0660 1 Degenkolb Engineers, 225 Bush St.,, Suite 1000, San Francisco, CA 94104, USA Email: kyu@degenkolb.com 2 University of California at San Diego, La Jolla, CA 92093-0085, USA Email: cmu@ucsd.edu CYCLIC PERFORMANCE AND RETROFIT DESIGN OF PRE-NORTHRIDGE STEEL MOMENT CONNECTION WITH WELDED HAUNCH Qi-Song YU 1 And Chia-Ming UANG 2 SUMMARY The effectiveness of using welded haunch scheme to rehabilitate pre-Northridge steel moment connections was investigated through cyclic testing of two full-scale specimens-one of them incorporated a lightweight concrete slab. Experimental results demonstrated that with the low- toughness E70T-4 groove weld in place, welding a triangular haunch beneath the beam's bottom flange significantly improved the seismic performance of steel moment connections. When a concrete slab was present, brittle fracture of the welded joints was prevented. A simplified model which considered the force equilibrium and deformation compatibility between the steel beam and the haunch was developed to predict the stress level in the beam flange groove welds. A design procedure for the welded haunch connection was proposed. INTRODUCTION Steel special moment-resisting frames (SMRFs), believed to be capable of dependable and ductile response during strong earthquakes, have long been considered a premier lateral force-resisting system. Unfortunately, widespread brittle fractures in or around the beam bottom flange to column flange groove weld were observed in more than 150 steel SMRF buildings after the January 17, 1994 Northridge earthquake. Among the concerns regarding the poor performance of these connections is the ability to economically rehabilitate steel moment connections in existing buildings. RESEARCH OBJECTIVES AND SCOPE The objective of this research was to investigate the effectiveness of using welded haunch for connection modifications. Two full-scale specimens were tested. One specimen (designated as NIST-2) was bare steel and the other one (designated as NIST-2C) incorporated a lightweight concrete slab. Based on the research findings, design procedure for the welded haunch modification was also developed [Yu et al. 1997]. TEST SPECIMENS AND TESTING PROCEDURE The specimen geometry and test setup are shown in Figure 1. Two nominally identical pre-Northridge steel beam-to-column subassemblies were first constructed for modification. See Table 1 for actual material properties. The moment connections (see Figure 2) were designed in accordance with the 1985 Uniform Building Code [Uniform 1985]. Each specimen was constructed and inspected using techniques similar to those used in pre-Northridge construction. Beam flange groove welding was performed with a 0.12-in. diameter E70T- 4 electrode (Lincoln 3M); steel backing and weld tabs were left in place. One-in. diameter A325 slip-critical high-strength bolts were fully tightened using the turn-of-nut method. The geometry and welding details of the welded haunch for modifications are shown in Figure 3. A 0.072-in. diameter E71T-8 electrode (Lincoln NR-232) with a specified Charpy V-Notch toughness of 20 ft-lb at -20° F was used for making all the welds for connection modifications. The groove welded joint of the beam top flange was left in its pre-Northridge condition. A 6-1/4-in. thick, 8-ft wide lightweight concrete slab was incorporated in the composite specimen in order to simulate a common practice of composite floor slab construction in