I have problems with a model that have around 30 supports.
It runs and I can’t see any result because it never converges.
I’m changing to zero the friction multiplier on the "Load Case Options" tab of the "Static Load Case Editor" and the problem is the same.
I have also tried changing the friction factor in the CII configuration but it doesn’t work either.
Then, I erased the input values of the friction coefficient, support by support in the restraint list, and the iteration converge.
Does anybody know how I can do to fix it in Caesar?
Include the friction values is important for me.

Then either the problem isn't "friction", or you didn't change the multiplier on all of the load cases.



See the 2nd part of the sentence above.



Start adjusting the "friction settings" on the "Computation Controls" tab of the Configuration Module. Make a small change, try the run, if it doesn't converge, change the configuration again. NOTE, changing these configuration settings requires you to go back to input and re-Error Check the model.

You can also find additional information on this subject if you use the [Search] option, to the right, just above the Calendar. Search for "convergence".

Search for "convergence" as Richard states. convergence problems can also be rooted in +y and Gap supports. Hit "F2" while CII is iterating to see which supports are involved and the "status" of these supports to focus in on the convergence issue. Sometimes it is a bit of a learned art, but the lessons learned come quick.

You must make sure how you support swill look in real.
If the pipe has PTFE plates underthem then friction factor is zero.but only if u have a metal to metal contact use 0.3 which never happens anyway.
second it may meen your system u designed is unbalanced so CII dont accept the supporting matrix so double check that.

Ack! Zero friction!

Most of us prefer to use a PTFE - to - steel friction factor of 0.1 or so.

To get back to the original thread, it's also possible to interrupt the solver and single-step through the iterations. This will tell you WHICH supports are not converging, and you can fuss with them manually. My experience has been that usually the "problem" supports are those relatively far from terminal points, so making drastic changes to them has little or no impact on nozzle loads.

Note that gradually ratcheting down the global friction factor is NOT conservative for loads on terminal points. You could easily end up with a steel-on-steel support adjacent to a nozzle with a friction factor of .15 or so. This is hardly a valid assumption when trying to evaluate loads on rotating equipment.

Tips.

*Friction and gaps are not best friends, do you really need both???
*The over use of gaps cause unstable models
*Gaps on vertical lines are especially prone
*With some Judicious trimming of gaps and friction, one can (usually) make the model work.
*Be careful what you do and dont add friction/gaps too,,,,,,,,,
*Its usually only one support which tips over the convergence process, although finding it can be tricky. trim gaps and supports until model runs, and then once the offending support is identifed, re-code supports until the best solution is found.