Orifice plate thickness to achieve critical flow

[MarkkraM]

When checking my orifice calculation method (after reading denniskb's thread) I came across a statement saying that choked/critical flow does not occur accross thin, sharp edged, orifice plates. For chocking to occur the thickness of the orifice plate must exceed the orifice diameter.

Is this true and why?

 

[hacksaw]

you are correct, neither thin or thick plates exhibit critical flow characteristics because of the entrance losses.

Thick plates do offer a discharge coefficient that is nearly independent of reynolds number and tend to approach critical flow behavior better than a thin plate. You only approach critical flow in really thick plates i.e. T/d>>>D, because you achieve pipe flow prior to the final expansion, but you still have to deal with the entrance losses.

There are corrections for the "critical flow" behavior of thin/thick plates (T/d~1) in Millers Flow Handbook, but they only increase the complexity of the calcualtion rather than improve performance.

A flow nozzle does achieve true critical flow. Some of the meter manf. can supply them.

 

[JoelleLaw]

Choking must occur, as the vena contracta has a minimum size equivalent to that of the orifice.

As pointed out to me, the following link supports this idea:

http://www.optimal-systems.demon.co.uk/flow-orifice.htm

 

[hacksaw]

Joel,

For thin plates it only occurs as a limiting condition as acknowledged in the web-site you've referenced.

The issue raised by Mark is that critical flow does not generally occur with plates that fall in the thin-plate category. True critical flow conditions can occur only for true nozzles with a contoured inlet.

 

[25362]

To Hacksaw, if I'm not mistaken the "0.5-0.6 pressure ratio" ROT doesn't apply to flow through nozzles which may reach critical (sonic) flow already at a 5-10% pressure drop. Right or wrong ? Please comment.

 

[davefitz]

A good discussion on this topic is thread798-51260, or get a copy of the following thechnical paper on choked flow thru square edged orifices:
author A.J. Ward-Smith , Int J Heat Fluid V1 no 3 pp123-132 1979.

Basically the discharge coeficient is related to the ratio of the plate thickness divided by the orifice diameter. For a truly sharp edge t/d=0, the Cd=1.0 . For a t/d in the range 1<t/d<3, Cd=0.81 . For very thick plates t/d >7 the Cd is calculated using the Fanno relationships for frictionally choked flow in piping.