Cohesive crack propogation

[FEMbiaatch]

I'm working on the explicit simulation of needle insertion into a thermoplastic polyurethane foil (0.4mm thick) using a cohesive (orphan mesh elements) zone modeled between two parts of a phantom corresponding to the outer diameter of the needle tip. While using an intricate bevel needle, the leading cutting edges transition from outer to inner edges as the needle penetrates. Thus, the fracture doesn't completely follow a circular path (outer diameter) as predefined, and I can observe continuum elements without cohesive traction-separation behavior failing according to the damage variable (D). Is there a way to include cohesive elements in a reasonably large region instead of a defined crack path? Or any other way to avoid damage to other continuum elements? Thank you.

 

[FEA way]

There can be only one layer of cohesive elements through the thickness and you would have to change the constitutive response from traction-separation to continuum (standard material models) in order to account for finite thickness. You could also consider different modeling techniques, depending on your requirements.