peglor
Mechanical
- Sep 10, 2002
- 109
I have a ProMech model that seems to give me the same displacement at several grid refinements, but gives quite drastically varying peak stresses. The stresses are linked to point features, but the behaviour of the model at these points is of interest for the results.
I get the feeling that I can keep refining the mesh as much as I like and the peak stress will continue to rise while the run summary assures me I have achieved convergence to 1 or 2% in the model. This leads to two questions:
1. Should I continue refining my mesh more and more, eventually achieving convergence (When the change from mesh to mesh is less than the stated convergence for the analysis possibly)?
2. What does the convergence criterion used by ProE mean in terms of the quality of the results?
I get the feeling that I can keep refining the mesh as much as I like and the peak stress will continue to rise while the run summary assures me I have achieved convergence to 1 or 2% in the model. This leads to two questions:
1. Should I continue refining my mesh more and more, eventually achieving convergence (When the change from mesh to mesh is less than the stated convergence for the analysis possibly)?
2. What does the convergence criterion used by ProE mean in terms of the quality of the results?