LIGGGHTS® - User Forum

Hi there,

During a scratch simulation which consist of conical tip and flat specimen, should i update ensemble(nve/sphere) after increasing force statement of tip ?

I would like to simulate a scratch simulation in micro scale, which tip and substrate make a contact and slid (constant velocity).
And during the sliding process i want to increase a normal load for tip ( E.G, 1 mN to 100 mN).

I am a beginner in using LIGGGHTS and I am currently working on calculating the fabric tensor for my simulations. I noticed that LAMMPS has a dedicated compute fabric command for this purpose (https://docs.lammps.org/compute_fabric.html). However, it seems that LIGGGHTS does not have a similar command.

Could anyone please advise on how to calculate the fabric tensor in LIGGGHTS? Are there specific procedures or workarounds available for this?

Thank you in advance for your assistance.

I am trying to replicate a box including spheres and multi-sphere particles with periodic boundaries. When try to run the model after replicating, I get this error "ERROR on proc 0: Internal error in multi-sphere method (../multisphere.cpp:472)". Note that when I removed the multi-sphere particles from the model, the error disappeared. When I made the boundaries non periodic, the error still occurred. Here is a minimum working example:

atom_style sphere
atom_modify map array sort 0 0
boundary p p p
newton off

communicate single vel yes

Hi, hope you're all doing well.

It is very common to have many variables defined throughout the input code in LIGGGHTS. Is there a way I could create a file named "variables.txt", assign all my variables in this text file, and then read it within my input code?

Thanks in advance.

Hi,

Hi Guys

Has anyone had issues installing LIGGGHTS on Ubuntu 24.04 LTS?

My installation seems to go okay however there is an error wit regard to vtk which prevents any simulation from running. Refer to the error below.

./lmp_auto: error while loading shared libraries: libvtkCommonCore-8.0.so.1: cannot open shared object file: No such file or directory

I have tried compiling vtk separately however the issue persists. Noting that I have done the same installation on the Raspberry Pi5 running Ubuntu without any issues.

I am trying to install LIGGGHTS on a HPC and I got the following error
Creating list of contact models completed
make[1]: Entering directory'/home1/10176/padesina/LIGGGHTS-PUBLIC/src/0bj_auto'
Makefile:945: *** 'Could not determine suitable appendix of VTK library with VTK_INC="-I/home1/10176/padesina/vtk/source/build/include", VTK_LIB="-L/home1/10176/padesina/vtk/source/build/lib64/" and VTK_APPENDIX="-vtk-9.3"'. Stop.
I noticed LIGGGHTS cannot find the "vtkVersion.h" file as in the attached make_auto.log file. Any ideas on what to do will be appreciated.

Hi Guys

I believe that it should be possible to compile LIGGGHTS to use the GPU by modifying the makefile and using the Kokkos library.
Has anyone managed to do this as yet?

Hello everybody,

Hi everyone,

I was using LIGGGHTS_PFM for a while, and after that, I realized that using LIGGGHTS-WITH-BONDS, based on Potyondy and Cundall's paper, is better for my simulations. I used the same model for an uniaxial compression test, which had no errors in LIGGGHTS_PFM, as input in LIGGGHTS-WITH-BONDS, but it cannot correctly calculate the average stress and strain. Despite the fact that those commands were original LIGGGHTS' commands, not specified only in PFM, can someone help me with this matter? I will attach the parts of the code that LIGGGHTS-WITH-BONDS cannot recognize.