Lean-Burn Combustor Simulator for an Engine-Component Test Facility: An Experimental and Computational Study

Abstract

Abstract The computational and experimental assessment of a lean-burn low-NOx combustor simulator for an engine component test facility is presented. The Engine Component Aero-Thermal (ECAT) facility is a full-scale engine-parts facility, designed for the study of the aero-thermal performance of fully cooled high-pressure nozzle guide vanes (NGVs). The facility operates with non-dimensionally matched engine conditions in terms of Reynolds number, Mach number, and coolant-to-mainstream pressure ratio. The combustor simulator is designed to replicate lean-burn conditions of swirl and temperature distortion upstream of the nozzle guide vanes. The purpose is to allow the study of flow capacity, aerodynamic performance (with film cooling), and thermal performance (overall effectiveness) in the presence of engine-realistic inlet distortions. The ECAT combustor simulator design builds on the work of Hall and Povey, who developed a full-scale low-speed atmospheric-pressure combustor simulator for the preliminary design of similar simulators later implemented in the Oxford Turbine Research Facility and ECAT. This pilot facility produced non-dimensional lean-burn combustor-exit conditions closely matched to target profiles representative of modern aero engines. This design was modified and scaled for compatibility with the annulus-line and higher-pressure operating conditions of the ECAT facility. In the present study, detailed experimental measurements with multi-hole probes and thermocouples (pressure profile and temperature field) are presented and compared to results from Reynolds-Averaged Navier–Stokes Simulations. Additional simulations were performed to understand how the elevated back-pressure and vane potential field affect the non-dimensional profiles of pressure loss, residual swirl, and temperature at the combustor–turbine interface. This is perhaps the most comprehensive study to date of a combustor simulator in an engine-scale research facility, providing a unique insight into the known challenges of simulator design, scaling issues when moving from low to high Reynolds number, and limitations of computational fluid dynamics in this flow environment. The results, which will serve as boundary conditions to understand the impact of this flow on components in future studies in this facility, demonstrate the fidelity with which lean-burn target conditions can be replicated in a non-reacting environment.