doi:10.3850/978-981-08-7724-8_11-05

ABSTRACT

The majority of lives lost in fires result from inhalation of toxic fire effluents. Estimation of the build up of toxic gases is required for performance based design. Preliminary modelling work using FDS 5 is reported for the room corridor test. This involved sensitivity analysis using the range of parameters found for wood. The greatest sensitivities were found for the gas temperature at different locations, the CO and CO2 concentrations and the heat of combustion. All other dependencies seem to be without deterministic content or strongly non-linear, suggesting that errors in the value for the heat of combustion will generate a significant error in the predicted temperature. It is therefore important to determine this property accurately, in conditions appropriate to a real fire. For the other six input parameters, the influence on the results is not as clear as for the heat of combustion. Even though the accurate prediction of fire toxicity is not yet state of the art, best estimates add an important dimension in the fire protection design of buildings.

Keywords: Toxicity, Simulation, FDS, Hazards, Sensitivity.

© 2011 CPD Unit, University of Leeds