PSI - Issue 54

Nikolai Kashaev et al. / Procedia Structural Integrity 54 (2024) 361–368 Kashaev et al. / Structural Integrity Procedia 00 (2023) 000 – 000

362

2

technology trends of the future (Srivastava et al., 2022). However, the introduction of additive manufacturing into industrial practice depends on the sufficient reliability of manufactured parts in terms of their fatigue performance (Zerbst et al., 2021). With this regard, a better understanding of material behavior in the case of cycling loading and the influence of possible manufacturing defects is of high importance. Surface defects such as local notches or undercuts as well as internal defects such as pores, lack of fusion defects, or hot cracks are decisive for the failure behavior of additively manufactured components (Zerbst et al., 2021; Akgun et al., 2021). It was recently reported that the fatigue life of Ti-6Al-4V manufactured by WAAM greatly depends on the failure source, since samples with embedded porosity show much higher fatigue limit and overall fatigue performance compared with samples with surface porosity (Akgun et al., 2021). To suppress the surface defects, e.g. mechanical (Bagehorn et al., 2017) or chemical polishing (Persenot et al., 2020) were successfully used for additively manufactured parts, whereas laser surface remelting (Fomin et al., 2018b) and residual stress engineering by means of laser shock peening (Kashaev et al, 2020) were implemented in laser beam-welded components. When the stress concentration at the surface is reduced, fatigue cracks tend to start at internal material local defects such as porosity (Gong et al., 2015; Li et al., 2016; Cao et al., 2018; Fomin et al., 2018). In this sense, some authors have concluded on a non-competitive hierarchical fatigue failure model that can explain the fatigue life behavior of AM components (Li et al., 2016; Cao et al., 2018). Surface roughness and defects are considered the most critical defects that give rise to early fatigue failure, then internal porosity triggers the failure and finally in samples without surface defects and porosity, the failure is initiated in microstructural features, reaching out usually the fatigue performance of equivalent wrought samples. The application of hot isostatic pressing (HIP) to AM parts can completely remove internal defects and therefore, dedicated surface treatments together with HIP are usually required for the most critical applications (Zerbst et al., 2021). For the evaluation of fatigue life in the case of short-crack growth from internal defects, approaches in which the dependence of the short-crack growth threshold on the crack length is taken into account are required. In principle, this dependence can be expressed either in terms of the threshold for the stress - Kitagawa-Takahashi (KT) diagram (Kitagawa and Takahashi, 1976) or in terms of the threshold for the stress intensity factor range,  K th - the cyclic R curve (Tanaka and Akiniwa 1988; Zerbst et al., 2016). In a previous study for the laser-welded Ti-Al-4V joints, it was demonstrated that the KT diagram is a good fatigue evaluation tool only when the fatigue limit is the focus of the analysis (Fomin and Kashaev, 2020). The KT diagram cannot provide information regarding fatigue crack growth rates for a given failure. However, using the cyclic R -curve, it was possible to consider the crack growth rates in the case of internal pores, and thus not only the size of defects but also the distance of defects from the surface were considered for probabilistic fatigue life prediction (Fomin et al., 2018a; Fomin and Kashaev, 2020). The aim of the current study is to investigate to which extent the reliability assessment of a component in the presence of internal defects, which has already been validated for laser beam-welded Ti-Al-4V joints, can be transferred to additively manufactured structures. Wire and arc additive manufacturing (WAAM) as an additive process was used to produce Ti-6Al-4V bulk structures. Fatigue specimens were extracted from these structures to characterize the fatigue behavior and to adjust the fatigue life prediction model. 2. Experimental 2.1. Material and WAAM process In this work, a wall of Ti-6Al-4V was manufactured using the cold metal transfer (CMT) Fronius TransPuls synergic 4000 CMT R power source, and Robacta Drive CMT W/F++/6.25m torch from Fronius International GmbH. The torch was attached to a 6-axis KUKA robot (KR 16 KS model with a KRC2 controller from KUKA AG). The shielding gas used in the torch was He. The gas flow was set at 15 l/min. The wall was manufactured in a closed chamber of 400 mm × 250 mm × 300 mm to avoid the oxidation of the wall during the manufacturing. The chamber was previously filled with Ar 99.9999 % for a period of 15 minutes. The gas flow inside the chamber was 15 l/min and it was maintained during the whole material deposition process. This method assures less than 0.18 wt.-% oxygen concentration in the part. The oxygen content was measured by hot gas extraction (HGE) analysis using a LECO system (LECO Corporation) with a helium carrier gas and carbon crucibles. The measurement was performed