A reduced-scale mass fire experiment

Abstract A previous correlation of flame heightdata has suggested that continuous flaming may not be possible over flat continuous fuel beds when the parameter q ″ D 1 2 (or, more appropriately, its nondimensional parent) is smaller than a limit, a condition possibly important in some mass fire situations. In this parameter, q ″ is the heat release rate per unit area and D is the diameter, or effective diameter, of the fuel bed. Below the limit, which can be interpreted as a lean flammability limit, the flame cover is expected to break up into flaming and nonflaming areas. The flame breakup hypothesis was examined in a fire experiment with a 7.32-m-square, flat, and horizontal array of wood-fiber boards impregnated with a controlled amount of heptane as ignitor. The flaming pattern began developing nonflaming areas near the predicted value of q ″ D 1 2 , 21 s after ignition, as the fire decayed from an initial state of relatively high intensity. With the aid of auxiliary experiments it was possible to show that the extinguished areas in the flaming pattern were due to dynamic effects of the fire and not attributable to local burnout of the fuel bed.