Kinematic Coupling applied to a hollow cylinder - will not converge

[Techsan123]

Here is what I have:

I modeled a hollow cylinder and fixed one of the bases. I created a Ref Point on the longitudinal axis of the cylinder just outside the opposite base. I then applied a kinematic coupling to this base/RP combo and under constraints checked all DOF. Unlike the BC's, this should allow the model to move in any direction. I then applied a moment to this Ref Point about the longitudinal axis.

My material is Hyperelastic and I am using a Tet Mesh. I have already adjusted the step increments to allow the processor more time and increments to converge.
If the moment is small enough the analysis will converge however, I am looking for the cylinder to actually deform and twist/buckle as a rubber hose would if you were to apply the same actions as described above. When the moment is large the analysis will just abort even with the adjusted step increment.

Why won't this happen with the described method and conditions that I am using? Is it the mesh or material maybe? Any help is greatly appreciated.

 

[Drej]

Do you have NLGEOM activated within the step?


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[Techsan123]

Yes it is on.

 

[bassmanjax]

Have you applied any other constraints to the ref point?

Martin Stokes CEng MIMechE

 

[Techsan123]

No, the Ref Point has no Constraints or BC's.

 

[bassmanjax]

Then I'd try fixing all the other degrees of freedom on the ref point, other than the rotation you want, to zero and see if that works.

Martin Stokes CEng MIMechE

 

[Techsan123]

No this does not work but thanks.

 

[Techsan123]

I will attach my file.

 

[bassmanjax]

One thing I would also try is to use a linear-elastic material model instead of the hyperelastic model that you are using. If the model works with the linear-elastic material, then you know that the hyperelastic material is suspect.

Martin Stokes CEng MIMechE

 

[bassmanjax]

Another point I just remembered..

If it's a simple cylinder, why use tets? You could easily use bricks or even shell elements if the diameter/thickness ratio is sensible...?

Martin Stokes CEng MIMechE

 

[bassmanjax]

Update - job runs to completion with linear-elastic material. I think you need to look at the hyperelastic material you are using.

Martin Stokes CEng MIMechE