TOPICS > Nation

Design Flaw Cited in Investigation of Minnesota Bridge Collapse

January 15, 2008 at 6:20 PM EST
Loading the player...
The National Transportation Safety Board revealed findings Tuesday from its probe of last year's deadly Minnesota bridge collapse, citing a design flaw with steel "gusset plates" used to build the structure. NTSB chief Mark Rosenker details the agency's investigation.
LISTEN SEE PODCASTS

TRANSCRIPT

JIM LEHRER: Next, what was behind that big bridge collapse in Minnesota? Ray Suarez has that story.

RAY SUAREZ: It was the middle of a busy rush hour last August when the Interstate 35 bridge in Minneapolis collapsed into the Mississippi River.

Thirteen people were killed, and more than 100 were injured. The eight-lane bridge was also carrying the weight of heavy equipment and materials for a repaving project at the time of the collapse.

This video from an Army Corps of Engineers surveillance camera captured how the bridge collapsed while the cars were still on it.

Questions about the condition of the steel deck truss bridge and its age — it was 40 years old — quickly led to new inspections of bridges and highways throughout the country.

It was one of more than 77,000 aging bridges that had been labeled as “structurally deficient.”

For months, federal investigators have been looking into the accident. At a news conference today, they confirmed a critical part of the problem was a design flaw with the bridge itself.

Explaining bridge design

Mark Rosenker
Chair, NTSB
This is a fracture-critical manufactured bridge, which if one critical element breaks, the entire bridge will collapse. And that is what happened here.

RAY SUAREZ: For more now on the findings about that flaw and the wider investigation, we turn to Mark Rosenker, chairman of the National Transportation Safety Board.

And your investigation turned up one critical component?

MARK ROSENKER, Chair, NTSB: The critical component that we found -- and we announced today -- is that there was under-design, under-sizing of the gusset plates. And there were approximately eight positions where that occurred.

RAY SUAREZ: What is a gusset plate? What is it there to do underneath the bridge?

MARK ROSENKER: Well, simply stated, it is a component of the bridge which ties the beams together. It is normally considered a stronger part of the bridge, even stronger than the beams that it ties together.

In this case, calculations were made that provided for an undersized, under-strengthened, if you will, gusset plate which could not hold the loads.

RAY SUAREZ: And this bridge was designed in a way so that if this one piece, the gusset plate, fails there's nothing to keep the whole thing from falling down?

MARK ROSENKER: There's no redundancy. This is a fracture-critical manufactured bridge, which if one critical element breaks, the entire bridge will collapse.

And that is what happened here. The U-10, the under-sized gusset, broke. And when it broke, the entire bridge collapsed.

RAY SUAREZ: So you'd found several of these plates failed, broken, when you brought up the pieces from the river, is that right?

MARK ROSENKER: That's correct. That is correct, approximately 16 of them, joints. But we found and believe that it is the U-10 gusset plate which was the first to go.

Some factors still unknown

Mark Rosenker
Chair, NTSB
The question is, what was different on that day, on August 1st at around 6 o'clock, that was different than the day before, the weeks before, the years before?

RAY SUAREZ: So by putting this attention, this focus on the gusset plates, are you ruling out the structural steel, the concrete, the other parts that make up a bridge like this one?

MARK ROSENKER: This is what failed. It was undersized. It was under-designed. And as a result of that, weights over the years, the improvements that were made from 1977 and 1998, put enough weight and pressure on it that ultimately it was going to come down.

What we're going to try to continue to find out over the next series of months is, what actually broke the proverbial camel's back here, the weights that were on there, conditions of the bridge?

A whole host of factors are being studied at this point to understand why the U-10 gusset plate failed at that time.

RAY SUAREZ: Now, when you released this information, the results of the investigation, you stepped back from calling this the cause of the collapse. Why, if you're so sure that these plates failed, are you not ready yet to say, "This is why the bridge fell down"?

MARK ROSENKER: Well, it is why the bridge fell down. It broke. It could not support the weights. The question is, what was different on that day, on August 1st at around 6 o'clock, that was different than the day before, the weeks before, the years before?

That's what we're trying to understand: What was the straw that broke the camel's back at that time?

Other bridges

Mark Rosenker
Chair, NTSB
When we uncovered the design flaw in this bridge, is there possibility of one somewhere else? It's a long shot, a very long shot.

RAY SUAREZ: Do we know if there are a lot of other bridges in the United States designed in this way, with this under-strength, undersized gusset plate?

MARK ROSENKER: We don't believe this is a systemic issue. But when we uncovered the design flaw in this bridge, is there possibility of one somewhere else? It's a long shot, a very long shot.

But when, in fact, you can do the recalculations necessary to understand exactly what the safety margins are that were built into the bridges that are currently standing, then we believed -- and we made this recommendation to the Federal Highway Administration -- that when, in fact, changes to the weights of a bridge, when you're going to make enhancements, when, in fact, you're going to make operational changes, allowing additional traffic to come over the bridge, then we believe it is prudent to do the recalculations on all of the elements, with particular emphasis in examining the gusset plates, to make sure that the margin of safety has been built into it.

RAY SUAREZ: Because this isn't something -- and correct me if I'm wrong -- that would turn up in a normal inspection. This bridge was episodically inspected. And this isn't even something that the engineers would be looking for, am I right?

MARK ROSENKER: That's right. There is an assumption that when the design is provided to the states and provided to the builders that the design is safe and has the appropriate margin built into it.

Now, this one, for whatever reason, we believe there were some miscalculations done in the original engineering drawings that provided for this to happen. The bridge was built to the drawings that were specified.

RAY SUAREZ: Now, if it turns out that there are other bridges in a similar situation, can they be fixed, retrofitted, or do they have to be rebuilt or taken out of service?

MARK ROSENKER: It depends upon the condition of the gusset plates. It depends upon the beams. Some can be reinforced. Each are unique; each have to be assessed with the calculations done with the appropriate engineering.

Government recommendations

Mark Rosenker
Chair, NTSB
There are 465 bridges of a similar design that we saw the I-35 bridge. There are another 12,500 bridges that have gusset plates that need to be examined, if improvements are to be made.

RAY SUAREZ: What is your agency, what is the federal government telling states around the country in order to be sure of where we stand now with the tens of thousands of bridges in the United States?

MARK ROSENKER: Ray, we made a recommendation today to the Federal Highway Administration.

That recommendation was that if, in fact, you are going to do any changes to the weight of your bridge or change operationally the amount of traffic or the weight to the traffic that's going to be coming by, a specific recalculation of all of the elements, with focus on gusset plates, needs to be done before any work is commenced.

There are 465 bridges of a similar design that we saw the I-35 bridge. There are another 12,500 bridges that have gusset plates that need to be examined, if improvements are to be made.

This is what we did today in recommending to the Federal Highway Administration that, when these processes, when you finally decide that you're going to make an improvement, recalculations of the structure must be done if you're going to maintain a safe infrastructure.

RAY SUAREZ: Now, that site has been cleaned up. In fact, there's already reconstruction on a new crossing of the Mississippi.

Do you have everything you need to be able to plausibly build a chronology, based on the investigations that were concluded in November, when you released the site to the state? Do you know everything you need to know at this point?

MARK ROSENKER: We believe we have a very good picture and good documentation of what's necessary to continue on with this investigation.

The analysis aspects continue. What we're attempting to understand right now is, what were the forces that were different on August the 1st at around 6 o'clock, that were different, that created enough weight to actually break those gusset plates and bring them down?

We'll learn that; I'm confident we will. And we'll have that when we determine our probable cause in the upcoming months.

RAY SUAREZ: Chairman Rosenker, thanks for joining us.

MARK ROSENKER: It's a pleasure.