Simulate Early, Simulate Often... In Rhino
I am currently working on a problem and figured I would see if anyone has any insight as to how to model things better for the purpose of an FE study.
I am working on a hydraulically powered arm. This arm revolves around a pivot pin. At one end of the arm is a weight. At the other is another pin that a hydraulic cylinder connects to. (simplification....)
I am trying to model the stress concentrations in the design, but am finding that I am getting concentrated stresses that are artifacts of the meshing and of the way the loads and constraints are applied, which is making the results rather difficult to exploit.
Looking at the results it looks to me that there is a big concentration at the restaint boundary, which is expected since the restaint is being terminated rather abrubtly. It also looks to me like there is only a limited number of elements being used, and some of the stress concentrations are clearly related to the element boundaries, and to the nature of the fixity of the restraints.
Is there a better, more accurate, way to model a pin and to apply more realistic loads and restaints?
I posted this request perhaps a little fast - I was looking for a resolution setting but hadn't found it. I have founded now, so I suspect I will be able to get smoother results by increasing the resolution.
I am, however, still interested in ways to model the behavior of pin - i.e. no shear at the surface....
Can you do a hand calculation to determine the reaction force at the pin where you are currently applying the restraint? Then apply your manually computed reaction force as a load on the appropriate faces of the interior of the pivot. For numerical stability, you will need a restraint somewhere. Can you put it at the opposite end of the link? Something like this: (your pivot represented by the middle hole)
This will move the restraint away from your region of interest and may give you a clearer picture of the stresses there.
Thank you for your thoughts. Your approach should indeed lead to a better picture around the points in question.
This post is inspired by another question I received via email. The following approach to pin connections may be an option in some cases.
Scan&Solve approximation:The cylindrical faces have been split into quadrants that are "mostly tangent" to the coordinate directions. This allows partial restraints to be assigned to these faces, allowing small rotations.
Thanks - this is food for thought. Although an approximation this approach may help to lessen the local impact of the over-specified restraints I am using now.