i am currently running a 3D fluidized bed using PIC method, but it showed an error message as BC_Uw_s(3,1) should not be specified for DEM/PIC simulations, how could I solve this?
Error from check_data/check_bc_walls.f:557
Error 1101: Invalid input for boundary condition: 3
BC_Uw_s(3,1) should not be specified for DEM/PIC simulations.
Thanks!
The setting for BC_Uw_s is here:
It’s labeled as “Wall U-velocity”. You can always use the “search” feature to find the control associated with an MFiX keyword.
Note that for PIC simulations this field is disabled (“greyed out”)
Do you have a value in that field? If so how did you enter it?
Please attach your project files - go to the main menu and select “Submit bug report” then upload the resulting .zip file.
– Charles
Hello Charles,
Many thanks for your reply! i do not have a value in that field and I have attached the zip file, please have a look at it~ Thanks
3dfluidbed_2023-02-23T125802.509700.zip (138.7 KB)
Hi @TY1
This is a very interesting case - I found a few different problems while investigating it. Thanks for the report.
First - let me note that I was not able to reproduce the error you reported, regarding BC_Uw_s should not be defined, so I think the file you uploaded might not be the same version that produced that error. In the file 3DFLUIDBED.LOG there is no mention of BC_Uw_s, and any fatal errors should be echoed in that file.
If I run the case on Linux, I get a “Required input not specified” message for nmax_g:
This is a setup error - it happens because specied equations are enabled for the fluid phase, but there are no fluid species defined:
Note that nmax_g is the number of fluid species. It normally gets set by the GUI when you add/subtract fluid species, but since none were added, it never got set - this is a GUI bug that we will fix for the next release. But, since you have no fluid species defined, the easiest thing to do is uncheck Enable species equations in the Fluid pane and you will avoid this error. (This also matches the error message in the attached LOG file. When I run on MacOS I do not get the popup, but the error is in the LOG file).
After getting past this little problem, however, the simulation still won’t run for me - it proceeds very slowly, and then after a few minutes I see:
MFIX WITH CARTESIAN GRID IMPLEMENTATION.
RE-INDEXING IS TURNED OFF.
INITIALIZING VELOCITY NODES...
MFiX process has stopped
Error: Solver crash!
Unfortunately, this Solver crash! message comes with no further details. Trying again, while running top, I see that the solver uses more and more memory until it has used all available RAM and the operating system kills the process. (You can enable Resource monitors under Settings in the main menu and you will see memory use. Or you can use a system utility like top).
Looking around a bit more - I find this:
We should probably make this warning more prominent - you have to be on the right tab and sub-tab to see it. But I believe the problem is that your particles are too small and you are trying to track too many of them. You might be able to get away with this on a system with a lot of memory, or a cluster. Or you may want to consider another approach, like CGP. Others may have more suggestions.
– Charles
Hello Charles,
Thanks a lot for your reply! I reran the file and the outcome was the same as yours, I guess I accidentally resolved the error regarding BC_Uw_s. Also thank you so much for providing the other suggestions about this simulation, they really helped a lot.
May I ask if there is any tutorial about the PIC method? I was required to do the simulation using this method but not many resources can be found online.