PSI - Issue 75

Jörg Baumgartner et al. / Procedia Structural Integrity 75 (2025) 120–128 Jo¨rg Baumgartner / Structural Integrity Procedia 00 (2025) 000–000

122

3

Reference to ISO6520-1

Imperfection designation

Limits for imperfections for quality levels

No.

t (mm)

C63

B90

B125

1.12 505

Incorrect weld toe, weld toe angle for fillet welds

≥ 0 . 5 see Table 1 see Table 1 see Table 1

Incorrect weld toe, weld toe radius

– 5052

> 3 not specified

≥ 1mm

≥ 4mm

Table 2: Definition of incorrect weld toe and weld toe radius according to Annex in ISO 5817 [5]

Fig. 1: Endurable nominal stresses of a transverse sti ff ener with di ff erent weld profile under 3-point-bending load [19]

for digitizing weld surfaces: laser line scanners (LLS) and fringe light projection (FLP) sensors. While LLS systems deliver single 2D scans that can be stacked into a 3D image, FLP systems generate a consistent 3D image within a limited area. There is currently no standardized method for evaluating geometrical parameters from 2D or 3D scans. A recent round-robin study involving 20 participants examined the determination of the weld toe radius r and weld toe angle α using 3D scans. The results revealed significant deviations among participants for welded joints [16], while lower deviations were observed for idealized specimens with simpler geometries [12]. Investigations by [7] indicate that di ff erent algorithms for calculating these parameters yield comparable results, provided the same range of points is selected. This suggests that the determination of the Area of Interest (AoI), i.e., the weld toe region, is crucial, as it is likely the primary factor influencing the results of the measurement of these local geometrical parameters. For this reason, this investigation primarily focuses on two aspects: first, the analysis of welded joints within larger 3D scans; and second, the determination of the weld toe based on 2D profiles extracted from 3D scans.