DESIGN CONSIDERATIONS OF FRICTION DAMPER FOR SEISMIC RESPONSE REDUCTION IN FRAMED BUILDINGS

Abstract

Experimental studies have shown that friction dampers are very effective in reducing seismic response in building frames. They also require less maintainance compared to other energy dissipation devices and are much more economical and simpler to install. A few methods for the design of friction damper for seismic response reduction are available but are not widely adopted in the national building design codes because of the significant variation in the proposed optimal values of the design parameters of friction dampers. The studies conducted so far are either experimental investigations or numerical investigations on a small subset of ground motions. This makes the adoption of design procedures by national building design codes difficult. In this thesis, we address three main questions related to the design of friction dampers: (i) How to numerical model the friction damper? (ii) What are the design parameters of the friction dampers and their optimal values? (iii) What is the number of friction dampers required in a building and how to allocate them to have maximum efficiency in seismic response reduction? Numerical simulations are performed using a suite of 50 ground motions subjected to single-degree-of-freedom (SDOF) and multi-degree-of-freedom (MDOF) systems with varying system parameters and connected with friction dampers. The numerical simulations on SDOF systems are used to identify the design parameters of the friction dampers. Past studies have used slip load in the friction slider as a design parameter to be optimized. Some studies have considered the slip load as a constant value or as a function of damper stiffness while some have considered it as a function of predominant period of the ground motion. We propose, a new parameter, slip factor (sf ), as a design parameter. It depends on the stiffness of the friction damper, sliding force in the friction damper and characteristics of the ground motion. It is shown that the slip factor is very effective in predicting the possibility of slip in the friction damper for a given ground motion. The slip factor also shows a very high correlation with the energy dissipation and hence serves as a better parameter to be considered in design than just slip load. Several system performance indices such as - relative performance index (RPI), normalized peak relative displacement (Rd) etc, are evaluated with varying damper parameters to understand the behaviour of the friction damper. A new system performance index nameky, damper performance index (DPI), is also proposed which not only accounts for the energy dissipated by damper but also considers the system response. The slip factor proposed for the SDOF systems is then modified to suit the MDOF systems so that it can be used in multi-storey building frames. It is shown that the optimal value of the slip factor in MDOF systems remains same as that of for SDOF systems. Also the energy dissipation characteristics of the friction dampers when it is connected at different stories in the frame is presened. Based on this a new index - storey dissipation potential index (SDPI), is proposed which predicts the maximum capacity of energy dissipation at each storey. The SDPI is shown to have a very high correlation with the maximum energy dissipation potential of the storeys, especially at the lower half number of storeys. It is recommended to allocate the dampers in a descending order of SDPI. Effect of such allocation on system response and energy dissipation is shown in the last part using a few examples. The method of allocation of dampers using SDPI is also compared with the well known available method called simplified sequential search algorithm (SSSA). The limitations of SSSA method in damper placement are highlighted through example case studies which also show that it is possible to achieve similar response reduction with fewer dampers placed in accordance with the proposed SDPI method. Hence, the peoposed allocation method is more efficient than the SSSA while also being simpler, intuitive and less resource demanding. The effect of inelasticity in the frame elements is also investigated and shown in for the case of a 2D two bay frame and a 3D frame with plan irregularity. The slip factor sf and SDPI are shown to work well in many different examples with frame elements as linear elastic, inelastic, with irregularity along height and irregularity in plan.