ASME BPVC Section VIII Div 2 - 2004 & 2010 Comparison - Design by Fatigue Analysis 1

[VLks]

I have read the ASME Section VIII Division 2, Part 5 Design by Analysis Requirement, 5.5 Protection Againts Failure from Cyclic Loading and ANNEX 3.F Design Fatigue Curve.

In ANNEX 3.F, there is table 3.F.1 Coefficient for fatigue curve 110.1 - Carbon, Low Alloy, Series 4XX, High
Alloy Steels, and High Tensile Strength Steel for temperature not exceeding 700 degree Fahrenheit, with ultimate tensile strength lower than 80 ksi. Is it right to assume these materials can sustain very high stress up to several times its ultimate tensile strength?
the interpolation of fatigue curve 110.1, the ASME Section VIII Division 2 2010 ed. ANNEX 3.F using equation
N =10^X

where

X = (C1 + C3.Y + C5.Y^2 + C7.Y^3 + C9.Y^4 + C11.Y^5) / (1 + C2.Y + C4.Y^2 + C6.Y^3 + C8.Y^4 + C10.Y^5)

Y = (Sa / Cus)(Efc / Et)

while the ASME Section VIII Division 2 2004 ed. Mandatory Appendices, Appendix 5, Design Based on Fatigue Analysis Table 5.110.1 using this equation for interpolation of N

N / Ni = (Nj /Ni)^[log(Si/S)]/[log(Si/Sj)]

My question is, which one of those equation gives more accurate interpolation of N in fatigue curve 110.1?
Because when I did interpolation for 12,500 psi, the ASME Section VIII Division 2 2004 gives 1,000,000 cycles and the ASME Section VIII Division 2 2010 gives around 2,000,000 cycles.

Really appreciate your help, thanks in advance,
Regards.

 

[TGS4]

Based on the 2006 Addenda that I am looking at - eye-balling 12,500psi of Sa puts it definitively at 1e6 cycles from Fig 5-110.1.1 - so we agree here. When I use the formula from Annex 3.F and the data in Table 3-F.1 from the 2011 Addenda, for an Sa value of 12.5 ksi, I calculate an N value of 994,880. That's close enough to 1e6 for me. Not sure where you went wrong, but the error appears to be yours.

Is it right to assume these materials can sustain very high stress up to several times its ultimate tensile strength?

In the sense that the local peak stress RANGE (and hence the peak stress amplitude, which is algebraically half of the range) (P+Q+F) can be higher than Suts, you are correct. However, that is not true for general membrane or membrane-plus-bending. Does this answer your question?

 

[VLks]

TGS4, Thank you very much for the reply.
Really appreciate it