Simulate Early, Simulate Often... In Rhino
Symmetry test of sphere on plate
Thank you first of all for the Spring model which is a great tutorial and good starting point!
I have been experimenting with SnS trying to get a handle in particular on how to get forces into more combined shapes.
As I haven't spent much time as yet I may have missed something obvious so please excuse me but I wanted to share the results of what I thought would be a simple test - a sphere on a plate.
In general the solution looks fine in that the stresses and deformations are consistent with what would be expected. However I did notice that there does seem to be an asymmetrical nature to the solution shown in rhino-sns-test.png attached. The plate is restrained symmetrically on all four sides and a scalar load is applied on the sphere. When deformed the plate appears to show higher stress on the top and left than the bottom and right and the base of the plate does not appear to deform parallel to the axis of deformation. Maybe there is a simple explanation that someone would share with me?
Also I was wondering is there a mechanism to provide a loading on a specific point rather than on a surface? For instance in my sphere model I would like to be able to "push" the sphere laterally with a vector force from one side...
Thank you for what looks like a very promising tool - I look forward to getting to grips with it better and seeing how it develops :)
Views: 202
- Attachments:
-
-
rhino-sns-test.png, 282 KB -
SnS_Report_002.png, 41 KB
-
Rhino-SNS-Test.3dm, 2.8 MB
-
Replies to This Discussion
-
Permalink Reply by Vadim Shapiro on
-
Thank you for posting this example. Actually, this test is not nearly as simple as one may think. The plate is very thin relatively to the size of the sphere, and the area where sphere touches the plate may also require very fine resolution. In other words, simple geometry does not mean simple stress analysis.
Some asymmetry should be expected with any numerical method that does not explicitly accounts for it. Scan&Solve computes displacements and stresses numerically on a grid of space that is not aware of the problem's symmetry and therefore does not respect it. Increasing resolution should make asymmetry much less noticeable.
As for your last question: Scan&Solve does not allow point loads because they are not physically meaningful. But you can always "cut" a small face on the sphere, or split the sphere into several faces, and apply loads only to some of these faces.
Hope this answers some of your questions. Thank you again for sharing!
© 2024 Created by Michael Freytag.
Powered by
