I'm having troubling constraining t 2

[Angle C]

I'm having troubling constraining this simple static analysis (see attached image). The sheet metal acting as a latch is fixed to ground (door assy) at one end, a force is applied downward in the middle bending the metal strip, and because of the lengthened path retracts the latch at the other end. I constrained the right side (the sliding side) with a planar constraint. The simulation runs, but no translation of the side that should slide occurs. The material only deforms under the button. I think I may know what the problem is but don't know a work-around. My thought is that the planar constraint over constrains what would happen in the real world. Some of the metal strip has to be pulled downward under the button, but at the start of the simulation this material is part of the surface with the planar constraint. Surely, this type of problem in common is simulation, but for me this is a new one. Anyone have any suggestions?

 

[GregLocock]

You are using a non linear FEA aren't you?

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[FEA way]

Which software do you use ? And could you say which surfaces exactly are featured in that planar constraint (What does this constraint actually do ? It seems to be a term specific for your software.) ? It would be best to mark these surfaces in the attached picture. I suspect that they belong to the gray part on the right but I'm not sure.
My first guess is that you should use contact interaction with frictionless or frictional sliding allowed instead of that constraint but let's clear out the terms first.

 

[GregLocock]

The metal strip needs to be able to slide over the point of the green bit, and the latch needs to be able to slide to the left. But if you are using linear FEA then nothing will move.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[rb1957]

and to clarify you need geometry (large displacement) non-linear (not material (plasticity) non linear).
this is because you want the model to update node positions based on displacements.
linear FEA assumes small displacements (and doesn't update node geometry).

Consider loading the model with an enforced displacement.

another day in paradise, or is paradise one day closer ?