View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001371 | OpenFOAM | Bug | public | 2014-08-13 09:00 | 2014-08-14 09:54 |
| Reporter | Assigned To | ||||
| Priority | normal | Severity | minor | Reproducibility | always |
| Status | resolved | Resolution | fixed | ||
| Platform | Linux | OS | OpenSuse | OS Version | 13.1 |
| Summary | 0001371: Possible over-restrictive condition on normals for use with symmetryPlane? | ||||
| Description | When using a custom mesh, I get the following error: --> FOAM FATAL ERROR: Symmetry plane 'slider_sym' is not planar. At local face at (-2.7 -9.24062e-08 0.174755) the normal (-9.54539e-24 -1 -5.28785e-07) differs from the average normal (-7.26287e-09 -1 -5.84412e-07) by 3.14713e-15 Either split the patch into planar parts or use the symmetry patch type Of course, I can use symmetry instead of symmetryPlane, but is 3.14e-15 not slightly over-restrictive to determine that a local face normal is not aligned with the plane average? Kind regards, F.T. Pronk. | ||||
| Tags | No tags attached. | ||||
|
|
It actually compares&prints the magSqr, not the mag so the actual error is in the order of 1e-6, 1e-7. This is the range at which you might start accumulating truncation errors and e.g. start getting parasitic flow in one of the other directions. Maybe you have to project your points onto the xz plane (perhaps using flattenMesh) and using binary writing. |
|
|
Dear Mattijs, Thank you for your reply. If the actual error is in the order of 1e-6, then I understand that face alignment might start to become a problem, so we can drop the issue. As for the flattenMesh utility, I'm using an enGrid mesh (courtesy of Oliver), and running it out of the box did not work as no normal could be determined from the patches. FlattenMesh is described as working on 2D meshes, does that mean it expects a frontAndBack type of patch in the boundary file? Could I get away with this issue by running a symmetry boundary instead of symmetryPlane? Kind regards, Francois. |
|
|
using a symmetry instead of symmetryPlane would help (if you can live with the fact that your planes are not fully flat) |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2014-08-13 09:00 |
|
New Issue | |
| 2014-08-13 15:14 |
|
Note Added: 0003204 | |
| 2014-08-14 09:30 |
|
Note Added: 0003205 | |
| 2014-08-14 09:53 |
|
Note Added: 0003206 | |
| 2014-08-14 09:53 |
|
Status | new => resolved |
| 2014-08-14 09:53 |
|
Resolution | open => fixed |
| 2014-08-14 09:53 |
|
Assigned To | => user4 |
