Some Errors and Questions

MFiX
#1

Hi,
I appreciate if you answer the following questions:

1- I use the Mixed wall BC and set the partial-slip wall type for the solid phase to use the JJ bc and no-slip wall for the fluid phase. when I save my data and reopen it again in GUI, the wall type has been changed to the Free-slip while the JJ partial-slip BC is yet ticked. Is it OK?

Th *.mfx file shows this for one of the walls:

Boundary condition 3: Left wall

bc_type(3)  = 'PSW'
bc_x_e(3)   = 0.0
bc_x_w(3)   = 0
bc_y_s(3)   = 0
bc_y_n(3)   = 0.7
bc_z_b(3)   = 0
bc_z_t(3)   = 0.015
bc_jj_ps(3) = 1
bc_uw_g(3)  = 0.0
bc_vw_g(3)  = 0.0
bc_ww_g(3)  = 0.0
# Solid 1
  bc_c_theta_m(3,1)  = 0.0
  bc_hw_s(3,1)       = 0.0
  bc_hw_theta_m(3,1) = 0.0
  bc_uw_s(3,1)       = 0
  bc_vw_s(3,1)       = 0
  bc_ww_s(3,1)       = 0

2- I am able to get reasonable values of solid velocity from .RES file using postmfix but it shows O(10^38) numbers in paraview. What is the problem?

3- When I want to use restart_2 to use a data as initial condition in GUI, it deletes the .RES file and I am not able to uncheck the box.SO, I will use the terminal to run my case. How should I use GUI in this case?

4- Do different values of .RES file write intervals and .SP write intervals OK?
Thanks,
Mohsen

#2

Hi Moshen -

It’s OK to post multiple questions like this but in the future you might find you get better responses if you make one post per issue with a more descriptive topic line - it’s also a little easier to track that way.

  1. This looks like a bug in the GUI. Can you attach your entire case file?

  2. No idea, can you attach the .RES file if it’s not too big?

  3. This is a known bug in the GUI, will be fixed in upcoming 21.2 release.

  4. It should be OK to use different write intervals for these - try it and if it leads to any problems, let us know.

– Charles

#3

Dear Charles,
Please find the attached RES file. I can use postmfix to time average the V_s between 20 to 69 seconds but paraview shows nonesense values for the particle y-velocity after t=30s.
Also, If you check the boundary condition of the walls in GUI, you will find out that FS is shown on the wall BC for the solid particles.
-Thank you
LAVERMAN-2D.zip (1.8 MB)
4.mfx (18.6 KB)

#4

Hi Charles,
I faced the same issue with the .Res file in another case. What do you think of the issue? postmfix shows in range values but paraview shows the values with 10^32 order of magnitude for the solid y-velocity.
As you can see in the attached consecutive contours, the gas velocity magnitude suddenly crashes. (40.59 and 40.60 seconds

40.59
40.591380×541 35 KB

40.60
40.601380×541 20.8 KB

)
LAVERMAN-2D.zip (2.1 MB)

#5

This is a Paraview issue. Some versions can’t display data if the file is very large (above 1 or 2 GB). The transient data is stored in the .SP* files (not the .RES restart file, which is only a single snapshot). I ran your case up to 60 seconds and the .SP3 and .SP4 files are over 3 GB. My version of Paraview (5.6) on Linux can handle these files just fine though. Either try another (newer) version of Paraview or only load one variable in Paraview (you can uncheck variables you want to load upon opening the file). You can also decrease the frequency of the file writing.

We also recommend using the vtk files instead of the legacy .SP* files if you want to visualize transient data. This will store a series of (relatively) small files rather than a single large file.

1 Like