PSI - Issue 5

Marek Smaga et al. / Procedia Structural Integrity 5 (2017) 989–996 Marek Smaga et al. / Structural Integrity Procedia 00 (2017) 000 – 000

995

7

the beginning of the high cycle fatigue regime. The benefit of a martensitic surface layer on fatigue life can clearly be seen at both test temperatures.

1,E+07

1,E+07

10 7

10 7

(b)

 a = 180 MPa

(a)

 a = 270 MPa

MSL p

10 6

10 6

1,E+06

1,E+06

N f

N f

ASL p MSL t, f=0.15

MSL t, f=0.15

MSL t, f=0.35

MSL t, f=0.35

MSL p

ASL t, f=0.15

1,E+05

1,E+05

10 5

10 5

ASL t, f=0.15

ASL p

10 4

1,E+04

10 4

1,E+04

1 2 3 4 5 6 7 8 9 0.7 0.8 2 3 4 5 6 7 8 R z in µm

1 2 3 4 5 6 7 8 9 0.7 0.8 2 3 4 5 6 7 8 R z in µm

Fig. 6. Number of cycles to failure for specimens with different surface morphologies fatigued at (a) AT and (b) 300°C.

At AT, the MSL p specimen achieved even the ultimate number of cycles N U = 2×10 6 without failure, while the ASL p specimen with comparable surface roughness (see Fig. 5) but turned without CO 2 snow cooling, i.e. purely austenitic microstructure, failed at N = 2×10 5 . Furthermore, an increase of the roughness parameter R z by a factor of 3 and 10 for the MSL t, f=0.15 and MSL t, f=0.35 specimens, respectively, had no significant influence on fatigue life and both specimens achieved N f > 10 5 , i.e. in the range of conventionally turned and polished specimens. This result clearly indicates that the well known detrimental effect of surface roughness on fatigue life can be suppressed by martensitic surface layers. Accordingly, specimens without martensitic surface layer and higher roughness, e.g. ASL t, f=0.15 , reach smaller number of cycles to failure (see. Fig. 6a). At 300 °C all specimens with martensitic surface layer obtained higher number of cycles to failure compared to the specimens with purely austenitic microstructure independent of surface roughness. The highest number of cycles to failure was achieved for the specimen MSL t,f=0.15 with R z = 1.8 µm. Interestingly, at 300°C, the specimen with martensitic surface layer after mechanical and electrolytic polishing achieved a smaller number of cycles to failure compared to the specimen with as-turned surface morphology resulting from the same turning parameters.

3.4. Cyclic deformation behavior

Fig. 7. Development of plastic strain amplitude versus number of cycles during fatigue tests at (a) AT and (b) 300°C.

To analyse the influence of surface morphology on cyclic deformation behavior of metastable austenitic stainless steel, the plastic strain amplitudes are plotted versus the number of cycles for all investigated morphologies in Fig. 7. It can be clearly seen that the ASL p specimen achieved the highest plastic strain amplitude during the whole fatigue process at both testing temperatures. At ambient temperature, the initial cyclic softening was followed by significant cyclic hardening due to the gradual formation of deformation induced  ´-martensite, as it was proved by in-situ magnetic sensor measurements Smaga et al. (2017a). At 300°C, the longer period of initial cyclic softening was followed by a saturation state up to end of fatigue life only for the ASL p specimen. All other specimens show slight cyclic hardening – see Fig. 7b. The possible reason for this behaviour could be a local formation of deformation