13th International Conference on Fracture June 16–21, 2013, Beijing, China -2- growth behaviour and their interaction with the applied biaxial loads should be investigated. With this motivation, experimental test was conducted on a cruciform specimen made of aluminium alloy 2198-T8 containing a butt weld joint and subject to biaxial loads [18], which is briefly summarised in Section 2. This paper focuses on the modelling of crack growth behaviour. 2. Summary of fatigue tests Four cruciform specimens were manufactured and tested under biaxial stresses [18]. Geometry and dimensions are shown in Fig. 1. The specimens were made of aluminium-lithium alloy 2198-T8 and joined by the friction stir weld (FSW) process. The panel skin has nominal thickness of 1.6 mm and the weld area has a nominal thickness of 3.1 mm. The top side of the thicker pad-up area measures 50 mm wide and the bottom side (specimen flat surface) is about 62 mm wide, making the average width of the pad-up 56 mm. Specimen COI-BIAX1 and COI-BIAX3 had weld seam orthogonal to the material rolling direction, whereas the weld path in COI-BIAX2 and COI-BIAX4 was parallel to the rolling direction. Initial crack was introduced parallel to the weld seam orientation and located at the specimen geometric centre. It is in the thermal-mechanical affected zone (TMAZ) on the so-called weld retreating side, about 5 mm from the weld nugget centre. The weld centre is not placed in the specimen geometric centre. Details are in Table 1. All tests were performed at applied biaxial stress ratio k = 1 and maximum nominal stress 100 MPa acting on the 250-mm-wide loading arm area with the thickness of 1.6 mm. The cyclic stress ratio was 0.1 in both load directions. Details of test measured crack growth lives and growth path can be found in [18]. Test results are presented in Section 4 in comparison with predicted crack growth behaviour. Further studies were undertaken by performing numerical modelling on various k ratios (k=0-2) to investigate the influence of biaxial stress ratio on crack growth rate and trajectory. Fig 1. a) Test specimen and strain gauge locations (thickness of pad-up and griping area is 2.8 mm; skin pocket thickness is 1.6 mm); b) FE model of ½ specimen and local model of the crack tip. Table 1 Test specimens and loading condition (applied stress y = 40 kN) Specimen No. Weld orientation Applied load ratio (k) COI-BIAX1 Weld orthogonal to rolling direction 1 COI-BIAX2 Weld parallel to rolling direction 1 COI-BIAX3 Weld orthogonal to rolling direction 1 COI-BIAX4 Weld parallel to rolling direction 1
RkJQdWJsaXNoZXIy MjM0NDE=