Fatigue strength of tapered partial-length cover plates
Abstract
Cover plates have been widely used to increase the flexural capacity of bridge beam members at high moment locations. A particular detail that has been used in Indiana involves a partial-length tapered cover plate welded to the beam flanges. Due to the severe stress concentration that occurs near the end of welded cover plates, fatigue cracks often develop at the toe of cover plate end welds. The cover plate end detail has one of the lowest fatigue resistance and is classified under Category E of the AASHTO fatigue specifications. The research study involved both experimental and analytical modeling. The primary objectives of the experimental study were to evaluate the number of loading cycles required to develop detectable cracks at the cover plate ends and to evaluate the performance of several methods of repairing the cracked cover plate ends. Three repair methods were investigated: a friction type bolted splice plate connection, air-hammer peening, and a combination of the previous two called the partial bolted splice connection. The main objective of the analytical study was to estimate the fatigue strength of cracked cover plate ends repaired with one of the repair methods studied. A total of 33 specimens at roughly 1/3 scale were tested under constant amplitude load-controlled cycling to fulfill the previously mentioned objectives. Detectable cracks were found to develop after a number of loading cycles equivalent to Category E design life. The bolted splice plate connection was the most effective repair method. Specimens repaired with the bolted splice connection achieved Category B design life after repair. If peening is applied to non-cracked cover plate ends, an additional number of loading cycles equivalent to Category $\rm B\sp\prime$ design life could be applied after repair. The fatigue life of repaired cover plate ends was successfully predicted using a crack propagation program. The propagation program was more successful in estimating the fatigue life of details repaired with the bolted splice and the partial bolted splice connections than for peened details. In general, the estimated lives were in between $2\over3$ and 1.5 times the experimental lives for the bolted and partial bolted splice plate details.
Degree
Ph.D.
Advisors
Bowman, Purdue University.
Subject Area
Civil engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.