is such simplification method on fatigue evaluation acceptable?

[YuJie_PV]

hi experts,

​i am checking a FEA REPORT of a buffer tank designed to asme viiii-1. in the meantime, the buffer tank is subjected to cyclic loading, thus VIII-2 part 5 is referred to evaluate the risk of fatigue failure.
and now i have following inquiries:

1, is it mandatory to check other failures mechanism, such as plastic collapse (part 5.2) , local failure (part 5.3)and ratcheting (part 5.5.6 ), for the tank as specified in VIII-2 part 5, while the vessel has been performed with the calculation per VIII-1 for shell, heads, nozzles, etc? and why?


2, the operating pressure fluctuates from 0.6~1.2MPa. the author of FEA report apply a pressure of 0.6(1.2-0.6=0.6)MPa to the model to get the equivalent stress range, based on which fatigue is evaluated per VIII-2. is such a simplification acceptable? i know it violates the equation in part 5..5.3 of ASME VIII-2 related to the calculation of effective alternating equivalent stress.
i just wonder if such simplification may actually get an approximate result not deviating too much from the results when fully following equation of 5.5.3 (stress under 1.2Mpa-stress under 0.6MPa).
have you ever encountered such practice before?

thanks for any insight on the issues.

 

[YuJie_PV]

hi all,
i just turned to my colleagues with experiences of FEA for decades. HE replies as follows:

1. YES, it's mandatory. the requirement of DBA in ASME VIII-2 Part5 is a whole, and could not be applied partially.
2. such simplification method (using pressure range to get the stress range) is quite commonly exercised in the field of vessels FEA.

intuitively, i just quite doubt his statement, but i have no proof to reject him.
i've seached the topic in the forum, but found no similar inquires.
i know there are many experts extremely experienced in the field, i'd like to hear your insight. thanks

 

[TGS4]

1. No. Protection against plastic collapse, local failure, and buckling is achieved by the Design-By-Rules of VIII-1. Going to Part 5 for a fatigue analysis only is permissible. Protection Against Ratcheting might need to be demonstrated to validate the fatigue analysis.

2. Maybe - provided that in the 0.6MPa-1.2MPa cycle you don't have stress reversals that might not occur in the 0-0.6MPa cycle. However, for fatigue following the Structural Stress Method, such a simplification would not include the mean stress effects, and therefore would be invalid.

 

[YuJie_PV]

Thanks so much, TGS4, so glad to see your comments.
1. regarding the first issue, i just quoted your comments to argue against my colleague, but he dismissed me with 46-4(c) in appendix 46 of VIII-1, which states:"(c) All of the failure modes listed in Division 2, Part 5 shall be considered." i am frustrated again.

2. the second issue;
a) i have nozzle loading as well, which are inputted as being constant during pressure cycle, which i believe would divert the stress range from being proportional to the pressure range.
b) i'd like expand a little, if i may, @TGS4, you mentioned stress reversal, does that mean if vacuum condition exist during pressure cycle, say -0.1MPa~1.2MPA, such simplification (use 1.3 MPa to get the stress range) will not be acceptable?

appreciate any response, thanks .

 

[IdanPV]

I am not and expert as TSG4 is, but here is my 2 cents regarding this issue:
1. As far as I understand it, 46-4(c) is only applicable when using Div. 2 to establish the thickness and other design details of a component for ASME VIII-1 pressure vessel. If you use Div.2 only for Fatigue Assessment, you don't have to demonstrate protection against other failure modes using Part 5.

2. In my FEAs, I sometimes used the pressure simplification of "internal pressure+FV" and sometimes I use internal pressure only.
And it depends on whether the part is affected by a FV condition or not.
Note that if you chose to neglect the FV condition you shall justify it.

 

[YuJie_PV]

@IDANPV, thank you. I've learned so much from your posts before. You, along with TGS4, JTE, and PREX, are among those I admire the most. I hope my knowledge will be sufficient to contribute to the forum someday in the future.

Just one last question: Can I stand on solid ground to reject such a simplification method directly? and simply ask the author of FEA to precisely follow 5.5.3 of ASME VIII-2.

Thanks.

 

[IdanPV]

Why you want to reject such simplification?
I think that TSG4 has answered your question and you should use your engineering judgement and experience to decide.

From my experience, both are expectable but it depends on the situation.

 

[YuJie_PV]

Thanks IdanPV

 

[TGS4]

You have received good advice from IdanPV. I will note for you that the rules of 46-4 talk about the failure modes needing to be “considered”, not necessarily “analyzed” or “performed”. For non-english speakers that subtlety can be missed. Essentially that means that some engineering judgement needs to be applied. The word choice here was very deliberate on the part of the Code Committee. But as IdanPV noted, that is only for the purposes of setting the thickness.

Again, some simplifications may be appropriate. Except in situations where a mean stress is needed, such as the Structural Stress Method.

 

[YuJie_PV]

@TGS4, Thanks for clarification, i think i've learned your meaning.
except for the Structural Stress Method you are referring to, which i am not familiar with.
i'll go through the related paragraphs in Part 5, i think i could figure it out myself, thanks again.