Potential Flaw Is Found in Design of Fallen Bridge

[RamboLiberal]

Source: NY Times

Investigators have found what may be a design flaw in the bridge that collapsed here a week ago, in the steel parts that connect girders, raising safety concerns for other bridges around the country, federal officials said today.

The Federal Highway Administration swiftly responded by urging all states to take extra care with how much weight they place on bridges when sending construction crews to work on bridges. Crews were doing work on the deck of the Interstate 35W bridge when it gave way, hurling rush-hour traffic into the Mississippi River and killing at least five people.

The National Transportation Safety Board’s investigation is months from completion, and officials in Washington said they were still working to confirm the design flaw in the so-called gusset plates and what, if any, role it had in the collapse.

Still, in making public their suspicion about a flaw, the investigators were signaling they consider it a potentially crucial discovery and also a safety concern for other bridges around the country. Gusset plates are used in the construction of many bridges, not just those with a similar design to the one here.

“Given the questions being raised by the N.T.S.B., it is vital that states remain mindful of the extra weight construction projects place on bridges,” Secretary of Transportation Mary E. Peters said in a statement issued late today.



Read more: http://www.nytimes.com/2007/08/09/us/09bridge.html?_r=1&hp&oref=slogin

[illinoisprogressive]

1. How awful. I still feel bad for those in Minn.

[FogerRox]

2. Maybe NYT might have read my post @ D-KOS last thursday

http://www.dailykos.com/story/2007/8/2/221827/0328

The Minneapolis bridge that fell wednesday, the road surface, or road deck was wider than the main bridge structure or main truss. I think this may be very unusual.
The road deck itself required its own wider trusses on top of the main truss because the road deck was wider than the main trusses. Vehicle traffic is normally within the main trusses, not extended outside of the main trusses. This is noticable in this pic from Wiki

.having a roadway wider than the truss is definitely a problem as it violates the basic strength of the truss design, IMHO makes for a weaker design... Think of it in the context of load transfer..... seems very ineffective to me. IMHO one would have to have the road deck be supported by a separate truss system prependicular to the main trusses to properly transfer load from the edge of the 8 lane road deck/trusses to the main trusses. As built this design does not include this road deck truss, thusly cannot transfer load effectively to the main truss.

And you are so right, the twisting stresses were not accounted for in the design. Coupled with the lack of proper load transfer from the road deck to the main truss, along with poor welds during construction and poor maintence..

...well, there is a multi faceted set of problems that led to failure.

http://www.dailykos.com/story/2007/8/2/221827/0328

[Kolesar]

4. The deck of the Main Avenue Bridge in Cleveland is wider than the "truss"

This is a long truss span that is high enough for Great Lakes freighters to pass underneath. It has I-beams extending out that support the outside lanes and a sidewalk, too, IIRC.

[FogerRox]

6. similar yes, but not as bad

http://www.newsnet5.com/news/13837297/detail.html

[Kolesar]

13. You found a picture of the Main Avenue Bridge--way cool

How about:


Or these ones I found on a local photographer's website. I went with the small image to conserve bandwidth. I dare you not to go to his website. I double-dog dare you:


http://flickr.com/photos/70205638@N00/tags/cleveland/

[bluerum]

3. It does not take an engineering genius to realize that the design was deficient.

IMO structures like these should be built with multiple redundant safety features. To defer extended safety margins because of the added expense is irresponsible and elitist. The design specs should be required to include survivability under absolute worst case scenarios.

[jayctravis]

5. I looked at pictures that CNN posted before the collapse...

The bridge looked like it was supported on four concrete columns. The part where the bridge rested on the columns was actually very very small, and it looked like metal sheets had been put under there...as if to "even it out". It made me think of when you prop a wobbly table up with cardboard under one leg.

[Endangered Specie]

10. This is not uncommon...

Go look in a parking deck and you will see that the entire set of T-beams (actually double-T) are supported by a tiny extension of concrete (kinda looks like an upside down cowbell) from the columns. You might think it looks flimsy but the extension is so heavily packed with steel reinforcement, and is probably the most over designed part of the structure!

Keep also in mind that not all concrete is the same, it could be a much higher strength concrete then usual, or have heavy reinforcement. Those metal 'sheets' (plates) are indeed there to distribute the load to the concrete, thats also standard practice.

My point is that appearances are deceiving, and no reasonable structural analysis can be conducted by pictures alone.

[FogerRox]

7. The i35w MN bridge didnt even have a next to best case scenario.

A failure of one single component could led to the entire bridge failing.

[Endangered Specie]

11. That is the case with alot of structures, especially connections.

[Endangered Specie]

9. You got a few things wrong:

Safety factors for things like bridges and buildings are set by the code, they cannot legally be curtailed. They also are pretty hefty. Civil engineers frequently apply more safety anyway.


Also, the design specs *DO* require survivability under worst case scenarios. It is also the engineers job to ensure this, as the design specs are 'minimums' and engineering judgment should be applied to make sure more is not needed.

In any event, civil engineers tend to be *very* conservative... So much so that we would not survive in other engineering fields because our conservatism would put us out of business. (note, Im not talking politically Im talking about design and technical practice). Usually when something falls down (building or bridge) its generally a result of bad construction, not a bad design... and in the cases where the design is the problem, its usually a gross error, not an engineer trying to 'squeeze the numbers'.

I can go into more detail if you want...

[bluerum]

14. Thanks for the insight into the world of civil engineering. However we

still have one "well designed" but collapsed bridge sitting in the Mississippi.

It is of course possible that poor construction was a factor. So poor that even with all the over-designing and redundant safety features and load support that the bridge still collapsed?

Perhaps the bridge was in such poor condition and over loaded that all the well designed safety features were overwhelmed and short circuited.

Possibly a combination of all these (and more) circumstances led to catastrophic structural failure.

I suppose that we have some chance of learning what happened - if the people at fault can't whitewash the findings.

The bottom line is that when I look at a collapsed bridge sitting in a river I have a hard time convincing myself that the design was robust.

[Tesha]

15. Remember: Even bridges are built by the lowest bidder.

Redundancy costs money.

Tesha

[unkachuck]

8. finally....

"Investigators have found what may be a design flaw in the bridge that collapsed here a week ago,..."

....and it only took them 50 years to determine there was a flaw in it's design?....

....you know, a few weeks ago a suspect running from police was charged with the deaths of a TV crew after the crews helicopter crashed while shooting the incident....

....why can't federal and/or state elected officials, whose actions or inactions lead to people dieing, be similarly charge with the consequential deaths?....a public bridge 'falling down' is unacceptable and inexcusable....

[Endangered Specie]

12. Unlike the UL with electrical engineers, the civil/structural profession

does not have an 'overseeing' body. The Professional Engineer who approves the design has the personal responsibility to ensure the design is safe and adheres to code. That in affect is the oversight.