An experimental investigation has been carried out on the influence of grooved casing treatment on the stall margin of a compressor rotor. Tests were conducted with two rotor builds having different solidities (but all other parameters identical) so that one of the rotors exhibited a wall, or casing, type of stall, while the other showed a blade stall. It was found that the casing treatment, when compared to the solid casing, was very effective in increasing the stall margin of the wall stall configuration, whereas there was little or no change in the stall point of the blade stall configuration. Detailed relative frame measurements of the rotor exit flow field were also taken as part of the program. These showed that in a wall stall situation, the use of casing treatment produced a substantial decrease in the relative total pressure defect, compared with the solid casing, while this was not true for the blade stall type of blading. The results of the experiments, both in overall measurements and in the detailed relative frame traverses, support the hypothesis that casing treatment is effective only in a situation in which a wall stall exists.

A nonlinear compressible flow model has been developed for the prediction of compressor performance and stability with a circumferential flow distortion. This model uses multiple parallel compressor segments and accounts for deviations from undistorted compressor performance. It is applicable to large-amplitude inlet distortions of total pressure and/or temperature, as well as circumferential variations in exit static pressure. The basic model requires the undistorted performance characteristics for each blade row; however, a modified version based upon the overall compressor performance gives an accurate approximation when detailed blade row characteristics are not available.

No abstract