Simulating a glass shelf bracket

[JetForMe]

I'm trying to simulate a cantilevered aluminum glass shelf bracket that mounts to a glass wall. Here's my model: http://a360.co/1Sd3Zky

The bracket consists of two aluminum discs with a connecting tube of aluminum between them. They sandwich the glass wall. An aluminum (or steel) rod goes through that, and the shelf will rest on the rod.

I want to make sure the torque introduced on the glass wall by the bracket doesn't break the glass there. So I set up the simulation and put a force on the end of the rod. The model I built is a bit simple, in that the disks, tube, and rod are all a single solid.

I run the simulation and see deflection in the rod (by the way, I couldn't figure out how to get a downward force centered over the end of the rod; it's actually off to one side), but I don't measure any stress/strain/contact pressure in the glass around the bracket. This leads me to believe I did something wrong in the modeling.

Can anyone help me out? Thanks!

[innovatenate]

To center the Force, try applying the load to the face instead of the Edge of the rod.

As far as the Contact Pressure results, they will only show on one side of the contact surface. This has been confirmed as the as-designed behavior of the solver. Please let me know if you have any questions!

[JetForMe]

Thanks, I had tried to do that and couldn't make it work, but I tried again and it's doing what I expect now.

But I still feel like my simulation is wrong. I'm expecting to see strain in the glass around the bracket, and I'm not seeing any.

[innovatenate · Accepted solution]

I'm seeing some strain, but the default mesh settings produce some rough meshing around the hole.

You could try changing the Mesh Settings:

This gives better results:

I hope that helps!

[JetForMe]

Ah, thank you. I see that now, too.

I realize the contacts were set up as "bonded" (it was done automatically by f360). I tried to change them to sliding, but it didn't seem to do what I wanted. There are two contacts, but I would have expected more. The bracket is not bonded to the glass. I don't know how much this affects the result.

I also realize: I don't know what the stress and strain limits can be before I risk breaking the glass. I don't think f360 knows this, either, does it? What are the units of the resulting stress and strain results, and how could I compare those to known values for glass, to determine if I have enough safety margin?

Thanks.

[innovatenate]

Due to glass being brittle, I would probably use the Utltimate Tensile Stress setting in the Material dialog. 

You'll find values for these in the Properties>> dialog.

The below link describes in further detail why this option would be correct. 

http://beinginventive.typepad.com/being-inventive/2012/03/safety-factor-of-brittle-and-ductile-mater...

(Note: Inventor and Fusion share similar solvers)

I hope that helps.

Thanks,