Simplification of 3D Solids to 2D Shells for Pressure Vessel

[m_ridzon]

In ANSYS, I have been asked to take a 3D pressure vessel model and midsurface it to a shell model for FEA. See the section view in the image below. My red outline shows a change in thickness. Does anyone know of a clever way to handle this when simplifying to shells?

(before anyone asks, no an axisymmetric model is not suitable for this situation)

 

[TGS4]

See Figure 5-A.9 in Annex 5-A of ASME Section VIII, Division 2. I believe that ASME PTB-1 has some good references for this...

 

[centondollar]

Just make the shell thickness tapered in the tapered part? Keep the neutral plane constant and adjust thickness equally on top/bottom surface of the shell. Sufficiently accurate for practical calculation and you avoid complications related to eccentric neutral planes (bending and normal force not decoupled etc.).

 

[cbPVme]

What failure modes and what loads are you assessing? Why would axisymmetric not be suitable here? This is a fairly simple model, why not just use solid elements?

 

[m_ridzon]

What failure modes and what loads are you assessing? Why would axisymmetric not be suitable here? This is a fairly simple model, why not just use solid elements?

Creep Fatigue per ASME Sec. 3, Div. 5 is being performed. My colleagues are more adept at that process than me. But they have made it clear that we are not doing an axisymmetric model. I'm fairly sure it has to do with the fact that other components that I did not show in my screenshot will render the situation non-axisymmetric.