Continuing the discussion from Error prompt. DT ＜DT_MIN. Recovery not possible!:
Hello everyone,
I am running a 3D fluidized bed with heat transfer in SQP. The calculation fails to converge at the initial step. I tried to adjust the time step size and some physical parameters, but still getting the same issue: DT< DTmin. Could you please check my case file so that i can get some clue for not converging my simulation? I have attached my case file here.
fbar4.mfx (16.2 KB)
Your case appears to be missing the particle input file. Can you please attach your full case setup?
(GUI→Menu→”Submit Bug Report”→Upload .zip)
I am running this on the HPC cluster, and particle input files are there. I have attached the full case file.
AT100v1.8.zip (6.1 MB)
It looks like the particles you are seeding from your input file are causing overlapping issues with the wall.
Using gdb:
Program received signal SIGFPE, Arithmetic exception.
…/mfix/model/des/des_thermo_cond_mod.f:586
586 R_in = sqrt(2.0D0OLAP_ACTUALRpart-OLAP_ACTUAL*OLAP_ACTUAL)
print Rpart
0.006055
print OLAP_ACTUAL
0.0136
Which results in a negative value passed to the square root leading to the error.
I would recommend taking a second look at your particle inputs file and relax the spacing from the walls. For a quick test, I increased the min/max bounds of your geometry in both the x- and z-directions (-0.05, 0.2) and the simulation is running.
Thank you for your suggestion. I will try to relax the space from the walls.