Contact Problem

[spaelicke]

Hello,

In this problem I have a cylinder and a rectangular plate.

Here is step 0, where nothing is done to either of them. The cylinder is resting on the surface of the plate.

In Step 1, a force is applied to the right side of the cylinder, while the left side of the cylinder is fixed. This shows the von Mises stress:

In Step 2, the force that was applied to the cylinder is removed, and the plate is moved upward a small distance in order to block the cylinder from returning to its original position.

I defined a contact interface between the cylinder and the plate. Here is a picture that’s zoomed in on the contact area:

I originally wanted to see the stress distribution in the contact area, but I’m guessing that this analysis doesn’t show that because I need an Advanced License. Is that correct?

Thank you,

Steven

[DenisJaunin]

Hello Steven,

I do not understand your explanations 1 and 2.

Could you do display the location of the load and the constraints of fasteners on your images.

Kind regards.

Denis.

[346gnu]

Steven,

You have a sharp edge in contact with the cylinder.

You are unlikely to get a sensible answers. The edge is a zero area.

Use a small round on the edge and consider a uniform meshing method (mapped?) as getting good pressure distributions from tets is nigh on impossible

In reality the edge (steel?) will yield and be burred over giving a finite rather than zero contact area where the material must yield and work harden. The cylinder will also show material has permanently moved. Repeated application of the load could polish surfaces or even cause fatigue cracks depending on material and myriad other factors.

If you are going to put a round on the edge then the problem becomes 2 crossed cylinders with different diameter. You may be better off carrying out a Hertzian stress hand calc. You can get the force to hold the cylinder up from a simple hand calc (or constraint reactions in you existing model). Keep the round on the edge to help the software get stiffness (force) answers and proceed by ignoring the results at the contact.

Regards

[gkoch]

Hello Steven,

have you tried the suggestions from Charles?

Let us know how it worked.

And don't forget to mark Charles' post as Correct Answer, if it helped you solve the issue.

Thanks,

Gunter

[spaelicke]

Hi everyone,

Sorry about the late response. I originally created this assembly as practice because I’m new to Creo Simulate. In this case, I was particularly interested in improving my understanding of multistep analyses and problems that involve contact.

Charles - thank you for your excellent explanation of what would really be going on in this problem. At this time, I’m just trying to practice and learn more about multistep problems and contact problems, but if I ever need to return to this case, I’ll try your suggestions.

I might post another discussion with different questions about multistep analyses soon.

Thank you,

Steven