CFD-DEM Simulation with Thin internal device – Mesh Size Limitation

Dear MFIX Support Team,

Hello. I am currently working on a CFD-DEM simulation of a spouted bed that we have built in our laboratory. However, I have encountered a challenge and would greatly appreciate your guidance.

In our setup, there are draft plates installed inside the bed, each with a thickness of 1 mm, which matches our experimental device. The particles used in the simulation have a diameter of 3 mm. In order to accurately capture the effect of these thin plates, I need to refine the mesh to a thickness of around 1 mm near the plates (I am using control points for mesh refinement).

However, when I try to run the simulation, I receive a Error from MFIX stating that the mesh size is smaller than the particle diameter, which is technically correct. I understand this limitation, but I was wondering if there are any suggested workarounds or shared experiences from the community on how to deal with such thin internal structures in CFD-DEM simulations.

Any suggestions or recommendations would be highly appreciated.

Thank you very much for your time and support.

Do you use an .stl as input, or build it in MFiX GUI? I don’t think you need a small cell size here to account for the thin walls. If you use “large” cells, then they’re split with the .stl model, supposing that’s what you’re using. But I also guess it could happen that even the split cells can be small - in that case you just have to make sure that your original cell size is sufficiently large.

Also, images of you model and setup will increase you chance of getting help. You can paste screenshots into the text input field, it’s quite easy.

1 Like

The mesh size cannot be smaller than particle size, in fact to be safe make it >=2.5d. But this is not a problem, because cutcell method can be used if needed. Copy and paste your entire input files including *mfx, *csv, *stl, ant *f you modified etc (zip it all), so we could reproduce the simulation and advise you.

1 Like