Blade channel flow in simulated radial flow turbomachine

Hansen, Robert C.

Permalink

https://hdl.handle.net/2142/112020 Copy

Description

Title

Blade channel flow in simulated radial flow turbomachine

Author(s)

Hansen, Robert C.

Issue Date

1967-07

Keyword(s)

• Blade Channel Flow
• Turbomachines

Abstract

The flow of a liquid through the blade channels of a radial-flow turbine is difficult to study both because of the complex nature of the flow and because of the difficulty of access for making observations and measurements within the rotating system of blades. This study utilized a stationary model which simulated insofar as possible such a flow while greatly simplifying the experimental problems. The channel flow patterns were observed in this model, and the static pressures were measured with a monometer while the flow speeds were measured with a hot-wire anemometer. This novel experimental model was a 6-blade segment of an 18-blade annular, radial-inflow cascade. One wall was plane, the other was contoured so that the circumferential area between was constant at all radii. Three rotational speeds (zero, intermediate, and high) were simulated by different model blade shapes. Differences in the flow patterns among the three blade shapes were minor. All exhibited three important characteristics: (1) a turbulent boundary layer which separated from the blade suction surface rather early in the channel, while the turbulent boundary layer on the blade pressure surface remained very thin and attached throughout the blade length; (2) the flow retained much of its linear momentum and did not diffuse appreciably in the divergent channels, and (3) a secondary flow developed along the walls between the blade surfaces. The effect of (1) and (2) was that a sizeable separation eddy existed along the blade suction surface downstream from the flow separation point. A review was made of the ideal flow theory as applied to liquids in turbomachines, and the importance of the assumptions made in applying this theory was noted. The calculated blade surface boundary layer development based on the ideal flow prediction of blade surface velocities was grossly different from that observed in the experimental model. The difference was presumed to be associated with the curvature of these surfaces, which was neglected in the boundary layer calculation. The theory of secondary flows in bends due to the transverse vorticity in the approach flow was extended and applied to the developing wall boundary layers within the curved blade channel. Agreement between the calculated secondary flow and that observed in the model was good. The consequences of this study are summarized as follows: (1) The shapes of a radial-flow turbine’s walls should be designed on the basis of the total flow velocity and attainable diffusion rather than on the current method based on considering only the radial component of the velocity, (2) secondary flows due to upstream and internally developed transverse vorticity can be predicted reasonably well, and (3) blade surface boundary layer development and separation needs further study; consideration of the surface curvature is indicated.

Publisher

Department of Theoretical and Applied Mechanics. College of Engineering. University of Illinois at Urbana-Champaign

Series/Report Name or Number

• TAM R 297
• 1967-0616

ISSN

0073-5264

Type of Resource

text

Language

eng

Permalink

http://hdl.handle.net/2142/112020

Sponsor(s)/Grant Number(s)

Caterpillar Tractor Company

Copyright and License Information

Copyright 1967 Board of Trustees of the University of Illinois

Owning Collections