A common bridge type is the steel-concrete bridge where the concrete deck is built over steel girders. In many earlier designs the bridge type was often built as non-composite, which means that the concrete deck and the steel girder has no shear connection at the steel-concrete interface and therefore bend as individual components. With the increased traffic loads of today some of the existing non-composite bridges have insufficient bending capacity, and therefore they must either be replaced or strengthened. To replace a bridge and construct a new one has many downsides, it is time consuming, expensive, and it consumes a lot of finite resources. Therefore, it is better if the bridges could be strengthened instead. Non-composite steel-concrete bridges can in some cases be strengthened by installing shear connectors that enable composite action between the concrete deck and steel girder. To enable full composite action, many shear connectors need to be installed (10-15 per meter). In some cases, full composite action is not needed to achieve a sufficient load capacity. Therefore, to save time and money and reduce material usage, it could be favourable if the amount of shear connectors could be lowered. The concept of using less shear connectors than required for full composite action is known as partial composite action and is defined as a ratio η that can vary between 0 and 1,0. If the ratio is 0, the structure is non-composite and if it is 1,0, it is fully composite. For every ratio between, the structure is partially composite. Partial composite action is not allowed by the standard for new composite bridges in Europe, EN 1994-2, which instead requires full composite action for new bridges. Since the conventional shear connector type, Welded Headed Stud (WHS) is impractical for post-installation this can yield large costs. This thesis therefore analyses the efficiency of strengthening non-composite bridges with partial composite action by post-installation of the shear connector type Coiled Spring Pins (CSPs), which is more suitable for post-installation compared to WHS since the installation can be made from underneath the bridge deck. The thesis consists of a theoretical study about composite action with a focus on partial composite action. In addition to the theoretical study, a case study is performed on an existing non-composite steel-concrete bridge, the bridge over Yxlö channel, which is situated south of Stockholm in Nynäshamn municipality. In the case study, hand-calculations to calculate the moment capacity for the bridge and the bending stresses in the bridge is made. In addition, a linear Finite Element-analysis (FE-Analysis) is made to evaluate the bending stresses in the cross-section. Further, in the FE-analysis, the horizontal slip and shear flow at the steel-concrete interface is evaluated. The calculations in the case study are made for 10 different degrees of shear connection from 0 - 1,0 with increments of 0,1. The results from the hand-calculations showed that partial composite action an efficient strengthening method, especially for lower degrees of shear connection. The moment capacity in the mid-section of the bridge could be increased between 16 and 41 % for shear connection ratios between 0,4 and 1,0, when applying plastic properties. If elastic properties were used, the increase in moment capacity for the same interval and section was 13 – 21 %, which shows that if it is possible to use plastic properties, the moment capacity could be increased more.  The results from the stress analysis in both the Hand- and FE-calculations showed that the stresses were reduced efficiently, especially for the top flange of the steel girder, where the stresses reduced 75-85 % for shear connection ratios between 0,4 & 1,0. The reduction of the stresses in the bottom flange were not as efficient, but still a reduction of 15 – 20 % is possible for shear connection ratio between 0,4 – 1,0. The overall conclusion from the thesis is that partial composite action can be an efficient strengthening method, and that non-composite bridges like the Yxlö Bridge could be strengthened with CSPs and have an effective increase of the bending moment capacity. This way the allowed axle- and bogie load on the bridge could be increased which could extend the technical life length of the bridge and reduce the need for new bridges.

Many of the existing steel-concrete bridges may need to be strengthened, as heavier vehicles areallowed on the Swedish roads. These bridges could possibly be strengthened by post-installingshear connectors. The shear connectors may enhance the load-bearing capacity through a higherdegree of composite action between the steel and concrete interface.For post-installing of shear connectors, it is advantageous to use a method that allows forinstallation from underneath the bridge as it avoids disrupting the traffic flow. The authors havehence focused on a shear connector called coiled spring pin (CSP); a sheet of metal rolled intoa coil. It’s inserted by hydraulic jacking into a pre-drilled hole and maintained in position dueto radial spring force, avoiding the need for welding.Information and data are collected from beam tests performed at Luleå technical university, theEurocodes and literature.This study investigates and identifies the behaviour and characteristics of a partial compositegirder reinforced with CSPs. The study compares the results obtained from the laboratory testsand the FEM-simulations. Furthermore, this research examines the factors that contribute to theaccuracy of the FEM models and investigates the influence of the CSP placement on the overallload-bearing capacity.Both the FEM simulations and laboratory tests indicate that the girders exhibit strength benefitsfrom applying CSPs. An optimal position for the connectors could not be determined, as theresults presented in the simulations was not proved by the laboratory tests. The simulationsindicate benefits with central placed CSPs, in contrast to the laboratory test where no differencesfrom the placement were shown, although only two test setups were used.

Deliberate and accidental explosions along with the heightened risk of loss of life and property damage during such events have highlighted the need for research in the behaviour of materials under high strain rates. Where an extensive body of research is available on steel and concrete structures, little to no details on how to address the design or retrofitting of wood structures subjected to a blast threat are available. Studies reported in the literature that focused on full scale light-frame wood structures did not quantify the increase in capacity due to the dynamic loading while the studies that did quantify the increase mostly stems from small clear specimens that are not representative of the behaviour of structural size members with defects. Tests on larger-scale specimens have mostly focused on the material properties and not the structural behaviour of subsystems. Advancements in design and construction techniques have greatly contributed to the emergence of taller and safer wood structures which increase potential for blast threat. This thesis presents results on the flexural behaviour of light-frame wood stud walls subjected to shock wave loading using the University of Ottawa shock tube. The emphasis is on the overall behaviour of the wall subsystem, especially the interaction between the sheathing and the studs through the nailed connection. The approach employed in this experimental program was holistic, where the specimens were investigated at the component and the subsystem levels. Twenty walls consisting of 38 mm x 140 mm machine stress-rated (MSR) studs spaced 406 mm on center and sheathed with two different types and sheathing thicknesses were tested to failure under static and dynamic loads. The experimental results were used to determine dynamic increase factors (DIFs) and a material predictive model was validated using experimental data. The implications of the code are also discussed and compared to the experimental data. Once validated, an equivalent single-degree-of-freedom (SDOF) model incorporating partial composite action was used to evaluate current analysis and design assumptions. The results showed that a shock tube can effectively be used to generate high strain-rate flexural response in wood members and that the material predictive model was found suitable to effectively predict the displacement resulting from shock wave loading. Furthermore, it was found that current analysis and design approaches overestimated the wall displacements.

Deliberate and accidental explosions along with the heightened risk of loss of life and property damage during such events have highlighted the need for research in the behaviour of materials under high strain rates. Where an extensive body of research is available on steel and concrete structures, little to no details on how to address the design or retrofitting of wood structures subjected to a blast threat are available. Studies reported in the literature that focused on full scale light-frame wood structures did not quantify the increase in capacity due to the dynamic loading while the studies that did quantify the increase mostly stems from small clear specimens that are not representative of the behaviour of structural size members with defects. Tests on larger-scale specimens have mostly focused on the material properties and not the structural behaviour of subsystems. Advancements in design and construction techniques have greatly contributed to the emergence of taller and safer wood structures which increase potential for blast threat. This thesis presents results on the flexural behaviour of light-frame wood stud walls subjected to shock wave loading using the University of Ottawa shock tube. The emphasis is on the overall behaviour of the wall subsystem, especially the interaction between the sheathing and the studs through the nailed connection. The approach employed in this experimental program was holistic, where the specimens were investigated at the component and the subsystem levels. Twenty walls consisting of 38 mm x 140 mm machine stress-rated (MSR) studs spaced 406 mm on center and sheathed with two different types and sheathing thicknesses were tested to failure under static and dynamic loads. The experimental results were used to determine dynamic increase factors (DIFs) and a material predictive model was validated using experimental data. The implications of the code are also discussed and compared to the experimental data. Once validated, an equivalent single-degree-of-freedom (SDOF) model incorporating partial composite action was used to evaluate current analysis and design assumptions. The results showed that a shock tube can effectively be used to generate high strain-rate flexural response in wood members and that the material predictive model was found suitable to effectively predict the displacement resulting from shock wave loading. Furthermore, it was found that current analysis and design approaches overestimated the wall displacements.