You can create the stl file with any software. Then you import the stl file in the MFiX GUI and MFiX will generate the mesh. Please see the tutorials in the documentation for some examples. There is no other software that can generate the MFiX mesh in advance.

Thank you for your reply. I would also like to ask how the GUI interface generates particles of two different diameters and the number of particles it wants。

I encountered the following error

Previous MFiX run is resumable. Reset job to edit model

MFiX process has stopped

DES geometry pre-processing complete.

Initial DES Particle array size: 4

Message from set_bc0.f:564

Outflow BC defined: IJK_P_g remaining undefined.

Generating initial particle configuration:

Phase 1: Number of particles to seed = 254

Phase 2: Number of particles to seed = 31

|-------------------------------------------------------------------|

IC Region: 1 |
---|

Phase |

ID |

------- |

1 |

------- |

2 |

------- |

Message from des/generate_particles_mod.f:692 |

Initial condition 1, Phase 1: Successful seeding |

Message from des/generate_particles_mod.f:692 |

Initial condition 1, Phase 2: Successful seeding |

Total number of particles in the system: 285 |

Beginning DEM settling period: 0 steps. |

DEM settling period complete. |

Message from des/set_bc_dem_mi.f:273 |

Error 1200: Unable to impose a uniform solids velocity on BC 2. |

Setting all solids to the highest velocity results in a large error |

in the solids volume fraction. |

```
Max Inlet Velocity: 0.0000E+00
Total BC_EP_s Error: 2.0000E-02
|-----------|--------------|--------------|--------------|
| Phase | BC_EP_s | Calculated | ABS Error |
|-----------|--------------|--------------|--------------|
| 1 | 1.0000E-02 | 0.0000E+00 | 1.0000E-02 |
|-----------|--------------|--------------|--------------|
| 2 | 1.0000E-02 | 0.0000E+00 | 1.0000E-02 |
|-----------|--------------|--------------|--------------|
```

Reading mesh stats: /home/dyfluid/MFIX/Particle inertial migration/1/MESH_STATS.LOG

Error: Solver crash!

Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.

Backtrace for this error:

#0 set_bc_dem_mi_mod_MOD_set_bc_dem_mi

at des/set_bc_dem_mi.f:317

#1 set_bc_dem_mod_MOD_set_bc_dem

at des/set_bc_dem.f:57

#2 main_MOD_initialize

at main.f:374

#3 run_mfix

at mfix.f:132

#4 main_MOD_run_mfix0

at main.f:79

Floating point exception (core dumped)

Uploading: set_bc_dem_mi.f…

Uploading: set_bc_dem.f…

Please upload your project file(s).

Here are the four related project files. Could you please help me find out what the problem is

set_bc_dem_mi.f (19.7 KB)

set_bc_dem.f (2.1 KB)

mfix.f (14.1 KB)

main.f (27.6 KB)

I need the `.mfx`

file. Please go to the GUI’s main menu and select “Submit bug report”. This will save a zip file with all necessary files. Please attach this zip file. Why do you need to modify the source code?

Hello, the problem I’m having right now is DT < DT_MIN. Recovery not possible! But I can’t find the cause of the problem, I shared my error report, hope you can help me

1_2023-08-08T103414.113454.zip (11.0 MB)

Please try to increase the number of pressure equations iterations (Numerics>Linear solver tab) to see if it helps. You can also try to extend the stl geometry in the z-direction so it overlaps with the MFiX domain box and apply wall BCs along the front and back planes. This will give you a cleaner mesh which may improve convergence.

Hello, how to modify the number of iterations of the pressure equation? Because I just contact mfix software, there are still many things I do not understand, I hope you can inform me, thank you

Is it the value in the following picture? I don’t know whether to change from the gui or from the code file

Everything can be done from the GUI. When you press “Save” the GUI writes out the `.mfx`

file.

How to “extend the stl geometry in the z direction so that it overlaps the MFiX domain box and applies the wall bc along the front and back planes”.I don’t really understand what to do

When you generate your stl file, scale it in the z-direction so it goes from z=±0.05m (instead of the current ±0.04). Then reload the stl file in the GUI.

You will also need to set wall BCs along the z=zmin and z=zmax planes.

Continuing the discussion from MFIX geometry meshing:

Thank you very much for your help. After I modified it according to your suggestion, it can run normally. But when I studied particle separation, neither particles nor droplets appeared. The simulation results did not change from beginning to end. How can I solve this problem?

1_2023-08-13T120532.468826.zip (20.2 MB)

Hi @xiaoyingtao - I’m trying to run this case to investigate the particle injection, but I can’t get it to run at all - I get this backtrace

```
Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation.
Backtrace for this error:
#0 calc_resid_mod_MOD_calc_resid_u
at calc_resid.f:800
#1 u_m_star
at solve_vel_star.f:322
#2 solve_vel_star_mod_MOD_solve_vel_star
at solve_vel_star.f:134
#3 iterate_MOD_do_iteration
at iterate.f:255
#4 run_fluid
at mfix.f:188
#5 run_mfix
at mfix.f:142
#6 main_MOD_run_mfix0
at main.f:84
```

after just a few seconds of run time. Is this case running for you? Did you upload the correct version of the file?

I’ve solved the particle injection problem. Thank you for your earnest answer. I would like to ask why particles seem to move in air in vtk when fluids are set in GUI? Is there any way to see the fluid

Can you please upload the current version of your project? I can’t get it to run at all.

With DEM, the fluid phase is continuous, while the solids phase is discrete. You can visualize the fluid phase variables as cell data. In your case this is stored in the “BACKGROUND_IC” vtk files.

Thank you very much for your help, which has solved a lot of problems that have been bothering me for a long time. If you encounter the problem of particle blockage, should you adjust the particle diameter or the particle speed? Both of these methods seem to pause my project and need to start from scratch. Is there any way to solve the particle blockage problem without losing the currently running project process?