118

Figure 5.6.14 – Southwest Gusset (End of Test)

5.7

HSS-16 – Bolted Splice Connection

5.7.1

Specimen Overview

Specimen HSS-16 was a very unique test to not only the experiments detailed in this chapter but also to the prior tests (Johnson 2005, Herman 2007). This test used a one sided bolted connection to transfer the forces from the brace to the gusset plate shown in Figure 5.7.1. The potential benefit to using this connection would be that the brace would not have to be field welded to the gusset plate. The problem of fitting the brace in between the gusset plates would also be avoided. Instead, the brace would be shipped to the sight with the extension plate already welded on it, and then field bolted to the gusset plate. The bolts in this experiment were 1-1/8” A490 slip critical class A bolts with oversized holes in the extension plate and the gusset plate. Also, the extension plate is cut out

119

with a notch in it as proposed by Packer (2007), so that the brace will have its entire cross section for one third of the length of the connection to the extension plate, thereby eliminating the problem of net section.

Figure 5.7.1 – Specimen HSS-16 Connection Detail

The frame failed by fracture of the net section at the 7/8 inch extension plate. The applied load ranged from -146 kips to 301 kips for a total range of 447 kips. The drift was measured from -2.86% to 3.03% for a range of 5.89%. Table 5.7.1 shows the performance states that were noticed at the associated drift ratio. Figure 5.7.1 shows the hysteretic response.

120

Table 5.7.1 - HSS-16 Peak Results

Figure 5.7.2 – Specimen HSS-16 Force-Drift Response

121

5.7.2

Initial Drift Range (Max/Min Range from 0% to 1.25%)

Visible bending of the SW connection was noticed at a drift ratio of -0.16% shown in Figure 5.7.3. This displacement was located at the extension plate, not at the center of the brace where it typically occurs. Since hinging occurred at the extension plate, B1, B2 and BC were not calculated for this specimen. Figure 5.7.4 shows that the eccentricity of the one sided connection was too much for the extension plate to resist and a hinge formed here, between the end of the brace and the edge of the gusset plate. The extension plate and gusset plate continued to bend with added deflection shown in Figures 5.7.5 and 5.7.6. Hinge lines formed in the gusset plate at a drift ratio of -0.45% shown in Figure 5.7.6.

Figure 5.7.3 – Visual Bending of SW Gusset and Extension Plate (-0.16%)

122

Figure 5.7.4 – Bending of SW Gusset and Extension Plate (-0.29%)

Figure 5.7.5 – Bending of SW Gusset and Extension Plate (-0.37%)

Figure 5.7.6 – Bending of Plates, Hinge Lines on Gusset (-0.45%)

123

During the middle stages of the initial drift range, a crackling sound (similar to the sound of a light rain) was heard coming from the gusset plate. This was the same sound that was heard when the bolts were tightened during fabrication using load indicating washers. It is believed that the sound during the test was coming from these load indicating washers, at the interface of the gusset plate. In cycle 21 at a drift range of 0.31% and at a load of 206 kips, the bolts in the SW gusset plate connection slipped. This created a very load bang and the slip itself measured 3/8 inch from the displacement of the whitewash shown in Figure 5.7.7. A much softer bang came on the compression half of the cycle and was presumed to be a smaller slip back in the other direction.

Figure 5.7.7 – Bolt Slip (0.31%)

The following compression cycle showed severe damage to the base metal at a drift of -0.49%. A six inch crack formed at the SW gusset plate base metal at the weld to the column. This is shown below, in Figure 5.7.8.

124

Figure 5.7.8 – Base Metal Fracture (-0.49%)

Because of the base metal crack and added deformation demands, the gusset plate started to show additional yield lines. Y3 yielding was noticed in the gusset plate at the beam, column, and at the end of the extension plate. These lines appeared to be hinge lines where the gusset plate was hinging, however they could also have been a result of the onset of local buckling of the gusset plate at the beam which wasn’t noticed until the moderate drift range.

Figure 5.7.9 – Y3 of SW Gusset Plate (-0.49%)

125

5.7.3

Moderate Drift Range (Max/Min Range from 1.25% to 2.75%)

The yielding of the extension plate started to show through flaking of the whitewash at a drift ratio of 0.67%, as shown in Figure 5.7.10. State Y3 yielding was reached by this plate at a drift ratio of 1.08%.

Figure 5.7.10 – Y1 of SW Extension Plate (0.67%)

Initial yielding was spotted on both gusset plates for the first time while the brace was in tension. At a drift of 0.81%, the system reached an applied load of 223 kips causing the gusset plates to yield shown in Figure 5.7.11. At the same time, the SW gusset plate buckled at the beam shown in Figure 5.7.12. It is important to note that this plate buckled only after a long crack (13”) developed in the base metal of the SW gusset plate at the column (making the connection eccentric).

126

Figure 5.7.11 – Y1 of NE Gusset Plate (0.81%)

Figure 5.7.12 – B1 of SW Gusset Plate (0.81%)

5.7.4

Severe Drift Range (Max/Min Range > 2.75%)

The bolts on the NE gusset slipped with a loud bang at a load of 276 kips, at a drift ratio of 1.27% (Figure 5.7.13). Over the next couple of cycles, the frame continued to make loud banging sounds. These sounds were presumed to come from the bolted connections at the gusset plates slipping, but it was unable to determine exactly where the slips were occurring because the potentiometers were dislodged whenever the connection slipped. These loud sounds occurred only during the tensile half of the cycles. Yielding in the SW gusset plate reached a stage of Y5 at a drift ratio of 1.43% (Figure 5.7.14).

127

Figure 5.7.13 – Bolt Slippage at NE Gusset (1.27%)

Figure 5.7.14 – Y5 of SW Gusset Plate (1.43%)

Yielding became visible in the framing elements during this stage. At a drift ratio of 1.67%, yielding in the beam and column of the NE corner occurred shown in Figure 5.7.15.

128

Figure 5.7.15 – Y3 of NE Framing Elements (1.67%)

Throughout the test, the crack in the gusset plate base metal at the southwest column continued to increase in length. At a drift ratio of -2.00% this crack tore the entire way through the gusset to column connection (Figure 5.7.16). At -2.31% drift, the crack started to tear the weld (or possibley, the base metal) at the beam, next to the column, during the tension half of the cycle. This also caused considerable local damage to the beam flange at this location. Figure 5.7.17 shows this damage. This crack grew rather slowly after this point and reached a total of nine inches long by the end of the test at a drift ratio of -2.86%. The total weld length left connecting the gusset plate to the SW corner of the frame at this drift ratio was 16 inches.

Figure 5.7.16 – Complete Tear of Base Metal at SW Gusset (-2.00%)

129

Figure 5.7.17 – Crack Develops in Weld at Beam (2.28%)

A crack in the base metal at the weld connecting the NE gusset plate to the column measured two inches at -1.98% drift. At a drift ratio of -2.31%, the crack grew to nine inches long. Figure 5.7.18 shows this condition. At cycle 37 with a drift ratio of -2.66%, the crack measured 17 inches long, and at -2.86% drift, the crack fractured over the entire length of the gusset plate to column connection (Figure 5.7.19). A table summarizing the weld and base metal damage is shown in Table 5.7.2.

Figure 5.7.18 – Base Metal Crack at NE Column (-2.31%)

130

Figure 5.7.19 – Base Metal Complete Fracture at NE Column (-2.66%)

Table 5.7.2 – HSS-16 Weld/Base Metal Damage Summary

At -2.66%, even with the base metal at both of the gusset plates totally fractured at the column, the extension plate at the SW corner started to crack shown in Figure 5.7.20. This crack continued to propagate through the plate, first on the south side of the plate at a drift of -2.86%, while a smaller crack also developed on the north side of the plate. Both conditions are shown in Figure 5.7.21. The tension cycle after the compression cycle which caused the cracks shown in Figure 5.7.21, caused the plate to fracture the entire way though marking the end of the test at 3.31% drift.

131

Figure 5.7.20 – Crack in SW Extension Plate (-2.66%)

Figure 5.7.21 – Cracks in Extension Plate (-2.86%)

5.7.5

Specimen Summary

The drift ratio of 5.89% is the largest amount of displacement that any of the frames tested (including the tests from thesis 1 and 2) experienced. However, the test results are less acceptable for seismic design than most others observed in this research program, because the frame did not develop its design resistance. There is still a great

132

deal to learn from this test. To understand why this frame was able to withstand such a large drift capacity could lead to new designs incorporating some of the same ideas, and hopefully leading to frames with similar, or even higher, drift capacities. These ideas will be discussed in chapter nine. Some characteristics of the frame were sacrificed in order to attain such a large range of displacement. Most notably, the force range was much smaller than previous experiments. Since the connection buckled before the brace did (the connection was weaker than the brace), the frame had an overall lower compression capacity, -146 kips. With the bolts slipping, and the welds tearing in the initial drift range, the tension capacity was also reduced, to only 301 kips. This gives a range of 447 kips, which is more than 100 kips lower than HSS-12. Damage to the base metal of the gusset plate occurred very early in this experiment. The drift ratio at which it occurred was -0.49%. In performance based design, the performance levels of immediate occupancy and operational are appropriate guidelines for this drift ratio.

5.8

HSS-17 – 3/8” Tapered Plate

5.8.1

Specimen Overview

Specimen HSS-17 uses a 3/8 inch tapered gusset plate with 3/8 inch fillet welds for the gusset plate welds. This specimen has the same geometry as the other two specimens with tapered plates (HSS-10 and HSS-13) except the thickness of the gusset plate has changed from 1/2 inch to 3/8 inch.