#63 Bugs in wall distance caluclation

[David Hora]

Hi all

There are two bugs in the wall distance calculation. The first bug comes from an incomplete application of the following fix and exists since 2014:

https://bugs.openfoam.org/view.php?id=91

https://github.com/OpenCFD/OpenFOAM-1.7.x/commit/9d7fa9f3e0d8972ad0c5d31190d939a4556503b4#diff-b5981067ca501896e4249039f3bc7700a388685bde122301b5b30a8e74df7515

https://sourceforge.net/p/foam-extend/foam-extend-4.1/ci/35af19983b1cf427f31491a135fb07536762efaa/

The second one results in the pattern which is shown in the attached file and is independent from the coupled patches (cyclic, ggi).

Kind regards
David

[David Hora]

since 2010, not 2014. Sorry.

[David Hora]

I forgot that we had a discussion about the first bug. Unfortunately with different conclusions. The original bug report can be found here:

https://sourceforge.net/p/openfoam-extend/ticketsfoamextendrelease/52/

The second bug is only present in foam-extend-4.1. Version 4.0 is not affected.

[David Hora]

Bug 2 is caused by changes in src/meshTools/cellDist/cellDistFuncs.C in commit 7fe57e. Using the file from foam-extend-4.0 solves this issue.

https://sourceforge.net/p/foam-extend/foam-extend-4.1/ci/7fe57ece38bdff18da617fb8bdf118ab6aafc251/

[Henrik Rusche]

Hi David,
I will deal with Bug 2 that is apparently caused by 7fe57e under this ticket.
The original code can fail badly in parallel. It's also quite inefficient. That's in nutshell why I changed it. There are a few (messy) comments in code that try to explain the issue.
Can you please upload the case to reproduce wallDist_noCyclic.png. Thanks.
Henrik