PSI - Issue 57

J. Baumgartner et al. / Procedia Structural Integrity 57 (2024) 618–624

620

J. Baumgartner et al. / Structural Integrity Procedia 00 (2023) 000–000

3

material thickness and the failure location; weld toe or weld root, Figure 1. Since no support e ff ects are considered when maximum stress for radii r ref 1 mm are used [5], the FAT-classes depend in this case on the notch type, resp. opening angle. In addition, a di ff erentiation between stresses normal to the weld and shear stresses was added. A table for all three reference radii r ref = 1 . 0 mm , r ref = 0 . 3 mm and r ref = 0 . 05 mm is provided.

r ref

a)

1.00 mm

1

prior recommended for use

0.30 mm

2

0.05 mm

3

t

optionally correction necc. due to potential change of load path

1

2

3

5

10

20

50 mm

30

r ref

b)

1.00 mm

1

Applicable, but larger radii recommended

0.30 mm

2

0.05 mm

3

t

1

2

3

5

10

20

30 50 mm

Fig. 1. Choice of the reference radius depending on the sheet thickness and failure location

In addition, the slopes of the design S-N curves of welded thin joints are modified. Evaluations of fatigue data show [2, 6] that in average, the slopes values k σ and k τ increase with decreasing thickness. Therefore, for thicknesses t < 7 mm, shallower slopes of k σ = 5 and k τ = 7 are recommended.

2.3. Post-weld treated joints

Two additional improvement techniques have been included in the IIW-recommendations: Weld profiling and HFMI-treatment. Weld profiling is similar to weld toe grinding; however, the whole weld is mechanically ground [9]. This leads, in comparison to weld toe grinding, to higher endurable stresses, since the typical undercut and subsequently a stress raiser is avoided. For the nominal stress approach, a bonus factor of 1.4 is recommended, compared to a factor 1.3 for weld toe grinding. For HFMI-treated joints, the recommendations from [3] are integrated in a compact version, allowing an increase in endurable stresses depending on the material strength. In addition to the bonus factors that lead to an increase in endurable stresses, the slope of the S-N curve is changed depending on the improvement technique. For weld toe grinding, weld profiling and TIG-dressing, the slope is set to k σ = 4, for HFMI-treated joints, the slope is set to k σ = 5. The shallower slopes lead to a pronounced increase in fatigue strength at high number of cycles ( N > 2 · 10 6 ). For all improvement techniques, the increase in FAT-classes is given for all three stress-based approaches, nominal, structural and notch stresses.

2.4. Influence of wall thickness

It is well known that the endurable nominal and structural stresses decrease with increasing wall thickness, e.g., [12]. This e ff ect is typically covered by power function with an exponent n that depends on the specimen type. The influence of the wall thickness on the endurable nominal and structural stresses is now better captured for transverse attachments and cruciform joints. The basis for the improved assessment is a data collection evaluated in