Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Open => Answered
Bruno Chareyre proposed the following answer:
In fact when it is a new bug better start a new thread (if on launchpad), or
even better report an "issue" directly on gitlab:
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Bruno Chareyre posted a new comment:
Thanks.
Re-posted here: https://gitlab.com/yade-dev/trunk/-/issues/214
Bruno
--
You received this question notification because your team yade-users is
an answer contact
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Answered => Open
Rohit John is still having a problem:
Dear Bruno,
I have found one more bug. The angular momentum value of the clump
returns (0,0,0) even though inertia is not 0 and angular
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John posted a new comment:
Dear Bruno,
Thanks for your explanation. I will look into it. How do I open a bug
and what information should I put into it?
Kind regards,
Rohit
--
You received this question
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Open => Answered
Bruno Chareyre proposed the following answer:
Ah... the clumping step is using positions+mass to deduce total mass and
center of mass while "position" of the cylinder is one of its
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Bruno Chareyre posted a new comment:
Very interesting. I'm unsure where to improve that in source code though.
Mass-inertia assignements are done mainly in some python helper function. As
you found out you can
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Dear Bruno,
I think I found a solution. I found that setting the mass of the
grdiConnections to 0 seemed to fix the problem of the triangular object
rotation.
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Answered => Open
Rohit John is still having a problem:
Dear Bruno,
Thanks for you swift reply.
> Best method is to download the sources with git [1], then you update with "cd
> trunk; git pull".
-
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Bruno Chareyre requested more information:
with newton damping 0.2 and
base_pos = [
[0.011, 0.011, 0],
[0.011, -0.011, 0],
[-0.011, 0.011, 0],
[-0.011, -0.011, 0],
]
The solution is stable
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Bruno Chareyre proposed the following answer:
I tried you example (nice one!) and I did not see something obviously
wrong.
It looks a bit like a ball-on-a-tip problem. This triangular object has
to fall one way
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Open => Answered
Bruno Chareyre proposed the following answer:
Thanks for more feedback.
> I did not download a new version of yade. But I pasted the fix given
in [1]
If you did it carefully it
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Expired => Open
Rohit John is still having a problem:
Dear Klaus Thoeni,
I used the fix mentioned in [1] to fix the problem I was facing. And it
does fix the dynamics my simulation (see code in the
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Open => Expired
Launchpad Janitor expired the question:
This question was expired because it remained in the 'Open' state
without activity for the last 15 days.
--
You received this question
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Dear Klaus Thoeni,
I was playing around with the pfacet and I found a discrepancy in mass
of the nodes in a pfacet. I calculated the mass of the nodes using the
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Klaus Thoeni posted a new comment:
Thanks for that. You are right. It is even more obvious is you use
spheres instead of the cylinders and and extent them over the whole area
of the pfactes. The spheres in
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John posted a new comment:
In the comment above, I have given the code which highlights the bug.
I've modified the mikado example [1]. The newton damping is set to 0,
but you can still see this bug when it
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Dear Klaus,
I was testing the mikado example [1], and I found an bug. This example
has many vertical cylinders on a pfacet square. In [1], the lengths are
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John posted a new comment:
Dear Klaus,
> Apparently there seems to be a bug in the distribution of the forces onto the
> gridNodes [1].
- Thanks for this information. Please let me know if I can help you
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Klaus Thoeni posted a new comment:
Hi John,
apparently there seems to be a bug in the distribution of the forces
onto the gridNodes [1]. I have to look into it but maybe once fixed it
might also fix part of
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Dear Klaus Thoeni,
Thank you for your input on my suspicions regarding the rounding error.
Is it possible for you to give a value (or a range of values) for
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Klaus Thoeni posted a new comment:
Yes, most likely the instability comes from numerical rounding errors
and the error made by the integration of the equation of motion.
Now, you are right. Newton damping is
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Answered => Open
Rohit John is still having a problem:
Dear Klaus Thoeni,
Thanks for your insight into the matter. I tried your solution and it
worked. However, project involves studying the
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Status: Open => Answered
Klaus Thoeni proposed the following answer:
He John,
you can find an example on how to use gridConnections and pFacets here
[1]. Just as a reminder, the end nodes of a
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Description changed to:
Hello,
I am simulating the interaction between a brush (bristles made of grid
nodes and grid connections) and an object (made of pfacet). The bristles
are arranged to form a square at
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Hello,
Please disregards, the comment I made above (#2), now I realize that I
need Law2_ScGridCoGeom_FrictPhys_CundallStrack() to handle the
interaction between
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
Hello,
I found that the engine Law2_ScGridCoGeom_FrictPhys_CundallStrack(), was
part of the problem. This got rid of the rotation that was caused at the
start
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Description changed to:
Hello,
I am simulating the interaction between a brush (bristles made of grid
nodes and grid connections) and an object (made of pfacet). The bristles
are arranged to form a square at
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Rohit John gave more information on the question:
I found that the order of defining the pfacet faces is affecting the rotation.
For
pfacet_nodes = [
[nodes[0], nodes[2], nodes[1]],
[nodes[0],
Question #695558 on Yade changed:
https://answers.launchpad.net/yade/+question/695558
Description changed to:
Hello,
I am simulating the interaction between a brush (bristles made of grid
nodes and grid connections) and an object (made of pfacet). The bristles
are arranged to form a square at
29 matches
Mail list logo
