[swtbkim]

Hi all!

I'm a new user of pointwise, and using FLUENT solver.

The thing is that FLUENT does not accept a mesh file whose skewness is higher than 0.98.

As all you knows, unstructured T-Rex makes meshs of very high skewness.

It doesn't a problem??


In fact, I'm suffering from that problem
and also think that the problem may be another thing, not T-Rex.

Thanks in advance!

[swtbkim]

To make the question more precise,


I know I can control maximum skewness of T-Rex before clicking Initialize.

What I'm curious is that T-Rex is highly structured boundary layer, so FLUENT can accept it even though skewness is higher than 0.98, or I should control it less than 0.98 for importing to FLUENT.

Thank you.

[tcarrigan]

Hello Sangwoo,

Have you tried combining the anisotropic tetrahedra generated by T-Rex on export? There is an option when exporting to FLUENT to "combine anisotropic tetrahedra." This will take the tets and combine them to form triangular prisms.

It's important to combine for a number of reasons, one of them being to reduce the equiangle skewness. An anisotropic, right angled tet has very high equiangle skew because of the small acute angle. However, once combined to form prisms, this is no longer an issue.

If after combination you still find you have cells with skewness over 0.98, I suggest you use FLUENT's tools to visualize where these problems are located. Once you know where the problem cells are you can go back to Pointwise to fix them. Often it's just a matter of making modifications to the surface mesh that resolves the problem.

[swtbkim]

Thanks for your reply

I tried to export as you suggested, but I couldn't find that option.
.cas file is directly exported without any option when I click File>Export>CAE.
If I change the solver to cgns, it asks just options of Data Precision.


What can I do??

[swtbkim]

Oh I found

But it does not work..
I guess it's because the geometry is too harsh

[tcarrigan]

What do you mean when you say "it does not work?"

If combination of the boundary layer elements is occurring, I suggest identifying the problem areas in FLUENT and then fixing them in Pointwise. There are a few settings you can use when running the T-Rex block solver to help resolve poor quality elements. Please see the following video, http://youtu.be/24bKZrpt5E4?t=40m57s.

[swtbkim]

Hi Travis.

I meant that I still had high skew elements after checking 'combine anisotropic tetrahedra'.

and this video is perfectly helpful!!
I didn't know that there is such a good material in Youtube.

Thanks a lot!