Numerical analysis of welding considering phase transformation

Abstract

This work presents the design of a new shear connector and the corresponding results obtained on push-out tests. This new shear connector consists on a steel rib with indented cut shape that provides resistance to longitudinal shear and prevents transversal separation between the concrete slab and the steel profile (uplift). Adding to this, the connector openings cut makes easier the arrangement of transversal reinforcement bars. The installation of the connectors is simple and requires only common welding procedure. Due to its load capacity, the indented connector is able to replace a group of stud bolts. Its structural behaviour is analyzed and compared with other existing connectors, like the stud bolt and the Perfobond. The influence of different geometrical and mechanical aspects on the ultimate load capacity and ductility is assessed. The performed studies indicate that the proposed indented connector presents a good mechanical performance, associated with constructive and economical advantages. the steel beam upper flange (Figure 1b). This connector was designed to fulfil the need of a connector that could only mobilize elastic deformations for service loads. The main disadvantage of Perfobond connector is the difficulty to position the slab lower reinforcement, when the steel bars have to cross the connector openings. Figure 1– (a) stud connector; (b) Perfobond connector; (c) CR connector. This work summarizes the design and tests results for the proposed shear connector, called CR (Figure 1c). CR connector has an indented cut form that constitutes a good alternative to Perfobond connector, because it provides an easier disposition of reinforcement bars. It presents a symmetric cut, with trapezoidal saliencies and re-entrant angles, which provide resistance to longitudinal shear forces and prevent the transversal separation between the steel profile and the concrete slab (uplift). The concrete positioned inside the connector’s apertures works as concrete dowels, with a similar behaviour to the one obtained with Perfobond connectors. The experimental tests results obtained for CR connector are critically analysed and compared to the ones obtained for Perfobond and stud, regarding the maximum load capacity and the connection ductility. Like Perfobond, CR connectors present the following advantages when compared to stud connectors: they can be large scale produced, they can assume particular forms and shapes, they are easily welded to the steel profile with no need of special equipment, and the welding task can be performed either at site or at an industrial unit. In terms of load capacity, CR connector provides equivalent resistance to a group of studs. Several tests performed with CR connectors by the authors of this work and with Perfobond connectors by other investigators showed that both the connection load capacity and ductility are influenced by the concrete strength and the slab transversal reinforcement (Leonhardt et al. 1987, Oguejiofor et al. 1994). Therefore, it is possible to control CR connection load capacity properly choosing concrete strength and reinforcement rate. On the other hand, when stud connector failure is governed by shearing, an increase on concrete strength has only a small influence on the connection load capacity. 2 EXPERIMENTAL TESTING 2.1 Push-out tests Push-out tests were used to study CR connector behaviour, in order to establish the load-slip relation. According to Eurocode 4 (2004), the push-out specimen consists on a steel beam section held in vertical position by two identical concrete slabs, as showed in Figure 2. Beside the specimen geometry, Eurocode 4 (2004) also defines the test procedure. The initial phase of the test is characterized by 25 cycles of loading and unloading, between load values of 5% and 40% of the predicted ultimate load. Following this, the test is controlled by deformation, with measurements of slip between the steel profile and the concrete slab at a constant rate. Lateral displacement between slabs is also measured. The test proceeds until failure, and slip is measured until the load value is at least 80% of the ultimate load. The slip capacity δu, measured in a push-out specimen, should be considered as the maximum slip correspondent to the characteristic load PRk, as shown in Figure 2. The characteristic load PRk is taken as the smaller failure load divided by the number of shear connectors and reduced of 10%. The characteristic slip capacity δuk is considered equal to 0,9δu. (a) (b) (c)