PSI - Issue 57

Mladen Lukić et al. / Procedia Structural Integrity 57 (2024) 550 – 559 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

557

8

the conditions of application of the method is given. This local method requires the stress to be computed by FE analysis, the modelling rules and analysis being given in prEN 1993-1-14 (2022). For this method, the weld toes and weld roots are explicitly modelled by FE analyses as shown in Figure 8. To account for the variation of the weld shape parameters, and non-linear material behaviour at the notches, the actual weld contour at the weld toe and weld root is replaced by idealized shapes. Only welded details with plate thicknesses t ≥ 5 mm are covered by this method to avoid the weakening of cross sections of plates in the case of thin plates. Additionally, the fatigue verification of parent metalaway from weld toes or of welded details with mild notches is excluded where a mild notch is defined as having an ENS less than 2 times the nominal stress. This can for example be the case at weld toes in transverse butt welds.

Fig. 8. Weld angle and rounding of different weld types: (a) butt weld, (b) fillet weld. Example of a fillet weld on the right.

As notch stresses can be represented either by principal stresses or by von Mises stresses, both design alternatives according to each hypothesis are given: • Principal stress (PS), as the maximum difference of the principal stress range; • Equivalent von Mises (EVM), as the maximum von Mises equivalent stress range calculated from the range of notch stress components. The range of notch stress components is used to consider any partial compressive stress state that would otherwise not contribute to the range, due to the strictly positive nature of the von Mises stress. To stay coherent with the resistance and verification, each method has its fatigue resistance curve (both with slope coefficients m 1 = 3 and m 2 = 5) 6. Recommendation for magnification and stress concentration factors (Annex D) When determining the nominal stresses, all additional stress raising effects not considered in the detail category tables have to be taken into account. Annex D contains recommendations for magnification factors k 1 and stress concentration factors k f for the calculation of modified nominal stresses for selected stress raising effects. The magnification factors k 1 for lattice girders made of hollow sections considering effects from secondary moments have been moved into Annex D of prEN 1993-1-9 (2022) for ease of use. Recommendations for stress concentration factors k f for transitions in width or thickness of flanges in I-section girders are given as shown in Figure 9. These factors take the additionalstress raising effects due to plate membrane stresses and plate bending stresses into account. For the application of the factors k f , additionalrequirements for the welding between web and flange, formulated in Annex D, are to be met. The given stress concentration factors k f only cover the stress raising effects for girders without cope holes. In case of girders with cope holes, further investigations have to be performed, e.g. in combination with prEN 1993-1-14 (2022). Furthermore, recommendations for stress concentration factors k f are given for constructional details with transverse butt welds of plates that exhibit thickness transitions and no vertical support of the plates. Stress raising effects resulting from misalignment due to fabrication ≤ 5% of plate thickness are already considered by the classified constructional details of the tables for the nominal stress method. In case of a further intentional eccentricity, the stress raising effects have to be taken into account by appropriate nominal stress modification, such as the given stress concentration factors k f .