Investigation of cracking due to restraint forces in Swedish concrete slab frame bridges

Abstract: When concrete slab frame bridges are designed with modern linear 3D FE-models large tensile restraint forces will arise in the transversal direction of the bridge from the temperature and shrinkage loads. These restraint forces are much larger than those that are accounted for in the older 2D models. This leads to that the amount of transversal reinforcement, calculated in the 3D FE-calculations, is much higher then what has been standard earlier. This thesis investigated whether or not cracking due to restraint forces is a problem for existing Swedish concrete frame bridges. This was done through modelling existing bridges with 3D FEM. Two different reinforcement models were applied and the calculated amount of reinforcement was compared to the existing reinforcement. Furthermore different types of concrete slab frame bridges were inspected. The different types were bridges with open or closed bottom plate, bridges where the valve was skewed or not and bridges that either crossed a pedestrian/bike path or a car road. Finally the bridges were investigated ocularly to help decide if restraint cracking is a real problem or a problem for the 3D FEM. The results of this investigation showed that cracking was common for bridges that crossed pedestrian/bike paths; however, the crack width was usually within the design limits. Furthermore the results of the comparisons between the two reinforcement models showed that there was a large difference depending on how the restraint forces are accounted for. The standard Eurocode method, where no special regards are taken to the restraint forces, could give an overestimation up to 2000%. The other method, where special regard is taken to the restraint forces, only gave an overestimation up to a maximum of 100%.