Bond and cracking of reinforced concrete

Ribbed reinforcement is described as “high bond” in Eurocode 2 (EC2) and within the code serviceability checks make no allowance for variations in either the ductility or bond characteristics of these bars. In this work, this matter is explored, and the crack development and behaviour of concrete beams reinforced with various types of ribbed steel bar are investigated, using both numerical and experimental approaches. The objective of the experimental approach is to undertake a series of experiments to compare the performance of beams made with standard reinforcement with that of beams formed with a new high-ductility bar produced by CELSA UK. The relationship between the bond strength and the rib pattern of reinforcing steel was studied and the behaviour at SLS load levels of RC beams with reinforcement of different rib patterns in flexure is discussed. The cracking of beams was monitored both visually and using a non-destructive Digital Image Correlation system to trace in-plane deformations and strains on the surface of the specimens. The test results showed that specimens with bars which had the highest relative rib area (fR value) exhibited the smallest crack spacing and crack width. A numerical model was developed to explore the crack development of reinforced concrete beams under flexural loading. The model employed a non-linear material model for concrete and a smeared crack approach. In order to address the well known numerical stability problems, associated with softening models, a non-local gradient method was used. Crack widths cannot be obtained directly from such models, due to the diffuse nature of non- local simulations, therefore a post-processing procedure was developed to allow the crack characteristics to be calculated. Several numerical examples are presented to illustrate the satisfactory performance of the model. In addition, a series of numerical simulations of the BOND AND CRACKING OF REINFORCED CONCRETE Simon H.C. Chan Page vi experimental beams tested in the present study were used validate the numerical model and conversely, to provide confidence in the consistency of the experimental results