Experimental study on slotted rectangular and square hollow structural section (HSS) tension connections

Abstract

An experimental study has been carried out on slotted hollow structural section (HSS) connections with and without welding at the end of the gusset plate. A total of 30 slotted square and rectangular HSS specimens, with 16 different connection configurations, were investigated for effects of weld length ratio, slot orientation, gusset plate thickness, slot opening length, and end welding. Experimental results show that the net section efficiency of the specimen with no end welding was greater than unity for a weld length ratio (L/w), where L is weld length and w is the circumferenctial distance between the longitudinal welds, as low as 0.8, and full net section efficiency was achieved with L/w ratio as low as 0.76 for a square HSS specimen with end welding. These results support findings from other research that show provisions to account for the effect of shear lag in slotted HSS connections are overly conservative in both Canadian and American design standards. Recommendations are provided to improve on provisions to account for shear lag in the Canadian design standard.

Analysis of the Shear Lag Factor for Slotted Rectangular HSS Members

“…The geometric and material variables for the test specimens are listed in columns 2 through 8 of Table 1, and the experimental rupture load, P e , is listed in column 9. The specimens tested by Zhao et al (2008), Korol et al (1994), andYeomans (1993) were loaded statically, and the specimens tested by Han et al (2007) and Yang and Mahin (2005) were loaded cyclically to simulate seismic loading.…”

“…The relevant geometric parameters for slotted HSS connections are the wall thickness and weld length. Wall thickness measurements for the 30 ASTM A500 Grade C specimens tested by Zhao et al (2008) resulted in a mean measured-to-nominal thickness ratio of 0.924. Using a design wall thickness equal to 0.93 times the nominal wall thickness according to AISC Specification Section B4.2, the mean measured-to-design thickness ratio is 0.994, with a coefficient of variation of 0.00710.…”

“…Because the same parameters are relevant for the tensile rupture limit state, V G = 0.050 was used for the analysis in this paper. The slightly more conservative value of ρ G = 0.994, which was based on the wall thickness measurements by Zhao et al (2008), was used in the analysis.…”

“…For five of the projects discussed in this paper (Zhao et al, 2008;Han et al, 2007;Yang and Mahin, 2005;Zhao et al, 1999;Zhao and Hancock, 1995), 20 data points are available with coupons extracted from the flat portions of the HSS walls. These tests-on ASTM A500 Grade B, ASTM A500 Grade C, and similar international gradesresulted in ρ M = 1.12 and V M = 0.0405, with only a small variation between grades.…”

Structural behaviour of inner set bolted joints in steel tubes under tension

“…Cheng and Kulak (2000) also suggested that the reduction of the effective net area (due to shear lag) would be eliminated for slotted CHS connections if a minimum weld length ( L w ) of 1.3 times the tube diameter ( D ) was used. Further experimental studies were performed (Wilkinson et al, 2002) on gusset plates slotted into RHS, including those by Martinez and Packer (Martinez-Saucedo et al, 2006; Packer, 2010) on slotted CHS gusset plate connections as well as those by Willibald and Packer (Willibald et al, 2006; Zhao et al, 2008) and Martinez-Saucedo and Packer (2009) on slotted HSS (hollow square section).…”