Learning Forum

[Astrid]

Hi,

I'm trying to decrease the computational time of my simulations by modelling only half of the 3D model using a symmetry region as displayed below. The idea is to press a spherical tip onto a piece of bone by imposing a vertical remote displacement and recording the reaction forces induced. I first tried with only one step which worked and gave me the same reaction force than for the whole model. However, when I'm increasing the number of steps (here I'm trying 5), the model with the symmetry is not converging anymore (whereas the full model is converging).

[peteroznewman]

Increase the computational efficiency an order of magnitude by slicing through the geometry on another plane to get down to 1/4 of the original. Now Copy/Paste the faces in the XY plane to create surfaces. Delete the solids. Open a new Static Structural model in Workbench and on the Geometry cell, set the Analysis Type to 2D. Then open Mechanical. In Mechanical, set the 2D Analysis Type to Axisymmetric.
Use Y = 0 on the bottom edge to support it, leaving X free.
Use a force on the top edge of the 1/4 sphere. You don't need any other constraints because the axisymmetric model automatically keeps the sphere on the centerline Y axis.
Now you can use very small elements at the contact point.

[Astrid]

The problem is that I want to increase the slope of the upper surface of my cube so the model would not be axisymmetric anymore. I've already worked with the axisymmetric model as you just described but now I want to evaluate the impact of increasing the slope on the reaction forces.