Hi All,
Fix rigid/small behaves very differently than fix rigid in v3.0.1. It seems like rigid bodies created by fix rigid/small can exert forces on neighbor particles, but the rigid bodies themselves do not move at all. For some reason, it looks like bodies created by fix rigid/small are not being time integrated. In the source code, the integration schemes look the same... Anyone have the same experience?
FYI, my comm cutoff is large enough, I have tried several comm cutoffs with no change in results.
Thanks,
Zamir
EDIT: Attached a test case
I put videos of the test case on youtube:
Video of test case using fix rigid (behaves properly):
http://youtu.be/lmpR2TABFtY
Video of test case using fix rigid/small (behaves unexpectedly):
http://youtu.be/aWdSVxs1LjE
| Attachment | Size |
|---|---|
 test_case | 410.21 KB |
zamir | Sun, 05/18/2014 - 17:08
Bug?
`Hi All,
I've been trying to figure what is wrong here, but I am thoroughly stuck. I think it is a bug.
Zamir
zamir | Mon, 05/19/2014 - 05:50
I put links to videos in the
I put links to videos in the original post which illustrate the problem. The two videos should look exactly the same, but they do not because fix rigid/small does not work properly.
zamir | Tue, 05/20/2014 - 06:23
update
Hi All,
I peeked into the source code a bit and I found a few interesting things:
1) fix_rigid_small.h is not included (#include) in pair_gran.cpp or fix_wall_gran.cpp. However, fix_rigid is included.
2) PairGran, FixWallGran, FixGravity, and FixTemplateMultisphere are not declared as friend classes in fix_rigid_small.h. It seems like they should be. Correct me if I'm wrong.
I hope someone can help me out here, or at least tell me whether I'm heading in the right direction...
Thanks,
Zamir
Nafets | Mon, 06/23/2014 - 11:56
Have you solved your problem?
Because i've got the same problem when i use rigid/small.
ckloss | Thu, 06/26/2014 - 12:09
Hi fix rigid/small is
Hi fix rigid/small is currently not connected to the granular features - use fix rigid instead
Best wishes
Christoph