How to import .msh file generated by other softwares

MFiX
#1

Dear All,

I want to know whether we can import .msh file generated by other (non)commercial FEM softwares. And since we have to set the regions before mesh under SMS mode, whether some features configured in region will be embedded in the .msh generated by Mfix so as to incur differences from .msh file generated by other softwares.

And if the above can be done, whether the rest of simulation procedure would be the same, and what is the possible difference?

The last question I want to ask is how to tackle with the model with delicate struction the mixture of gas-particles will pass through like tubes and holes with diameters as small as 0,5mm under the chamber of about 50mm length in X-Y-Z directions to simulate interaction between gas and solid particles as small as (1e-5, 1e-4). Do you think such kind of relative scales of different materials (domain scale 1000 times than delicate struction, delicate struction 10-100 times solid particles) is feasible for MFIX? And what is the best numerical method (TFM/DEM/PIC) and mesh technique(small cell/area tol, snap tol, dot product tol, intersection tol) for this situation? I would quite appreciate if you could provide me some precious insights based on your rich experience on the multiphase fluid simulation of varying scales. Thank you.

#2

The .msh file format is specific to MFiX and there is no other software that support it.

This geometry may be too challenging for a Structured mesh code with cut cells such as MFiX. It is difficult to locally resolve tiny geometric details, especially internal tubes with thin walls. As for the method, it depends on the number of particles and particle size. DEM is typically prohibitively slow when the particle count is large (say more than a few millions) and/or the particles are small (say below 200 microns). In that case TFM or PIC are preferable.

#3

Thank you Jeff. I am currently trying to do meshing for chamber with delicate structures. Sometimes, the quality of generated mesh by either control point or uniform method is bad, and sometimes, the “index, dimension 1, above upper bounds” type error would happen for some configuration of cell numbers in each direction or points position for control point method.

And as you say, the cut-cell method is challenging for this kind of model, whether the other meshing methods can be selected. I hereby attach the .stl and .mfx files. Thank you.

chamber_delicate_structure1.mfx (8.3 KB)
chamber_delicate_structure2.STL (766.9 KB)

Resolution of converted .stl from CAD software and related tricks to do meshing