3d Contact modeling between two cylinders Question

[sokn]

Hello,

I am modeling a cylinder undergoing a four point bend test in abaqus. In the model there are 2 contact interactions defined between the test fixture and the cylinder undergoing bending. These contact surfaces (of the fixture and cylinder) both are smooth cylindrical shapes. In the cylinder I have made grooves on the bottom to apply the boundary conditions along a line rather than a point.

No matter how I seem vary the contact model, I get errors. I have attempted both surface-surface interactions, and node to surface interactions. I have made a similar model only with a rectangular cross-section that converged to a solution with <1% error from hand calculation. I believe the problem to have something to do with the contact definition between two cylindrical surfaces, but am at a loss on how to correct the issue. I would greatly appreciate any help that can be offered.

I have attached my cae model.

Thanks,
Scott

 

[compositesFEAguru]

My initial guess is that you may be having overclosure issues. One thing to try is to select "Adjust only to remove overclosure" under the Slave Adjustment tab in you Interaction dialog box. Try refining the mesh as well. A third thing to try is replace your load application rollers with analytically rigid surfaces.

Hope one of these help,

Firehole Composites

http://www.firehole.com

 

[IceBreakerSours]

No normal behavior is defined in the interaction.

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[rstupplebeen]

The first thing that I would do is utilize the 2 symmetry planes that you have. This will help with many of the rigid body motions that can be present in the test. After this carefully construct your boundary conditions fixing the bottom of the fixture or as compositesFEAguru mentioned use rigid surfaces. I would also model the top fixture to avoid applying a concentrated force. I hope this helps.

Rob Stupplebeen

https://sites.google.com/site/robertkstupplebeen/

 

[sokn]

Thank you for the various responses, I greatly appreciate your help. I will play around with the normal behavior, thanks for the suggestion.

rstupplebeen--When you say I should be using rigid surfaces, is that different than how I have the model defined currently? I applied a rigid body constraint to the "top fixture". I am not attempting to simplify the model using symmetry planes because this is just a precursor model to an irregular geometry that is not symmetric but is similar to a cylinder.

Thanks again,
Scott

 

[rstupplebeen]

Either as you have it or preferably with analytically rigid surfaces as compositesFEAguru suggested.

Rob Stupplebeen

https://sites.google.com/site/robertkstupplebeen/