LIGGGHTS® - User Forum

Hi everyone,
I hope you are all well.
I have a question about Liggghts ability to fix the particles in their position but update their velocities. We know liggghts has the following command:
fix 3 wall nve/noforce
which does the opposite thing I need, changing particles position but fixing their velocity. I need to keep the particles in their initial position, but update their velocities based on forces. Now I am using the following command:
fix integr all nve/sphere

Hello all,
Is the Hertz model given in model same as Hertz Mindlin?

Kindly let me know.

Regards,
Tarun

Hello,

briefly about my problem, i have a cylinder where i fill particles and move this cylinder to mix the particles. I quickly notice that even with small numbers of particles (about 10,000) it takes a long time to simulate. To better see where the most time is spent I ran the script with modify_timing verbose.

Hi everyone I am new to LIGGGHTS and I am getting for scale factor 10 particle position x,y,z 10 times bigger and for scale factor 100 position 100 times bigger compared to Scale factor 1. It is obvious. But I want dump files again to scale factor 1 from 10 or 100. I think that dump_modify is not suitable for dump custom command. I am also uploading pictures of dump files for factor 1, factor 10, and factor 100. If anyone has an Idea about that it is very useful for me and I will be very thankful if you can share it with me. Thanks in advance.

Dear all,

i want to evaluate the number of contacts of a group of particles with themselves. Basically i want to know if a group of particles stays "intact" all the time or gets "fractured" over time. For this i found the interesting compute "group/group" command. Unfortunatly when i use the command i get the error message:

"Pair style does not support compute group/group"

In the documentation it says there are no restrictions to this compute, so i'm curious what the reason can be. I'm using the pair_style gran model hertz ...

Hi Everyone,
I am a newbie using LIGGGHTS. My problem would be similar to a steady state flow of spherical particles through a hopper domain with conical bottom, so I took it as a start. I have been able to run a simulation where I inserted the particles, and let them fall under gravity (picture attached)
However, I have not been able to run it in a "Recirculation" mode where the particles leaving the domain would be reinserted from the top. Can any body help me out with a means to do that.

My issue is that I have ran three different codes keeping everything same but only these 3 commands different in each code

fix gravi all gravity 9.81 vector 0. 0.0 -1.0.
fix p all addforce 0 0 -9.81
fix p all setforce 0 0 -9.81

So, the code with gravity force command acts like the particle comes down very slowly but in their 2 commands i.e. (add-force and setforce) the particles leaves the box quiet early, My question is that in all three cases the particles should act similarly as the force is same in three cases? Is the issue seems to be in fix?

Hello,
I am just starting to look into simulation with liggghts for my studies.

Is there any way to output the distribution of CPU cores in space and is it possible to change the number of CPU cores after startup?
As an example I want 4 cores for filling and 16 for further steps like mixing.

thanks for your help

Phil

Does anyone install LIGGGHTS successfully on a mac with M1 (apple silicon)? I could not as VTK-8.2 could not be installed. Thanks.

Hello guys, apologies if the question is repeated here.
I have been trying to simulate two types of particles through inclined surface. I don't want to use `.stl` file to define surface for particles insertion points.

I want to use cylindrical region and insert particles from that region.

When I use the `insert.stl` then it works but when I remove insertion cad and use region with

`fix ins nve_group insert/rate/region seed 32452867 distributiontemplate pdd1 &`

It gives error.