[qtian]

Dear all,

I am trying to understand how to choose the parameter for nNonOrthogonalCorrectors. I searched " nNonOrthogonalCorrectors" in the message board and read every thread. However, for some reason, I can not find much information about it.

Someone happened to mention that use "checkMesh" to find information about mesh non-orthogonality and skewness.

Here is the information I got from "checkMesh"

Mesh non-orthogonality Max: 56.7059 average: 8.27139
Non-orthogonality check OK.

Face pyramids OK.

Max skewness = 76.5705 percent. Face skewness OK.

Minumum edge length = 0.0026. Maximum edge length = 0.0312349.


My question is how can I decide the nNonOrthogonalCorrectors value, based on above informaiton? is there any special formula?

Thanks for your help. Happy new year!!

Quinn

[msha]

there is some info in programmers guide of the code in tutorials section

[schmidt_d]

Quinn,
I suppose that this is a matter of opinion and experience. Usually CFD codes use a deferred correction approach where the orthogonal part of the Laplacian is handled implicitly while the non-orthogonal part is treated explicitly. Generally, spatial errors in CFD dominate temporal errors. So I would say that if your code is running and stable, you are using enough non-orthogonal corrections. Start with 1 for nNonOrthogonalCorrectors and go from there. To learn more, look at H. Jasak's thesis and also Peric and Ferziger's CFD book.
-David