Document Type



Doctor of Philosophy


Structural Engineering

First Adviser

Roy, Sougata


Single support bar (SSB) modular bridge joint systems (MBJS) are exceedingly being used for accommodating thermal movements in medium and long span bridges exceeding 27 in. (700 mm). These systems, which are typically comprised of steel and polymeric components, exhibit complex dynamic response when subjected to successive impacts by each crossing truck axle. This repeated dynamic loading generates many amplified stress cycles at the welded and bolted connections within the system, rendering them susceptible to accelerated fatigue cracking in service. As noted from observed cracking in field installations, the center beam (CB) to support bar (SB) connection is the most critical detail; however, the behavior and fatigue resistance of this connection in SSB systems was not well understood. Additionally, the dynamic response characteristics determined from limited field measurements on SSB systems indicated a wide range of possible amplifications dependent on the joint size and properties and the speed of the crossing vehicle.Considering the above, comprehensive experimental and analytical research was performed to characterize the dynamic behavior of typical SSB MBJS under wheel loads and determine the fatigue resistance of bolted CB-SB connections in SSB MBJS, which were expected to be more fatigue resistant than their welded counterparts. The study included static and fatigue testing of full-size SSB systems in the laboratory, characterization of suitable material models for the nonlinear hysteretic polymeric components, static analyses of the tested system, dynamic analyses of a previously field-tested system, and parametric three-dimensional Finite Element Analyses of many differently-sized systems subjected to dynamic loading under different service conditions.Through the fatigue testing of 14 full-scale SSB MBJS assemblies, the infinite life fatigue resistance (Constant Amplitude Fatigue Threshold) of the bolted CB-SB connections was determined as 16 ksi (110 MPa), equivalent to fatigue Category B in the seventh edition of the AASHTO LRFD Bridge Design Specifications. These connections were previously limited to fatigue Category D by Appendix A19 in the third edition of the AASHTO LRFD Bridge Construction Specifications. Experimental and analytical studies highlighted the importance of proper joint compaction and bolt pretension on the fatigue performance of the connections, and demonstrated that the past common practice of replacing the polymeric components with steel discs within the systems during the fatigue tests drastically reduced the fatigue resistance of the CB-SB connections.Dynamic time-history analyses of a field-tested MBJS system (I70/I25 Flyover Ramp MBJS tested by Dexter et al. [1997]) were performed using two new load pulse models (derived through convolution and tributary area methods) and three existing load pulse models for representing the wheel loading on the CBs. The results of analyses using these load pulses applied to a simplistic beam element model of the system exhibited good correlation with the field tests; however, comparison with more field observations is required to verify the new load pulse models. Additional dynamic time history analyses were performed with more sophisticated representations of the systems. All analyses correlated well with the field results, demonstrating that the dynamic behavior of SSB MBJS can be adequately simulated using time-history finite element analyses.Using the results of the dynamic analyses as a basis, a factorial parametric study was designed which included as variables: the joint size; the gap opening; the precompression gap for CB-SB connection assembly; and the vehicle speed. The results were normalized by a dynamic interaction parameter (DIP), which was defined as the pulse duration (including the effects of the vehicle speed and the joint characteristics) divided by an equivalent natural period of the MBJS. The equivalent natural period was estimated for each parameter combination independently from the analysis results, using a generalized single-degree-of-freedom idealization of the SSB MBJS. The DAFs obtained from the simulations and the respective system DIPs allowed the generation of a shock spectrum for SSB MBJS that represented the variation of the DAF as a function of the system parameters and the operating conditions. The results provided insight into the dynamic response characteristics of SSB MBJS, and allowed the development of a practical design equation for determination of joint-specific DAFs.Updated testing and data interpretation protocols for SSB MBJS, particularly for load reversal and infinite life, were developed based on the overall results of the study. Design guidelines were developed for regular application, and the AASHTO LRFD Bridge Design and Construction Specifications were revised based on the key findings.