Effects of global tensile residual stresses on the fatigue strength of welded high-strength steel joints

• Effects of global tensile residual stresses on the fatigue of welded joints. • Numerical welding simulations revealed differences under varied boundary conditions. • Experimental fatigue tests on welded joints under elevated mean stress conditions. • Successful inclusion of welding boundary conditions in fatigue assessment. Large-scale engineering as-welded structures typically generate high tensile residual stresses due to the structural stiffness restricting the contraction of weldments. These tensile residual stresses can often reach the yield strength of parent material. In small-scale specimens, similarly high tensile residual stresses locally at the weld toes can be achieved by appropriately selecting the joint type and using sufficiently large dimensions. However, it is not yet known whether different primary (global) and tertiary residual stress components significantly influence the local fatigue strength of welded specimens, or whether they should be considered in fatigue assessments. The objective of the current work is to experimentally and numerically investigate the residual stresses and fatigue strength of welded-around longitudinal gusset joints made of high-strength 690QT ship steel under various mean stress conditions, representing different boundary constraints in welded structures. First, welding deformations and residual stresses are numerically estimated using an experimentally validated heat source model and sequentially coupled finite element model. Herein, different boundary conditions are applied to study the evolution of local residual stresses and their components at the fatigue-critical weld. Secondly, fatigue tests are experimentally performed on similarly welded longitudinal gusset joints under different mean stress conditions that reflect the simulated residual stress fields. The numerical simulation results reveal clear differences in residual stress gradients and components across the different boundary conditions. Differences in the improvement factors between the constant minimum stress test series with respect to the constant mean stress condition were found to be 1.52 (no elevation) and 1.31 (elevated global mean stress). The findings of this work highlight the importance of considering boundary conditions in the evolution of welding-induced residual stresses, as well as their inclusion in fatigue assessments.