simulation divergence under high pressures

Submitted by rob1 on Fri, 12/16/2016 - 11:52

Hello,

I am a user of LIGGGHTS and I am applying a high pressure (10% of the Young's modulus) through a servo wall.
After some time I have the loss of particles (from 10 to 100 out of 12000), but the simulation runs well for more than half its duration.
Why there is this problem? Is it solely due to the choice of the timestep?

I am choosing a timestep suitable for my properties, indeed I am doing a check and it is always positive and the simulation runs without problem at the beginning.

Thank you and have a nice day,
rob1

Forums:

ckloss's picture

ckloss | Mon, 01/02/2017 - 13:36

Hi Rob,

the pressure dependance of the time-step criteria is something that the equations do not take into account!
The maximum overlap should be 0.5% up to a couple % of the diameter

Best wishes
Christoph

rob1 | Wed, 01/11/2017 - 10:33

Dear Cristoph, thank you for your answer.
Anyway I still do not understand what to do: since I do not want to lower the pressure, can I simply use a smaller timestep?

Thanks a lot and have a nice day,
rob1