PSI - Issue 7

I. Milošević et al. / Procedia Structural Integrity 7 (2017) 327 –334 I. Milosˇevic´ / Structural Integrity Procedia 00 (2017) 000–000

331

5

Table 1. Fatigue tests and fractography results for D 4 specimens. Nr. σ norm

√ area

Cycles

ECD

µ m

µ m

1 2 3 4 5

1.01 1.01 0.97 0.99 0.99

13.940.600 8.386.270 13.554.600 4.767.440 15.794.000

9.57

10.80 60.43 18.44 38.28 35.54

53.56 16.34 33.93 31.50

D 7 . 5 fatigue tests (Figure 6) showed a smooth transition between surface and subsurface crack initiation. Surface initiated cracks could be investigated beyond 10 7 cycles. A wide scatter in the TA was observed.

0 . 7 0 . 75 0 . 8 0 . 85 0 . 9 0 . 95 1 1 . 05 1 . 1 1 . 15 1 . 2

1.04

1.55

Data Points Murakami

1.01

1.50

1.45

0.97

1.40

Normalised stress, σ norm , - (log.)

0.93

1.35

0.90

S a,norm , - (log.)

90%

1.30

0.86

S a /(HV+C2), - (log.)

Surface Subsurface None

1.25

0.82

1.20

10%

0.78

1.15

10 4

10 5

10 6

10 7

10 8

3

10

50

Number of cycles, N (log.)

√ area, µm (log.)

Fig. 6. S / N data for specimens D 7 . 5 at T = 20 ◦ C , R = − 1. Fig. 7. Murakami approach - D 7 . 5 data The same type of defects was observed on D 4 and D 7 . 5 fracture surfaces. Defect sizes ( √ area ) were measured from 7.12 to 15.03 µ m. The average defect size is 10.98 µ m. The data is given in Table 2. In Figure 4 and Figure 7 the defect based √ area model showed similar results. The normalised fatigue limit S a , norm decreased with increasing defect sizes. A large scatter can be observed.

Table 2. Fatigue tests and fractography results for D 7 . 5 specimens. Nr. σ norm

√ area

Cycles

ECD

µ m

µ m

1 2 3 4 5 6

0.97 0.93 0.93 0.93 0.93 0.90

9.297.220 26.779.200 18.028.200 16.878.500 22.212.000 16.395.700

7.12

8.04

15.03 11.65 10.44 10.77 10.87

16.96 13.15 11.78 12.15 12.26

Fatigue results at 350 ◦ C were presented in Figure 8. Fatigue strength decreased as expected due to higher temper atures. Most specimens failed originating from slip planes, which are outlined in Figure 9 (a,b). Except one, every specimen failed due to internal initiated cracks. Another three specimens failed due to inclusions. No real transition area was found due to the absence of multiple crack initiation sites. Murakami’s √ area approach could not be applied because of low number of significant data.

5. Discussion

The discussion refers to fatigue results and fractographical analyses of D 4 and D 7 . 5 specimens at room temperature. Fatigue tests at T = 350 ◦ C showed a di ff erent behaviour without a major influence regarding nonmetallic inclusions.