Design of Single-Sided Fillet Welds Under Transverse Loading
Date
2021-09-01T14:26:37Z
Authors
Thomas, Justin
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In North American steel design codes, a “directional strength enhancement” (or “sinθ”) factor is used to increase the predicted strength of fillet welds subjected to tension-induced shear. CSA and AISC code committees have expressed concerns about this factor being potentially unsafe for single-sided fillet welds; however, only cautionary, but vague, statements to address this issue exist in codes.
An experimental program was therefore developed at Dalhousie University to test 40 single-sided fillet welds in cruciform connections subjected to branch axial tension. Using ultimate loads, a first-order reliability analysis was performed to determine the inherent safety index (β) of North American fillet weld design procedures. The results show that current provisions meet/exceed the target safety index (i.e. β = 4.0) specified by North American codes (e.g. CSA S16 and AISC 360) provided that: (i) the “sinθ” factor is not used and (ii) tension at the weld root is avoided.
Description
In North American steel design codes, a “directional strength enhancement” (or “sinθ”) factor is used to increase the predicted strength of fillet welds subjected to tension-induced shear. CSA and AISC code committees have expressed concerns about this factor being potentially unsafe for single-sided fillet welds; however, due to a paucity of physical tests on such welds, only cautionary, but vague, statements to address this issue exist in codes.
An experimental program was therefore developed at Dalhousie University to test 40 single-sided fillet welds in cruciform connections subjected to branch axial tension. The connections varied the fillet weld size, branch-plate thickness, and loading eccentricity, to investigate the effects of these parameters on the weld strength. Using ultimate loads, a first-order reliability analysis was performed to determine the inherent safety index (β) of North American fillet weld design procedures over a practical range of live-to-dead load (L/D) ratios for elements in steel buildings (0 ≤ L/D ≤ 3). The results show that current provisions meet/exceed the target safety index (i.e. β = 4.0) specified by North American codes (e.g. CSA S16 and AISC 360) provided that: (i) the “sinθ” factor is not used and (ii) tension at the weld root is avoided. A new theoretical model for single-sided fillet welds is also introduced to account for weld eccentricity (i.e. induced stress due to the bending of the weld).
Keywords
Structural engineering, Steel structures, Connections, Welding, Codes, Design