Abstract
Yields of toxic combustion products from burning materials vary considerably between different stages and types of fire, eg fully ventilated, vitiated. It is important that any laboratory studies of such phenomena be carried out as close as possible to real fire conditions. The Purser furnace has been developed to widen the range of fire types amenable to laboratory-scale investigation. This paper describes Salford?s Purser furnace and reports the findings of a systematic assessment of the experimental parameters that affect the performance of the Purser furnace.