Issue 41

D. Nowell et alii, Frattura ed Integrità Strutturale, 41 (2017) 197-202; DOI: 10.3221/IGF-ESIS.41.27 198 Figure 1 : Typical image of a fatigue crack, during in-situ loading in the SEM Our work focuses on measuring the variation of crack opening with distance along the line of the crack, since this quantity is straightforward to measure. Indeed, if one employs a direct measure of displacement, such as digital image correlation (DIC), then the crack flanks are the largest displacements available, once rigid body motions have been excluded. In analyzing our work, we started off by using simple elastic models [1], and showing that the technique can be used to calculate stress intensity factors by (for example), plotting the variation of crack opening against  r, where r is the distance from the crack tip. If the displacements vary as  r, then it is clear that the strains (and therefore stresses) vary as 1/  r, and hence the K-value may readily be extracted. Later, [2] our work extended to include a simple elasto-plastic model due to Pommier and Hammam [4]. This was shown to give a slightly better fit to the experimental results, but the absolute value of the plasticity parameter  (effectively the crack tip opening displacement) was found to be sensitive to the choice of crack tip position. However, it is clear from both approaches that there a considerable amount of crack closure present, and this may be considered as a residual (negative) stress intensity factor, which may be summed with the applied K to give the actual stress intensity experienced by the crack. An example of this is shown in Fig. 2, which is data obtained for a fatigue crack growing under cyclic loading and observed in a scanning electron microscope. Figure 2 : Variation of measured and calculated stress intensity factor with load, for a typical DIC experiment [5] In Fig.2., the residual K is apparent as an (almost constant) offset between the theoretical and experimental lines, once the crack is open.