Srs.,

Let's say I have the following combination:

L1: W+P
L2: W+P+U1 (U1 is an acceleration - uniform load)
L3: L2-L1 (Stress variation caused by U1)

Occasional stress as B31.3 was already checked and not shown above.

This uniform load is expected to occur 65.000 cycles but causes a small variation in stress, no more than 10 to 20MPa maximum. Based on ASME B31.3 rules and allowables, is it possible to show that this variation represents infinite life and need not be considered in fatigue calculation (miner's summation considering other variable loads)?

Regards.

If you are going to do this "by the book" I guess it depends on the greatest displacement stress range (SE).
Your value for ri here will be Si/SE or 20MPa/SE.
Of course, that ratio is taken to the fifth power to convert your 65k cycles into equivalent cycles for your SE evaluation. If SE is a low 40MPa, then ri is 0.5 but (ri)^5, then, is only 0.03. So this adds only 2032 cycles to your count for SE.
Not significant.

Dave,

I'm applying the eq. 1d of B31.3 just for secondary loads to analyze fatigue.

Regarding to mechanical loads that varies over the piping lifetime, since the cycles stresses variations are all below the material endurance limit they are been ignored in fatigue evaluation.

For example, according to table 3-F.1 of ASME VIII Div.2 the endurance limit for carbon steel is 7kPSI (48MPa) for a polished bar and considering that the stress calculated by B31 or Caesar code formula is only one-half of the theoretical stress and that de ASME fatigue curve is amplitude, i'm comparing the stress variation directly with this 48MPa. Is it correct?

The idea of the endurance limit is not to waste time couting the cycles that are negligible for the piping fatigue life.

Regards.

B31.3 is working on a new appendix to assess high cycle fatigue. So what is their take on "high cycle fatigue"? It would apply to systems that see at least 100,000 cycles AND have a stress RANGE of at least 20 MPa.
So, that work is similar to what you suggest - that there is a certain stress range magnitude to make this significant - but you have a different magnitude.
I don't think you will see the term endurance limit in B31.3 for these range calculations. Footnote 4 in 302.3.5(d) gives the minimum value for f as 0.15 for "an indefinitely large number of cycles".

Ok thank you.

If you take the allowable as 1,25f(Sc + Sh) from Appendix P and considering Sc=Sh=2/3Sy (low temperature):

Sa = 1,25 x 0,15 x 4/3y = 0,25Sy

Any displacement stress range below min[endurance limit ASME VIII ; 0,25Sy] represents infinite life or indefinitely large number of cycles.