Observations of Large Scale Turbulence in Corner-Wall Experiments

Abstract

Abstract It is recognized that small scale test methods are unable to predict the fire growth potential of interior finish materials. This has led to an interest in the use of full scale room fire tests. One of the reasons for the effectiveness of these full scale experiments is their ability to reproduce a turbulence similar to the one expected under actual fire conditions. This paper describes some of the flame morphologies observed in full scale fire experiments, and their origins are ascribed to certain turbulent conditions. The corner placement of the ignition source is shown to produce a more severe exposure to the walls than a side wall exposure. The spread of fire to the ceiling is shown to follow the corner between the two walls, and the subsequent flame spread at the ceiling is in the narrow region at the intersection between the walls and the ceiling, termed in this paper the "T" pattern flame morphology. In cases where the walls and/or the ceiling are combustible, it has been observed that the "T" pattern begins to burn first, and it is only afler this "T" burning is established that the remainder of the ceiling is ignited and begins to burn. A special flame morphology is shown which appears to be important to the flame spread under the ceiling of a compartment. A mechanism is presented lo explain the observed morphology in which a vortex structure originally formed on one wall is interrupted by fluctuations in the buoyant plume at the corner, is captured by the flow on the other wall, and is then swept along with the flow at that wall. The significance of this turbulence is discussed, and it is suggested that it be taken into account when developing compartment fire models.