Effect of Flexural Cracks on Web-Shear Cracking of Prestressed Concrete Continuous Members

Abstract

Shear tension failure in thin-webbed prestressed concrete elements is yet to be predicted in an accurate manner. The analytical method uses assumptions to simplify the theory. This makes it less accurate. The calculated resistance to web-shear cracking is based on the principle tensile stress reaching the tensile strength of concrete. Failing in shear tension is more risky than failing in bending, since it often happens abruptly without any warnings. Previous research shows however that shear tension failure often occurs before the principle tensile stress reaches the assumed tensile strength of concrete. This means that the resistance is being overestimated. Comparisons between experimental results and theoretical models have shown that the coefficient of variation for the shear force at web-shear cracking for experiments including flexural cracks is significantly larger than for experiments without flexural cracks. It is assumed that flexural cracks are present when principle tensile stress in the outer fiber is higher than tensile strength in the flange. This indicates that the presence of flexural cracks influence the stress distribution in the areas where web-shear cracking occurs. To find the influence of flexural cracks on web-shear cracking a well documented experiment about continuous prestress concrete members failing in shear tension is used for a case study. Namely the thesis ‘’The Influence of Axial Load and Prestress on The Shear Strength of Web-Shear Critical Reinforced Concrete Elements’’ written by Liping Xie in 2009. Three of the specimen are chosen to be researched, between these three the only changing variable is the amount of prestress. All three fail under shear tension, however one has no observed flexural cracks, one has the flexural cracks and the web-shear cracks occur simultaneously and the last one shows flexural cracks before web-shear cracking. Three analyses are performed per specimen: analytical analysis, linear finite element analysis (LFEA) and a non-linear finite element analysis (NLFEA). These analyses are compared to each other and to the experimental results. An outstanding result is that for every load in every specimen it holds that: σ1_ANA > σ1_LFEA ≥ σ1_NLFEA. The maximum principle tensile stresses calculated with an analytical method are consistently higher than principle tensile stresses that follow from a LFEA and NLFEA. This means that performing a simple analytical analyses gives a lower shear resistance than the LFEA and the NLFEA. All three analyses indicated flexural cracks for the beam in which no flexural cracks were observed. However the NLFEA showed that these were cracks of a maximum crack width of mm at the moment of web-shear cracking. These widths are not visible by eye. All specimen showed web-shear cracks under a shear load lower than predicted, meaning all analyses overestimate the shear tension resistance. A sensitivity analysis is performed to find the influence of shear reinforcement and of the tensile strength of the concrete. Conclusions are that the shear reinforcement has little to no influence up until the moment of web-shear cracking. All analysis overestimate the resistance to web-shear cracking. In this thesis a sensitivity analysis is performed to the influence of the tensile strength of concrete. It is concluded that if the tensile strength of the concrete is reduced by 40% the NLFEA still results in shear forces larger than the shear force at web-shear cracking during the experiments. Another conclusions of this research is that the presence of micro flexural cracks reduce the principle tensile stresses in the adjacent web areas. The specimen in which no flexural cracks were observed during testing shows flexural cracks conform all analyses it is impossible to compare specimen uncracked in bending with specimen cracked in bending. Up until reaching the tensile strength of the concrete the LFEA and the NLFEA are logically equal. When the load is further increased differences between the LFEA and the NLFEA occur. It is assumed that this difference is the result of the presence of the flexural micro cracks. The specimen with the smallest flexural micro cracks showed also the smallest difference between σ1_LFEA and σ1_NLFEA From this it can be concluded that larger cracks result in a larger reduction of σ1. It is expected that if the micro flexural cracks evolve to significant flexural cracks that the principle stresses are reduced even more, resulting in a underestimation of the actual shear tension resistance.