The surface stresses in the vicinity of the weld of a hopper corner joint model and a perpendicular joint model calculated by t x t shell models are compared with those calculated by fine solid models. Based on the results, selection method of a point to represent the weld toe in HSS evaluation is discussed. As results, the followings are found:
Under conditions chosen, solid stress of the solid model with weld beads at a distance of D from the weld toe agrees with shell stress at a distance of D from the plate surface intersection of the solid model without beads apart from the shell element nearest to the structural intersection.
Under conditions chosen, shell stress of the element nearest to the structural intersection underestimates solid stress substantially when shell and solid stresses are compared by the procedure of the preceding paragraph.
It is recommended to select the point directory below the intersection of the plate surfaces as the point to represent the weld toe in shell analysis. In this case, the stress of the nearest element to the intersection should not be used in stress extrapolation.
The approaches to the fatigue strength assessment have been further developed during recent years (e.g. Fricke, 2003). In addition to the conventional nominal stress approach, local approaches such as the (structural) hot-spot stress (HSS) approach have reached the stage of practical application. Regarding HSS approach, experimental and analytical procedures have been derived for its determination by extrapolating the structural stress outside the localized notch-affected zone to the weld toe (e.g. Niemi, 1995). In most classification society rules, HSS is evaluated by shell FE models in which the weld geometry is neglected. In these cases, special care is needed to select an appropriate point to represent the weld toe for stress extrapolation.
