Interaction diagrams for square concrete columns confined with fiber reinforced polymer wraps

Abstract

Experimental investigations are conducted to study the increase in axial capacities of short concrete columns due to confinement with fiber reinforced polymer (FRP) wraps. Experimental evidences gathered in this work and supported also by a few past studies suggest that concrete made from aggregates with lower unit weight and higher porosity offer lower modulus of elasticity. However columns made of such concrete must have larger dilational property under load. This fundamental perception motivated the current research to evaluate the attainable axial capacity enhancement in concrete columns made of different aggregates available in Bangladesh. In this context, confined and unconfined concrete column specimens made using stone, brick, recycled stone and recycled brick aggregates were tested under uniaxial compression. Carbon-FRP and glass-FRP bonded to the surface of concrete with epoxy offered different confining effect on the specimens. To measure the dilation effect, simultaneous measurements are taken by using a digital data acquisition system and image analysis technique. The vertical load and displacement histories obtained from the load cell of a computer controlled universal testing machine is synthesized with the strain measurement results gathered from analyzing of high definition still images. High speed (60 frames per second) video camera was used to capture digital images. In this process, lateral strains and Poisson’s effects are measured and the confinement due to dilation of concrete for different aggregate types are evaluated and compared. To assess the Poisson’s ratio in concrete with different aggregates, numerical trials on nonlinear finite element solid model of columns are conducted. In the numerical trials, the experimental measurements on load-vertical strain and load-horizontal strain are matched with the numerical results. It is seen that upon loading, the confinement effect in concrete is mobilized due to the dilation property. Concrete under load begins to dilate and hence gets cracked. The dilation effect in brick aggregate concrete was found distinctly larger than the stone aggregate concrete. The Poisson’s ratio of stone aggregate concrete was also found convincingly lower than brick aggregate concrete, recycled brick aggregate concrete and recycled stone aggregate concrete by FE analysis. The values of Poisson’s ratio of the concretes were estimated at 0.25, 0.35, 0.37 and 0.40, respectively. This direct observation suggests that the confinement effect due to dilation of brick and recycled aggregate concretes are higher than stone aggregate concretes. The modulus of elasticity of stone aggregate concrete was also measured higher than brick and other recycled aggregate concretes with a lower confined compressive strain. Based on the measurements, a set of linear relations between confined compressive strength to unconfined compressive strength and confined compressive strain to unconfined compressive strain of concrete columns are suggested for each of the aggregate types. The stress-strain model considering the Poisson’s ratio and modulus of elasticity is proposed for stone, brick, recycled stone and recycled brick aggregate concrete. The values of strain enhancement coefficient and confinement effectiveness coefficient were found to be larger for stone aggregate concrete. The P-M interaction diagrams are constructed to illustrate the enhancement in consequential bending moment capacity of column due to larger peak confined compressive strains. The enhancement in moment capacity due to FRP confinement of stone aggregate concrete column was distinctly found to be lower than brick and recycled aggregate concrete columns. Furthermore, experimental, analytical and numerical procedure developed in this work is applicable for constructing rational P-M interaction diagrams for confined short concrete columns, in general. P-M interaction diagrams drawn using the suggested procedure can be applied for the design of confined columns for both axial load and bending moment that fall above the line connecting the origin and balanced point of P-M interaction curve.