C-K5K5 Welds of plates and branches to HSS
PDF page 607 · AISC 360-22
Section K5 consolidates all the welding rules for plates and branch members to a main (or through) HSS member into one section.
Due to differences in the relative flexibility of a main member, loaded by a force normal or inclined to its surface, and a branch member carrying a membrane force parallel to its surface, the transfer of load along the length of a weld is highly non-uniform. Thus, sometimes only a portion of the total weld length is effective in an HSS connection.
One option to prevent progressive failure of the weld and establish ductile behavior of the joint for simple T-, Y-, and K-connections is to design welds at their ultimate strength to develop the branch member yield strength. This requirement is presumed to be satisfied when using matching filler metal and any of the prequalified joint details in AWS D1.1/D1.1M (AWS, 2020) for T-, Y-, and K-connections. For fillet welds loaded at 90° to the axis of the weld, it can be shown that this requirement is met—in most cases—if the effective throat of the weld is ≥ 0.95 times the branch member thickness, for branches with and 70 ksi (480 MPa) electrodes. The wall of a rectangular HSS branch that is loaded in tension is an exception, because the fillet weld directional strength increase factor does not apply (see Section J2.4), and in this case the required effective throat of the weld is ≥ 1.43 times the branch member thickness.
As an alternative option, welds to an HSS main member may be designed as “fit for purpose” to resist branch forces that are typically known in HSS truss-type connec- tions by using what is known as the “effective length concept.” Many HSS truss web members are subjected to low axial loads and in such situations, this weld design philosophy is ideal. However, the nonuniform loading of the weld perimeter due to the flexibility of the connecting HSS face must be taken into account by using weld effective lengths. Suitable effective lengths for plates and various HSS connections subjected to branch axial loading and/or moment loading, are given in Tables K5.1 and K5.2. These provisions are based on full-scale HSS connection and truss tests, supplemented by numerical modeling, that studied weld failures (Frater and Packer, 1992a, 1992b; Packer and Cassidy, 1995; McFadden and Packer, 2014; Packer et al., 2016; Tousignant and Packer, 2015, 2017a, 2017b, 2018, 2019).
Effective lengths used to determine weld sizing for rectangular HSS T-, Y-, and cross-connections with moments and for overlapped connections are based on a rational extrapolation of the effective length used for design of the branch member itself. Diagrams that show the locations of the effective weld lengths, most of which are less than 100% of the total weld length, are shown in Table K5.1 for connections to rectangular HSS and Table K5.2 for connections to round HSS.
For rectangular HSS, the effective length concept of weld design recognizes that a branch-to-main member connection becomes stiffer along its edges, relative to the center of the HSS face, as the angle of the branch to the connecting face and/or the width ratio (the width of a branch member relative to the connecting face) increase. Thus, the effective length used for sizing the weld may decrease as either the angle of the branch member (when over 50° relative to the connecting face) or the branch member width (creating width ratios over 0.85) increase. Note that for ease of calculation and because the error is insignificant, the weld corners were assumed as square for determination of the weld line section properties in certain cases.
When the welds in rectangular HSS overlapped joints are adequate to develop the strength of the remaining member walls, it has been found experimentally that it is permissible to eliminate the weld on the “hidden toe” of the overlapped branch, provided that the components of the two branch member forces normal to the chord substantially balance each other. The “hidden toe” should be fully welded to the
chord if the normal components of the two branch forces differ by more than 20%. If the “fit for purpose” weld design philosophy is used in an overlapped joint, the hidden weld should be completed even though the effective weld length may be much less than the perimeter of the rectangular HSS branch. This helps account for the moments that can occur in typical HSS connections due to joint rotations and face deformations, but are not directly accounted for in design.
The design weld size in all cases shown in Table K5.1, including the hidden weld underneath an overlapped member as discussed in the foregoing, is the smallest weld throat around the connection perimeter. Adding up the strengths of individual sections of a weld group with varying throat sizes around the perimeter of the cross section is not a viable approach to HSS connection design.