Effects of short cracks on fatigue life calculations

Application of the EBSD technique to describe the initiation and growth behaviour of microstructurally short fatigue cracks in a duplex steel

“…The fact that microstructurally short cracks can grow far below Δ K th and at higher rates than predicted for a certain value of Δ K according to the long crack theory (as schematically shown in Fig. 1) implies that there is some uncertainty with non‐conservative fatigue‐life predictions (Suresh & Ritchie, 1984; Kaynak et al ., 1996).…”

A damage mechanics based approach to structural deterioration and reliability

“…As the short crack approaches a, in length, its behaviour and growth rate also approach those of long cracks (Kaynak et al, 1996). Once a long crack develops, fatigue damage growth enters the more familiar and widely documented domain of stable crack growth described by fracture mechanics.…”

“…However, no unique definition of the threshold crack length exists in the fracture mechanics community (Kujawski and Ellyin, 1992) and this approach, like its'macroscopic counterpart, is unable to provide any estimate of the residual strength at the end of the crack initiation period. (ii) Attempts also ' (as illustrated in Kaynak et al, 1996). Neither any standard definition nor any reliable method of measurement is known to exist for the initial length a0 which is supposed to be only as large as the grain size ( w 10 -100,um).…”

“…Moreover, short cracks may grow at rates higher than those for long cracks depending on the stress ratio (Kaynak and Ankara, 1992). Ignoring the short crack growth stage, or using long crack growth rate parameters (e.g., m and C in Paris-Erdogan law) for short crack growth "can lead to potential dangerous over-prediction of (fatigue) life" (Kaynak et al, 1996). The threshold crack length, a,, below which LEFM (and consequently the Paris-Erdogan law) is not valid, may be approximately estimated (Dowling, 1993) with the help of two macroscopic quantities S , and AKth (both for R = -1):…”

“…The crack length ranging from grain diameter to about 50-100pm is used, depending on the material and physical scale of interest." However, Martin and Wirsching (1991) used a value of a, = 0.5mm for structural welds and Kaynak et al (1996) chose a value of l m m for En7A steel. Murtaza and Akid (1995), on the other hand, chose a, = 120pm for BS250A53 steel.…”