API 530 or ASME 2 D to determin fatigue life at high temperatures 1

[McT178]

When designing for fatigue life of a pressure vessel at high temperatures, should the allowable be taken from API 530 or from ASME section 2 part D? I have seen this done both ways. In some situations I have seen Larson-Miller calculations that us API 530 to determine a temperature, and then allowable for that temperature is then obtained from ASME sec2 part D. I see that ASME is usually a lower allowable, so maybe this is just a conservative approach. Also, are the allowable stresses in API and ASME for pressure vessels only? For instance would a support system inside a pressure vessel also be designed to the the allowables (Table A-1) or would it be more appropriate to design the support structure using the yield stresses in ASME Table Y-1?

 

[TGS4]

First question: what is "high temperature" for you? Specifically, what material and what temperature?

 

[McT178]

Well we can be slightly off the charts and extrapolate, but stainless steel (304H) at 1300 to 1500 deg F would be common.

 

[TGS4]

You are well into the creep regime for 304H, which makes this a creep-fatigue problem. There are no rules in ASME Section VIII, Division 2, Part 5 for this type of evaluation. API 530 is for creep only - not fatigue. Same with the values in ASME Section II, Part D.

You will need to ensure that you don't have creep-ratcheting, first, before you worry about fatigue.

Fatigue is a totally different beast from creep, and creep-fatigue is an order of magnitude more difficult than either one on its own.

 

[LSThill]

TGS4 (Mechanical)
Thank you
steve

 

[McT178]

TGS4 - The API 530 allowables I am asking about are from the Larson-Miller S/n charts (Figure F-12 for example). I believe the stresses are to rupture. My understanding of this was that API 530 values are rupture stresses and ASME II D values are 1% creep in 100,000 hours.

 

[TGS4]

McT178 - I suspect that you are right regarding the basis of the allowable stresses. it still doesn't change that you were asking about creep-fatigue interaction failure and the allowables are related to creep failure.

The Figure that you refer to (F.12) is a stress-temperature-time graph. Unless I missed something, there is nothing related to number of cycles.

Do you understand the difference between creep (time-dependent) and fatigue (cycle-dependent) and ratcheting (a cycle-dependent failure mechanism)?

 

[McT178]

You are right, I incorrectly asked about fatigue life when I meant creep life. I guess I inadvertently jumped ahead because the allowable stress obtained is eventually used in a ratcheting analysis.

I think I need to explain further. The system is designed for 3 different temperatures and a certain amount of time is allowed for each temperature. Solving all 3 equations for 100% life will provide a sort of average design temperature. This temperature is what is then used to determine the allowable. It is at this part that the allowable stress is taken from ASME and not from API. I guess my real question is towards the difference between API and ASME. For some materials the values are basically equal (chrome-moly) and for others (304H), API is significantly higher. Is this due to the brittleness of the material at the higher temperatures? Does 1% creep and 100,000 hours for chrome-moly equal failure because the material is brittle.?