PSI - Issue 18

Matilde Scurria et al. / Procedia Structural Integrity 18 (2019) 586–593 Matilde Scurria, Benjamin Möller, Rainer Wagener, Thilo Bein/ Structural Integrity Procedia 00 (2019) 000–000

590

5

similar results showing higher values of the Young’s modulus in the case of the XZ direction have been also found by testing of specimens manufactured from as printed bulk material and polished to the final shape by Scurria et al. (2019). According to the results achieved so far, it can be stated that additively manufactured Inconel ® 718, for cyclic deformations limited to the linear-elastic regime, presents an anisotropic behavior. In particular, the Young’s modulus is higher when the loading direction is rotated by 45° with respect to the build direction, compared to the configuration in which they are aligned. Furthermore, in the linear-elastic regime, the Young’s modulus is not affected by the heat treatment. With the increasing of the maximum strain amplitude to  a,t = 0.6%, the hystereses enlarge for the XZ direction, while, for Z, they are still quite narrow. In this case, the local plasticity, caused by gradients of the local properties or by geometrical imperfections, is a valid explanation for this phenomenon. In fact, for polished specimens and high values of the plastic strain portion, the cyclic material behavior is isotropic, as shown by Scurria et al. (2019). Since there are no test results that describe the cyclic properties of Z specimens subjected to  a,t = 0.6% and heat treatment ‘A’ (Figure 3d), it cannot be stated that, even for this maximal strain amplitude, along the Z-direction, the cyclic material behavior is independent of the heat treatment. However, it can be deduced from the results of Figure 3f that the hystereses mainly overlap for  a,t = 0.8%. In conclusion, along the build direction Z, the cyclic behavior is independent of the heat treatment. The situation is very different, if the properties are evaluated along the XZ direction. In this case, with the increasing of the strain amplitude, the hystereses “enlarge” or “stretch”, depending on the applied heat treatment. In general, as a consequence of heat treatment ‘A’, the hystereses are wider and cross the axes of the ordinate in correspondence with higher values of the stress. However, in correspondence with the same total deformation, the behavior in tension of ‘A’, represented by the first quadrant, results in a lower value of the stress compared to ‘B’ and ‘C’. Regarding ‘B’ and ‘C’, while ‘C’ tends to behave like ‘A’ for higher values of the strain amplitude, ‘B’ stretches the hystereses to higher values of the stress, in both the first and third quadrants.

Fig. 4. Cycles to crack initiation N i of the tests as a function of the strain amplitude

Finally, while it was not possible to find a direct correlation between the number of cycles to crack initiation and the cyclic material behavior from Figure 3a, 3b, 3d and 3f, additional information are given by the diagrams of Figure 3c and 3e. In Figure 3c, the lower value of the stress corresponding to the maximum total strain of  a,t = 0.6% in the case of the material state ‘A’, leads to a longer life, while, for the states ‘B’ and ‘C’, the higher maximum stress corresponds to a lower number of cycles to crack initiation. In Figure 3e, not only does a higher value of the maximum stress correspond to a lower number of cycles to crack initiation (’B’), but also, for comparable values of the latter, the material, with a more enlarged hysteresis (‘A’), presents a lower fatigue life compared to ‘C’. The results are summarized in the diagram in Figure 4 for the three different strain amplitudes (  a,t = 0.4%, 0.6% and 0.8%) on the ordinate and the number of cycles to crack initiation N i on the abscissa, both on a logarithmic scale. Here, it can be remarked that the Z specimens generally result in a longer fatigue life, even if this difference is much