Hi all, I am trying to set up a simulation where particles are inserted in an extreme of a rectangular pipe. I have inserted them either with constant and gaussian velocity but when I visualize the results some of them start their motion with an extremely high velocity.
From the other hand I have been struggling trying to put a movile surface that compress the particles (along the z axis). So far, I have found that I should use servo command, but when I add to the DEM file the folowing instruction:
fix servo all mesh/surface/stress/servo file ../DEM/cad/movingwall.stl type 1 com 0. 0. 0. ctrlPV force axis 0. 0. 1. target_val 10 vel_max 1. kp 0.02
error messege interrupts the simulation (log file is attached). I will be thankful to recibe any suggestion.
Feder.
(I attached the set up and a video of the simulation for further references).
| Attachment | Size |
|---|---|
 new.tar_.gz | 1.6 MB |
 log_base_case_mondalib.txt | 274.07 KB |
Ermek Asylbekov | Thu, 03/29/2018 - 08:08
No stl Files
Hi Feder,
since you didn't give us the stl-files I couldn't rerun the case (in the future try to give us a working example). I didn't see any terrible mistakes in the code by looking through. The DEM timestep might be a little to large (20% of the Rayleigh step would be 1e-6) but I don't think that that is the problem. Since it only occurs at some particles and not all of them its really hard to tell without running the case by myself.
Regarding the moving plane:
Knowing the error message might be very helpful ;)
I would solve this via move/mesh command like (replace [fixID] by yur ID etc):
fix [fixID] all move/mesh mesh [mesh_name] linear [vx] [vy] [vz]
downside on this method is that its not controlled by the force applied.
If you upload the whole (working) case(s) I could look into it.
Regards,
Ermek
federicogaspar@... | Tue, 04/03/2018 - 20:47
Ermek
Hi Ermek! Thanks for taking a time to answer me. I attached the full case in the previous messege (dont know why I could not upload any file here) as well as the log file in the case of the error when I try to run with the servo command.
j-kerbl | Wed, 04/04/2018 - 14:42
Hi Feder,
Hi Feder,
I've taken a look at your case and the error message. So far my findings are inconclusive. I could not reproduce the segfault.
I've taken your case, uncommented the servo wall, removed the zwalls fix and added the following line below the servo:
fix walls all wall/gran model hertz tangential history mesh n_meshes 2 meshes cad1 servo
If you do have another problem here, please add an "echo both" command at the top of the LIGGGHTS input script and post the log again.
Cheers,
Josef
federicogaspar@... | Fri, 05/04/2018 - 17:31
Me again.
Hi! I been trying to take back this case and I realized that the problem with the insertion of particles was related with the periodic boundary conditions, but I still can not run this case with the servo command. Although, (taking into account your suggestion) I give a step forward, the simulations runs but few steps later the following error comes up:
ERROR: Mesh (id servo): Mesh elements have been lost / left the domain. Please use 'boundary m m m' or scale/translate/rotate the mesh or change its dynamics (../multi_node_mesh_parallel_I.h:680)
Indeed, when I visualize the case, four or five particles leave the domain.
At last I would like to ask how should I visualize the moving wall in paraview. I observed in an example case that people added the folowing command
dump dumpstl all mesh/stl 200 post/dump*.stl
but I could not reproduce their results.
P.S.: I attached the case file and the log in the first comment.