Integral precast girder-to-cap connections for accelerated bridge construction in seismic regions
Accelerated bridge construction (ABC) is increasingly desired and needed, due to the aging transportation infrastructure across the United States and the always-growing demand placed on our nation's highway system. Precast concrete is a common way to incorporate ABC techniques. Advantages over typical cast-in-place concrete methods include speed of field construction, improved quality control, and decrease in detoured traffic during construction, among others. However, precast concrete structures have not been used to their full potential in high seismic regions, due to the deficiency of precast concrete connections in past earthquake events. The California Department of Transportation (Caltrans) is eager to incorporate ABC methods if connections suitable for high seismic regions can be developed. Therefore, a study has been conducted to investigate the inverted-tee cap beam and I-shaped girder bridge system for its viability for implementation by Caltrans. A large-scale experimental investigation of the bridge system was conducted, verifying that the system has excellent potential for such use. The study identified an as-built connection detail that has been previously incorporated by Caltrans as being capable of providing an integral moment girder-to-cap connection. However, the study also introduced an improved connection detail utilizing grouted unstressed strands, similar to those used in post-tensioning applications, that has the promise of providing an even better connection alternative. A follow-up large-scale experimental study was conducted to provide a detailed investigation of the improved detail. In addition, the follow-up study was used to quantify the performance of another new girder-to-cap connection detail utilizing looped strands and dowel bars. Both connection details were verified to be very constructible and to provide excellent seismic performance, even when subjected to vertical acceleration demands significantly beyond typical design recommendations. Along with connection behavior, these experimental studies were used in conjunction with analytical approaches to investigate current approaches related to load distribution in integral bridges. This work showed that current recommendations are overly conservative in the amount of the column seismic moment that is required to be carried by adjacent girders in the superstructure. A better distribution model, based on the relative stiffness of the superstructure components, is proposed that matches well with the analytical and experimental results from this study and three other large-scale experimental seismic studies. Finally, analytical approaches for the incorporation of vertical acceleration effects were considered, and the results were used to verify the observed experimental performance of the proposed girder-to-cap connection details.