A smorgasbord of factors, including welding problems and construction misfits, contributed to the cracking of “stringers” on the Blanchette Bridge in St. Louis in July 2005, according to a recent report issued by University of Missouri-Rolla researchers.

Blanchette Bridge carries Interstate 70 across the Missouri River, connecting St. Louis and St. Charles counties in eastern Missouri. Constructed in 1958, the bridge approaches are supported with three large plate girders alternated by smaller stringers. Bridge inspectors with the Missouri Department of Transportation discovered a vertically cracked stringer – a continuous beam that supports the bridge deck – during a routine inspection in July 2005 and asked UMR researchers to help pinpoint its cause.

Dr. W.N. “Nick” Marianos, a research professor in UMR’s Center for Infrastructure Engineering Studies, led the team’s year-long investigation, which included site visits and field tests. The team concluded that a combination of factors converged to lead to the cracking. Although the elements had several sources, most occurred in the design phase.

The bridge’s troubles began in 1979, when the westbound bridge’s original reinforced concrete roadway deck was replaced with a steel grid deck system. This rigid system was placed on top of the original larger plate girders and smaller stringers, creating a unequal weight distribution.

“If a structure is given the choice between giving loads to stiff supports or less rigid supports, the load will tend to go to the larger supports,” Marianos says. “Loads were shifted as the big plate girders essentially sucked the weight away from the stringers, causing an unequal weight distribution. This variance was one factor in the beam’s cracking.”

Marianos’s team found that while the lighter steel decking helped improve the bridge’s overall capacity by reducing the amount of strength the bridge had to use to carry itself, the 20-year-old stringers and plate girders rebounded inconsistently when the heavy concrete deck was removed.

To compensate for the differences, construction workers first added shim plates to the top of some of the stringers in an effort to get many of them up to the right elevation. Yet even with the addition of shim plates, workers occasionally couldn’t get the grid to sit on all four supports.

“Discussions with one of the contractors’ project engineers revealed that a number of these misfits occurred during the construction, and that the typical solution to the problem was parking a piece of construction equipment on the panel, forcing it to seat itself on all of the shim plates for welding,” Marianos says. “Apparently, this ‘fix’ was not a common occurrence, but was not rare, either.”

The UMR team also found that welding problems contributed to the crack. Shim plates were added in a piecemeal fashion to run the length of the bridge. Although designers were careful not to place open joints in the shim plates in places that would undergo negative bending, they didn’t instruct the contractor and welders how to handle welds at the shim plate open butt joints.

“The workers just welded across the joints, something that by modern practices isn’t done,” Marianos says. “In several of these locations, the grid was being forced to touch the stringers by parking a truck on the grid. The stringers tried to bounce upward once the weight was removed, which stressed the weld.”

Temperature cycles during the day also loaded the weld.

“The top of the bridge deck is more exposed to the sun and heats up faster in the mornings than the bottom does, which drags the stringers upward,” Marianos explains. “All of these factors lead to the fact that for a significant amount of time, some of these stringers are bending backwards from what the designers thought they would be.

Traffic loading was also a contributing factor and caused structural fatigue at the shim joints. “Since fatigue damage is constantly accumulating, more cracks can be expected in the future,” Marianos says.

The UMR team presented MoDOT with four alternate strategies for addressing the bridge’s fatigue damage and vulnerability. The options ranged from continuing normal inspections to performing extensive retrofits aimed at strengthening the stringers and protecting against the consequences of stringer cracking.

“Limited retrofits, aimed at removing crack initiation locations, and more frequent inspections is the approach our team believes best balances structural safety and economy,” Marianos says. “The nice thing about this grid deck is that it’s beefy. If a stringer does crack, it can carry between inspections. MoDOT has some flexibility because although this is an item they need to be aware of and understand, it doesn’t require immediate action.”

Kenneth Foster, MoDOT supervising bridge inspection engineer, says inspectors will continue to monitor the bridge for future cracking but any retrofits would likely be held off and combined with any future rehabilitation the bridge may undergo.

“Luckily, this is a unique situation for us,” Foster says. “ We won’t have this problem carrying on to other bridges.”