| doi:10.3850/978-981-08-6218-3_CC-Th032 |
 Final Paper PDF
|
LONG-TERM BEHAVIOUR OF COMPOSITE STEEL CONCRETE SLABS: AN EXPERIMENTAL STUDY
S. Shayan1, S. Al-Deen1, G. Ranzi1,a and Z. Vrcelj2
1School of Civil Engineering, The University of Sydney, Sydney, Australia.
agianluca.ranzi @sydney.edu.au
2School of Civil and Environmental Engineering, The University of New South Wales, Sydney, Australia
EXTENDED ABSTRACT
This paper presents experimental results aimed at investigating the influence of time effects on the long-term behaviour of composite steel concrete slabs. The main advantage of composite flooring systems relies on the ability of the profiled sheeting to act as permanent formwork and, once the concrete has hardened, as external reinforcement. In particular, this work intends to gain insight into the nonlinear shrinkage distribution which occurs in composite slabs. In fact, this non-uniform profile develops through the slab thickness because of the lack of moisture egress at its bottom side due to the presence of the sheeting. Two sets of experiments have been prepared to study this behaviour. The first set of tests focuses on evaluating the actual non-uniform shrinkage profile from three types of slabs cast with no steel reinforcement which consist of solid slabs; solid slabs with their bottom face sealed using a plastic sheet and composite slabs. The occurrence of shrinkage has been monitored by measuring the variations of strains at different levels of the crosssection. Temperature readings have also been recorded. These samples were able to move freely as placed on a large number of steel balls. The second set of tests has been carried out on four simply supported beams. Two slabs have been kept unloaded and subjected to shrinkage effects only while the remaining two have been subjected to a sustained load after four weeks from casting. Only representative results of these tests are reported and concluding remarks on the observed shrinkage profiles are provided.
The experimental tests carried out are sub-divided into two sets. The first set of experiments includes the long-term testing of nine shrinkage slabs. For clarity only the detailed results of three typical tests representative of the three types of specimens prepared are reported. These are referred to as SS1, SS2 and SS3. The size of each slab is 1200 mm x 1200 mm x 125 mm. Shrinkage development is measured over time on samples placed on steel balls. This arrangement attempts to eliminate possible restraints provided by the supporting floor. Regarding the samples, SS1 consists of a solid slab, SS2 is a solid slab with a plastic sheet attached to its bottom face to avoid moisture exchange to take place and SS3 is a composite slab. The only difference between SS2 and SS3 relies on the fact that the latter sample possesses an internal restraint provided by the steel deck. The results observed during these tests are very interesting as they demonstrate how the final deformation in a solid slab is uniform despite the fact that the actual shrinkage profile is non-uniform. Despite this it is acceptable to simply use a uniform shrinkage profile for design purposes. In the case of SS2 the total deformation remains linear while generating a clear curvature. Values of shrinkage measured on the top face are around 20-30% greater than those observed for SS1. The values recorded at the bottom face remained within 100 microstrains. The difficulty in measuring the strains in such a situation is that the shrinkage curvature would tend to bend the slab upwards while its self-weight would oppose such deformation. This implies that the total deformations are a combination of these two contributions mixed to different magnitudes with time. The profile observed for SS3 appears to be very similar to the one obtained for SS2 despite the additional restraint.
The second set of experiments consists of four simply supported slabs of length 2.6m. These have been labelled as CS1, CS2, CS3 and CS4 respectively (where CS stands for Composite Slab). Slabs CS1 and CS2 have been subjected to a sustained load at 27 days after concrete casting while the companion slabs have been kept unloaded.
Overall it can be concluded that the presence of the steel deck or a sealed face produces a shrinkage induced curvature not present for a solid slab with no reinforcement. This effect is counter balanced, when comparing the response of solid and composite slabs, by the fact that the overall axial loading is smaller than the one observed in fully exposed slabs. For design purposes it has been suggested to use a linear varying shrinkage profile with a top value of approx. 1.25 times the free shrinkage strain commonly available in codes and a corresponding bottom value between 0–25%, whichever produces the worst scenario. It is also suggested that this approach needs to be verified by additional long-term tests.