Flaw growth monitoring as an aid to lifetime prediction : Silk, M.G.; Whapham, A.D.; Hobbs, C.P. International Journal of Materials & Product Technology, Vol. 4, No. 3, pp. 215–231 (1989)

Abstract

The error in the determination of flaw size using accurate non–destructive testing (NDT) techniques is no longer likely to be the limiting factor in the classification of flaws. Constructors of new plant still call for increasing precision in nondestructive flaw sizing techniques and can justify this on commercial grounds. However, this trend towards the detection and removal of ever–smaller flaws diverges from the needs of the operators of existing plant. Here the application of the same concepts would result in severe economic penalties in terms of repair and shutdown costs, particularly because the majority of such repairs are going to be unnecessary. Other things being equal, the pressure to apply more stringent inspection standards will continue. Inspection standards used for new plant, or for particularly stringent inspection tasks such as NRPVs, will tend to be transferred into the more general inspection field. As the inspection of any large engineering structure is likely to result in the identification of a flaw, this can cause problems in continuing operation unless a reliable alternative to repair or removal is available. An alternative option is to monitor the behaviour of the flaw over a period and to place direct limits on flaw growth. This approach has had little success in the past because the precision of the NDT techniques was so poor that flaws could be close to failure before growth was established. More recently attempts have been made to 'fingerprint', but unless this is carried out with the newer, more accurate techniques, it is difficult to see how changes in the fingerprint can be interpreted with safety. Now, however, both the time of flight diffraction (TOFD) technique and the alternating current potential drop (ACPD) technique might achieve precisions of 0.25 mm or better in estimating flaw growth. The monitoring approach can then be shown to have economic benefits without affecting the essential structural safety argument for the plant. This paper reviews the justification for monitoring and presents experimental flaw–monitoring data.