contact problem - too large penetration.

[vadims]

I have problem with contact modelling – too large penetration. Warning message say: „Convergence has been achieved in spite of large penetration. If this message is repeated frequently, we recommend either increasing penalty stiffness <FKN>, or enlarge penetration tolerance <FTOLN>”. Changing those factors does not change anything. Can somebody help?

 

[jiligeo]

Which values did you use for FKN and FTOLN?

 

[Stringmaker]

I'd suggest you ramp your load more gradually. Try issuing the following:

KBC,1 ! ramp loading
NSUB,20,100,1 ! large number of substeps
SOLC,ON

I've had this happen before and have always been able to correct it by changing a few small things. If penetration is still unacceptably high you will want increase FKN.

 

[cbrn]

Hi,
also make the contact stiffness update at each equilibrium iteration. Though this requires the most of computation time with regards to other possible settings, this ensures also the most precise results.

Regards