The practice of characterizing the growth of fatigue cracks on the basis of fracture mechanics primarily relies on laboratory fatigue tests on specimens containing ‘long’ flaws which are typically tens of millimeters in length. There are, however, a number of fatigue-critical engineering components, such as turbine discs and blades, whose design requires an understanding of the propagation of fatigue cracks of significantly smaller dimensions. In the majority of investigations where continuum approaches have been adopted for the characterization of small fatigue flaws (of size range from a fraction of a millimeter to several millimeters), it has been shown that the growth rates of small flaws can be significantly greater than the corresponding rates of long flaws when characterized in terms of the same nominal driving force (see Fig. 15.1). The direct application of laboratory data (derived from experiments on long fatigue cracks) to design against the failure of safety-critical components containing short flaws can, therefore, lead to dangerous overestimates of fatigue lives. Research effort in this area has led to an awareness of the apparently anomalous behavior of short fatigue cracks and has provided possible ways in which many seemingly conflicting viewpoints of total-life and defect-tolerant fatigue approaches can be rationalized in a unified fashion.

To focus attention on the practical significance of the ‘short crack problem’, consider the effect of crack growth characterization on the estimated fatigue life of a commercial alloy, Fig. 15.2.