Hanging spheres

[Anton Gladky]

After playing with some ESyS simulations. I have found that in some cases spheres hang and tremble on mesh-generated surfaces (see screenshots). I think it is an abnormal behavior.

[Diego Peinado Martín (diego-peinado)]

I did not find this behaviour yet. I'm experimenting with viscous walls, and having other problems. I also modified the standard behaviour of one sided walls to two sides walls, and with this behaviour, can occur what you relates.

Please can you post your script to reproduce it?
THanks,

[Anton Gladky (gladky-anton)]

I have attached the script. It takes a little bit long to finish the script.

[Anton Gladky (gladky-anton)]

I have the same picture after "night simulation"

[Dion Weatherley (d-weatherley)]

Hi Anton,

Thanks for the comprehensive bug report. This looks very similar to the problem we encountered with the funnel flow exercise at the workshop in Aachen.

Steffen and I have discussed this issue and will give it high priority for a bugfix before the next release. The first thing will be to identify a small test case that displays this behaviour so we can more easily probe what exactly is happening to the hanging particles. If you can think of a test case or are able to learn some more about the forces etc. experienced by the hanging particles in your simulation, please post it here. Would be enlightening to know if hanging particles are touching edges, corners or faces of triangles.

Cheers,

Dion.

[Anton Gladky (gladky-anton)]

Ok, I will try to "localize" the problem, using just 2-3 triangles and 1 sphere.

[Anton Gladky (gladky-anton)]

I have created a small script with this problem

http://geom.com.ua/bug.tgz

Hope it will help to fix it.

[Anton Gladky (gladky-anton)]

Archive is in attachment

[SteffenAbe (s-abe)]

Fixed in svn trunk (r1278).

[Anton Gladky (gladky-anton)]

I can confirm, that the problem is completely resolved in r1278. Thanks a lot!