PSI - Issue 19

Kim Bergner et al. / Procedia Structural Integrity 19 (2019) 140–149 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

143

4

medium cycle fatigue regime by means of a regression calculation based on the maximum likelihood method and the S-N curve parameters at the kneepoint N k and the limit number of cycles N lim = 10 7 [5]. The scatter band is assumed to be constant in the medium cycle fatigue regime and is adapted for the high cycle fatigue regime until the limit number of cycles N G = 1∙10 7 . For this, the slope k* in the high cycle fatigue regime has to be specified.

2.3. Metallographic analysis

Fatigue tested bending specimens were investigated by stereomicroscopy and pictures were taken of the fracture surface as well as specimens being chemically etched and probed by light microscopy for phase analysis. To check whether, in addition to the presence of the casting skin, the rim zone proportion also has an influence on the fatigue strength, the proportions of the rim zones on the specimen cross section were determined by fracture surfaces analysis (Fig. 4). Subsequently, the bending specimens were divided into four categories, according to their determined rim zone proportion (Table 2) [6].

Table 2. Rim zone proportion categorisation of the specimen cross section

Fraction of the rim zone on the specimens cross section

Low [%]

Medium [%]

High [%]

Rim zone foothill

2 - 11

12 - 21

22 - 31

-

Fig. 4. Fracture surface with 10 % rim zone proportion

2.4. Strain-controlled fatigue tests The fatigue analysis, for the rim zone and for the bulk material, requires the fatigue strength and the stress-strain behavior. Due to the low yield strength of the rim zone, the strain-life approach is used for the fatigue assessment. Accordingly, the fatigue strength and the cyclic stress-strain behavior are determined in strain-controlled fatigue tests with un-notched specimens of mainly ferritic nodular cast iron EN-GJS-400-15. The specimens were removed from cast blanks, shown in Fig. 2, and had a width of w = 6 mm and height of h = 0.4 mm, which led to a highly stressed volume [7, 8] of HBV 90% = 24 mm³. At room temperature, the fatigue tests were carried out for alternating loading, R ε = -1, in a servo-hydraulic test rig with maximum loads of 2.2 kN under constant amplitude loading. All specimens were loaded until crack initiation or until the achievement of the limit number of cycles of N G = 1·10 7 . Through the strain-controlled tests, as a first step, cyclic stress-strain curves according to Ramberg-Osgood (1) [9] and strain-life curves according to (2) Coffin [10], Manson [11], Basquin [12] and Morrow [13] were determined. = , + , = + ( ′ ) 1 ′ (1)