[montju]

Hi all,

I'm currently trying to generate a C-mesh around a NACA aerofoil with a single block going from the trailing edge of the pressure side to the trailing edge of the suction side. Behind the trailing edge, there should be two blocks. The block and vertex ID's are shown in the image "blockMesh-vertices.jpg".

The boundary layer block around the aerofoil is block 0, the downstream blocks are 3 (suction side) and 4 (pressure side) - disregard the rest of the image as it's redundant for this problem.

Even though the blocks are generated the way I intended them to be, the faces connecting block 0 with blocks 3 and 4 show some weird behaviour --> "naca0012-mesh1.jpg", "naca0012-mesh2.jpg"

When I run checkMesh, I get the following output --> "checkMesh.txt".


It seems to me that even though the blocks are generated and I checked multiple times that I follow the right-handed rule and all other definitions for the hex blocks as specified on

 /https://www.openfoam.com/documentation/user-guide/4-mesh-generation-and-conversion/4.3-mesh-generation-with-the-blockmesh-utility

, blockMesh is unable to do this kind of block or am I missing a point somewhere and didn't properly define the block? I also attached the blockMeshDict file to recreate the problem.

I would really appreciate some feedback if what I'm trying to do is even possible with blockMesh and if so, how I would possibly fix this issue.


Cheers, Sebastian

[Tobi]

Hey Sebastian,


it looks like that your ordering of the blocks is wrong in the areas you mentioned. It seems that you have some flip of the volumes (checkMesh should give you - probably - negative volumes - simply test it). For a easier analysis, simply take the block which is wired and create the mesh for only this block. Doing so will give you a better understanding of the problem.

I guess the numbering of the blocks is wrong (some small mistake).


Furthermore, as I discussed with Philipp already, it might be also a good option to use snappy for your investigations.

Tobi

[Tobi]

Here one tweet that shows that snappy can handle your problem Ver well.

https://twitter.com/joegigeo/status/...R36SBa2nA&s=19

However, i would further optimize the mesh around the airfoil.

[montju]

Hi Tobi,


thanks for taking a look - initially I really thought I had the block ordering correct but the more I look at it the more I'm doubting myself again.


In the meantime I already looked into snappyHexMesh, but the twitter post you attached finally convinced my that I snappy will most likely be the better option for this problem, especially with all the different airfoils and AoA I'd like to simulate.


Sebi