PSI - Issue 38

Antti Ahola et al. / Procedia Structural Integrity 38 (2022) 457–465 Ahola et al. / Structural Integrity Procedia 00 (2021) 000 – 000

463 7

(a)

(b)

1000

1000

100 Nominal stress range Δ σ nom (MPa)

180 140 160 125 112 100 FAT

100 Nominal stress range Δ σ nom (MPa) SSFW SSFW Box girders: I sections:

DSFW (K-butt) DSFW SSFW (Beveled)

FAT80

Gurney and Maddox (1973) Knight (1979)

10

10

10 5

10 4

10 6

10 5

10 4

10 6

Cycles to failure N f

Cycles to failure N f

Fig. 6. Experimental CA fatigue test data for the longitudinal welds: (a) UHSS (S960) grade I section and box girder specimens with the longitudinal SSFW and DSFW (Skriko et al., 2021), and (b) I section specimens with double-sided intermittent welds from the studies by Gurney and Maddox (1973) and Knight (1979). 4. Discussion In this study, an overview on the fatigue strength capacity of SSFW joints in the transverse (non-load-carrying) and longitudinal load configurations was carried out. Experimental data from the published literature was extracted for the study, and the findings were supplemented with the new experimental fatigue test results on the intermittent welds and DSFW joints made of S355 mild steel grade. In the current design standards, neither the design category nor failure mechanism of the SSFW joints have been widely assigned although the FAT80 class is proposed in the design standards for the transverse attachment joints (DNVGL-RP-C203, 2016; EN 1993-1-1, 2005; Hobbacher, 2016). The crane standard EN 13001-3-1 (2018) differentiates between the SSFW and DSFW joints by giving a class FAT80 for SSFW, and FAT90 for DSFW joints. The experimental data showed that the SSFW joints failed from the weld root, while the DSFW joints fail from the weld toe. In the SSFW joints, the weld root has higher stress concentration than the weld toe due to the flank angle, and weld root is thus the critical location of these joints. In the S355 grade, the DSFW joints had reasonably high fatigue strength, Fig. 3a and Table 1 (in the specimens, an elastic pre-alignment during the welding was applied to compensate welding angular distortions). In the S1100 specimens, the difference in the fatigue strength capacity of SSFW and DSFW was lower, approximately one FAT class. Based on these findings, the recommendation by the crane standard for the SSFW and DSFW joints is justifiable. In the transverse weld configurations, intermittent welds can be applied without a significant decrease in the fatigue strength capacity in the case of non-load-carrying joint based on the experimental data (see Fig. 4). Melaku and Jung (2017) investigated the vertical web stiffeners of box girders welded with staggered and chained intermittent DSFWs employing the hot-spot stress method. In their study, staggered intermittent welds resulted in lower hot-spot stress than the chained welds, however both intermittent welds resulted in higher hot-spot stresses than continuous fillet welds. In the context of this paper, only intermittent SSFW joints were fatigue-tested, and further experimental verification for intermittent DSFW joints should be carried out. In the longitudinally loaded intermittent fillet welds, a clear drop in the fatigue strength capacity was obtained based on the experimental results, see Fig. 6b. This is an expected results as intermittent welding produces high stress peaks in the start and stop positions of fillet welds, and should be thus avoided at least in the critical locations of structures (Blodgett, 2008). Furthermore, in the intermittent welds, the consideration of functionality against other failure mechanisms, such as stability of welded sheet between in the intermittent welds is important (Khedmati et al., 2007). When comparing the results of SSFW and DSFW in the longitudinal load-carrying welds, no significant difference was found although in the design standards, lower fatigue