Bolt failure at Big Dig: An anomaly?

Boston's Big Dig tunnel project doesn't have a "black box" – like a plane's flight data recorder – to pinpoint why several tons of ceiling tiles fell and fatally crushed a motorist July 10.

But even in the absence of such evidence, experts suggest that the epoxy-and-bolt system used to fasten the concrete slabs overhead is so commonly used that its failure seems unique to the Big Dig.

"That technique is used extensively," says Jerome Connor, a structural engineering professor at the Massachusetts Institute of Technology (MIT) in Cambridge. Epoxy is a high-strength adhesive that often requires mixing on-site before installation. "It's the fact that they only used a limited number of bolts – there was a very low margin of safety," says Dr. Connor, who is not personally involved in the investigation, and notes that it's premature to draw conclusions about the cause of the tragedy.

Massachusetts Gov. Mitt Romney (R) told reporters on Monday that more than 1,100 of the tunnel's bolt assemblies were installed with epoxy. Highway officials plan to reinforce all of the epoxy assemblies with non-epoxy anchors.

For a design like the Big Dig's ceiling, Connor says, engineers often add safety "redundancies," or identical, extra supports capable of holding far more weight than the structure actually demands. In other words, there should have been enough epoxy-and-bolt anchors to hold the ceiling panels in place even if a few of them failed.

In the I-90 Connector tunnel, he contends that too few anchors were used to hedge against failure. "They didn't have enough capacity to carry the load," says Connor. "It was a chain reaction, it occurred very suddenly. There was no room – no margin for error."

But the accident defies what has long been a tried-and-true method for securing panels to tunnel ceilings. Indeed, the Massachusetts Transportation Authority last week enlisted European construction firm Hilti Corp. to install a similar model of load-bearing bolt anchors – but ones that won't rely on high-strength adhesives such as epoxy.

"Obviously there are checks and balances in the construction process," says Marty Schofield, vice president of product safety at Hilti. "Given the history of adhesive anchors in general and construction as well, it would tend to indicate that this was some kind of an isolated situation."

Some bolts from the ceiling wreckage have shown "indications of very little adhesive having been applied," says Mr. Schofield. An accident caused by improper installation or errors in mixing the epoxy, he says, would vindicate the tunnel's design and designers.

That helps explain why the extensive probe – which includes criminal investigators – is focusing on construction errors instead of flawed design. Massachusetts Attorney General Thomas Reilly has already subpoenaed construction documents to determine whether contractors flouted safety protocol. On Monday, Mr. Reilly told reporters that some documents reflected a "substantial dispute" among engineers over whether the anchor system was adequate to hold the weight of the three-ton ceiling panels. The I-90 Connector was slated for routine inspections in August – one month after the ceiling collapse killed Boston resident Milena Del Valle.

Even as investigators scrutinize the construction history of the $14.6 billion project, others are reviving criticism of its management. A chief complaint: Massachusetts lacked adequate supervision of private contractors. "Was this the right management structure for a large project?" asks David Luberoff, executive director of the Rappaport Institute for Greater Boston. Bechtel/Parsons Brinckerhoff, the joint venture leading the Big Dig, was involved in both the design and construction efforts – an arrangement that some observers say may have complicated oversight.

Engineers say a key lesson is constant vigilance. "If you use a procedure that is subject to difficulties, then you certainly need to play it very safe by testing," says Herbert Einstein, a professor of civil and environmental engineering at MIT. "If something is very common, and the consequences are something like an accident, then you need to test a hundred times, a thousand times. We're responsible for public safety."