PSI - Issue 68

S.R. Raghuraman et al. / Procedia Structural Integrity 68 (2025) 769–775 S.R. Raghuraman et al. / Structural Integrity Procedia 00 (2025) 000–000

771

3

The research project is split into two parts. The first part includes initial fatigue tests and encompasses LITs, CATs as well as the primary loading experiments in the HCF regime that were performed on a servo-hydraulic fatigue testing system (Shimadzu Deutschland GmbH, type EHF-U 50 kN). The tests were carried out stress-controlled with a sinusoidal load-time function and a stress ratio of R s = -1 at room temperature (RT). Round specimens with a test section diameter of 6 mm, a gauge length of 22 mm and a total specimen length of 175 mm were subjected to LITs and CATs at a test frequency of 5 Hz. To characterise the fatigue behaviour, a tactile extensometer (Shimadzu Deutschland GmbH, type Dynastrain) with a measuring length of L 0 =15 mm was used for the mechanical stress-strain-hysteresis measurement. In addition, supplementary non-destructive testing (NDT) methods based on temperature using an IR camera (Micro-Epsilon Messtechnik GmbH & Co. KG, type thermoIMAGER TIM QVGA HD) and electrical resistance measurements using the 4-wire method (Toellner, type TOE 7621) were applied to detect the material response as a result of cyclic loading. Ex-situ investigations with a magnetic Barkhausen noise measuring system (QASS, type µmagnetic) provided additional information regarding the current damage state, in order to evaluate the influence of primary loading. The stress amplitude of the primary loading σ a, PL and the associated extent of primary loading in terms of number of cycles were determined on the basis of a trend S-N curve provided through the fatigue life prediction method StressLife. Light-, confocal- and scanning electron microscopic examinations were carried out to quantify the development of damage characteristics in the surface area, which can particularly be assessed in terms of change in roughness arising from the occurrence of intrusions and extrusions, leading to microcracks as well. For this purpose, microstructural investigations of different fatigue stages were compared to analyse the extent of the induced surface damage. In the ensuing part of this research, specimens were subjected to secondary loading with higher testing frequency, whereby a distinction was made between specimens with and without reconditioning. Specimens of the initial state were also tested as a reference. For this purpose, a resonance pulsator (Russenberger Prüfmaschinen AG, type GIGAFORTE 50 kN) was used, which enables a loading at 1 kHz up to the transition to the Very-High-Cycle-Fatigue (VHCF) regime until failure. Based on different numbers of cycles to failure of specimens pertaining to various conditions, the RP could be estimated. 3. Experimental results An LIT and two CATs were carried out at a test frequency of f = 5 Hz, from which input variables for the lifetime prediction according to StressLife can be calculated. The procedure for calculating a trend S-N curve is described in detail by Starke (2019) focussing on an unalloyed SAE 1045 steel (1.1191, C45E) in a normalised condition. The same procedure was equivalently carried out for the SAE 4140 steel within this work. The data extracted from the LIT and CATs facilitated the generation of the trend S-N curve according to StressLife as depicted in Fig 2 (a). Based on these insights, the fatigue limit of the material at 10 7 cycles can be estimated to σ a = 575 MPa. The red data points represent the conventionally determined numbers of cycles to failure. It can be stated that these data points validate the estimated trend S-N curve with good agreement. From the trend S-N curve, σ a, PL = 720 MPa was selected as the stress amplitude for primary loading and its corresponding number of cycles to failure is represented by the black data point. In order to quantify the damage and its development, CATs were carried out at σ a, PL = 720 MPa up to specimen failure.