Experimental and Numerical Investigation of End Zone Crack Control Strategies During Detensioning of Pretensioned Concrete Bulb-Tee Girders
Files
TR Number
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Prestressed concrete bulb-tee girders (PCBTs) are commonly employed in bridges to allow for efficient design and construction due to their cross-sectional properties and offsite constructability. However, PCBTs are susceptible to end zone cracks which develop in the web due to high vertical tensile stresses caused by prestress force transfer from the strands into the concrete at beam ends. This issue begins during strand detensioning and is exacerbated when the beams are lifted from a precast bed. Typically, these cracks are small enough that they do not impact the strength limit state; however, they pose serviceability concerns by providing potential avenues for water/chloride ingress, which may lead to premature corrosion of the reinforcement and reduce the service life of a structure. Despite being designed in accordance with current practice, prestressed girder end zones often result in highly congested reinforcement details and may still develop crack widths that exceed serviceability thresholds, requiring costly repairs such as epoxy injection or surface coatings. Overall, the objective of this dissertation was to analyze end zone cracking in PCBTs and to provide practical solutions that the Virginia Department of Transportation can implement to combat reinforcement congestion and excessive crack widths. An experimental and numerical program was conducted with the goal of providing a design methodology that more reliably ensures crack width standards are met while reducing reinforcement congestion. First, the investigation focused on experimental testing of concrete prism specimens reinforced with traditional deformed bars and welded grid reinforcement (WGR) designed to mimic half of a PCBT web with the goal of evaluating the performance of WGR versus traditional deformed bars. Following this, five 45 in. deep PCBTs end zones were fabricated to experimentally evaluate the effects of reinforcement amount and type (WGR versus traditional deformed bars) on end zone cracking. End zone cracking behavior was evaluated immediately after prestress release and after moving the beams from the casting bed to storage. In addition to experimental evaluation, nonlinear finite element analyses were conducted to evaluate the impact of various design choices on end zone cracking. Overall, more than 500 total models with girders ranging in depth from 29 in. to 93 in. were considered to evaluate the impact of strand pattern, strand debonding, reinforcement amount, reinforcement layout, reinforcement diameter, concrete material properties, and PCBT web width. These models were validated against experimental data. Key findings showed that demand reduction strategies were most effective at controlling cracks overall, with strand debonding and splayed draping being the most effective in reducing end zone cracking compared to traditional strand patterns. In addition, capacity modification strategies (e.g., adding more end zone reinforcement) are effective at controlling end zone cracking; however, these strategies are much more effective when combined with demand reduction strategies to control cracks and minimize reinforcement congestion. In addition, the use of welded grid reinforcement proved to be less effective at controlling cracking compared to traditional deformed bars, as welded cross wires acted as crack formers. Following this, an empirical design equation for PCBT end zone reinforcement was developed using a synthetic database of end zone cracking results from the finite element models. This equation provides designers with a simple tool to address factors impacting both end zone capacity and demand, integrating the effects of beam depth, concrete release strength, concrete density, strand debonding, and splayed draping. In addition, it allows for the input of a desired serviceability limit, as the user can define maximum crack width limits based on exposure conditions. This methodology provides a conservative design envelope that enables designers to avoid the burden of costly end zone crack repairs after detensioning.