DISPLACEMENT BASED DESIGN OF R.C. INFILLED FRAME BUILDINGS

Abstract

The unreinforced brick masonry panel is provided as infill in a large number of RC frame buildings in many parts of the world. There is no separation between the masonry infill panel and the RC frame. In many cases, the infill panels in the ground storey are omitted to provide the space for parking or commercial activities. These buildings are known as open ground storey buildings. Traditionally, the infilled frame buildings are designed as bare frame buildings although in practice the infill panels are not separated from frames to avoid the interaction between the RC frames and the infills during earthquakes. The response of buildings in past earthquakes has proved that the infills are affecting the earthquake behaviour of the building frames. The behaviour of building frames during earthquake is influenced by the high strength and stiffness of the infill panel until it fails. The proportioning of the members and earthquake resistant measures based on the analysis of bare frames has limited validity in presence of the infill panels. The open ground storey buildings are known for their fragile performance during many past earthquakes. Till date most of codes around the world suggest the force based design method for buildings. Researchers have reviewed the force based design methods and pointed out their limitations. Recognizing the limitations of force based design, it seems rational to adopt a seismic design method directly addressing the displacements and deformations in the structure right from the beginning of the design process. The displacement based design procedures are developed to overcome the limitations of force based design methods. In the present study, a displacement based design procedure for earthquake resistant design of masonry infilled RC frame has been proposed. This procedure is also applied to earthquake resistant design of open ground storey buildings. The inter storey drift is an important consideration for the electrical and mechanical in components often referred as non-structural elements. The non-structural elements are classified as acceleration sensitive or displacement sensitive components by FEMA 356. The displacement sensitive components include the pipelines carrying gases and fluids for air conditioning or for many other purposes such as oxygen pipes in case of hospitals. The proposed displacement based design procedure is especially useful for the buildings for which effective drift control is necessary for the functionality after the earthquake because the buildings are designed to deform by limiting interstorey drift under specified level of seismic excitation. The seismic behaviour of the infilled frame depends on the relative strength of the infill panels and the frame. Observations of experimental studies suggested that the weak infill panel bounded by strong frame has demonstrated ductile behaviour of over all structure without any brittle mode of failure. The force deformation behaviour is controlled by the frame. Although the infill panels degrade completely at higher drifts, the frame remains stable and able to resist the gravity load. The strong infill panel-weak frame has exhibited brittle mode of failure by shear failure of columns. The structure is not stable after reaching to limiting deformation. The strong frame infilled by the strong panel show large strength enhancement compared to the bare frame. The load deformation behaviour is controlled by the strength of the infill panel. The infill constituting the infilled frame considered in the study, is assumed to be weak such as made up of hollow tile bricks or burnt clay bricks and the infill is constructed in full contact with the frame at all boundaries. The infills are acting as diagonal brace when subjected to lateral forces. At drift angle of approximately 0.3-0.5%, initial failure of infill occur by crushing at corners in contact with frame and separation occurs at other corners (i.e. tension diagonal). This may accompany by severe diagonal cracking parallel to the strut and sliding shear failure along horizontal bed joints. After this, the structural

action changes from a truss or braced frame to frame action with the infill playing no further part in the response. This happens normally around a drift of 1%. After achieving approximately 1% drift, the infilled frame behaves as a bare frame. The open ground storey buildings are observed to fail by formation of storey mechanism due to occurrence of plastic hinges at top and bottom of the open storey columns. The drift in the open storey is observed larger then rest of the stories. When the frame elements are proportioned by strong column-weak beam rule, the seismic behaviour of open ground storey buildings are found to be similar to the uniformly infilled frame buildings. A non dimensional displacement profile for the infilled frame buildings is proposed which is significantly different than the displacement profile of a bare frame building. The drift in the bottom stories of the infilled frame buildings is large as compared to the upper stories. This behaviour of the infilled frames is observed during many analytical and experimental studies. The buildings of four, eight and twelve storey height having three bays in each direction are designed as bare frames by the Direct Displacement Based Design (DDBD) procedure as suggested in literature. The infill panels are modeled as diagonal struts. Based on the results of the pushover analysis of the infilled frame buildings, an empirical non dimensional equation for displacement profile has been suggested, considering the effect of the length to height ratio of the infill panels. The compressive strength of infill masonry is assumed 4.0 Mpa in this study, which is commonly observed strength of masonry in many parts of the world. In DDBD procedure, a multi-degree of-freedom (MDOF) structure is represented as an equivalent linear single-degree-of-freedom structure (SDOF) having secant stiffness at the maximum displacement response. The characteristics of the equivalent SDOF structure are derived using equal work principle. The damping in the equivalent SDOF structure is the sum of the elastic and the hysteretic dampings. The effective period is found from the displacement spectra for the total damping corresponding to the design displacement. The base shear is calculated from the secant stiffness and design displacement. The DDBD procedure is suggested for a infilled frame buildings. The equivalent single degree of freedom system characteristics are calculated for the infilled frame building considering the assumed sizes of the beams and the columns and based on proposed displacement profile. The displacement profile is the displacement shape of the building when storey drift in any of the stories reaches the design drift, the drift in other stories may be equal to or less than the design drift. The energy dissipation in the infill panels are accounted as the hysteretic damping, which is averaged with the hysteretic damping of the frame. The total damping is the sum of the elastic and the hysteretic damping. The displacement spectrum is corrected for the total damping and the effective time period is found from corrected displacement spectrum corresponding to design displacement. The effective stiffness and design base shear are calculated from effective time period and design displacement. The proposed procedure is used for seismic design of the open ground storey buildings with minor changes. The life safety limit state is assumed to occur at 2.0 % drift under MCE condition according to IS:1893:Part-I:2002 and the serviceability limit state at 0.5 % drift under DBE condition. It is assumed that the yield displacement for the infilled frame and the bare frame is same. The assumption is verified from results of pushover analysis of buildings with different heights and bay width for 115 mm thick masonry panels having compressive strength of masonry 4.0 MPa. The comparison of yield displacements from the bilinear approximation of the push over curves for bare and infilled frame buildings show that the consideration of same yield point is quite acceptable when the weak infill panel and

strong frame assumption is fulfilled. The results of nonlinear dynamic analysis performed by the software SAP 2000 is validated through comparison with the experimental results of pseudo-dynamic test on a four storey full scale RC frame published in literature. The building was tested as bare frame and uniformly infilled frame. The frequency, maximum top floor displacements and time history displacement response for two levels of seismic excitation for bare and infilled frame buildings closely match with the experimental results. The buildings of four, eight and twelve storey heights are now designed by the proposed DDBD procedure. Different arrangement of the infill panels have been considered such as the uniformly infilled frame and the open ground storey. Same plan having three equal bays in each direction is considered for all the buildings. Reinforced concrete frame made up of beams and columns is infilled by the unreinforced masonry panels of 115 mm thickness and compressive strength 4.0 MPa. The ground storey columns are assumed as fixed at base. The structure is assumed to be located in seismic zone V according to IS:1893-Part-I:2002 with zone factor of 0.36. The buildings are assumed to be resting on the medium soil (Type-2). The buildings are designed for 2 % drift. The uniformly infilled frame and open ground storey buildings are designed for the earthquake loading along shorter direction of the building. The buildings are designed for the life safety limit state under MCE condition and checked for the serviceability limit state under DBE condition. The performance of designed buildings is assessed by non-linear dynamic analysis by applying eight spectra compatible acceleration time histories. The three dimensional models of the buildings are analysed by SAP-2000. The beams and columns are modeled as two hinge elements. The floors are modeled as rigid diaphragm. The beam hinges are moment hinges and elastic flexural rigidity is calculated by dividing the yield moment by Vll yield curvature. The column hinges are P-M interaction hinges. The flexural rigidity of columns is calculated by yield moments divided by yield curvature. The infill panels are modeled as diagonal struts according to FEMA 356. The nonlinear time history analysis is carried out for time histories corresponding to the MCE level earthquake excitations, which reflects the performance of the structure at the life safety limit state. As the buildings are designed for assumed displacement profile considering the secant stiffness at effective displacement and energy dissipation during the nonlinear response, the maximum displacement response at floor level will directly reflect the effectiveness of the design method. The average of maximum floor displacement and drift are in good agreement with proposed displacement profile for the uniformly infilled frame and open ground storey buildings. The displacement response of the open ground storey buildings observed is similar to the uniformly infilled frame buildings. The drift in the open storey is higher than the drift in the ground storey of the uniformly infilled frames buildings. The failure by formation of storey mechanism is avoided by capacity design of columns. The infilled frame buildings are designed according to the design codes in the practice. Most of the codes all over the world suggest force based design method (FBD) as far as earthquake resistant design is concerned. The buildings are designed according to provisions of the Indian codes IS:1893:Part-I:2002 and IS:13920:1993. The performance of the designed buildings is assessed by nonlinear dynamic analysis by applying eight spectra compatible acceleration time histories. The spectrum compatible time histories are generated corresponding to the recorded time histories of the earthquakes. The average of maximum floor displacement and interstorey drift for the building designed by force based design method is found to be less than respective results of the building designed by DDBD. Vlll The DDBD of an eight storey existing building is carried out as an example. The building has eight equal bays in one direction and three unequal bays in other direction. The building is designed for earthquake loading in both the directions in plan. The performance of the designed building is evaluated by nonlinear dynamic analysis under the action of eight spectra compatible time histories. The average of maximum floor displacement and drift from time history analysis are in good agreement with proposed displacement profile. The displacement profiles for the longitudinal and transverse directions are very much different due to different configuration of infill panels in both the directions. The top displacement in the transverse direction is less due to infill panels of high aspect ratio as well as large span beams are required to be designed for gravity load cases above third floor. The configuration of infill panels can be effectively used to control the top displacements. Sufficient number of bays provided with infill panels with preferred aspect ratio can influence the top displacement significantly. Of course, functional requirement of building is also need to be considered. Although the proposed displacement profile is based on four, eight and twelve storey buildings having varying bay width, application of DDBD using this displacement profile in case of eight storey building show very good agreement. Therefore, the proposed displacement profile is applicable to other buildings with different number of bays and heights. A DDBD procedure has been proposed for infilled frame buildings which incorporate the strength, stiffness and displacement characteristics of the composite structural system. The non dimensional displacement profile proposed for the infilled frame buildings is applicable to the buildings of varied height, number of bays and a practical range of length to height ratio of the infill panels having the compressive strength of 4.0 MPa. It is assumed that the infilled frame consisting of the strong frame and weak infill panels. The IX seismic behaviour and the procedure may differ for other combination of relative strength of the frame and the infill panels.