Browsing by Author "Gerber, Johannes Daniel"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Alternative wall-to-slab connection systems in reinforced concrete structures
    (Stellenbosch : Stellenbosch University, 2016-12) Gerber, Johannes Daniel; Van Zijl, G. P. A. G.; Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering.
    ENGLISH ABSTRACT: In the construction of multi-storey buildings, the reinforced concrete (RC) core or shear walls normally precede the construction of the RC floors. A system is therefore required to connect the floors to the already cast walls. There are many different floor-to-wall connection systems available in South Africa, but their behaviour and capacity are not always fully understood, especially when the moment capacity of the joint is to be utilised in order to design more economical structures. This research investigation focusses on four systems. The first is the conventional way of casting walls and floors by using continuous starter-bars that protrude through the formwork. The second system makes use of pre-bent site-installed starter-bars that are fixed within the wall, straightened after the wall is cast and then lapped with the slab reinforcement. The third system is also based on the principle of pre-bent starter bars, but they are pre-assembled inside a steel box to speed up the installation process and ease access to the bars once the wall is cast. Another system, that is not as popular in South Africa yet, is the use of cast-in anchors with mechanical couplers. The anchors are cast into the concrete wall and, once the form work is removed, threaded continuation bars are screwed into the couplers to make the connection between the floor and wall. The design procedure for the continuous starter-bar system is well understood and documented in design codes, but not enough information is available on the design procedure for the other systems. Certain practical aspects of the installation process are also not fully understood. Cold bending and straightening of the starter-bars are inevitable in both bend-out systems. Previous research shows that this cold working of the reinforcement can reduce the yield stress, ductility and E-modulus of the steel. In order to investigate these findings, a series of tensile tests are conducted on Y10 and Y12 reinforcement (high yield, deformed reinforcing bars). The results indicate that a significant reduction can be expected in both the yield stress and modulus of elasticity of the steel. A low cycle fatigue test further con rmed that a reduction in the ductility of the steel is also present. The next phase in the research investigation is to construct full scale models of the four connection systems. The systems are installed in four separate joints between a cantilever oor and a free-standing wall. The systems are designed to all have the same moment and shear capacity and are based on a typical wall-to-slab connection found in an o ce block. The construction of the models are done in such a way as to mimic the construction methods in practice as closely as possible. The entire process is done in a controlled environment, with a best practice approach adopted. Once the models are constructed, a load is applied at the tip of the cantilever oor in order to generate a bending moment and shear force at the wall-to-slab joint. The displacement of the specimen is monitored throughout the test. The structural performance is subsequently compared in terms of tip displacement, crack widths and ultimate resistance. The e ect of the cold bending on the starter-bars is clearly visible, as both the responses of the bend-out systems are less desirable than the results from the continuous starter-bar system. The experimental phase is followed by numerical analysis of the connection systems. Once the FE model is calibrated and veri ed with the data from the experimental phase, a sensitivity study is conducted. The e ect of either varying the steel or the concrete properties on the structural performance of a slab-to-wall connection is investigated. The FE results shows that the structural performance is signi cantly more sensitive to a reduction in the yield stress of the starter-bars, than to the use of a lower concrete grade. Finally, the systems are compared in terms of their constructibility, availability, material cost and their structural performance. Some recommendations on quality control matters and best-practice principles are also discussed. It is concluded that all the alternative connection systems can be successfully implemented in a moment-fixed wall-to-slab connection. The site-installed bend-out system performed the best structurally, but other practical considerations and project specific aspects might play an even bigger role in deciding which system to use in practice. Further, in order to ensure that the bend-out systems perform on the same level as the conventional system, it is recommended that the design should be conducted with a set of modified steel properties to allow for the negative effect of the cold working on the starter-bars.