Why the strain varies with mesh size?

[leess8888]

Hi,

I have calculated a case that the material bears surface traction.

When I did the mesh sensitivity, the output max log strain increased from 0.133 to 0.741 with the element size decreasing from 0.5 to 0.03?

I can't understand this result. Can you help explain the theory inside this problem?

Thanks

Wyn

 

[rb1957]

I imagine you're discovering a stress peak with the finer mesh that the larger mesh didn't detect.

Remember the stress results for a QUAD are sensitive to the stress gradient, and are possibly most accurate at the element centroid.

I imagine you're running linear FEA. a question to ask yourself is "when do i violate linearity?" and then "does it matter?"

another day in paradise, or is paradise one day closer ?

 

[leess8888]

Thank you for your message.