This study concerns helium smoke simulators which have been used for simulating smoke spread in aircraft cabins during inflight tests. The smoke from cabin fires is buoyant due to higher temperatures, compared to the surroundings. However, the smoke from commercially available smoke generators is nonbuoyant and an addition of helium is employed to make it buoyant so that the actual circumstance is closely represented. The dependence of buoyancy on the concentration of helium and Ihe validity of this approach to represent thermal buoyancy effects are important questions that need to be answered. In this technique, helium is added to smoke from commercially available smoke generators in order to provide buoyancy in the flow due to the lower density of helium. The mixture is then used to simulate the flow of smoke and toxic gases in the cabin of an aircraft, or in other enclosed spaces, without employing heated air to obtain the buoyancy. Since buoyancy plays a very important role in the flow and spread of smoke and hot gases, it is important that the simulation include this effect in order to closely represent the actual flow An experimental investigation is undertaken to explore the validity of simulating thermally buoyant gases, from a fire, wilh a helium and air mixture. The experimental flow consisted of a ceiling jet with a corner flow. This same configuration was used in previous studies which were directed at thermally buoyant flows. Thus, a basis for comparison between the two buoyant flows can be established. Parameters, such as the volume flow rate of the air and helium and the initial helium concentration, were varied over wide ranges. Temperature profiles from the thermally buoyant case were compared to helium concentration profiles. In addition, velocity profiles from the two studies were compared to determine the flow spread. From these results, the characteristics of the buoyant helium-air mixture were obtained. It is found that, though this mixture can be used to qualitatively model thermally buoyant flows, large quantitative differences exist between the two and must be taken into account for an accurate simulation.
Read full abstract