Tcoll/DT_Solid in SQP

Hi @cgw, Dr.@jeff.dietiker, I’m checking if I can increase simulation time-step so that it runs faster – I’m simulating dense, slow moving aspherical particles such as settled bed at UMF. Default, Tcoll/DT_Solid is 50. Reducing this, say 25 or 20, can increase overall simulation time-step? Thank you,

You will need to experiment to see if it helps. Reducing Tcoll/DT_Solid will make the simulation run faster but you may miss collisions, which will most likely make the simulation fail due to excessive overlap.

Thank you, Dr. @jeff.dietiker. I’d give a try by reducing this input value. It is not possible to track all collisions in my setup as I’m running it in batch mode but I’d look for “simulation-failure due excessive particle-particle overlap” error.

Is there any other indication to look for, which suggests this value is too low?

Best,

Hi team, continuing on this query, I tried looking to reduce Tcoll/DT_Solid using text editor on HPC but all in vain. Could you let me know what is the variable name when using text editor? Best,

it is dtsolid_factor. You can hover over a keyword in the GUI and it will show you the name of the keyword along with it’s description.

You can also use the search icon
search
to search for keywords and their descriptions

Thank you,

I think, if dtsolid_factor is unedited from its default set value of 50, this variable does not show up in text editor

#_______________________________________________________________________

DEM - Advanced

nfactor = 0
des_neighbor_search = 4
neighbor_search_n = 20
neighbor_search_rad_ratio = 1.0
#_______________________________________________________________________

I edited this value on GUI and saved mfx file. Text editor now shows
#_______________________________________________________________________

DEM - Advanced

nfactor = 0
des_neighbor_search = 4
neighbor_search_n = 20
neighbor_search_rad_ratio = 1.0
dtsolid_factor = 25.0
#_______________________________________________________________________

→ Can adding this line dtsolid_factor = 25.0 using text editor is sufficient or one has to save mfx file only after editing this value in GUI only?

Best,

You are correct, default values are not saved in the .mfx file.
Entering the value in the GUI or adding the line in the file using text editor will have the same effect. We encourage users to use GUI rather than edit directly because it is easy to make mistakes and mess up the file. But if you are careful it is fine.

Following up on this, what monitors, write-outs can be used (besides full domain gas and solid phase(s) VTKs) to check simulation worthiness when dtsolid_factor = 20.0? This increases time-step for faster progress but at the same time, we don’t want simulation-failure.