International Journal on Architectural Science, Volume 5, Number 4, p.91-98, 2004 91 ASSESSING FIRE BEHAVIOUR OF COMMON BUILDING MATERIALS WITH A CONE CALORIMETER C.L. Chow DCL Consultants Limited, Hong Kong, China W.K. Chow, N.K. Fong, Z. Jiang and S.S. Han Department of Building Services Engineering, The Hong Kong Polytechnic University, Hong Kong, China (Received 30 August 2004; Accepted 10 September 2004) ABSTRACT Fire behaviour of selected samples of building materials including wood and plastic materials such as polyvinyl chloride (PVC) and poly(methyl methacrylate) (PMMA) commonly used in the market were assessed. Both thermal aspects and smoke toxicity were studied by testing the samples of wood, PVC and PMMA in a cone calorimeter. PVC was found to be very toxic as it has the smallest value of smoke potency. Not much smoke was given out in testing wood with a cone calorimeter. PMMA has a higher value of smoke potency, appeared to be not so toxic. A flashover heat flux at floor level of 20 kWm -2 was applied. It was found that under such heat flux, both wood and PMMA were ignited, but PVC was very difficult to ignite. Therefore, the thermal effects of burning PVC by an accidental fire would not be so bad. However, when the heat flux was increased to 50 kWm -2 , smoke would give problems upon ignition of the materials. 1. INTRODUCTION Combustible materials such as polyvinyl chloride (PVC), Poly(methyl methacrylate) (PMMA) and wood are widely used as building materials and consumer products. The fire behaviour of burning those materials should be watched. Although PVC materials are much more difficult to burn, earlier studies on smoke toxicity of those samples [1-3] indicated that burning PVC will give a very low value of toxic potency, denoted by LC 50 . On the other hand, though PMMA is easier to burn, it has a higher value of LC 50 and appeared to be not so toxic. Not much smoke was given out in testing wood with a cone calorimeter. This point should be considered carefully as the number of arson fires appears to be increasing. Obviously, full-scale burning tests [e.g. 4,5] are good for understanding the actual fires. However, huge resources are required for studying so many fire scenarios. Another alternative is to study the fire behaviour of the samples by a cone calorimeter [e.g. 6,7]. Both thermal aspects and smoke emission can be assessed. The results of the heat release rate per unit area of the component materials can be used to deduce the heat release rate of the actual arrangement, say in a retail shop, using the theory available in the literature. Convolution theorem [e.g. 8,9] can be applied in studying furniture fires from the heat release rate per unit area curves measured from a cone calorimeter with some assessed burning areas deduced from the furniture calorimeter. Smoke measurement and gas analysis would give both information for designing fire detection systems [e.g. 10] and estimating the smoke toxicity in real fires [e.g. 11,12]. With advanced instruments such as Fourier Transform Infrared Spectrometer (FTIR) [e.g. 13], toxic species in smoke can be measured and quantified upon burning the materials. It is observed in some past fires that the burning of those combustibles under a post-flashover fire would be very different from that before flashover [e.g. 14]. There are interests in assessing the fire behaviour of those materials. Testing the samples under high heat fluxes might give a very different picture and such tests would be useful in fire hazard assessment. For example [9], the floor materials of a room might be exposed to a heat flux of 20 kWm -2 , the vertical wall material to 35 kWm -2 and the ceiling mounted material to 50 kWm -2 . In addition, the results from smoke measurement should also be useful in assessing toxicity [e.g. 11,12]. In fact, the fire behaviour of plastic materials for consumer products should be studied at least by a cone calorimeter as proposed earlier [15,16].