Issue 35

Chahardehi et al, Frattura ed Integrità Strutturale, 35 (2016) 41-49; DOI: 10.3221/IGF-ESIS.35.05 41 Focussed on Crack Paths Fatigue crack growth under remote and local compression – a state-of-the-art review A. Chahardehi (http://orcid.org/0000-0002-7823-7484) Atkins Energy, Euston Tower, 286 Euston Road, London NW1 3AT, UK amir.chahardehi@atkinsglobal.com A. Mehmanparast (http://orcid.org/0000-0002-7099-7956) Cranfield University, UK a.mehmanparast@cranfield.ac.uk A BSTRACT . There is an ever increasing need for accurate understanding of the fatigue crack growth behaviour in major engineering materials and components. With the move towards more complex, probabilistic assessments, the traditional ‘safe’ or conservative approach for prediction of fatigue crack growth rate may no longer be attractive. Current codes and standards tend to be ambiguous about the treatment of compressive stress cycles: on the one hand code guidance on fatigue crack initiation may be non-conservative, while assessment of crack propagation may be inconsistently conservative. Where codes are non-conservative they could lead to dangerous assessments. The current paper provides a critical review of state-of-the-art in literature and a study of current code implications. K EYWORDS . Fatigue crack growth; Compressive stress; Closure; R-ratio; Paris law; Design codes. I NTRODUCTION ost common engineering predictions of fatigue crack growth (FCG) rate demand a high level of accuracy. The need for accuracy may be perceived to be of two apparently distinct natures: on the one hand the scientist and the academic is interested in deeper understanding of the fundamental behaviour of materials in order to progress what may be termed the ‘frontier of science’, while on the other hand the engineer is interested to utilise this, or any usable understanding in the most effective way in practical applications. In offshore applications such as for Oil & Gas platforms, where cost saving considerations merit it, the move towards risk-based inspections and complex probabilistic assessments relies fundamentally on accurate knowledge of the fatigue crack growth behaviour. Nevertheless deterministic predictions of fatigue crack growth for safety applications have traditionally relied on inherent conservatisms in the Paris law coefficients used in the ‘design’ fatigue crack growth laws (fracture mechanics-based fatigue assessment), often taking the prediction curve two standard deviations away from the mean of data. M