yade-users team mailing list archive

[Václav Šmilauer <eudoxos@xxxxxxxx>]

> Currently in Ig2_Box_Sphere_ScGeom, sphere-box contacts are seen as 
> contacts between the sphere and another sphere twice bigger (the big one 
> representing the box).
> I'd like to change that and make the fictious sphere the same size as 
> the real one, so that box-sphere contacts will have the same stiffness 
> that contacts between equal spheres (preserves somehow a symetry, I 
> never knew why this factor 2).
> 
> Please let me know if this change would create problems for you, and/or 
> be prepared to see a small change in the rigidity of the boundaries in 
> the triaxial test.

IMHO in longer perspective, you would be better off implementing ScGeom
for Wall+Sphere and get rid of Box altogether, at least for the purposes
of boundary conditions.

You could also assign some fictitious thickness to Wall as well, getting
soft membrane-like behavior (Wall can interact in one direction only,
there would be no problem of spheres going through to the other side).

Cheers, Vaclav