How to relax configuration with overlapcheck no

Submitted by bern on Thu, 09/19/2019 - 15:34

Hello everyone,

I want to simulate with a volume fraction of 0.4288 but this would lead to unreasonable high numbers for maxattempt to work.
So deactivated overlapcheck, but then the manual says the overlapping configuration has to be relaxed.
My problem: there is no word on how to do this.

Can anyone tell me how to relax properly? ;)

Thank you in advance.

Forums:

Daniel Queteschiner | Mon, 10/07/2019 - 15:32

The actual solution to your problem heavily depends on your case setup.
You may be able to use a larger insertion region and let particles settle under gravity (most natural solution).
You may insert particles using a smaller size than their actual diameter and use fix adapt to scale them up slowly. (beware of artificial stresses)
If you insert particles without any overlap check, the artificial overlaps introduce excessive energy that needs to be removed from the system in one form or another. You could for example reset the particle velocity repeatedly using the set command or use an integrator that limits the maximum particle velocity like fix nve/limit (at the end of relaxation unfix that integrator and use the proper one, e.g. fix nve/sphere).
If you are up for some coding, there exist various dense packing methods, e.g. Lozano et al., Comput Math Appl 71 (2016), 1586-1601, that can achieve much higher volume fractions than the methods available in LIGGGHTS-PUBLIC.

Daniel Queteschiner | Fri, 10/11/2019 - 14:35

When scaling up particles it may happen that there is not enough space for evasive movements and thus the particles may get jammed.
This is less likely to happen if you for example pour particles into a container. It always depends on the specifics of the acutal case of course.

saman | Fri, 10/11/2019 - 19:00

Thank you for the answer.
I had another question from you.
How can it be controlled?(beware of artificial stresses)