Skip to main content

CROSS Safety Report

Incompetent design of simple steel beams

Report ID: 1183 Published: 24 May 2023 Region: CROSS-UK


Overview

This report concerns designs for simple steel beams submitted to building control bodies under Part A of the Building Regulations. The reporter, a checking engineer for a building control body, was not able to accept the effective length and restraints assumed in a number of submitted designs. Most originated from individuals who used proprietary structural design computer packages without, in the view of the reporter, a sufficient understanding of the subject.

Key Learning Outcomes

For property owners, clients and commissioning architects:

  • Steel beams and other structural elements should be designed by suitably qualified and experienced (SQEP) civil and structural design engineers
  • All structural design should be signed off by an appropriately experienced chartered civil or structural engineer
  • Be aware that the adequacy of a structural design submitted to a building control body is the responsibility of the originator - do not rely on the building control review

For structural designers:

  • Designers must understand the principles of the problem at hand (and relevant design codes) before using software
  • Structural design should be undertaken in accordance with the design standards as stated in Approved Document A in England and Wales, and the equivalent in Scotland and Northern Ireland
  • Be aware that health and safety legislation places duties on all designers to ensure they do not put people at risk of harm

For building control bodies:

  • Assess compliance of structural designs against the design codes stated in Part A of the Building Regulations in England and Wales, and the equivalent in Scotland and Northern Ireland

Full Report

Find out more about the Full Report

The Full Report below has been submitted to CROSS and describes the reporter’s experience. The text has been edited for clarity and to ensure anonymity and confidentiality by removing any identifiable details. If you would like to know more about our secure reporting process or submit a report yourself, please visit the reporting to CROSS-UK page.

 

A reporter, who is a structural engineer checking building regulation applications submitted to a local authority in a UK city, has become increasingly concerned about the number of designs, submitted to show compliance with Part A of the Building Regulations, that they have judged as inadequate. Over the last year, the reporter’s attention has been drawn to the number of issues they have had to raise in respect of steel beam designs for low rise domestic extensions, loft conversions, flat conversions and similar projects. The reporter’s main concerns are that:

  • some users of computer programs 'fill in boxes' without a sufficient understanding of structural design
  • the effective length of steel beams designed under BS 449 and BS 5950 has been incorrect
  • steel beams have been designed as fully restrained when they should have been designed as unrestrained

The reporter has assessed calculations for steel beams supporting timber floors and roofs that have shown the beam as fully restrained between supports, however, when the reporter has assessed the effective length based on the full span, the beam is judged as inadequate. The reporter has also had instances of beams supporting masonry walls being designed as fully restrained when, in the reporter’s view, they should be designed as unrestrained.

The reporter says most of these issues have arisen when designers have incorrectly used proprietary structural design packages. The reporter believes a significant driver is the use of structural design programs by unqualified designers, and that it is very easy to purchase a software program and, with little training, prepare and submit design calculations. In the experience of the reporter, these concerns do not apply to steel elements designed under Eurocode 3, as there is greater guidance in this code, and it is not commonly used by unqualified designers.

The reporter goes on to say some local authorities do not deploy checking engineers and some approved inspectors only check larger schemes, and that this lack of checking could lead to incorrect designs being accepted. The reporter finds at least one incorrect design each week and considers that a similar picture across other regions would suggest many incorrect designs could be progressed every week and, ultimately, structures built in accordance with those designs.

many incorrect designs could be progressed every week and, ultimately, structures built in accordance with those designs

The greater number of issues the reporter has noticed concern designs to BS 449, which they believe is a simple code that lends itself to being used by unqualified designers. Although the reporter has also found that some experienced engineers are unaware of the revision to clause 26a of BS 449 in November 1995 concerning loading to top flanges and beam end fixity. The reporter understands that BS 449 is no longer supported by BSI and that, although not cited under Part A of the Building Regulations, building control bodies are unable to refuse its use. Nevertheless, the reporter suggests that BS 449 could be ‘deleted from use’. The reporter has found fewer issues relating to applications designed under BS 5950 as they believe Table 13 (in that code) gives clearer advice regarding the effective lengths of beams.

The reporter believes that around 30 to 40% of structural applications originate from designers who are not members of any professional engineering body and, therefore, do not become aware of design issues through journals or other related information sources. Communicating with this group of people is difficult.

The reporter makes several suggestions that they consider could improve the situation:

  1. The new Building Safety Regulator could consider a requirement that only qualified engineers should submit calculations
  2. Software packages for single beams designed under BS 449 and BS 5950 should use a default effective length of 1.2L and treat beams as unrestrained
  3. There should be more guidance relating to the lateral restraint of beams. The only material the reporter could find regarding lateral restraint to steel beams was P360 Stability of Steel Beams and Columns published by the Steel Construction Institute. This document provides details of how beams can be restrained but the reporter believes good site supervision is required to achieve the details on site
  4. The reported issues should be raised with all engineers checking building regulation applications
  5. CROSS should consider how the reported issues could be communicated to the wider industry beyond its readership

Expert Panel Comments

Find out more about the Expert Panels

Expert Panels comment on the reports we receive. They use their experience to help you understand what can be learned from the reports. If you would like to know more, please visit the CROSS-UK Expert Panels page.

The reporter is right to be concerned about the use of structural design software by persons who are not suitably qualified and experienced. Those undertaking structural design must be competent to do so. Incompetence can cost lives; the capacity of an unrestrained beam can be significantly less than when restrained, meaning that in the cases described possibly dangerous under-design could be taking place.

the capacity of an unrestrained beam can be significantly less than when restrained, meaning that possibly dangerous under-design could be taking place

CROSS receives many reports about structural designs being undertaken by persons who appear not to be competent, particularly in respect of work undertaken in the domestic residential market. CROSS-UK report 1132 - Inadequate design for basement works, published in 2022, considered issues of designer and checker competency for structural alterations to a townhouse. Persons who issue designs that are incompetent not only risk lives but are likely also not meeting their legal obligations.

The reporter suggests a requirement that only qualified engineers should submit calculations; CROSS agrees with this suggestion and once again recommends that all structural designs be signed off by an appropriately experienced chartered civil or structural engineer. Clients, and other persons appointing structural designers, must satisfy themselves as to the competency of proposed designers prior to appointment. As a minimum, a structural designer should be expected to be a member of a professional body that regulates structural designers, normally the Institution of Structural Engineers or the Institution of Civil Engineers.

a structural designer should be expected to be a member of a professional body that regulates structural designers

Some of the reporter’s concerns relate to designs undertaken to BS 449 and BS 5950. Steelwork designers should note that neither BS449 nor BS 5950 appear in any of the lists of Codes, Standards and References in Approved DocumentA. BS  449 is not maintained by BSI and BS 5950 has been withdrawn by BSI. Approved Document A does however state: ‘There may be alternative ways of achieving compliance with the requirements and there might be cases where it can be demonstrated that the use of withdrawn standards no longer maintained by the British Standards Institution continues to meet Part A requirements.’

Whilst BS 449 and BS 5950 are still used by some designers, CROSS recommends that structural design be undertaken in accordance with the current design standards as listed in Approved Document A. Superceded codes (such as BS 449 and BS 5950) are useful when assessing existing structures designed to previous codes, but new structures should be designed to the current codes since these represent best practice and are updated as required. Designers should also note that professional indemnities may only cover designs carried out to codes and standards stated in Approved DocumentA.

Use of software by persons who are not competent

The use of software to produce designs for simple steel beams may, in some cases, be happening because the user is not competent in structural design. Persons who do not know how to design a simple beam may be thinking an automated design process will take all matters into account. However, inputs determining effective lengths and buckling parameters must be determined by the user, and unless the user is suitably qualified and experienced then, unknowingly to them, their design may be unsuitable. Designers using structural design software should have enough experience and knowledge to anticipate the software outputs, for example, beam sizes, and recognise any outputs that do not ‘feel right’.

Designers using structural design software should have enough experience and knowledge to anticipate the software outputs

Inappropriate use of software by persons who are not competent has been reported to CROSS on a number of occasions, including, for example, CROSS-UK report 989 – Dangerous design of a retaining wall, published in 2021. This report concerned what would have been a dangerous structure, liable to failure during or soon after construction, designed using software.

Restraints and lateral torsional buckling

The concepts of buckling and restraint are actually quite difficult and require sound engineering judgement. Competent structural engineers know that stability relies critically on restraint conditions. Some codes use words to describe different types of restraint, whereas some designers argue that pictures would be far more helpful and intuitive. For example, BS5950-1: 2000 Table 13 does not include illustrations, although Table 14 for cantilevers does. 

As the reporter has rightfully noted, this gap is 'plugged' by guides, including P360 Stability of Steel Beams and Columns. The reporter has suggested that default settings for effective lengths and restraints in software should be set at values that would produce more conservative designs; this may be helpful in some cases but is no substitute for competency. Competent designers will also know that deflection, or other criteria, may be the overriding matter determining final beam selection rather than bending stress.

Ensure all designs are checked before submission to Building Control

Structural engineering is a safety-critical profession. Structural design, regardless of the use of software or not, must go through appropriate checking processes, either within the design organisation or an independent organisation, as part of completing validation of the design.

As very clearly illustrated by the reporter, the checking of engineering designs submitted under building control processes has many benefits including, not least, the prevention of unsafe structures being erected. Building control bodies have a responsibility to review all designs for compliance with Part A and its equivalents across the UK but, nevertheless, this does not absolve the designer from any of their responsibility to produce an adequate design.

Building control bodies have a responsibility to review all designs for compliance with Part A and its equivalents across the UK

The Institution of Civil Engineers published Submission of structural engineering data for approval under Part A of the building regulations in 2015. The paper offers guidance to structural engineers to ensure they maintain an adequate standard, examining the components of an acceptable submission to building control bodies. It looks at essential competency standards that should be followed to help engineers protect the health, safety and welfare of those in and around buildings.

CROSS will continue to take opportunities to communicate to relevant parties its wider concerns about unsuitably qualified and experienced people undertaking structural design.

Submit a report

Your report will make a difference. It will help to create positive change and improve safety.

Our secure and confidential safety reporting system gives professionals the opportunity to share their experiences to help others.