Do not know the reason for high run-time of the MFiX-SQP case

MFiX How to
#1

Hello,

I hope you are doing great.

I am trying to run a MFiX-SQP case (
sample_SQP_case.mfx (11.0 KB)

with the particle_input.dat file (particle_input.dat (80.7 KB))
However, the case is running without an error in the Ubuntu system, it is taking unusually high run-time for just 528 particles, (an estimated time in days) .
Can you please let me know from the attached files what the reason for it could be, and also how I could make it work to run faster?
I am looking forward to hearing from you soon.

Sincerely,
Sumanth Penmetcha

Installation problem for MFiX 24.1.1 and 24.2 on OSX
#2

The time step is 6E-9 s. Please try to lower the Youngs modulus to see if this helps. You should also use the grid-based des search, not the N-square.

#3

Hello,

Thank you for your response.
Can you please tell me how do I change the time step, and how does that affect the solver run time?
I am looking forward to hearing from you soon.

Sincerely,
Sumanth Penmetcha

#4

The solids time step is a function of the material properties and is computed automatically. If you lower the Young’s modulus, this will decrease the time step. It is not a linear relationship, you will have to experiment with it. The run time is directly related to the time step (number of DEM iteration equals the final time divided by the solids time step).

1 Like
#5

Hello Jeff,

Thank you so much for your solution, I will try to use that method to solve the case.
Instead of a particle_input.dat file, I am trying to use automatic particle generation to see if that might also help.
Also, can you please tell me if there’s a way to know the mass of each SQP particle generated in MFiX through the “automatic particle generation”?

Once again, thank you for your help.

Sincerely,
Sumanth Penmetcha

#6

The SQP designer shows you the volume of the particle, you can find the mass by multiplying by density.

image
image753Ă—515 84.2 KB

You can also save the particle mass in the vtk file:

image
image580Ă—943 120 KB

#7

Hello Jeff,

Thanks for your response.
What if I am enabling the automatic particle generation using the following particle size distribution.

image
image855Ă—511 21.9 KB

and I am generating around 500 particles, and I need to find the mass for each of them separately, as the bounding diameter might vary for each of them due to the distribution, what is the better way to calculate the mass for each particle?
Also, do you recommend reading the vtk file from Paraview? or through postmfix as suggested in this link, 7.2. Post Processing — MFiX 24.2.1 documentation

Also, the particle_output.dat file prints out the diameters, SQP a,b,c parameters, so if that information is enough to calculate the volume of each individual particle, what’s the formula to calculate the volume for SQP particle, that mfix uses?

Any suggestion would be beneficial.
Thank you so much.

Sincerely,
Sumanth Penmetcha