Comparing Appendix 13 results with Finite Analysis Results(ANSYS) 1

[LY.Yan]

Good Day to all,
I am trying to understand my vendor's ANSYS report by comparing the values for Membrane, Bending and total stresses between Appendix 13 and ANSYS. The Finite element was build with no tube ligaments and only internal pressure as loading condition. However the values for membrane stress are way off. I would like some advice if the linearized line is correct or what might most likely be reason for large high membrane stress in FEA. The total stress does not seem too far off with appendix 13. Thank you for your time and advice.

L.Y Yan

 

[IdanPV]

Hello,

First, take a look at clause 5.2.1.1 of ASME VIII-2:

For components with a complex geometry and/or complex loading, the categorization of stresses requires significant knowledge and judgment. This is especially true for three-dimensional stress fields. Application of the limitload or elastic–plastic analysis methods in 5.2.3 and 5.2.4, respectively, is recommended for cases where the categorization process may produce ambiguous results.

In case of an rectangular vessel, my suggestion is to try an elastic-plastic FEA in accordance with 5.2.4.

Regarding your question, it seems that the linearization method which your vendor chosen is not correct. Make sure he works with ASME VIII-2, Part 5 Annex 5-A.

 

[DriveMeNuts]

Yes, your linearisation line is most certainly off. It should pass through the thickness of the material. Try first passing it through the midspan of the flat side where there will be no peak stresses. You will be able to then split membrane and bending. Then do it at the corner where there will be added peak stresses.

The membrane stress can be calculated using a hand calculation, avoiding the use of FEM. Your FEM will produce an identical result to a hand calc. This is a statically indeterminate structure, so your hand calc will require some tricky maths.

This is not a complex geometry so elastic analysis is appropriate (App 13 is an elastic hand calc). Using elastic plastic is pointless for the purpose of comparing FEM to App 13, as App 13 uses elastic Membrane-Bending theory. Comparing elastic-plastic results to App 13 is like comparing a donkey to a nuclear warhead.

 

[LY.Yan]

Good Evening,
Thank you for the replies. The latest values i received was really promising with the advice here. I do have a larger discrepancy at the corners comparing it with the mid of the plates. Can i gather that might be due to proximity of the edges. Thank you again.

L.Y Yan

 

[DriveMeNuts]

The SCL is correct.
You have used linearised Stress intensity. You should be plotting 'linearised normal stress' in the 'Y' axis.
The membrane stress should be almost exactly the same as App 13.

 

[IdanPV]

What I meant to say is that if one is using Elastic-Plastic analysis he doesn't need to mess with linearization and categorization of stress.
And the post processioning of the results is more simple.

 

[mechengineer]

I think that FEA or hand calculate (App 13) is to obtain the actual stresses in the vessel and it may be comparable. Elastic analys and elastic plastic analysis are used to determine the 'allowable' stress that nothing related to what is the actual stress in the rectangular vessel.