Modelos de análisis del comportamiento en fractura de las juntas secas postensadas de estructuras construidas por dovelas

Zusammenfassung

Dry joint are generally used in the construction of prestressed segmental bridges due to its advantages such as time and cost saving. The shear capacity of the joints plays an important role in the behavior of the whole structure. However, the formulations of the diverse design codes lead to very different values of the shear strength of multiple-keyed joints. In this thesis three different finite element models have been developed to analyze the fracture behavior of post-tensioned dry keyed joints under shear loading. In the first model, the Brittle Cracking model implemented in Abaqus has been used for the material. Secondly, the joint has been modeled with the Total Strain Rotating Crack model of Diana. The principal advantage of this model is that it does not require the definition of a shear retention factor, which needs still further investigation while the influence on the results is not negligible. Finally, the Sequentially Linear Analysis is applied to model the joint. This supposes a significant advance because the results include the post-peak behavior with snapbacks, avoiding convergence problems, typical in brittle structures. It has been developed a numerical model of a joint with different number of keys and prestressing stress, and a regression equation of the obtained results has been presented. This equation estimates the shear capacity of the joint depending on the number of keys and the prestressing stress (until 3 MPa) for a characteristic concrete compressive strength of 50 MPa. The results have been compared with the existing formulas proposed by different design codes and authors, turning out that the ATEP formula underestimates the shear capacity of the joint, and the AASHTO formula overestimates it in the case of multiple keys and low prestressing force. The results show that the average shear stress transferred across the joint decreases as the number of keys increases, but this effect declines with higher prestressing stress in the joint, becoming negligible for 3 MPa. It can be concluded that for higher prestressing stress the behavior is independent of the number of keys. Consequently, if the existing formulation to estimate the shear capacity of dry keyed joints is applied for prestressing stress superior to 3 MPa, it would not be necessary to include any factor taking into account the dependence on the number of keys. In the appendix chapter is presented the research carried out in order to design the test set-up to obtain experimentally the shear strength of dry keyed joints.