is it allowed to avoid fatigue analysis by raising design pressure?

[YuJie_PV]

hi, all,

my colleague come to me with a question, that he have a tubular exchanger with the channel side subjected with cyclic loading. the operating pressure ranges from 0 to 0.3MPa, and the design pressure is set at 0.6MPa。the operating temperature is not high, around 170C。

according to the screening criteria of ASME VIII-2, it's required to perform fatigue analysis to all the components of tube side.

when he came to me with this issue, i advised him to raise the design pressure of tube side to no less than 1.5MPa (pressure variation not exceeds 20% of design pressure), so that screening criteria of method A in 5.5.2.3 of ASME VIII-2 could be applied to achieve the exemption of fatigue analysis then. BTW，such pressure increase would not add too much cost considering the diameter of the exchanger is so small, around 450mm .

but i quickly realized i may have made a mistake because ASME VIII-2 has following statement in 5.5.2.1(c) ："If the specified number of cycles is greater than (10)^6, then the screening criteria are not applicable and a fatigue analysis is required."

i think my suggestion is reasonable but it seems conflicting with the Code. i am confused， so my question is: if the pressure variation is 0~0.3MPa, and the total number of such cycles is greater than (10)^6, is it allowed by ASME code to raise design pressure enough large to avoid fatigue analysis ?

Thanks in advance.

 

[DriveMeNuts]

I read that as, if cycles exceed 10^6, then don't bother with the effort of screening for fatigue exemptions. A comprehensive fatigue assessment is required.

 

[YuJie_PV]

thanks DrivemMeNuts,

i am aware of the philosophy same as you implies, but i still think there is some ambiguity regarding to subparagraph 5.5.2.1(c) on how to define if a cycle shall be considered in a fatigue analysis. For example, let me hypothetically reset my operating variation to be 0~0.05MPa, and design pressure to be, say 5.0MPa. in such condition, am i still required to perform fatigue analysis as long as the cycle count is more than 10^6?

i actually have read the thread "

https://www.eng-tips.com/viewthread.cfm?qid=431553",

in which the issue was talked about already by these experts, but i am still not convinced.
even the most infinitesimal pressure variation shall be handled with fatigue analysis as long as the count is large enough? if the answer is NO, where does that separating point lie?

thanks, any insight appreciated.

 

[TGS4]

You raise a very interesting point. And this is actually a point currently being discussed at the Code Committee (WG-DBA) level. Effectively asking the question: what constitutes a "significant" cycle for the purposes of the screening assessment. And since the Method B screening has been updated in the 2023 Edition, this brings the question into closer focus.

Unfortunately, there is not any any consensus - yet. But stay tuned.

For the time being, however, the Code words are clear - if your cycle count exceeds 10^6, a fatigue analysis is mandatory. But don't let that frighten you - a fatigue analysis can be performed relatively easily...

 

[XL83NL]

I think the same ‘bypass’ (at least, the question the OP is asking) can be found in EN 13445-3 (looking at the 2021 eft, issue 1), under para 5.4.2, eq. 5.4-2?

The line below the equation outlines, I think, the procedure the OP is looking for;

In the above equation, Pmax is the maximum permissible pressure Pmax calculated for the whole vessel (see 3.16) in the normal operating load case (see 5.3.2.1). For simplification, Pmax may be replaced by the calculation pressure P.

In the EU, design pressure is usually P or PS, and MAWP can be considered Pmax.

Huub
- You never get what you expect, you only get what you inspect.

 

[YuJie_PV]

@TGS4, thanks for clarification, i am fully clear now.
@XL83NL, thanks. it's quite beneficial to make comparison between ASME and EN code. The equation you pasted seems simple.
Per the equation, i assume if fatigue issue due to temperature change exists, no "bypass" is available anymore in EN 13445, is it correct?
Thanks

 

[XL83NL]

Do you have a copy of EN 13445-3? For temperature it says;

Cyclic thermal loads can be neglected if:
- for start-up and shutdown cycles, the number shall not exceed 2 000 and the rate of temperature change at the surface shall be less than 60 °C per hour for ferritic steel sections. The designer can specify a higher rate of surface temperature change based on favourable/good industry experience and practice;
— if the requirements of Annex U are satisfied for operating conditions.

If these conditions on pressure and thermal loads are met, then no fatigue analysis is necessary and the standard requirements of non-destructive testing given in EN 13445-5 shall be applied.

If these conditions cannot be met, then a fatigue assessment is necessary according to either Clause 17 or Clause 18

Para 5.4.3 is for vessels under testing group 1, 2 and 3. For group 4 it has less stringent requirements.

Huub
- You never get what you expect, you only get what you inspect.

 

[YuJie_PV]

@XL83NL, thanks for explanation. i am trying to go through these paragraphs in EN 13445, thanks.

 

[davefitz]

Don't forget, the thermal stress increases by the square of the wall thickness in the case of fast startup or shutdowns, so increasing the design pressure implies a thicker vessel whihc then implies a higher thermal stress.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[YuJie_PV]

@davefitz, that's an insight, thanks.

 

[XL83NL]

Good point Dave. Hadn’t thought of that.

The cases where we see the methods of EN 13445 as beneficial are typically those where there is excess thickness since for example schedule pipe is used.

Huub
- You never get what you expect, you only get what you inspect.