2d flow over 1d object

[looo]

Also with your file it doesn't work for me. Maybe a python3 error...

edit: same problem also with python2. So maybe I am doing something wrong.

I open your file. Double click the "Slice-Mesh"-object in the treeview and click on the "Mesh"-button. Once done I click "paraview". Is there anything I do wrong?

[thschrader]

I open your file. Double click the "Slice-Mesh"-object in the treeview and click on the "Mesh"-button. Once done I click "paraview". Is there anything I do wrong?

No, thats ok. I did the same and it still works.
Do you use a cfmesh-version older than 2 month?
There where some changes:
https://sourceforge.net/projects/cfmesh-cfdof/

[looo]

Do you use a cfmesh-version older than 2 month?

I installed cfdof and cfmesh via addon-manager last week. Did you install cfmesh manually? Is there a way to check the version of cfmesh?

[thschrader]

You can check the „version“ in the Readme.
But from time to time I have a look at github/sourceforge
to see if there are changes in the files.

My workflow for installing cfd-of is:
Download bluecfd-core/cfmesh/hisa from.
Install bluecfd-core manually
Install cfmesh/hisa via preferences/cfd by using option „Choose existing file“.
But in the end there should be no difference between my workflow and the standard-installation.

cfmesh_version.JPG (34.1 KiB) Viewed 726 times

[looo]

I played a bit with the number of boundary-layer-elements / first-cell-height parameters and found a working combination. To me it seems this sometimes works and sometimes not.

with the boundary-layers the velocity-distribution at the wall looks good:

Bildschirmfoto von 2019-01-28 13-16-29.png (189.06 KiB) Viewed 719 times

[oliveroxtoby]

I played a bit with the number of boundary-layer-elements / first-cell-height parameters and found a working combination. To me it seems this sometimes works and sometimes not.

with the boundary-layers the velocity-distribution at the wall looks good:
Bildschirmfoto von 2019-01-28 13-16-29.png

If you make a 3D mesh instead of 2D, you'll see that it is a bit unpredictable as to whether cfMesh ends the boundary layers at the wall or wraps them around onto the front/back patch. Since the 2D mesh is currently produced by extruding the front patch of the 3D mesh, you can be left with no boundary layers if it did wrap it around. I'll see if I can modify it to rather produce 2D meshes directly with cfMesh's cartesian2DMesh tool.