Citation

Woodburn PJ, Britter RE. Fire Safety J. 1996; 26(1): 35-62.

Copyright

(Copyright © 1996, Elsevier Publishing)

DOI

unavailable

PMID

unavailable

Abstract

This paper concerns sensitivity studies of computational fluid dynamics (CFD) simulations of a fire in a tunnel. The simulations were of an experimental fire in a tunnel carried out by the Fire and Thermofluids Section of the Health and Safety Executive, Buxton, UK. The fuel used in the experiment was kerosine, the heat output rate was 2.7 MW and the tunnel was longitudinally ventilated. During the period of the experiment studied in the simulations, there was an upstream layer of approximate length 11 m. Simulations were carried out for two areas of the tunnel: the area around the fire and the area downstream of the fire. This paper describes the simulations of the area around the fire, whilst the accompanying Part II paper describes the area downstream of the fire. In the fire area simulations, the upstream propagating smoke layer length was found to be sensitive to the ventilation velocity, the ventilation velocity profile, the turbulence model used and the heat input rate. This case, in which the fire did not extend over the width of the tunnel, gave an upstream layer at higher ventilation velocities than those found in the literature. While reduction of the heat input rate to allow for radiative heat transfer from the flame caused a significant change in results, neither radiative heating of the tunnel ceiling nor the distribution of the fuel across the fuel pan had a significant effect on the results.