Dougal Drysdale

Variation of heat release rate with forced longitudinal ventilation for vehicle fires in tunnels

Many tunnels are equipped with longitudinal ventilation systems to control smoke in the event of a fire. However, the influence of such ventilation on fire development and fire spread has rarely been considered. This paper presents the results of a study using a Bayesian methodology to estimate the effect of forced longitudinal ventilation on heat release rate (HRR) for fires in tunnels. The behaviour of car and heavy goods vehicle (HGV) fires with a range of forced ventilation velocities is investigated. Results are presented and the implications are discussed. It has been found that forced ventilation has a great enhancing effect on the HRR of HGV fires, but has little effect on the HRR of car fires.

Critical heat and mass transfer at pilot ignition and extinction of a material

Traditionally, pilot ignition of a combustible solid is associated with a critical heat flux to the surface and a surface temperature necessary to produce the volatiles. More fundamentally, it can be interpreted in terms of a critical mass flow of volatiles sufficient to support a nascent flame capable of losing heat to the surface without the flame temperature being reduced to a value below which the flame is extinguished. Measurement of the critical mass flowrate at pilot ignition provides a method of quantifying ignition and extinction conditions in terms of basic properties of the solid and its decomposition products.

Nonlinear dynamics of flashover in compartment fires

Flashover is a phenomenon whereby a room fire undergoes a rapid increase in size and intensity. Such a transition is suggestive of a nonlinear process. We therefore seek to apply modern geometrical and computational techniques of nonlinear dynamics to a simplified model of fire growth to investigate flashover and other instabilities occuring in compartment fires. We present here a simplified model of fire growth in a compartment and conduct a preliminary analysis of the dynamics exhibited, in terms of both transient simulations and quasi-steady evolution manifolds, for variations in controlling parameters.

Flame spread on inclined surfaces

Abstract Experiments have been carried out to investigate the effect of slope on the rate of upward spread of flame over a combustible solid. There is a clear difference between the mechanisms of spread for thin and thick fuels. With the latter, a change in mechanisms is found to occur at an inclination of 15–20°. This is greatly enhanced if entrainment of air from the sides is restricted.

Flammability of plastics II: Critical mass flux at the firepoint

Abstract The piloted ignition of five common thermoplastics and three fire-retarded modifications has been studied by exposing horizontal samples to irradiance levels in the range 12 – 35 kW/m 2 . By monitoring the sample weight continuously, it has been possible to determine the minimum rate of evolution of decomposition products necessary to support sustained flaming. The results are significantly lower than values reported elsewhere: it is suggested that this may be attributed to a lower convective heat transfer coefficient at the exposed surface in the present apparatus.

Numerical modelling of early flame spread in warehouse fires

Abstract An important European research initiative ‘Major Hazards Arising from Fires in Warehouses and Chemical Stores’ has recently concluded; Edinburgh University has contributed both experimental and theoretical determinations of the rate of upward flame spread on corrugated fibreboard packaging. The theory presented here is a development of a recently published thermal model of upward flame spread; the model is in the form of a Volterra integral equation which is solved for the time-dependent flame spread velocity. The main advance over the previous theory lies in the modelling of burnout behind the pyrolysis region. The original theory, with no burnout, always predicts self-extinguishment of the flame front when a realistic, empirical, non-linear flame height correlation is adopted; the inclusion of burnout in the new model overcomes this restriction and allows the flame spread to proceed indefinitely at a steady-state. The numerical solution algorithm of the model is particularly attractive since it permits heat release data from cone calorimeter tests to be used directly as input. The FORTRAN coding of the model includes routines which perform the necessary I/O sequences automatically; some minor post-processing of the cone calorimeter data is occasionally required prior to running the model but this too has the potential for future automation.

Heat transfer analysis of the composite slab in the Cardington frame fire tests

Abstract The structural modelling of the Cardington Frame fire tests as part of the Department of Environment, Transport and Regions funded Partners in Technology project has highlighted the importance of the temperature evolution both temporally and spatially in determining the structural response. Restrained thermal expansion/contraction and thermal bowing are the main driving force behind almost all the structural phenomena witnessed in the tests. The four British Steel fire tests carried out on the 8-storey composite steel and concrete building at Cardington have provided a wealth of information about the temperatures in the fire atmosphere and the protected and unprotected steel. Unfortunately, there is considerably less information on the temperatures attained in the concrete slab. In Tests 1–3, the temperatures through the depth of the slab have been recorded only at a few points and in terms of the structural modelling this has been just about adequate. There were no temperatures recorded in the slab in Test 4 (Office demonstration test). The finite element, adaptive heat transfer program HADAPT has been used to model the heat transfer to the composite steel and concrete slab. HADAPT is a 2D adaptive heat transfer code capable of carrying out a nonlinear, transient, thermal analysis. The code models moisture evaporation from the pores of the concrete by assuming a phase change in the region of 100°C. The measured concrete temperatures in Tests 1–3 have been used to calibrate the model which has then been used to predict the slab temperatures in Test 4.

An experimental evaluation of critical surface temperature as a criterion for piloted ignition of solid fuels

Abstract With a view to developing a simple engineering method for the prediction of piloted ignition, the validity of the critical surface temperature criterion for piloted ignition is examined experimentally for seven thermoplastic materials. The results indicate that the surface temperature at piloted ignition for each material studied varies by ±15 K or less. As such, the surface temperature criterion appears to be suitable for engineering calculations. Analysis of time-surface temperature histories shows that the radiant heat source temperature has a significant effect on the material heating over the range of source temperatures utilized (700–1050 K). The variations in material heating are of sufficient magnitude to cause changes in ignition times by a factor of two or more for different heat sources present in typical fire scenarios. Our current level of understanding of piloted ignition is shown to be insufficient to support extrapolation procedures to determine the minimum incident radiant flux required for piloted ignition. An experimental approach to determination of the minimum radiant flux required for piloted ignition is demonstrated to be feasible.

Using cone calorimeter data for the prediction of fire hazard

The Cone Calorimeter is an apparatus capable of providing quantitative data which are relevant to the response of combustible materials to fire conditions. Several fire models have been developed to use such data to predict various aspects of fire behaviour. This paper considers how these data are used to model upward spread of flame on flat vertical surfaces and proposes that a modified test procedure may be more appropriate.

Experimental and theoretical models of flashover

A comparison between experimental data and both the qualitative and quantitative nonlinear behavior predicted by a fire growth model is made for a model compartment (0·4 m cube). The predictions are in relatively good agreement with the data, and are consistent with the expected nonlinear behaviour and in particular the occurrence of flashover jumps.