PSI - Issue 75

Ralf Glienke et al. / Procedia Structural Integrity 75 (2025) 474–488 Ralf Glienke et al / Structural Integrity Procedia 00 (2019) 000 – 000

476

3

strength already differs in the FAT values (112 or 125, or C1 or C), but also in the slope parameters for the S-N curve (m 1 = 3, 3.45 or 5) as well as in the thickness reduction factor with exponents (n = 0 or 0.05 or 0.1). For conventionally transverse loaded butt welds, a modified thickness reduction may be applied according to some design codes, if the distance between the weld transition points L T is known at the design phase. According to updated DNV-RP-C203 (2024), a two-slope S-N curve exists for high quality butt welds with a flatter slope parameter (m 1 = 3.45) and a knee point at N = 3 ∙ 10 6 . Current findings indicate that the size effect according to EN 1993-1-9 (2010) is regulated too conservatively, especially as the linear misalignment is limited to 2 or 3 mm at maximum, see Glienke et al. (2024b), Bartsch et al. (2024), Seidel (2025). For thermally cut edges , DC 125, 140 or 160 applies depending on the standard, whereby machine-controlled cutting without repair welding is assumed. However, the fatigue data base from EN 1993-1-9 partially refers to the 1950s and 1960s, whereby the condition of the edges and the material strength are often not documented, see Feldman et al. (2019). The fatigue strength of the free edges depends on the residual stresses at the cut edge, the hardness, and the surface roughness, see Diekhoff et al. (2020). More recent investigations from Grimm et al. (2024), Lipiäinen (2022) and Schröder et al. (2025) show the potential of post-treatment and the consideration of edge quality as well as steel grade for fatigue strength improvement. In contrast to Eurocode 3, this is already considered in EN 13001-3-1 (2019). For non-welded constructional details (flange neck of a seamless rolled ring flange, holes in the tower wall or base material), the standards can only provide lower limit values for the detail categories, as these must already be valid for steel grade S235. IIW Recommendations (2024) allow a higher FAT class to be used if an applicable code is available or fatigue tests are carried out. The FKM-Guideline (2020) consistently follows a different approach. The fatigue strength verification of welded joints is based on the nominal stress concept with classified constructional details or with the use of hot spot or effective notch stresses. For non-welded constructional details, a material-based approach is used, which calculates the amplitude of the fatigue component limit for a specific mean stress at N D =10 6 based on the material fatigue limit at zero mean stress, considering the component design (notch effect and support effect) and manufacturing (surface roughness, corrosion protection, surface treatment factor). DC 160 and 180 apply to the base material in accordance with EN 1993-1-9 (1 st or 2 nd generation). In contrast to the material fatigue limit (FKM-Guideline (2020)), these values were derived for real components (beams) with rolled inclusions, as well as the possibility of falling tools during assembly or markings for tracking during production (Hobbacher (2010)). For components with holes subjected to normal stress , DC 90 (m 1 = 5) applies according to (FprEN 1993-1-9 (2024)), regardless of the hole size (local stress gradient) and the component dimensions for steel grades from S235 upwards. If the holes are not drilled, i.e. laser-cut or punched, DC 50 (m 1 = 3) applies. Recent articles (Bartsch et al. (2023), Glienke et al. (2024c), Kalkowsky et al. (2025a), Kalkowsky et al. (2025b)) show the potential for the fatigue strength assessment of non-welded components based on the FKM approach. Already the consideration of the steel grade S355 in the approach, which is the focus of application for wind turbine towers, would reduce conservatism compared to Eurocode 3 design. Higher fatigue resistances can be expected in tower production with fully mechanised welding processes, which lead to higher seam qualities, machine-guided thermal cutting with post-treatment and consideration of the material strength in the design of components with non-welded constructional details. As the tower section is blast-cleaned internally and externally after manufacture, the question of the effect of blast-cleaning on fatigue strength arises. 2.2. Post-weld treatments and modification of the S-N curve The use of post-weld treatment processes aims to extend the crack initiation life of welded joints. Thermal methods are used to reduce residual tensile stresses (stress-relief annealing) or to reduce the notch effect (TIG dressing of the seam transition). In addition to reducing the geometric notch effect (grinding), mechanical surface treatments aim to generate residual compressive stresses and/or work hardening at the potential crack location. This is achieved by shot peening or high frequency mechanical impact (HFMI) (Nitschke-Pagel et al. (2006)). As a result of successful post weld treatment, the S-N curve is rotated towards flatter slopes and thus to higher fatigue resistances, see Fig. 2. The improvement in fatigue strength is given by the factor k Imp and can be applied to nominal, hot spot and notch stress detail category. The post-weld treatment processes are considered differently in the various standards, whereby only stress-relief annealing and HFMI post-weld treatment are included in the 2 nd generation of EC 3 (FprEN 1993-1-9