Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Xin Li posted a new comment:
Hi Klaus,
I reran the simulation with hertz-mindlin contact law (with rolling stiffness)
and GlobalStiffnessTimeStepper(safetycoefficient=0.8), everything was ok.
Thank you!
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Klaus Thoeni posted a new comment:
I believe this was related to an issue in the GlobalStiffnessTimeStepper
which was fixed here [1]. Please can you confirm that everything runs as
expected with yadedaily so we
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Xin Li posted a new comment:
Hi Bruno,
> Zero exactly?
Yeah, actually from the 3D view of the simulation, no strain can be observed
before timestep droped to 0 even though no errors are reported. But, I
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Bruno Chareyre posted a new comment:
> the time step suddenly drops to 0
Zero exactly?
--
You received this question notification because your team yade-users is
an answer contact for Yade.
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Xin Li posted a new comment:
Hi Anton,
I have create an issue with the same information about this problem on
gitlab.
--
You received this question notification because your team yade-users is
an answer
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Xin Li posted a new comment:
Hi Bruno,
Thank you for your answer! I have changed the time step setting to
GlobalStiffnessTimeStepper(), this program can continue to run. But, the
time step suddenly drops to 0
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Anton Gladky posted a new comment:
It would also be good if you compile yade with debug symbols, so we can
get more information about the crash.
Could you also please create an issue on gitlab [1] with this
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Bruno Chareyre posted a new comment:
Hi,
It could be that a high torsional stiffness needs a smaller timestep, so that
you get a classical numerical instability if you keep it the same.
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Karol Brzezinski posted a new comment:
Sorry, but I don't know how to help with 'ktwist'. I don't have any experience
with twisting yet. Maybe you should start another topic.
My comment about system
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Status: Answered => Solved
Xin Li confirmed that the question is solved:
Thank you very much Karol!
After canceling the setting of torsional stiffness of hertz-mindlin
contact law, this simulation started
Question #700411 on Yade changed:
https://answers.launchpad.net/yade/+question/700411
Status: Open => Answered
Karol Brzezinski proposed the following answer:
Hi,
please note that krot is not a problem. The simulation fails if ktwist is
nonzero (but you prescribe 'kr' to both 'krot' and
New question #700411 on Yade:
https://answers.launchpad.net/yade/+question/700411
python3.8:
/builds/yade-dev/trunk/deb/yadedaily/pkg/common/InsertionSortCollider.cpp:778:
bool yade::InsertionSortCollider::spatialOverlapPeri(yade::Body::id_t,
yade::Body::id_t, yade::Scene*, yade::Vector3i&)
12 matches
Mail list logo