Reply to discussion by S. Guravich and S. Skarborn on “Double-angle shear connections with short outstanding legs”Appears in the Canadian Journal of Civil Engineering, 35(8): 786–795.

The authors thank S. Guravich and S. Skarborn for their thoughtful discussion. First, the authors would like to emphasize that while the bending of the outstanding legs is a significant contributor, it is not the only source for the rotational ductility of double-angle shear connections. Shortening of the outstanding legs does reduce the rotational capacity of the connections. However, for the tested connections in the study, they still possessed a rotational capacity at the level of 0.03 radians. As pointed out by S. Guravich and S. Skarborn, column face widths less than 203 mm may be affected. The paper gave 152 mm as a column face width that would definitely create the problem rather than may create the problem. For face width 203 mm, the fillet welds can often run onto the corner radius of the hollow structural section column to eliminate the need for shortening the outstanding legs. The initial thought for group C specimens to adopt filled flush flare bevel groove welds (FBGW) was to reflect such a fact that ‘‘in general, there is a tendency for flare groove welds to be oversized and filled flush’’ (quote from Packer and Frater 2005). This tendency was likely due to the following two reasons: (1) the CISC Handbook (CISC 2006) only listed filled flush for flare bevel and flare V-welds in its Welding Practice in Part 6; and (2) the concept of a flare bevel fillet weld was introduced for the first time only recently in CSA W59 (Packer, and Frater 2005) and in Supplement No.1 (January 2005) to CSA-S16-01 (CISC 2006). Some extra work on the FBGWs was reported by the first author in a subsequent paper (Gong 2008). For the tested connections, the average corner radius was 22 mm, which was considerably less than the maximum value 36 mm given by the CISC Handbook. The subsequent study found that the weld face width was a better characteristic dimension than corner radius to predict the effective throat of the FBGWs. The authors believe the groups A and B specimens, which used fillet welds, also provided useful information for the design of FBGW using unaltered outstanding legs. For example, corresponding to 6 mm fillet weld, the required effective throat for FBGW is 4.3 mm. To avoid further stress concentration, a weld return can be made at the top of the angles and then ground. In addition, the designer has the option to increase the effective throat of the FBGW. The unique merit of the unaltered outstanding leg is that it is able to provide extra space for stronger FBGW, if necessary. The authors agree that the paper would have been enhanced if additional connections using smaller size flare bevel fillet welds had been tested. The authors will consider doing additional testing subject to financial support being available. As for other preferred solutions, such alternative connection types were beyond the scope of the paper.