The Statistical Nature of Fatigue Crack Propagation

Abstract

A statistical investigation of the fatigue crack propagation process was conducted. Sixty-eight replicate constant amplitude crack propagation tests were conducted on 2024-T3 aluminum alloy. From the experimental data, the distribution of N was found to be represented best by the 3-parameter log-normal distribution. Six growth rate calculation methods were investigated and a modified secant method was found to introduce the least amount of error into the growth rate data. No one single distribution was found to describe the distribution of da/dN. A prediction scheme using the distribution parameters of da/dN was partially successful.

Delamination growth in polymer-matrix fibre composites and the use of fracture mechanics data for material characterisation and life prediction

“…Whereas the paper by Virkler [6] is widely acknowledged as the seminal study on the variability (i.e. scatter) in FCG rates in metals, for composites that honour goes to Murri [4].…”

“…Indeed, one potential problem with merely certifying via a single FSFT is the large scatter that is often seen in delamination growth [2], even for relatively simple Mode I (opening-tensile) loaded fatigue tests on polymeric-matrix fibre composites [4,5]. For metals the seminal work on the variability in measured fatigue crack growth (FCG) rates is the paper by Virkler et al [6]. The seminal work on the variability observed in FCG rates, via delamination growth, in composites is [4].…”

Estimating the statistical properties of crack growth for small cracks

“…Figure 4 also provides a comparison between several studies including the Artley, et al study, a study by Virkler, et al 4 , the fastener hole quality study, and a study by Shaw 3 and LeMay3.…”

“…The materials employed in these studies were 7075-T6 Aluminum (Artley, et al 2) 2024-T3 Aluminum (Virkler, et al 4 …”

Variability of Fatigue Crack Initiation Life in Flux Cored Arc Welded API 2W Gr.50 Steel Joints

“…Although the fatigue crack growth rate data of welded structures should be accurately determined and made available, very few data are available for welded structures from the stress range viewpoint. However, it is widely recognized that fatigue crack initiation and propagation are fundamentally probabilistic phenomena [5][6][7][8][9][10]. Risk assessment and remaining fatigue crack growth life prediction require accurate probabilistic fatigue crack growth models.…”