yade-dev team mailing list archive

[luc scholtes <lscholtes63@xxxxxxxxx>]

> Interesting!


> I'm not sure you need to save/load, but perhaps I missed the point.
>

Not sure too, but the fact is that I need at least 1 timestep to identify
spheres on joint through their interactions with facets and then I should be
able to delete all the joints facets as well as the concerned interactions.
Thought that save/reload was the easiest way...


> Did you think about using clumps for the groups of element not traversed by
> a joint plane?
>

Yes, but, from what I understood, there are no interactions between elements
belonging to a same clump (?), and I want to be able to inititate cracks
inside the unjointed/intact parts of the material.


> For modified normal, what is your pane, writing a new Ig2_Sphere_Sphere_XXX
> which would test if spheres share a common plane before assigning normal?
> I'm not sure what is the best way here.
>

The identification is probably not the most difficult part here as each
concerned spheres are identified with the concerned joint plane normal. I
thought that writing a new Ig2_XXX could not be the best way as the
interacting geometry is already computed in the existing Ig2_XXX... I was
thinking about doing it directly in the contact law (?) which is obvious for
the normal direction (access through the Dem3DofGeom) but less for the shear
direction... Here is attached a snapshot of the kind of thing I would like
to do.

If you have any suggestion, it would be great

  Luc

Attachment: Joint.png
Description: PNG image