CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS (http://www.cfd-online.com/Forums/ansys/)
-   -   Eulerian phase "screen" boundary condition (http://www.cfd-online.com/Forums/ansys/112188-eulerian-phase-screen-boundary-condition.html)

blacksoil2012 January 23, 2013 12:17

Eulerian phase "screen" boundary condition
 
Hi,

I'm doing particles+water multiphase flow in Fluent using Eulerian approach. In reality, the outlet is with a screen; that is, only water comes out, but not the particles. When I set this to be "pressure-outlet" boundary, I can not find a way such that the particle phase will be prevented from going out at the screen. Does anyone know how to do this?

I am aware that if my particles are simulated as DPM, then it can have separate boundary condition different from that of the continuous phase.

Thank you very much!

blacksoil2012 January 24, 2013 17:36

Anybody had such experience? I assume when people were doing CFD that involves filters or screens, they may encounter this situation.

One thing that popped up was that in CFX, it offered "degassing" BC, while gas can escape from the domain while water is having a free surface. Although it's not the thing I wanted (separate BC for granular phase and primary liquid phase in an Eulerian multiphase simulation), at least it provided separate BC for different phases. Unfortunately, Fluent seems unable to provide the option of setting up different type of BC for different phases on a boundary.

bunny rabbit January 29, 2013 23:20

Why don't you apply porous jump condition to the outlet. It works. I am using it for the same purpose.

blacksoil2012 January 30, 2013 01:53

Quote:

Originally Posted by bunny rabbit (Post 404924)
Why don't you apply porous jump condition to the outlet. It works. I am using it for the same purpose.

Thanks for your input. I see that "porous jump" boundary condition is only used for interior interfaces to simulate "screen" or "filter". Does that mean for my actual outlet boundary (which is actually a screen), I have to add extended computational domain to enclose that outlet so that I can apply porous jump BC on it?


All times are GMT -4. The time now is 12:27.