Limiting conditions for a sustained flame over condensed fuels : Analysis by experiments and stagnant layer theory

Abstract: Ignition and extinction characteristics of a material determine whether a fire will be initiated and grow. In fire safety engineering, where modellers deal with a multitude of materials, simplified thresholds are commonly used to determine when ignition and extinction occur.Such a threshold is the critical mass flux, which describes when sufficient fuel vapour is produced to support sustained flaming.A related criterion is the critical heat release rate.This thesis is an investigation of how these two thresholds are correlated to environmental and fuel specific properties. This assessment is accomplished with small-scale experiments and stagnant layer theory. Experimental equipment were fine tuned to enable quantification of each investigated influence on the threshold criteria. A cone calorimeter sample set-up was modified so that the sample material could be exposed to a more uniform heat flux and the influence of external heating on the critical mass flux could be better quantified. Porous gas burners have been used previously to emulate burning of condensed fuels. In this thesis, such burners were used so that mass transport at the limiting points could be held at a constant rate. A set of burners with different diameters were used to provide insight on the influence of the convective heat transfer coefficient on the critical mass flux. Both the cone calorimeter test results and the gas burner results yielded information about the chemical heat of combustion of the fuels. It is shown that the critical mass flux, but not the critical heat release rate, is a function of the chemical heat of combustion. The thresholds are also dependent on the sample size, which is seen in the correlation for the convective heat transfer coefficient. Although the external heat flux influences the mass loss rate history it does not significantly change the amount of pyrolysates needed for sustained flaming. The findings are summarized in an engineering correlation that can be used to find approximate values for fire modelling.