No abstract

The performance deterioration of a high-speed axial compressor rotor due to surface roughness and airfoil thickness variations is reported. A 0.025 mm (0.001 in.) thick rough coating with a surface finish of 2.54–3.18 rms μm (100–125 rms μin.) is applied to the pressure and suction surface of the rotor blades. Coating both surfaces increases the leading edge thickness by 10 percent at the hub and 20 percent at the tip. Application of this coating results in a loss in efficiency of 6 points and a 9 percent reduction in the pressure ratio across the rotor at an operating condition near the design point. To separate the effects of thickness and roughness, a smooth coating of equal thickness is also applied to the blade. The smooth coating surface finish is 0.254–0.508 rms μm (10–20 rms μin.), compared to the bare metal blade surface finish of 0.508 rms pm (20 rms μin.). The smooth coating results in approximately half of the performance deterioration found from the rough coating. Both coatings are then applied to different portions of the blade surface to determine which portions of the airfoil are most sensitive to thickness/roughness variations. Aerodynamic performance measurements are presented for a number of coating configurations at 60, 80, and 100 percent of design speed. The results indicate that thickness/roughness over the first 2 percent of blade chord accounts for virtually all of the observed performance degradation for the smooth coating, compared to about 70 percent of the observed performance degradation for the rough coating. The performance deterioration is investigated in more detail at design speed using laser anemometer measurements as well as predictions generated by a quasi-three-dimensional Navier–Stokes flow solver, which includes a surface roughness model. Measurements and analysis are performed on the baseline blade and the full-coverage smooth and rough coatings. The results indicate that adding roughness at the blade leading edge causes a thickening of the blade boundary layers. The interaction between the rotor passage shock and the thickened suction surface boundary layer then results in an increase in blockage, which reduces the diffusion level in the rear half of the blade passage, thus reducing the aerodynamic performance of the rotor.

Testing over a range of Reynolds numbers was done for three NACA 65 Profiles in cascade. The testing was carried out in the VKI C-1 Low Speed Cascade Wind Tunnel; blade chord Reynolds number was varied from 250,000 to 40,000. A semiempirical theory is developed which will predict the behavior of the shear layer across a laminar separation bubble. The method is proposed for two-dimensional incompressible flow and is applicable down to short bubble bursting. The method can be used to predict the length of the laminar bubble, the bursting Reynolds number, and the development of the shear layer through the separated region. As such it is a practical method for calculating the profile losses of axial compressor and turbine cascades in the presence of laminar separation bubbles. It can also be used to predict the abrupt leading edge stall associated with thin airfoil sections. The predictions made by the method are compared with the available experimental data. The agreement could be considered good. The method was also used to predict regions of laminar separation in converging flows through axial compressor cascades (exterior to the corner vortices) with good results. For Reynolds numbers below bursting the semiempirical theory no longer applies. For this situation the performance of an axial compressor cascade can be computed using an empirical correlation proposed by the author. Comparison of performance prediction with experiment shows satisfactory agreement. Finally, a tentative correlation, based on the NACA Diffusion Factor, is presented that allows a rapid estimation of the bursting Reynolds number of an axial compressor cascade.

The performance deterioration of a high speed axial compressor rotor due to surface roughness and airfoil thickness variations is reported. A 0.025 mm (0.001 in.) thick rough coating with a surface finish of 2.54–3.18 RMS μm (100–125 RMS microinches) is applied to the pressure and suction surface of the rotor blades. Coating both surfaces increases the leading edge thickness by 10% at the hub and 20% at the tip. Application of this coating results in a loss in efficiency of 6 points and a 9% reduction in the pressure ratio across the rotor at an operating condition near the design point. To separate the effects of thickness and roughness, a smooth coating of equal thickness is also applied to the blade. The smooth coating surface finish is 0.254–0.508 RMS μm (10–20 RMS microinches), compared to the bare metal blade surface finish of 0.508 RMS μm (20 RMS microinches). The smooth coating results in approximately half of the performance deterioration found from the rough coating. Both coatings are then applied to different portions of the blade surface to determine which portions of the airfoil are most sensitive to thickness/roughness variations. Aerodynamic performance measurements are presented for a number of coating configurations at 60%, 80%, and 100% of design speed. The results indicate that thickness/roughness over the first 10% of blade chord accounts for virtually all of the observed performance degradation for the smooth coating, compared to about 70% of the observed performance degradation for the rough coating. The performance deterioration is investigated in more detail at design speed using laser anemometer measurements as well as predictions generated by a quasi-3D Navier-Stokes flow solver which includes a surface roughness model. Measurements and analysis are performed on the baseline blade and the full-coverage smooth and rough coatings. The results indicate that coating the blade causes a thickening of the blade boundary layers. The interaction between the rotor passage shock and the thickened suction surface boundary layer then results in an increase in blockage which reduces the diffusion level in the rear half of the blade passage, thus reducing the aerodynamic performance of the rotor.

A model of the spanwise variation of the 3-D flow effects on deviation is proposed for middle-stage rotors and stators. This variation is taken as the difference above or below that predicted by blade element theory at any spanwise location. It was found that the stator variation is strongly affected by the end-wall boundary-layer thickness as well as camber, solidity, and blade channel aspect ratio. Rotor variation was found to depend on end-wall boundary layer thickness and tip clearance normalized by blade span. If these parameters are known or can be calculated, the models provide a reasonable approximation to the spanwise variation of deviation for middle compressor stages operating at low to high subsonic inlet Mach numbers.