End-region behavior of precast, prestressed concrete I-girders employing 0.7-inch diameter prestressing strands

Pretensioned concrete girders are currently fabricated using 0.5- or 0.6-in. diameter prestressing strands. In recent years, however, it has become of interest to employ larger-diameter 0.7-in. diameter strands to reduce the number of strands and improve the efficiency of pretensioned concrete members. Such a transition requires a considerable initial investment that needs to be justified based on the benefits obtained. Furthermore, the use of 0.7-in. strands would increase the stresses within the end-region of pretensioned elements, which could lead to undesirable cracking and impact the serviceability of the girders. The work presented in this thesis consists of 1) a comprehensive parametric investigation to evaluate the benefits and limitations of using 0.7-in. strands in pretensioned bridge girders, and 2) a full-scale experimental study to investigate the behavior of pretensioned concrete girders with 0.7-in. strands at the time of prestress transfer. The parametric investigation was accomplished by designing thousands of bridge girders with different span lengths, concrete release strengths, and transverse spacings. The results showed that the most noticeable benefit of 0.7-in. strands over 0.6-in. strands was a reduction of up to 35 percent in the number of strands. However, the difference in the total weight of prestressing steel was insignificant. Increasing the release strength of concrete, at least to 7.5 ksi, was found essential to observe benefits in design aspects other than the number of strands. The experimental investigation involved the fabrication of two Tx46 and two Tx70 specimens at the Ferguson Structural Engineering Laboratory. All specimens employed 0.7-in. strands on a 2- by 2-in. grid and the standard detailing currently used for girders with smaller-diameter strands. The observed crack widths in the specimens upon prestress transfer did not exceed those typically observed in Tx-girders with smaller-diameter strands. Therefore, the use of 0.7-in. strands does not seem to trigger a need to modify the end-region detailing in Tx-girders. However, noticeably greater bursting and spalling forces were observed in the end regions of the specimens compared to the demands predicted by AASHTO LRFD provisions. The measured 24-hour transfer length from the specimens also exceeded estimates by AASHTO LRFD and ACI 318-14 provisions.