Alasdair’s Engineering Pages © A. N. Beal 2016    

Alasdair’s Engineering Pages

The Structural Engineer, Vol. 73. No. 10, 16 May 1995


Codes for the 21st Century: BS£££, EC999 or Codes-u-like?

A. N. Beal BSc CEng MIStructE MICE, Thomason Partnership

The price of progress

Structural codes of practice have always been controversial but today the biggest issue for many practising engineers is not philosophy but cost. In tune with the spirit of the times, BSI has ‘built up its core market’, i.e. locked customers into QA schemes which require them to buy every new code edition or revision which is published, is ‘publishing more codes than ever before’, i.e. dividing up codes into innumerable parts and issuing ‘new editions’ instead of amendments and, it seems, is intent on charging its captive customers more money for its codes than ever before.

Where in the past a code might have been expected to last, with a few judicious amendments, for 20 years or more - 42 years unamended in the case of that wonderful old warhorse CP2 - today the norm is about 5 years. More replacements mean more expense - and, if truth be told, more photocopying as many engineers ‘opt out’ of their responsibilities under copyright law. While the Institution of Structural Engineers sells the public its ‘Gold Book’ Recommendations for Permissible Stress Design of Reinforced Concrete Building Structures complete with design charts for only £25, BSI now charges over £105 for BS81I0, with design charts extra.

A fine example of the new ‘spirit of enterprise’ can be found in the latest draft amendment to BS8110. This proposes to transfer guidance on specifying concrete out of BS8110 into BS5328 - making one considering strength and the other durability. If the amendment is adopted, it will appear as a ‘new edition’ of BS8110, to replace the ‘old’ editions in use at present - and because the ‘new’ BS8110 will no longer cover design for durability, engineers will also need to buy BS5328, a modest little 4 part document costing £137, as well.

If BSI manages to hold prices at 1994 levels, the cost of the codes needed to design a concrete structure will rise from £105.50 to £242.50 - almost ten times the price of the equivalent guidance from the Institution of Structural Engineers! If design for durability stays in BS8110, and many would argue that it should, there would be no need for most of the proposed amendments and the remainder could easily be covered by an amendment slip costing (say) £10, saving each existing user over £200. The total sum of money involved is considerable: if we assume that there are (say) 10,000 copies of BS8110 in use, many owned by non-members of BSI, then the proposed amendment could cost the profession about £1-2M altogether. Not bad for a routine technical amendment?

Harmonisation, limit state, Eurocodes and all that

In the ‘bad old days’, structural engineers had only one set of Codes of Practice for design: CP111 for masonry, CP112 for timber, BS449 for steelwork, CP114, 115 and 116 for reinforced, prestressed and precast concrete, CP117 for composite members and CP118 for aluminium. Steel bridges were designed to BS153 (a nice slim volume) and there were CP2004 and CP2 for foundations and retaining walls They were generally well written, practical, concise - and technically consistent with one another.

Now things are much better. For a start, there is more choice - either limit state or permissible stress design, except for timber and foundations, where promised limit state Codes are still ‘under development’. Then there are manuals: when each new BS appears, bodies like IStructE. BCA and SCI rush to publish booklets translating its content into something better suited to the needs of practising engineers. To add interest, each new limit state Code often presents a new interpretation of limit state theory, perhaps dividing the safety factor into a different number of partial factors, or choosing different values for them, or maybe just presenting them for use in a way which is different from other Codes. In this way, the individuality of the Code committee is expressed and consumer interest is maintained.

One of the more refreshing changes in modern Codes is that simplicity is no longer seen as an overriding necessity. Today’s drafters are happy to incorporate the complex formulae generated by researchers into their new Codes - encouraged by computer software companies which can be relied on to offer engineers programmes to make things easy again (at a price). Those who press for simpler Codes are often insensitive to the very real hardship these would cause in the software industry.

However, for the cognoscenti, Codes such as BS8110, BS5950 etc. are distinctly passé. The future lies in Europe and its exciting new Eurocodes and Euronorms. Not only do these offer many interesting new variations on the limit state theme and a reassuringly non-tabloid presentation but they also bring the unexpected bonus of a completely new language for technical matters. The visionaries and innovators among us can now express their creative thoughts in terms which will set them apart from the common herd. Not for them the outmoded concepts of ‘forces’, ‘moments’, ‘shears’ and ‘stresses’ - these engineers of tomorrow will speak of actions (either direct, indirect, permanent, variable, free or fixed), transverse or tangential action effects, persistent and transient situations and much more.

Abandoning the language, ideas - even the sign conventions - of the past, they will lead the way to a brave new world. Perhaps one day they may even learn how to calculate nominal reinforcement in accordance with Eurocode 2 clause 4.4.2. In design offices up and down the country, engineers wait with barely concealed anticipation, eager for the day when the present draft ENVs become full Eurocodes and present UK Codes are consigned to the dustbin of history.

Sadly, progress has been less impressive in sorting out the various theoretical problems which beset limit state theory [1]. No doubt one day these will he solved and engineers using limit state Codes will he able to analyse problems such as overturning, continuous beams and laterally-loaded walls in a rational way. However, until then, limit State aficionados must just muddle along as best they can - and ignore the siren voices which call for a return to permissible stress design.

Codes for the modern world

The political and economic upheavals of the 1980s have made surprisingly little difference to the content and style of structural Codes. It still seems to be assumed that design and construction is securely under the control of independent consulting engineers - high priests of the construction industry who can be relied on to balance the interests of economy and quality in an intelligent manner for best results. Codes such as BS5950 were intended to impose as few restrictions as possible on these highly skilled professionals, leaving them to judge when to push stresses or deflection to the limit and when to exercise more caution. It was always assumed, of course, that one engineer would have overall responsibility for the structure as a whole, coordinating the designs of subcontractors etc.

Deregulation, fee-cutting and the design-and-build revolution have changed all that. Today’s engineer is often employed by a contractor (or subcontractor), usually on a limited brief. ‘The structure’ may be a series of separate tendered packages, each built by a different subcontractor, each with its own engineer and then connected together on site as well as can be managed. All too often, design is carried out by engineers of limited experience and ability, working to a very tight programme, on a very low fee - and under intense commercial pressure. The only controls on technical standards are Codes of Practice and Building Control

The present generation of Codes was never intended for this situation. Their relaxed safety limits were intended to be used with care, not treated as routine commercial targets, and their often complex guidance can be easily misused and misunderstood by engineers working under pressure. The Steel Construction Institute Advisory Desk has reported a large increase in queries about odd applications and interpretations of clauses in BS5950 - exposing loopholes in a document which was never intended to serve as the technical rulebook governing an intense commercial contest.

For that purpose, quite a different style of document is needed - something closer to BS5950’s predecessor, BS449. This was originally drafted in the 1930s, in commercial circumstances surprisingly similar to those which prevail today. The 1920s had seen the development of a highly competitive design-and-build market and the commercial pressures this placed on designers were creating problems - ‘cut designs’ would be prepared to win contracts at the lowest possible price, undercutting the work of more responsible designers. In this free-for-all, regulation was needed to stop things going badly wrong.

The Codes of the 1930s were drafted to bring some order to the situation, setting common technical standards to safeguard the interests of the public and ensure fair competition. Codes drafted for this purpose need to be technically sound, based closely on existing practice, to be clear, concise and simple - and to be as free of ambiguity and loopholes as possible. Engineers brought up in better times may find this unpleasantly restrictive but if technical standards are to survive in the rough and tumble of the marketplace, some restrictions on designers’ freedom are inevitable.

Codes like this may not be the kind of documents that researchers and Code committees would like to write but our Code drafters must take account of the real commercial world their documents will be used in - and the role they are required to play in it. There is no point in drafting complex, exotic Codes designed for use in circumstances which no longer exist. The Codes we need for the next century will be humbler beasts - simple, practical documents drafted to ensure that standards of safety and serviceability are maintained, despite the commercial pressures acting on the designers of today.


1.  Beeby, A. W., Beal, A. N. and Taylor, J. C.: ‘γ-factors - a second look’, The Structural Engineer, Vol. 72, No. 2, 18 January 1994.

The original copy of this paper is available from