Hi everybody,
I am trying to model a cylinder-like shape with a high length-to-diameter ratio.
The multi-sphere particle could look like this: ooooooo
So, the model would include a huge number of spheres, when I try to model several thousand of those particles.
As far as I know, liggghts also supports the application of ellipsoids in granular models, via the ASPHERE package.
With respect to computational time:
- Would it be a good idea to use an ellipsoid instead of the multi-sphere approach?
- Is there a limit, when contact detection between ellipsoids is always (or likely) to be quicker than that for a certain number of spheres?
With respect to modelling in liggghts:
- Can I add different ellipsiod shapes at the same time? (different factories would be fine)
- Can the ellipsoid-ellipsoid contact include rolling friction?
Thank you very much in advance.
Best regards,
MiRa
ckloss | Sun, 11/16/2014 - 15:46
Hi MiRa,
Hi MiRa,
the ellipsoidal potential is an MD potential so it can not be recommended for simulation of macroscopic granular particles.
Best wishes
Christoph Kloss
MiRa | Mon, 11/17/2014 - 15:22
Multi-sphere versus ellipsoid
Hi Christoph,
thank you for your reply!
It is not possible to switch off or neglect the MD potential? I believed that the interaction between particles was defined by the pair_style?
Best regards,
MiRa
MiRa | Mon, 11/17/2014 - 15:31
edit
edit:
"The granular pair styles are used with spherical particles." [1]
Which means they do not work with ellipsoid particles, I guess?
Best regards,
MiRa
http://lammps.sandia.gov/doc/Section_howto.html#howto_14