Welds Behind the Plywood

The Corner Cantilever

Mara: Midtown Manhattan, nineteen seventy-seven. William LeMessurier's newest tower is finished: Citicorp Center, fifty-nine stories. And it stands in a strange way. St. Peter's Lutheran Church held the corner of the site, and the church was staying. "There couldn't be a column in that corner," LeMessurier said in an interview. "My solution was to move the columns to each side." So the tower rides on four stilts, one at the middle of each face. Each stands nine stories tall. That left the church its corner, with the tower's own corners reaching out over it, some seventy feet.

Eli: A listener wants to know whether it's true that a student's question revealed a New York skyscraper could blow over in a storm.

Mara: That tower is the skyscraper in question. The story runs from a senior thesis at Princeton to a summer of secrets in 1978. And one piece of it is still argued over: whose question actually set things moving. Today's angle is the engineer who blew the whistle on himself.

Eli: Stilts in the middle of each face. So the corners have nothing underneath.

Mara: Nothing at all. So LeMessurier had to steer every load inward. He designed chevron-style diagonal bracing on every face, carrying the wind and the weight in to those columns.

Eli: So the whole tower leans on four columns and one bracing system.

Mara: Which is exactly what Diane Hartley started poking at. In 1978 she was a Princeton undergraduate, writing her senior thesis on this tower. She worked the wind loads herself, including quartering winds: wind coming in on the diagonal, pushing on two faces at once instead of one. By her math, that case came out heavier. As she put it, "When it came time to double check the quartering wind analysis, which I had determined would yield a 42% higher loading. I didn't see those calculations."

Eli: A student catching what the designers missed. That's a lot for a thesis.

Mara: And she didn't push it. In June 1978, with the thesis already overdue, she called LeMessurier's office. She reached a junior engineer, not LeMessurier. By her account, he "convinced" her that her "calculation was not correct, and the building was inherently stronger." She was a student, so she deferred. "I footnoted that conversation and turned the thesis in."

Eli: So her doubt stopped one desk short of the man who designed it.

Mara: Maybe. LeMessurier remembered a student's call that June too: "June 1978, I'm in my office. I got a call from a student at some– I do not know the school." Other accounts name a different caller, Lee DeCarolis, a student at a New Jersey engineering school, relaying his professor's doubts about those mid-face columns. How much of Hartley's own math ever reached LeMessurier is still uncertain. Two students, two calls, one tower on stilts. Eli, the numbers are yours.

The Sixteen-Year Storm

Eli: And at that point LeMessurier sounded sure of himself. He explained that the church's corner put the supports at the middle of each wall, and he said his design made the tower particularly resistant to quartering winds. Accounts differ on whether one student called or two. Either way, after those questions he sat down and ran the diagonal case again, on paper.

Mara: Which loads the chevrons: steel upside-down V's, eight stories tall, funneling force to the stilts.

Eli: And on the diagonal, the tension in some of those braces didn't drop. It went up forty percent.

Mara: Forty percent is a lot. But buildings carry spare margin for exactly that.

Eli: That margin was thinner than anyone thought. According to historical accounts, LeMessurier learned of the bolted joint substitution only after construction had concluded. The joints on those braces were supposed to be welded. The steel contractor used bolts instead, on the view that the building didn't need the extra strength of welds, and bolts saved real money. With bolts, the diagonal wind hit harder still. The load on the bolts themselves jumped a hundred and sixty percent.

Mara: And his own team had sized those bolts only for wind hitting straight on.

Eli: So in late July he flew to London, Ontario, to the wind-tunnel experts in Canada. He asked them to weigh his new numbers against their original data, and told them to tell him the truth, not to go easy if it came out wrong. Their answer: the stress from quartering winds might be even worse than his latest calculations.

Mara: The man who moved columns for a church, now testing his own design.

Eli: On July 28th, 1978, at his summer place in Maine, he found the weakest joint, on the thirtieth floor. If that one gave way, the whole structure would follow. Bolts like these could fail in quartering winds of about seventy miles an hour. Then he ran New York weather records. A storm that strong could come as often as once every sixteen years. That's without the tuned mass damper running: a four-hundred-ton concrete block up top, a counterweight against the sway.

Mara: Michael Greenburg, author of The Great Miscalculation, says it would topple sideways and could collapse into other buildings.

Eli: Possibly setting off a domino effect, he says. Which brings it back to the thesis. One reading, held by Hartley and by the National Society of Professional Engineers: her inquiry is what triggered the recheck.

Mara: Eli, LeMessurier told it another way. In his account, it was the New Jersey student's call, and a lecture he was preparing for his Harvard students, that led him to compute the diagonal winds. Robert J. McNamara holds that reading too. Which call sent him back to the numbers isn't settled.

Eli: Either way, by the end of July, a full office tower could fail in a storm New York might see within sixteen years. And only he and one wind-tunnel expert in Canada knew the whole of it.

Across Lexington Avenue

Mara: On that island in Maine, he weighed his options. Silence was one. By accounts of that summer, so was suicide: steering his car into a bridge abutment on the Maine Turnpike. He rejected it as a coward's way out. As he later put it, "I made all these calculations, then stopped and said, 'This is real bad.' I made up my mind right then and there, I'll have to do something about it." He called his lawyer and his liability insurer and told them the whole story.

Eli: An engineer reporting his own mistake to the company that insures him.

Mara: The lawyers brought in a second structural engineer, Leslie Robertson, to evaluate his findings. Then, on August 2nd, LeMessurier and the tower's other designer walked into the bank's executive offices across Lexington Avenue. He opened by saying he had a real problem for them. He laid out the flaw and the fix for the bank's chairman, Walter Wriston, and its executive vice-president. Wriston agreed to go ahead. "I just didn't give them a choice," LeMessurier said later. He also recalled that both men treated him remarkably well.

Eli: Quick approval, and a building still filling with office workers every morning.

Mara: So the fix would run as a confidential program, pulling in city building officials, insurers and emergency planners, all inside a tower that stayed open for business. Eli, the night shift is yours.

Torches in the Dark

Eli: The plan was blunt: cover the bolted joints with welded steel plates, each two inches thick. There were more than two hundred joints to do. So every night, after office hours, carpenters came in, boxing the chevrons in plywood and tearing out drywall. Welders took over through the night, with the building's fire alarms shut off. Then cleaners cleared the mess before the first secretaries arrived.

Mara: Flying into LaGuardia one Sunday night, LeMessurier watched the torches working up the tower.

Eli: Publicly, Citicorp said a design review had led engineers to recommend extra welding on some wind-bracing connections, and that "there is no danger." Then, on August 9th, The New York Times came calling. LeMessurier braced himself and rang back just after six. He got a recording. A strike had just shut down the Times and every other major paper in the city.

Mara: That same evening.

Eli: Behind the scenes, the police worked out a plan to evacuate the neighborhood. LeMessurier said that's what scared him: "there was this plan to evacuate all the people who lived within 10 blocks." The Red Cross had about two thousand workers on standby, and weather services watched for storms around the clock. Then, before dawn on September 1st, the news everyone dreaded: Hurricane Ella, off Cape Hatteras, heading for New York. The welding was only partly done. An emergency group met early that morning. Ella turned out to sea.

Mara: The welding finished in October 1978. By then the tower could stand up to a seven-hundred-year storm, with winds so severe they have only a one-in-seven-hundred chance in any given year, even without its damper. LeMessurier compared himself to a parent diving in after a drowning child. "This was my baby. I really believed that if something wasn't done, it really was going to fall down."

Eli: Meanwhile tenants and neighbors sat inside that risk, through a hurricane scare, untold.

Mara: That's where the readings split. LeMessurier said the danger was kept from the public to avoid scaring people. That reading, held by him and by the bank's chairman, Walter Wriston, says secrecy kept midtown calm while the city's emergency plans stood ready. In 1998 the National Society of Professional Engineers Board of Ethical Review saw it the other way. It granted that avoiding panic can be legitimate, but said months of exposure, behind an untested evacuation plan, weren't an acceptable substitute. In its words, "matters of public health and safety must take precedence."

Eli: And after the city's newspapers came back, no further stories appeared.

Mara: Nearly two decades passed before Diane Hartley heard about it. In about 1996 she was upstairs with her young son when her husband started yelling that her thesis building was on TV. Watching, she came to see her undergraduate thesis as what had averted a disaster.

Eli: Eighteen years after a junior engineer told her the building was fine.

Mara: So the listener's question gets a yes, with a twist. Student questions sent LeMessurier back to his numbers, and his own recalculation found the bolted braces that could bring the tower down in a storm. The stilts he set beside a church in 1977 now stand on welds made behind plywood, at night, while the papers were dark.

Eli: What's still open is what reached him. Testimony or letters from the junior engineer who took Hartley's call would show which of her points got through. So would archives showing whether LeMessurier, in later lectures, folded her call into Lee DeCarolis's inquiry.

Mara: A subject to read into is chevron bracing.

Eli: And keep following technology with us.

Mara: And that's the story. I'm Mara.

Eli: And I'm Eli. Thanks for listening, and until next time.