Cyclic Tests on Large Scale Steel Moment Connections

Moment resisting frames are widely used in earthquake resistant steel construction. The beam-to-column connections play a crucial role in the performance of these frames. For the popular detail of the welded flange - bolted web type moment connection, the 1988 UBC requires supplemental web welds between the shear tab and the beam web for beam sections where Z_f/Z ≤ 0.70. Z is the plastic modulus of the beam, and Z_f is the plastic modulus of the beam flanges only. Past tests have suggested that inadequate participation of the beam web connection is transferring moments may have a detrimental effect on inelastic deformation capacity of the beam, particularly for beam sections characterized by a low Z_f/Z. The supplemental web welds are intended to increase the flexural participation of the web connection for such sections.

An experimental investigation was conducted on large scale beam-column subassemblages to collect additional data on the effect of the Z_f/Z ratio and the web connection details. A total of eight specimens was tested. Three different beam sections were used: W24x55 (Z_f/Z = 0.61), W21x57 (Z_f/Z =0.67) and W18x60 (Z_f/Z = 0.75). Web connection details included bolted webs, bolts with supplemental web welds, and a single specimen with a fully wleded web connection. All connections were to the flange of a W12x136 column. The panel zone of the column was sufficiently strong so as to force inelastic deformations to occur primarily as flexural yielding of the beam at the connection. All specimens were loaded cyclically to failure. Specimens were constructed by a commerical structural steel fabricator and inspected by an independent welding inspection film. Inspection included ultrasonic testing of all comoplete penetration groove welds.

Results of the testing program showed highly variable performance among the eight specimens. The primary criterior for judging performance was the plastic rotation developed by the beam prior to connection failure. All connections failured by fracture in the vicinity of the complete penetration beam flange groove welds. Plastic rotations developed by the beams varied from ± .002 radian to ± .015 radian.

The tests showed no clear influence of the Z_f/Z ratio or web connection detail on beam plastic rotation capacity. Rather, the highly variable quality of the complete penetration beam flange groove welds appears to have dominated the response of the specimens. Nonetheless, the specimens with supplemental web welds developed somewhat larger plastic rotations than their counterparts without web welds. Consequently, no change is recommended at present in the UBC detailing requirement pertaining to supplemental web welds.

The eight specimens tested in this program showed highly variable and unpredictable behavior, and developed plastic rotations that were judged as poor to marginal for severe seismic applications. Tests by other investigators have also shown similar variable results. Some of this variability can be attributed to the influence of the Z_f/Z ratio and web connection details. However, a great deal of the variability also appears to be related to the performance of the beam flange groove welds.

The final recommendation of this investigation calls for a review of current industry practice for seismic steel moment connections. The results of this and previous test programs leads to questions on the reliability of the welded flange - bolted web detail for severe seismic applcations. A thorough review of design and detailing practices, as well as welding and quality control issues is needed.