Error introduced by particle-particle overlap during contact

Submitted by aperelma on Tue, 10/08/2013 - 21:00

Just starting my learning curve in discrete element method.

I am having trouble wrapping my head around the concept of the particle overlap during the particle to particle collisions. I feel the change in void fraction would introduce a good amount of error.

Is there some good literature on the concept?

Thanks

Forums:

cstoltz | Wed, 10/09/2013 - 04:07

The overlap is only a small fraction of the radius of the particle (unless your stiffness is too soft or timestep too large). It shouldn't introduce much error.

If it helps, picture a tennis ball bouncing off the court. When it hits the court, it actually deforms a bit and you end up with an interface between the ball and the court. In a DEM simulation, when two spherical particles collide, we can't really deform the sphere itself, so we let them overlap a bit to create a contact surface. It also lends itself to the contact actually having a duration, as opposed to a hard-sphere event-driven DEM code in which contacts are instantaneous events.

My two cents,
Chris