PSI - Issue 17

Meike Funk et al. / Procedia Structural Integrity 17 (2019) 183–189 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

186

4

D a

0,1445

U max 1.96 · deviation (95% of the data) mean

(b)

4

(a)

0,1444

0,1443

3

0,1442

U max (mV)

2 D a (µm)

0,1441

0,1440

1

0,1439

0

2

4

6

8

10

12

0

2

4

6

8

time (min)

a (mm)

Fig. 2. (a) Scatter of a thermal stabilized DCPD signal. (b) Δa -values for ΔU scatter = 0.1µV and U 0 = 0.14419mV

2. Results

The SEN-specimens were fatigued under fully reverse conditions with a loading amplitude of 60 MPa. In order to visualize the crack front on the fracture surface, periodic overloads were applied every 50,000 cycles. Additionally, the crack length was recorded time-synchronously to the experiment using the two potential probes. A correlation between the two recorded signals of the DCPD and the crack length and shape determined on the fracture surface allows conclusions about the propagation of the initial cracks.

2.1. Detection of long cracks

On the fracture surface, the marker-loads visualize the crack front at defined cycle numbers, and allow the tracking of the crack path through the material until failure. Figure 3a shows the crack surface of a specimen with the crack front evolution marked by overloads. The crack front is nearly straight and parallel to the notch root. The crack surface shown in figure 3b is inclined, the crack length on the right hand side is longer than on the left hand side. The differences between the crack lengths on both surfaces are increasing with the crack length. This changes should reflect in the run of the measured potentials.

Fig. 3. Fracture surfaces of specimens with crack fronts marked with overloads. (a) Crack front parallel to the notch root, (b) inclined crack front.