About simulating rock-like material

Submitted by xiuhan on Mon, 07/01/2013 - 11:42

Hi all, I wanna simulate brittle material like rock, which means the bonds between particles should start from a bonding model with certain strength(both in normal and shear direction), then after loading (acting external force), the bonds should break and then the particles can move freely so that a collision model should be applied.
Does anyone have similar problem, how you solve it with liggghts, what contact model you used?
really appreciate any reply.

ckloss's picture

ckloss | Fri, 07/12/2013 - 14:15

We have an early-stage bonded particle-model.
We use it for research project and collaborations to develop it further, but it's not ready for release at this point.

Christoph

richti83's picture

richti83 | Tue, 07/15/2014 - 08:26

a basic implemention is allready there:
https://github.com/Polyun/LIGGGHTS-PUBLIC/tree/master/bondspackage

For me it was only stable for single-core operation and bonded particles should never been lost (you need to make plane -walls at the simulation box borders)
Also the dissapation factor in bond_gran.cpp line 286 is critical, because bond-force is lost over time -> I set it to 1.0 (no damping) to solve this.

I can not give more support because I'm not the author of this code, but I used it as starting point for my own implementation.

Best,
Christian.

I'm not an associate of DCS GmbH and not a core developer of LIGGGHTS®
but I have 7 years user experience using LIGGGHTS®, ParaView and LINUX