The Relevance of Interaction Rules in Evaluating the Growth of Multiple Fatigue Cracks

Abstract

ABSTRACT Flaws introduced during fabrication or developing in service rarely occur singly. Contained in the popular codes used to assess the integrity of structures are rules defining how arrays of multiple flaws should be treated. The current US and UK codes adopt the same philosophy, - simplification of the flaw array into a single larger flaw of assessable shape. Recent experience in using these rules has suggested that in many cases, they may result in assessments that are excessively conservative, possibly resulting in premature rejection or repair of plant. Although a number of solutions for Fracture Mechanics parameters relevant to interacting flaws have been derived, little has been done to investigate the mechanisms of crack interaction. This paper describes a series of fatigue tests performed to investigate by direct observation how crack growth rate and shape development are affected during interaction. The results are encouraging in that only weak effects have been observed with no indication of grossly increased crack growth rates. A methodology for assessing simple flaw arrays is proposed which assumes individual growth until crack tip contact or overlap occurs, at which point crack shape is adjusted to a smooth semi ellipse and crack growth evaluation continued. INTRODUCTION Manufacturing or service-induced flaws rarely occur in isolation and are often clustered together in critical areas of a structure or plant. A common problem therefore in assessing plant integrity or remanent life is the determination of the behavior of groups of flaws and their influences on one another. In structures subject to fatigue loading, or where failure by fracture is a possibility, Fracture Mechanics methods are increasingly used to calculate the growth rate, shape development and stability of crack-like flaws. It is the influence of the proximity of another flaw on the effective value of the stress intensity factor operating at a crack which will modify the way in which it grows. The problem of flaw interaction is dealt with specifically by two of the currently used assessment codes which give guidance on the use of Fracture Mechanics, - ASME section XI Division 1:1986 (1) and BSI PD 6493:1980 (2). In these documents, multiple flaw arrays meeting specified criteria are "recharacterised" into single flaws of simple, easily handled shapes for which standard solutions exist. The criteria for interaction are nominally and loosely based on a particular increase in stress intensity factor (typically 20%) in the ligament between flaws, although the solutions used to calculate this are not given or referenced in the documents. FLAW RECHARACTERISATION. The rules governing the amalgamation of flaws are broadly similar in both documents, the rules in ASME XI generally being based on flaw depth or height and those in BSI PO 6493:1980'on flaw length. Recharacterisations are defined for coplanar and non co-planar surface breaking and embedded flaws. Each of the codes contains cases not addressed by the other.