PSI - Issue 57

Lewis Milne et al. / Procedia Structural Integrity 57 (2024) 365–374 Lewis Milne et al. / Structural Integrity Procedia 00 (2019) 000 – 000

366

2

data in the VHCF regime is prohibitively costly and time consuming using conventional fatigue testing methods at 10-100Hz, and as such accelerated fatigue testing methods are necessary to produce cost-effective VHCF data. A commonly used accelerated fatigue testing method is ultrasonic fatigue testing (UFT), which resonates test specimens at 20kHz allowing for fatigue data to be produced up to 1000x faster than using conventional test methods. This rapid fatigue testing is known to bring some challenges, however. Most notably, the increased test frequency leads to a phenomenon known as the frequency effect, where the increased testing rate and reduced test duration can influence the observed fatigue results. For low-carbon steels, the fatigue life observed through UFT is typically far longer than the fatigue life at standard frequencies, due to the strain rate sensitivity of the dislocation glide mechanisms within the body-centred- cubic (BCC) α -ferrite regions (Mughrabi et al. 1981). Klusák et al. (2021) observed an increase in fatigue life of 100x at equivalent stress amplitudes for S355J2 steel, whereas Bach et al. (2018) observed an increase in fatigue limit of 58% at ultrasonic frequencies for C15E steel. Similar discrepancies are commonly observed across a number of low-carbon steels in literature (Gorash et al. 2023; Nonaka et al. 2014; Tsutsumi et al. 2009). Until the effect of frequency on the fatigue behaviour can be fully quantified, the usability of UFT to evaluate VHCF data for BCC materials remains limited. Attempts to quantify this frequency effect have thus far been limited, however some investigations have been carried out: Gorash et al. (2022) evaluated the average discrepancy between the finite life region of SN curves produced at 20kHz and 20Hz for the structural steel S275JR. This provided a simple, but crude method of evaluating the strain rate sensitivity for a given material. Bach et al. (2018) proposed a frequency sensitivity parameter, , based on the Hart strain rate sensitivity equation (Hart 1967). The definition of this proposed parameter, as presented in (Bach et al. 2020) is given in equation 1, where is the stress amplitude and is the test frequency. Using this parameter, the frequency sensitivity for a range of ferritic-pearlitic carbon steels was evaluated, and it was observed that the frequency sensitivity appears to correlate strongly with the ferrite volume % (Bach et al. 2018). = ∆ ln( ∆ln ) (1) The aim of this investigation is therefore twofold: Firstly the VHCF curves for two commonly used equivalent structural steels grades, S355JR and Q355B, will be evaluated through the use of UFT and compared to the corresponding conventional frequency fatigue curves. Secondly, the models proposed above will be adapted in order to produce methods which can be used to quantify the frequency sensitivity and to try to relate the UFT data to the conventional frequency fatigue data for the two materials.

(a)

(b)

Figure 1 – Micrographs taken at a magnification of x500 for (a) Q355B and (b) S355JR