Implementation of varying thickness of a pressure vessel

[Fyn_Anderson]

Hi everyone,

recently I started using Abaqus.
I am designing a pressure vessel with a cylindrical part and a dome at each ending approximating a hemisphere.

Now, I have come to the first issue I don't know how to solve by myself:
The difficulty lies in the domes. As the objective is to get as close to the real vessel as possible I have to take into account the variation of thickness along the radius. Due to the winding process a lot of material gathers around the bosses. In other words the number of lamina rises as the circumference decreases.
How do I approximate this phenomenon?

This is my first post so please try not to be too harsh regarding my lack of forum etiquette.

Thank you very much!

 

[DriveMeNuts]

It sounds like more of a 3D CAD question than an Engineering Question.

 

[mskds545]

I don't understand what the question is. If you know how the thickness varies, and it is axisymmetric, I would expect it to be easy to draw. If it is not axisymmetric, the artwork might get difficult, but I agree with DriveMeNuts that that is a CAD question. Bigger picture, though, do you have a meaningful material model for this region? It sounds like you are modeling a filament-wound composite pressure vessel, which is a very difficult thing to do from the ground up even for an expert. Since you discuss varying thickness, I would guess you are modeling it as a single continuum with effective anisotropic properties and not modeling the actual windings. The effective material properties in these areas will be different than the effective material properties in the cylindrical region. The way you've described the geometry, the effective properties will probably change continuously along the heads. Do you have a good model for this? Are you following established methods (ideally codified, but at least well documented in the literature)? If you are, what methods specifically are you using, and what makes your application different at the heads than more typical cases? Are you designing something that will proven by proof test, or a final product? If you provide more details of what you are trying to do and a more specific question, you're more likely to get help (though not likely from me, since I've done very little work with filament-wound composites).

-mskds545

 

[Fyn_Anderson]

Thanks for the answers. It wasnt really a CAD problem as the geometry was already implemented. We solved the issue though. As you said mskds545 we chose continuum shell elements and did not model the actual windings. The laminate theory allows for this simplification. The simulation won't be compared to real pressure vessel behavior but only to a prestudy.