- Home
- Documents
*Experimental and numerical analysis of concrete slabs ... concrete slabs prestressed with composite*

prev

next

out of 12

View

0Download

0

Embed Size (px)

Experimental and numerical analysis of concrete slabs prestressed with composite reinforcement R. Sovják1, P. Máca1, P. Konvalinka1 & J. L. Vítek2 1Experimental Centre, Faculty of Civil Engineering, Czech Technical University in Prague, Czech Republic 2Department of Concrete and Masonry Structures, Faculty of Civil Engineering, Czech Technical University in Prague, Czech Republic

Abstract

The behaviour of concrete slabs prestressed with glass fibre reinforced polymer (GFRP) composite bars is investigated in this paper. The main advantage of GFRP bars is their high strength/self-weight ratio. On the other hand, Young’s modulus is very low compared to steel reinforcement, which is the main cause for unacceptable deflections. To eliminate such deflections and to utilize the high tensile strength of GFRP bars it is very useful to pretension the bars. During the experimental work a set of three concrete slabs 4.5 m long was casted. Each slab was prestressed with four GFRP bars. The slabs were subjected to four points bending. Each specimen was subjected to ten tow-load cycles and afterwards loaded until failure. Deflection under each loading point and in the middle of the beam span was recorded. Moreover, stress at the lower and upper surface was measured. The experimental procedure was modelled numerically in finite element nonlinear software. Brick elements were used for meshing and the full Newton-Raphson method was used for calculation. Special bond slip-law (GFRP-concrete) is involved based on the experimental results of pull-out tests. A stress-strain diagram and the stress and crack development were plotted. The experimental and numerical results show a statistically important relationship. A large deflection typical for GFRP reinforced slabs is observed, as well as very early crack propagation. The serviceability limit state (SLS) is exceeded much earlier than bar rupture is observed. For this reason, it is recommended that the design of GFRP reinforced structures is governed by SLS criteria. Keywords: composites, concrete, GFRP, numerical modelling, prestressing.

Computational Methods and Experimental Measurements XIV 83

© 2009 WIT PressWIT Transactions on Modelling and Simulation, Vol 48, www.witpress.com, ISSN 1743-355X (on-line) doi:10.2495/CMEM090081

1 Introduction The service lifetime of reinforced concrete structures is in many cases determined by the durability of the reinforcement itself. For this reason much attention is being paid to improving the resistance of the reinforcement to aggressive environments. Because steel bars are currently the most used type of reinforcement, several options of improving its corrosion resistance have been developed. One way is to improve the properties of the concrete itself by decreasing its permeability, increasing concrete cover and waterproofing the concrete. The other way is to use epoxy coated bars. However, it has been proven that none of these measures or their combination can eliminate the long- term risk of steel corrosion [1]. The ultimate way to provide corrosion resistance is to use bars made from stainless steel, but this solution is highly expensive and not always possible. Therefore, in recent years non-metallic reinforcement has gained a great deal of interest from many researchers. Fibre-reinforced polymer (FRP) reinforcement is by nature corrosion resistant. Therefore it can be successfully used in highly corrosive environments such as bridge decks, off-shore structures and slabs in chemical factories where high corrosion resistance is required. Furthermore, FRP reinforcement has other great properties such as magnetic transparency, thermal non-conductivity and generally higher tensile strength than steel. The behaviour of FRP bars, however, is different than that of steel and is highly dependent on the type of fibre and the production process. FRP reinforcement is linear-elastic up to failure and its elastic modulus is typically lower than that of steel. For instance FRP bars that are reinforced with glass fibres (GFRP) have typically a modulus of elasticity of 15 to 25% of steel. Furthermore, there are FRP bars with many different surface textures available on the market and every producer makes a unique type of bar. For this reason the design of structures reinforced with FRP bars must be approached with maximal caution. As mentioned above the low elastic modulus is the biggest issue that needs to be considered when designing concrete structures that are reinforced with GFRP bars. The main problem is that the overall stiffness of GFRP reinforced concrete member decreases significantly after the concrete cracks in the tension zone. From the cracked section analysis [2] it can be easily calculated that deflection and crack widths will be much larger for concrete member reinforced with GFRP bars compared to that reinforced with steel. One of the methods how to eliminate unacceptable deflections on serviceability limit state (SLS) is to pretension the GFRP reinforcing bars. This paper describes both experimental and numerical analysis of concrete slabs that are reinforced with GFRP pre-tensioned bars. Many researchers reported [3, 4, 10] that pretensioning of the GFRP bars help to decrease deflections and crack widths.

1.1 Research significance

Composite reinforcement is relatively new material and it has a potential to be widely applied in structures where special properties such as corrosion resistance

84 Computational Methods and Experimental Measurements XIV

© 2009 WIT PressWIT Transactions on Modelling and Simulation, Vol 48, www.witpress.com, ISSN 1743-355X (on-line)

or magnetic wave transparency are required. In fact, FRP reinforcement has been implemented in several structures all over the world. However, composite materials, including FRP, are very dependent on the production technology and process. Therefore every producer creates a unique product which is very similar to other products but not exactly the same – it is like with finger prints. FRP bars consist of two main components - fibres and matrix. Every component can be chosen accordingly to the specific needs of the customer. This gives the much needed adjustability and variability of building materials. Moreover the final characteristics of the entire rod will be different than characteristics that would be obtained by simple summation of properties of fibres and matrix. This effect is called synergism [5]. Based on these facts it is very important to conduct extensive research in this specific area in order to describe, clarify and completely understand behaviour of concrete structures reinforced with FRP reinforcement which is available locally.

2 Material characteristics

Before the loading tests were performed exact mechanical characteristics of materials used during this research were determined. This is very important as it helps to eliminate the number of variables. Elastic modulus, modulus of rupture (MOR) and compressive strength of concrete were measured. Cylinders were used for the modulus of elasticity measurement and 150 mm cubes were used for the ultimate compressive strength determination. The average results of these measurements are summarized in Table 1. Furthermore, basic properties of GFRP were determined such as modulus of elasticity and axial coefficient of thermal expansion. The tensile strength was determined by the producer and was not checked by the researchers. These material characteristics are compared with standard steel properties in Table 2.

Table 1: Material characteristics of used concrete.

Size of test specimen Average value Compressive strength cubes 150 mm 36,55 MPa MOR beams 100x100x400 mm 6,08 MPa Modulus of elasticity cylinders Ø150x400mm 30,8 GPa

Table 2: Material characteristics of GFRP bars in comparison with standard steel.

GFRP Steel B500 Tensile strength [MPa] 650 500 Modulus of elasticity [GPa] 40 210 Axial coefficient of thermal expansion [1/°C]

6e-06 12e-06

Stress-strain diagram Linear-elastic Bi-linear with hardening

Computational Methods and Experimental Measurements XIV 85

© 2009 WIT PressWIT Transactions on Modelling and Simulation, Vol 48, www.witpress.com, ISSN 1743-355X (on-line)

3 Experimental procedure

Experimental work was done in laboratories of Experimental Centre in the faculty of Civil Engineering, Czech Technical University, Prague. Three slabs reinforced with prestressed GFRP bars were tested in four point bending test. The slabs were 4.5 m long with a clear span of 4 m. The loading points were approximately in thirds of the span. The setup of the experiment is shown in Figure 1 and the cross section of the slab is shown in Figure 2.

Figure 1: Slab dimensions.

Figure 2: Cross-section.

Each slab was reinforced with four GFRP bars with diameter of 14 mm. The bars were pretensioned to stress 215 MPa. This stress level corresponds to one third of the ultimate tensile strength of the bar as provided by the producer. At both ends of the slab three stirrups (at 50 mm centres) were inserted to provide reinforcement for the lateral stresses induced by the pre-tensioned bars. It can be seen in Table 1 that the concrete used to cast the slabs had average compressive strength of 36 MPa and the concrete cover was 43 mm.

3.1 Pretensioning procedure

The pretensioning procedure was relatively simple. A wooden mould was constructed on the floor of the lab and both stirrups and GFRP tendons were inserted inside. GFRP tendons were 6 m lo