CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [ICEM] Octree surface mesh issue (http://www.cfd-online.com/Forums/ansys-meshing/84198-icem-octree-surface-mesh-issue.html)

siw January 23, 2011 04:18

[ICEM] Octree surface mesh issue
 
2 Attachment(s)
I've made three meshes around an aircraft: coarse, medium and fine with the Octree method in the usual manner (http://www.cfd-online.com/Forums/ans...procedure.html).

The surface meshes on the nose and tailplane are shown in the images: coarse on the left, medium mesh on the upper right and fine mesh on the lower right in each image.

I set the octree sizings on the medium mesh using: 1/8, 1/4, 1/2, 1, 2, 4 etc and used a global scale factor of 1.5 for the medium mesh and 2 for the coarse mesh.

The coarse mesh looks okay but why does the medium mesh show regions which are too coarse (see nose apex and fuselage at the tailplane mid-chord)? The scale factor should not allow this too happen.

Thanks

PSYMN January 30, 2011 16:27

Looks right to me...
 
I am just looking at still images, without the ability to measure, etc. What makes you think there is a problem?

Have you checked the edge lengths (there is a quality metric for that or you could just measure a few manually to compare). I did that roughly by eye and it looks about right to me.

siw January 31, 2011 02:46

2 Attachment(s)
Hi,

I could not see any metric for edge length so I measured a few elements on the nose and on the fuselage on in the mid-chord region of the tailplane and they are bigger on the medium mesh(!)

I've marked the images where the larger elements are occuring on the medium mesh to highlight where ICEM looks to be getting it wrong.

Thanks

siw February 4, 2011 04:25

Does anyone else think there's no problems with my intermediate mesh?:confused:

BrolY February 4, 2011 07:27

Could you post others pictures with your geometry only ?
I would like to see your curves.

You could also check the number of nodes that ICEM puts on your curves for all your models.

PSYMN February 4, 2011 08:49

I see it, I just don't understand it...
 
Hey SIW, I see it now at the nose, I just have no idea why it would do that. I will think about it.

Simon

siw February 13, 2011 12:46

Simon, any thoughts?


All times are GMT -4. The time now is 07:52.