PSI - Issue 57

Gustav Hultgren et al. / Procedia Structural Integrity 57 (2024) 428–436 Hultgren & Barsoum / Structural Integrity Procedia 00 (2023) 000 – 000

434

7

potentially causes these slower fluctuations observed in the simulated values. Consequently, it would become possible 155 to disregard the rapid variations observed in measured geometrical traits, such as the weld toe radius, by employing a 156 low pass filter. This filtering approach might help mitigate the impact of fast variations in the radius estimate, which 157 could likely result from the challenges in radius estimates for small radii. 158 Concluding remarks 159 The present study highlights the potential struggle of directly utilising existing analytical equations on idealised 160 measurement data from real welds in the as-welded condition to determine the stress concentration factors. The 161 transferability of analytical relations to actual measurement data needs to be further investigated as the following 162 dissimilarities are currently seen between the two: 163 • The rapid variations in the measurements of the weld toe radius are not reflected in the simulation results, 164 where the oscillations are slower. This difference shows that a low pass filtering procedure could be used 165 for weld toe radius measurements. 166 • The oscillation disagreement between the geometrical traits and the simulated stress could be either that 167 the measurement error generally increases for small radii, that the effect of the surrounding weld profiles 168 smoothens out fast variations or a combination of the two. 169 Acknowledgements 170 HIAB AB is gratefully acknowledged for producing, scanning and testing the specimens presented within the study. 171 The material used in the experimental investigation was provided by SSAB, for which the authors are grateful. 172 Winteria AB is also acknowledged for providing its quality assurance system and continuous discussions regarding 173 weld quality aspects. 174 Funding 175 SSAB financially supported the research. Their support is gratefully acknowledged. Figure 5. Comparison between the probabilistic local stress and analytical expressions for the stress intensity factor. Values outside the equations' validity range are presented as black markers. The colour indicates the weld tow radius as shown in the legend. Note that the colour scale is not linear.

176 177 178