CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Velocity inlet boundary condition for porous medium (

Chander March 8, 2012 11:08

Velocity inlet boundary condition for porous medium

I am using porous medium model in my simulation and I want to share an observation with you and need confirmation on whether what I have done is correct.
I wanted to specify a spatially varying mass-flow rate (varying along one direction y) on the inlet to my porous medium domain such that the total mass-flow entering the porous domain is M (say). For that I calculated velocity as a function of y such that it leads to the required mass-flow distribution along y and the total mass-flow also sums up to M. I imposed this velocity profile on the inlet surface of the porous medium. The velocities in the velocity profile were superficial velocities ( I had calculated them to be so) and I also selected superficial velocity in the options under the porous model tab in CFX-Pre.

Now when I analysed the result file and calculated the total mass-flow on inlet, I found it to be less than M. I found that the new total mass-flow M' was almost equal to porosity*M. That meant that CFX was treating the velocity at inlet boundary as true velocity and not superficial velocity (while it should have treated it as superficial velocity).

So I changed the velocities on the velocity inlet boundary condition by dividing each velocity by porosity so as to impose profile of true velocity instead of superficial velocity. And this corrected the problem as CFX post reported total mass-flow at inlet to porous medium as equal to M.

Has anyone else also encountered this issue? Shouldn't CFX have treated the velocities at porous medium inlet as superficial velocities or am I missing anything here?

evcelica March 8, 2012 23:07

Did you define your porous domain as superficial or true velocity? If true velocity I would expect the result you saw.
*EDIT* Nevermind, I just re-read that you did use superficial.

That's why I like to use mass flow, its pretty hard to mess that up.

Chander March 9, 2012 11:52

Thanks for your reply!
However, in my case I need to specify a velocity profile only to get a massflow profile.

evcelica March 11, 2012 22:18

I just noticed there is a check box for sources of porous domains which reads "multiply by porosity". If you had this box checked that would explain your results. If not, then I have no clue and would believe it to be a bug.

All times are GMT -4. The time now is 21:21.