Simulation of drying wheat in fluidized bed

MFiX
#1

I want to model the drying of wheat in fluidized bed. I choose H2O_cr as moisture in wheat and gas H2O in air. I set the initial mass fraction of H2O in wheat 0.25 and in air 0.015. for the boundary condition I set the mass transfer of gas species to 0.015. Now I have 2 questions:
1- in the outlet which is pressure outlet, I do not know the mass fraction of h2o, then what should I do?
2- Do I need the thermal properties of wheat like heat of formation, molecular weight and viscosity?
Also, when I enter some values for the wheat properties and run the simulation, I encounter this message and nothing happen, no run and no error even!!!

INFO check_data/check_solids_common_all.f:326
Message: 2000 The energy equations are being solved (ENERGY_EQ) and
the constant solids specific heat is undefined (C_PS0). Thus, the
thermochemical database will be used to gather specific heat data on
the individual gas phase species.

INFO check_data/check_solids_common_all.f:344
Message 2001: One or more species molecular weights are undefined and
the solids phase species equations are solved (SOLVE_EQ( 1)). The
thermochemical database will be used in an attempt to gather missing
molecular weight data.

INFO check_data/check_solids_common_all.f:360
Message 3000: Searching thermochemical databases for solids phase 1
species data.

1: Species: H2O_cr
Checking for H2O_cr in mfix.dat :: Found!
2: Species: wheat

What are these about?dryer-1st.mfx (15.0 KB)

2 Likes
#2

Hi. I’m reviewing unreplied questions on the MFIX forum and I found your question. Sorry for the delayed reply.

The ‘searching thermochemical database’ messages are routine. They do not indicate any problem - this is just MFIX searching for, and finding, the definitions of your species.

When I run your case in MFIX 20, I get an error from the solver. MFIX version 19 was not as good about catching and reporting solver errors, this has been improved in version 20.

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

Backtrace for this error:
MFiX running: elapsed time -1:59:59  
#0  0x7feb94cfec02 in ???
#1  0x7feb94cfde45 in ???
#2  0x7febe180d67f in ???
#3  0x7feb9520112d in physical_prop_cps
	at /home/cgw/Work/NETL/mfix/model/physical_prop.f:495
#4  0x7feb952017ee in physical_prop_
	at /home/cgw/Work/NETL/mfix/model/physical_prop.f:85
#5  0x7feb950709a8 in __calc_coeff_mod_MOD_calc_coeff
	at /home/cgw/Work/NETL/mfix/model/calc_coeff.f:341
#6  0x7feb95057fdb in __main_MOD_initialize
	at /home/cgw/Work/NETL/Cases/dryer-1/build/pymfix/main.f90:436
#7  0x7feb951e6bce in run_mfix_
	at /home/cgw/Work/NETL/mfix/model/mfix.f:132
#3

At line 495 in physical_prop.f we find

  C_PS(IJK,M) = C_PS(IJK,M) + X_s(IJK,M,NN) * &
       (lCp * RGAS / MW_s(M,NN))

so this looks a lot like a zero-division error - you have not set a non-zero molecular weight for the ‘wheat’ species and this is probably the cause of the error.

We are always working on improving error detection and handling in MFIX. We should probably have a check for non-zero molecular weight to avoid this situation.

Thanks for the bug report & sorry for the delayed response.

  • Charles
1 Like