Seismic Retrofit Of Bridges Supported On Hollow Core Prestressed Concrete Pile Columns

Download or read book Seismic Retrofit of Bridges Supported on Hollow Core Prestressed Concrete Pile-columns written by Samuel Turner and published by . This book was released on 2020 with total page 215 pages. Available in PDF, EPUB and Kindle. Book excerpt: The Washington State Department of Transportation (WSDOT) began a bridge seismic retrofit program in 1991 in order to address the earthquake risks associated with state-owned bridges. The majority of these bridges were constructed in the 1960s, prior to the development of modern seismic design standards during the 1970s and 1980s, and one class of bridges built during that era represents a particular cause for concern. In that class of bridges, circular, hollow-core precast, prestressed concrete piles were driven into the ground, with sufficient length left projecting above ground level to form the columns. Hence, they are referred to in this thesis as "pile-columns." Cap beams were then cast in place on the pile-columns and were connected to them by means of concrete plugs with longitudinal reinforcement cast into the tops of the columns. Many of these hollow pile-column bridges form part of the state's "Lifeline Routes" for seismic events and are consequently slated for retrofit in the near future. Previous, but limited, experimental research have shown that hollow, prestressed concrete specimens show minimal ductility under cyclic lateral loading and fail suddenly, because the wall of the column spalls both inwards and outwards. However, cyclic lateral loading performed during an earlier phase of this research program on an as-built scaled model specimen of the column-to-cap-beam connection showed better ductility than expected. In this portion of the research program, reported in this thesis, experiments were conducted to better understand the behavior of the hollow core pile-columns under pure bending (relevant in the pile region below-grade) as well as combined bending and shear (relevant in the column region above-grade). Finally, a column-to-cap-beam connection retrofitted with a carbon fiber jacket was tested under cyclic lateral loading. Under elastic conditions, the peak moment demand in the pile-column occurs at column-to-cap-beam connection; the moments below-grade are approximately half as large. As a result, the retrofit concept for these bridges is concentrated on that connection. The tested retrofit option involved jacketing the as-built plug region in a carbon fiber wrap on the basis that that procedure is simple and reliable, it slightly lowers the peak moments below-grade, and, if a fiber jacket could be used over only the upper part of the column, it would avoid invoking environmental restrictions for bridges over waterways. Concerns with this, and any other retrofit option, include vulnerability to combined bending and shear failure of the hollow cross section just beyond the end of the plug, as well as susceptibility to pure bending failure of the hollow pile below-grade. Three experimental programs were conducting during this investigation. The first and second were conducted to determine the pure bending, and combined shear and bending, strengths of a hollow pile-column section. The results of these tests aided in calibrating the transverse load capacity of hollow prestressed concrete specimens under two separate load configurations. The third experiment was conducted on a scaled cantilever column-to-cap beam connection, retrofitted with a fiber jacket around the plug region, under reversed cyclic displacement. The results of the third experiment showed that the retrofitted pile-column performed slightly better than the as-built specimen, and that the critical mode of failure was not internal spalling, as suggested by previous researchers, but rather by column cracking, followed by strand debonding. Review of the test results from the as-built specimen concluded that that specimen, too, had failed by strand debonding. Debonding failure proved to be non-ductile, because the strands buckled during the half-cycle following debonding, then fractured as they re-straightened. It was further concluded that the fiber jacket had provided little benefit because strand debonding, and not concrete confinement, was the primary failure mechanism. A third retrofit option was then developed, but limitations on resources prevented it from being tested. Strand debonding can occur only after the column wall cracks, which requires high net tension stress, due primarily to bending. A thick steel jacket, made composite with the concrete components, would increase the flexural stiffness of the region and reduce the bending stress in the concrete column wall, thereby preventing cracking. A methodology for selecting the jacket thickness was developed, and it is recommended that this retrofit option be tested and, if the test is successful, it should be adopted.