Issue 35

G. Kullmer et alii, Frattura ed Integrità Strutturale, 35 (2016) 368-378; DOI: 10.3221/IGF-ESIS.35.42 372 hexahedral mesh containing the crack front. Thereby for every node at the crack front, first the energy release rates G I and G II for the respective crack opening modes are calculated. To evaluate the stress intensity factors K I and K II the mean values of the energy release rates G I and G II over the thickness of the specimen and the equations valid for the state of plane stress EG K I I  (1) EG K II II  (2) are used. Since plane mixed mode loading is existent, the crack deflection angles are calculated with the maximum tangential stress criterion after Erdogan and Sih [4]. V ERIFICATION OF THE CRACK GROWTH SIMULATIONS efore the execution of the actual numerical simulations of CT-specimens with different changes in stiffness, the simulation model and the modified version of ADAPCRACK3D are verified. Therefore, simulations with the intended different values of Young´s modulus, the different thicknesses and a constant orientation angle α = 90° are conducted. Due to symmetry mode II should be zero and the crack should grow straight through the changes in stiffness independent of Young´s modulus or the thickness of the change in stiffness. The numerically determined resulting crack paths for all cases are almost straight with slight deviations from zero in y-direction. The extent of the deviations is for the CT-specimen with a homogeneous partition P3 with default Young´s modulus and default thickness as big as for the CT-specimens with different changes in stiffness. Already numerical inaccuracies due to an asymmetric FE-mesh, typical for using the free mesh option, cause slight deviations from a straight crack path. The maximum deviation from the straight line is in all cases for a simulated additional crack length of more than 20mm less than 0.2mm. Thus, the simulation model and the modified version of ADAPCRACK3D are suitable for the intended investigation. Moreover, crack growth simulations with the CT-specimen with a change in stiffness with Young´s modulus double the default Young´s modulus and an orientation angle α = 45° are executed with crack growth increments of 0.2mm, 0.5mm und 1mm. Because the resulting crack paths are almost the same, the fixed default crack increment of 0.5mm for the further simulations is chosen, since this uniform crack increment has turned out to be a reasonable compromise between effort and accuracy. C RACK PATHS FOR DIFFERENT ORIENTATIONS OF AN INCLUSION he simulated crack paths for different orientations angles of a change in stiffness modelled as an inclusion and the alignment of the inclusion for the extreme orientation angles α = 30° as well as α = 90° are presented in Fig. 7. For the stiff inclusion, Young´s modulus is double the default Young´s modulus and for the compliant inclusion, Young´s modulus is half the default Young´s modulus. Figure 7 : Simulated crack paths with variable orientation of the inclusion, a) stiff inclusion, b) compliant inclusion. B T b) a)