Effect of blockage-fire distance on buoyancy driven back-layering length and critical velocity in a tunnel: An experimental investigation and global correlations

Experiments are conducted in a model tunnel to study the effect of a vehicular blockage at the upstream of the fire source on the buoyancy driven back-layering length and critical velocity in a longitudinal ventilated tunnel. The relative distance between the vehicular blockage and fire source are ranged in 1 m ∼ 6 m. It is found that with no blockage, experimental data on back-layering length and critical velocity can be well collapsed by Wu model and Li model, respectively. However, with the increase in blockage-fire distance, both the back-layering length and critical velocity first decrease then approach to constants similar to those with no blockage. The modified Wu model and Li model amended by cross-sectional blockage ratio proposed by Lee, which does not include the factor of blockage-fire distance, still fail to predict experimental results for different blockage-fire distances. Thus a dimensionless modification coefficient η is proposed and correlated non-dimensionally with the normalized blockage-fire distance to account for this effect. Finally, global models are proposed to predict back-layering length and critical velocity including factors of both cross-sectional blockage ratio and blockage-fire distance, which are shown to well collapse the experimental measurements in good agreement. •Fire experiments carried out in a 72 m long reduced scale model tunnel.•Critical ventilation velocity data are obtained with blockage effect.•Smoke back-layering length data are obtained with blockage effect.•Blockage-fire distance factor is included to modify previous models.•New global models developed in good agreement with the measured data.