This Is AuburnElectronic Theses and Dissertations

Laboratory Testing of Transverse Joints in Precast-Panel Bridge-Deck Replacement Systems




Matthews, Jamieson

Type of Degree

Master's Thesis


Civil Engineering

Restriction Status


Restriction Type

Auburn University Users

Date Available



Out of the 614,387 bridges in the United States, 56,007 are structurally deficient (FHWA 2016). In many instances this structural deficiency relates to the deck of the bridge. Deck degradation is very common in bridges over 25 years old (Biswas 1986). Many decks fail several years before the rest of the bridge superstructure and substructure reach the end of their useful life (Bettigole and Robison 1997). An effective solution to this problem is to rapidly replace the bridge deck with precast panels. An important aspect of any deck system is the behavior of its transverse joints. The long-term success of a deck replacement project is dependent on the durability of the deck joints. The Alabama Department of Transportation does not have any standards to test or evaluate the performance of transverse joints in precast-panel bridge deck systems. This thesis describes the development of test methods and performance criteria to compare behavior of various precast-panel transverse joints and determine their acceptability. The test methods and performance criteria were implemented on a proprietary deck system in the laboratory. The two joint types tested for the Exodermic® deck system were the unreinforced shear key joint and the staggered hook reinforced joint. The transverse joints were tested in quasi-static and fatigue loading for positive bending, negative bending, and a newly developed test of shear reversal under constant positive bending. Data were collected and analyzed to determine the in-service and long-term performance of the joints. The results were based on capacity, midspan deflection, and crack opening of the specimens. A practical test method for future implementation was produced successfully. However, correlation between the quasi-static and fatigue test results could not be determined due to the small range of specimens tested.