Lehigh auditions a new deck for a landmark bridge

With two towers reaching 700 feet into the sky and a span that stretches nearly a mile, the Verrazano-Narrows Bridge is a familiar beacon to travelers driving from New Jersey to New York City.
As the starting point for the annual New York City Marathon, the bridge is also a landmark for the world’s serious runners, more than 40,000 of whom filled the bridge for this year’s race on Sunday, Nov. 7.
The Verrazano-Narrows, the largest suspension bridge in the U.S., opened in 1964. Five years later, to accommodate increased traffic, a lower deck was added.
Now, New York City’s Triborogh Bridge and Tunnel Authority (TBTA) has decided to replace the original, upper deck—a grid of steel beams overlaid with concrete—with a steel orthotropic deck.
TBTA’s decision, says Sougata Roy, was necessitated in part by security measures taken after Sept. 11, 2001, when TBTA restricted truck traffic to the upper deck for safety reasons. The resulting heavier traffic loads accelerated damage to the upper deck. TBTA retrofitted the deck with new concrete, but with limited success.
Roy, a senior research scientist in Lehigh’s ATLSS (Advanced Technology for Large Structural Systems) Center, is testing a full-scale prototype of the orthotropic deck that will be used as the Verrazano-Narrows’ replacement deck. The experiments are taking place in ATLSS’ structural testing lab, whose test floor and fixed reaction walls, among the world’s largest, impose multidirectional loads that simulate the demands structures sustain from traffic, wind and earthquakes.
An asymmetry that enables more efficient use of materials
Roy, the project’s principal investigator, is being advised by former ATLSS director John W. Fisher, who supervised stress tests in the 1990s on prototypes of orthotropic decks for New York City’s Williamsburg and Bronx-Whitestone Bridges.
“Orthotropic decks, properly designed, are the only decks that can enable bridges to achieve 100 years of service,” says Roy. “That’s the lifespan suggested by the U.S. Federal Highway Administration. Our goal is to verify whether it can be done.”
An orthotropic deck has different stiffnesses in perpendicular directions. The stiffening ribs of the deck plate run parallel to the bridge’s length. Transverse plates, or diaphragms, run perpendicular to the ribs and are spaced intermittently, providing less stiffness transversely than longitudinally.
This asymmetry, says Roy, enables an orthotropic deck to distribute loads from passing trucks (source of the fatigue damage to steel bridges) to a bridge’s structural supporting elements with more efficient use of material. This allows for a more lightweight deck, which reduces stress to the deck and the suspension cables that support it.
But the asymmetry also increases the complexity of the stresses in the deck’s components and requires sophisticated analysis of the deck and the welds connecting the bridge elements. If welded connections are not designed and fabricated correctly, says Roy, cracking and crack propagation can result.
Two connections pose particular challenges, says Roy. Cutout areas, where ribs pass through the transverse diaphragm plate, rotate across the diaphragm in response to truck loading and develop a complex combination of in-plane and out-of-plane stresses. Secondly, the connection between enclosed ribs and the deck plate must be made from outside. The heat applied for rib-to-deck plate welding requires careful calibration: Too much can cause melting through the thin ribs. Too little can result in inadequate penetration, a poor connection and premature cracking.
Using HPC to model complex stresses

Roy and his group are conducting tests on a prototype deck panel which is about one-quarter the size of a typical span, measuring roughly 25 feet by 25 feet, and which is integrated with floor beams and stringers to simulate actual bridge conditions.
Six hydraulic actuators positioned above the deck impose loading forces in sequence to mimic the passage of truck traffic. Three actuators beneath the bridge deck account for the global deformation imposed by the simulated truck loads.
Roy’s group is also using Lehigh’s high-performance computing (HPC) capabilities to model and analyze the entire test setup and better understand the deck performance.
“HPC enables us to more accurately predict the system performance of a deck,” says Roy. “The physical results allow us to calibrate our HPC models. After we monitor actual stresses on a prototype, we can develop a framework for designing and simulating the behavior of bridge decks with computational modeling and without expensive lab work.
“Our hope is that the finite element analysis and experimental simulations will help us develop specifications for the robust design of new decks that can last a century.”