PSI - Issue 39

Wei Song et al. / Procedia Structural Integrity 39 (2022) 214–221 Author name / Structural Integrity Procedia 00 (2020) 000–000

219

6

(a)

(b)

1.0

1.0

p/t = 0 p/t = 0.1 p/t = 0.2 p/t = 0.3 p/t = 0.4 p/t = 0.45

Weld toe

p/t = 0 p/t = 0.1 p/t = 0.2 p/t = 0.3 p/t = 0.4 p/t = 0.45

Weld root

0.0 0.2 SED for weld toe ∆ W (Nmm/mm 3 ) SED for weld toe ∆ W (Nmm/mm 3 ) Weld toe 0.4 0.6 0.8 0.2 0.4 0.6 0.8 1.0 (c) 0.0

0.0 0.2 SED for weld root ∆ W (Nmm/mm 3 ) 1.0 (d) SED for weld toe ∆ W (Nmm/mm 3 ) Weld root 0.4 0.6 0.8 0.2 0.4 0.6 0.8 0.0

p1

p1

0.2

0.4

0.6

0.8

0.2

0.4

0.6

0.8

axial misalignment amplitude e/t

axial misalignment amplitude e/t

p/t = 0 p/t = 0.1 p/t = 0.2 p/t = 0.3 p/t = 0.4

p/t = 0 p/t = 0.1 p/t = 0.2 p/t = 0.3 p/t = 0.4

p1

p1

0

0.0

0

0.5

1

2

3

5

0

0.5

1

2

3

5

angle misalignment a (°)

angle misalignment a (°)

Fig. 4 Effect of weld penetration on SED values under different misalignments. (a) SED values at weld toe for different penetration ratios considering axial misalignments; (b) SED values at weld root for different penetration ratios considering axial misalignments; (c) SED values at weld toe for different penetration ratios considering angular misalignments; (d) SED values at weld root for different penetration ratios considering the effect of angular misalignments.

3.2. The expression of fatigue life considering the misalignment effect Fig. 5 shows the logarithmic S- N curves against the nominal stress range, ∆W, hot spot range and ENS range. Weld toe failure for axial and angular misalignments specimens and weld root failure was plotted in the same figure. The slope (m) is determined by all specimens and T σ is the ratio of fatigue strength with 10% and 90% survival probability. Comparing the fatigue class in Fig. 5(a), all the experimental test data above FAT63 which recommended in IIW standard, indicating the validity of experimental data and the reliability of the fatigue class, local details were not considered into nominal stress, and the points in the scattering band cannot illustrate the effect of characteristic parameters including dislocation. Lazzarin et. al summarized a large number of experimental data on LCWJs and proposed the SED dispersion band, m=1.5 in Fig. 5(b), which is satisfied with the experimental data. The maximum principal stress obtained from FEM is used for hot spot stress and effective notch stress, FAT100 and FAT 225 were plotted separately in Fig. 5(c)-(d), these methods illustrate the validity of using different evaluation parameters. Considering that the dispersion of the three methods is greater than the nominal stress, it may be caused by the measurement errors of individual specimens. Although there are different types of misalignments, the fatigue strength is within the allowable range of the design strength.