Experimental and Numerical Investigations into Optimal Partial Concrete Filling of FRP and Steel Tubular Poles

Abstract

Glass fibre-reinforced polymer (GFRP) tubular poles can be superior to conventional poles, in that they are lighter in weight and more durable. Thin-walled tubular poles, however, tend to fail in flexure by local buckling, before fully utilizing the high tensile strength of GFRP. Increasing the wall thickness would solve this problem, but at a significant material cost. A simple and economical solution is to partially fill the tube with concrete. The aim of this study is to establish the optimal length of concrete filling that is required to achieve the highest moment capacity at a minimum dead weight in cantilevered GFRP and steel poles. The study comprises experimental and numerical phases. Six 3660 mm long and 220 mm in diameter GFRP tubes of 4.15 mm wall thickness as well as four 1855 mm long and 114 mm in diameter steel tubes of 3 mm wall thickness, were filled with concrete of varying lengths, ranging from zero to a 100% of the span. The tubes were tested to failure in cantilever bending. The completely filled tubes achieved nearly double the strength of the hollow ones. Furthermore, it was found that the optimal ratio of concrete filling length was 0.34 and 0.46 of the span, for the GFRP and steel tubes, respectively. This is defined as the minimum filling length required to achieve the capacity of the completely filled tube. Numerical models have been developed to predict the behaviour of partially concrete-filled GFRP and steel tubes as well as the optimal filling ratio. The models incorporate other models developed for hollow and completely filled tubes and account for the slight non-linearity of multi-layer GFRP tubes, concrete, and steel plasticity. An important feature of the models is their ability to account for ovalization and local buckling of the hollow part of thin tube. The models were successfully validated and used in a parametric study to investigate the effects of key parameters, namely diameter-to-thickness (D/t) ratio, GFRP laminate structure and steel yield strength. It was shown that the optimal filling ratio increases as D/t ratio is reduced or as more GFRP fibres become oriented longitudinally. However, it was unaffected by the steel yield strength.