Hello everyone, I’m rocky into mfix and I’m trying to generate the mesh to my system with a figure stl that was imported, but somehow the mesh is made inside the figure (where I need it ) and outside as well (I don’t need it there). I tried changing the normals directions but it doesn’t work. And also, the other time the program made a mesh into the walls of the system which does not sense in my study. Anyone knows how to fix it? or should I build a simplified figure with the tools of mfix? Below there are the imahes of how is the mesh
Hi Paula, please upload your files if you need help. Easiest way is “Submit bug report” from the main menu which will create a ZIP file you can upload here.
Hi, here you can find the zip file
I don’t fully understand your geometry and what is the role of the cone (there seem to be a small gap between the cone and the stl file. We typically do not include the wall thickness in the stl file, unless both internal and external flow is modeled. There will be fluid cells in the regions where the stl normals point to. Here the stl file has a small wall thickness with normals pointing inside and outside the vessel, this is why you have fluid cells on both sides. If you just need internal flow modeling, you will need to modify your stl file so you only have the inner shell of the geometry. If you are new to MFiX, my recommendation is to get familiar with the meshing and setup with simpler geometries so it is not overwhelming.
Hello, I used a simplified shape, but now it won’t let me run it. This is the error:
Message from set_icbc_flags.f:196
Error 1100 (PE 0) : No initial or boundary conditions specified inthe following cells:
I J K
2 2 1
3 2 1
4 2 1
5 2 1
6 2 1
7 2 1
8 2 1
9 2 1
10 2 1…
Error from set_icbc_flags.f:211
Fatal error reported on one or more processes. The .LOG file may contain additional information about the failure.
final23804_2026-04-23T172135.596676.zip (44.0 MB)
There is an issue with not having the default walls. Enabling default walls fixes the error. What BC do you want at the z=zmin plane?
I tweaked the mesh with a default NSW at z=zmin, and turned off k-epsilon for troubleshooting. Running DEM with 100 micron particles in going to be very slow and even slower with Hertzian model. I switched to LSD to get you started. Please try the attached.
Hello, thank you. It runs now; however, it takes some time. At the beginning, it didn’t allow me to write the particle data output file (there was some error with the file) at a certain point during the run, but I removed that part and now it works fine.
This was the error that I received:
WRITING VTP FILE : SOLIDAPI_0004.vtp
WRITING VTP FILE : SOLIDEXP_0004.vtp
float overflow in vtp_MOD_write_scalar_in_vtp_bin at des/vtp_mod.f:2192
By the way, I was thinking of running this case in MFiX-Exa so that the calculation could be faster. Do you know anything about it? Is it a good idea?
Do you have more details about this error message? A float overflow suggests some quantity became unphysical, but I cannot tell without more context (please attach your setup if you need further assistance).
Please feel free to use MFiX-Exa if you want. It is a different code so you will need to redo the setup from scratch.