PSI - Issue 19

Yukitaka Murakami et al. / Procedia Structural Integrity 19 (2019) 113–122 Yukitaka Murakami et al./ Structural Integrity Procedia 00 (2019) 000–000

114

2

1. Introduction Additive manufacturing (AM) is expected to be a promising new manufacturing process for components having complex geometry. The advantages of additive manufacturing (AM) have been emphasized, especially for high strength or hard steels, which are difficult and costly to manufacture by traditional machining to complex shapes. However, the disadvantage or challenge of AM is the presence of defects and surface roughness which are inevitably produced by the manufacturing process. Many papers on fatigue properties have been published and S - N data have been reported. Possibilities of applicable models for defects have been also discussed from the viewpoint of fracture mechanics. However, the strength level of AM materials is currently far to the ideal goal which must be recognized in the discussion of fatigue properties. Without strict and reliable quality control of components regarding defects and surface roughness, we cannot positively admire the advantages of AM as the new technology. In this paper, the method of quantitative evaluation of defects with complicated shapes and configurations, is explained from the viewpoint of small 3D cracks. First of all, the ideal fatigue strength (goal) to be attained by AM is discussed. Use of Fatigue-Grades from 5 (ideal strength) to 1 (lowest grade) is proposed for mutual comparison of fatigue performance of AM materials. The factors which decrease fatigue strength and degrade tensile properties are quantitatively analyzed. It is verified that the fatigue limit of AM materials is determined by the presence of nonpropagating crack emanating from defects. The reasons are made clear for fatigue fracture from surface defects even by tension-compression fatigue test as due to the difference of population of defects and increase in stress intensity factor for surface crack compared to internal cracks. Practical guides will be presented for the fatigue design and development of high quality AM materials, based on the combination of the 3D defect analysis, the statistics of extremes on defects, and the  area parameter model. It is shown that fatigue notch effect in AMmaterials is influenced by probability of presence of defects at notch root. 2. Where is the goal of fatigue strength for AM materials? How can we reach the goal, i.e. the ideal fatigue strength? We know there are many influential factors. But we should recognize where the goal of the strength is. Figure 1 shows the robust relationship for commercial steels between fatigue limit and Vickers hardness HV , up to HV ~400 (Garwood et al (1951), Nishijima (1980)). The relationship is expressed as Eq(1).  w = ~ 1.6 HV (1) Where,  w is fatigue limit in MPa and HV is the Vickers hardness in kgf/mm 2 . For HV >~400 (Garwood et al (1951)), the fatigue limit drops and has a big scatter due to the presence of nonmetallic inclusions. Up to HV ~400, these nonmetallic inclusions in commercial steels are mostly nondamaging due to small size (Murakami and Endo, T (1980), Murakami (2002)). This is the useful information when we discuss the allowable size of defects in AM materials. There is a theoretical background of the relationship between HV and fatigue limit of smooth specimen. The relationship is related to the correlation among cyclic stress strain curve (CSSC), tensile strength  u and HV (Murakami (2019)). 3. Current experimental results on AM materials and the ideal fatigue limit The data plotted in Fig. 1 shows the current fatigue performances in terms of the fatigue strength and HV for Ti 6Al-4V (Masuo et al (2018) and additional data) and Ni based superalloy 718 (Yamashita et al (2018)). Figures 2 (a) and (b) shows the S - N data for Ti-6Al-4V. Only the specimens with Hot Isostatic Pressing (HIP) followed by surface polish can attain the ideal fatigue strength expected by the robust relationship of Fig. 1. The fatigue limit of specimens without HIP and without surface polish are less than 30% of the ideal fatigue strength. If we test more specimens, we will have much lower fatigue strength. The reason can be explained by the surface roughness (Fig. 4) and the presence of large defects observed at fatigue fracture origins as shown in Fig. 3. As shown in Figs. 1 and 2, only the fatigue limit of the specimens with HIP + Surface polish can attains the ideal fatigue strength. On the other hand, the S - N data of specimens without surface polish and without HIP are far below