Web Content Display Web Content Display

Advanced Fan and Compressor Development Studies


SPONSOR: NAVAIR

 

Transonic Fan & Compressor Studies

A study was completed to examine the effects of low-profile vortex generators on fan shock-boundary layer interaction at M=1.4 using a two-passage cascade simulation. Effects on losses and turning angle were obtained with wedge probe wake surveys.

A transonic stage designed by N. Sanger using 3D-viscous codes was installed in the reactivated transonic compressor test rig. The baseline geometry was also used as the point of departure for a swept rotor study using Bezier surface representations of the geometry.

Concurrently, the pressure-sensitive paint (PSP) technique is being developed in the laboratory to provide a rotor-flow diagnostic to augment LDV and high-response pressure transducer techniques. All techniques will be used to validate the Sanger stage and, later, more advanced designs which incorporate sweep.

Stage calculations were performed on the Sanger compressor using Chima's (NASA Glenn) SWIFT code which uses various averaging techniques between blade rows. Energy averaging was used with good overall prediction of the stage performance.

 

Off-Design and Compressor Cascade Stall

Second generation CD blades, designed by Gelder of NASA LeRC, were installed in the Low Speed Cascade Wind Tunnel (LSCWT) and tested at design. The measured blade surface pressure distribution agreed well with the design distribution. These blades were designed with half the solidity of the previous Sanger stator (for NASA LeRC Rotor 67A), and with elliptic leading and trailing edges so as to minimize losses and leading edge separation bubbles. At off design a Reynolds number study was performed on the blades and the effect of Reynolds number on the formation of a midchord separation bubble was quantified.

The refurbished fan and motor for the cascade was recently used to verify the measurements taken previously before the motor failed. In addition to performing midspan LDV measurements five-hole pressure measurements were also taken in the wake of the blades to determine the loss coefficient and distribution. Inlet total pressure profiles were also obtained with a 21 port rake probe ahead of the blades. Blade surface flow visualization was also performed with titanium-dioxide and kerosene to view the flow separation on the suction surface of the blades.

Endwall flowfield measurements were performed in the Low Speed Cascade Wind Tunnel. Pitchwise surveys of the wake were performed with a five-hole probe starting at mid span and then progressing toward the north wall (through which LDV measurements are taken) in decreasing increments. Thus an overall loss distribution has been obtained for the second-generation controlled-diffusion blades at 40 degrees inlet flow angle (4 deg. above design incidence). Next off-centerline surveys were conducted with the two component LDV for about 40% span. Three-dimensional viscous flow calculations of the whole blade passage were also performed in an attempt to compute the complex separated flow over the blades.