AS 2885.1 generally provides all the background information you need to develop a simple fracture control plan (at least for simple cases). But one aspect that has not been well covered is guidance on fracture initiation control. It isn’t the most critical aspect of a fracture control plan but is mentioned in a few places, of which the following are the most significant:
- 4.8.2(b)(ii) – the pipe longitudinal weld seam and other pipeline components must have adequate toughness to minimise likelihood of fracture initiation (with a note that the toughness to arrest a propagating fracture is higher than required to avoid initiation, so for the pipe body, as distinct from the seam weld, propagation control ensures imitation control).
- 188.8.131.52 – reiterates need for fracture initiation control in the seam weld and says it shall be determined by “a recognised method”, which is not very helpful to the uninitiated.
The method usually used to estimate the toughness required to control fracture initiation is set out in Section 2.3 (and particularly 184.108.40.206) of Fracture Control Technology for Natural Gas Pipelines (c. 2001), PRCI catalog number L51846e (referenced in an earlier post). But sometimes it has been misunderstood so it seems worth presenting it here.
The key is that as Charpy V-Notch toughness (CVN) increases so does the through-wall critical defect length (CDL), up to a point. For high CVN values the CDL asymptotically approaches a plateau. What we want to achieve is a high CDL (ie. high tolerance to serious mechanical damage) without specifying an unreasonably high Charpy value. A CDL 80% of the maximum (plateau) value has been deemed high enough to give adequate fracture initiation control. So all that is necessary to is select the CVN corresponding to the CDL that is 80% of the maximum. Better illustrated graphically:
In practice the shape of the curve means that the toughness required to control fracture initiation is usually quite low, as in the example here and as noted by AS 2885 paraphrased above.
The forthcoming revision of AS 2885.1 will at least mention this approach, but the mention is necessarily brief so even when it is published I hope this expanded explanation is useful.
For the record, the example curve above is for a DN 400 pipeline, MAOP 10.2 MPa, X70 steel, design factor 0.72 and hence 6.0 mm wall thickness, and the Charpy values are full-size equivalent. Critical defects lengths were calculated using the equations in AS 2885.1 Clause 4.8.5.