Issue 38

X. Yu et alii, Frattura ed Integrità Strutturale, 38 (2016) 148-154; DOI: 10.3221/IGF-ESIS.38.20 149 surveillance aircraft, severe multiaxial loads may exist at some of the primary structural locations. Multiaxial fatigue analysis, including fatigue crack growth (FCG) analysis under non-proportional mixed mode I and II loads, is needed for these locations. At present, experimental FCG investigations are predominantly undertaken for mode I or proportional mixed mode loads. Only a few studies [1-8] focus on non-proportional mixed mode loads. Limited results on A106-93 mild steel [3] reveal that a long and stable shear mode FCG – which is significantly different from the commonly understood open mode FCG – could be produced by non-proportional loads. To illustrate this difference, Fig. 1 compares the FCG behaviour under two load cases: (i) proportional load case TD1, under which, the FCG deviated into a direction that can be approximately predicted by the maximum tangential stress (MTS) criterion [9], and striations were found along the deviated crack path; (ii) non-proportional load case TD2, under which the FCG deviated to a direction that is about 60° off the MTS prediction, and dimples were found along the deviated crack path. For this particular case, the deviation angle is approximately predictable by the maximum shear stress (MSS) criterion [10]. The shear mode FCG as shown in Fig. 1 (d) is more than 8mm long along the deviated path. Thus, it is not a short- distance propagation in shear mode occurring in an early stage of crack growth as discussed in [7]. Nor is it a transient behaviour occurring at a short-distance coplanar growth following the change of load mixture as reported in [11]. The persistence of the shear mode FCG as presented in Fig. 1 (d) indicates the complexity of underlying mechanism under non-proportional loads. It means that a stage II fatigue crack can propagate in other than open mode growth, and in such a case, the crack path is absolutely unpredictable by the commonly accepted MTS criterion. Nevertheless, the above finding is so far only based on experimental results for A106-93 mild steel. It is not clear whether this finding also applies to other materials, such as aluminium alloy (AA) 7075-T651, which is typically used for aircraft structures, as no similar tests have been reported for the latter in the open literature. Hence, the purpose of this study is to investigate FCG behaviour in AA7075-T651 under non-proportional mixed mode I and II loads. In particular, it aims to clarify whether the long and stable shear mode FCG also occurs. For the sake of comparison, the specimen design and majority of the mixed-mode load cases used in the present study are the same as reported in [3]. Figure 1 : FCG from a mode I pre-crack, under proportional (a-c) , and non-proportional (d-f) , mixed-mode loads [3, 12]. (SSY in the figure refers to small scale yielding). S PECIMEN AND TEST PROCEDURES he fatigue tests of the present study were performed on an Instron 8500 servo-hydraulic tension-torsion biaxial fatigue test machine with a capacity of ±250 KN and ±2000 Nm. The test machine is located in the Structural Laboratory, School of Civil Engineering at the University of Sydney. This is the same test machine as the one that T