Experimental Investigation of Diffuser Hub Injection to Improve Centrifugal Compressor Stability

Results from a series of experiments to investigate whether centrifugal compressor stability could be improved by injecting air through the diffuser hub surface are reported. The research was conducted in a 4:1 pressure ratio centrifugal compressor configured with a vane-island diffuser. Injector nozzles were located just upstream of the leading edge of the diffuser vanes. Nozzle orientations were set to produce injected streams angled at $−8,$ 0, and $+8$ degrees relative to the vane mean camber line. Several injection flow rates were tested using both an external air supply and recirculation from the diffuser exit. Compressor flow range did not improve at any injection flow rate that was tested, and generally diminished as injection rate increased. Compressor flow range did improve slightly at zero injection due to the flow resistance created by injector openings on the hub surface. Resistance and flow range both increased as the injector orientation was turned toward radial. Leading edge loading and semivaneless space diffusion showed trends that are similar to those reported earlier from shroud surface experiments that did improve compressor range. Opposite trends are seen for hub injection cases where compressor flow range decreased. The hub injection data further explain the range improvement provided by shroud-side injection and suggest that stability factors cited in the discussion of shroud surface techniques are valid. The results also suggest that a different application of hub-side techniques may produce a range improvement in centrifugal compressors.