An experimental study of the self-extinction mechanism of fire in tunnels

Abstract

It is widely accepted that temperature decay along tunnels in fire conditions is mainly due to the heat loss to the surrounding walls and the ceiling. A previous study found fires could self-extinguish in a 1:20 reduced-scale tunnel made of steel plates and fire-proof glass and measuring 20.8 m long, 0.45 m wide and 0.23 m high (Wang et al., 2019). Self-extinction was presumed to be caused by significant heat loss at the boundaries, which led to the descent of the smoke layer and subsequently blocked the supply of fresh air. A series of experiments with fires with heat release rates of 2.8 kW, 5.6 kW, 11.2 kW and 16.8 kW were conducted in a tunnel made of fireproof board and measuring 20.8 m long, 0.45 m wide and 0.23 m high. As the heat loss at the boundaries was negligible, it was expected that self-extinction of the fires could not occur. However, self-extinction of the fires was observed, contrary to the predictions made. Furthermore, the times to reach self-extinction for the fires at different heat release rates in the two tunnels made of materials with distinctive thermal properties were basically consistent. By analyzing the velocity at different distances and scrutinizing the flow patterns using a laser sheet, it was found that the smoke velocities decreased with an increase in distance away from the fire and reduced to approximately zero at 8–9 m, irrespective of the fire sizes. While the hot smoke moves toward the portal driven by buoyancy, the same amount of makeup air flows back to the fire seat as smoke or makeup air flowing through the portals are approximately zero. The flow in the tunnel is longitudinally bi-directional and internally circulated, which persists throughout the experiment and eventually leads to the self-extinction of the fires.