Investigation of Tip Clearance Modeling Techniques for a Transonic Compressor Rotor

Abstract

Computational fluid dynamics (CFD) codes are becoming an integral part of the design and analysis process involved with creating and improving upon new engine designs. This necessitates the investigation and development of accurate modeling techniques for flow simulations with a quick turn around time of typically 48 hours. The present paper is focused on increasing the fidelity of compressor rotor simulations by examining three rotor tip clearance modeling techniques. The first approach models the tip clearance region as a loss-less, periodic, un-gridded region as first proposed by Kirtley et al. The second approach is a modification of this technique to study the vena-contracta effects. The tip clearance region remains un-gridded, but, the physical radial depth of tip clearance is gradually reduced to the smallest constriction typically seen in the tip clearance because of flow phenomena such as the shroud and blade-tip boundary layers. The final approach is a completely gridded tip clearance region of full depth to verify the vena-contracta approach as well as to determine if any increase in fidelity is achieved through this computationally costly procedure. These three tip clearance modeling approaches are applied to the NASA transonic compressor rotor, Rotor-35, in a rotor only configuration and the predicted operational ranges are compared to available LDV data. Span-wise performance characteristics such as total pressure ratio and total temperature ratio are compared at a near peak efficiency and at a near-stall operating point. Tip-vortex resolution and predictions are also examined. The merits and demerits of the three approaches are discussed and recommendations are made for a viable approach in terms of accuracy and computational resources.