DEM case with reactions diverge

MFiX How to
#1

Hi, guys,
I want to use the DEM with 2 gas reactions and 2 solid reactions to simulate a 2-D fluidized bed.

case description:
It is a simple case with a gas inlet orifice in the bottom, and pressure outlet in the top. The reacting particles(biomass) is injected from the orifice. Initially there are 98(for testing purpose, I set the number small to avoid spending too much time) sand particles serving as bed materials.

my problem:
It is weird that when I only add 2 gas reactions, the simulation is good; when I
only add 2 solid reactions, it is good as well; HOWEVER, when I add both 4 reactions, it kept diverging.

Here is the case set-up:
silane_pyrolysis_pic_3d_2019-10-18T174245.298872.zip (630.9 KB)

Best regards

1 Like
#2

I have a few suggestions,

  1. You should do a solid phase check in usr_rates_des.f and make sure only the biomass reacts, not the sand. (PM should be 1).
  2. The injected biomass temperature is very high.
1 Like
#3

thank you
Could you be more specific about how to make sure only the biomass reacts?

It that means adding a condition statement before defining the des_rates() variable? I.e., something like this:

if ( PM==1)
des_rates(Rs_CO2)=…

#4

HI,
I try to remove the sands and see whether it will diverge or not. Then it stops diverging, but I cannot see the product gas being generated.
Here is the new case file
gas_and_solid_2019-10-22T153911.004027.zip (321.7 KB)

#5

In the usr_rate_des.f, when you calculate the heterogeneous reaction, add if(PM==1).

#6

According to your run log file, GAS_AND_SOLID.log, you only ran it for 0.12710 s and there are only 2 biomass particles, (Total PIP=2), and the particles temperature should still be low, so the reaction is slow now, so my suggestions is run the case longer.

#7

I modified my case.
gas_and_solid_2019-10-23T160010.167358.zip (72.7 KB)
And I tried run the case to 1s, but it is very weird that if I try to inject the CO or CH4 gas from the bottom inlet, it will diverge quickly.

#8

Turn the stiff_chemistry on (stiff_chemistry = .TRUE.)
There are few more things:

  1. heat of reaction: unit is Joules/kmol
  2. Since you use SI unit, make sure the unit of the homogeneous reaction rate in usr_rates.f is kmol/s/m3, and the heterogeneous reaction rate in usr_rate_des.f should be kmol/s.
2 Likes
#9

Following your suggestion, it did stop diverging. Thanks.

But currently the problem is I cannot witness any solid reactions.(the last two reactions)
%E5%9B%BE%E7%89%87
gas_and_solid_2019-10-24T123600.908880.zip (20.6 MB)

#10

I am not quite familar with the stiff_chemistry, but I found only if turned on the stiff_chemistry, it will not diverge.

#11

By the way, I am trying to set up the case with the same simulation conditions of this paper.
CFD-DEM_Xiaoke_2014_Norway.zip (2.3 MB)
I want to start with a simlified version of it : no radiation, no turbulence, small number of particles and inject the product gas of pyrolysis directly.

#12

I found my mistakes:
use 1/12.0 rather than 1/12 in the des_rates defining line