mesh convergence analysis

[Geros]

Hello

I have a part that consists of many bodies and i want to run a mesh convergence analysis.
This analysis has to run in the entire model or in a part of it?

 

[Naffoussi]

You could perform a static analysis by applying 1g in the three directions separately only on the part of interest. And compare the stiffness and stress obtained to judge if your model mesh convergence will gives you an accurate results.

Seif Eddine Naffoussi, Stress Engineer

http://www.Innovamech.com

33650 Martillac û France

 

[corus]

There are many ways to check if your mesh has converged. Firstly if you're applying a pressure load then you'd expect the stress normal to the surface to be at that pressure value. Secondly, in regions sufficiently remote from features, then the stress may be calculated analytically, for example in a cylindrical shell wall. Thirdly, you may have measurements of stress or strain in certain regions which would validate your model. Where you have a stress concentration which is of interest, such as a fillet radius, then it's wise to refine your mesh in that region. Refining the whole mesh might not be possible due to memory constraints, or run times.

 

[IceBreakerSours]

Additionally, you could look at strain energy as a function of # of elements.

Are you new to this forum? If so, please read these FAQ:

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[stressebookllc]

Also look at your strain energy error value epsilon. Needs to be really small, in the order of E-10 or so.

http://www.stressebook.com

Stressing Stresslessly!