Grant 2: Hub-girder bolt assembly without an interference fit in bascule bridges

Trunnion-Hub-Girder (THG) assemblies of bascule bridges in Florida were found to fail during assemblies of the Christa McAullife and Brickell Avenue Bridges.   Furthermore, minute surface cracks and shrink defects were observed in the hubs after the trunnion-hub assemblies were installed in the girders on the Miami Avenue Bridge.  Since such an assembly can cost more than $100,000, and more expense and time in delays and reassemblies, it is a problem that needs to be addressed further.

  To address this issue, the FDOT instituted a combined analytical and experimental study in 1998.  The analytical part (Phase I) of this study used a finite element program that determines transient temperatures and stresses, critical stresses and critical crack lengths in the THG assembly.   These parameters were mainly studied for first finding whether the critical stresses exceeded the yield strengths or allowable stresses of the materials used for the THG assembly.   Also, since there were two different assembly procedures that could be used to develop the final THG assembly, the critical stresses and critical crack lengths were compared to find the optimum assembly procedure.  This finite element program has an interactive graphical user interface, in which a range of parameters can be defined and altered, and hence, can be used for thermal stress and fracture resistance analysis on any future THG assembly built in Florida or the rest of the country.

The experimental part (Phase II) of the previously study was done to confirm the finite element results and also develop specifications for the assembly procedure, which previously had been not available.  This part of the study involved instrumenting a full-scale trunnion, hub, and girder, and recording temperature and stress at various locations throughout both assembly procedures.  This phase was completed during Summer 2001 and clearly validates all of the results from Phase I.

The results from Phases I and II of the aforementioned research to study the trunnion-hub-girder assemblies of bascule bridges indicate that a serious problem can arise during one of the assembly procedures of the THG.  The problem, a high probability of cracks developing in the hub (and possible catastrophic failure), exists when a trunnion-hub assembly is cooled in liquid nitrogen for shrink-fitting into the girder during one of the assembly procedures; namely, when the trunnion is shrink fitted into the hub and then the trunnion-hub assembly is shrink fitted into the girder.

REQUESTED SERVICES:  To do away with cooling a trunnion-hub assembly, elimination of shrink-fit at the hub-girder is proposed.  By eliminating the interference fit at the hub-girder, new problems arise because the bolts now transmit most of the loads from the bridge through the girder, then to the hub (and backing ring), then to the trunnion and finally, to the bearings.  For the hub and bolts to now be able to transmit the load, numerous parameters need to be studied and defined or specified.  Some of these parameters include: girder thickness, type of bolts (that is, grade and material), bolt sizes, number of bolts, number of bolt circles, bolt material, hub flange thickness, and hub flange diameter, to name a few.