Analysis of Multiple Co-Planar Surface Breaking Flaws Loaded in Tension Under Cleavage Failure Conditions

In procedures such as R6 and BS 7910 for assessing the structural integrity of structures, complex multiple flaws located in close proximity to each other are generally characterised as one, larger, single flaw. Recent studies have shown that the current characterisation rules may be non-conservative under some circumstances. Concurrent experimental and analytical programmes are being undertaken in order to further investigate this potential non-conservatism for situations where the possibility of cleavage failure has to be taken into account when assessing structures or components containing multiple flaws. The analytical work has involved inelastic finite element modelling in conjunction with numerical analysis based on the weakest link methodology for cleavage initiation to determine the probability of cleavage failure. This has allowed the probability of failure for the situation of multiple adjacent flaws to be compared with that for the single characterised flaw to determine whether the characterisation rules are conservative. Initial results from the cases studied so far indicate that: • For flaws that do not touch, the probability of cleavage failure for the single characterised flaw is higher than the original flaws, confirming that the characterisation process is conservative in this case (perhaps significantly so). • For low aspect ratio (depth/length ≪ 1.0) flaws in contact, the probability of cleavage failure for the single characterised is higher than the original flaws, confirming that the characterisation process is conservative in this case. • However, for high aspect ratio (depth/length ∼ 1.0) flaws in contact, the probability of cleavage failure for the single characterised flaw is lower than the original flaws, indicating that the characterisation process may be non-conservative in this case. This paper covers the initial stages of a much larger programme. The analyses presented in this paper are in the process of being extended to allow for crack-tip constraint effects. Early results from these further analyses indicate that the loss of constraint between interacting flaws may be a significant factor and could eliminate at least some of the perceived non-conservatism for the high aspect ratio flaw. Further work is also underway to examine the effect of any localised coalescence of defects (a re-entrant region); the current analyses consider defects that are only just in contact. Further papers are therefore expected.