Bridge collapse shows post-WWII engineering aproach to be wrong

Published 6 August 2007

Engineers on the Brooklyn Bridge (opened in 1883) did not have sophisticated methods to calculate loads, so they made their best guess — and then multiplied; in the aftermath of WWII engineers believed they could save on materials by accurately predicting traffic pattenrs and loads; Minnesota bridge collapse shows that approach to be wrong

Remember Senator Ted Stevens’s (R-Alaska) “Bridge to Nowhere”? The bridge would cost $315 million dollars, of which the federal government will pay $223 million. The bridge — officially called the Gravina Island Bridge — will replace the ferry that currently connects the city and borough of Ketchikan, Alaska (population 13,125) to developable land on Gravina Island (population 50) and improve access to Ketchikan International Airport. The bridge would replace or augment a seven-minute ferry ride from Ketchikan to its airport (by “augment” we mean that it would cut the seven-minute ferry ride to a four-minutes car ride). Opponnents of this ultimate pork-barrel project note that the $223 million federal contribution to the bridge would cost approximately $16,990 per person in the Ketchikan Gateway Borough if the costs were split amongst them. Opponents, among them Senator John McCain (R-Arizona) have also noted that the federal contribution represents $4.5 million for each of Gravina Island’s residents.

Now, there are 594,709 bridges in the United States (and if the Gravina Island Bridge is built, there will be 594,710). Many of these briges — the Minneapolist bridge which collapsed last week was one example — were built during an era when designers were confident they knew enough about bridge strength and weight loads that they could build bridges lighter and cheaper. Engineers now know more about the frailty of bridges and the punishment they take from heavy trucks, strong currents, and the occasional errant barge which slams into a support column. Unfortunatley, this knoweldge come a bit late in the game: Many of the U.S. 594,709 bridges were built during the 1950s and 1960s, a time when designers did not fully understand the effects of metal fatigue or other challenges. The Washington Post’s Eric Weiss quotes Mal Kerley, Virginia’s chief engineer, to say that “Maybe we out-thought ourselves for a little while,” referring to postwar bridge-building when the interstate highway system was created. “What has happened over time is that we learned things.”

Here is an example of not out-thinking oneself: In 1883 the Brooklyn Bridge opened to horse-and-buggy traffic. Today it carries more than 144,000 vehicles a day. Engineers on the Brooklyn Bridge and other spans did not have sophisticated methods accurately to calculate loads, so they made their best guess — and then multiplied. The result was over-engineered, overbuilt, over-robust structures which will likely outlast their younger brethren. “The smaller the amount of knowledge you have, the bigger the factor of safety you use,” said Joseph Yura, emeritus professor of civil engineering at the University of Texas. “In those days, you would do the calculations, and the chief engineer would just up everything by one size.” Today, again, bridge designers are being more cuatious and conservative: Spans are built with stronger materials, redundant support systems and super-strength steel that can withstand the pounding of thousands of vehicles a day.

In between then and now, however, there was a period during which different approaches prevailed. After the Second World War, for example, with steel prices high, engineers thought they could “minimize materials and maximize stresses,” Kerley said. The result was structures such as the Frederick Douglass Bridge over the Anacostia River in Washington, D.C. Built in 1950, the Douglass Bridge sits on only two main horizontal steel beams bolstered by vertical supports. The main beams were so thin that engineers at the time attached numerous vertical “stiffeners” to keep the beams from bending. Yes, the stiffeners required a lot of manual welding, but “steel was more expensive and labor was less expensive” at the time, said Kathleen Penney, the District Department of Transportation’s deputy chief engineer and director of a $27 million refurbishing of the bridge. She told Weiss that if the Douglass Bridge were built today, the steel used would be 25 percent thicker, there would be more than two main beams in case one failed, and there would be far fewer welds and connections to minimize the number of corrosion and fatigue points.

Collectively, there was more confidence in using less material,” Penney said. “Since that time, we’ve had a lot of experiences that pushed us back to more conservative approaches.”