NSC Archives

Projects and Features

Eurocode 3: A personal view

Eurocode 3 will soon be upon us. Most organisations have not made adequate preparations for it. This paper was given at the BCSA conference in November 2003.

This is a personal view of what I believe the introduction of EC3 might mean for the steelwork community in the UK and to supplement this with my own thoughts on what should be done in order that the transition run as smoothly as possible. I am even prepared to be so rash as to provide some indicative time scales, although a period of over 30 years of personal engagement with the process of preparing Codes of Practice has demonstrated all too clearly that the only aspects of time scale on which one can be certain is that all estimates will be significantly exceeded.

Many of you will be familiar with at least some of the key dates in the development of EC3. Activity effectively started in 1975 – although the UK as a body took relatively little interest. This was in spite of the fact that the chairman of the drafting panel, Professor Patrick Dowling was, of course, the occupant of the British Steel Chair at Imperial College. Early work was largely undertaken by enthusiasts, with very little tangible support. The drafting panel started from the ECCS Recommendations and produced an initial draft in 1983. It was not difficult to spot the joins between the different sections and to work out which member of the panel had been principally responsible for which of them. The process then moved into a more serious pan- European engagement with a draft for use – the so called ENV document – appearing in 1992. The subsequent conversion process into the final EN document began in earnest a few years ago and the first package of five parts is scheduled to appear very shortly. Interestingly, the November 2001 special issue of the ICE’s Civil Engineering on Eurocodes gave the date as February 2003. The first issue of Eurocodes News that appeared in July 2003 wisely did not give a date. It is intended that this package contain:

  • Part 1.1 General Rules and Rules for Buildings
  • Part 1.2 Structural Fire Design
  • Part 1.8 Design of Joints • Part 1.9 Fatigue
  • Part 1.10 Material Toughness and Through Thickness

Although the final length of each document will not be clear until we actually see the printed version, it is already apparent that more material is being provided on the design of joints than is given to the whole of the Part 1.1.

Important points to note during this gestation period are: the steady growth in the number of countries involved in the process, an inexorable increase in the range of topics being covered (countered very recently by realisation that the length of at least some of the documents was becoming unrealistic), the growing importance of supporting standards for products and an increasing realisation of the interaction between the introduction of the Eurocodes and the National legislative position in the various countries.

Despite an initial official reluctance to engage with the process, the UK has had a surprising large influence on the content of EC3. Not only was Patrick Dowling the first and longest serving chairman of the drafting panel, he was ably assisted in the role of technical secretary by Colin Taylor of the SCI. In addition, many prominent members of the structural steelwork community in the UK have contributed (often in a relatively invisible fashion) to the development of many parts of EC3 and support has been provided by Corus, BCSA, SCI, BRE and others. Interestingly, the UK has had an even stronger influence on the Composite Code EC4, largely through the efforts of Roger Johnson and David Anderson at the University of Warwick.

The result is that the documents are far closer to the sort with which we in the UK are familiar than might have been expected. Of course, individuals, no matter how talented or how forceful, cannot be expected to secure UK preferences on every occasion. Thus the axis system, notation, aspects of the rather pedantic style and organisation, as well as certain of the technical provisions are undoubtedly not of a style that we would choose. It is of interest that those in UK Government with responsibility for our legal framework i.e. the building regulations, are now far more personally engaged than was the case some 20 years ago.

It is also of interest to observe the pattern of provision of government/industry resource – principally payment for the time of suitable people – to undertake work in support of the development of Eurocode 3 at various times.

Initially, little help was provided; it was almost as if the official view was one of ignoring a potential nuisance. More recently, however, relatively large amounts of resource have been provided e.g. to undertake calibration, produce sets of worked examples, prepare a simplified and more restricted version of the ENV document and to produce the Institution of Structural Engineers’ Manual. All these were, of course, undertaken with the express aim of encouraging usage of the ENV for real projects during the trial period. I know of almost no examples of this – except for a couple of contrived applications, in which parallel designs were paid for and parts of designs for which rules were available in EC3 (but not in BS 5950) and the Code was used as some sort of reference.

Given the extremely modest degree of engagement with the ENV by the Industry, it is questionable whether the amount of resource devoted at that time would not have been much better saved and spent in support of the introduction of the final EN document.

In Eurocode jargon we have now reached “Stage 49” for the initial package of EC3 material. This means that the documents are going through the final editing process, the technical content having been agreed, and we can expect to see the published versions very shortly. We already have final published versions of: EC1 dealing with Loading and EC0, dealing with load combinations and load factors.

Working with the Eurocode package on even the simplest of structures would be expensive in the sense that a large number of separate documents will be required and the purchase prices are not low. Nor would it be straightforward in the sense that the organisation of EC3 does not lend itself to the conduct of specific design tasks, since material has been arranged in a very different way from the British norm.

Many of you will have heard me state that I do not see the “average designer” working directly from the Eurocode documents. They are, quite simply too voluminous and too complex. Thus some form of “interpretive assistance” will, in my view, be essential. This raises the obvious question of how might this reasonably be arranged? One view is that the power of modern IT systems ought to figure more prominently. The SCI has an embryo programme of work in this area; “Steelbiz” is an obvious vehicle to use and the feasibility of operating in this fashion was first demonstrated nearly a decade ago ¹. To be effective, however, activity of this type should be undertaken on a national scale, using the best available resources and should be co-ordinated in such a way that we do not arrive at six slightly different versions of the most obvious and most attractive item, with nothing being done in support of more difficult areas. Unfortunately, history suggests the opposite. To appreciate this, consider the various organisations that ought, in some way, to be involved:

  • BSI – essentially publishes and sells the Code document; importantly it owns the copyright.
  • SCI/BCSA/Corus – support the market; the first two need to be resourced to undertake work in this area.
  • UK experts – most, if not all of the people most able to contribute are fully employed in their “day jobs”.
  • Government – typically concerns itself with “correcting wrongs” and not “establishing rights”.
  • Software developers – will undoubtedly wish to upgrade their products but on the basis of an established and agreed interpretation of the new regime.
  • Design community – on past history will delay adoption of EC3 until the last possible moment on the basis that the great majority will see little if any benefit in effecting the changeover.
  • Owners and insurers will wish to see their “product” adopt what they believe to be the most up-to-date, competitive and safe approach to design.

Of course, significant pockets of activity already exist and these were collected together in an interesting illustration for use at the recent SCI Council Awayday by Graham Owens in Fig 1. What is illustrated graphically by this picture is the way in which different organisations are pursuing different objectives to different time scales with only the minimum of interaction. Where such interaction is occurring, it is largely as a result of the individuals concerned taking it upon themselves to liaise with other known and therefore potentially relevant activities. To take some examples:

I understand from David Moore at BRE that they are concentrating on two ODPM funded initiatives; the preparation of:

  • A Companion Design Guide
  • A set of worked examples

The former is aimed at senior design engineers and will include various items of information.

The second document will illustrate the application of: Part 1.1. , 1.2, 1.8, 1.10 and Part 5 – Piling.

It is accepted that in producing these worked examples a number of “problems” will be encountered in the sense that some form of interpretation of the way in which the new procedures should be implemented – especially when not all material is provided in the Code itself – will arise.

“Organisation” of EC3 Support Activities
(courtesy of SCI)

Taking the specific example of a software developer, Alan Rathbone of CSC has been kind enough to share some thoughts with me on their likely work programme. He has emphasised the paramount need to have an agreed authoritative approach to the actual implementation of all the relevant technical provisions. This is not, of course, a task that any software developer would wish to undertake alone nor is it likely that any of them would have anything approaching the necessary in-house specialist technical expertise to do it. They are also acutely aware of issues of timing and the unpredictable nature of the changeover to EC3 by different companies and in different sectors of the steel construction market.

The Institution of Structural Engineers has agreed to produce a revised version of its Manual of Design to EC3 and will be working in association with the BCSA and the SCI to prepare this. It is not a simple task consisting of picking the sections of the current document required for the most frequently encountered design situations, as coverage in EC3 of many topics is somewhat different from the equivalent treatment in BS 5950. For example, guidance on selecting appropriate values for the effective length of columns, several aspects of the design of portal frames etc, simply are not present. An exercise undertaken on behalf of the BSI Sub-committee dealing with BS 5950 and EC3 produced a list of 64 items where users of EC3 who were accustomed to working with BS 5950 would not find an equivalent provision. Some “authority” will have to establish the accepted UK view on matters of this type.

It is therefore of interest to speculate on the cost of transition. I put this question to Derek Tordoff of the BCSA and here is a slightly modified version of his response:

  • The value of the UK steel construction industry output per annum is about £2,500m.
  • If total design costs are assumed as 8% this would give a figure of £200m per annum.
  • Taking the directly code-related work as (say) 50% of this activity gives a figure of £100m per annum.

I leave you to speculate on how much less efficient that activity would be as a result of inadequate preparation to use EC3; even a 10% loss would equate to £10m per annum. If we were to have a structural failure that was directly attributed to misapplication of EC3, followed, by a period of extreme caution i.e. our industry’s equivalent of the Hatfield rail crash and the ensuing aftermath, such an estimate would be but a fraction of the eventual total cost.

Put this way it does, of course, seem logical that a major national programme should be put in place, with its cost being covered by the savings from not doing it. Identifying and persuading those responsible for the latter to contribute to the former is, of course, far from straightforward.

My main purpose is to leave you with this thought. We are scheduled to hear of recent successes within the steel construction Community in this country – arguably the most vibrant and successful steel construction community possessed by any nation worldwide. Do we wish to see that compromised through inadequate preparation for an inevitable change? Do we wish to experience an extended transition period that seems likely to be more disruptive than the conversion from permissible stress to limit states design that accompanied the introduction of BS 5950 in 1985? If the answers to these questions are no – as I firmly believe they should be – what is our community prepared to do to correct the situation and how can we mobilise the assistance of others in this task?


  1. Byfield, M.P. and Nethercot, D.A., “Can Codes of Practice be both Comprehensive and User-Friendly?” Fourth Pacific Structural Steel Conference, Singapore, Vol.1, Steel Structures, ed. N.E. Shanmugan and Y.C. Choo, Pergaman, 1995, pp.29-40.

Prof. David Nethercot, President of The Institution of Structural Engineers. Head of Department of Civil and Environmental Engineering, Imperial College London.

Print Friendly, PDF & Email

Share this post

Related Posts


Click on the cover to view this month's issue as a digimag.