Improving fire modelling is a key issue to design efficient safety measures for a safe people evacuation in case of fire. Such an analysis should consider the different impacts of fire on people as temperature, visibility but also toxicity. Most of the standard curves used in tunnel fire studies are based on quite old fire tests without any detailed toxic gas qualification. Very few fire tests were published in that way. Based on those few tests, some standard fire emission factors are available in the literature. The objective of this paper is to review those emission factors considering the different toxic species and dealing with using recent cars. A method is then proposed to define a carbon monoxide equivalent emission factor to consider the different species through their specific threshold. Such an approach can be easily introduced into fire codes. To meet this objective, two series of tests were performed. The first concerns individual combustible materials of cars as plastics and tyres. The second focusses on full car burning tests including a detailed smoke analysis. Those two series of tests lead to an analysis of the smoke toxicity and a comparison of emission factors with standard ones.

No abstract

In the event of a fire, composite pressure vessels behave very differently from metallic ones: the material is degraded, potentially leading to a burst without significant pressure increase. Hence, such objects are, when necessary, protected from fire by using thermally-activated devices (TPRD), and standards require testing cylinder and TPRD together. The pre-normative research project FireComp aimed at understanding better the conditions which may lead to burst, through testing and simulation, and proposed an alternative way of assessing the fire performance of composite cylinders. This approach is currently used by Air Liquide for the safety of composite bundles carrying large amounts of hydrogen gas.

When a fixed roof storage tank is surrounded by a pool fire, and when the vents are not correctly dimensioned, the inside pressure can raise until the mechanical limits of the structure. The aim of this study is to develop a tool that will allow to follow the variations of the pressure and temperature in the gas phase, as a function of the size of the vent. This will help the sizing of new tank's vents or evaluating the risk of already build structures. The model also gives the temperature's distribution in the liquid phase, at each time step. In case of undersized vents, this information will be convenient to describe the phenomenological aspect of the accident after the rupture of the tank, i.e., to qualify and quantify the fire ball generated and its effect on persons and structures. Currently, if few analytical models exist to quantify the fire ball and its effects after the rupture, none are able to describe with accuracy the prerupture phenomena. The tool developed takes into account the two phases of the accident and allows to study complete and consistent scenarii. © 2010 American Institute of Chemical Engineers Process Saf Prog, 2011