PSI - Issue 35

Vera Friederici et al. / Procedia Structural Integrity 35 (2022) 106–114 V. Friederici et al. / Structural Integrity Procedia 00 (2019) 000–000

107

2

Nomenclature R m

tensile strength 0.2 % yield strength Youngs modulus

R p0.2

E

A g

strain before reduction in area

A

strain at fracture

∆ K ∆ K th K max K min

stress intensity factor range

threshold stress intensity factor range maximum stress intensity factor minimum stress intensity factor

N C m S a

number of cycles

coefficient of Paris equation coefficient of Paris equation

stress amplitude

stress

σ

A Specimen

highly loaded surface area of specimen highly loaded surface area of ring structure

A Ring

N 50, Ring

cycles to reach 50 % probability for failure of ring structure cycles to reach 50 % probability for failure of specimen

N 50, specimen

m L

scatter parameter of Weibull-distribution

1. Introduction In wind turbines, slewing bearings are used to adjust the angle of the blades (pitch) for different wind speeds and thus control the power and loads of the wind turbine. Lifetime calculations of slewing bearings have previously considered speed, bearing load, and temperature (GMN, 2020). However, the change to risk-based inspections and complex probabilistic estimations makes sense for applications such as wind turbines for cost-saving reasons (Chahardehi, 2016). These estimates are fundamentally dependent on accurate knowledge of fatigue crack growth behavior. Until now, the prediction of fatigue life still depends very much on empirically determined material properties (Pugno 2006). Therefore, when designing against fatigue using the "safe-life" approach, i.e. for a practically infinite service life, the empirically determined SN-curves are still of interest. The basic fatigue SN-curve is affected by several parameters such as mean stress, roughness, notch, and dimensions. Thus, in order to better understand the life of a bearing, surface corrosion must also be taken into account, since surface defects resulting from pre-corrosion greatly reduce the fatigue limit (Adasooriya, 2019). If the design of components also allows for the initiation and propagation of cracks, the crack propagation behavior of the material must also be considered when determining the service life. Nonmetallic inclusions, local hardness (Murakami, 2002), and corrosion then also have a significant effect on the failure behavior (Álvarez, 2021). Crack propagation behavior can be described by a linear fit to the log-log plot of da/dN versus ∆ K , the well-known Paris' law: = ∗ (1). Crack propagation curves for positive stress ratio R can easily be obtained (ASTM E-547). However crack propagation on compact tension specimens according to ASTM E-647 is only possible for positive stress ratio values. Consideration of compressive stresses on crack growth has great relevance in the assessment of the service life of complex components (Chahardehi, 2016). Additionally, negative stress ratio ( R = -1) also corresponds to the loading condition during rotation bending or push-pull test. In order to be able to use the correct input data for the simulation