yade-dev team mailing list archive

[Bruno Chareyre <bruno.chareyre@xxxxxxxxxxx>]

> Thanks for the suggestion of getting ride of plastic dissipation; it 
> was indeed the source of error. I attach 2 figures, one with plastic 
> dissipation, one without (frictAngle=0); in the second case, the 
> energy is well conserved (total relative to the max is somewhere 
> around 1e-5, which is pretty good). Bruno, any idea why the 
> dissipation term in Law2_ScGeom_FrictPhys_CundallStrack gives 
> erroneous (over-estimated) values?
I was looking at that just now. Having energy conserved without friction 
is a good thing, but it is a really weak test unfortunately.
No friction means no rotations of bodies (*1)... basically a 1D mass 
spring system (modulo collisions generating horizontal motion).

I don't know how to track energy creation better. What I can recall is 
plastic dissipation is correct in quasistatic loading (attached energy 
bilan in periodic triax), so the problem must be linked with some 
dynamic behaviour (most probably rotations, see *1) more than the 
dissipation function in CundallStrack itself.

Cheers.

Bruno

Attachment: energy_comparison.png
Description: PNG image