yade-users team mailing list archive

[Jan Stránský <question688993@xxxxxxxxxxxxxxxxxxxxx>]

Question #688993 on Yade changed:
https://answers.launchpad.net/yade/+question/688993

    Status: Open => Answered

Jan Stránský proposed the following answer:
>> what values have you tried?
>O.dt=10e-8

what about 1e-10? 1e-12? 1e-15? 1e-20?

> If stiffness reaches less than 10e5, then particles are passed through
facet.

what are the striffness values of both polyhedra and facets?
You can decrease stiffness of polyhedrons and increase stiffness of facets artificially (e.g. 1e300, the harmonic average in contact law makes it ok)

> Can you please provide a script (two bodies as is described)
reproducing this behavior?

thanks, but it is not really a good script:
- there is a lot of not needed facets
- the two polyhedrons may not interact at all. E.g. with
      Dolomite.young = 5e7
      Bound.young = 1e300
      O.dt=10e-5
   the simulation behaves quite OK (because the polyhedrons does not interact)

for some other values, the interaction polyhedrons interact and "bounces
excessively".

> it is a typical symptom of too large time step (making the simulation
unstable) and I don't believe it is not overcome with decreasing the
time step to a **sufficient** value

this still holds. It seems O.dt=1e-8 is still too large.

cheers
Jan

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.