The flow through a compressor cascade with tip leakage has been studied experimentally. The cascade of GE rotor B section blades had an inlet angle of 65.1º, a stagger angle of 56.9º, and a solidity of 1.08. The final turning angle of the cascade was 11.8º. This compressor configuration was representative of the core compressor of an aircraft engine. The cascade was operated with a tip gap of 1.65%, and operated at a Reynolds number based on the chord length (0.254 m) of 388,000. Measurements were made at 8 axial locations to reveal the structure of the flow as it evolved through the cascade. Measurements were also made to reveal the effects of grid generated turbulence on this flow. The data set is unique in that not only does it give a comparison of elevated free stream turbulence effects, but also documents the developing flow through the blade row of a compressor cascade with tip leakage.
Measurements were made at a total of 8 locations 0.8, 0.23 axial chords upstream and 0, 0.27, 0.48, 0.77, 0.98, and 1.26 axial chords downstream of the leading edge of the blade row for both inflow turbulence cases. The measurements revealed the formation and development of the tip leakage vortex within the passage. The tip leakage vortex becomes apparent at approximately X/ca= 0.27 and dominated much of the endwall flow. The tip leakage vortex is characterized by high streamwise velocity deficits, high vorticity and high turbulence kinetic energy levels. The result showed that between 0.77 and 0.98 axial chords downstream of the leading edge, the vortex structure and behavior changes.
The effects of grid generated turbulence were also documented. The inflow turbulence had an intensity of 3.1%, a longitudinal integral scale equivalent to 26% of the upstream projected blade spacing, and spectral and correlation properties consistent with the expectations of homogeneous isotropic turbulence. The results revealed significant effects on the flow field. The results showed a 4% decrease in the blade loading and a 20% reduction in the vorticity levels within tip leakage vortex. There was also a shift in the vortex path, showing a shift close to the suction side with grid generated turbulence, indicating the strength of the vortex was decreased. Circulation calculations showed this reduction, and also indicated that the tip leakage vortex increased in size by about 30%. The results revealed that overall, the turbulence kinetic energy levels in the tip leakage vortex were increased, with the most drastic change occurring at X/ca= 0.77.
The behavior of grid turbulence flowing through the cascade outside the viscous regions was compared with basic predictions using Rapid Distortion Theory (RDT). Near the passage center the free-stream decay of turbulence stresses and spectra is substantially modified by the deceleration and turning of the flow. Near the blades the turbulence is additionally modified by the blocking of normal velocity fluctuations. The evolution of turbulence through the cascade, including its spectrum and space time correlations, appears quantitatively consistent with RDT effects superimposed over a background of viscous decay. This suggests that, at least in circumstances similar to those modeled here, RDT may offer a viable approach for calculating the modification of turbulence passing through a blade row.
This work was performed using the Virginia Tech Low Speed Cascade Tunnel. The authors acknowledge the support of the Office of Naval Research, in particular Drs. Ronald Joslin and Patrick Purtell, for their support under grant N00014-99-1-0230 and N00014-03-0199. Single point experimental data from this study is available for download here.
Please direct any questions or comments concerning the data or other aspects of this work to William Devenport at devenport@aoe.vt.edu.