T_K
Mechanical
- Aug 31, 2017
- 5
I was reviewing what would be involved in qualifying a set of nuclear class 1 bolts for fatigue by using experimental stress analysis in accordance with ASME BPVC Section III Appendix II Subsection II-1500. I follow the basic method for the testing, but I'm a bit at a loss with calculating Ks and by extension Kn. This defines the level of loading and number of cycles required, so Ks/Kn are fairly important.
Ks is made up of 5 factors (Ksc, Ksf, Ksl, Kss, Kst). Ksc and Kst are the only ones I'm stuck on.
Ksc is a "factor for the differences in design fatigue curves at various temperatures"
Kst is a "factor for the effect of test temperature"
Kst has an equation (SaN at test temp / SaN at design temp), where SaN equals Sa from the applicable fatigue curve at N cycles. For the sake of this equation I'm assuming N is the number of specified service cycles, not the chosen number of test cycles, though the Appendix is silent on the definition of N (though it does define Nc, Nd, Nt and Ntmin).
My current theory is that I calculate Kst using the appropriate curve (I-9.4 for bolting) by looking up Sa at N on the curve and then apply the modulus of elasticity ratio using the E value at test temperature and at design temperature to get a different Sa value. This seems like the most logical way of accounting for test vs design temperature and seems in line with my understanding of how the fatigue curves were created using strain data then multiplied by "E" to get the nominal stress curves at room temperature. If this is correct, I don't know why they wouldn't make this more clear which they easily could.
As for calculating Ksc I really don't know where to go. The definition makes it seem related to temperature effects, but that seems like a double count with Kst. There's no equation given, only the definition. Section II-1520(g) also states that there's 5 factors, one each for size, surface finish, cyclic rate, temperature, and number of tests. All of the 5 are accounted for except "cyclic rate" and that doesn't seem related to the definition of Kst or Ksc, though from nomenclature I would have assumed it would be Ksc. My only path forward right now it to take the definition at face value and double count the temperature factor by including it twice, which seems definitely wrong.
Has anyone been through II-1500 testing before? How did you calculate all the factors that make up Ks (particularly Kst and Ksc).
Ks is made up of 5 factors (Ksc, Ksf, Ksl, Kss, Kst). Ksc and Kst are the only ones I'm stuck on.
Ksc is a "factor for the differences in design fatigue curves at various temperatures"
Kst is a "factor for the effect of test temperature"
Kst has an equation (SaN at test temp / SaN at design temp), where SaN equals Sa from the applicable fatigue curve at N cycles. For the sake of this equation I'm assuming N is the number of specified service cycles, not the chosen number of test cycles, though the Appendix is silent on the definition of N (though it does define Nc, Nd, Nt and Ntmin).
My current theory is that I calculate Kst using the appropriate curve (I-9.4 for bolting) by looking up Sa at N on the curve and then apply the modulus of elasticity ratio using the E value at test temperature and at design temperature to get a different Sa value. This seems like the most logical way of accounting for test vs design temperature and seems in line with my understanding of how the fatigue curves were created using strain data then multiplied by "E" to get the nominal stress curves at room temperature. If this is correct, I don't know why they wouldn't make this more clear which they easily could.
As for calculating Ksc I really don't know where to go. The definition makes it seem related to temperature effects, but that seems like a double count with Kst. There's no equation given, only the definition. Section II-1520(g) also states that there's 5 factors, one each for size, surface finish, cyclic rate, temperature, and number of tests. All of the 5 are accounted for except "cyclic rate" and that doesn't seem related to the definition of Kst or Ksc, though from nomenclature I would have assumed it would be Ksc. My only path forward right now it to take the definition at face value and double count the temperature factor by including it twice, which seems definitely wrong.
Has anyone been through II-1500 testing before? How did you calculate all the factors that make up Ks (particularly Kst and Ksc).