follows:
a) the first step should be checked as required for EXC3/4 but using the same torque conditions as
given in Table 1 of this Published Document; and
b) for inspection after the second step, rotation angle values given for the combined method should
be replaced with:
1) 0° (in lieu of 15°); and
2) 60° (in lieu of 30°).

14.4.2 Torque method (see BS EN 1090-2:2018, 12.5.2.5)

Inspection of tightening by the torque method should conform to the requirements for EXC3/4 in
BS EN 1090‑2:2018, Table 25.

14.4.3 Combined method (see BS EN 1090-2:2018, 12.5.2.6)

For the combined method, checking of the first step should conform to the requirements for EXC3/4.

14.4.4 HRC method (see BS EN 1090-2:2018, 12.5.2.7)

For the HRC method, checking of the first tightening step should conform to the requirements
for EXC2/3/4.

14.4.5 Solid rivets (see BS EN 1090-2:2018, 12.5.3.1)

The sequential sampling plan for inspection of solid rivets for hot riveting should conform to the
requirements for EXC2/3.

14.5 Erection
14.5.1 Inspection of trial erection (see BS EN 1090-2:2018, 12.7.1)
The checks needed during trial erection vary according to the type of bridge and the method
of erection. The following checks should be carried out, where relevant, but do not represent a
comprehensive list:
a) dimensions critical to assembly to other parts of the structure;
b) overall horizontal and vertical alignment and twist;
c) evidence of correct re-establishment of alignment of units in staged trial erection;
d) temperature differentials in box girders;
e) fit-up of bolted joints and site welding preparations;
f) alignment of bolt holes; and
g) identification marks of members and their orientation, including packer plates.

14.5.2 Survey methods and accuracy (see BS EN 1090-2:2018, 12.7.3.1)

The survey of the completed structure should conform to the requirements for EXC3/4. This should
include the recording of dimensional checks at the acceptance of the structure.

26 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

 Table 4 — Minimum extent of supplementary NDT of shop welds in steel grades up to and including S355 and QSC F56

Weld type OrientationA) Thickness (t) or throat (a)B) Proportion of joints tested
mm Magnetic particle (MT) Ultrasonic testing (UT)C)
or penetrant testing (PT)
Butt Transverse t<8 100 Not applicable
8 ≤ t ≤ 20 20 50D)
PUBLISHED DOCUMENT

t > 20 50 100
Longitudinal t<8 50 Not applicable
8 ≤ t ≤ 20 5 0
t > 20 10 10
Fillet Transverse t ≤ 20 5 No requirement
t > 20, a ≤ 10 10 0F)
t > 20, 10 < a < 15 20 5F)
t > 20, a > 15 20 10F)
Longitudinal As per transverse fillet but reduced by 2 levelsE)
A)
Transverse applies to all welds orientated within 60° of the longitudinal axis of members, except for connection zones where all orientations are deemed to be transverse.
Connection zones are all locations within 200 mm of a main structural connection, loading point or support position.
B)
t = the nominal maximum parent metal thickness in the joint; and

a = the nominal fillet weld throat dimension (including any specified penetration).
C)
Testing conditions to BS EN ISO 17640 should be as follows: testing level B, evaluation level DAC – 14 dB.
D)
100% for single-sided butt where no access to root side.
E)
See 14.3.1 for explanation of change in level.
F)
The main purpose of these inspections is to provide assurance that there are no systematic embedded imperfections which are either not permitted, such as solidification or HAZ
cracking, lamellar tearing etc, or which would also contribute to a resulting loss of required cross section in the weld throat or leg (fusion boundary) zones, such as LORP, LOSWF,
laminations, etc.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

27
PD 6705-2:2020
28
Table 5 — Adjustments in proportions of supplementary NDT for conditions other than those covered by Table 4 and Table 6

Condition Change in levelA)

Site welded joints +1
B)
PD 6705-2:2020

Automatic and robotic welded joints −1

Standard proprietary products manufactured on a mass production basis B) −1
Steel grades above S355 and below S500 +1
Steel grades S500 and above +2
QSC F36 (relative to F56) −1
After a non-conformance with acceptance criteria in Table 9 or Table 10 +2 (minimum, depending on degree of severity of non-
conformance)
Applicable to all joints of similar type tested in the same production batch

To be maintained until the cause has been identified and rectified and defects eliminated
A)
See 14.3.1 for explanation of change in level.
B)
Not applicable to transverse butt welds of F90 and above.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT
 Table 6 — Minimum extent of supplementary NDT of shop welds in steel grades up to and including S355 and QSCs F71 to F140

Weld OrientationA) Thickness (t) B) or Proportion of joints tested

type throat (a)
mm
%

Magnetic particle (MT) or Ultrasonic testing (UT)C) Radiographic testing

penetrant testing (PT) (RT)D)
F71 F90 F112 and F140 F71 F90 F112 and F140 F90 F112 F140
PUBLISHED DOCUMENT

F) F)
Butt Transverse t<8 100 100 100 n/a 20 100 20 100 100E)
8 ≤ t ≤ 20 50 100 100 100 100 100F) 0 20 100E)
F)
t > 20 100 100 100 100 100 100 0 50 100E)
Longitudinal t<8 100 100 100 n/a 20F) 100F) 10 50 100
8 ≤ t ≤ 20 20 50 100 10 20 100F) 0 10 100
F)
t > 20 20 50 100 10 20 100 0 10 100
Fillet Transverse t ≤ 20 20 100 100 No requirement No requirement No requirement No requirement
t > 20, a ≤ 10 50 100 100 0 20 20
t > 20, 10 < a ≤ 15 100 100 100 20 0 50
t > 20, a > 15 100 100 100 50 100 100
Longitudinal As per longitudinal butt weld
A) B) A) B)
and as per Table 4, and .
C)
Testing conditions to BS EN ISO 17640 should be as follows:

QSC F71 and F90: testing level B, evaluation level DAC −14 dB;

QSC F112: testing level C, evaluation level DAC −17 dB;

QSC F140: testing level C, evaluation level DAC −20 dB; and

transverse indication scans required for longitudinal welds.

Where phased array technique is to be used, the principles of BS EN ISO 13588 should be applied. Testing level D should apply where the test procedures, the equipment and the
operator’s competency should be subject to independent validation for suitability for assessment of Table 10 requirements.
D)
Applies to in-line butt welds only. BS EN ISO 17636 test class B should be used. RT not required if phased array technique used for UT.
E)
Not relevant for designs based on Table 8.3 in BS EN 1993‑1‑9:2005.
F)

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

Phased array required if RT not used.

29
PD 6705-2:2020
30
Table 7 — Weld acceptance criteria for visual inspection

Main criterion Imperfection type Dimensional parameter Joint type Weld Acceptance limits according to QSC Remedial action in event of
All dimensions in mm
A) B) B), C)

type non-conformanceE)
Description BS EN ISO 6520‑1 Symbol BS EN ISO 5817 F56 F71 to F140
PD 6705-2:2020

[ ] applies to longitudinal welds only D)

ref no. ref no.

[For F36, see F)]
Overall joint Weld location error — Dr — All Fillet Dr ±10 [±10] As per F56 Refer to welding co-ordinator
geometry for remedial action
Incorrect weld type — Dr — All All NP [NP] As per F56 Refer to welding co-ordinator
for remedial action
Inadequate weld — Dr — All All Dr −0 [−5] As per F56 Add extra weld length

h
length
Linear misalignment 5071, 5072 3.1 In-line butt Butt h ≤ 0,2t [0,3t] See Table 8 Remove existing weld, realign
joint, remake preparations and
h≤4 [5]
check with MT or PT, reweld

h
to AWPS,
507 — Cruciform All h ≤ 0,4t [0,4t] See Table 8 check with UT with MT or
h≤6 [6] PT, reweld to AWPS, check

β
with UT

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

Angular 508 — In-line butt Butt β ≤ 2° [3°] See Table 8 Correct using approved
misalignment procedure, check with MT or

h
PT
G)
Root gap 617 3.2 Lap, tee, Fillet h≤2 [3] See Table 8 Refer to welding co-ordinator

h
cruciform for remedial action
Excess weld 502, 503, 504, 1.9, 1.10, 1.11, All All NL B) [NL] B) As per F56 Correct by grinding if any
dimensions 512, 5214 1.16, 1.21 functional limits are required B)
PUBLISHED DOCUMENT
 Table 7 (continued)

Main criterion Joint type Weld Remedial action in event of

All dimensions in mm
Imperfection typeA) Dimensional parameterB) Acceptance limitsB), C)according to QSC
type non-conformanceE)
Description BS EN ISO 6520‑1 Symbol BS EN ISO 5817 F56 F71 to F140
[ ] applies to longitudinal welds only D)

ref no. ref no.

[For F36, see F)]
PUBLISHED DOCUMENT

Surface notches Crack 100 to 106 — 1.1, 1.2 All All NP [NP] As per F56 Remove imperfections by
grinding to approved
Lack of fusion 401, 506 — 1.5, 1.13 All All NP [NP] As per F56 excavation shape, check with

h
PT, reweld to AWPS,
Lack of penetration 402 1.6 All Single- NP [h ≤ 0.1t] As per F56 [NP] refer to welding co-ordinator to
(unspecified) sided determine cause
[h ≤ 1]
butt
Large cavities 2015, 2016, 202, — 2.6 All Butt NP [NP] As per F56

h
510 2.7, 1.15
Undercut, under 501, 509, 511, 1.7, 1.8, 1.14, All All h ≤ 0,5 [1] See Table 8 Remove by grinding, repair to

d
fill etc. 515, 517 1.17, 1.19 AWPS
Pore 2017 1.3 All All d≤2 [2] See Table 8 Remove to depth of 3 mm,
repair to AWPS
Toe angle 5051 α 1.12 All All α ≥ 90° [90°] See Table 8 Correct with high speed rotary
burr or reweld to AWPS
Damage 601, 603 to 606 — 1.22 All All NP [NP] As per F56 Grind out to smooth profile,

h
repair by AWPS if h > 1
G) G)
Loss of cross- Insufficient weld 5213 1.20 Tee, lap, Fillet NP [h ≤ 0,1a] As per F56 Increase size using same AWPS

h
section throat, aH) cruciform [h ≤ 1] G)
Insufficient leg 521 — Tee, lap, Fillet NPG) [h ≤ 0,1z] G) As per F56
H) G)
length, z cruciform [h ≤ 2]
Multiple con- 201, 202, 501, h, l, d, lp 4.2, 4.1 All All See I) See I) As per F56 Increase cross sectional
forming surface 509, 511, 515, area using methods above,
imperfections 5213 depending on imperfection
type
Surface conditions Deposits 602, 613, 614, 615 — 1.23 All All Not permitted if surface to receive See Table 8 Remove by grinding or blast
corrosion protectionJ) cleaning

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

31
PD 6705-2:2020
32
Table 7 (continued)

Main criterion Imperfection type Dimensional parameter Joint type Weld Acceptance limits according to QSC Remedial action in event of
All dimensions in mm
A) B) B), C)

type non-conformanceE)
Description BS EN ISO 6520‑1 Symbol BS EN ISO 5817 F56 F71 to F140
PD 6705-2:2020

[ ] applies to longitudinal welds only D)

ref no. ref no.

[For F36, see F)]
A)
The three digit BS EN ISO 6520‑1 designation is deemed to include all four digit sub-categories.

B)
Symbols are as defined in BS EN ISO 5817 or BS EN ISO 6520‑1.

Dr = the dimension (or weld type) specified on the drawings. NP = not permitted. NL = no limits specified for FFP; functional limits may be applied for a particular application.

C)
Where more than one limit is given for an imperfection for a given QSC and weld orientation, all limits should apply.

D)
Longitudinal welds are those not deemed to be “transverse”; see Table 4 A).

E)
AWPS = approved welding procedure specification for repair, qualified in accordance with 9.2.2.

F)
For joints in any orientation to F36, the imperfection limits given for longitudinal welds [ ] for F56 are applicable.

G)
Where a root gap h has been observed in a tee, cruciform or corner joint, the nominal required fillet weld dimensions, as measured with a weld gauge, should be increased as follows, unless compensating
penetration beyond the root has been proven: minimum required throat size a + 0,7 h; minimum required leg length z + h (on affected leg only). See also H).

H)
Both a and z measurements should be checked, irrespective of which has been specified on the drawings. They are related as follows:

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

           Angle between fusion faces 120° 110° 100° 90° 80° 70° 60° See also G).

           z/a                                                2,0 1,74 1,56 1,41 1,31 1,22 1,15

I)
All permitted imperfections resulting in loss of cross-section should be summed and assessed as follows:

Ʃhl + Ʃ0,5dt ≤ 4,5t or [≤ 9t]. Ʃhl should include subsurface imperfections assuming h = 3; see Table 10, K). Measurement length lp = 100 mm. For fillet welds, “a” should be substituted for “t”.

J)
It should not be permitted on uncoated steels conforming to BS EN 10025-5 and BS EN 10088-1.
PUBLISHED DOCUMENT
 Table 8 — Weld acceptance criteria for visual inspection for QSCs F71 to F140 where limits differ from those for F56 in Table 7

Imperfection type Acceptance limits according to QSC

All dimensions in mm
A) B), C)
Remedial action in event of
non‑conformance

F71 F90 F112 F140

[ ] applies to longitudinal welds only D)
PUBLISHED DOCUMENT

Linear In-line h ≤ 0,15t [0,2t] h ≤ 0,1t [0,2t] h ≤ 0,05t [0,2t] h ≤ 0,05t [0,2t] As per Table 7
misalignment butt h≤4 [5] h≤3 [4] h≤3 [4] h≤3 [4]
Cruciform h ≤ 0,3t [0,4t] h ≤ 0,2t [0,4t] h ≤ 0,1t [0,3t] h ≤ 0,1t [0,2t]
h≤5 [6] h≤4 [6] h≤3 [5] h≤3 [3]
Angular misalignment β ≤ 2° [3°] β ≤ 1,5° [2°] β ≤ 1° [2°] β ≤ 0,5° [2°] As per Table 7
E)
Root gap h≤2 [2] h≤2 [2] h≤1 [1] h ≤ 0,5 [0,5] As per Table 7
Undercut, underfill h ≤ 0,3 [h ≤ 1] NP [h ≤ 1] NP [h ≤ 0,5] NP [h ≤ 0,5] As per Table 7
Porosity d ≤ 1,5 [1,5] d≤1 [1] NP [NP] NP [NP] As per Table 7
Toe angle α ≥ 110° [90°] α ≥ 150° [90°] α ≥ 165° [110°] α ≥ 175° [110°] As per Table 7
Deposits Not permitted if surface NP [NP] NP [NP] NP [NP] See F)
to receive corrosion
protection I)
Variation in root or cap Δh ≤ 3 G) [Δh ≤ 3] Δh ≤ 2 G) [Δh ≤ 2] Δh ≤ 1 G) [Δh ≤ 1] Δh ≤ 0,5 G) [Δh ≤ 0,5] See H)
longitudinal profile B)
A) A)
Imperfection types as designated in Table 7 .
B)
Symbols are as defined by BS EN ISO 5817 or BS EN ISO 6520‑1.

NP = not permitted. Δh = the maximum variation in cap or root profile measured along weld axis over any length of 3 mm.
C)
Where more than one limit is given for an imperfection for a given QSC and weld orientation, all limits should apply.
D)
Longitudinal welds are those not deemed to be “transverse”; see Table 4 A).
E)
Correct fillet weld sizes as per Table 7 G).
F)
For F112 and above, remove with high speed rotary burr.
G)
Transverse welds outside connection zones are exempt from this restriction; see Table 4 A).
H)
Correct by tapering slope to shallower angle or repair to AWPS.
I)

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

It should not be permitted on uncoated steels to BS EN 10025-5 and BS EN 10088-1.

33
PD 6705-2:2020
34
Table 9 — Weld acceptance criteria for magnetic particle and penetrant testing

Imperfection type Acceptance Indication Acceptance limits according to QSCA) Remedial actions in event of non-
standards pattern conformanceC)

F56 F71 F90 to F140

[ ] applies to longitudinal welds with longitudinal linear indications only B)
PD 6705-2:2020

D)
[For F36, see ]
Surface notches BS EN ISO 23278 and Isolated Level 2 [Level 3] Level 1 [Level 2] NP [NP] Remove by high speed burr machining
identified in Table 7 BS EN ISO 23277 with machining marks in longitudinal
and Table 8 direction. Repair to AWPS if
Grouped E) NP [NP] NP [NP] NP [NP] non‑conformance with Table 7 and Table 8
depth requirements (h).
A)
NP = not permitted.
B)
Longitudinal welds are those not deemed to be “transverse”; see Table 4 A).
C)
AWPS = approved welding procedure specification for repair, qualified in accordance with 9.2.2.
D)
For joints in any orientation to F36, the imperfection limits given for longitudinal welds [ ] for F56 are applicable.
E)
“Grouped” should be deemed to apply to any indications separated by less than 2,5 mm.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT
 Table 10 — Weld acceptance criteria for ultrasonic testing with limited optional radiographic testing
Imperfection typeA) Dimensional Location Acceptance limits according to QSCB), C) Remedial action
parameterB) in cross- in event of non-
section or conformanceE)
[ ] applies to longitudinal welds only D)

Description BS EN ISO 6520‑1 Symbol BS EN ISO 5817 throat F56 F71 F90 F112 F140
ref no. ref no.
[For F36, see F)]
PUBLISHED DOCUMENT

Cracks 100 to 106 — 2.1, 2.2 Full depth NPH) [NP] H) NPH) [NP] H) NPH) [NP] H) NPI) [NP] I) NPJ) [NP] J) Refer to welding

h, l
co-ordinator to
G) H) I) I) J) J)
Pores , 2011, 2014, 2.3, 2.5, 2.6 to Within h≤3 [3] h≤3 [3] NP [h ≤ 3] NP [NP] NP [NP] determine
G)
inclusions , 2015, 2016, 202, 2.13 6 mm l ≤ 10 [20] l≤5 [10] NP H) [l ≤ 5] NP I) [NP] I) NPJ) [NP] J) cause. Remove
cavities G), lack 203, 301 to 304, of any by grinding
of fusion, lack of 401, 402 surface to approved
penetration excavation shape.
I)
Deeper h≤3 [3] h≤3 [3] h≤3 [3] NP H) [h ≤ 3] NP [NP] I) Check with MT
than 6 mm l ≤ 1,5t [3t] l ≤ 10 [1,5t] l≤5 [20] NP H) [l ≤ 5] NP I) [NP] I) or PT. Reweld to
from any AWPS.
surface

H, L
h, l, lp, 4.2 Full depth Σl ≤ 1,5t K) [3t] K) Σl ≤ 1,5tK) [3t] K) Σl ≤ 1,5tK) [3t] K) NP [Σl ≤ 1,5t] K) NP [NP]
lp = 100 [100] lp = 100 [100] lp = 100 [100] NP [lp = 100] NP [NP]
H≥6 [6] H≥6 [6] H≥6 [6] NP [H ≥ 6] NP [NP]
L ≥ 10 [10] L ≥ 10 [10] L ≥ 10 [10] NP [L ≥ 10] NP [NP]
L)
Uniformly 2012, 2013 — 2.3, 2.4 Full depth Not permitted if obstructs detection or evaluation of imperfections above NP [NP] L)
distributed or
clustered porosity

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

35
PD 6705-2:2020
36
Table 10 (continued)
Imperfection typeA) Dimensional Location Acceptance limits according to QSCB), C) Remedial action
B)
parameter in cross- in event of non-
section or conformanceE)
[ ] applies to longitudinal welds only D)

Description BS EN ISO 6520‑1 Symbol BS EN ISO 5817 throat F56 F71 F90 F112 F140
PD 6705-2:2020

ref no. ref no.

[For F36, see F)]
A)
Imperfection types as designated in Table 7 A).

B)
Symbols are as defined by BS EN ISO 5817.

NP = not permitted. H = the clear gap between adjacent imperfections measured in through thickness direction. L = the gap between ends of adjacent imperfections measured along weld axis.

C)
Where more than one limit is given for an imperfection for a given QSC and weld orientation, all limits should apply. If surface breaking imperfections are detected by UT, the criteria given in Table 7,
Table 8 and Table 9 apply.

D)
Longitudinal welds are those not deemed to be “transverse”; see Table 4 A).

E)
AWPS = approved welding procedure specification for repair, qualified in accordance with 9.2.2.

F)
For joints in any orientation to F36, the imperfection limits given for longitudinal welds [ ] for F56 are applicable.

G)
Radiographic testing may be used to assist in interpretation of these imperfections.

H)
Rejection level DAC −14dB.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

I)
Rejection level DAC −17dB.

J)
Rejection level DAC −20dB.

K)
If permitted surface imperfections resulting in loss of cross-section also exist, the criteria in Table 7 I) also apply.

L)
Also, not permitted when checked by radiographic testing.
PUBLISHED DOCUMENT
PUBLISHED DOCUMENT PD 6705-2:2020

Table 11 — Production tests on run off coupon plates

Weld type QSCs Material grades Test type Testing rate

In-line transverse butt weld F56 and above S355 to S460 ML, NL Charpys 1 in 10
in tension S460 to S690 Q Tensile 1 in 5
S420 to S690 QL, QL1 Tensile and Charpys 1 in 5
F90 and above All UT + 3 macros A)
1 in 5
Longitudinal butt or fillet F90 All UT + 3 macros B) 1 in 20
F112 and above All UT + 3 macros B)
1 in 5
A)
These tests are additional to those described in rows above.
B)
A tack weld should be included in the coupon plate if used in production and at least one macro should be taken at each
end of the tack weld.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 37

PD 6705-2:2020 PUBLISHED DOCUMENT

Annex A (informative)
Background to the development of European and
International execution standards and their relationship to
previous British execution standards

A.1 The European Committee for Standardization (CEN)

BS EN 1993‑1‑1, Eurocode 3: Design of steel structures – Part 1-1: General rules and rules for
buildings has been prepared by CEN Technical Committee 250, subcommittee 3 (CEN/TC250/SC3).
BS EN 1993‑1‑1 states that the design rules are dependent on the steelwork being fabricated and
erected in accordance with BS EN 1090, which was anticipated to contain a specification for the
execution of steel structures.
The responsibility for drafting the execution specification was given to a dedicated CEN Technical
Committee (CEN/TC135).
The general design rules for structural steelwork were published by CEN as a Prestandard (ENV)
which was published by BSI as a draft for development in 1992 (DD ENV 1993‑1‑1). The design
rules were subjected to a two-year trial period by the Member States, followed by submission of
comments which were taken into account by the project teams in their preparation of the final parts
of BS EN 1993.
ENV 1090 was published in six parts. The general specification for structural steelwork including
buildings (Part 1) was published by CEN in 1996. In 1998, BSI published it together with a UK
National Application Document (NAD) as DD ENV 1090‑1, to enable designers to use it in conjunction
with other related British Standards. ENV 1090‑2 to ENV 1090‑6, which provided supplementary
rules for cold formed thin gauge members and sheeting, high strength steels, hollow section
structures, bridges and stainless steels, respectively, were produced by CEN over a four-year
period. These later parts only provided modifications and additions to the general specification in
ENV 1090‑1. Parts 1, 3, 4 and 6 were published by BSI as DDs.
BSI Technical Committee B/525/10 produced a comprehensive NAD, CEN ENV 1090‑5: Execution of
steel structures – Part 5: Supplementary rules for bridges, which was available for industry, but not
formally published by BSI as a DD.
Following the ENV period, CEN/TC135 decided to merge the six separate parts of ENV 1090 into
one new part (EN 1090‑2) covering all structural steel application areas, to be drafted by one
working group. Another working group was set up to draft EN 1090‑3, an equivalent specification for
aluminium structures. A third working group was set up to prepare a new Part 1, covering conformity
assessment for both steel and aluminium.
One of the main conclusions from the ENV comments was that there were wide variations in
steelwork specification practice amongst the Member States and across different steelwork
applications. Therefore, a degree of flexibility needed to be built into the standard if it was to be
acceptable to most Member States. CEN/TC135 introduced the concept of an execution class for this
purpose, with four levels which could be selected by the specifier. It was decided that the choice of
execution class should be related to the level of reliability required by the designer for the proposed
design. The selection of level of structural reliability was a design matter and some guidance had
already been provided by CEN/TC250 in EN 1990:2002+A1:2005, Annex B, Eurocode: Basis of
structural design, which sets out the fundamental principles and terminology on which the other

38 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Eurocodes are based. A new term “service category (SC)” was also introduced into BS EN 1090‑2,
which depended on whether or not the design of the structure (or component) was considered to be
susceptible to fatigue loading. No method for quantifying which SC applied in any structural situation
was given. The execution class was intended to be dependent on the SC.
Therefore, BS EN 1090‑2 could only be used if the reliability level (related to CC), the SC and its
relationship to the execution class was known. BS EN 1990:2002+A1:2005, Annex B establishes
principles and recommends a method of adjusting the reliability level based on varying the load
factor and the degree of independence of the design checker and inspection supervisor. However, it
does not cover the concepts of execution class or SC.
The final draft of EN 1090‑2, prEN 1090‑2, was put out by CEN for public consultation in 2005. By
this time the general steel design rules, EN 1993‑1‑1, had already been out to public consultation,
amended, finally approved by CEN/TC250 and published by CEN. A large number of comments were
received on prEN 1090‑2. These included particular concerns from the UK about the technical basis
of a proposed relationship, given in an informative annex, between the reliability level (expressed
as CC), the SC and the execution class. The UK put forward detailed proposals for improving the
technical basis by quantifying the SC and relating the variable technical requirements directly to that
parameter. This was based on principles already used in BS 5400 and in the International standard
for steel structures, ISO 10721 (see A.2).
At final vote, the UK voted negatively to the document, together with two other Member States. The
UK’s main concern was that, even if the informative recommendations for determining execution
class were followed, there would be situations where either the design resistance values assumed in
BS EN 1993 would be invalid, or some of the requirements would result in an economic penalty and
represent a potential barrier to trade.
In 2011, a minor amendment of EN 1090‑2 was issued by CEN. This was followed by a proposed
revision which was issued by CEN for Enquiry in 2015. This was rejected by CEN Members as not
being ready for formal vote. A further proposal was issued by CEN for another Enquiry in 2017,
which was approved and subsequently voted for publication in 2017. A significant change was the
inclusion of important new or updated reference standards. Another change was the removal of the
previous Annex B giving guidance on selection of execution classes. Guidance on this subject was then
introduced in a more simplified form in BS EN 1993‑1‑1:2005+A1:2014, Annex C. However, the main
UK concerns about incompatibilities with BS EN 1993 remained, which resulted in an abstention vote
by the UK in 2017.
In view of the specialist technical issues involved, this Published Document offers expert guidance
to potential specifiers of bridge steelwork. These take account of the target safety levels assumed in
BS EN 1993, including the values of the partial resistance factors and other nationally determined
parameters in its National Annexes.

A.2 International Organization for Standardization (ISO)

During the 25-year development period for the Eurocodes, ISO Technical Committee 167 drafted an
international standard for steel structures, ISO 10721.
Two subcommittees were set up to do the drafting. Subcommittee 1 was responsible for ISO 10721‑1,
Steel structures – Part 1: Materials and design and Subcommittee 2 for ISO 10721‑2, Steel structures
– Part 2: Fabrication and erection. ISO/TC167 was well supported by members from Europe, North
America, Asia and Africa in particular. Much of the development work produced by the Eurocode 3
project teams was taken into account during the drafting of ISO 10721‑1. The two subcommittee
and the plenary meetings were held at the same venues and times, so that there was a consistent
relationship between the design rules and the execution specification. The UK was a major

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 39

PD 6705-2:2020 PUBLISHED DOCUMENT

contributor throughout the work of the two subcommittees and provided experts in design of steel
buildings, bridges, towers and masts. BSI also held the secretariat for Subcommittee 2.
ISO 10721‑1 was published by ISO in 1997 and ISO 10721‑2 was published in 1999. One of the
important features of the two standards was the linking of the execution requirements for welding
in ISO 10721‑2 to the static and fatigue design rules in ISO 10721‑1. This was done by using a
parameter similar to the QSC. This approach was consistent with the method used to link the
execution requirements in BS 5400‑6 to the design rules in Part 3 and Part 10 of that standard. A
similar system was adopted in BS 8118 and in PD 6705‑3.
International standards are primarily aimed at assisting countries or regions which do not have their
own standards. It is CEN policy to use ISO standards as a basis for European standards whenever
possible and, in many cases, they are adopted without amendment. BSI did not adopt ISO 10721 as
a British Standard as the UK has its own equivalent standards. However, ISO 10721‑2 has been used
as a reference standard in BS 5950 for structural use of steelwork in buildings which are likely to be
subjected to significant fatigue loading. It was also used in the drafting of the UK NAD for ENV 1090‑1
and ENV 1090‑5. However, it was not used in the development of EN 1090‑2.
ISO/TC 167 reviewed its standards for structural steelwork in 2017. Both ISO 10721‑1:1997 and
ISO 10721‑2:1999 were confirmed and remain current.
The recommendations in this Published Document have taken account of the principles adopted
by ISO 10721.

A.3 Comparison of main features of BS EN 1090-2:2018 and BS 5400-6

BS 5400‑6 was superseded by BS EN 1090‑2 in 2008 and withdrawn in 2010, but may still be in use
in some specifications and is in use in connection with assessment of existing bridges. Some of the
most significant differences between the two standards as they relate to bridges are detailed below.
a) Scope of application: BS EN 1090‑2:2018 applies to all types of structural steelwork, whereas
BS 5400‑6 was specific to bridges.
b) Reference standards: BS EN 1090‑2:2018 refers normatively to more than 200 new European
standards or parts of standards, many of which have been published in the last 20 years. Many
replace earlier British Standards, some of which were still referenced in BS 5400‑6.
c) Material grades: The range of steel grades for plates and sections has been increased from
S460 in BS 5400 to S690 (or S960 with special provisions) in BS EN 1090‑2:2018. BS 5400‑6
only covered hot rolled structural hollow sections, whereas BS EN 1090‑2:2018 covered
both hot and cold rolled structural hollow sections. BS 5400‑6 covers cast iron whereas
BS EN 1090‑2:2018 does not.
d) Product forms: BS EN 1090‑2:2018 covers high strength cables whereas BS 5400‑6 did not.
e) Execution methods: BS EN 1090‑2:2018 allows laser and plasma cut holes with conditions.
These processes were not covered by BS 5400‑6.
f) Weld procedure qualification: BS 5400‑6 required weld procedures to be qualified by testing.
BS EN 1090‑2:2018 allows other methods to be used to qualify procedures.
g) Personnel qualification: BS EN 1090‑2:2018 has qualification requirements for welding co-
ordinators, welders and weld inspectors. BS 5400‑6 did not address this.
h) Quality management: BS EN 1090‑2:2018 addresses the subject of quality documentation,
including quality plans, in a more explicit manner. BS 5400‑6 did not address this, although
recording of testing was covered in some of the referenced standards.

40 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

i) Quality control: BS EN 1090‑2:2018 is based on FPC which is on-going, and testing is not
specific to any particular contract or structure, unless project specific requirements are added.
BS 5400‑6 did not address FPC and the specified scope of testing was contract-specific.
j) Acceptance levels: As a result of the FPC basis for quality control [see i)], BS EN 1090‑2:2018
target quality levels need to be high enough to cover the most quality-sensitive design situations
allowed by BS EN 1993. The BS EN 1090‑2:2018 weld acceptance levels are based on arbitrary
criteria, some of which are not necessary for structural integrity. BS 5400‑6 quality levels were
fitness-for-purpose (FFP) based and provided the minimum which was structurally acceptable
for each contract.
k) Weld testing: BS 5400‑6 defined the scope for each method of testing to be used for different
elements of the bridge and the acceptance criteria were specified according to the method of
testing used. BS EN 1090‑2:2018 does not specify these matters. Coupon testing of selected butt
welds was required by BS 5400‑6, but is optional in BS EN 1090‑2:2018.
l) Non-conformances: The rate of non-conformance of welds is likely to be higher with the FPC
acceptance levels. BS EN 1090‑2:2008+A1:2011 stated that this is not necessarily a cause for
rejection provided the imperfection can be accepted on an FFP basis, which is not defined
(BS EN 1090‑2:2018 is now not so clear on this point). BS 5400‑6 levels, being FFP based,
normally result in a lower rejection rate, but where non-conformances occur, repair was
normally required without further consideration.
m) Preloaded bolting: A wider range of bolt types and tightening methods is given in
BS EN 1090‑2:2018 than in BS 5400‑6 which referenced BS 4395 and BS 4604 (now both
withdrawn). Tension control bolts (HRC bolts) and load indicating washers are covered in
BS EN 1090‑2:2018.
n) Geometrical tolerances: BS EN 1090‑2:2018 has more comprehensive requirements for checking
geometrical tolerances than BS 5400‑6. BS EN 1090‑2:2018 also includes tolerances needed for
fit-up and other purposes which are not directly related to the resistance of components.
o) Surface treatment: BS EN 1090‑2:2018 covers the application of surface coatings for corrosion
protection, which was not covered in BS 5400‑6.

Annex B (informative)
Method of determining QSC and guidance for use in
drafting specifications

B.1 General principles

B.1.1 Objective
The objective of the QSC system is to ensure that imperfection qualities and methods needed
for assurance of their quality are correctly specified, taking into account the full range of design
conditions permitted in BS EN 1993, and that quality control and inspection resources are deployed
most efficiently.
In order to achieve this objective, the issue of QSC should be considered during the design process.
By restricting the stress levels in some of the most highly cyclically stressed components, it might

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 41

PD 6705-2:2020 PUBLISHED DOCUMENT

be possible to achieve economies in execution by limiting the maximum required QSC throughout
the structure.

B.1.2 Technical issues

The prime issue concerns the prevention of fracture of structural elements subjected to tensile
stresses in service due to the presence of unspecified manufacturing imperfections, particularly
those involving loss of cross section or introduction of notches. The NA to BS EN 1993‑1‑8 and
NA to BS EN 1993‑1‑10 make some allowance for such imperfections in their recommended values
of ϒM, such that the risks of ductile and/or brittle failure respectively provide the reliability level
recommended in BS EN 1990.
However, any variable loading in service results in cyclic stresses which can increase the size of
an imperfection by stable crack growth. The rate of crack growth depends on the stress range and
frequency. The imperfection size should not exceed the “safe” size allowed for by ϒM in the design
rules during the design life.
Fracture mechanics principles applied to steel dictate that the rate of growth of an imperfection
per cycle is proportional to between the third and fifth power of the cyclic stress range. The size of
a manufacturing imperfection should be sufficiently small that it does not grow above the safe size
during the design life. As a consequence, the size of a manufacturing imperfection should typically be
inversely proportional to approximately the fourth power of the stress range for a given number of
cycles in the design life. The quality of manufacture and the controls needed to assure the quality are
therefore highly dependent on the degree of stressing expected at the imperfection concerned.

B.1.3 Description of system

The QSC system involves rating parts of the bridge in terms of their service stressing requirements.
This is designated by an “F” number, where the higher the F number is, the higher the stressing
demands in service are. There are six levels, as follows:
F36: This can only apply to parts of a bridge where low levels of static and cyclic stressing can exist.
Its use is therefore limited in practice, see B.3.3.
F56: This is expected to apply to most parts of a bridge. It assumes full static stressing with a
limited degree of cyclic stressing. The quality levels should be readily achievable during execution
and require a normal level of testing using straightforward procedures to provide assurance of
conformance.
F71 and F90: These levels extend the scope of F56 to cover specific parts of certain types of bridge
where the level of cyclic stressing can become increasingly significant. The reduced size of initial
imperfection limits permitted means that the risk of non-conformances is higher than for F56 and
the extent of testing required is consequently increased. In some cases, restrictions might need to be
imposed on the use of certain types of execution process.
F112 and F140: These levels only apply in special circumstances in which the cyclic stressing level
exceeds that covered by F90. Details where such cyclic stress levels are permitted by the fatigue
design rules are limited. The imperfection limits are so small that there is a high risk of non-
conformances and, as a consequence, often expensive and time consuming repairs are necessary.
Furthermore, the capability of normal testing procedures to evaluate such imperfections becomes
increasingly less reliable.
It is important that the specifier selects the correct QSC level for each part of the bridge. If certain
parts which require a higher level than F56 are not identified, then those parts will be at higher
risk of not achieving the specified design life. Conversely, if large areas of the bridge are identified

42 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

as requiring a higher QSC than is actually needed, then the extent of testing and possible correction
work could be significantly increased without justification.
The scope of use of the QSC system covers the quality and/or testing requirements of a number of
execution processes, see 5.5.
The following clauses provide essential advice for applying the system:
• B.2.1 Definition of QSC levels;
• B.2.2 Calculation procedure for determining required QSC level; and
• B.3 Use in drafting execution specifications.

B.1.4 Previous and existing uses of QSC system

A similar system to the QSC system was used in BS 8118‑2:1991 (structural aluminium), in
ISO 10721‑2:1999 (which was reviewed and confirmed in 2017) (structural steel) and in
BS 5400‑6:1999 (steel bridges). In these standards, the designation “minimum class (requirement)”
was used instead of “QSC”.
In BS EN 1090‑2:2008, Annex B the term “service category” was used to distinguish between “static”
and “fatigue” service conditions, but without any quantitative differentiation. In PD 6705‑2:2010,
the principles of the previous systems were used, together with the new designation term
“quantified service category (QSC)”. The same principles and term are used in PD 6705‑3:2009
(structural aluminium).

B.2 Determination of QSC

B.2.1 Definition of QSC levels
The QSC is dependent on the orientation and intensity of the static and cyclic stressing at a cross-
section of a member or in a joint. The QSC is defined in Table B.1 in terms of the static utilization
factor k and the minimum required fatigue strength ΔσCmr.
Table B.1 — Definition of QSC levels

Level of service QSC Static utilization factor kA) Minimum required value of
stresses reference fatigue strength ΔσCmr
Static Cyclic N/mm2
Reduced Very low F36 Tension k ≤ 0,6 Δσ Cmr ≤ 36
Shear k ≤ 0,8
Compression k ≤ 0,8
Full Low F56 — k ≤ 1,0 Δσ Cmr ≤ 56
Full Significant F71 — k ≤ 1,0 56 < Δσ Cmr ≤ 71
F90 — k ≤ 1,0 71 < Δσ Cmr ≤ 90
F112 — k ≤ 1,0 90 < Δσ Cmr ≤ 112
F140 — k ≤ 1,0 112 < Δσ Cmr ≤ 140
A)
The static utilization factor, k, is the ratio of the ultimate limit state (ULS) design action on the member
cross-section or joint divided by its design resistance. Where the resistance depends on an interaction
formula for more than one stress mode, a pro-rata value of k should be used as the limit. This should
apply to all static failure modes at the joint or member in question.

ΔσC is the reference value, in units of N/mm2, which is used to denote the Δσ-N-curve
where the fatigue strength (stress range) at 2×106 cycles has the same value in N/mm2
(see BS EN 1993‑1‑9:2005, 1.3 and 1.4). The same numerical value is used to designate the detail
category (DC), used for design purposes for specifying the maximum permitted Δσ‑N‑curve for a

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 43

PD 6705-2:2020 PUBLISHED DOCUMENT

particular geometry of detail, method of manufacture and stress orientation. The detail categories in
BS EN 1993‑1‑9, Clause 7 vary from DC 36 to DC 160. The severity of stress raising features in each
detail, which are specified by the designer, reduces with increasing DC number. The parameter ΔσCmr
is defined in the same terms as for ΔσC above. However, the suffix mr is used to denote the minimum
required Δσ-N-curve which would still satisfy the fatigue limit state criteria in BS EN 1993‑1‑9 for the
particular cyclic stressing conditions at the location being assessed. This minimum required curve is
denoted by the F number, as opposed to the DC number (which is often higher). Guidance on use of
ΔσCmr is given in B.2.2.
In Table B.1 the QSC F numbers are used to denote convenient ranges of ΔσCmr values, the F number
having the same number as the ΔσCmr number at the top of each range. Fracture mechanics has been
used to determine practical limits for unspecified manufacturing imperfections which are compatible
with the stressing conditions defined by the highest ΔσCmr curve in each F category.
NOTE While the “utilization factor” k may be used as a convenient parameter for defining imperfection limits for
the static limit states, it is not appropriate for the fatigue limit state. See Annex C, C.23, Note.

B.2.2 Calculation procedure for determining required QSC level

The procedure for determining the appropriate QSC level(s) makes use of the data already
obtained in the course of the fatigue assessment of the bridge. This normal design process can be
summarized as follows.
a) Evaluation of the applied stress ranges throughout the bridge in accordance with the
requirements of BS EN 1993‑2 and its UK National Annex. These are generally expressed as
the equivalent constant amplitude stress ranges ΔσE,2. for 2 million cycles. In some cases, the
applied stress ranges can consist of a spectrum of stress ranges and cycle numbers. Nominal
stress ranges apply, except in a case where a stress concentration factor kf is required, when the
modified nominal stress ranges apply.
b) Evaluation of the maximum permitted fatigue strengths ΔσC at potential fatigue initiation sites,
using the detail category tables in BS EN 1993‑1‑9, starting with the highest stressed regions
first. In some connections more than one stress direction might need to be considered.
c) Use of the results of a) and b), to apply the verification criteria in BS EN 1993‑1‑9, as amended by
BS EN 1993‑2 and its UK NA, taking into account the UK recommended partial safety and damage
equivalent factors.
d) In the event of a non-conformance being found in c), the detail category may be improved or the
applied stress range reduced until conformance is achieved.
Once the design process has reached stage d), the QSC level(s) can be determined using the
following procedure.
1) For bridges where there is no specific fatigue loading requirement, a QSC of F56 may be deemed
to apply to all parts.
2) For bridges other than in 1), all details which have been categorized according to b) as having a
DC of 56 or less may be deemed to require a QSC of F56.
3) All parts of the bridge other than those that are deemed to require a QSC of F56 according to 2),
which are assessed in accordance with the method in c), but using a fatigue strength (given in N/
mm2) of ΔσCmr = 56, and are found to conform, may be deemed to require a QSC of F56.

44 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

4) If any parts of the bridge do not conform to 3), but are found to conform when using a fatigue
strength (given in N/mm2) of ΔσCmr = 71, those parts may be deemed to require a QSC of F71.
5) In the event of any non-conformances with 4), the same process is repeated using successively
higher values of ΔσCmr as per Table B.1 until all parts of the bridge conform and the full scope of
QSCs is established.

B.3 Use in drafting execution specifications

B.3.1 General principles
The QSC requirements for every part of the bridge should be specified and this information should be
made available in the clearest and most convenient form to the parties responsible for the execution
processes. Experience has shown that many types of bridge do not normally require a QSC above F56.
In other types of bridge, higher QSC levels are only likely to be required in specific zones, the rest still
only requiring F56. Provided that the higher QSC zones are clearly identified, the remaining zones
may then be deemed to be F56 without being individually identified. The recommended procedure
for doing this is given in B.3.2.
Table B.2 gives some general guidance on where the highest QSC levels might be found in different
parts of different types of bridge. Some of the higher QSCs might only apply to very local zones of a
member or component. In some designs the QSC might be restricted due to the difficulty of avoiding
the use of a low DC feature in a member.
Table B.2 is consistent with three general principles. The first is that, the higher the proportion of live
loading is relative to the self-weight loading, the higher the QSC is likely to be. The second is that, the
higher the fatigue load is relative to the ULS live load, the higher the QSC is likely to be.
NOTE For example, railway bridges are likely to have higher QSCs than highway bridges.

The third principle is that the higher the steel grades used, the higher the QSC is likely to be. This is
due to the stress levels being higher while the fatigue strength and crack growth properties, which
are independent of grade, remain the same.
Table B.2 should not be used as the basis for defining the QSC requirements for any particular design,
which should always be done using the procedure in B.2.2.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 45

PD 6705-2:2020 PUBLISHED DOCUMENT

Table B.2 — General guide to where highest QSCs might be found in common types of bridge

Highest QSC ApplicationA)

Pedestrian Highway Railway
F56 Most ordinary types • Supporting towers and Main girders on medium to
columns long span bridges with heavy
• Main girders on medium to ballasted or concrete decks.
long spans
• Composite bridges with
multiple girders
F71 Special cases where • Cross girders Main girders on short span
very susceptible • Short span heavy ballasted or deck
to crowd or wind non-composite bridges bridges.
induced oscillation
F90 Steel decks with thin road Main girders, cross girders and
surfacing steel decks on short span light
weight steel bridges.
F112 and F140 Special features only Certain designs where F90
might not be sufficient.
A)
This table gives indicative guidance only, assuming steel grades up to S355. For higher grades, the
categories are likely to increase proportionately to the steel grade. The actual maximum QSCs should be
checked for each specified design.

B.3.2 Recommended procedure for specifying required QSCs

The QSC level(s) should be specified for all parts of a bridge on the construction drawings. The
recommended procedure to be applied to each drawing is as follows.
a) Where F56 applies to all parts on a drawing, add the following general note in a conspicuous
position on that drawing stating: “QSC F56 applies to all parts”. (Where F56 applies to all parts
of a bridge, the general note may be added instead to the General Notes drawing for that series
of drawings).
b) Where there are some parts on the drawing which require QSC(s) above F56, add a general note
in a conspicuous position stating: “QSC F56 applies to all parts except where a higher QSC is
indicated”. The zones where the higher QSCs are required should be identified using the method
of indicating the QSC level and direction as shown in Figure B.1.
The use of F36 for a general category should not be used except where considerations of stiffness/
deflection as opposed to strength are an overriding factor throughout the design (see also B.3.3).

B.3.3 Use during execution

In the event that there is a non-conformance with a F56 weld acceptance criterion in Table 7,
Table 9 or Table 10, in a part of the bridge which is not highly stressed, an option is to reassess the
imperfection according to the criterion for F36, should the stress levels at the imperfection be found
to meet the F36 requirements of Table B.1. If successful, this can be an alternative to carrying out a
repair if the latter is difficult or time-consuming.

46 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Figure B.1 — Method of designation of QSCs for specific zones on drawings

NOTE 1 The arrow symbol indicates the QSC (F number) and the direction in which the cyclic stresses are
applied. This information defines the quality requirement for the adjacent details.

NOTE 2 There could be applications where more than one QSC is required for an individual joint, depending
on the directions of stressing. In this case the QSC is indicated for each direction.

Annex C (informative)
Explanatory notes on selected recommendations

C.1 Thickness tolerances

Thickness tolerance class can have two main effects. Too negative a tolerance can result in a reduction
in both static and fatigue resistance due to increase in stresses through loss of cross section. The most
severe effect is on the buckling resistance of semi-compact plate elements. Too positive a tolerance
can result in a higher steel dead weight. Both these effects are most pronounced for thinner plates.
Class A negative tolerances are adequately covered by that part of the UK ϒM value which is needed
for geometrical deviations, (the other part being for deviations in mechanical properties). Class A
positive tolerances are adequately covered by the UK ϒG,j value.
Tolerance Class D might not be adequately covered by the UK ϒM value for the lower thickness.
Tolerance Class B and Class C might not be adequately covered by the UK ϒG,j value for the
upper thickness.

C.2 Surface conditions

Weld repairs on steel products are a potential source of notch-like flaws, which can have an adverse
effect on life when cyclic stresses are high.
Welding of steels in accordance with BS EN 10025‑6 needs to be subject to special controls to avoid
impairment of mechanical properties.
There is also a higher risk of introducing hydrogen cracks through local weld repairs.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 47

PD 6705-2:2020 PUBLISHED DOCUMENT

C.3 Internal discontinuities

Where cyclic stresses are high in the through-thickness direction, there is an increased risk of
initiating a fatigue crack from a planar internal discontinuity normal to the direction of stressing.
BS EN 1993‑1‑9:2005 provides detail categories up to DC 80 for normally welded joints of this type
when the transverse material is thin. Thin material is generally no less susceptible to lamellar flaws
than thick material.
Such joints could be subjected to high tensile weld shrinkage stresses which can dominate over any
applied compressive stresses.
In the region of load bearing stiffeners/diaphragms, high local stresses could be present, which could
reduce the load carrying capability if laminations are present. Bearing stiffeners on one side of a web
are potentially more at risk as the full shear force in the web has to pass through one surface only if
there is a mid-plane discontinuity.

C.4 Z Testing
Steels from modern mills are likely to have adequate through-thickness properties due to the low
level of impurities (particularly sulphur), compared to the limits permitted by the product standards.
Constructors normally know when it is necessary to request through-thickness testing or restrictions
on impurities from a particular supply source. Guidance to assist in such decisions is given in
BS EN 1011‑2:2001, Annex F and PD 6695‑1‑10.

C.5 Steel castings

There are two main differences between cast steel products and wrought steel products, which
need to be taken into account in design. The first is that the moulds for cast products are usually
more complex in shape. This can lead to problems arising from restricted flow paths leading to
premature solidification, differential shrinkage, segregation, cracking, porosity and residual stress. In
design, differences in thickness should be kept to a minimum and sharp corners should be replaced
with generous radii wherever possible. For this reason, it is preferable to use product samples for
destructive testing, as opposed to extension pieces or separate cast items.
The second difference is that cast products are not subject to subsequent mechanical working as,
by definition, occurs with all wrought products, which are mostly subjected to multiple rolling
processes. The benefits of rolling are firstly that the grain structure is improved, giving better
mechanical properties, particularly in the most common direction of primary stress. In the absence
of working, some casting designations might require heat treatment for achievement of the required
mechanical properties. Secondly, the action of rolling closes up many internal discontinuities in the
original ingots for wrought products and aligns them in a direction least injurious to the mechanical
properties in the main direction of stress. For this reason, inspection for defects in castings is
particularly important, especially if they are subjected to significant cyclic stresses in service.
Apart from proprietary products, steel castings can sometimes be useful in replacing complex welded
connections where local stress concentrations can result in low fatigue strengths, particularly where
a large number of identical connections occur in a bridge. In this case, attention to detail design
can enhance the fatigue strength by reducing the stress concentrations at the member junction and
relocating the welds to more accessible areas where the geometry is more uniform. The acceptance
levels for casting flaws should be specified in accordance with Table 3. This should include the entire
volume of the casting, particularly in the regions of the remaining stress concentrations.

48 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

C.6 Welding consumables for weathering steels

Avoiding unplanned loss of throat through the corrosion of unprotected welds is essential to their
long term design resistance. The recommended consumable options in BS EN 1090‑2:2018, Table 6
have been proven to give comparable corrosion resistance in weld metal to that in parent weathering
steels. Matching filler wires can sometimes give rise to a higher risk of weld metal cracking due to the
higher weld metal copper content.
Designers commonly specify grinding butt weld caps flush after welding to reduce the risk of ponding
of water on exposed surfaces. When C-Mn consumables are used in the main body of multipass welds,
the weld metal should be left at least 3 mm below the surface planes, both on the sides and the ends
of the weld, prior to depositing the weather resistant capping runs.

C.7 Preloaded bolting assemblies

The risk of thread stripping, both during tightening and under service loading, is reduced by using
the recommended products. This risk is higher with thinner nuts and when nuts have to be tapped
oversize where hot dipped galvanized assemblies are used, which is a very common requirement
in UK bridges. HV nuts are thinner than HR nuts. In the event that a stripping problem occurs on
property class 8.8 bolts, the use of property class 10 nuts might overcome this.
A passivation process might be needed after hot dipped galvanizing to enable site paint coats to
adhere adequately to the exposed parts of the bolt assembly. The lubrication during manufacture can
be applied to the nut after the etchant has been washed off, the nut dried and the threads tapped.
The use of lubrication in the nut threads and on the nut bearing face reduces the tightening torque,
usually reduces the variation in k factor and reduces the risk of bolt breakage during tightening.
Electroplating and acid pickling of grade 10.9 bolts carries a risk of hydrogen embrittlement. Heat
treatment procedures to mitigate against this are used, but might not always be reliable.

C.8 Hardness of free edge surfaces

The hard thin layer of metal close to the thermally cut edge has a microstructure with very reduced
ductility which is susceptible to microcracking in the event of subsequent plastic straining, whether
during execution or in-service. The design fracture toughness requirements of the adjacent
unaffected material is normally sufficient to resist fracture from pre-existing microcracks, but not
necessarily from a crack which initiates in a hardened zone under high applied tensile strain and
which needs to be arrested. Pre-existing microcracks can also cause premature initiation of fatigue
cracks in locations subject to high cyclic stressing. The recommended limitations given in 8.4 aim to
reduce these risks.
The relaxation of the hardness limits for machine plasma cut edges in certain conditions in 8.4 is
based on the superior quality of surface finish usually achieved with this process. To assist with
interpretation of the conditions in 8.4c), typical examples showing the zones which are governed
by plain and stress raising features are illustrated in Figure C.1, together with the relevant stress
direction for assessing the QSC limits. For elements whose design is governed by stiffness or control
of deflection, e.g. intermediate transverse web stiffeners, F36 may be assumed to apply, except where
any applied stress requirements show that a higher QSC applies.
There is normally a hardness gradient in the first millimetre of material from the cut edge. The zone
of any excess hardness is normally less than this, depending on the cutting process.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 49

PD 6705-2:2020 PUBLISHED DOCUMENT

Figure C.1 — Definition of plain edge and stress raising zones for machine plasma cut edge surfaces

View A View B
Key
Plain edge (PE). Quantified service category based on stresses parallel to edge
Stress raising zone (SRZ). Quantified service category based on stresses parallel to arrow
Weld toe datum
Tangent point datum
Arrow on obscured SRZ

C.9 Flame straightening

The procedures needed for correcting different types of distortion needs to define the locations,
extent and sequence of application of heat to the component being corrected. These are significantly
different for the three modes of distortion which most commonly need correction.
The recommendations for controls on flame straightening are governed by the need to control
the maximum temperature of the steel and the time spent at temperatures close to the maximum.
Account has been taken of the guidance in the steel product standards about the potential
deterioration of material properties when steel products are subjected to sustained thermal
treatments post supply, such as stress relief and hot forming. Allowance has been made for the fact
that the cumulative time spent close to maximum temperature during the flame straightening process
is typically not as long as for the other two thermal processes.

50 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Allowance has also been made for the fact that the delivery condition is an option for the steel
supplier, but not for the purchaser, for the most commonly used steels in bridges. Modern steel
making practice is using the condition +M (thermomechanical rolling) more frequently to obtain the
required mechanical properties, which enables benefits in chemistry to be made. Condition +M is
more sensitive to post supply thermal treatments than conditions AR and N which are in Group 1).
For this reason, condition +M has been placed in Group 2), together with steels whose mechanical
properties depend on cold forming or quench and tempering and which are also more sensitive to
post supply thermal treatment.
The infra-red (IR) method of measurement of surface temperature of steel is not recommended in
view of its potential inaccuracy in measurement due to variations in the emissivity of steel surfaces at
high temperatures.
The criteria for the range of qualification of a qualified flame straightening procedure test is that
the cumulative time at temperature should not exceed that in the qualification test. For practical
purposes, this is defined in terms of the material thickness and the number of heating-cooling cycles.

C.10 Execution of holing

The practical difficulties of ensuring that the local hardness and quality of thermally cut holes in
production according to BS EN 1090‑2:2018, 6.6.3a) are such that the process should be restricted
to joints which are stressed well below the limits permitted in BS EN 1993‑2, BS EN 1993‑1‑9 and
BS EN 1993‑1‑10.

C.11 Quality management requirements

The weld quality management requirements of BS EN ISO 3834 are an important element in the
assurance of adequate welding quality in the final bridge. The comprehensive requirements of
BS EN ISO 3834‑2 are usually adequate for controlling the most stringent quality conditions covered
by BS EN 1993‑2 and BS EN 1090‑2. BS EN ISO 3834‑1:2005, Annex A compares the requirements
for BS EN ISO 3834‑2, BS EN ISO 3834‑3 and BS EN ISO 3834‑4. Many of the requirements for
BS EN ISO 3834‑2 and BS EN ISO 3834‑3 are similar. BS EN ISO 3834‑4 has few specific requirements.
For the smaller and lighter range of bridges using simple materials and constructional details, the use
of constructors certified to BS EN ISO 3834‑3 might be adequate. This avoids unnecessary rejection of
potentially competent tenderers.
Bridge constructors certificated to BS EN ISO 3834 are limited by the type of construction prescribed
by the certificating body, which approves the full details of the welding management system in
greater detail than is covered in BS EN ISO 3834.

C.12 Tack weld procedures

The main risks from the presence of tack welds are hydrogen cracking due to too low a heat input,
moisture and high restraint at the assembly stage. If such cracks are not detected prior to welding of
the joint and the root pass does not fuse the whole of the tack, they can remain in the final weld and
can even extend during the later passes. These cracks are mainly detrimental to the performance of
transversely stressed joints.
The other risk, which is more prevalent with automatic welding, is that lack of fusion can occur at the
beginning or end of the tack weld due to the sudden change in joint profile. This is detrimental to the
performance of longitudinally stressed joints, e.g. web-to-flange joints.

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 51

PD 6705-2:2020 PUBLISHED DOCUMENT

Whilst the size of such flaws is not usually large enough to cause a fracture, where cyclic stressing
requirements require a QSC of F71 or above, there is a risk that the fatigue life can be reduced.
Whilst this level of stressing might only be required occasionally, it is preferable to have an approved
weld procedure in place to eliminate this risk if the welding organization is qualified for QSC F71 or
above (see 9.1).

C.13 Weld procedure qualification

Acceptance criteria for procedure qualification testing should be set higher than that required for
normal production. This ensures that there is some margin for unavoidable deviations in production,
which can be easier to control in a procedure test. In some cases, the criteria might be the same,
but they should not be lower than the acceptance criteria required from production. This principle
applies to all forms of testing in all structural materials.

C.14 Welding co-ordination

Whilst BS EN ISO 14731 gives general principles for the role and qualification of welding
co‑ordination personnel, the full details are determined as part of the certification conditions, defined
by the BS EN ISO 3834 certification body for each individual constructor (see 9.1). The certification
body takes account of any limitations on the type of bridgework to be undertaken by the organization
when considering the requirements for the degree of authority, the previous experience and the
technical training of the welding co-ordination personnel employed for that work.

C.15 Temporary attachments

Risk of damage during removal and the difficulty of repair to the quality required for the highest
fatigue stressing situations are the reasons for the recommendation in 9.4.3. Note that the part
of the attachment weld metal which remains below the member surface constitutes part of the
permanent material.

C.16 Permanent steel backing

The addition of a permanent steel backing can reduce the detail category of a butt weld. It could also
introduce a potential corrosion initiation. These matters should be checked by the designer.
Unwelded butting backing components orientated parallel to the stressing direction constitute a very
severe notch with a fatigue detail category less than ΔσC = 56 N/mm2. Depending on cross section,
they could also increase the risk of brittle fracture.

C.17 Welding of bolted joint components

Welding of heat treated nuts, bolts or washers can temper the material and reduce their mechanical
properties. The high carbon levels can result in hydrogen cracking where small weld deposits are
used on the larger component sizes.

52 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

C.18 Fit-up of preloaded bolted joints

Adequate fit-up quality in the snug-tight condition is essential for preloaded joints, otherwise the
structural performance might be impaired. This can take various forms dependent on the joint
geometry and loading type as demonstrated in the following examples.
a) On double covered slip resistance connections, differences in the inner element thickness
on each side of the joint causes gaps on the thinner side which can lead to a greater share of
clamping force on the thick side [see Figure C.2 a)]. The slip resistance of the thinner side is
reduced. This can result from adverse rolling tolerances.
b) A more severe condition of type a) can occur when the splice plates are integrally welded on one
side of the joint, e.g. in the diagonal to chord gusset joints in a truss, resulting in loss of clamping
force [see Figure C.2 b)].
c) In double covered slip resistance joints where there is misalignment between the elements on
each side of the joint in members which are stiff relative to each other, e.g. in splices between
plate girders or between box chords in a truss, there can be gaps on opposite sides of the joint.
When the bolts are fully tightened, the reduction of the misalignment can result in overstressing
of the adjacent welds [see Figure C.2 c)].
d) In joints subjected to cyclic tension forces across the joint plane, e.g. end plated cross girder to
main girder connections, the design could depend on the faying surfaces around the bolt holes
remaining in compressive contact under all conditions of applied loading to prevent undue
cyclic tension in the bolts. For this to be achieved, the surfaces need to be in contact at the
snug-tight stage.
Figure C.2 — Examples of effect of lack-of-fit on clamping force and forces in components in pre-loaded joints

a) Mis-match between plate thickness

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 53

PD 6705-2:2020 PUBLISHED DOCUMENT

Figure C.2 (continued)

b) Insufficient packing between c) Differences in depths of plate girders

chord gusset plates and diagonal
Key
F Total applied bolt force on each side of joint
S Out-of-plane shear force in splice plate due to lack-of-fit
C Net clamping force on faying surface
R Net reaction force in component (tension positive, compression negative)

C.19 Preloading force

The minimum specified preloading force should not be lower than that given in BS EN 1090‑2:2018,
Table 18, otherwise the design assumptions made using BS EN 1993‑1‑8 could be invalid.

C.20 Torque method

The torque method is likely, in practice, to give a lower pre-tension than the methods involving a final
nut rotation. This is because the method does not take advantage of any increase in the bolt strength
above the minimum required by the product standard. It is also wholly dependent on the torque-
tension k value, which has to be conservative to avoid bolt overstretching (over which there is no
control as there is with the turn methods).
The torque method is the only safe method where bolts might need to be undone and retightened in-
service, as the turn methods usually involve incremental stretching.

54 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

C.21 Combined method

The recommended increase in turn angles is to allow for an extra margin for:
a) out of parallel of grip surfaces;
b) angular misalignment of bolt head axis or nut thread axis allowed by product standards; and
c) out of alignment of ply holes in stack.
In view of the high reliance on the k value for prevention of the bolt stretching prior to the application
of the final turn, it is important to take account of the fact that k values can be affected by storage time
and conditions.

C.22 Part turn method

The recommendations have been based on the BS 4604‑1 method, which has been widely used in
the UK, together with the bedding torque values for the Highways Agency Specification for Highways
Works, Series 1800 [1].
The advantage of this method over the combined method is that the bedding tension is typically in
the region of 20% to 30% of the ultimate bolt strength, whereas, in the combined method, it is likely
to be more than 70%. Therefore, there is no danger of stretching the bolt at step 1 and no need to use
a K1 or K2 bolt assembly.
The range of grip lengths in Table 3 is based on those given in BS 4604‑1.

C.23 Scope of weld inspection

The scope of weld inspection in BS EN 1090‑2:2018 is based on the principles of FPC, which is
not specific to any one contract passing through the works at any time (see BS EN 1090‑2:2018,
12.4.2.3). For FPC, the supplementary NDT methods are not defined and only the principles for
selection of method in BS EN ISO 17635 have to be followed (see BS EN 1090‑2:2018, 12.4.2.6).
BS EN ISO 17635 does not give any advice on suitability for verification of different imperfection
types and sizes.
In specific contracts, the extent and methods of testing of specific joints can be defined
(see BS EN 1090‑2:2018, 12.4.2.3). Table 4, Table 5 and Table 6 of this Published Document define
the methods and extent for a wide range of conditions, based on the following principles.
a) The methods and extent are dependent on the degree of difficulty in detection and evaluation
of the acceptable imperfections, their expected risk of occurrence and their adverse effect
on resistance.
b) The acceptance limits for imperfections which are considered are based on fitness-for-purpose
criteria, given in Table 7, Table 8, Table 9 and Table 10.
c) The acceptable size of imperfections becomes smaller with increasing QSC (see B.1.2 and C.24),
and this has a major effect on the method and extent of inspection required.
d) The methods are chosen according to the joint type, size and orientation with respect to the
critical stress direction, as well as the type of imperfection measurement required.
BS EN 1090‑2:2018, Annex L provides a proposal for relating the scope of and method of weld
inspection with the particular service requirements for a given type of weld. Whilst the aim is sound
in principle, the UK raised a number of concerns about its technical correctness, its usability and the
resulting extent of inspection proposed. Firstly, inspection requirements are primarily differentiated
on the basis of fatigue utilization which is fundamentally flawed in principle (see Note). Secondly,

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 55

PD 6705-2:2020 PUBLISHED DOCUMENT

they are also differentiated on the basis of a subjective consequences criterion. Thirdly, each weld
is classified in terms of detailed stress parameters specific to that weld, which then has to be
classified individually. Lastly, the extent of inspection is typically well in excess of that specified for
FPC purposes in BS EN 1090‑2:2018, Table 24, including a frequent requirement for radiography.
For these reasons the method is not considered to be an acceptable alternative to the QSC based
recommendations in this Published Document.
NOTE BS EN 1090‑2:2018, Table L.1 differentiates weld inspection class on the basis of whether or not the
“fatigue utilization” or calculated fatigue life is more or less than four times the required life. In terms of stress
this is equivalent to a “fatigue utilization” of approximately 70% of the allowable stress range (for typical bridge
spectra). However, fracture mechanics shows that the tolerable imperfection sizes are substantially reduced as
the applied stress range increases (see B.1.2). Thus, the amount of inspection required needs to also increase as
the level of cyclic stressing increases. A given “fatigue utilization” (e.g. k) produces a wide range of applied stress
ranges, depending on the detail category, so it is not a technically acceptable parameter for defining weld quality or
inspection requirements. For example, if a value of k = 0,7 were to be selected for differentiating between a stricter
and a more relaxed set of imperfection limits, a DC 100 detail stressed to, for example, 65 N/mm2 would be in the
“more relaxed” category (k = 0,65), whereas a DC 50 detail stressed to, for example, 40 N/mm2 would be in the
“stricter” category (k = 0,8). This is the reverse of what the fracture mechanics would predict.

C.24 Weld acceptance criteria

The weld acceptance criteria in BS EN 1090‑2:2018, 7.6.1 for routine factory production control
(FPC) are intended to maintain a target standard of quality suitable for general structural steelwork.
The UK proposed that this should be comparable with the criteria required for qualifying weld
procedures and welders which are mainly in accordance with BS EN 5817, level B with some features
of level C. It is normal practice in industry for FPC levels to be set at a level marginally higher than the
level needed for fitness for purpose (FFP), so that the risks of FFP non-conformances are minimized,
(particularly if only a small statistical sample of production welds is required to be tested). The
UK drew attention to the fact that this principle had not been applied and that there were too
many instances where the criteria in BS EN 1090‑2:2018, 7.6.1 were not compatible with the FFP
criteria necessary to support the recommended partial factors in BS EN 1993, particularly in the
case of EXC1, to some extent in EXC2 and occasionally in EXC3. Conversely, some of the criteria are
not necessary for FFP purposes. In view of the fact that many structural steel welding facilities are
currently certified to a particular FPC EXC and might be producing structures other than bridges, it
is not proposed to recommend alternative FPC criteria specifically for bridges, provided that the FFP
criteria in this Published Document are adhered to for all project specific production of bridge welds.
BS EN ISO 5817, which forms the basis for the FPC acceptance criteria specified in
BS EN 1090‑2:2018, 7.6, makes it clear that not all of the criteria are assessable by NDT. This raises
the question of how such criteria can be verified in the final product, i.e. whether the specification
requirements are capable of practical implementation. In other cases, some of the criteria bear little
relation to the safe performance of the component and can be unnecessarily onerous.
BS EN 1090‑2:2018, 7.6.2 provides new proposals for introducing more severe acceptance criteria
for welds subject to fatigue. Whilst the aim is sound in principle, the UK raised concerns about its
technical correctness, its usability and the resulting consequences for required production effort.
Firstly, the assumed fatigue severity is based on the maximum permitted strength for the detail
category (DC) concerned, not the actual cyclic stress level in the weld, which in most bridges is
substantially lower than the DC permits (see Annex B). Secondly, every weld needs to have its
DC(s) labelled, (presumably in more than one direction). Thirdly, the acceptance criteria are still
predominantly based on the FPC criteria in BS EN 1090‑2:2018, 7.6.1, so some criteria might not be
sufficient for ULS let alone fatigue. Fourthly, the criteria in BS EN ISO 5817:2014, Annex C are based
on different ΔσC-N-curves and DCs from those in BS EN 1993‑1‑9. They only consider individual

56 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

imperfections (not multiple or combined with different types), do not address partial factors and are
subject to a warning that there might be difficulties in establishing NDT criteria for some of them.
For these reasons, the method is not considered to be an acceptable alternative to the QSC based FFP
system in this Published Document.
Therefore, FFP based acceptance criteria have been provided in Table 7, Table 8, Table 9 and Table 10
for assessing welds in bridges to ensure safe performance and avoid unnecessarily stringent testing
and quality requirements. The recommendations in Table 7, Table 8, Table 9 and Table 10 are based
on the principles established in BS 5400‑6, ISO 10721‑2 and the proposed FFP based specification
clauses submitted to CEN/TC135 by the UK at the time of enquiry. The main features are listed below.
a) The acceptance criteria are dependent on the QSC.
b) The acceptance criteria only include those requirements which are essential for structural
performance.
c) The acceptance criteria for each inspection method are given in separate tables in terms
appropriate to each method.
d) Relaxation of imperfection limits has been given for longitudinal welds.
e) Differentiation in limits for buried imperfections is made according to their position in the cross
section. Restrictions are also given on the minimum separation distance between imperfections.
f) Reference is made to BS EN ISO 6520‑1 and BS EN ISO 5817 for imperfection terminology,
definitions, reference numbers and dimensional symbols whenever they are given in
those standards.
g) The acceptance criteria in Table 7, Table 8, Table 9 and Table 10 represent the physical limits
which are acceptable for performance purposes. No allowance has been made for measurement
error. In cases of doubt, particularly in the assessment of a buried imperfection to Table 10, the
imperfection should be rejected and its true dimension verified by excavation for the purposes of
calibration of the NDT measurement technique.
Some of the more important features of the FFP criteria in Table 7, Table 8, Table 9 and Table 10 and
their comparison with the FPC criteria in BS EN ISO 5817 are given below.
Profile limitations given in the FPC criteria are not included unless they have a negative effect on
resistance. This includes excess weld metal BS EN ISO 5817 imperfection numbers 1.9, 1.10, 1.11
and 1.21. Such stress concentration effects are generally controlled by toe angle. These FFP toe angle
limits lead to less restriction for low QSCs but are tighter for higher QSCs. Similarly, there is no control
stipulated for asymmetry of leg lengths. These can be controlled for FPC purposes, e.g. distortion, cost
of consumables or, in some cases, design considerations, e.g. clearance, drainage. However, for FPC
level B, in particular, the limits are very restrictive and leave a small margin against undersize non-
conformances.
The FFP limits for surface notches tend to be more restrictive than allowed by FPC level B,
particularly as the QSC increases.
The FFP criteria do not permit undersize fillet weld throats, which is the same as for FPC level B.
FPC level C does allow sub-size throats. The reason for not allowing sub-size throats for FFP is that it
is not practicable to assess the internal condition of fillet welds by NDT, and some allowance has to
be made for the undetected presence of such imperfections, such as lack of root penetration, lack of
side wall fusion, root gap, porosity, HAZ and solidification cracking etc., all of which can reduce the
effective throat.
The FPC criteria do not give a limit on leg length. In fillet welded tee, cruciform and corner joints,
which are stressed transversely, the failure plane is close to or on the fusion boundary of the stressed

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 57

PD 6705-2:2020 PUBLISHED DOCUMENT

element. For asymmetric welds with convex caps and root gaps the limits given, even for level B, do
not prevent serious shortfalls on leg length even if the throat measurement is within limits. For this
reason, strict controls on leg length are included for FFP.
For FFP assessment of volumetric imperfections, reliance is put on ultrasonic testing, except where
radiographic inspection might be needed for the highest QSCs (see Table 10). Characterization of
types of internal imperfection is not required except in the case of cracks. A limit of 3 mm in height
is assumed, although this might mean restricting length in some cases. This is considered to be a
minimum height for UT assessment purposes. Some of the FPC imperfections have height limits
below 3 mm, which cannot be assessed by commercial NDT techniques.
The FPC criteria for buried imperfections do not distinguish between near surface and deeper
zones, the former being more critical on an FFP basis. Neither do the FPC criteria for multiple
imperfections (imperfection number 4.2 in BS EN ISO 5817) give any limits on minimum clearance
between adjacent imperfections. For these reasons, the FPC criteria for buried imperfections can
be inadequate.
The detailed FPC criteria for porosity can only be assessed using radiography, which is not commonly
used for structures. The FFP criteria for porosity have been simplified for this reason.
In summary, some of the FPC range of criteria can be either excessive or inadequate when judged on
an FFP basis, depending on the imperfection type and the required QSC.
The above system of specific inspection and FFP acceptance criteria means that:
1) the customer has a clearly defined level of assurance that satisfactory performance in service can
be achieved for each bridge contract; and
2) the constructor is able to estimate the production and testing costs for each bridge contract at
the time of tender.

Annex D (informative)
Full references of Parts of normative standards not
provided in BS EN 1090‑2:2018, Clause 2

D.1 General
The following normative standards are listed in BS EN 1090‑2:2018 as having all their individual
Parts included as a normative requirement, but without the full reference of each Part being provided.
This is not consistent with other normative standards where only some of the individual Parts are
normative but where the full reference is provided. The full list of missing references is given in D.2 to
D.7 in the order in which the parent standard appears in BS EN 1090‑2:2018, Clause 2.

D.2 Steels (see BS EN 1090-2:2018, 2.1.1)

BS EN 10149, Hot rolled flat products made of high yield strength steels for cold forming
Part 1: General technical delivery conditions
Part 2: Technical delivery conditions for thermomechanically rolled steels
Part 3: Technical delivery conditions for normalized or normalized rolled steels

58 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

BS EN 10163, Delivery requirements for surface condition of hot-rolled steel plates, wide flats
and sections
Part 1: General requirements
Part 2: Plate and wide flats
Part 3: Sections
BS EN ISO 9445, Continuously cold-rolled stainless steel – Tolerances on dimensions and form
Part 1: Narrow strip and cut lengths
Part 2: Wide strip and plate/sheet

D.3 Mechanical fasteners (see BS EN 1090-2:2018, 2.1.4)

BS EN 14399, High-strength structural bolting assemblies for preloading
Part 1: General requirements (2015)
Part 2: Suitability for preloading (2015)
Part 3: System HR – Hexagon bolt and nut assemblies (2015)
Part 4: System HV – Hexagon bolt and nut assemblies (2015)
Part 5: Plain washers (2015)
Part 6: Plain chamfered washers (2015)
Part 7: System HR – Countersunk head bolt and nut assemblies (2018)
Part 8: System HV – Hexagon fit bolt and nut assemblies (2018)
Part 9: System HR or HV – Direct tension indicators for bolt and nut assemblies (2018)
Part 10: System HRC – Bolt and nut assemblies with calibrated preload (2018)
BS EN 15048, Non-preloaded structural bolting assemblies
Part 1: General requirements
Part 2: Fitness for purpose
BS EN ISO 6789, Assembly tools for screws and nuts – Hand torque tools
Part 1: Requirements and methods for design conformance testing and quality conformance testing:
minimum requirements for declaration of conformance
Part 2: Requirements for calibration and determination of measurement uncertainty

D.4 Welding (see BS EN 1090-2:2018, 2.3)

BS EN ISO 3834, Quality requirements for fusion welding of metallic materials
Part 1: Criteria for selection of the appropriate level of quality requirements
Part 2: Comprehensive quality requirements
Part 3: Standard quality requirements
Part 4: Elementary quality requirements
Part 5: Documents with which it is necessary to conform to claim conformity to the quality requirements
of ISO 3834‑2, ISO 3834‑3 or ISO 3834‑4
BS EN ISO 14554, Quality requirements for welding – Resistance welding of metallic materials

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 59

PD 6705-2:2020 PUBLISHED DOCUMENT

Part 1: Comprehensive quality requirements

Part 2: Elementary quality requirements
BS EN ISO 15609, Specification and qualification of welding procedures for metallic materials – Welding
procedure specification
Part 1: Arc welding
Part 2: Gas welding
Part 3: Electron beam welding
Part 4: Laser beam welding
Part 5: Resistance welding
Part 6: Laser-arc hybrid welding
BS EN ISO 17660, Welding – Welding of reinforcing steel
Part 1: Load-bearing welded joints
Part 2: Non load-bearing welded joints

D.5 Testing (see BS EN 1090-2:2018, 2.4)

BS EN ISO 6507, Metallic materials – Vickers hardness test
Part 1: Test method
Part 2: Verification and calibration of testing machines
Part 3: Calibration of reference blocks
Part 4: Tables of hardness values
BS EN ISO 17636, Non-destructive testing of welds – Radiographic testing
Part 1: X- and gamma-ray techniques with film
Part 2: X- and gamma-ray techniques with digital detectors

D.6 Erection (see BS EN 1090-2:2018, 2.5)

ISO 4463, Measurement methods for building – Setting-out and measurement
Part 1: Planning and organization, measuring procedures, acceptance criteria
Part 2: Measuring stations and targets
Part 3: Check-lists for the procurement of surveys and measurement services

D.7 Corrosion protection (see BS EN 1090-2:2018, 2.6)

BS EN ISO 2063, Thermal spraying – Metallic and other inorganic coatings – Zinc, aluminium and
their alloys
Part 1: Design considerations and quality requirements for corrosion protection systems
Part 2: Execution of corrosion protection systems

60 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

BS EN ISO 8501, Preparation of steel substrates before application of paints and related products –
Visual assessment of surface cleanliness
Part 1: Rust grades and preparation grades of uncoated steel substrates and of steel substrates after
overall removal of previous coatings
Part 2: Preparation grades of previously coated steel substrates after localized removal of
previous coatings
Part 3: Preparation grades of welds, edges and other areas with surface imperfections
Part 4: Initial surface conditions, preparation grades and flash rust grades in connection with high-
pressure water jetting
BS EN ISO 8502, Preparation of steel substrates before application of paints and related products –
Tests for the assessment of surface cleanliness
Part 2: Laboratory determination of chloride of cleaned surfaces
Part 3: Assessment of dust on steel surfaces prepared for painting (pressure-sensitive tape method)
Part 4: Guidance on the estimation of the probability of condensation prior to paint application
Part 5: Measurement of chloride on steel surfaces prepared for painting (ion detection tube method)
Part 6: Extraction of soluble contaminants for analysis – The Bresle method
Part 9: Field method for the conductometric determination of water-soluble salts
Part 11: Field method for the turbidimetric determination of water-soluble sulfate
BS EN ISO 8503, Preparation of steel substrates before application of paints and related products –
Surface roughness characteristics of blast-cleaned steel substrates
Part 1: Specifications and definitions for ISO surface profile comparators for the assessment of abrasive
blast-cleaned surfaces
Part 2: Method for the grading of surface profile of abrasive blast-cleaned steel – comparator procedure
Part 3: Method for the calibration of ISO surface profile comparators and for the determination of
surface profile – focusing microscope procedure
Part 4: Method for the calibration of ISO surface profile comparators and for the determination of
surface profile – stylus instrument procedure
Part 5: Replica tape method for the determination of the surface profile
BS EN ISO 8504, Preparation of steel substrates before application of paints and related products –
surface preparation methods
Part 1: General principles
Part 2: Abrasive blast-cleaning
Part 3: Hand- and power-tool cleaning
BS EN ISO 12944, Paints and varnishes – Corrosion protection of steel structures by protective
paint systems
Part 1: General introduction
Part 2: Classification of environments
Part 3: Design considerations
Part 4: Types of surface and surface preparation
Part 5: Protective paint systems

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 61

PD 6705-2:2020 PUBLISHED DOCUMENT

Part 6: Laboratory performance test methods

Part 7: Execution and supervision of paint work
Part 8: Development of specifications for new work and maintenance
Part 9: Protective paint systems and laboratory performance test methods for offshore and
related structures

62 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

Bibliography
For dated references, only the edition cited applies. For undated references, the latest edition of the
referenced document (including any amendments) applies.
Standards publications
BS 4395 (withdrawn), Specification for high strength friction grip bolts and associated nuts
and washers for structural engineering – Higher grade bolts (waisted shank), nuts and general
grade washers
BS 4604‑1 (withdrawn), Specification for the use of high strength friction grip bolts in structural
steelwork – Metric series – Part 1: General grade
BS 5400 (withdrawn), Steel, concrete and composite bridges
BS 5400‑6 (withdrawn), Steel, concrete and composite bridges – Part 6: Specification for materials and
workmanship, steel
BS 5950 (withdrawn), Structural use of steelwork in building
BS 8118 (withdrawn), Structural use of aluminium
BS EN 1011‑2:2001, Welding – Recommendations for welding of metallic materials – Part 2: Arc
welding of ferritic steels
BS EN 1779, Non-destructive testing – Leak testing – Criteria for method and technique selection
BS EN 1993‑1‑1, Eurocode 3: Design of steel structures – Part 1-1: General rules and rules for buildings
BS EN 1993‑1‑8, Eurocode 3: Design of steel structures – Part 1-8: Design of joints
BS EN 1993‑1‑10, Eurocode 3: Design of steel structures – Part 1-10: Material toughness and through-
thickness properties
BS EN 10340, Steel castings for structural uses
BS EN ISO 3834‑1:2005, Quality requirements for fusion welding of metallic materials – Part 1: Criteria
for the selection of the appropriate level of quality requirements
BS EN ISO 3834‑3, Quality requirements for fusion welding of metallic materials – Part 3: Standard
quality requirements
BS EN ISO 3834‑4, Quality requirements for fusion welding of metallic materials – Part 4: Elementary
quality requirements
BS EN ISO 14731, Welding coordination – Tasks and responsibilities
CEN ENV 1090‑5, Execution of steel structures – Part 5: Supplementary rules for bridges
DD ENV 1090‑1 (withdrawn), Execution of steel structures – Part 1: General rules and rules for
buildings (together with United Kingdom National Application Document)
DD ENV 1090‑3 (withdrawn), Execution of steel structures – Part 3: Supplementary rules for high yield
strength steels
DD ENV 1090‑4 (withdrawn), Execution of steel structures – Part 4: Supplementary rules for hollow
section structures
DD ENV 1090‑6 (withdrawn), Execution of steel structures – Part 6: Supplementary rules for
stainless steel
DD ENV 1993‑1‑1 (withdrawn), Eurocode 3: Design of steel structures – Part 1-1: General rules and
rules for buildings (together with United Kingdom National Application Document)

© THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED 63

PD 6705-2:2020 PUBLISHED DOCUMENT

Other publications
[1] HIGHWAYS AGENCY. Manual of Contract Documents for Highways Works – Volume 1:
Specification for Highways Works – Series 1800 – Structural Steelwork. November 2005.

64 © THE BRITISH STANDARDS INSTITUTION 2020 – ALL RIGHTS RESERVED

PUBLISHED DOCUMENT PD 6705-2:2020

THIS PAGE DELIBERATELY LEFT BLANK

NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAW

British Standards Institution (BSI)

BSI is the national body responsible for preparing British Standards and other
standards-related publications, information and services.
BSI is incorporated by Royal Charter. British Standards and other standardization
products are published by BSI Standards Limited.

About us Reproducing extracts

We bring together business, industry, government, consumers, innovators For permission to reproduce content from BSI publications contact the BSI
and others to shape their combined experience and expertise into standards Copyright and Licensing team.
-based solutions.
The knowledge embodied in our standards has been carefully assembled in Subscriptions
a dependable format and refined through our open consultation process. Our range of subscription services are designed to make using standards
Organizations of all sizes and across all sectors choose standards to help easier for you. For further information on our subscription products go to
them achieve their goals. bsigroup.com/subscriptions.
With British Standards Online (BSOL) you’ll have instant access to over 55,000
Information on standards British and adopted European and international standards from your desktop.
We can provide you with the knowledge that your organization needs It’s available 24/7 and is refreshed daily so you’ll always be up to date.
to succeed. Find out more about British Standards by visiting our website at You can keep in touch with standards developments and receive substantial
bsigroup.com/standards or contacting our Customer Services team or discounts on the purchase price of standards, both in single copy and subscription
Knowledge Centre. format, by becoming a BSI Subscribing Member.
PLUS is an updating service exclusive to BSI Subscribing Members. You will
Buying standards automatically receive the latest hard copy of your standards when they’re
You can buy and download PDF versions of BSI publications, including British revised or replaced.
and adopted European and international standards, through our website at
bsigroup.com/shop, where hard copies can also be purchased. To find out more about becoming a BSI Subscribing Member and the benefits
of membership, please visit bsigroup.com/shop.
If you need international and foreign standards from other Standards Development
Organizations, hard copies can be ordered from our Customer Services team. With a Multi-User Network Licence (MUNL) you are able to host standards
publications on your intranet. Licences can cover as few or as many users as you
wish. With updates supplied as soon as they’re available, you can be sure your
Copyright in BSI publications documentation is current. For further information, email [email protected].
All the content in BSI publications, including British Standards, is the property
of and copyrighted by BSI or some person or entity that owns copyright in the Revisions
information used (such as the international standardization bodies) and has
formally licensed such information to BSI for commercial publication and use. Our British Standards and other publications are updated by amendment or revision.
We continually improve the quality of our products and services to benefit your
Save for the provisions below, you may not transfer, share or disseminate any
business. If you find an inaccuracy or ambiguity within a British Standard or other
portion of the standard to any other person. You may not adapt, distribute,
BSI publication please inform the Knowledge Centre.
commercially exploit or publicly display the standard or any portion thereof in any
manner whatsoever without BSI’s prior written consent.
Useful Contacts
Storing and using standards Customer Relations
Tel: +44 345 086 9001
Standards purchased in soft copy format:
Email: [email protected]
• A British Standard purchased in soft copy format is licensed to a sole named
user for personal or internal company use only. Subscription Support
• The standard may be stored on more than one device provided that it is accessible Tel: +44 345 086 9001
by the sole named user only and that only one copy is accessed at any one time. Email: [email protected]

• A single paper copy may be printed for personal or internal company use only.
Knowledge Centre
Standards purchased in hard copy format: Tel: +44 20 8996 7004
• A British Standard purchased in hard copy format is for personal or internal Email: [email protected]
company use only.
Copyright and Licensing
• It may not be further reproduced – in any format – to create an additional copy. Tel: +44 20 8996 7070
This includes scanning of the document.
Email: [email protected]
If you need more than one copy of the document, or if you wish to share the
document on an internal network, you can save money by choosing a subscription BSI Group Headquarters
product (see ‘Subscriptions’).
389 Chiswick High Road London W4 4AL UK