Orifice plate using rasInterFoam
 

I'm trying to validate a simple orifice plate model for use in a larger assembly, but it is not behaving as I would expect. As you can see below the 'tail' from the orifice adheres to one side of the duct, rather than remaining symmetrical. Is this expected? Duct is 25mm wide, air, inlet velocity = 0.5m/s.


http://lh3.ggpht.com/_fdWk3HSvzHc/Sc...Screenshot.png

This might be the Coanda effect at work, where a jet attaches itself to a boundary. There are papers out there that describe this phenomena.

Hi,
Is this result from steady state simulation?
Maybe transient run shows fluctuation of this tail up and down?
I would guess, that's not impossible for this kind of flow to show such a behaviour.
Kerstin

Long time ago this has been. But I am experiencing something similar at the moment. The strange thing is: While using an incompressible solver, everything looks fine and qualitatively similar to ISO 5167-2 results.

But if I use rhoSimpleFoam instead of simpleFoam I get kind of two layers, one behaving normally in the outer half of the flow and something close to a jet in the middle half of the flow...

Any ideas concerning that?
Could you maybe tell which solver you used at that time, if it was compressible or incompressible, if you solved the "problem" or accepted the flow to be different from what one would have expected?