Seismic retrofit of masonry-to-timber connections in historical constructions

摘要

Proper structural connections play an important role in ensuring seismic loads distribution and developing global damage mechanisms of structures. In unreinforced masonry buildings, effective connections between masonry walls and timber floors or walls through the use of anchors can prevent the occurrence of out-of-plane mechanisms and promote "box-like" behavior. This thesis aimed at studying and developing seismic retrofit solutions for connections between masonry walls and timber structural elements, such as floors and timber framed walls, for historical unreinforced masonry buildings built in Lisbon, from the late 19th century to the beginning of the 20th century. However, similarities concerning constructive details of connections found on unreinforced masonry buildings across different countries, bring the possibility of expanding the solutions under study to other contexts. The tasks carried out focused on characterizing the tensile response of the connections, through an experimental campaign followed by analytical development of the results. Quasi-static monotonic and cyclic tests were carried out on pairs of injection anchors, to be applied on masonry-to-timber elements connections, and on wall-to-floor connections retrofitted with a system of steel tie rod with anchor plate. The pairs of injection anchors failed by combined cone-bond failure, with strength capacity and shape of the envelopes being highly influenced by the boundary conditions of the masonry wall. Hysteretic loops' shape is mainly influenced by bond slip at the grout/masonry interface, causing strength and stiffness degradation, and pinching. Unstrengthened wall-to-floor specimens' failure mechanism was nails pullout, while for the strengthened ones was failure of the bolted connection between steel angle and timber joist. The hysteretic behavior of the latter shared similarities with the one of the injection anchors, but it was controlled by slip at the bolted connection. There were considerable increases in maximum pullout force and dissipated energy, from unstrengthened to strengthened wall-to-floor connections, and ductility was kept high as well. From the experimental results, strength prediction formulae were studied and adapted to better fit the results obtained and idealized curves were developed based on force-displacement backbone curves. Existing behavior models provided good approximations to the experimental results but further validation needs to be implemented. The trilinear curves provided better approximations to the backbone curves, and ASCE/SEI 41-06 approach revealed to be very conservative. Retrofit design and acceptance criteria regarding displacements, forces and behavior factors were proposed.