EXPERIMENTAL PERFORMANCE OF LVL-CONCRETE COMPOSITE FLOOR BEAMS David Yeoh 1 , Massimo Fragiacomo 2 , Andrew Buchanan 3 , Bruce Deam 3 ABSTRACT: This paper reports the outcomes of short-term collapse tests performed on eleven LVL (laminated veneer lumber)-concrete composite floor T-beams. Different variables such as span length (8 and 10 m), connection and concrete type, and design level (well- and under-designed, in terms of connector numbers) were investigated. During the tests, mid-span deflection, connection slips and strains were measured. Connection types investigated include triangular and rectangular (150 mm and 300 mm long) notches cut in the timber and reinforced with a coach screw, and modified toothed metal plates pressed on the edge of the LVL joists. All of the beam specimens were designed using the effective bending stiffness or γ-method, in accordance with Annex B of Eurocode 5. All well-designed beams provided more than 95% composite action even though there were relatively few connectors (e.g. six 300 mm long notches on the 8 m span beam). The beams with 300 mm rectangular notched connection exhibited the best performance, with high stiffness and strength beyond the ultimate limit state load level and, requiring fewer connectors along the beam. The triangular notch was found to be a viable alternative, with more connectors but easier and faster to cut than a rectangular notch. Metal plate connectors provide a practical construction possibility, but the beam stiffness was found to rapidly deteriorate beyond the ultimate limit state load level. KEYWORDS: Composite structures; Connectors; LVL; Timber construction; Timber-concrete; Wood. 1 INTRODUCTION 123 Timber-concrete composite (TCC) system is a construction technique used for strength and stiffness upgrading of existing timber floors and new construction such as multi-storey buildings and short-span bridges. By combining two different materials it is possible to exploit their best qualities since the timber is positioned in the tension region of the composite section while the concrete is used in the compression region. The presence of timber, due to its lower density in comparison with reinforced concrete, decreases the weight of this flooring system, implying several advantages: (1) higher efficiency in terms of load carried per self-weight; (2) better seismic performance derived by less structural mass; and (3) lower carbon footprint of the building when compared with concrete, due to the advantage of carbon stored in the timber. The advantages given by the concrete slab are: (1) larger thermal mass and fire 1 David Yeoh, Faculty of Civil and Environmental Engineering, Universiti Tun Hussein Onn Malaysia, Johor, Malaysia. Email: david@uthm.edu.my 2 Massimo Fragiacomo, Department of Architecture, Design and Urban Planning, University of Sassari, Italy. Email: mfragiacomo@uniss.it 3 Andrew Buchanan and Bruce Deam, Department of Civil and Natural Resources Engineering, University of Canterbury, Christchurch, New Zealand. Email: andy.buchanan@canterbury.ac.nz and bruce.deam@canterbury.ac.nz resistance; (2) better acoustic separation; and (3) good structural performance in seismic regions since the floor behaves as a rigid diaphragm. All the aforementioned advantages can only be achieved if the composite system is structurally effective by means of a stiff and strong shear connection system. Quite a number of short-term collapse tests have been performed to date on TCC floor beams [1]-[3]. Collapse tests are important to quantify the actual composite action of the system, the load-bearing capacity and the failure mechanisms. There is in general a close relationship between the collapse load and the failure mechanism, and the type of connection system. A push- out test of the connection should always precede a beam collapse test in order to obtain important information on the mechanical properties of the connection. Lukaszewska et al [1] tested five 4.8 m span full scale TCC floors of triple T-section glulam joists tested to failure in 4-point bending. The concrete slab was prefabricated off-site with mounted connectors. Three specimens had lag screws surrounded by steel pipes whilst two specimens had metal plates nailed to the glulam joists. Composite action of 60% and 30% were achieved in the beams with lag screws and metal plates, respectively. Ceccotti et al [2] tested double 6 m span glulam T-beam, with 18 corrugated rebars glued to each beam with epoxy resin. Beam was twice loaded and unloaded prior to 4-point bending collapse test after a 5-