This work forms the first phase of a continuing initiative aimed at reducing fire deaths in residential dwellings in New Zealand (NZ). Loss of life in residential buildings dominates NZ annual fire death statistics. Few items within these buildings have the potential to bring about untenable conditions as swiftly as upholstered furniture. It is a major goal of safety research - and this work in particular – to better assess the hazard of furniture fires. Especially, in respect to our ability to predict this hazard. The heat release rate of a burning item is acknowledged as the most important property in fire hazard analysis. As a starting point, this work includes a critical review of reaction to fire calorimetric techniques. These techniques are the basis of heat release rate measurement. The technique of oxygen consumption calorimetry is subjected to a comprehensive uncertainty analysis. This includes a detailed example of the application of this analysis to a common Standard Test Method. A less favoured calorimetry technique based on thermochemistry is redeveloped. Its usefulness as a calibration tool in respect to oxygen consumption calorimetry is explored. This is helpful as the thermochemistry technique is independent of oxygen concentration measurement, which in tum is the crucial parameter in oxygen consumption calorimetry. The combustion behaviour of dozens of small-scale furniture composites and 13 full-scale furniture items are tested using the above principles. The experimental programme used the newly commissioned cone and furniture calorimeters. The characterisation of these apparatuses appear in this work. The experimental results are used to validate the applicability of widely published European furniture fire models, to NZ items. This study shows that these existing techniques, while comprehensive, do not predict with goodness the combustion behaviour of NZ furniture. However, the NZ data set is small and the direction of future initiatives are detailed.
展开▼
