Experimental and Numerical Investigation of Different Flame Types Inside a Laboratory Scale RQL Combustion Chamber

Abstract

Abstract RQL (rich burn, quick quench, lean burn) combustion chambers are common in modern aero engines due to their low NOx emissions and good stability. The rich primary zone leads to lower flame temperatures and in combination with the lack of oxygen, the NOx production is low. The mixing of the secondary air must be quick in order to avoid stoichiometric conditions and at the same time must ensure the oxidation of the soot produced in the fuel rich primary zone to keep soot emissions to a minimum. However, the design of such a combustion chamber is complicated due to the complex interaction between the swirling primary flow and the jets of the secondary airflow. In this paper, we present a new test rig, which was designed to study combustion processes inside RQL combustion chambers at atmospheric conditions. The test rig features liquid kerosene combustion and a realistic quenching zone as well as good access for optical and conventional measurement techniques. For realistic engine like conditions the combustion air is preheated to 600 K and the fuel–air equivalence ratio in the primary combustion zone is set to be between Φ = 1.66 and Φ = 1.25, resulting in an overall thermal power between 80 kW and 110 kW. To get insights into the complex flow field inside the combustion chamber unsteady RANS simulations of both the reacting and the non-reacting case were performed using OpenFOAM. The turbulent flow field was modeled using the k-ω-SST model and the combustion was simulated using the Partially Stirred Reactor model. The experimental investigations showed two stable flame types for the same operating conditions with considerable differences in the visible flame structure and soot radiation. The flow field of both of these flame types were measured using a 1.5 kHz 2D PIV System. The numerical simulations showed good overall agreement with the experimental results but could not represent the change in flame type. In order to understand the underlying effects of the flame change the OH* chemiluminescence was recorded and the two-phase flow near the nozzle exit was investigated. This showed that the change in flame structure might arise due to spray dispersion of the pilot fuel nozzle and the recirculation of the secondary air into the primary zone.