Cannot get proper contact between cylindrical parts using surface-to-surface

[drennon236]

I am trying to get proper contact using surf-to-surf and finite sliding. I am in dynamic, implicit step. Below you can see how my contact is done. I have tried in the first step fixing the inner cylindrical concrete part and the outer steel cyliner part, while applying small amount of gravity. Then fully release one part in the second step to get contact. I have also tried several other BCs and load conditions, quasi-static application and moderate dissipation. Below there is no BC's or load applied. All I want to do is to get contact between the two cylindrical surfaces. I have also surface smoothing turned on.

Surface-to-surface

Initially there is sticking in my model, but then there is less and less contact for some reason. There is no load, BC's or deformation in any direction. Should it not always remain sticking then? Picture below is for inner concrete part.

 

[FEA way]

Have you tried modeling this with small penetration between the cylinders and then using either strain-free adjustment or interference fit to resolve this penetration ?

 

[drennon236]

I have tried "Ajust only to remove overclosure" and "Specify tolerance for adjusment zone" options but this gave me the results seen above. Also tried interference fit for which the results are shown below. All of this was done with small penetration.

Penetration and interference fit gives displacement when there are no loads or BC's. Is there anything else I could try? Feel like Ive tried all options and that this should be a fairly simply problem. I can mention that small sliding with initial clearence gives sticking between the parts for all steps, but small sliding is not valid in my case.

 

[drennon236]

There is deformation on the top as you can see, I am not sure where it comes from. But I can use boundary conditions to have no displacement and I get good contact - any idea what I should do next?