Galloping Gertie: How a Steady Wind Twisted the Tacoma Narrows Bridge Apart

On a gray Thursday morning in November 1940, a brand-new bridge in Washington State began to twist like a wrung-out towel.  One side of the roadway rose while the other fell, again and again, until the deck tilted as much as 45 degrees.  For about an hour, people on shore watched and filmed it.  Then the middle of the bridge tore loose and dropped into Puget Sound.  The Tacoma Narrows Bridge had been open for just four months, and locals already called it Galloping Gertie.  Here is how a record-setting bridge shook itself apart in a steady wind, why the usual textbook explanation is not quite right, and how its failure changed every long bridge built since.

A ribbon across the Narrows

The Tacoma Narrows is the tight spot where Puget Sound squeezes between the city of Tacoma and the Kitsap Peninsula.  For decades, people on the peninsula depended on ferries, and local boosters had pushed for a bridge since the 1920s.

Washington’s state bridge engineer, Clark Eldridge, drew up a conventional suspension bridge.  Its roadway would be stiffened by steel trusses about 25 feet deep, a kind of open lattice hung beneath the road.  It would cost about $11 million.  The federal Public Works Administration, which was paying much of the bill, wanted something cheaper.  It made its money conditional on the state hiring outside consultants, and the bridge was redesigned by Leon Moisseiff, a respected New York engineer who had worked as a consultant on the Golden Gate Bridge.

Moisseiff was a champion of a newer idea called deflection theory.  In plain terms, it says that on a very long suspension bridge, the heavy main cables do most of the work of holding the deck steady, so the deck itself can be lighter and more flexible.  He replaced the deep trusses with solid steel plate girders only 8 feet deep.  The result was slimmer and far cheaper.  The bridge cost $6.4 million and took 19 months to build.  When it opened on July 1, 1940, its 2,800-foot main span made it the third-longest suspension bridge in the world, behind only the Golden Gate and George Washington bridges.  It was also just two lanes and 39 feet wide, which made it unusually narrow and shallow for its length.

A bridge that made people seasick

The bridge started moving before it was even finished.  In May 1940, as the steel deck went up, riveters noticed it bouncing in the wind, and according to the Washington State Department of Transportation, some chewed on lemons to fight off nausea.  That bounce gave the bridge its nickname.

After opening, it only got more famous.  The roadway would sometimes rise and fall in a wind as light as 4 miles per hour, with waves rolling along the deck that could lift and drop cars by several feet.  Drivers watched the cars ahead of them disappear into a dip and pop back up.

Engineers tried tie-down cables anchored to concrete blocks on shore, angled cable stays near mid-span, and hydraulic buffers near the towers.  None of it stopped the motion.  The state also hired Frederick Burt Farquharson, an engineering professor at the University of Washington, to study the bridge with scale models in a wind tunnel.  By early November, his tests had turned up a dangerous twisting motion, and he concluded that the solid girders were catching the wind.  He suggested adding streamlined fairings or deflectors along the sides of the deck, and state officials began drafting a contract to install them.  They ran out of time.

The morning it twisted

On November 7, 1940, the wind over the Narrows was blowing about 38 miles per hour by 7:30 a.m. and about 42 miles per hour by 9:30.  The deck was galloping in its usual way, with waves a few feet high.  Then, at about 10:03 a.m., the motion changed.  The bridge began to twist.  One edge of the roadway rose while the other dropped, and the center line stayed nearly still, as if the deck were rolling around it.  Within minutes, according to WSDOT’s timeline, the roadway was tilting as much as 28 feet one way and then the other.

A few people were caught on the bridge.  Leonard Coatsworth, a newsman at the Tacoma News Tribune, had driven out with his daughter’s cocker spaniel, Tubby, in the back seat.  The car lurched out of control, and he jumped out.  He later described what happened next: “Around me I could hear concrete cracking.  I started back to the car to get the dog, but was thrown before I could reach it.  The car itself began to slide from side to side on the roadway.  I decided the bridge was breaking up and my only hope was to get back to shore.”

Others tried to save the dog.  Howard Clifford, a News Tribune photographer, went out but had to turn back.  Professor Farquharson, who had come out to film the bridge’s motion, did reach the car and opened the door.  Tubby, terrified, bit his finger, and Farquharson gave up and staggered back to safety.  According to the family, Tubby had only three legs.

At about 10:30, a chunk of concrete fell from the center span.  At 11:02 a.m., a section of roadway about 600 feet long broke free and plunged into the Narrows.  By 11:08, the rest of the center span had followed, and Coatsworth’s car went with it.  No person died that day.  Tubby was the only victim.  The state later paid Coatsworth $450 for his car and $364.40 for its contents.

It wasn’t quite resonance

For decades, many physics textbooks used the Tacoma Narrows Bridge as the classic example of resonance, the effect you get when you push a child on a swing at just the right moment each time.  The idea was that the wind pushed the bridge in a steady rhythm that happened to match its natural swaying rhythm, so each push added a little more motion.  It is a tidy story, and it is mostly wrong.

In a 1991 paper in the American Journal of Physics, engineers K. Yusuf Billah and Robert Scanlan pointed out that the wind that day was fairly steady, not pulsing.  Measurements reported by Farquharson put the deadly twisting at about 12 cycles a minute, or one full twist every five seconds.  That did not match the rhythm of the swirls of air that peel off a blunt object in the wind, which at that speed would have come about once a second.

What happened instead is called aeroelastic flutter.  Once the deck began to twist, the twisting itself changed how the wind flowed over and under it, so the wind pushed harder in the direction of the twist.  Normally, a vibrating structure loses a bit of energy with each swing and settles down.  Here, each twist made the next one bigger, and no perfectly timed push was needed.  Above a certain wind speed, a steady breeze was enough.  The solid, flat-sided girders made it worse, because they blocked the wind like a wall instead of letting it pass through the way an open truss does.

The federal board that investigated the collapse, which included the famed bridge engineer Othmar Ammann and the aerodynamics expert Theodore von Kármán, reported in March 1941.  Its main findings were that the bridge was too flexible, that the solid girders and deck acted like an airplane wing, creating drag and lift, and that engineers did not understand wind forces well enough and should test long suspension bridges with models in wind tunnels.

The film everyone has seen

Part of the reason Galloping Gertie is so famous is that the collapse was caught on film.  Barney Elliott and Harbine Monroe, who owned The Camera Shop in Tacoma, filmed it with two cameras, including the failed attempt to rescue Tubby.  Their footage went out in newsreels around the world, and in 1998 the Library of Congress added it to the National Film Registry.  Farquharson filmed it too, and frames from his 16mm film, like the one at the top of this page, became part of the engineering record.

One odd detail: many copies of the famous footage run faster than real life.  One camera was running at 16 frames per second, but copies were often made as if it had run at 24, so the bridge seems to twist about 50 percent faster than it really did.

What engineers changed

The collapse sent bridge designers back to the drawing board.  Wind tunnel testing became a normal step for long suspension bridges.  Deep, open trusses came back, and later designers developed streamlined box-shaped decks that slip through the wind more like a wing.  Other bridges were retrofitted.  New York’s Bronx-Whitestone Bridge, which had a similar slim plate-girder deck, got 14-foot steel trusses added along both sides in 1943.

The bridge’s engineers took it hard.  Eldridge blamed the federal money men, telling reporters, “The men who held the purse-strings were the whip-crackers on the entire project.”  Moisseiff died in 1943, his reputation badly damaged.  Like the Boston molasses flood two decades earlier, the disaster became a lesson that engineering students still study.

World War II and wrangling over money delayed a replacement for almost a decade.  When it opened on October 14, 1950, the new bridge reused the old piers but had four lanes and a deep, open truss under the roadway.  Farquharson tested its design in his wind tunnel, and local promoters called it Sturdy Gertie.  In 2007, a second suspension bridge opened right beside it to carry eastbound traffic.

What’s left of Gertie

Salvage crews spent more than two years taking the wreck apart.  Steel was precious during the war, and the cables and remaining steel were sold as scrap.  The job still lost money.  The state spent about $646,661 on salvage and got back about $295,726 for 7,000 tons of scrap steel.

The fallen center span still lies on the floor of the Narrows, where it has become an artificial reef for fish and other sea life.  In 1992 the wreck was listed on the National Register of Historic Places.  A chunk of the old roadway is in the collection of the Washington State History Museum in Tacoma.

The bottom line

Galloping Gertie was not brought down by a hurricane or an earthquake.  It fell in a steady wind of about 40 miles per hour, because its slim, solid deck could be twisted by the very air flowing past it, and each twist made the next one worse.  Nobody at the time fully understood how wind and a flexible structure could feed each other.  The lesson, recorded on film and in every bridge built since, is that for a long, light structure, the wind is not just a load to carry.  It can become part of the machine.

Build a bridge, or read the whole story