PSI - Issue 13

Fuminori Yanagimoto et al. / Procedia Structural Integrity 13 (2018) 116–122 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

118

3

Table 1 Chemical composition of the steel employed (mass%) C Si Mn P S 0.14 0.41 1.45 0.017 0.003

vTs [ ℃ ] -40

Table 2 Mechanical properties of the steel employed

Yield stress at room temperature [MPa]

Tensile strength at room temperature [MPa]

Thickness [mm]

30

364

540

2.2. Specimen and experimental procedure

The configurations of the employed specimens and tab plates are shown in Fig. 2. The specimen width was 1,500 mm. The specimens were welded to tab plates. The specimen had side groove whose length was 200 mm on one side of the specimen to make the crack propagate straight. Considering the distance between pins, the reflected stress wave may influence when the crack reached about 600 mm. Thus, the crack was intended to be arrested before it reached 600 mm. Because the yield stress of the employed steel was 364 MPa at the room temperature, the applied stress was set to 315 MPa. The temperature of the specimen should be kept isothermal. However, the isothermal region started 280 mm from the top of the specimen because it is needed to keep the temperature near the crack initiation point low enough to cause brittle fracture. Therefore, there were temperature gradients between the top of specimen and the beginning of the isothermal region. These temperatures were controlled using the system to spray the liquid nitrogen and compressed air in cooling bath.

1,000 (Wide2016) 440(Wide2017)

950(Wide2016) 1,270(Wide2017) 760(Wide2016) 800(Wide2017)

950(Wide2016) 1,260(Wide2017)

unit:mm

1500

1500

760(Wide2016) 745 (Wide2017)

Side groove

Pin hole

300

1000

1500

Test plate

Φ 385

32

30

32

50

50

290

290

Fig. 2 Specimen geometry

2.3. Preliminary crack arrest tests and model prediction

The experimental conditions were determined using the numerical model based on the local fracture stress criterion (Shibanuma et al., 2016a). At first, to obtain usual crack arrestability of the steel, temperature gradient crack arrest tests using normal width specimens were carried out. The experimental procedure followed WES2815 (The Japan Wedling Engineering Society, 2014). The specimen width was 500 mm. The temperature gradients were in a range between 0.26 and 0.33 ℃⁄mm . Some of the crack arrest experiment results were provided by the previous study (Shimada et al., 2017). The local fracture stress of the employed steel was identified as 4,410MPa by comparing the model calculation and the experimental results. The model calculation and experimental results are shown in Fig. 3. Although there are good agreement between the model and the experiments as long as the arrest temperatures are lower, the model calculation does not consist with the experimental results in higher crack arrest temperature cases. This inconsistency was not found in the validation of the model (Shibanuma et al., 2016b).