Ratio between contact stabilization energy and the strain energy

MarioS

Hello everybody,

could someone please help me with the topic contact stabilization energy, what should be the allowed ratio between the contact stabilization energy and the strain energy in hypermesh? Where can I find this info?

For example I am doing a pretension analysis of a 3D bolt, there are contacts with friction. And I am getting this curve (see attachment 1) with the seetings of CNTSTB (see attachment 2). I don´t understand why is the contact stabilization energy so high at the end of the analysis if I set the S1 parameter to 0.01?

 

Thank you in advance.

Mario

1.PNG

2.PNG

RobHoglund

Hi Mario,

Generally you want the contact stabilization energy to be quite lower than the strain energy in the model to better match the physics at the end of the step- perhaps 10 times less.  Since we are introducing extra energy into the system, the less stabilization, the better.  The contact stabilization energy is effectively computed from the introduced damping for the contact when CNTSTB is used, based on reference stiffness.  So this value may require some experimentation.

You can try reducing S1 parameter further, all the way to 0.0 to reduce stabilization at the end of the step, and if there is trouble with convergence there but it solved with S1 = 0.01, increase to some very small amount.

If your contact type is S2S or you're using LGDISP, the CNTSTB parameter should be effective.  You may also try removing CNTSTB and switching the contact sets as well, which can also help with convergence, and verify that the model solves without friction.

Thanks,

Rob H.

Community-Pasted-Attachment-1617216811647.png