Learning Forum

[naveen28]

Hi,

I am running an static structural analysis on a plastic part (HIPS material). I am getting varying stress based on their mesh size (used sphere of influence). So, I conducted a mesh independent study and below is the image of it (Also attached the image).

 

I want to know if the above study is enough to conclude on the mesh size? Please clarify with explanation.

 

Thanks,

Naveen

[peteroznewman]

Hi Naveen,

The plot shows evidence of a singularity in your model. Do you have a sharp interior corner?

Please show a zoomed in view of the geometry in the region of the maximum stress.

Read this discussion on Mesh Refinement Study. The plot should level out at small size elements.

[naveen28]

Hi Peter,

Please refer below image for the maximum stress region

[peteroznewman]

The geometry has no sharp interior corner, so if there is a singularity, it's not caused by geometry.

Please show the smallest element size on that same view. It might be the case that you just haven't got the element size small enough to converge on the "true" stress.

How is that part being loaded?  Is there contact?  If so, please show the other part and its mesh.

[naveen28]

Hi Peter,

Below is the image with smallest mesh size of 0.01 in (at the high stress region, used sphere of influence).

There is no contact with other parts at this region. 

[peteroznewman]

Continue making new points on the graph, and throw away all the results except for 0.1 and plot the data for 

0.01


0.006667


0.004444


0.002963


0.001975


0.001317

These sizes use a factor of 1.5 reduction. When the graph levels out, you can stop plotting points.