CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Pointwise & Gridgen (https://www.cfd-online.com/Forums/pointwise/)
-   -   High skewness for a 2D unstructured domain in Pointwise (https://www.cfd-online.com/Forums/pointwise/157386-high-skewness-2d-unstructured-domain-pointwise.html)

Stefanodm01 July 28, 2015 05:29

High skewness for a 2D unstructured domain in Pointwise
 
2 Attachment(s)
Hi everybody,

I am a student at the first experience with meshing softwares like Pointwise, so I apologise if the question I am about to ask may look to many of you as a trivial one. However, I cannot seem to find a solution to my problem in previous threads on this forum nor anywhere else.

I have to mesh the model of a particular kind of power MOS, composed of multiple layers of different materials.

When I try to create an unstructured 2D domain for the suface laying in the XY plane of one these layers, however, Pointwise generates triangles with such a high skewness as to be recognised as degenerated when I try to initialize a block that includes said domain.
This issue does not present itself for any of the other surfaces of my model, not even for the ones which have the exact same rectangular geometry and are delimited by connectors with the same number of points.

Do you know why this happens and what can I do to solve the problem?

I am attaching a couple of screens in order for you to better understand the situation. In such screens, the different structure of the meshes for two identical surfaces should be clearly visible.

I thank you in advance for your time and support.

jchawner August 2, 2015 16:57

Hi Stefano:

It's really difficult to tell what's going on with your mesh from the pictures so all I can do is cite generalities.

Triangles in your 2D domain will be skewed if you use a connector spacing that's inconsistent with the geometry of the domain's perimeter. In other words, if you have a highly concave geometric region and only but a few grid points in that area, your cells will be skew. This analogy extends to 3D.

My advice would be to Examine the minimum included angle metric on each of the domains on the block with the skew cells. Find where the most extreme values are. Look at the connector grid point distribution in that area. Adjust accordingly.

Good luck.


All times are GMT -4. The time now is 04:47.