CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Question about I-grid (http://www.cfd-online.com/Forums/main/107159-question-about-i-grid.html)

Anna Tian September 19, 2012 04:48

Question about I-grid
 
Hi,

Could you please show me a picture of I-grid?

gfoam September 19, 2012 09:47

1 Attachment(s)
Attached you have a picture of an I type mesh. Regards
Gonzalo

Anna Tian September 20, 2012 02:43

Quote:

Originally Posted by gfoam (Post 382560)
Attached you have a picture of an I type mesh. Regards
Gonzalo

Why we need to use the I-block for the blades for that case?

gfoam September 20, 2012 07:35

I thinkyou must use it because you can get a mesh with high quality in that geometry. An the clls generates have faces which are almost normal to the local flux, so that you can get a converged solution easier and faster. Moreover whith hexahedral or cuadrilateral cells you can get a mesh with high resolution with fewer cells than a mesh with triangles or tetrahedrons. Regards.
Gonzalo

Anna Tian September 20, 2012 09:36

Quote:

Originally Posted by gfoam (Post 382749)
I thinkyou must use it because you can get a mesh with high quality in that geometry. An the clls generates have faces which are almost normal to the local flux, so that you can get a converged solution easier and faster. Moreover whith hexahedral or cuadrilateral cells you can get a mesh with high resolution with fewer cells than a mesh with triangles or tetrahedrons. Regards.
Gonzalo

Thanks for your answer.

But one drawback of that I-block is the non-uniform mesh at each side of your blade. When your inlet angle is even larger, that drawback will be more critical. How to solve that problem in this case?

gfoam September 20, 2012 10:11

Hi:
well, you use I-type and other grids like that when you have periodic BC's at the top and bottom face, but if you have high AoA flow at the inlet, i don't know if th condtion of periodicity is still satisfied. So in that cases you must simulate the entire domain and not only a part of this. But I don't have much experience with turbomachinery simulations and therefore I'm not sure about it. Regards.
Gonzalo


All times are GMT -4. The time now is 13:03.