3D fatigue crack growth in mixed mode II + III

VERONIQUE DOQUET; GRACIELA BERTOLINO; ELIAS MERHY

Otawa

Conferencia; 12th INTERNATIONAL CONFERENCE on FRACTURE; 2009

ICF

using different setups. However, crack growth has to be envisaged in 3D, as in real engineering problems. Due to fatigue of high-speed train wheels, elliptical cracks initiated on subsurface defects grow in mode II+III with a variable KII/KIII along their front. Mixed-mode threshold and kinetics are needed to predict their evolution. But to the authors best knowledge, only Hellier et al [1-2] Murakami et al [2] and Pokluda et al [3] tried to obtain such mixed-mode data or at least mode II and III data on the same material, for comparison.Pokluda et al [3] concluded that mode III crack growth is slower than mode II when ΔKII=ΔKIII. However, it is not sure that their experimental device generates equal ΔKII and ΔKIII at the points where data were collected. Conversely, Murakami et al. [2] concluded that the mechanisms and thresholds for mode II and mode III were essentially the same.Hellier and Corderoy [1] developed an asymmetrical 4 points bending device in which a specimen with an edge notch can be turned around a horizontal axis so as to vary continuously KII/KIII from zero to infinity, at least in principle, since 3D finite element computations reported below show that pure mode II or III cannot be obtained, except at mid-thickness. This device is however interesting, provided the gradient in mode mixity along the crack front is taken into account in the analysis. A similar device was thus built and used to investigate proportional mixed-mode II +III crack growth in a titanium alloy.