misconvergence with constant force convergence value?

[cbrn]

Hi,
I'm facing a non-convergence with a strange characteristic:
after some equilibrium iterations, the Force Convergence Value tends to a constant value, higher than the Criterion of course.
The Workbench model is cyclically-symmetric, the cyclicity is enforced via CPCYC. It also contains surf-to-surf contacts and, unfortunately, also node-to-surface MPCs due to the presence of remote forces.
I suspect that the MPCs are not correctly handled in this cyclic analysis.
Have anyone of you already come across this kind of non-convergence?

Thanks in advance,
Regards

 

[ansysfreak]

Hi

I had the same problem with the cyclic command.
But I think the development will solve this problem in version 12.0.
They recommended I use the CPCYC command with remote force, my problem was solved.
If you suppres the remote force, does the analysis converge then?

Garry

 

[cbrn]

Hi,
thanks for your help.
I know CYCLIC must not be used inside WB v.11.0; moreover, the fact that it "seemed" to work in v.10.0 was purely casual, in fact it was never documented.
However, I'm using CPCYC here. The fact is that the Contacts/Targets and thus also the MPCs are processed AFTER the Commands snippet, so I've got no chance to correctly handle the pilot nodes and also the nodes at the cyclic sector's "edges" which are created by WB itself. I'm now about sure that it's what is puzzling the solution. In addition, in my case it will contaminate also the solution of the Contacts because after all the algorythm is the same.
If I suppress the remote force the solution converges.

Regards