France, Davison, and Kirby 1999
Three series of experiments on beam-to-column connections of HT and CFT columns were presented in a series of papers (France et al. 1996, 1999a, 1999b, 1999c). The girders were connected to the steel tube wall by the flow-drill technique. Both moment connections and simple connections were tested. The stiffness and strength properties of the connections were investigated and the effect of concrete filling on the connection performance was discussed.
Experimental Study, Results and Discussions
With the advent of the flow-drill technique, it was possible to use standard bolts for thin-walled tubular columns. In this technique, the tube wall is heated and displaced into the column. It then becomes thick enough to form threads for bolting. The flow-drill technique is advantageous for thicknesses up to 0.492 in..
The first test series included simple connections to hollow square tubes through flow-drill technique. The main parameters of the tests were end plate type, beam size, tube thickness, and bolt spacing. Some specimens were tested under axial load. The test setup consisted of a simply supported and horizontally placed column with a vertical wide flange beam at the middle. A shear force was applied cyclically at the top of the beam excluding the specimens having partial depth end plates. These specimens were subjected to monotonic shear loading at the beam tip . Either partial end plates or flush end plates were welded to the beam end and then bolted to the steel tube face. Partial end plate connections exhibited small moment capacity when compared to the plastic moment of the beam and they were classified as pinned connections. Flush end plate connections could be regarded as semi-rigid joints rather than simple joints as they exhibited significant moment capacities. The specimens with variable end plate thicknesses did not show any significant difference in behavior. However, larger wall thickness and higher beam depth caused an increase both in strength and stiffness of the connection. When the bolt spacing was increased, the strength and stiffness of the specimens was found to improve. The level of axial force influenced the response by decreasing the post-yield stiffnesses of the specimens.
In the second test series, moment connections to hollow square tubes using the flow-drill technique were tested. The test setup consisted of a horizontally placed column having a vertical girder at the middle. The end plate and girder sizes were selected to ensure that the steel tube face would govern the strength and stiffness response of the connections. The main parameters of the test were steel tube thickness and the grade of the steel tube. Flush end plates and extended end plates were used between the girders and steel tubes. The girder ends were subjected to monotonically applied shear force. The extended end plate connections underwent large rotations mainly due to the deformation of the steel tube wall, which became less as the thickness of the steel tube increased. After the bolts were removed, it was observed that the end plates were undamaged and the bolt threads deformed considerably. The specimens either failed by bolt pull-out or failed at a load level close to that. The compression zone of the specimens experienced yielding and outward web buckling. The amount of web buckling was larger for smaller tube thickness. In general, the extended end plate specimens with large tube thickness had higher strength but less ductility compared to the ones with small tube thickness. When the response of the extended end plates having different steel grades were compared, it was found that a 30% increase in steel yield strength caused an approximately 27% improvement in moment capacity at a 0.015 radian rotation level, beyond which the connection had signs of extensive yielding. For the specimens with flush end plates, the deformation was mainly at the column face. These specimens exhibited ductile response but less moment strength than the corresponding extended end plate specimens. The difference in strength at 0.02 rad. rotations was 63% and 70% for the specimens with 0.315 in. and 0.394 in. wall thicknesses, respectively. For the specimens with flush end plates, the moment strength improved approximately 42% when the thickness increased from 0.315 in. to 0.394 in.
In the third series of the experimental study, simple and fully-restrained connections of square CFTs manufactured using the flow-drill technique were tested. The same test setup was utilized as in the previous tests. The specimens were subjected to monotonic shear load applied at the cantilever girder-end. The D/t ratios of the columns varied between 20 and 31.7. The range for the measured yield strength of the steel was 46.1 to 61.9 ksi and the compressive strength of concrete was either 6.29 ksi or 7.32 ksi. For the flush end plate specimens, which were classified as simple connections, ordinary bolts were inserted before the casting of concrete. The ordinary bolts were then removed and replaced with strain-gauged bolts after the concrete had hardened. For the extended end plate specimens, ordinary bolts were inserted before pouring the concrete and these bolts were used throughout the test. The flush end plate specimens had different size beams as the main parameter. It was found that the specimens with larger size beams had more strength and stiffness. For each of the flush end plate specimens, yielding was observed across the whole compression flange of the beams. However, in the case of the equivalent hollow steel tube connections tested earlier, yielding had localized at the tips of the compression flanges. In addition, it was observed that the axis of rotation was at the compression flanges since the concrete core prevented the web of the columns from local buckling, and this caused an increase in initial stiffness. At the end of the tests, it was found that yielding at the column flanges had not extended to the webs. For some of the flush end plates, a sudden change in stiffness was observed in the transition region from elastic response to the nonlinear response while the equivalent hollow tubes achieved a smooth transition. This was attributed to crushing of the concrete in the steel tube. When the concrete was investigated after the test, it was found that crushing took place at some of the bolt regions as a result of bolt rotation. The two extended end plate specimens had different yield strength values for the steel tube. At the initial stages of the loading of the first specimen, a sudden drop in moment capacity at a rotation value of 0.0035 rad. was observed, and this was attributed to bolt slip. The loading then continued up to the failure of the outermost bolts at the tension region. When the specimen was examined after the test, it was found that there was no damage at the end plates. Yielding occurred at the girder webs and at the girder flanges in the compression zone. The steel tube did not deform at the compression region, while extensive deformation took place at the tension region. When the response was compared to an equivalent unfilled specimen, improvements in strength and stiffness with reduced ductility were observed. The second extended end-plate specimen had higher strength steel than the first one. It exhibited greater moment capacity and initial stiffness. The end plate underwent some slight deformation and extensive yielding in compression flange and web of the girder. The column face deformation was similar to the first specimen.
Reference
France, J. E., Davison, J. B., and Kirby, P. A. (1996). “Experimental Testing of Flowdrill Connectors with Concrete-Filled SHS Columns,” Tubular Structures VII, Proceedings of the Seventh International Symposium on Tubular Structures, Farkas J. and Jarmai, K. (eds.), University of Miskolc, Hungary, August 28-30, 1996, pp. 333-340. France, J. E., Davison, J. B., and Kirby, P. A. (1999a). “Strength and Rotational Response of Moment Connections to Tubular Columns Using Flowdrill Connectors” Journal of Constructional Steel Research, Vol. 50, pp. 1-14. France, J. E., Davison, J. B., and Kirby, P. A. (1999b). “Strength and Rotational Stiffness of Simple Connections to Tubular Columns Using Flowdrill Connectors” Journal of Constructional Steel Research, Vol. 50, pp. 15-34. France, J. E., Davison, J. B., and Kirby, P. A. (1999c). “Moment Capacity and Rotational Stiffness of Endplate Connections to Concrete-filled Tubular Columns with Flowdrilled Connectors” Journal of Constructional Steel Research, Vol. 50, pp. 35-48.