surface orientation airfoil/domain always misaligned
 

Hello,

I've created a flow domain including an airfoil to do a quasi 2D flow simulation (as I couldn't figure out whether there's a way to create a 2D mesh with boundary layer cells which is then extruded into 3D - just one layer thick).

My problem is, that the surface orientation is always wrong. Either the airfoil is correct OR the domain is correct but I am not able to change the surface orientation of just one of them to align them as needed for meshing.

I need a volume mesh between the outside of the airfoil and the inside of the domain of course. As you can see on the picture, the back (inside of the flow domain) has the wrong surface orientation.

How can I change the surface orientation of individual elements (airfoil or domain)?

Klaus

Greetings Klaus,

According to this tutorial: https://github.com/enGits/engrid/wik...ing_Geometries - section "Importing Geometries":
Best regards,
Bruno

not always...
 

thank you for your feedback, but I didn't use Blender and "Correct surface orientation" changes the surface orientation of all faces so I end up with the same problem the other way around.

Klaus

Hi Klaus,

Mmm... if the surfaces are already roughly consistent in the orientation, you should be able to control which side the volume mesh should be generated, as explained near the end of the section "Setting Up the Boundary Conditions": https://github.com/enGits/engrid/wik...ary_Conditions

There is also a utility in OpenFOAM named surfaceOrient. You can check how it's used by running:
If this doesn't work either... I don't have an STL file handy to test this with :( If you could provide one, I or anyone else could have a look into this.

Best regards,
Bruno

Hi Klaus,

I agree with Bruno's suggestion and in addition you can also try this option from Jens Höpken's blog, which can be quite helpful:

http://www.sourceflux.de/blog/preparing-geometry/

Another way around could be to correct the surface orientation of your "airfoil.stl" file and then later on merge it with your bounding box defining the computational domain extent.

Good luck,
Taxalian

Hi,

one quick question: does it make a difference on which direction the normal vectors are pointing, whether it´s an external or an internal flow? Or does it only matter that the direction of all vectors is consistent (all out, all in)?

Thanks in advance!

Quick answer: Yes, it has to be consistent, at least along a particular surface. See Figure 10 on this wiki page for more details on how you can later assign the boundary conditions depending on the orientation: https://github.com/enGits/engrid/wik...and-enGrid-1.2

Hi,

thanks!