ABSTRACT
Abstract A test program in which flush end-plate connections of variable plate thickness are loaded under cyclic displacements of variable amplitude was undertaken as part of a study of the behaviour and modelling of semi-rigid partial-strength connections. This paper presents the details of the testing regime and describes the initial observations of the experimental work. The moment-rotation hysteresis curves are determined and initial conclusions are drawn on the effect of the plate thickness on the cyclic response of the flush end-plate connections. The yield and ultimate rotational capacities are predicted using the Eurocode 3 T-stub method and a second method; these being compared to the experimental results. It is intended that one of these methods will form the basis of a new cyclic response prediction model. This model will allow the seismic behaviour of connections using flush end-plate connections to be predicted from their material and geometrical properties. Keywords: flush end-plate connections, moment capacity calculation, semi-rigid partial-strength, cyclic response testing
1 Introduction
In recent years, there has been a shift away from the full-strength, heavily-welded connections conventionally favoured hi seismic engineering towards semi-rigid moment connections. These moment connections can display many attractive features such as high ductility and energy dissipation characteristics. One such connection is the end-plate connection for which an abundance of design guidance is available [1]. In particular, the flush end-plate connection, which is classified as a partial strength connection, may be designed to yield under strong ground motion. Considering the brittle failure of many connections inspected after the Northridge and Kobe earthquakes, this ability to yield at a specified load in a controlled manner offers
obvious advantages. However, to ensure that the correct yielding mode is obtained, carefiil selection of design details such as end-plate thickness, bolt size and bolt spacing is required. While the existing guidance for wind-moment design includes these as considerations for ductility requirements, it is unlikely that the ductility provided will be adequate for seismic engineering applications where a much higher ductility demand is experienced.
