Constrain DOFs

Submitted by cgrohs on Tue, 09/07/2010 - 15:30

Is it possible to contrain the rotational degrees of freedom for granular particles?

Christian

Forums:

ckloss's picture

ckloss | Tue, 09/07/2010 - 17:01

Hi Christian,

sure, this is possible - the simplest possibility to constrain it is to use fix nve instead of fix nve/sphere, because fix nve does not update omega.

Christoph

cgrohs | Wed, 09/08/2010 - 09:20

Hello Christoph,

Thank you for the quick answer! Sometimes it's much faster to ask than to read the manual ... :)
I'll have a try as soon as possible.

Christian

marketos | Wed, 09/08/2010 - 17:50

Hi Christian,

Even though it is possible I think that you should be very careful when you do this, as the solution for your grains will no longer satisfy rotational equilibrium! The forces calculated in this way for example would create a stress tensor inside the grain that is not symmetric.

If my suspicions are correct and you want to do this so as to limit the rotations of spheres, I think a much better solution would be to implement a rotational spring in LIGGGHTS, ie a spring that applies a moment proportional to the rotation.

George

ckloss's picture

ckloss | Thu, 09/09/2010 - 13:36

Hi george,

yes, you are right. I forgot to note that, thanks.
I think it depends on the application - if the motion of the particles is physically constrained it would be a good idea to _model_ this constraint instead of just not integrating rotational motion.
On the other hand, if it is just used for keeping the particles away from rolling in a numerical experiment (e.g. to produce some special initial packing) the method of using fix/nve seems appropriate
Christoph