[vainilreb]

Hi everyone,

I might have discovered a bug in chtMultiRegionFoam.

My model consists of 4 solid regions, those being a heating layer plus 3 layers of different materials, arranged around a fluid region.

This fluid region is quite complex but - for first understanding - may be reduced to an inlet, a porous zone and an outlet. The fluid has to pass the porous zone on its way from inlet to outlet.

I created the stationary flow field first, using chtMultiRegionSimpleFoam and everything looks fine. There is a backpressure generated by the porous zone, so the pressure upstream of the porous zone is higher than downstream.

The problem is:

When I start chtMultiRegionFoam from the stationary result, it kind of flips the pressure distribution along the porous zone. It takes only several time steps, let's say an overall time span of a millisecond, until the pressure downstream is much higher than upstream. A few milliseconds later I end up with backflow through the center of the porous zone which creates a large eddy system inside the porous zone.

I've tried almost every imaginable variation of boundary settings, but the result stays the same.

Might this be a bug? Please tell me if you need more information. I'd really appreciate your help on this, it has already cost me weeks of try and error...

Thanks a lot.

Robert

[vainilreb]

I have tried the same setup using rhoPimpleFoam and do not get the same two eddies, but a nice flow field. I'm afraid the problem could be caused by chtMultiRegionFoam solving for p_rgh, although I set gravity to (0 0 0). Are there any further differences between rhoPimpleFoam and chtMultiRegionFoam (regarding the flow solving process)?

[vainilreb]

Please guys, I need help.
What's the matter? Too little information? No clue?

[Adri_12]

Originally Posted by vainilreb

Please guys, I need help.
What's the matter? Too little information? No clue?

Hi Robert,

I am starting a case really close to yours, with a 3 region flow air/porous zone/air and I am wondering :

1. How to consider the porous region instead of a classical solid region definition.
2. Could you share your case or explain your steps/choices to achieve a good case parametrization (fvSolutions in constant/porous/ folder?)
3. Have you solved your backflow problem ?

Thanks !

Adrià

[hdotyao]

Originally Posted by Adri_12

Hi Robert,

I am starting a case really close to yours, with a 3 region flow air/porous zone/air and I am wondering :

1. How to consider the porous region instead of a classical solid region definition.
2. Could you share your case or explain your steps/choices to achieve a good case parametrization (fvSolutions in constant/porous/ folder?)
3. Have you solved your backflow problem ?

Thanks !

Adrià

Hello Adrià,


I am doing a similar case using chtMultiRegionFoam with two fluid region, and both are employed with porous media. I have also backflow problem when I have a mass source for a fluid region... Have you solve your backflow problem?


Best,
Yao