Nonlinear modeling of pressure vessel

[alinalin]

I use COSMOSWORKS to model a pressure vessel (like cylindrical shape) made of fabric having a large force applied on the top partial flat surface and internal pressure inside. The thickness is very small, and diameter-thickness ratio is around 4*10^4.

I first use linear static analysis and it showed large displacement. Then, I checked the option of large displacement and run. It showed error: can only complete 25% load step, fail...

Now I try nonlinear analysis. I use different options of fixed time increment and it still showed the error that the load step is too big. I use shell mesh and the mesh is very coarse.

I can only try to decrease the time step now and every time it runs for a long time and show the error again. Is there any way that I can solve this problem?

Thanks for any input!

 

[peyto]

Sounds like maybe excessive deflection? Have you done any hand calcs to see if you are in the right ballpark with regard to loading and geometry of your part? I would double check your constraints and then see if I could get it to run with a much smaller load.

 

[alinalin]

Thanks for the reply.

I can not use analytical formulas to calculate because there is a large force applied on the top. I changed the thickness and found that even linear static analysis is working. That's the problem of excessive deflection, which is also too big for nonlinear analysis.

 

[corus]

It may have buckled in some way and so large deflection goes off to infinity.

corus

 

[ktop]

You are of course using a shell model for this.

The heads of cylindrical vessels frequently will buckle in compression when there is pressure inside. Do a run with a small load and look for circumferential compressive stresses. If this is happening you have to export your model and run it in geostar using a shell element with a wrinkling option.

KTOP