CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Mixing in porous media (

vishyaroon April 17, 2009 12:34

Mixing in porous media
How good (are trustable) are predictions of mixing of two fluids in porous regions (any software - Fluent, STAR)? I know from the basic formulation of porous media, that the constants are defined for computing the pressure drop across the region, but can we believe the results of mixing of two fluid streams in the region?

otd April 17, 2009 21:19

Are mixing liquids, gases, both?

harishg April 18, 2009 01:29

Trial and error and literature survey is what you need to do.

vishyaroon April 18, 2009 16:13

I am looking at two gases that mix as they pass through a porous zone.

otd April 19, 2009 14:11

Sorry I didn't read your question closely. It appears you are asking about the ability of Fluent to do your problem.

Have you asked the question on the Fluent forum or, perhaps, asked the Fluent support staff directly?

andy2o April 19, 2009 17:49

Consider these two examples of domains which could be modelled as porous medium flows, but have very different mixing characteristics:

1) Flow through many parallel micro-channels - which will permit zero cross-flow mixing because the flow is confined to the individual channels.
2) Flow through a random bed of spheres, where fluid will be spread as it hits each sphere, causing cross-flow motion which will help mixing.

I guess your application lies somewhere between these extremes. The point is that a drag correlation only computes the drag. It does not distinguish between these two different mixing scenarios. Therefore you need to supply a model for the effective momentum and mass diffusion caused by the porous medium as well as the drag force to complete your model of a porous medium. This diffusion model will need to take the form of correlation / approximation based on experimental data (or perhaps very detailed CFD) just like the drag coefficients do.

As always: The results will be only as good as the correlation! So you can only trust the results if you trust your correlation and validate your model!!

Best regards,

vishyaroon April 20, 2009 08:41

Thank you all for replying to the question. I guess as Andy mentions I'll have to do a detailed CFD or get some correlataions based on experiments.

All times are GMT -4. The time now is 21:45.