PSI - Issue 13

Meike Funk et al. / Procedia Structural Integrity 13 (2018) 279–284 Author name / Structural Integrity Procedia 00 (2018) 000 – 000

283

5

3.3. Correlation of step length and lifetime

The correlation between the length of the step prior to the crack coagulation and the cyclic lifetime for the different experimental conditions on the loading levels 70 and 80 MPa is shown in figure 4. The length and the height of the step are correlated, therefore the examination was focused on the length of the step. Beside some outliers a linear dependence between the length of the step and the cyclic lifetime is visible. Especially for the specimen tested at 80 MPa without overload a nearly perfect linear relation is found. However, Specimens tested at a loading level of 70 MPa without overload show two different linear dependencies. Specimens fatigued with a maximum load of 70 MPa show step lengths up to nearly 3 mm, but specimens fatigued with 80 MPa just up to 1.2 mm. This indicates that the higher load leads to a stronger driving force for coalescence, and the steps are flattening out. The specimen with laser cuts tested at a load level of 70 MPa also show a nearly perfect linear dependence between the cyclic lifetime and the length of the step. Especially the specimens with laser cuts on different levels show the highest cyclic lifetimes and the longest step length within this group. This clearly indicates that the location of the crack initiation sites plays an important role for the cyclic lifetime.

70MPa without overload 70MPa with overload 70MPa with lasercut

2.8

80MPa without overload 80MPa with overload

1.20

2.4

two levels

1.05

2.0

0.90

two levels

1.6

0.75

1.2

0.8

0.60

two levels

length of the step [mm]

length of the step [mm]

0.4

0.45

0.0

0.30

60,000

70,000

80,000

90,000

100,000

110,000

90,000

105,000 120,000 135,000 150,000 165,000

cyclic lifetime

cyclic lifetime

Fig. 4. Linear relationship between the step length and the cyclic lifetime for specimen tested with a maximum stress of 80 and 70 MPa.

4. Discussion

The experiments show that the location of incipient cracks and the presence of steps strongly influence the cyclic lifetime. Investigations of the fracture surface and the notch root of broken specimens allow conclusions on the cyclic lifetime. The presence of steps significantly enhance the cyclic lifetime. When cracks are initiated in different distances to the notch root, they coagulate late and the cyclic lifetime is enhanced. The later the individual cracks coagulate, the higher is the lifetime (figure 4). When the cracks propagate in the same plane, the total lifetime is reduced due to an early coagulation of these cracks. The results clearly indicate that the scatter of the cyclic lifetime is determined by differences in the crack initiation scenarios. Not in every case, a clear relationship between the fracture surface and the cyclic lifetime is evident. There are several examples with a missing explanation for the present lifetime-fracture surface relationship. For example, the specimen loaded with σ max = 70 MPa with the highest lifetime shows a very smooth fracture surface without a significant step at all. Potentially, this specimen shows a retarded crack initiation due to the lack of adequate crack initiation sites. This assumption is supported by the potential curve showing a plateau region in the first quarter of the cyclic lifetime. Moreover, the initiation of many cracks with only small steps in between can lead to a high cyclic lifetime. In this case, the number and location of individual cracks can determines the cyclic lifetime. Currently, a correlation between the amount of cracks and the lifetime cannot be confirmed.