PIC not running when including species

MFiX
#1

Dear forum,

I am facing an unknown error when trying to solve a simple PIC case with two gas species. I am not interested in the chemistry but the fact that there are two gas species in my model. I set everything and it can run if only one gas phase is used but, as soon as I enable species_eq(0), the simulation stops after the first time-step stating “Previous MFiX run is resumable. Reset job to edit model”.

I have tried to enable the energy equation, but then I get as many errors as particles XX in the system stating that “The initial species mass fraction for particle XX does not sum to one. Verify that the IC region containing this particle has the solids species mass fractions defined: IC_X_s(ICV,1,:).”

I suspect this is a rather simple question, but I am unable to see where my problem is and I would kindly appreciate any help. Please see below the case I am using.

Thanks!
Eduardo

PIC_model.mfx (18.0 KB)

#2

I’m seeing a segmentation fault, in both SMP and serial mode.

Program received signal SIGSEGV: Segmentation fault - invalid memory reference.

Whenever this happens, it means a serious error has occurred. Usually this is a bug in MFiX, occasionally it is due to out-of-bounds array access due to particles being far outside the domain.

Looking with the debugger:

Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x00007f8483797be0 in des_thermo::calc_cp_des (pnp=1)
    at /home/cgw/Work/NETL/mfix/model/des/des_thermo_mod.f:118
118	      lTs = DES_T_s(pNP)

(gdb) p pNP
$1 = 1
(gdb) p DES_T_s
$2 = <not allocated>

The offending line 118 is trying to index into an array which has not been allocated. This is an MFiX bug and will be fixed in the upcoming 21.4 release. Thanks for the report.

– Charles