IN SERVICE SEISMIC STRENGTHENING OF RC FRAMED BUILDINGS

TUGRUL TANKUT, UGUR ERSOY, GUNEY OZCEBE, MEHMET BARAN AND DILEK OKUYUCU Department of Civil Engineering, Middle East Technical University, 06531, Ankara, Turkey

An innovative non-evacuation retrofitting technique has been developed for reinforced concrete building structures, which constitute the major portion of the existing building stock in Turkey and in the region. The introduction of cast-in-place reinforced concrete infill walls, connected to the existing frame members, is known to be very effective in improving the overall seismic structural performance. However, this technique is not suitable for strengthening of the existing building stock, since it involves messy construction works and requires evacuation. The idea behind the proposed method is to transform the existing hollow masonry infill walls into strong and rigid infill walls by reinforcing them with relatively high strength precast concrete panels epoxy glued to the wall and dowel connected to the frame members in order to enhance the lateral stiffness of the frame. The panels to be assembled on the wall are small enough to be handled by two workers. About thirty quasi-static tests have so far been performed on one-third scale, one-bay two-storey reinforced concrete frames (two series consisting of twenty-one such tests will be reported in the proposed paper), besides three shake table tests on three-dimensional test units consisting of two threebay two-storey frames of similar properties connected with slabs to verify the efficiency of the developed strengthening technique under actual seismic action. In these tests, effects on the performance of the parameters such as (i) panel geometry, (ii) panel-to-panel connections, (iii) panelto-frame connections, (iv) internal or external applications, (v) lapped splices on column reinforcement, (vi) axial load level in columns and (vii) use of one-storey test frames have been investigated. Additional test series have been planned to study (viii) the effects of infill wall aspect ratio, (ix) the possibility of using lightweight high strength concrete panels and (x) the advantages of using fibre reinforced concrete panels. Test units have been designed and constructed to reflect the most common deficiencies observed in the local practice, such as strong beam-weak column connections, insufficient confinement, low-grade concrete (C12-C16) and poor workmanship. Realising the problems caused by insufficient lap length on the column reinforcement that may lead to premature failure in retrofitted structures, this factor was also included among the parameters studied. Various types of panels of manageable size and weight were designed in two categories; (i) nearly square panels arranged in three layers and (ii) full height narrow strips. Since masonry wall thickness is usually less than the frame member width and the wall is eccentrically placed on the external face of the frame, the two possible panel application modes were studied; (i) internal application where the panel layer was nicely surrounded and confined by the frame and (ii) external application where the panel layer had to be extended over the frame members and anchored to the external face of the frame. The method of connecting precast concrete panels together is a critical issue, since these connections need to be capable of transferring tension and compression as well as shear. Shear keys and welded connections on panel reinforcement were considered essential in the design of the first series of test specimens. However, tests on these specimens revealed that integration of the panels with each other and with the hollow brick wall could be achieved satisfactorily by epoxy alone. Indeed, welding of bars and shear keys was eliminated in all the following test series, leading to considerable simplification, and yet a very satisfactory performance could be obtained. Although it is obviously best to have epoxy-anchored dowels along the four sides of each panel, the possibility of reducing the number of panel sides having dowels has also been investigated to simplify the technique further and to improve its cost efficiency. 1
Advances in Earthquake Engineering for Urban Risk Reduction

Tests were performed under constant vertical load combined with reversed cyclic horizontal load simulating seismic action. Numerous deformation measurements taken were used in the evaluation of the specimen performance in terms of load-displacement, ductility, energy dissipation, stiffness degradation etc. A detailed discussion of the test results will be given in the full text. However, some of the major observations are summarised below:

The proposed retrofitting technique successfully serves the purpose of enhancing the lateral stiffness and thus improving the seismic performance of framed structures. Since strengthening by using panels with straight edges is equally successful, complicated connections with shear keys and welding can and should be avoided. Three layers of nearly square panels and full height narrow strips exhibit almost identical behaviour with a slight superiority of the strips, since a higher number of dowels is used. Although panels dowel connected to the frame along less than four (three or even one) edges can produce almost the same capacity, the use of dowels along the four sides is considered essential in the lower storeys of building structures. External applications are equally successful if a reasonable number (not more than number of dowels used in internal applications) of carefully spaced anchor bolts are used. Significant bond deterioration was not observed in test specimens having inadequate lap length (20) when the axial load was relatively high (20% of the capacity); however, some bond problems were observed under lower axial load level (10%). Almost identical observations could be made on the one-floor specimens, indicating that the use of this simpler type test specimen is also acceptable.

An overall interpretation of the test results, summarised in Table 1, indicates that the proposed seismic retrofitting technique can effectively improve the seismic performance of reinforced concrete framed structures.
TABLE 1: Performance improvement

Relative to masonry infilled frame Lateral load capacity Lateral stiffness Ductility Energy dissipation 2.5 times 3 times 2 times 3 times

Relative to bare frame 15 times 20 times 0.2 times 60 times

It is also evident that the proposed technique can suitably be applied to building structures in use with minimal disturbance to the occupants, and in that sense, the panel reinforced masonry infill wall technique deserves to be classified into the occupant friendly rehabilitation (OFR) category. Besides, its cost effectiveness is obviously very high. Cost of construction is estimated to be of the same order as that of the cast-in-place reinforced concrete infill. However, when the cost of evacuation (i.e., the moving out and moving in expenses together with the rent to be paid during the construction period) is taken into account, the cost effectiveness of the proposed technique becomes unrivalled. In parallel with the experimental work, analytical studies towards formulating a convenient design procedure have also been carried out. In one of the two approaches so far explored, panel reinforced masonry infill wall is modelled as a monolithic structural wall having an equivalent thickness reflecting the concrete strength and thickness of the panel as well as the strength and thickness of the masonry infill wall. The other approach is based on the pin connected diagonal strut of an equivalent width representing various properties of the panel, the masonry wall and the epoxy layer (a modified version of the Smith method). Both approaches seem to have the potential to serve as the basis of a 2
Advances in Earthquake Engineering for Urban Risk Reduction

suitable design procedure if calibrated using the reliable experimental data available. Case studies are being carried out to finalise the calibration process. On the other hand, the Turkish seismic code is currently being revised with the major issue being the introduction for the first time of an additional chapter devoted to rehabilitation. It is hoped that the panel reinforced masonry infill wall technique will be included in the code among the recommended seismic retrofitting methods.

Keywords: seismic strengthening; retrofitting; precast panels; infill walls; connections; diagonal strut

3
Advances in Earthquake Engineering for Urban Risk Reduction