PSI - Issue 2_A

Nobuo Nagashima et al. / Procedia Structural Integrity 2 (2016) 1435–1442 Author name / Structural Integrity Procedia 00 (2016) 000–000

1439

5

Table 2 It shows K p and, C p ( ε pa = C p / N f

Kp ), K

e and C e ( ε ea = C e / N f

Ke ).

K p

C p

K e

C e

0.49 0.61 0.45 0.38

0.28 0.54 0.15 0.10

0.13 0.08 0.07 0.24

0.005 0.006 0.009 0.014

Normalized steel ( σ B =450-600MPa) QT steel ( σ B =900-1200MPa)

FMS alloy ( σ B =950MPa) SUS304 steel ( σ B =618MPa)

Normalized steels QT steels FMS alloy SUS304

10 -3 Elastic strain amplitude, ε ea 10 -2

10 1 10 2 10 3 10 4 10 5 10 6 10 7 Number of cyclics to failure, N f (cycles)

Fig. 6. The elastic strain amplitude ε ea plotted against the number of cycles to failure N f. Each marks showing ,● FMS alloy, ○ SUS304 steel, ＋ normalized steels ( S25C, S35C, S45C ( σ B =450-600MPa)), and × quench tempered steels ( S45C, SCr440, SCM435, SNCM439 ( σ B =900-1200MPa)). Figure 5 shows the relationship between ε pa and N f. The ε pa - N f relationship can be given by ε pa = C p / N f Kp . This relationship, called the Manson-Coffin law, is the most fundamental relationship in low cycle fatigue. Table 2 shows K p and C p for the materials tested. Data from NIMS fatigue datasheets on carbon steel and low carbon alloys are plotted in the figure (NIMS Fatigue Data Sheet (FDS) Nos.38, 39, 44, 45, 52, 56) where ‘+’ indicates low-strength normalized pearlitic steel (N steel) (S25C steel (FDS No.38), S35C steel (FDS No.39), S45C steel (FDS No.44)), and ‘x’ indicates martensitic steel (QT steel) (S45C steel (FDS No.44), SCr440 steel (FDS No.45), SCM435 steel (FDS No.52), SNCM329 steel (FDS No.56)). A comparison of low-strength N steel and QT steel shows that, by the Manson-Coffin law, K p tends to be higher for higher strength. The ε pa - N f plots for FMS alloy and SUS304 steel in the present study show good linear correlation (Figure 7), indicating that the Manson-Coffin law holds for both materials. As shown in Figure 2, the FMS alloy has high strength, but the K p for the alloy is 0.45, which is less than the value of 0.49 for the low-strength N steel. Figure 6 shows the relationship between the elastic strain amplitude ( ε ea = ε ta - ε pa ) and N f ． The ε ea - N f relationship can be given by ε ea = C e / N f Ke . 3) Table 2 shows K e and C e for the steels tested. Data from NIMS fatigue datasheets on N steel and QT steel are plotted in the figure. The ε ea - N f relationship appears be divided into two types: one characterizing N steel, and one, QT steel. In addition, the values for the FMS alloy are plotted above those for the QT steel. This shows that if steels have a comparable cyclic life to failure, the elastic strain is larger for steels with higher strength. A possible cause of the greater elastic strain values for the FMS alloy is the pseudoelastic strain shown in the hysteresis loops in Figure 3. A large cyclic elastic strain means that the plastic strain is reduced at constant total strain amplitude, and consequently the accumulated plastic strain is reduced, resulting in longer low cycle fatigue life. Therefore, the ε ea - N f relationship is suitable for use as an index for evaluating the low cycle fatigue properties of pseudoelastic FMS shape memory alloys. Figure 7 shows the respective change in N and σ a (i.e., cyclic softening and hardening curves). A comparison of the respective cyclic softening and hardening behaviors shows that SUS304 steel exhibits hardening in the initial stage of cyclic loading at ε ta = 0.6% and 0.9%, but no significant change until failure; and exhibits cyclic hardening from the beginning at ε ta = 1.4% and 2.0%. The FMS alloy exhibits cyclic hardening from the beginning at ε ta = 2.0%. All the curves exhibit a similar trend at other values of ε ta . The strain hardening is suppressed in the initial stage of cyclic loading, and increases in the middle stage. The hardening is saturated at the threshold value of ε ta .