The trench effect is a combination of two well-understood but separate ideas: the Coanda effect from fluid dynamics and the flashover concept from fire dynamics.

The Coanda effect is the tendency of a fast-moving stream of air to deflect towards nearby surfaces. The fast-moving stream tends to experience a decrease in static pressure, which creates a pressure difference between those areas far from the wall and the wall itself. This bends the fast-moving stream towards the surface and tends to keep it attached to that surface.

Flashover is something that occurs when the majority of surfaces in a space are heated to the point at which they give off flammable gases that are hot enough to ignite themselves. Prior to flashover, flammable gases may be given off but are not hot enough to ignite themselves.

The trench effect occurs when a fire burns next to a steeply-inclined surface. The flames lie down along the surface, in accordance with the Coanda effect. The flames heat the material farther up: these emit gases, reach their auto-ignition temperature and then start burning, in accordance with flashover theory. The flames from these areas are themselves subject to the Coanda effect and blow a jet of flame up to the end of the inclined surface. This jet is sustained until the fuel is exhausted.

Background

The trench effect became known as a consequence of the scientific investigation of the King's Cross fire. The fire started on an escalator between the Piccadilly Line platforms and the ticket hall at Kings Cross St. Pancras tube station. The evidence of many eyewitnesses indicated that at the early stages, the fire in the escalator was obviously of manageable size: officers from the Fire Brigade and British Transport Police gave evidence that the fire appeared no larger than a large cardboard box burning. Those present were all taken by surprise when it suddenly changed into something that threw a sustained jet of flame into the ticket hall.

What seems clear is that in the early stages of the fire, the flames visible to anyone not standing on the burning escalator were a small part of the full story. The majority of the flames were lying down in the escalator trench; only a few protruded above the balustrade and were visible to observers. The lack of visible flames lulled the emergency services into a false sense of security. When the treads of the escalator flashed over, the size of the fire increased exponentially and set most of the ticket hall alight.

References

• Y. Wu & D. Drysdale, Study of upward flame spread on inclined surfaces HSE contract research report no. 122, 1996. (http://www.hse.gov.uk/research/CRR_htm/) ISBN 0717612899
• K. Moodie, The King's Cross Fire: Damage Assessment and Overview of the Technical Investigation Fire Safety Journal, vol 18 (1992) 13-33
• S. Simcox, N.S. Wilkes & I.P. Jones, Computer Simulation of the Flows of Hot Gases from the Fire at King's Cross Underground Station Fire Safety Journal, vol 18 (1992) 49-73
• K. Moodie & S.F. Jagger, Results and analysis from the scale model tests Paper presented at I Mech E seminar, The King's Cross Underground Fire: fire dynamics and the organisation of safety 1 June 1989; ISBN 0852987056
• A.F. Roberts, The King's Cross Fire: a correlation of the eyewitness accounts and results of the scientific investigation Paper presented at I Mech E seminar,The King's Cross Underground Fire: fire dynamics and the organisation of safety 1 June 1989; ISBN 0852987056

pl:Efekt okopowy