Fire Fighting

A further myth is that fire brigades will not enter a building clad with composite panels. This is totally untrue, a situation confirmed by our discussions with fire-fighters. The reality is that the fire brigade will perform a dynamic risk assessment on arrival at the fire - the result of which will determine their actions. Obviously, panel system collapse and internal flashover is the biggest fear of fire-fighters - this is a particularly important consideration in relation to internal panel systems.

The reality is that buildings clad in structurally supported (mechanically-fixed) external panels do not present a specific hazard, as the panels will not collapse until the structural steelwork frame collapses.

In an article published in Fire Prevention and Fire Engineers Journal in December 2002 the UK’s Building Research Establishment state ‘Fire statistics show that external claddings constructed from sandwich panels tend not to be a major fire risk, particularly if the chances of an arson attack can be reduced and its effect minimised.’

Center Parcs Fire

Recent misreporting, led to another damaging myth that the estimated £100m Center Parcs fire - possibly the largest fire loss in the UK for many years - was caused by composite panels, which was patently not the case. The true position was clarified by Paul Hayden, Assistant Chief Fire Officer of Suffolk County Fire Service in a report in the August 2002 issue of Fire Prevention and Fire Engineers Journal;

“The key contributory factor to structural fire spread was the extensive flat roof construction. Although investigators describe it as ‘sandwich’ construction, it was not constructed of pre-formed sandwich panels, but a built-up layer of plasterboard internal lining, polystyrene insulation, timber boards and topped off with roofing felt and bitumen”.

Sahib Foods
- Architects Professional Indemnity Cover

In a recent high court judgement on 3rd March 2003 the architect responsible for specifying the use of polystyrene (EPS) cored panels in a high risk cooking area has been found liable for the losses that occurred in a subsequent fire. The cost to Sahib Food’s insurers, Norwich Union, at the time was £17M.

The facts of this case are clear. Polystyrene panels were specified and used throughout the premises of Sahib foods in a refurbishment carried out in 1994 to 1995. A fire occurred in January 1998 that spread rapidly throughout the factory causing a total loss and led to the closure of the business. The source of the fire was a cooking area that involved frying operations using gas fired equipment.

One of the key issues influencing the judgement was that the architects conceded that at the time of the specification in 1994 they knew the same about the fire performance of composite panels as they know at the present time. In his judgement HHJ Peter Bowsher noted: ‘It is agreed that in 1994, the year with which we are concerned, knowledge of the risk of the use EPS panels in relation to food factory fires was developing in the architects profession……’.

This judgement was based on the particular facts of the case. The case relates only to the specification of polystyrene panels in cooking areas - universally acknowledged as high risk processes for many years. There have been a relatively large number of fires in similar circumstances in food factories. Statistics published by the Building Research Establishment state ‘Fire losses in food factories over the period 1991 to 1999 represent an average loss per year of £27M. Most of the food factory fires over this period resulted from cooking risks or malfunction of equipment’. It is interesting to note that insurer losses in food factory fires in 1998 and 1999 were £46M and £64M respectively. We believe that the vast majority of these losses were in factories with internal polystyrene panel systems.

In stark contrast - the historical fire performance and insurance loss statistics relating to the use of Kingspan polyurethane and polyisocyanurate panels in the external envelope has been excellent. There is no evidence whatsoever to suggest that Kingspan panel systems are not fit for the purpose that they are recommended and there is no link with higher insurer losses. This is proven by detailed insurer loss statistics and case studies on fires in buildings clad with polyurethane panels. In addition to fire safety the specification of Kingspan panels gives a number of other important benefits to the client.

The implications for Architects’ PI insurance cover are unclear but it important to understand that the judgement refers only to polystyrene panels in a high risk food sector applications.

The Non-Combustibility Myth

In recent years huge confusion has been caused by panel manufacturers and their raw material suppliers making claim and counter claim about the non-combustibility or otherwise of their own and their competitors panel systems and core materials. The simplest way of looking at this issue is to address the insulation core and the panel system separately.

The Insulation Core

The definition of non-combustibility according to British Standards can be found in BS 476 Part 4. This is a small scale furnace test where the sample is burnt at a high temperature and contribution to temperature rise and flaming is assessed. Polystyrene, polyurethane, polyisocyanurate and most rock fibre cores used in composite panels are believed to be combustible according to this test.

Rock fibre can be defined as combustible because of the relatively high levels of organic binders used to glue the fibres together.

The Panel System

In the case of polystyrene and rock fibre panels adhesives are used to adhere the steel facing to the insulation. In rock fibre panels the adhesive is normally polyurethane and levels of adhesive can be quite large to ensure a good bond and minimise the risk of delamination. As a direct result panels with a rock fibre core cannot be rated as completely non-combustible.

Section cut through the thickness of the panel after 30 minutes exposure showing that the char occurs only in the area of direct flame impingement and there is no fire propagation.

Ignitability of Insurer Approved PIR Insulant

One concern raised by insurers is the potential for exposed insulation in ‘real’ situations. The reality is that any insulation panel system is likely to be damaged or modified after installation and this may lead to exposure of the core – and example would be where a hole has been cut in the panel for new pipework, ductwork or electrical services. Such modifications are often not finished correctly and are perceived to provide an easy entry point for fire. The question is how big is the risk?

The fact is that insurer approved PIR is very difficult to ignite.

The images above cover the effect of a high intensity propane torch on exposed PIR after the metal facings have been removed.

The propane torch generates a temperature of over 1000°C. The images show the effect after 30 minutes exposure and are a good demonstration of how the PIR forms a strong carbonaceous char that protects the insulation core from ignition. When the burner is removed after 30 minutes the core self extinguishes.
Good practice dictates that any panel damage is repaired or site modifications are installed correctly – however, the propane torch test does demonstrate that insurer approved PIR does not ignite and the exposed core does not present a fire hazard.

Combustibility of Installed Panel Systems

The reality is that all panel systems have varying levels of combustibility because all contain combustible materials. The only reliable way of assessing panel system combustibility is to base a judgement on testing the entire system. Tests and standards such as LPS 1181, LPS 1208 and FM 4880 are ideal in this respect.

The varying levels and grades of performance defined by these tests allow the specific panel system performance to be matched to the specific risk.

The real issue is not directly related to the combustibility of the panel system or core - all have combustible elements. The real issue is related to how a specific system will perform in a real fire scenario and whether it acts as a non-combustible building element by not contributing to fire propagation.