Learning Forum

[karimi]

Hi

I am trying to model teh flow in this system, that I have a non-porous fluid coming and interacting with a porous fluid and then leaves it to a non-porous fluid, and eventually the outlet. My inlet is the mass flow rate and my outlet is a pressure. The model initialize pressty good but I am having a serious problem with convergence. My mesh is really good quality and fine enough, my geometry is smooth and nice too. And I am sure my BCs are all correct, I just do not know if I missing something here. Maybe I need to define an interface between the porous and non-porous regions or change teh solver? my fluid is incompressible and just simple water. Would appreciate any helps and ideas here? My mesh is way finer than this pic so I am sure there is no problem with the mesh. I went to even 20 million elements fine mesh but still the same problem.
 

[karimi]

Just an update, I used both Steady and Transient methods. Also, decreased the permeability to a half, however, none of these helped me to get this converged.

[Rob]

Without some idea of domain size, flow speed, mesh quality and many other factors it's difficult to comment. Some images of results (pressure & velocity are a good start) along with residuals may help. 

[karimi]

Thank Rob, so lets resolve this step by step. When I define a porous and a non-porous region, at their interfaces, Do I need to define a specific boundary condition like a porous jump or it can be treated just as a wall? 
Also, when I define a fluid component as a porous, do I still need to define a porous jump or one of those will do the job for me?

 

[Rob]

If you treat the boundary between porous & non-porous as a wall there will be no flow. The default boundary between the zones should be an interior, and I would generally leave it like that. About the only reason for using a PJ there would be to trap particles if the DPM model was being used. 

[karimi]

Rob,

I figured the conact between the porous and nonporus zone are being recognized as walls and this is why the flow cannot get out and I get divergence in continuity. However, when whatever I do I cannot transfer those contact regions/mesh interfaces to internal and fluent insist on recognizing them as walls. I am stuck and I do not know what to do with this? Appreciate your help

[karimi]

Rob, so if you see here my interfaces are being treated as wall and I think here is the error I get since teh flow cannot go from one region to another one. What should I do here? Is that correct? and whatever I do it cannot make the wall to internal

 

 

 

 

 

[Rob]

Check the interface contact area, I think you can right click on pair now and check. The walls are for when the surface pairs aren't in contact. I assume you didn't set "coupled wall" when you set them up?