PSI - Issue 19

Masanori Nakatani et al. / Procedia Structural Integrity 19 (2019) 312–319 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

315

4

Fig. 2. Shapes and dimensions (in µm) of the artificial defects: (a) ϕ 100-µm drill hole, (b) ϕ 200-µm drill hole, (c) ϕ 500-µm drill hole, (d) 200-µm, semi-circular, electrical discharge machining (EDM) notch, (e) focused ion beam (FIB) notch, (f) circumferential notch, (g) 800-µm, semi-circular EDM notch.

3. Results and discussion

3.1. Influence of defects on the fatigue limit of Alloy 718

The influence of defects on the fatigue limit of Alloy 718 was examined in detail in a previous publication by the authors (2019). The key research findings of the study, that is, the relationship between fatigue limit, σ w , and defect size, √ area , are detailed in Fig. 3. As evidenced by the figure, a marked difference in defect-size tolerance could be observed between the FG and CG specimens. In the FG specimen with a focused ion beam (FIB) notch (Fig. 2 (e), √ area of 35 µm), although cracking occurred at the defect site, the fracture propagated in the smooth region of the specimen, instead of at the defect. This indicated that the FIB notch did not negatively impact the fatigue limit of FG, whereas a drill hole (Fig. 2 (a), √ area of 93 µm) caused fatigue limit degradation. On the other hand, the fatigue limit of the CG specimen was not diminished, despite the introduction of drill holes with equal or larger diameters (Figs. 2 (a) and (b), respective √ area of 93 µm and 185 µm). Based on these facts, it can be inferred that grain size affected the threshold defect sizes of each material ( √ area 0, FG and √ area 0, CG ) , above which defects became detrimental to the fatigue limit. A similar conclusion was also reported in a study by Ono et al. (2015), whereby it was strongly maintained that the largest grain size dominates crack initiation and is one of the factors governing the alloy’s fatigue strength. In the defect-size regime where the fatigue limit was adversely affected, the fatigue limit decreased with an increase in defect size. In this regime, the applicability of the √ area parameter model proposed by Murakami and Endo was assessed by comparing the fatigue limits obtained through experiment and prediction. The fatigue limit was predicted by the following equation: (1) where σ w is the predicted fatigue limit in MPa and √ area is expressed in µm. The experimental values were in good agreement with the prediction in the √ area regime smaller than 185 µm, within a 10% error margin. In contrast, in the larger √ area regime, the increase in √ area led to a greater discrepancy between the experimental data and the predictions. The prediction using Eq. (1) was unconventional at √ area > 300 µm, perhaps due to the transition of the fatigue crack behavior from “small crack” to “large crack”. In the small -crack regime, the fatigue limit was determined by 1/ 6 w  = 1.43( 120) / ( ) HV area +