yade-dev team mailing list archive

[Chareyre <bruno.chareyre@xxxxxxxxxxxxxxx>]

I don't understand the problem Vaclav, sorry.
I thought the topic was the comparison between ScGeom and Dem3Dof in TriaxialTest. It implies the comparison is done using *_FrictPhys_Basic functors (i.e. the laws for non-cohesive materials - I call them "granular"). Did I miss something?

FOR TRIAXIAL TEST WITH FRICT_PHYS LAWS :
"geometric distance (radius1+radius2) should be preferred, definitely. InitialEquilibrium distance is only creating strange effects"

"For continuum-discrete approaches :
I understand some need that"

I don't think it is possible to get stable results with triaxial engines
and non-cohesive laws if  equilibrium distance is defined differently.
It will generate instabilities on the boundaries.

Me not liking continuum mechanics? I'm surprised.

-- 
Equilibrium distance different for SpheresContactGeometry and Dem3DofGeom_SphereSphere (TriaxialTest with facet vs. box walls)
https://bugs.launchpad.net/bugs/399963
You received this bug notification because you are a member of Yade
developers, which is subscribed to Yade.

Status in Yet Another Dynamic Engine: In Progress

Bug description:
Seems that there is too much friction with the walls, but inspecting the XML file gives nothing suspicious. UnbalancedForce sits constantly at 1.1407 after a small while, without going down.