PSI - Issue 75

Lewis Milne et al. / Procedia Structural Integrity 75 (2025) 419–425 L.Milne et al. / Structural Integrity Procedia 00 (2025) 000–000

420

2

ical for Very High-Cycle Fatigue (VHCF) regime. Above mentioned scatter of fatigue data is usually associated with the transition from HCF to VHCF. Di ff erent types of transition from the surface failure mode (HFC) to the internal failure mode (VHCF) are discussed by Shiozawa et al. (2001), who suggested a classification of the corresponding SN curve shapes using the concept of duplex SN curves. In the beginning of this research, comprehensive ultrasonic fatigue (USF) testing results have been published by Gorash et al. (2023) for S275JR + AR with insights into corrosion, mean stress and frequency e ff ects. Recently, the summary of the research outcomes were published by Gorash et al. (2025) with focus on the fatigue performance comparison of the steels S355JR + AR and S275JR + AR considering welding and surface e ff ects, specifically welding porosity and pitting corrosion. The availability of data on the fatigue behaviour of welded joints in the VHCF regime is limited was shown in the review by England et al. (2023). Also USF testing has been proven by England et al. (2023) to be a viable method to study the VHCF of welded joints in a reasonable timeframe. An important progress for USF testing of structural steel welded joints has been achieved by designing the sample that includes a realising joint geometry with a weld toe. It was applied to study the VHCF behaviour of S275J2 + N flux-core arc welded joints by England et al. (2025) using the USF testing method. An important part of the accelerated fatigue testing is the evaluation of the frequency e ff ect, which is critical for a wider adoption of the USF testing method as a reliable approach to investigate fatigue. The frequency e ff ect is a commonly encountered challenge in USF testing of low carbon steels, that was investigated by Milne et al. (2024) using two comparable grades of ferritic steels Q355B and S355JR + AR. More recently this investigation extended to include other steel grades (S275JR + AR, S275J2 + N and C45) and establishing the relation between frequency e ff ect and various material attributes like ferrite content, yield strength, UTS, etc. So this paper will report on the further progress in understanding and evaluation of frequency e ff ect with focus on recently obtained experimental results for high strain rate tensile testing of structural steel grades. The key factor that has previously been proposed by Milne et al. (2024) to cause the increase in fatigue resistance at ultrasonic frequencies is the increase in strain rate with increasing test frequency. Using the relations between sinusoidally applied force, cyclic stress and strain, we can evaluate that for a fatigue test of a material with an arbitrary stress amplitude and elastic modulus of 300 MPa and 210 GPa respectively, ˙ ε would be 0.127 s − 1 at 20 Hz and 127 s − 1 at 20 kHz. This shows that the strain rate during USF testing is significantly higher than at conventional frequencies and is well within the range where dynamic influences may take e ff ect. The influence of strain rate on the properties of Body-Centred Cubic (BCC) metals, such as ferrite, was previously described by Mughrabi et al. (1981). As the strain rate increases, the lattice friction stresses within the material will increase, which limits the thermally activated glide of dislocations. As such, a higher flow stress is necessary to overcome these frictional stresses and enable dislocation glide. It is therefore clearly important to account for the influence of the strain rate on the yield strength when testing BCC metals, such as ferritic steels, at increased frequencies. To investigate the influence of strain rate on the material properties of the steels of interest, high strain rate tensile testing was carried out in conjunction with the Henry Royce Institute in Manchester, using a Zwick HTM 5020 High speed Testing Machine with the specification available from Zwick Roell Group (2024), which is capable of carrying out tensile tests from 0.2 m / s to 20m / s. Other important features are the e ff ective piston stroke of 250 mm, system pressure of 280 bar and high-speed data acquisition with 80 MHz. High strain rate tensile tests were successfully carried out for the S275JR, S355JR, S275J2 and C45 materials following the procedure in the standard ISO 26203-2 (2011). Specimens were a custom dual gauge length dogbone specimen, as mandated by Henry Royce Institute and shown in Fig. 1a. Strain was measured using Digital Image Correlation (DIC). A Photron SA1.1 FastCam high-speed camera was used with a Nikon 24-85mm zoom lens. Two high-power Hedler Profilux LED lamps were used to light the specimen. The frame rate captured by the camera was varied from 10000 to 20000 fps, along with the resolution of the capture, depending on the rate of the test. For high-speed photography, the widest possible aperture was used with f / 3.5 at the wide end (24mm) and f / 4.5 at the telephoto end (85mm). The camera was then kept at a fixed distance and zoom level throughout all of the tests, to ensure this calibration did not vary. An image of the test set-up is shown in Fig. 1b. To enable DIC analysis of the specimens, a speckle pattern was painted on the specimen’s surface shortly before testing. A white base layer was applied using Dupli-Color Aqua Lackspray paint, followed by a black stochastic 2. Specimens, test set-up and procedure