Learning Forum

[rahmanmahfuz11]

Hi,

I have a composite tube of glass fiber and carbon fiber. I have set up the delamination problem in between these two layers by the Exponential for Interface Delamination Cohesive zone and I am twisting the two layers in two different directions on the open end. However, the solution always diverges whenever the twist angle reaches around 7.5 degrees regardless of how small the load stepping is. I am attaching the loading conditions here. z is the axis of the tube.

I would be grateful if anyone here could help me with this issue.

Thank you.

[Sean Harvey]

Just a few checks. Delamination is numerically unstable, so if your setup is correct, you may still run into some challenges. .. With that said please entertain my suggestion to check these:
Please check you have large deflection on.
Also, make sure you have auto time stepping with a large enough maximum substeps, probably in the 1000s.
Sometimes the issue is the material properties and units, so just a check on that.
In some loading/material combinations the delamination is unstable so this can happen.
Try on simple mesh of say two thin plates being pulled apart with displacement loading. I know your tube model seems simple in itself, but just a suggestion.
Now what you can do is if the above are checked, you can add some viscous regularization to try and see if that helps. This adds fictitious viscosity to the czm model. You would need a command object and specify c1 appropriately (see below). Too large and the results will be influence with too large a damping coef. Too small and it won't help. It could be that the non-convergence is due to other issues, but at least you can try this and see if it helps, especially for cases that goes unstable suddenly.

TB,CZM,,,,VREG
TBTEMP,22.0 ! Define first temperature
TBDATA,1,c1 ! Define damping coefficient at temp 22.0

For more information, you can search the Ansys help on viscous regularization. Hope this helps.

Regards Sean

[rahmanmahfuz11]

Thank you Sean. Other than viscous regulation I did everything & they are okay. I'll try the viscous regulation.