Skip to main content
May 13, 2015 12:47 PM UTC
Question

Re: Mesh Question

sjohnstone,

I ran your test model with my own boundary conditions and mat props, and as you can see in the below images, there is very little change in the solution and convergence. I made three separate 2D runs, 1) with default automesh, 2) autogem set to 1 for element size, 3) autogem set to .25 for element size.

Study 1:

study_default_vm_stress.JPG

study_default_vm_conv.JPG

Study 2:

study_auto_1_vm_stress.JPG

study_auto_1_vm_conv.JPG

Study 3:

study_auto_25_vm_stress.JPG

study_auto_25_vm_conv.JPG


    1 reply

    May 13, 2015 1:00 PM UTC

    The other thing I forgot to mention is a visual inspection of the stress contours.

    In Steven's first image (Study 1), the stress contours to the left-hand side are visibly jagged and sharp-cornered - to me this would immediately suggest that a finer mesh is required.

    In Studies 2 and 3, the contours look much smoother, so I would have less concern over the convergence.

    It looks like Study 2 underestimates the stress slightly compared to Study 3 - I make it 1895 vs 1916 pounds-mass per inch-square-second (approx), which is a 1.1% difference. Study 1 looks like 1870 (even more approx), which would be 2.4% less than the Study 3 value - still not bad for most engineering purposes.