I have discovered two errors in the fatigue calculations for PD5500 vessels.

Both involve the value of e in equation C-8:

1)

N = A (22/e)^m/4 (Sr x 2.09 x 10^5 / E)^-m (see table C.1 for m & A)

Reading paragraph C.3.2.3 it is clear that e should be the Maximum of 22 or the section thickness, where as PV ELite is using the corred nominal thickness, this is also indicate in the examples found in ANNEX W.

2)

e is defined in C.1.3 as the nominal thickness of section being considered including corrosion allowance. PV Elite is taking e as the corroded nominal thickness, which I beleive to be incorrect. See para 1.6.

Both of these errors are non-conservative, the first significantly increasing the calculated value of N for thinner walled vessels.

Jez