Report: Minneapolis bridge design firm dropped ball

WASHINGTON — Critically important load-bearing calculations were inexplicably never performed during the design of a major bridge in Minnesota, setting up the catastrophic collapse of the span last year, a federal investigation concluded Friday.

The error, attributed to the principal design firm failing to properly size steel plates, was the primary cause of the Interstate Highway 35W bridge collapse near downtown Minneapolis on Aug. 1, 2007, the National Transportation Safety Board said in its final report.

Minnesota and federal authorities contributed to the accident, the report said, by failing to properly review the bridge design and failing to detect during inspections damage that developed in the plates well before the accident.

Even more than a year after the accident, the lessons learned are slow to take effect, investigators warned. Current state and federal design-review processes may not be adequate to detect design errors, the board was told.

The safety-board report issued nine recommendations to the Federal Highway Administration and the American Association of State Highway and Transportation Officials.

The proposals included strengthening government oversight of design contractors and improving the training of bridge inspectors to identify problems similar to those uncovered after the Minneapolis accident, which focused national attention on the problem of structurally deficient bridges in the nation.

One-half-inch steel gusset plates that tied beams together on the I-35W bridge were half the thickness they should have been to maintain structural integrity, investigators determined. Gusset plates are expected to be stronger than the structural beams they connect.

Thirteen people died and 145 others were injured when the 1,907-foot bridge broke apart, crashed into the Mississippi River and spilled 17 vehicles into the water. Undersized gusset plates at eight joints on the main center trusses of the span were fractured and prompted the failure, investigators found.

But why the glaring design error occurred — sloppy drafting work, miscalculations or no number-crunching at all — wasn't conclusively pinpointed during the safety-board investigation.

Investigators said they assumed the anomaly was an error of omission.

Even after 40 years of service, the deck truss bridge would have been able to handle vehicle traffic and the weight of construction equipment and materials on the span at the time of the accident if the steel gusset plates were an inch thick, investigators said.

But as it was designed, the I-35W bridge was an accident waiting to happen. Unknown to state officials conducting routine inspections, the undersized plates in the middle of the span had become distorted, or bowed, due to structural fatigue over many years, finally giving way.

Investigators said they inferred from the evidence that not all the necessary calculations were done to decide what size plates to use. No documents revealing such computations were found in the files of the engineering firm that did the work, Sverdrup & Parcel and Associates, said Joseph Epperson, a safety-board expert in structural analysis.

Sverdrup & Parcel is no longer in business. The engineer who made the decision on the gusset-plate size died before the accident.

A replacement bridge built to current standards opened this year.