ASME Design by analysis - ELASTIC-PLASTIC STRESS ANALYSIS METHOD

[Bozo_Sam]

Hi there,

I have another question with regard to the ASME Section 8 Dev2 PArt 5. For the calculation of protection against collapse, we have three methods: Elastic stress; Limit load, and Elastic-plastic stress. Since I have good experience with nonlinear FEA I would like to perform Elastic-plastic analysis. What I am having a problem with is defining the limit for Elastic-plastic analysis. In the code it states that

"A global plastic collapse load is established by performing an elastic-plastic analysis of the component
subject to the specified loading conditions. The plastic collapse load is taken as the load which causes overall
structural instability. The concept of Load and Resistance Factor Design (LRFD) is used as an alternate to the rigorous
computation of a plastic collapse load to design a component. In this procedure, factored loads that include a design factor
to account for uncertainty, and the resistance of the component to these factored loads are determined using an elasticplastic
analysis (see Table 5.5).

On the figure, we can see that the LRFD has a factor of 2.4, which seems extensive.

Could someone please provide a bit of guidance on the limit for this type of plastic analysis

Thanks!

 

[TGS4]

The 2.4 factor is the design margin against ultimate tensile strength, as indicated by Appendix 10 in ASME Section II, Part D.

Btw, you have a very old version of the Code. The current Table 5.5 uses the Greek letter Beta, where Beta depends on the Class of Construction within Division 2.

Changes are constantly being made to the Code. Always use the most recent, especially for Part 5.

 

[Bozo_Sam]

Thanks, TGS4. Yes indeed I checked and I quoted the 2013 edition, but I do have a 2019 edition as well.

I am starting to see the advantage the elastic-plastic approach has over the elastic or limit load. The factor of 2.4 might seem extensive but the plastic limit stores a lot of strain energy of the structure.

I have yet to study in detail how to characterize the plastic material properties in accordance with ASME. I can see some long equations there =). Do you mayme know of any sheet that would make that any easier?

There is another topic that I am having a bit of trouble understanding. Since most of the pressure vessel designs include some form of sheet metal I would assume 2D shell elements are sufficient for the analysis. Yet I see shell elements really being used in pressure vessel analysis. Since elastic-plastic analysis takes longer to execute and doesn't benefit from linear superposition, I would try to use shell elements as often as possible. Is there a general rule of when shell elements can be used and when solids are necessary?

Thanks!

 

[TGS4]

The stress-strain equations in Annex 3-D are long. Setup an Excel spreadsheet or a Mathcad file. And use the 2021 Edition - changes were made to Annex 3-D in the 2021 Edition.

Shell elements can be useful for Protection Against Plastic Collapse, are less useful for Protection Against Local Failure, have good utility in Protection Against Collapse From Buckling, and have limitations in Protection Against Failure from Cyclic Loading: Fatigue. Use your engineering judgement.

 

[Bozo_Sam]

Thanks, TGS4. I prepared the sheet and it is actually pretty straightforward.

Thanks for also explaining the shell vs solid element use. I have a good idea of when shell would be appropriate, I just wanted to see if there are some specific rules.

I have another question with regard to the load scaling for elastic-plastic approach. While it is straightforward on how to scale the mechanical loads (multiply them with the needed factor) I have trouble interpreting the scaling of thermal loads. If we were to have the LNG tank for instance which operates at -165°C, hence the dT would be 185°C for the thermal loading analysis. Would we need to scale those for 2.4 also? So instead of having dT of 185°C, we would have 444°C. This seems a bit of an overstretches to me...

 

[TGS4]

You should factor the thermal loads as appropriate for the applicable load case combination. The only practical way of factoring the thermal loads is to factor the coefficient of thermal expansion, because you want to not change your temperature and the temperature-dependant material properties.