PSI - Issue 18

Ibrahim Al Zamzami et al. / Procedia Structural Integrity 18 (2019) 255–261 Author name / Structural Integrity Procedia 00 (2019) 000–000

260

6

the IIW and EC9, respectively. This is not surprising since the negative inverse gradients provided by the available design codes were determined by reanalyzing a large number of experimental data obtained by testing rigid and stiff welded joints, i.e. welded joints with a thickness much greater than 5 mm. In contrast, experimental fatigue curves determined experimentally when testing thin and flexible welded joints are characterized by a negative inverse slope varying between 3 and 6 (Sonsino et al. 2010). That is why Sonsino et al. (2010) recommends performing a fatigue assessment of thin welded joints through fatigue curves with the same endurance limit (at 2 million cycles) as the one provided by the standard codes and a negative inverse slope invariably equal to 5. By using the methodology recommended by Sonsino et al. to evaluate the fatigue strength of thin welded joints, the use of the modified EC9 design curves (grey dashed line in Fig. 3a-c) and the modified IIW design curves (black dashed line in Fig 3a-c) leads to a more conservative estimate of the fatigue life of aluminium-to-steel welded joints. In particular, according to the Wöhler diagrams of Figure 3a-c, it has been observed that the modified design curves of IIW provided conservative fatigue life estimates for all welded configurations. Similarly, it was observed that the modified curves of EC9 led to conservative predictions of fatigue resistance in all cases except the butt joints (Fig. 3a). For the non-load carrying fillet tee-welded joints (Fig. 3d), the steel plate is subjected to fatigue loading whereas the aluminium plate works as a stiffener. Looking at Fig. 3d, although the tee-welded joints used in the present investigation are classified as thin joints, it worth mentioning that the use of the negative inverse slope suggested by the standard codes results in a conservative fatigue life estimation of the aluminium-to-steel welded joints. In contrast, the use of the k value suggested by Sonsino et al. (2010) would result in a higher level of conservatism in estimating the fatigue lifetime of the tee-welded joint (Fig. 3d). In summary, by using the nominal stress approach, the fatigue behaviour of the butt-welded (Fig. 3a), cruciform welded (Fig. 3b) and lap-welded joints (Fig. 3c) can be assessed with a higher level of accuracy by treating the joints as conventional aluminium to aluminium welds and setting the negative inverse slope equal to 5 as suggested by Sonsino. Conversely, for the tee-welded joints (Fig. 3d) the hybrid-welded connections can be considered as standard steel-to-steel welded joints without the need to change the value of the negative inverse slope. 4. Conclusions The aim of this investigation was to provide a detailed investigation of the fatigue strength of aluminium-to-steel thin welded joints using the nominal stress approach. The key conclusions of the present investigations are the following:  The fatigue fracture of the aluminium-to-steel welded joints always initiates from the interface between the weld seams and the aluminium plate. Hence, the fatigue strength of aluminium-to-steel hybrid joints can be designed by treating these joints as conventional aluminium-to-aluminium welded connections. Except for the tee welded joints, the cracks initiates from the interface between the weld and the steel plate, and it can be treated as conventional steel-to-steel welded joints and use the IIW and EC9 to estimate the fatigue strength.  The negative inverse slope, k, suggested by Sonsino was seen to provide a conservative fatigue strength of the aluminium-to-steel welded joints using the nominal stress approach. Acknowledgements EWM® (www.ewm-group.com) is acknowledged for supporting the present research investigation. References Al Zamzami, I., Di Cocco, V., Davison, JB., Lacoviello, F., Susmel, L., 2018. Static strength and design of aluminium-to-steel thin welded joints. Welding World, 1–18. Al Zamzami, I., Davison, J.B., Susmel, L., 2019. Nominal and local stress quantities to design aluminium-to-steel thin welded joints against fatigue. Int J Fatigue 123, 279-295. Borrego, LP., Costa, JD., Jesus, JS., Loureiro, AR., Ferreira, JM., 2014. Fatigue life improvement by friction stir processing of 5083 aluminium alloy MIG butt welds. Theoretical and Applied Fracture Mechanics 70, 67–74.