newsteelconstruction.com

NSC Archives

Projects and Features

What is fire resistance?

Graham Couchman of the Steel Construction Institute charts the history of fire resistance in structural engineering, and concludes that designs, in the most part, have resulted in safe buildings, as failure due to fire is a rarity.

Introduction

Despite the fact that post-Grenfell much traditional practice is being questioned, hence the numerous recent articles in New Steel Construction, designers in the steel construction sector are very familiar with the concept of designing to achieve fire resistance. Normally sufficient passive protection is applied to ensure that ambient temperature design governs – there is no need to explicitly design for the fire limit state with reduced loads and reduced material properties. In this article Dr Graham Couchman of SCI considers what the concept of fire resistance actually means, in particular the use of standard time-temperature curves and standardised resistance periods, and why we try to achieve them. He concludes that whilst the concept of fire resistance is well established and easy to use, we should not be closed to considering other approaches.

The paper The rise and rise of fire resistance by Angus Law and Luke Bisby¹ provided much of the background presented in this paper, and is gratefully acknowledged. It was published in the Fire Safety Journal and the intention here is to take that knowledge to a wider/different audience, for whom it is equally relevant and valuable. Input from Dr Craig English of Semper is also gratefully acknowledged.

Some background

A key time in the development of the concept of fire resistance was 1903, when following a Fire Prevention Congress in London a paper was published that contained four key concepts. Firstly that the term ‘fire resisting’ was more appropriate for use in construction than ‘fire proof’. Secondly that systems should be classified according to whether they provided ‘temporary’, ‘partial’ or ‘full protection’. This concept was extended to the third concept of time periods, with resistance for at least 45 minutes, 90 minutes and 150 minutes respectively. Finally, it was proposed that fire testing should be standardised, in terms of duration of exposure, minimum temperature, required loading, and minimum specimen size.

‘Full protection’ has been interpreted as meaning the structure could survive burn-out of the fire compartment’s contents without intervention by fire and rescue services. Options for lower levels of protection were recognised as being practically (commercially) necessary. At the time these definitions were based on a combination of test and real fire experience, which may be a critical point where blurring between real situations and standardised tests started to occur. An obvious example is that the standard time-temperature curve we use in most testing today has temperature that increases up to an asymptote, whereas if contents have burned out then clearly at some point the temperature will start to drop. In 1928 Ingberg made an attempt to link the severity of a real fire to an equivalent period of exposure in a standard fire test – the concept of ‘equivalence’, which was recognised at the time as having limitations.

Figure 1: ISO standard ‘fire curve’

Legislation took hold of these concepts, and a century later they are still being widely used. Perhaps this is due to a lack of practical alternatives, but it is still very important to recognise the limitations of such an approach.

Application today

The background summary given above illustrates that the whole area of design for the fire limit state is a bit messy and confused. That confusion seems to be exacerbated in the minds of many by a further blurring, namely that between Building Regulations and Approved Documents (or their equivalent in other nations). Approved Documents were introduced in 1985, and provide ways in which compliance with the Regulations can be demonstrated, for example by testing a specimen in a standard fire test and achieving a stated resistance period. But Approved Document provisions are not the only way of showing regulatory compliance, and indeed in some cases they may even be inappropriate. In the past two years we have seen this dis-joint in the context of load bearing light steel framed walls – Approved Document B2 (AD-B) requires/allows such walls to be tested with a one-sided fire, but clearly some such types of wall could be exposed to two-sided fire (Figure 2) and simply satisfying the AD-B provisions is now recognised as not then being appropriate3.

Figure 2: Walls in red would be exposed to fire from one side only. Walls in blue could be exposed to fire from two sides.

The fact that periods of resistance recommended in AD-B vary according to building type seems sensible if they have a relationship with burn out of compartment contents. The fact the resistance period increases with building height appears to be illogical if a relationship with burn out is claimed – an apartment in a multi-storey building will not contain more calorific content than one in a three-storey structure so why does the resistance period go up? Law and Bisby suggest this may have less to do with logic and more to do with harmonising different regulations. However they also note that whether those creating the recommendations appreciated it or not, the adoption of longer periods for taller buildings does increase the effective ‘factor of safety’. There is logic to ensuring that taller buildings are more resistant to fires that are not ‘average’, because of the consequences.

It is worth adding that when sprinklers are provided the resistance period may reduce.

Alternatives and possible developments

The approach described above has been criticised for several obvious reasons:

  • The standard heating curve does not look like a real fire, particularly its lack of a cooling phase.
  • Test furnaces are difficult to control, and the thermal and mechanical boundary conditions are unrealistic.
  • The ‘equivalence’ method fails to take into account a number of relevant factors.

Perhaps less obviously, it has long been understood that methods given in typical guidance (AD-B, BS 9999, BS 9991)2,4,5 provide no explicit measure of building fire safety. The same is true of the deterministic approaches set out in fire engineering codes, such as BS 79746. Not knowing what safety level one’s fire design provides is the reason why the Hackitt review recommended the use of outcome-based approaches, and why safety cases are now being prepared for tall residential buildings in order to determine which of them is in a potentially unsafe condition (despite having quite possibly satisfied regulatory requirements).

For the currently very topical case of car parks, a simple alternative would be to consider the heat release rate of different vehicles and how the fire may spread between them7, and then be able to more accurately quantify the consequences such fires may have on the structure. Those consequences would lead to more informed decisions concerning the level of fire protection, if any, that is required to satisfy life safety, property and environmental objectives.

Despite the obvious logic and potential benefits, rather than the approach described above it seems likely that future developments in AD-B may include extending fire resistance periods for open sided car parks, and/or requiring sprinklers to reflect the greater fire risks associated with modern vehicles. A requirement for the use of sprinklers could reflect re-consideration of the purpose of Building Regulations – moving towards protecting assets as well as achieving the current objective of saving lives.

More complex fire engineering methods take a more realistic view of how structures behave in fire, not only in terms of fire load but by allowing for variables such as the size of compartments and their ventilation, and the criticality of different structural elements when considering time to failure. Risks should also be assessed in the context of the exit strategy for occupants, access for fire and rescue services etc. Significant savings may be made when such an approach is used, and some structures will more than warrant this level of investment in design.

Conclusions

Design using standard fires and resistance periods is convenient, and it could be argued that this approach has been shown to produce appropriate structures given that structural failures in fire remain a rarity. It is important however that designers, specifiers, clients and other stakeholders recognise that achieving a certain fire resistance period in a standard test in not always necessary or even appropriate. As we try to construct more ‘carbon efficient’ structures we should not be content to always use approaches we know to be conservative.

References

  1. The rise and rise of fire resistance. Law and Bisby. Fire Safety Journal Vol. 116, September 2020
  2. Fire safety: Approved Document B. Gov.uk
  3. P442: Design of loadbearing light steel walls exposed to fire on two sides. SCI, 2024
  4. BS 9999:2017. Fire safety in the design, management and use of buildings. Code of practice. BSI.
  5. BS 9991:2024. Fire safety in the design, management and use of residential buildings. Code of practice. BSI.
  6. BS 7974:2019. Application of fire safety engineering principles to the design of buildings. BSI
  7. Open car parks in fire. Ozcelik. New Steel Construction, September 2024.

Share this post

Related Posts

THIS MONTH'S MAGAZINE

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

Archives