yade-users team mailing list archive

[Karol Brzezinski <question706654@xxxxxxxxxxxxxxxxxxxxx>]

Question #706654 on Yade changed:
https://answers.launchpad.net/yade/+question/706654

    Status: Open => Answered

Karol Brzezinski proposed the following answer:
Hello Ketian,

As you noticed the real problem is that NaN interactions are produced
during your simulation. The breakage function needs to know the forces
acting on the particle in order to evaluate its strength. It is not
related to the particle breakage. If you want to test it, just remove
'cbcontroller' from engines and comment out the mainlogic() command.
Then just run the simulation. You will get a similar error, because of
NaN interactions. It is because TriaxialController also needs to get
proper (not the NaN) interactions to compute stress.

Most likely it is related to the rolling friction because I could
proceed with the whole simulation after turning off the 'IncludeMoment'.
I am not sure what is wrong with the rolling friction in this
simulation, but it is another issue (maybe you could try with nonzero
eta [1]).

Best wishes,
Karol

PS. I noticed that you have 10% of clumps in simulation, and you
evaluate only the clumps so the spheres cannot break. Consider
evaluating both spheres and clumps as in this example [2].


[1] https://yade-dem.org/doc/yade.wrapper.html#yade.wrapper.Ip2_FrictMat_FrictMat_MindlinPhys.eta
[2] https://gitlab.com/yade-dev/trunk/blob/master/examples/clumps-breakage/oedometric.py

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