Mesh size is bigger in the inner region than the background mesh box

[nuofei]

Hi! I’m trying to learn how to use the hex-dominant parametric, so please don’t suggest to me to use the standard mesher

As you can see from the picture, some mesh cells in the inner region is bigger than the background mesh box. I’m not sure why because I didn’t add any refinement to the background mesh box. How do I deal with this? Any tips?

Can you also comment whether my initial mesh looks good or not? If not, can you recommend the areas or refinements I need to improve? Total cells = 11.5 M, I add 3 region refinements (small wake, large wake and rotating zone), 2 surface refinements (whole turbine, rotating zone) and inflate boundary layers. Do let me know if you need to go through the project.

Thank you!

[igaviano]

Hi @nuofei,

Thanks for posting on the forum!

@nuofei

As you can see from the picture, some mesh cells in the inner region is bigger than the background mesh box. I’m not sure why because I didn’t add any refinement to the background mesh box. How do I deal with this? Any tips?

It’s tough to say why this is happening without taking a look at the simulation. Could you share it’s URL for us to take a closer look?

@nuofei

Can you also comment whether my initial mesh looks good or not? If not, can you recommend the areas or refinements I need to improve? Total cells = 11.5 M, I add 3 region refinements (small wake, large wake and rotating zone), 2 surface refinements (whole turbine, rotating zone) and inflate boundary layers. Do let me know if you need to go through the project.

From the pictures you’ve shared, the mesh looks visually good, but a better approach is to evaluate the quality by taking a look at the meshing log.

Maybe you could also benefit from the following tutorial:

• Hex-Dominant Parametric Meshing of Front Wing | SimScale

Best,
Igor

[nuofei]

Hi @igaviano , here’s my project link : SimScale Login

I checked the mesh log. Only volume ratio was quite high abt 140.

[nuofei]

From the tutorial you shared, I noticed the external flow volume wasn’t created in the CAD mode. Why did it not extract flow volume first? Is this the correct sequence when usin hex dominant parametric? Does that mean the body of geometry is not deleted and kept throughout simulation?

[igaviano]

@nuofei,

Yes, hex-dominant Parametric excludes the necessity of creating a flow region manually as the bounding box will do just that. Then, you’ll use the mesh itself as the domain instead of the geometry like it happens with the standard mesher.

[nuofei]

Ohh I see. So, if I extracted the flow volume, would it cause some issues?

[igaviano]

Yes, exactly @nuofei! By following the steps on the tutorial I’ve provided you should be able to generate a mesh without further issues

[nuofei]

Hi @igaviano , I have a rotating zone in my flow domain though. How should I proceed then?

[Ricardopg]

I’m a bit late to the party, but if you have a rotating zone when using hex-dominant meshes, you need to create cell zones via surface refinements (documentation).

Cheers