The initial performance of typical modular bridge expansion joints (MBEJS) has been good and most MBEJs have not had durability problems. However, in some cases, MBEJs have developed fatigue cracks and other durability problems. Causes of the problems include the following: a) the dynamic response of these systems was not fully understood; b) poor details were specified and fabricated; and c) some installations were poorly constructed. In addition, in today's construction process, MBEJs are typically procured by a low-bid process. This situation, combined with the fact that there are currently no nationally accepted design specifications for MBEJS, contributes to the problems. To address these problems, experimental and analytical research was performed and performance-based specifications and commentary for the fatigue design of MBEJs were developed. To determine the static and dynamic response of different MBEJ systems subjected to truck loading and to verify proposed structural analysis models, field tests were conducted at four sites. The key findings of the field tests were as follows: a) a MBEJ can be modelled as a two-dimensional frame by using beam elements; b) maximum measured amplified dynamic vertical moment range was 1.63 times the corresponding static moment; c) horizontal bending moment range measured under normal driving conditions was up to 23% of the amplified vertical bending moment; d) horizontal moment ranges measured during emergency braking tests were almost 50% of the amplified vertical moments; and e) the measured stress ranges from typical traffic were slightly less than the stress ranges predicted on the basis of the proposed upper-bound limit-state design loads. The static load and strain distributions were also measured in the laboratory and were found to be similar to distributions measured in the field. A practical test procedure was developed to determine the fatigue resistance of critical details. Fatigue tests of centerbeam- to-support-bar connections indicate that the fatigue resistance of full-penetration welded connections is equivalent to AASHTO category C, and the fatigue resistance of bolted connections is equivalent to AASHTO category D. Partial-penetration welded centerbeam-tosupport-bar connections have very low fatigue resistance and therefore are not feasible under the proposed design specifications. The proposed fatigue design specifications are designed to be integrated with the current load and resistance factor design specifications for bridge design. To be consistent with these bridge specifications and allow for regional differences in the bridge fatigue design loading or changes in the loading over time, the proposed MBEJ fatigue design specifications use the same fatigue design truck as the current AASHTO bridge specifications, the HS- 15 loading (after taking into account the load factor of 0.75 for fatigue). One important modification of the AASHTO truck loading for MBEJs is explicit recognition that the rear axles of the HS truck loading in the specifications actually represent tandem axles. For the purposes of comparison, a single axle of the HS- 15 tandem rear axles is equivalent to a "fatigue limit state" static axle load (i.e., an axle load that is supposed to be only rarely exceeded) of 107 kN (24 kips). The proposed dynamic amplification factor is conservatively set as 1.75. The recommended horizontal load range, acting simultaneously with the vertical load range, is equal to either 20% or 50% of the amplified vertical load for installations either on open highways or near facilities where continuous heavy braking is expected, respectively. A simplified method is proposed to estimate the distribution factor (i.e., the fraction of the design wheel load range assigned to a single centerbeam). (A)
Samenvatting