Dixon, Illinois – 1873
Sunday, May 3, 1873 was a special day for many residents of Dixon, Illinois and surrounding areas. It was a Sunday morning filled with spring sunshine. People dressed in their finest. There was an air of excitement for some. This was Baptism Sunday for members of the Dixon Baptist Church, the first such special Sunday of the year. A few hundred were in attendance. Several members, including the daughter of the pastor, Rev. J. H. Pratt, were to experience the rite of full immersion near the banks of the Rock River. Those in attendance could watch the ritual from the new bridge. This was the procedure Rev. Pratt had used on previous Baptism Sundays with even larger groups attending.
The bridge extended across the Rock River. The town of Dixon, population of about 5,000, had the bridge designed and constructed by L. E. Truesdell of Belvidere, IL at a cost of $75,000.
The main structure was made from cast iron except for the floor and floor beams. There were five spans, each 132 feet long, a total span of 660 feet. The spans rested on four piers made from heavy masonry. There was an 18-foot wide roadway with a 5-foot wide walkway on either side. The walkways had 3-foot high railings along the edges of the bridge.
In the late 1800s, Dixon was a rapidly growing town and an important crossing point for travelers moving north and south in Western Illinois. Until 1846, ferry service provided river crossing. Between 1846 and 1868, Dixon had at least eight wooden bridges that allowed pedestrian and animal-draw wagons to cross the Rock River. All the bridges failed, mostly from high water and floating ice jams that occurred each spring.
The city sought a better solution and a committee investigated several bridge designs. In a close vote, the aldermen approved a bid by L. E. Truesdell who had a 1856 patent for a lattice bridge design. The concept allowed manufacture of short, light pieces to be transported to poorly accessible bridge sites and assembled there. He had built other bridges in Illinois, but this would be the longest span yet.
Dixon celebrated the bridge completion with a dedication on January 21, 1869. A canon shot announced the bridge opening. There was a one-half mile long parade. There was a public load test that included about 45 tons of horse-drawn, loaded wagons and many citizens.
Four years later was the first Baptism Sunday of 1873. Not long after the ceremony began, with the choir singing and two of the members already baptized, some said they felt the bridge vibrate. The bridge tender ordered boys off their advantageous view from the trusses. He also ordered the other people off the bridge. Soon, a span collapsed into the river that was 15 to 20 feet below. The falling structure carried many people with it. Some fell directly into the water. Parts of the structure fell onto other victims, trapping them. A few horses that pulled carriages or wagons also fell into the river.
Rescue began quickly. A few individuals were able to navigate the flowing water and swim to shore. However, some victims trapped by structural elements were not recovered quickly. One horse rescue required two days and use of a raft anchored by rope to the shore. There were several heros and many stories from the disaster. Overall, records reported at least 45 fatalities and 56 injuries.
There is a long history of bridges collapsing and the loss of life. Safe design or lack of it is often a factor. Construction, assembly, maintenance or lack of maintenance and other factors can also affect individual cases. What caused the failure at Dixon is not fully known. The use of cast iron may have been a contributing factor, since casting flaws may not be detected during visual inspection. In addition, cast iron is brittle and subject to loading failure and flexural fatigue sooner than more ductile steel that replaced cast iron in bridge construction.
Many bridge failures identify a variety of causes. For example, in the I-35W bridge collapse in Minneapolis in 2007 some engineers pointed to an older design method for that bridge that did not incorporate “redundancy,” which reduces the potential for single-point failures.
The walkway collapse at Florida International University in March 2018 during final assembly raised design and construction issues for a relatively new method of bridge design that involves construction and assembly close to the final site.
Tacoma Narrows, WA – 1940 – “Galloping Gertie”
One of the most famous bridge failures was that of the Tacoma Narrows Bridge on November 7, 1940. It was a spectacular failure that was captured in numerous photos and movies found on web sites. The only fatality was that of a dog.
At the time, the bridge was the third longest suspension bridge compared to the Golden Gate Bridge in San Francisco and the George Washington Bridge in New York City. The total length was 5,939 feet and the longest span was 2,800 feet. It was 195 feet above the water of Puget Sound below. It connected Tacoma with Kitsap Peninsula in Washington. Its intent was to replace ferry service across Puget Sound.
The initial design by engineer Clark Eldridge included typical 25-foot-high trusses under the road to stiffen the roadway. In order to save money, engineer Leon Moisseiff offered an alternate design that used 8-foot-high girders and reduced the $11 million original cost to $8 million. The design had considered winds up to 120 miles per hour and estimated sideways deflection of no more than 20 feet.
Construction began in September 1938. The bridge opened to use on July 1, 1940.
Even during construction, there was concern because the deck would move up and down during moderate winds. Winds blowing through Puget Sound were not uncommon. Bridges were expected to move when there was a wind. But construction workers compared the up-and-down motion to that of a horse and gave the bridge the name “Galloping Gertie.”
After the bridge was open for traffic, some thrill seekers would drove over the bridge to experience the bounce or ripple, even in winds as low as 3 or 4 miles per hour. The waves were often 2 or 3 feet high and occasionally as much as 5 feet high. The bouncing sometimes lasted only a few moments and at other times for several hours.
Wind tunnel models by Professor Frederick Farquharson of the University of Washington determined that at times the models also showed a twisting motion. He concluded that such a twisting motion could lead to the failure of the bridge.
As the stormy season for Puget Sound began in October 1940, engineers created a “fix” to minimize the up-and-down motion. They connected cables from the deck structure to the supporting piers. However, it was not long before the cables that limited the up-and-down motion snapped and the restraints became useless.
On November 1, 1940 Professor Farquharson was at the bridge after it had been closed earlier that morning. He observed not only the typical bouncing action of the deck, but the twisting action he had observed in his wind tunnel tests. He made movies of the action and took photographs that captured the ongoing event that after an hour led to collapse of the middle span.
The fallen span remains at the bottom of Puget Sound today. The location of the sunken remains was placed on the National Register of Historic Places in order to prevent salvaging operations.
The early theoretical explanation that resulted was resonance. Resonance occurs when an object vibrates at it “natural” frequency. An external force will cause the motion of the object to increase or deflect more and more. The argument was that the high winds caused the ever increasing deflection of the bridge deck. More recent analysis disagrees because the amplitude of the up-and-down motion was not a function of the wind velocity as learned from detailed studies of the film capturing the bridge motion. Further analysis pointed to flutter as the cause. Flutter was known about for airplane design. The new subfield for this phenomenon in bridge design gained the name “bridge aerodynamics-aeroelastics.”
Regardless of the theory for the failure, the collapse of Galloping Gertie caused aerodynamics to become an important part of architectural and structural design.
After WWII a replacement bridge finally received funding. Following two years of construction the new bridge opened on October 14, 1950. It was the fifth longest suspension bridge and incorporated design changes resulting from lessons learned from “Galloping Gertie.”
2007: I-35W Bridge Collapse in Minneapolis, Minnesota
A more recent bridge collapse gaining national attention and addressed through many lawsuits occurred in Minneapolis on August 1, 2007. The 8-lane bridge spanned the Mississippi River. At the time, two lanes in each direction were closed because of construction. About 1,000 feet of the deck truss collapsed and 456 feet of the main span fell 108 feet into the 15-foot-deep river. The collapse involved 111 vehicles, with 17 being recovered from the water. 13 people died and 145 were injured.
The bridge was designed more than 40 years earlier and opened to traffic in 1967. The National Transportation Safety Board (NTSB) conducted a comprehensive investigation of the failure that included early design records. The report identified the likely cause for the bridge failure and NTSB offered many recommendations that were considered by the Minnesota Department of Transportation, which had authority for the bridge.
The NTSB report stated that the probable cause of the collapse was the inadequate load capacity of the gusset plates at key locations in the structure due to a design error by the principle design firm. Contributing to the gusset plate failures was an increasing load from bridge modifications over time, increased traffic and concentrated loads of construction materials delivered to the bridge on the day of the failure. In addition, NTSB cited as contributing factors the failure by the design firm to implement adequate quality control procedures to ensure that gusset plate calculations were performed and the inadequate design review by Federal and State transportation officials. Also contributing to the failure was inadequate inspections for the conditions involved in gusset plate failure. The NTSB report provided many recommendations to ensure bridge safety.
In addition, the concept of redundancy in bridge design was introduced some time after the I-35W bridge design occurred. Redundancy in bridge design prevents a collapse if there is a fracture or failure suffered by a bridge member. In engineering, redundancy is the duplication of critical components or functions of a system with the intention of increasing reliability of the system, usually through the form of a backup or fail-safe arrangement. Redundancy is important in safety-critical systems. For many years, aircraft design has incorporated redundancy in its numerous systems, often with three levels of redundancy.
The bridge was rebuilt with improved design features and redundancy and returned to service on I-35W.
- Natalia Belting, “1873 Dixon Bridge Disaster,” Champaign-Urbana News Gazette, December 6, 1981.
Galloping Gertie (Tacoma Narrows, WA)
Movies of the Failure
I-35W, Minneapolis, MN
- National Academies of Sciences, Engineering, and Medicine, 2014, Bridge System Safety and Redundancy, Washington, DC: The National Academies Press. https://doi.org/10.17226/22365)
- Redundancy analysis of the I-35 Bridge Collapse: https://www.thinkreliability.com/case_studies/root-cause-analysis-of-the-i-35-bridge-collapse/