Contact modeeling

[halston]

Hi all,

I am modeling an insertion of a solid pin to a mandrel-specimen assembly. specimen has a central hole and mandrel is inside the hole. Pin will be pushed down in the hole which is drilled in the mandrel.
I've defined two pairs of contacts, pin-mandrel and mandrel-specimen. I've used both normal and tangential contact between the surfaces. and then I run the model and get some results. But it seems that the stress distribution is not correct in the specimen. I just get to regions of stress and strain in the specimen and I'm sure that it's not correct.
I've attached a drawing of my problem. In the drawing 1 is the pin, 2 is the specimen and 3 is the mandrel.
I'm using isotropic tangential behavior with penalty friction formulation and nonlinear normal behavior with constraint enforcement method of penalty(standard) for both contact pairs.
Can anyone kindly help me with this?

 

[rstupplebeen]

check out remove overclosures in the contact definition. It is good for assemblies that are slightly penetrating or for modeling press fit situations. This looks like axisymmetric geometry so that should help with solution time and free DOFs. I hope this helps.

Rob Stupplebeen

 

[halston]

Hi Rob,

Thanks for your reply. I tried to remove aoverclosure by toggling on the "Adjust only to remove overclosure" in edit interaction dialog box. But it didn't work and I still get almost the same stress distribution in the specimen. Is it the right way to remove overclosure?
As you said, the geometry of my model is axisymmetric.
On top of that, when I increase the diameter of the pin to have a higher diametral interference, the solution will be aborted due to errors!
Any suggestion is highly appreciated.

 

[rstupplebeen]

Check your error messages. I assume there is something about poor element shape. Can you post your cae or a picture of the undeformed and slightly deformed mesh?

Rob Stupplebeen

 

[halston]

I checked my model again and I found the problem. Everything was ok, I just needed to remove the mandrel instance in the visualization module in order to see the correct stress distribution in the specimen.
But now I have another problem. I need to model slip between the mandrel and the specimen when I rotate the mandrel. I change the shear stress limit or elastic slip parameters but there is no change in my model. I should mention that the specimen at the beginning of the rotation is in plastic region and s33 has exceeded the shear stress yield.
I expect that the surface of the hole rotates with the mandrel until slippage occurs. It's not happening in my model now!

 

[rstupplebeen]

Do you have friction in the contact definition?

Rob Stupplebeen

 

[halston]

Yes, I do. I have defined isotropic friction with friction coefficient of 0.1.

 

[rstupplebeen]

More information is needed. Could you post your files or pictures?

Rob Stupplebeen

 

[halston]

Attached you can find my files. Please have a look and let me know how I can improve it. Thanks a lot!

 

[rstupplebeen]

Winzip does not like your file. Please re-zip it and resend.

Rob Stupplebeen

 

[halston]

I will attach them in two different steps and there is no need for winzip or winrar. Here is the .cae file.

 

[rstupplebeen]

I believe that your boundary conditions are off. Specifically CenterBC you have as Y symmetry. This constrains the part in the Y direction so it can not be compressed. That is why you have a large stress at the axis of symmetry. I hope this helps.

Rob Stupplebeen

 

[halston]

Can you please explain the relationship between the centerBC and slippage between the mandrel and the specimen surfaces? I can't understand that!

 

[rstupplebeen]

You are holding the center line in the Y (axial) direction instead of just the bottom of the part in Y. That is why there is a stress concentration on the center line.

Did you intend to the the indenter to have a dimple in the center?

I hope this helps.

Rob Stupplebeen