Self-contact in a Plate with a Hole

Question

Self-contact in a Plate with a Hole

(OP)

28 Feb 23 21:44

Hello,

I am modelling a plate with a hole in it, subjected to a compressive displacement load. There is an overlap of elements when the hole is compressed and I tried imposing a self-contact to the whole plate but the overlap still occurs. My model is a 3D part with shell elements. I tried reducing the model to 2D and used solid elements which seem to reduce the overlap to some extent. I have a few questions here and would be really great if someone could help me

Is reducing the model to 2D is the only way to avoid this overlap? Is this a right approach?
Am I applying self-contact correctly or am I missing something?
Why can't 3D shell edges be used to apply self-contact?

I also found a similar thread in this forum but there the issue was non-convergence. In my case I am getting solutions for different load cases but I am stuck with the overlap.

Thanks in advance!

Regards,
Vaishnav

Fig.1 Base Model with BC. The model is ZSYMM

Fig.2 3D model with overlap of elements in the top region of the hole

Fig.3 2D model with solid elements with much less overlap

FEA way (Mechanical) · Answer 1 · 2023-02-28 22:40:18

Most likely self-contact is detected here but there are some penetrations (probably due to coarse mesh). Also, it's best to use general contact, it can account for pretty much all possible contact scenarios, of course including self-contact.

Vaishnav Bhaskaran (Student) · Answer 2 · 2023-03-01 11:02:13

I tried with a finer mesh with general contact but still the overlap occurs. I also tried moving the hole closer to the load and the overlap is more prominent. Is this overlap unavoidable?

FEA way (Mechanical) · Answer 3 · 2023-03-01 12:01:07

Is it still 2D (plane stress) ? Make sure that thickness is properly accounted for by contact. You could also try scaling the penalty stiffness.

Vaishnav Bhaskaran (Student) · Answer 4 · 2023-03-02 11:13:01

No, it is still in 3D only. I have included the shell thickness in the contact. I tried both Penalty and Augmented Lagrange with scaling but no change in the penetration. Does it have anything to do with the shell element in this case? I have uploaded my model. Could you please check and let me know where am I going wrong?

https://files.engineering.com/getfile.aspx?folder=9dc78692-a6eb-4613-afcd-7

Come Join Us!

Posting Guidelines

Self-contact in a Plate with a Hole