Suppression of bluff-body stabilized diffusion flames

Abstract

The stabilization and suppression of a nonpremixed flame formed behind a backward-facing step in a small wind tunnel have been studies by impulsively injecting a gaseous fire-extinguishing agent (bromotrifluoromethane) into the airflow. Methane issued from a porous plate downstream of the step to simulate a pool fire in the aircraft engine nacelle. As the mean air velocity was increased, two distinct flame stabilization and suppression regimes were observed: rim-attached wrinkled laminar flame and wake-stabilized turbulent flame. In both regimes, as the agent injection period was increased at a fixed mean air velocity, the critical agent mole fraction at suppression decreased. In the rim-attached flame regime, the total agent mass at suppression was nearly constant at a fixed air velocity nearly independent of the agent mole fraction, injection period and step height. In the wakestabilized flame regime, the turbulent mixing process of the agent into the recirculation zone behind the step essentially determined the critical agent mole fraction dependence on the injection period. The total agent mass required for suppression increased with the mean air velocity and then leveled off to a level proportional to the step height as the transition from the rim-attached to wake-stabilized flame regime occurred. INTRODUCTION Fires in the aircraft engine nacelle, which encases the engine compressor, combustors and turbine, can be stabilized by a recirculation zone formed behind a clutter (tubes and boxes, etc.) under over-ventilated conditions [1-7, 9, 10]. The fuel sources are leaking jetfuel and hydraulic-fluid lines that can feed the fire in the * Research Engineer. Research Institute, Senior Member AIAA t Associate Research Physicist, Research Institute t Mechanical Engineer. Propulsion Sciences and Advanced Concept Division, Member AIAA Copyright © 1998 by the authors. Published by the American Institute of Aeronautics and Astronautics, Inc. with permission. form of a spray, puddle, or pool. Similar conditions may exist in fires in aircraft dry bays, ships, or land combat vehicle engine compartments. Suppression occurs when a critical concentration of agent is transported to the fire. After the fire is extinguished, reignition may occur as the fuel-air mixture makes contact with hot metal surfaces or sparks from damaged electrical circuits. Because of its superior effectiveness, halon 1301 (bromotrifluoromethane, CF3Br) has been used as a fire-extinguishing agent to protect aircraft engine nacelles and other compartments. As halon 1301 is replaced with a possibly less effective agent, the amount of replacement agent required for suppression over a range of operating conditions must be determined. Hence, it is not clearly known whether or not the flame extinguishing data [7, 12-14] using conventional methods such as a cup burner can effectively characterize bluff-body stabilized flames. The broad objectives of this study are as follows: (1) Determine difficult-to-extinguish cases by a parametric investigation using combinations of given geometric elements and experimental conditions. The parameters to be considered are (a) clutter configuration (backward-facing step, buffle plate, J-flange, cavity, and blockage ratio), (b) fuel and injection characteristics (fuel type: methane and JP-8; spray or pool), (c) air flow characteristics (velocity and temperature), (d) hot surface (roughness and temperature), and (e) suppression agents (agent type: CF3Br, CF3I, C2HF5 [HFC-125], C3HF7 [HFC-227ea]; temperature, supply vessel pressure and injection period). (2) Gain a better understanding of the fundamental mechanisms of flame stabilization and identify the critical parameters that are important to suppressing bluff-body stabilized flames. (3) Develop a phenomenological model that can be integrated into computational fluid dynamics models