Learning Forum

[Guangqi]

How can I mesh a quarter of cylinder symmetrically? In the picture, the mesh elements at the edge is different with the elements at the center. That may cause some dissymmetry in the final results.

https://csueducn-my.sharepoint.com/:ig/personal/guangqi_ouyang_csu_edu_cn/ESSarDju9dJLvMEvvULVn6UBoiBTvHX4L8M7afR3f68SIw?e=afMaJi

 

I do not know why I cannot upload the picture, so I upload a onedrive sharelink.

[peteroznewman]

 The site IT staff are working on fixing the error that prevents images from being inserted.

Unfortunately, the site has also inserted an emoji into your URL so I can't see that picture either.

[peteroznewman]

Inserting Images is now working.

[Guangqi]

This is the image. You can see that the meshs are not the same in circled. 

[peteroznewman]

The mesh does not have to be perfectly symmetric everywhere to have good results.

You should perform a mesh refinement study, reducing the element size and show that the results don't change.

 

[Guangqi]

Yes, the results don't change, but the results are not symmetric. Even though I have set all the boundary conditions and supports symmetric. What else I can do to improve it?

[peteroznewman]

Are the materials isotropic?  I see it mentions Composite Cure, which I know nothing about, but many composites are orthotropic.  Maybe the results should not be symmetric, especially when the results persisted after using smaller elements. Try to run the simulation without ANSYS Composite Cure Simulation (suppress it) and use Structural Steel instead. Are the results symmetric with steel?