I am new to Mfix-DEM (19.1.4) and want to investigate 3D particle movement through a geometry with a complex internal structure. I have not seen whether MFix can import external meshes.
To investigate I am trying to import a simplified geometry (with no internals) using a .STL file. Using a single .STL I am unable to properly select the inlet and outlet as BC`s, so decided to split the .STL into 3 parts (walls_external, inlet and outlet) and assigning the respective .STL to the correct regions. The background mesh is clipped to the internal domain, but the fluid boundary conditions gives errors.
Is there an alternative method to import complex geometries, or meshes?
I’m not exactly certain at what you are trying to do. Let’s assume that your simple geometry is a cylinder and you want flow into the bottom and outflow of the top. You want the cylinder to overhang the top and bottom of the domain extents, then assign those plans, mass inflow and pressure outflow boundary conditions.
The top is set as a pressure outlet, but get the following error. I found a work around by commenting out bc_p_g(3) and changing bc_type(3) from ‘CG_PO’ to ‘PO’ manually, but this only works if there is no overhang.
If you define a pressure outlet along a plane region that is located on the overall domain box, you should use ‘PO’, not ‘CG_PO’. It is not recommended to manually edit the .mfx file outside the GUI.
I suggest you to follow the above suggestion that Mfix need to extend the geometry above the inlet and outlet as Gbotha. The other suggestion from me is the quality of STL file that you export from the software. You need to mesh surface to create more domain in STL file before exporting to the STL file. I use Spaceclaim from Ansys to mesh and then export to STL in my case. I hope that my suggestion can help you to solve your problem.
Extending the STL file and using a plane as the inlet/outlet worked for the simplified case.
When the DEM particles are inserted ans allowed to settle in the domain, I get an error that particles were detected in ghost cells as in approaches the inlet BC. Is there a BC that allows the fluid phase to pass through, but stops the DEM particles?
