LIGGGHTS® - User Forum

Hi,
Does granular pair style output force info to compute pair/local? As LAMMPS doesn't have such capability.
Thanks,
Ram

Hi Christoph,
We can assign the properties of mesh wall/sphere from atom_type value. Can we assign density of that mesh wall/sphere like this way?
....
create_box 2 reg
group sphere type 2
set group sphere density 2500
...
..
fix cad all mesh/gran/stressanalysis stl/sphere.stl 2 0.1 0 0 0 0. 0. 0.
...
...
Will it change the density of sphere.stl?
Thanks
Ram

Hi Christoph,
Can I run two different simulations at same time on same system? Will there be any effect on the outputs of the simulations?
Thanks,
Ram

Hi Christoph,

In the class atom, the sphere positions are sorted, it exists the array j = tag[i] to indicate at the line i you will have the sphere number j. I did not find, if it exists an array such as : i = invtag[j].

This array invtag[] is important in my case, because I load the interactions pair (Si, Sj) and I have noticed that the values of x[Si] do not correspond to the S1 poistion loaded via my file *.sys.

Please, can you help me to find this array (invtag) or desactivate the sorting only when I want to load the interactions file.

Hi Christoph,
I want to modify source code which can calculate net force on whole mesh moving in granular particles. That mesh is spherical type. Currently, we can measure force on triangles of the sphere. Can you give me some hints for doing this?
Thanks,
Ram

Hello all,

I would like to simulate particles in a container that are stirred by to stirrers. The container is available in an stl file and so are the two stirrers. So, in total, I've got three stl files. I assume that splitting the whole geometry into 3 parts is necessary, since the two stirrers are supposed to have different angular velocities and the container is at rest.

While importing the second stl and and applying the fix wall/gran/hooke I get the following error message:

ERROR: There must be only one fix wall/gran with style 'mesh/gran'

Hi Christoph,

I am developing a new contact model. In my case as you suggested, I took the pair_gran_hooke_history like base.

For the first step, I only replaced the name of class PairGranHistory by PairLatticeHistory.

I have two files pair_lattice_history.cpp and pair_lattice_history.h. In the file .h, I change:

#ifdef PAIR_CLASS
PairStyle(lattice/history,PairLatticeHistory)
#else
#ifndef LMP_PAIR_LATTICE_HISTORY_H
#define LMP_PAIR_LATTICE_HISTORY_H

Thus into my simulation file, I can use this key word: pair_style lattice/history 1 0

Hi all,

I have no idea how liggghts generates stress-control compression tests, although I know that liggghts has a powerful engine to conduct strain-control ones. However, it is necessary to do stress-control if want to conduct drained triaxial test because I need to maintain the confined stress constant.

Could someone told me how to do stress-control compression or give me some suggestions?

Thanks!

Best,
Tong

Hi all,
I think there are some problems with the implementation of the tangential spring in the pair_style gran/hooke, especially for soft particles where the overlap between the particles gets high.
So far the relative rotational velocities between 2 particles are calculated by using the radius of the particles:

wr1 = (radi*omega[i][0] + radj*omega[j][0]) * rinv;

The force application point for the tangential forces also lies on the radius of the particles:

torque[i][0] -= radi*tor1;

Hi all!
The region command in liggghts is static but it becomes dynamic when we apply wiggle command. Is this possible that we apply the wiggle command but the region should remain static. Can anybody help?
sajjad