I have been running a study on a trailer coupling ball.

For the first run i used the default mesh density (not a draft quality mesh) ;

Results:    Deformation is as expected.

                   Max stress (node) = 729 N/mm^2

                   Max stress (element) = 582 N/mm^2

                   Max displacement = 1.605 mm

So I set a mesh control to refine the mesh in the area of interest. I increased the mesh density in that area until the difference between Node and Element stress values were within 10%.

Results:

1. Displacement

2. Node stress

3. Element stress

Can anybody explain how stress has increased if displacement decreased? This trend was seen gradually in the results as I refined the mesh.

Also, the max stress calculated is roughly 40% higher than the fatigue limit, But the component has passed a 2 million cycle physical test at this loading.

Any help would be greatly appreciated.