Accelerated bridge construction techniques taking advantage of prefabricated bridge elements and high-performance materials are being used more frequently for bridge replacement projects. They result in minimal road closure times and traffic interruption and in the reconstruction of long-lasting highway bridges. Longitudinal closure pour connections are an important deck-level component for modular bridge elements that are heavily stressed by traffic loadings and environmental effects and whose durability is a concern. To address cracking and leakage issues in such connections, the strength and failure modes of the longitudinal ultrahigh-performance concrete (UHPC) closure pour connection between adjacent prefabricated deck units were evaluated. First, specimens with and without a longitudinal UHPC closure pour connection were fabricated, instrumented, and tested. Finite element (FE) models were established to improve understanding of the behavior of the specimens under the loading condition. In addition, strut-and-tie models (STMs) were developed on the basis of FE model predictions to estimate the strength of the specimens. The jointed specimens were found not to have any cracks or leakage at the early stage but had lower cracking loads than did the jointless specimens. The strength and ductility of the jointed specimens were comparable with those of the jointless specimens. On the basis of the FE models and STMs, the ultimate strength of the specimens was accurately predicted.

Because of the potential for concrete crushing, the negative moment transfer mechanism of the transverse connections above the pier support is still a point of concern. The objective of this study was to evaluate the behavior of a transverse closure pour connection used for the constructed Little Silver Creek Bridge in Iowa and to validate its adequacy for modular bridge application. To evaluate the need for the complicated compression block, two transverse connection specimens with and without the compression block were designed, fabricated, instrumented, and tested under a negative bending moment. Finite element models were established to further understand the specimens’ performance, and hand calculations were performed to estimate their moment capacity. No significant difference was found between the crack patterns of the two specimens, and the diaphragm concrete tended to crush in the bottom regardless of the configuration. The established finite element models were sufficient at representing the structural behavior of the two transverse connection specimens. The connection with a compression block had higher yield and ultimate moment capacity than the connection without a compression block. To design both types of connections according to the classic theory of reinforced concrete design, an effective width equal to the width of the steel girder bottom flange and the centroid of the compression force at the bottom of the steel girder bottom flange can be assumed for the connections with and without a compression block, respectively. Hand calculations reasonably estimated the moment capacity of the two specimens, and the two connections were safe under the codified, factored loads.

No abstract