Use of Time-Accurate CFD for Inlet-Fan Interaction

Abstract

The Compressor Aerodynamics Research Laboratory at Wright-Patterson Air Force Base is currently investigating, experimentally and numerically, the effects of inlet distortion on transonic fan performance. The experimental investigation will quantify the flowfield for a cold-tested diffuser-fan with distortion screens, and a coupled diffuser-fan system. Each of these experiments will be mirrored numerically using the solver TURBO. The primary research goal of the work is to quantify the physical mechanisms for distortion-transfer and develop reduced-order models to account for performance and stall-margin loss in engine design. As inlet geometries become more aggressive, namely having smaller length to diameter ratios, the secondary flow physics induce potentially-harmful distortion at the engine fan face. The effects of distortion are typically only investigated experimentally, and at too low a temporal and spatial resolution to adequately quantify the mechanisms which attenuate or amplify total pressure and total temperature non-uniformities. Because of the imbalance of operating conditions between different passages of the turbo-machine, adverse effects have been noted on stall-margin which cannot be explained with traditional design methods. Recently large-scale simulations have been used to investigate total pressure distortion patterns on several full-annulus, multi-stage fan configurations. The current effort simulates a distortion-producing inlet coupled with a single-stage fan. This allows distortion of total pressure, total temperature, and flow angularity to be investigated. Preliminary numerical results have been obtained on the US Air Force Research Laboratory DoD Supercomputing Resource Center (AFRL DSRC) SGI Altix 4700 system for the coupled diffuser-fan system. Total pressure and temperature exhibit both a distinct circumferential variation and a counter-rotation shift. As the simulations are analyzed, a greater understanding of performance detriment, stall-inception, and distortion- transfer in an installed aircraft system will be gained. Finally, a brief discussion of continued efforts investigating blade-row interactions will be presented.