PSI - Issue 19

Masahiro Takanashi et al. / Procedia Structural Integrity 19 (2019) 275–283 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

278

4

2.3. Test conditions

Two types of tests were conducted. The test matrix is summarized in Table 2. The first test was intended to confirm the size effect on the fatigue life. The second test was intended to examine the mean stress effect. The target fatigue lives were set to either 1.0×10 5 or 2×10 6 cycles. The strain amplitudes corresponding to the target lives were calculated from the best-fit curve proposed by the DFC subcommittee (Kanasaki et al., 2013). It must be noted that to prevent a run-out (non-failure) test, the strain amplitudes were increased by 10 – 20%.

Table 2. Test conditions. Test ID

Material

Aim of test

Target life

Upper strain (%)

Strain amp. (%)

Mean strain (%)

1.0×10 5

CS1 CS2

Carbon steel Carbon steel

Size effect Mean stress Size effect Size effect Mean stress Mean stress

0.24 0.30 0.22 0.18 0.30 0.30

0.24 0.24 0.22 0.18 0.22 0.18

0

< 1.0×10 5

0.06

LAS1 LAS2 LAS3 LAS4

Low-alloy steel Low-alloy steel Low-alloy steel Low-alloy steel

1.0×10 5 2.0×10 6

0 0

< 1.0×10 5 < 2.0×10 6

0.08 0.12

3. Experimental results

Figure 2 indicates the relation between the strain amplitude and the failure life. Here, the failure life was defined as the cycle when the main crack grew to more than half of the plate thickness (45mm). The solid line is the best-fit curve proposed by the DFC subcommittee. The curve is drawn using the ultimate tensile strength of low-alloy steel, 597 MPa (see Table 1). All test data of the large-scale specimens showed longer lives than those given by the best fit curve, which was developed based on the small-sized specimens. The open and solid symbols denote the mean stress effect test (with mean strain) and the size effect test (without mean strain), respectively. As is obvious from Fig. 2, the fatigue lives with the mean strain are shorter than those without the mean strain for both materials.

Fig. 2. Fatigue test results.

Next, the crack size was investigated at N bfc , the fatigue life determined by the best-fit curve proposed by the DFC subcommittee. Figure 3 shows the crack size at N bfc in each specimen. In CS2 test with mean strain, the crack size was not obtained at N bfc (= 6.3×10 4 ) because of the crack coalescence. Many small cracks of the order of 1 mm were detected in the replica at 5×10 4 cycles. At the next observation, 6×10 4 cycles, crack coalescence occurred and the main crack went through the notch width. Except for CS2 test, the crack length at N bfc was measured with a