shell-solid contact

[james80]

hi,
I am working on shell-solid contact analysis.
I built a test model which consisted of a cylinder and reactangle
the cylinder was meshed with solid185 while the rectangle with shell63 with thickness of 0.1
I setup the contact using MPC algorithm with always bonded type contact.
In the first case, the shell and the solid elements were touching each other while in the other case there was a 0.05mm gap.
I applied a force on 1000N in z direction at the midnode and obeserved the displacement in z-direction.
Initially i thought that as long as the contact pairs are within the pinball region, it will not matter if the shells are in contact with solid or if there is a gap of 0.05mm.
But the deformation results were different for both the cases,
For case with the gap): it was 0.003554 while for the touching case it was 0.004006.
Can anyone think of any reason why this is happening?
I apreciate your help
james80

 

[cbrn]

Hi,

my understanding is that in the second case you have 0.05 mm more of elastic "pseudo-material" whose properties are derived from the underlying elements. It is really uncommon to have two components "bonded" while they are not touching (in the real world, how would this be possible?). The elems are considered "in contact" or "near contact" depending on the pinball region, yes, but the contact's properties are derived from the "mating" configuration, so moving the parts one with respect to the other have an effect on the results, I believe...

Regards