PSI - Issue 13

Ann-Christin Hesse et al. / Procedia Structural Integrity 13 (2018) 2053–2058 Ann-Christin Hesse et al./ Structural Integrity Procedia 00 (2018) 000 – 000

2054

2

1. Fatigue assessment of welded joints

For the fatigue assessment of welded joints, the nominal stress concept is a widely used approach. The fact that the concept is comparably easy to use and has proven its reliability in the past have contributed to the success of the nominal stress method. The method uses the stress in the net-section area in front of the weld. Standards provide a list of weld details and corresponding fatigue resistance values. The IIW fatigue recommendations for example use so called FAT-classes which correspond to the fatigue stress range at N = 2∙10 6 load cycles, at a survival probability of 97.7% and a slope of m = 3 (normal stresses, load cycles n < 1∙10 7 ). The FAT classes, according to Hobbacher (2009), were established using mainly arc welded samples but they apply to all fusion-welded components, which include arc welded as well as beam welded components, although the resulting weld geometry differs significantly. An alternative to the nominal stress method is the notch stress approach. This approach uses the highest elastic stress at the weld root or toe. In a finite element model the weld toe and weld root are substituted by a rounded shape with a defined reference radius. This is done to avoid infinite stress results. The first reference radius of 1 mm was proposed by Radaj (1990) and took micro-structural support effects in steels into account. Nowadays, other radii like 0.05 mm or 0.3 mm can be used as well as shown in Fricke (2008), depending on the sheet thickness. The notch stress concept has the advantage that weld details and components can be assessed, which are not incorporated in the nominal stress concept. Furthermore, the influence of different parameters of a welded joint on the fatigue strength (e.g. different alignments or incompletely filled grooves) can be estimated. The butt joint was chosen as a representative weld detail for the tests. This detail is frequently used in beam welded constructions. In the nominal stress concept, for examples in the IIW guidelines, this weld detail is represented with three different FAT-classes, depending on the execution of the weld seam and the weld geometry. If welding is performed from both sides, FAT 90 or FAT 80 apply. To use FAT 90, the joint has to be manufactured in flat position, the axial misalignment of the joint has to be lower than 5% of the sheet thickness and the weld reinforcement has to be lower than 10% of the sheet thickness. If those conditions are not fulfilled, FAT 80 can be applied if the axial misalignment is lower than 10% of the sheet thickness. If the butt joint is only welded from one side, FAT 71 can be applied if the root is checked by appropriate NDT methods. Hobbacher (2009) For the fatigue tests, the influence of different parameters (welding process, steel grade, sheet thickness t and axial misalignment) were investigated. For welding, laser as well as electron beam welding was incorporated in the test matrix, see Table 1. 2. Experimental

Table 1: Overview over the experimental matrix

No.

Welding process

Ax. Misalignment

S355J2+N

S960Q

3 mm

6 mm

10 mm

6 mm

10 mm

● ●

● ● ● ●

○ ○ ● ●

● ● ● ●

○ ○

2a 2b 3a 3b

Disk laser Disk laser

None 0.15∙t None 0.15∙t

Electron beam Electron beam

- -

- -

● Full S - N curve (at least 15 samples)

○ Reduced S -N curve (7 to 10 samples)

Table 2: Strength values of the base material determined by tensile tests S355J2+N S960Q 3 mm 6 mm 10 mm 6 mm 10 mm

Yield strength R eH resp. R p0,2 [MPa]

372 514

460 571

422 551

1000 1032

1021 1061

Ultimate strength R m [MPa]