minimize inefficiency in partial-admission turbine

[Allen3]

This is a question about aerodynamic friction in the context of a partial-admission impulse turbine.

The working fluid will be air or some other gas. In this particular turbine, the buckets will be U shaped and will be symmetric, i.e., if the rotor were simply spinning by itself in the air, it would tend to simply rotate the air, not drive it axially in either direction.

In order to understand how friction losses might be minimized as the rotor moves past the blocked portions of the stator, I'm considering as a mental exercise three different ways block the pressured air from flowing through some portion of the stator, in order to accomplish partial-admission. For the sake of this exercise, let's assume we are going to block one half of the stator.

Method one would be to place a smooth plate between the stator and the rotor that covers half of the stator. As the rotor turned, each bucket would move through the moving air from the stator for one-half revolution, and then past a plate for the second half.

Method two would be to place a plate on the entrance of the stator. As the rotor turned, each bucket would move through the moving air from the stator for one-half revolution, and then past the vanes of the stator, through which no air could flow, for the second half.

Method three would be to retract half of the stator. As the rotor turned, each bucket would move through the moving air from the stator for one-half revolution, and then past "open space" for the second half.

Does anyone have a feel for the relative friction losses that could be expected in each of these three scenarios? Which would have the most friction, which would have the least, and how significant might the differences be?

Thank you for any thoughts you could provide.

 

[stgrme]

I can't give you any absolute answers, but I can relate some information, based on my experience. I worked for the Westinghouse Steam Turbine-Generator Division through the 1970s.

Westinghouse built many industrial steam turbines, which employed control (first) stages with less than a 360 degree arc of steam admission. In order to reduce windage (friction) losses, these turbines usually incorporated baffles (or shielding) in the control stage in the portion of the circumferential arc where no flow was admitted. The baffles reduced the amount of circumferential flow and consequently windage losses. In downstream stages, the diaphragms (stationary vane elements) were also built with less than a 360 degree arc of flow passages. Typically, the arc of flow passages increased in each subsequent stage.

I hope this information helps.

Best of luck!

 

[Allen3]

Thank you for your feedback. Just a quick follow up question: were the baffles placed relatively close to the rotor, or at some distance? Was the surface of the baffle smooth (to promote laminar flow) or rough, dimpled or radially grooved (to promote the development of a shearing layer)? Thank you.

 

[stgrme]

I'm relying my recollection; I don't have any drawings. As I recall, the baffles were smooth plates. There were no close clearances or seals of any type.

Best of luck!