Arturo E. Schultz, Eray
Baran, and Catherine W. French
Department of Civil Engineering, University of Minnesota, 500 Pillsbury Drive SE, Minneapolis, MN 55455
Prestressed concrete through-girder pedestrian bridge systems consist of two prestressed concrete girders that support reinforced concrete cast-in-place floor beams at their bottom flange and a reinforced concrete cast-in-place deck placed on top of the floor beams. Mn/DOT Type 63 girder cross section is used on these bridges with a typical span of 125-135 ft. and a typical spacing of 12-15 ft. between the two girders.
Two issues have recently been raised regarding the prestressed concrete through-girder pedestrian bridge system, which has been widely used in the State of Minnesota. The first issue concerns the ductility of prestressed concrete girders in these bridges because the section that is typically used may be considered to be over-reinforced according to AASHTO LRFD Bridge Specifications. Response of the section, including neutral axis location, strand stress at ultimate capacity, and moment capacity, predicted by AASHTO Standard and AASHTO LRFD Specifications are compared with the sectional response determined from nonlinear strain compatibility analyses. Modifications are proposed to the AASHTO LRFD procedure to rectify the errors in predicting sectional response.
The second issue that was investigated in the study was regarding the strength and stability of prestressed concrete through-girder pedestrian bridges when subjected to striking by overheight vehicles. Three-dimensional full-scale finite element models of an entire bridge system as well as bridge subassemblages were used to evaluate the strength, stiffness, and ductility characteristics of the bridge system and connection details. Accurate representation of the bridge details in the finite element models were assured by utilizing the experimentally determined load-deformation characteristics of these connections in the finite element models.
Three series of laboratory tests were conducted in order to investigate the performance of currently used and proposed details to be used in the future construction for prestressed concrete through-girder bridges. Results from these tests were either directly incorporated in modeling of the behavior of the components in the finite element models or the experimental data was used to calibrate the subassemblage finite element models. Performance of a typical prestressed concrete through-girder pedestrian bridge system was analyzed through three sets of finite element analyses using the models of an entire bridge system. Each of these three sets of analyses aimed at studying a different aspect of the bridge system behavior.
The pull-out tests performed on steel inserts indicated that the type of inserts currently being used in prestressed concrete bridge girders in the State of Minnesota has the ability to undergo significant amounts of plastic deformation without a reduction in the load capacity. The ductile behavior of the steel inserts used in prestressed concrete girders was also confirmed by the results from the connection subassemblage tests. The connection subassemblage tests also revealed that the behavior of these inserts is significantly affected by the construction method followed during the fabrication of prestressed concrete girders. Results obtained from the girder end detail specimens indicated two types of horizontal load resisting mechanisms depending on the type of detail. During testing of the girder end detail specimens, large values of lateral displacements following the peak load capacities were measured with some level of residual load capacity.
The static lateral load finite element analyses indicated significantly different bridge response depending on whether or not the flexibility of the girder supports were included in the models. It was also determined that the load transfer mechanism among the bridge components depends on whether the girders were loaded at the exterior or interior face. Results from these analyses also showed that the lateral load and deformation capacities of the bridge system could be improved by increasing the ductility and strength of the connection between the girders and floor beams.
The dynamic lateral impact analyses that were performed in an attempt to determine the demand that would occur on the bridge system indicated relatively small impact durations. The dynamic analyses revealed a different deformation pattern of the bridge system than the deformation patterns observed in the static analyses. The damage in the bridge caused by the impacting body was observed to remain highly localized near the impact location for approximately half of the impact duration. As a result, the support flexibility of the girders did not have much effect on the dynamic behavior of the bridge, as opposed to the behavior observed in the static lateral load analyses. The Equivalent Static Force (ESF) values determined from the dynamic analyses were smaller than the static lateral load capacity of the bridge for the cases with flexible girder supports, while for the rigid supports the ESF values were still larger than the static lateral load capacities.
Results of the stability analyses indicated that the local girder damage that would occur in prestressed concrete through-girder pedestrian bridges due to striking of over-height objects may cause the failure of the bridge depending on the extent of damage that the girders would be subjected to. The bridge was determined to be more susceptible to failure when the impact damage occurs near the girder midspan than the girder ends. When only one of the girders was impacted, failure of the bridge would require slightly larger amount of damage in the girder section for failure than the damage level required for failure when both girders are damaged. The amount of “additional capacity” between the cases of single girder versus the both girders being damaged is due to load redistribution from the impacted girder to the other girder. Analysis results showed that the in the case of both girders being impacted, failure of the bridge would occur when approximately 15 percent to 40 percent of the girder web, depending on the location of impact, was damaged in addition to the entire bottom flange.
Department of Civil Engineering, University of Minnesota, 500 Pillsbury Drive SE, Minneapolis, MN 55455
Prestressed concrete through-girder pedestrian bridge systems consist of two prestressed concrete girders that support reinforced concrete cast-in-place floor beams at their bottom flange and a reinforced concrete cast-in-place deck placed on top of the floor beams. Mn/DOT Type 63 girder cross section is used on these bridges with a typical span of 125-135 ft. and a typical spacing of 12-15 ft. between the two girders.
Two issues have recently been raised regarding the prestressed concrete through-girder pedestrian bridge system, which has been widely used in the State of Minnesota. The first issue concerns the ductility of prestressed concrete girders in these bridges because the section that is typically used may be considered to be over-reinforced according to AASHTO LRFD Bridge Specifications. Response of the section, including neutral axis location, strand stress at ultimate capacity, and moment capacity, predicted by AASHTO Standard and AASHTO LRFD Specifications are compared with the sectional response determined from nonlinear strain compatibility analyses. Modifications are proposed to the AASHTO LRFD procedure to rectify the errors in predicting sectional response.
The second issue that was investigated in the study was regarding the strength and stability of prestressed concrete through-girder pedestrian bridges when subjected to striking by overheight vehicles. Three-dimensional full-scale finite element models of an entire bridge system as well as bridge subassemblages were used to evaluate the strength, stiffness, and ductility characteristics of the bridge system and connection details. Accurate representation of the bridge details in the finite element models were assured by utilizing the experimentally determined load-deformation characteristics of these connections in the finite element models.
Three series of laboratory tests were conducted in order to investigate the performance of currently used and proposed details to be used in the future construction for prestressed concrete through-girder bridges. Results from these tests were either directly incorporated in modeling of the behavior of the components in the finite element models or the experimental data was used to calibrate the subassemblage finite element models. Performance of a typical prestressed concrete through-girder pedestrian bridge system was analyzed through three sets of finite element analyses using the models of an entire bridge system. Each of these three sets of analyses aimed at studying a different aspect of the bridge system behavior.
The pull-out tests performed on steel inserts indicated that the type of inserts currently being used in prestressed concrete bridge girders in the State of Minnesota has the ability to undergo significant amounts of plastic deformation without a reduction in the load capacity. The ductile behavior of the steel inserts used in prestressed concrete girders was also confirmed by the results from the connection subassemblage tests. The connection subassemblage tests also revealed that the behavior of these inserts is significantly affected by the construction method followed during the fabrication of prestressed concrete girders. Results obtained from the girder end detail specimens indicated two types of horizontal load resisting mechanisms depending on the type of detail. During testing of the girder end detail specimens, large values of lateral displacements following the peak load capacities were measured with some level of residual load capacity.
The static lateral load finite element analyses indicated significantly different bridge response depending on whether or not the flexibility of the girder supports were included in the models. It was also determined that the load transfer mechanism among the bridge components depends on whether the girders were loaded at the exterior or interior face. Results from these analyses also showed that the lateral load and deformation capacities of the bridge system could be improved by increasing the ductility and strength of the connection between the girders and floor beams.
The dynamic lateral impact analyses that were performed in an attempt to determine the demand that would occur on the bridge system indicated relatively small impact durations. The dynamic analyses revealed a different deformation pattern of the bridge system than the deformation patterns observed in the static analyses. The damage in the bridge caused by the impacting body was observed to remain highly localized near the impact location for approximately half of the impact duration. As a result, the support flexibility of the girders did not have much effect on the dynamic behavior of the bridge, as opposed to the behavior observed in the static lateral load analyses. The Equivalent Static Force (ESF) values determined from the dynamic analyses were smaller than the static lateral load capacity of the bridge for the cases with flexible girder supports, while for the rigid supports the ESF values were still larger than the static lateral load capacities.
Results of the stability analyses indicated that the local girder damage that would occur in prestressed concrete through-girder pedestrian bridges due to striking of over-height objects may cause the failure of the bridge depending on the extent of damage that the girders would be subjected to. The bridge was determined to be more susceptible to failure when the impact damage occurs near the girder midspan than the girder ends. When only one of the girders was impacted, failure of the bridge would require slightly larger amount of damage in the girder section for failure than the damage level required for failure when both girders are damaged. The amount of “additional capacity” between the cases of single girder versus the both girders being damaged is due to load redistribution from the impacted girder to the other girder. Analysis results showed that the in the case of both girders being impacted, failure of the bridge would occur when approximately 15 percent to 40 percent of the girder web, depending on the location of impact, was damaged in addition to the entire bottom flange.
No comments:
Post a Comment