CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Tetrahedral vs Polyhedral (http://www.cfd-online.com/Forums/fluent/93083-tetrahedral-vs-polyhedral.html)

Rohini October 4, 2011 14:53

Tetrahedral vs Polyhedral
 
Hi all,

I am trying to validate the efficiency of polyhedral meshes in capturing wall shear stress in blood vessels. For this, I create tetrahedral meshes using gambit and convert them to polyhedral using fluent. However, when I am comparing the WSS distribution between polyhedral and the original tetrahedral meshes, the polyhedral seems to underestimate WSS. Has anyone had this problem before? Your suggestions are appreciated.

Thanks

Amir October 4, 2011 16:46

Quote:

Originally Posted by Rohini (Post 326676)
Hi all,

I am trying to validate the efficiency of polyhedral meshes in capturing wall shear stress in blood vessels. For this, I create tetrahedral meshes using gambit and convert them to polyhedral using fluent. However, when I am comparing the WSS distribution between polyhedral and the original tetrahedral meshes, the polyhedral seems to underestimate WSS. Has anyone had this problem before? Your suggestions are appreciated.

Thanks

Hi,

I also had few investigations concerning WSS in human tissues and upon my experience non of these types were good choices. In my case, WSS was very sensitive to boundary element especially smoothness of grid lines on boundaries. Finally I could validate my numerical results just with proper mapped hexahedral cells with smooth grid lines. I think the main issue of other element types is normal vector of boundary faces which may deviate from origin one. Another tip, note that in many softwares like FLUENT (at least in 6.3 ver.) wall traction is reported instead of WSS, so if you want tangential stress you'll need some post processing steps. Hope that help you.

Bests,

Rohini October 4, 2011 16:55

Hi Amir,

Thanks for your response. What exactly do you mean by smoothness of grid lines on boundaries? Is it the boundary layer thickness you are talking about? Also, I am using fluent 12.1 and I am able to visualize Wall shear in fluent itself. However even velocity distribution for some planes that visualized seemed to be lower in polyhedral case.

Amir October 5, 2011 09:20

Quote:

Originally Posted by Rohini (Post 326683)
What exactly do you mean by smoothness of grid lines on boundaries? Is it the boundary layer thickness you are talking about?

Hi,

No, I meant that slope of grid lines should be continuous @ boundaries when we compare structure grid @ boundaries. (something when you use multi-zone topology, this feature may occur). Unfortunately, wss results for any pave topologies @ boundaries were not satisfactory.
Quote:

Originally Posted by Rohini (Post 326683)
Also, I am using fluent 12.1 and I am able to visualize Wall shear in fluent itself. However even velocity distribution for some planes that visualized seemed to be lower in polyhedral case.

I know that, I just wanted to warn you about a bug which I saw in FLUENT ver 6.3 and I'm not sure it's removed in new versions or not. You can also see WSS in 6.3 versions but it's wall traction not wss!


All times are GMT -4. The time now is 00:59.