LIGGGHTS® - User Forum

Hi,

I am trying to run a simulation in Liggghts... when I run it by mpirun (with 6 threads) each processor use about much memory. The memory that have been used increased by time.

How could I reduce memory usage?

Any can help me with this: ''ERROR on proc 0: assertion failed''
_____________________________________________________________________________________________________________________
boundary m m m
units si

region domain block -3.7 2.6 -4.2 3.9 -4.5 1.95 units box
create_box 3 domain

neighbor 0.003 bin
neigh_modify delay 0 check no

Dear All
does anyone have reference for calculation of directional heat flux in the DEM?
The LIGGGHTS calculates heat flux based on distance of two particles in contact. I checked a code but I hesitate about the LIGGGHTS calculations.
Regards

Hi,

I need the properties of ROM ore (iron Ore), could anyone make it available?

Thanks.

Hi,

I was going trough the LIGGGHTS examples and had a problem after simulations 'cohesion' and 'insert stream'. I tried opening the post-files with ParaView 5.4.0-RC2 but the programm couldn't find the right reader for the post-dump files.
Apparently the dump-files are saved as dump_xxxx.cohesion and dump_xxxx.stream. (xxxx as a number like 8000)

What should i do to be able to view the results on ParaVew?

Thanks

Hello friends,

I am trying to use the mesh/tet command and so far I had no success.

I used 2 meshes in the study. The first mesh is a STL mesh. The second is a VTK mesh. Both meshes are in the same scale (mm). I scaled the meshes by 1e-3, because the unit system of the study is SI. If I run a test without the VTK mesh, everthing run ok. But if I run the study with the VTK mesh, liggghts return this error:

ERROR: VTK mesh file is incompatible with simulation box: One or more vertices outside simulation box (input_mesh_tet.cpp:298)

Hi

I'm trying to do a simulation with materials of different diameters (two materials). But I'm facing the following error:

"ERROR: Fix property/global (id m3): per-atomtype property matrix must be symmetric, i.e. N atom types require you to define N columns and N rows with N*N total values (../fix_property_global.cpp:158)"

In my settings are:

fix m3 all property/global coefficientRestitution peratomtypepair 2 0.8 0.8 0.8 0.8
fix m4 all property/global coefficientFriction peratomtypepair 2 0.15 0.15 0.15 0.15

Hi,

I can already count total number of contacts as richti83 suggested here: https://www.cfdem.com/forums/particle-contacts-using-contactatom

I'm interested in counting collisions only between atoms of different types (types 1 and 2). Is it possible?

Hello everyone,
I think, the references at the documentation of "gran model hertz model" are not correct.

I spent a whole day to have a look at the references, before using them at my thesis.
I discovered, that the references are containing similar models, but no one of them, uses this model.

More over, there are some questionable things at the documentation like the bullet point
rmu = coefficient of rolling friction

But rmu is no where to be found ...

Hi everyone,

I am managing to implement the mesh/stress/servo command in my simulation. In my case, I would like to use the command to compress a packing by a small vertical force, i.e. I just want to flatten the top surface of the packing rather than strongly compress it. In other words, I just want to use a small force to compress the packing, thus I use CtrlPV force and target_val 0.003 as the parameters.