PSI - Issue 13

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

280

2

length. The further crack propagation can be calculated using fracture mechanics, but this period covers only a minor part of the total cyclic lifetime. Previous experiments on SEN-specimens with a constant stress amplitude undertaken by Bär and Tiedemann (2017) have shown, that cracks are initiated at both edges of the notch and propagate in form of quarter ellipses before they coalesce, forming a continuous crack front. Higher loads favor multiple crack initiation in the notch root, these cracks appear in a half elliptical shape. The propagation of these short cracks is determined by the stress state in the notch root and the interaction of the individual cracks. Fatigue investigations of Bär and Wilhelm (2014) with periodic overloads have shown, that the cyclic lifetime can be associated with the fracture surface. If steps between the initiated cracks appear in the fracture surface, due to an extended individual crack propagation, the cyclic lifetime is enhanced. These steps occur, when the cracks are initiated in different distances to the root notch. In contrast to constant amplitude testing, the propagation of these experiments were strongly influenced by the periodic overloads. As shown by Hu et.al. (2009) the propagation of initiated cracks in the same level is accelerated, due to the interaction between the individual cracks. In this work the propagation of short cracks in SEN specimens is investigated with a focus on the number and location of the crack initiation sites in the notch.

2. Experimental Details

2.1. Material and testing equipment

All experiments were performed on SEN (Single-edge notched) specimens of an EN AW 7475-T761 clad sheet material with the dimensions of 2.8 x 12 x 80 mm. The specimens with a notch depth of a 0 = 1.0 mm were loaded in a special equipped hydraulic testing machine. A detailed description of the testing machine is given by Bär and Wilhelm (2014), Tiedemann and Bär (2014) or Bär and Tiedemann (2017). All experiments are performed under fully reverse loading conditions ( R = -1) with a frequency of f = 20 Hz.

2.2. Experimental details

To generate different crack scenarios, the specimens were grouped in seven different experimental conditions. For three different maximum stresses of 60, 70 and 80 MPa, specimens were loaded under the following conditions: a) constant load, b) constant load with one tensile-compression overload (OL) after 200 cycles with σ max = 240 MPa, c) constant load with introduced laser cuts with a depth of about 3 µm at both edges of the notch root. The laser cuts were introduced either directly in the notch root or in a certain distance to the notch root. In this case the specimens were fatigued with a maximum stress of σ max = 70 MPa. A high-resolution DC potential drop method enabled the in-situ measurement of the crack length as described by Bär and Tiedemann (2017). The system was improved by welding the wires for the potential drop measurement directly onto the specimen surface instead of using press-fitted pins. Due to this improvement all “starting effects” caused by deformation of the holes and subsequent changes of the conductivity of the press-fitted pins are now eliminated and the noise of the potential signal is halved. Therefore, the potential drop is rising constantly and the crack length can be calculated reliable right from the beginning of the fatigue experiment. This allows to detect an early crack propagation even below a change of the potential drop of 1 %. This corresponds to a crack length of some micrometers, assuming a continuous crack front. 2.3. DC potential drop method