LIGGGHTS® - User Forum

I simulated with multisphere in the liggghts-PFM version, acquiring a portion of the specimen via the displace_atoms command and delete_atom command. Both commands work, but after continuing to run, the intercepted portion goes back to its original position. displace_atoms doesn't seem to work, why is that? Is the particle location not updated correctly?

region bc block 0 1 0 1 0 1 units box side out
displace_atoms all rotate 0 0 0 0 1 0 30
delete_atoms region bc

Hello everyone,

Please be kind enough to explain me the meaning of following two errors.

WARNING: The following model does not yield consistent results with coarse-graining: cohesion model sjkr
WARNING: The following model does not yield consistent results with coarse-graining: rolling model epsd2

Hello everyone,

I'm currently working on a simulation involving a box where the particles inside undergo a volume expansion of 100%. My objective is to calculate the critical stress within the box by applying a compressive force, when the simulation box will alternately expand and contract.

Up to now, I have managed to simulate the expansion and contraction of the box and the application of force on its walls as separate processes. However, I'm facing challenges in combining these two simulations to run concurrently.

Hello mates,

Is there any paraview plugin for the LIGGGHTS_Flexible_Fibers available for newer paraviews? The one from Richti83 is for the Paraview 5.8.1 which is apparently too old for our IT-dept. in the company (only 4 years old bullshit). They are ignoring to download the one from Richti83:

https://github.com/richti83/ParaView_Reader_for_LIGGGHTS/tree/master/pre...

Any hint would be highly appreciated. Thanks in advance

HI,

I'm trying to implement a rigid body dynamics problem in LIGGGHTS. I am using variables as inputs for the fix move/mesh commands. For the linear/variable style there's no problem, however, the rotate variable style only accepts variables as inputs for the angular velocity omega. I would like to also use variables for the inputs of the origin and axis arguments. Is there a way of doing this?

Thanks for your help.

Best regards.
Miguel

Hello,

I want to use spheres and one superquadric (cylinder) in my simulation. However, it turned out to be illegal when I tried using two types of atom (atom_style granular; atom_style superquadric).

I know I can still use superquadrics to represent sphere (just set both "blockiness" as 2), but, in this case, the rotation is not updated and the angular velocity is always 0.

I use nve/superquadric (scheme 1) for integration:

fix integ bed nve/superquadric integration_scheme 1

So, how should I deal with either problems?

Thanks

Hello, everyone,

I have been working on a simulation that involves the random compaction through uniaxial pressing of particles with varying size distributions, using two different material types to create a composite. Up to this point, the process and simulations have been straightforward, and LIGGGHTS has been an invaluable tool.

Hello,

I am currently facing challenges in resuming a simulation that involves superquadric particles formed by gravitational settling.

1. My initial approach was to utilize the "read_restart" command in a subsequent script file. Unfortunately, this method does not allow for resizing the simulation box with the "change_box" command.

2. My second attempt involved the "write_data" and "read_data" commands. However, I discovered that these commands do not support superquadric particles.

Hi everyone,

I'm very new to LIGGGHTS, and I'm having troubles to create a simple simulation.

As a starter, I want to create a simulation where a single particle (metal particle) is settling in air and another simulation where a single particle is settling in viscous medium (like a resin).
When a particle falls in air, I just want to have force of gravity acting downwards on a particle.
When a particle falls in a viscous medium, I want to have drag force acting upwards on a particle, while force of gravity acting downwards on a particle.

Hi Everyone,

I am analyzing a sieve geometry for particle transfer. The CAD model of sieve plate is complicate with around 1 Mio. of tessellated faces. Hence, reading and "import and parallelization" of this taking huge amount of time. Is there any alternate way to speed up this time? I can't use the recently introduced model "massflow/mesh/sieve" fix command because most of modelling data require here is unknown.
I also tried converting the STL file to VTK using Paraview tool but it also did not work.
Any suggestion that I can try?

Regards