SHAKE-TABLE TESTS OF A REINFORCED CONCRETE FRAME RETROFITTED WITH HYSTERETIC DAMPERS CONNECTED USING AN IMPROVED JOINT STRUCTURE

In the last decades, the use of hysteretic dampers for the seismic upgrading of existing frame structures increased exponentially. Several types of hysteretic dampers have been used, one of them being the brace dampers that are installed in the structure like conventional diagonal members. In the case of reinforced concrete (RC) structures, the connection between the old RC beam-column joints and the damper devices has been pointed out as the key challenge of this retrofitting technique (Fardis, 2009). In fact, when the damper has the form of a conventional diagonal bar, it will be subjected to high axial loads and its influence on an existing, and probably damaged, beam-column joint is of major concern. Several solutions have been proposed for the joint structure in the past. One of them consists on using constraining members (referred also as “restraining plates” hereafter), independent from the damper brace element, that are fixed onto the concrete slab through steel anchor bolts. Steel anchors on concrete resist high shear loads, while longer anchoring height, larger number of anchors and heavier anchoring plates, are needed when high axial loads appear, making the retrofitting solution costly and non-competitive. This paper investigates an improved solution for the joint structure that is aimed at minimizing the axial forces acting in the anchor bolts. The proposed solution consists on a particular arrangement of interface materials with low friction properties and restraining plates. To investigate experimentally the efficiency of the proposed solution, dynamic tests were conducted with the 3x3m shaking table of the University of Granada. First, a three storey, three bay prototype RC frame structure was designed considering only gravity loads and conventional reinforcement detailing. The prototype structure represented pre-70 existing RC structures located in moderate seismic zones, such as the Mediterranean region. A high proportion of the building stock in these regions were under-designed before the appearance of seismic codes or by using rudimentary anti-seismic design criteria (Masi, 2004). Recent earthquakes have shown the poor performance of this type of structures (L’Aquila 2009, Lorca 2011), underlining the need of seismic assessment and retrofitting (Benavent-Climent et al., 2013). Second, from this prototype frame, a test structure was defined by applying scaling factors of 2/5, 1 and 1, for dimensions, stress and acceleration, respectively. Third, the test structure was subjected to several seismic simulations with the shaking table in order to induce moderate damage. Fourth, the damaged test structure was retrofitted with brace-type hysteretic dampers that were connected to the main frame using the improved joint structure. The specimen was subjected to a series of dynamic tests under increasing seismic simulations, which corresponded to very frequent, frequent, rare and very rare earthquakes. 1 FPU PhD student, Department of Structural Mechanics and Hydraulic Engineering, University of Granada, Spain, elenaoliver@ugr.es 2 Professor, Department of Mechanics of Structures and Industrial Constructions, Polytechnic University of Madrid, Spain, amadeo.benavent@upm.es