PSI - Issue 42

Denny Knabner et al. / Procedia Structural Integrity 42 (2022) 561–569 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

562

2

1. Introduction If it is technically impossible to manufacture components from one piece, component connections are necessary. Depending on the area of application, these component connections should fulfil two different basic functions: on the one hand, movements must be guided – these component connections are referred to as joints; on the other hand, forces and torques must be transmitted. Whereas joints are generally designed to be form-fitted (e.g., linear guides), force-transmitting connections can be form-fitted (e.g., dovetail connection), friction-fitted (e.g., shrink fit) or material-bonded (e.g., welding). There is clearly relative movement (sliding) at the contact surfaces of the components in joints resulting from the function of the joint to guide movement. Global sliding at the contact surface of force-transmitting connections would lead to failure of the function, since the value of the force (or torque) to be transmitted would be greater than manageable by the component connection. Nevertheless, localised microsliding movements (slippage) can occur in form-fitted (e.g., [1 – 6]) and friction-fitted (e.g., [7,8]) force transmitting connections. This is due to cyclical elastic deformation, differences in stiffness of the components and vibrations occurring under dynamic loads [9,10]. These microslips can lead to the phenomenon of fretting fatigue. In addition to effects such as wear or the formation of a debris layer, tribological mechanisms also lead to the formation of microcracks [11 – 13]. Due to external loads, these cracks can grow and thus lead to component failure. The problem is that, depending on the tribological parameters, the fretting-fatigue strength may be only 20% of the plain-fatigue strength or even less. Affected components designed in the classical way in relation to intrinsic material strengths fail unexpectedly and prematurely. This is because the fretting-fatigue strength depends mainly on the tribological parameters of slip amplitude and contact pressure , as demonstrated in the works of [14 – 19]. However, these parameters are not included in the classical calculation approaches. Figure 1 shows the relationships between fretting-fatigue strength and slip amplitude (a) and contact pressure (b). This applies qualitatively to all material pairings but differs in its quantitative expression. a) b)

Figure 1: a) – Relationship between fretting-fatigue strength and slip amplitude at constant contact pressure (according to [18]); b) – Relationship between fretting-fatigue strength and contact pressure at constant slip amplitude (according to [19])

The works of [20] and [21] compile existing approaches to fretting fatigue mentioned in the literature to date. However, most of those approaches fail to include the parameters slip amplitude and contact pressure, which means that the validity for cases of fretting fatigue must be questioned. Only the approaches of the category "fretting specific parameters" (according to [20]) calculate frictional work as a function of and . However, this method is unable to account for the U-shaped curve from Figure 1a. Furthermore, the method according to Ruiz [22] does not provide a threshold value from which a degree of utilization can be derived. Finally, the validation calculations in [20] were all carried out on laboratory setups at the known failure locations. For real components, however, the failure location is not known and must be correctly predicted, as must the degree of utilization at this point. [23]