Strange results in a displacement simulation

[andr333]

Hi all,

I am currently studying the displacement of a sheetmetal plate at my work.

For this specific case I have a certain load applied in specific spots, and I need to know to max displacement because I have a limit i cannot pass, in my case 0.9 mm.
When I run a simulation with this plate I have created I got a result of 0.93 mm which is not good enough, so I add a sheetmetal bar on the bottom of the plate, that in reality will be assembled together using rivets.

The problem is when I run the simulation with the bar, I get the same max displacement with almost the same distribution, and this assuming that the whole contact surface of the bar and the plate are bonded.

I cannot find any mistake in the way I have setup the simulation and assembly (i have discussed this with fellow colleagues at work, but this cannot be true, it just doesn´t make sense that adding a bar is not increasing the stiffness of the whole assembling.

Can anyone give some tips or advice on how to to constrain and set up 2 parts together?

 

[brep]

In Creo Simulate it's really each to mix a shell/beam model. There are many ways you can do this, but here's one: Start with the shell (surface) geometry and then create a surface region (in Simulate's interface) to "burn in" lines to your surface where you want the beams to be. If the surface is planar then you can just create the surface region as a sketch on that plane. Otherwise you can project the sketch onto the curved surface.

You can then assign beam properties to these (or indeed any curves/edges) in your model.

Make sure your units are correct for the beam properties so that they contribute meaningfully to the stiffness of the model. You can use the Review button in the beams dialog to see the inertial properties for the section you defined...

You can also see a graphical preview of the beam on the model once it is assigned, which helps to give you a visual check.

Finally, there are a ton of postprocessing options in SImulate so that you can check the beam stresses individually (look for the "Display Location") tab in the Results Window Definition dialog box...

Does this help?

PS you should be able to do a good sensitivity study on some of the beam parameters (thickness, web length, etc) to see what effect changes make. Then optimize....