Learning Forum

[hhayati]

Hello,

I am trying to simulate particle transport in a filter (filter fouling) and I used Porous Media in Fluent to simulate fluid flow.

The geometry has 3 zones-feed zone, membrane zone, and permeate zone. Fluent generates shadow walls associated to the interface between these zones. My issue is that apparently particles cannot deposit on the shadow walls, although I set "Trap" boundary condition for them! Particles cross the porous media and reach permeate zone, which is wrong!

Has anybody worked on this before? If so, would you let me know what the solution is?

Best,

[Rob]

If Fluent is creating walls at these locations you'll not get any flow either. Can you check the model set up? Please post a few images of the pressure & velocity field.

[hhayati]

@rob I have not had any issue with the flow domain. The following shows the pressure in feed and permeate zones and velocity in membrane zone.


and Fluent considers shadow wall for the interface between zones. The figure below shows the boundary conditions in Fluent.

[Rob]

You're looking at wall-18, wall-19 etc? Those are part of the interface zone and only exist if there are only cells on one side of the interface. If you have cells on both sides of the interface (ie 100% contact) the wall isn't present. A wall, wall-shadow pair is for a wall boundary with cells on both sides but also a conformal mesh.

[hhayati]

Yes, I have cells on both sides. And, I do have conformal mesh. Based on what you sad ("the wall isn't present"), then particles can not deposit on these interfaces, right? even if I set trap boundary condition for them.

[Rob]

Correct as there isn't a wall. If you remesh with a conformal mesh you can make those surfaces into a porous jump and those can trap particles.

[hhayati]

Thank you so much .
I am a bit confused.
can I consider the membrane zone as porous media and its boundaries as porous jump?

[Rob]

If the membrane is a porous media you set the properties on that. You use a PJ as the bounding surface and ensure resistance is zero, but use it to assign particle bc's.

[hhayati]

This is how I set up the PJ boundary condition, but it is not giving me the same results that I got when it was interface boundary condition!!!

[Rob]

Can you post the new results?

[hhayati]

Thank you @rob. It worked out for the fluid domain and gave correct fluid flow.
Another issue that I am facing now is that in particle deposition UDFs I need to introduce porous jump thread and face.
All the UDFs are developed for walls, but when I hook them to my current simulation, the particle tracking won't work! I could not find out how to introduce porous jump thread and face to the UDFs!
Would you let me know how I can do that?

[Rob]

You can use PJ zones, but it's a bit more advanced. However, we're also very limited in what we can cover on here, so all I can say is read the UDF manual.

[Fluent User]

Hello,

why to use a zone for permeate?

Best wishes

[Fluent User]

Hello, you have chosen the inlet region according to the startup length? How have you fixed the 3 regions (inlet, middle, outlet)? or is it only for post-proceeding? Best wishes