Breaking: Bridge Collapse in Minneapolis

KSTP TV 5/ME | 8/1/07 | Me

[ButThreeLeftsDo]

Just turned on the news. 35W bridge collapsed in the Mississippi River. Cars, trucks, semis.....

Fires burning, tanker trucks, at least one school bus, more than ten cars......

Just now breaking.......

[Palladin]

Until I see an actual picture, I’ll file that one under “myth”.

All I remember is Bush hugging Brown and saying “Helluva job, Brownie” while hundreds of people slowly drowned in their attics.

[spunkets]

You’re welcome.

[toast]

I already read that. I just didn’t want to admit to reading the Trib.

Click on the graphic of the bridge halfway down the page. Very informative view of where the investigation will probably focus.

[jeffers]

That's a good truss diagram that explains the basic design element of simple trusses well.

In this case however, it's not a simple truss. It's an arched truss, and in my opinion, has a strong cantilever element to it as well. Here's a picture of the bridge, looking northeast, before the collapse:



A simple truss has flat bottom and top chords, as indicated in your excellent diagram.

An arched truss incorporates the strength of an arch into the truss framework, usually as the top or bottom chord, sometimes in between the two.

All of the above can function as standalone units between piers.

The cantilever design comes into play when superstructure components on the shoreward side of the mainspan piers counterbalance all or part of the weight of the center span superstructure. If one span of a cantilever fails, the adjacent spans usually fail in sympathy, because the counterbalancing weight has been removed.

I agree that this bridge is not a true cantilever. There are no massive counterweights attached to the approach ends of the superstructure, there are no free ends in the center of the mainspan, and no pivot pin to join two cantilevers together.

However, it's not a simple truss either. There's an arch to contend with, and it is very clear, on two counts, that the spans are not independant.

Obviously, the northern approach span failed after the center span dropped. It failed just as you would expect half a cantilever to fail, by rocking shoreward in the absence of its counterweight. Just as obviously, the approach spans were not designed to be independent, as the superstructure ran over the piers on into the approaches without interuption at the piers as is normally seen in simple post and beam, post and plate, or post and truss design.

Without the engineering drawings, and a lot of processing time, I can't say to what degree arch type, truss type, and cantilever type load distribution mechanisms were in play in the static design, much less in the dynamic forces of daily use, and further still from the dynamics of failure.

The best I have to go on here are the images, statements from MnDOT to the effect that the bridge was designed such that failure of any major component would fail the entire superstructure, and the specifics of the failure sequence itself.

At this point, rather than labelling the design a truss, or cantilever, or arched truss, about the best I can do is call it a unitary design, somewhat like modern automobile frames are put together, putting several different load distribiuting mechanisms into play to create one overall structure. In fact, I can't even say whether the bottom chord was in tension or in compression. That arch is a pretty flat one, as arches go, but arches are in compression, as are the bottom chords of cantilevers. As you note, the bottom chords of trusses are in tension, and its beyond my abilities to say whether this design was more of a truss, arch, or cantilever.

It's hard to tell in the photos, but my impressions are that in your first photo, the failure of the bottom chord, as far as visibility allows are at least consistent with a compression failure, i.e., diagonal shear. That's not good enough to be conclusive, and I'm not engineer enough to analyze this design just from the pictures.

On the good side, from the article posted above, MnDOT is looking at the same general area we are, southeast side of the bridge. Between us and MnDOT we have one or more layers of usually clueless reporters, but we're in the same ballpark no matter how you look at it.

[brityank]

Thank you all for the educated discussion you've all provided, it is appreciated.

I just watched a blurb about the MNDoT Bridge List, and they showed some pictures of the structure of this one from (I believe) the 2005 report. [ It runs as "Bridge History" under KARE-11 OnLive ^ ] They showed missing bolts, severe corrosion, and misaligned members in the steelwork support structures.

I'm sure there are probably more than a single cause to the failure, just as there is a multiple cause for the dearth of maintenance to the infrastructures -- self-aggrandizing politicians, money-grubbing lobbyists, and an uninterested public.

[spunkets]

"He said inspectors have long been on the lookout for metal fatigue and cracking in the bridge because it was designed before engineers learned about dangers to bridges from fatigue cracking."

Thanks for the article. The existence and practical effects and consequences of metal fatigue's been known and understood since the 30s and that was definitely so post WWII. This bridge was built in '67. They're still passing the BS about how they didn't know.

It's interesting, how in the same article, the folks that know and are responsible for spending the money, say that there was plenty to spend on fixing dangerous faults. The others say flat out in the face of that, that there was none. The fact is that the engineers didn't appreciate and know what they had in front of them and froze. IOWs, it was over their heads. MNDOT designed and built the bridge in the first place. Now they're doing a song and dance over the poor results.

I can't see a politician, rep, or rat that would turn down $s after a warning that their bridge is a jalopy that's threatening to fall in the water. The reason's are obvious. Regardless of credentials, they're inept bureaucrats frozen in the state of ignorance, confusion and indecision. I doubt many will be entertained by their song and dance. They may be taken in by it. The Gov is probably fuming and the rats are certainly over joyed that this occurred in a rep admin.

[spunkets]

Thanks.

"it's not a simple truss either. There's an arch to contend with"

I know, I was just giving a simple explanation. The bottom chord is an approximation of a parabola. That allows all the triangular elements in the span to be reduced in size, as they occur in the truss away from the simple support. It simply reduces the overall weight of the truss, while maintaining the support capability in each element. The bending moments are the same for each element ina parabolic truss.

At the support post, the truss element(triangles) is large, because it supports half the weight out to the center. The next element can be smaller, because it supports less weight to the center. ect...

"On the good side, from the article posted above, MnDOT is looking at the same general area we are, southeast side of the bridge."

Right. I think they knew about this before, but froze.

[spunkets]

You’re welcome.

[Sweet_Sunflower29]

Oh heavens. I am so very sorry...

[jim_trent]

> “The existence and practical effects and consequences of metal fatigue’s been known and understood since the 30s and that was definitely so post WWII. This bridge was built in ‘67. They’re still passing the BS about how they didn’t know.”

Yes and No. Fatigue was not really known until metal airplanes. It was not well known in the aviation industry until after WWII — after several fatigue failures. It was not used elsewhere because it was expensive to design for that (mostly, it could not be designed for — parts were made and tested to failure) and most things were heavier than airplanes and did not need to worry about fatigue.

The information on fatigue was filtering down to the highway industry in the mid to late 1960’s. I had some of the early articles written on it at the start of my engineering career. A lot of it was VooDoo. I saw a newer article on it a while back that is much more usable. Over-the-road highway trailers were just starting to get light enough back then that fatigue started to be a problem — particularly in gasoline tankers (they went from a tank on a frame to a monocoque structure with the shell actually carrying the load and they also became totally welded at about that time).

Fatique information had not made it to the bridge industry even then. They had a bridge at Sioux City over the Missouri river that was built sometime in the 1970’s or so that started fatigue cracking within a year of its completion. It had to be replaced in just a few years.

Back then, the engineers thought that bridges would never see enough cycles to cause fatigue cracking (at least, that is what they are teaching in bridge inspection courses). Obviously, they were wrong. A combination of using calculators and computers to pare down the weight to the absolute minimum and the increased use of welding was the major causes (note that riveting and bolting structures, which is rarely used now, means fatigue cracks are less likely to start and will end at the next joint instead of cracking completely through the structure).

As far as you statement that politicians would NEVER fail to correct a problem — you have obviously never dealt with politicians. Unless you can give them and EXACT date and time when it will fail, they are not interested. Sometime soon is not good enough. Read my “All About” page here if you want to read about my experiences on that.

[spanalot]

“Reckonin” and “cypherin” are my favorite words.

[spanalot]

I see that there is a navigable canal and lock near the end of the bridge that seemd to drop first - wonder if there were any collisions over the year - a dent in the steel tube could drastically reduce its compressive strength - the lower chord is in compression, right?

[luvie]

Gotcha....yew betcha!

[sodpoodle]

Interesting map of USA showing percentage of deficient or obsolete bridges by state

http://www.msnbc.msn.com/id/20093413/?GT1=10252

apologies if this has already been posted.

[spunkets]

"Yes and No. Fatigue was not really known until metal airplanes. It was not well known in the aviation industry until after WWII "

Short... Yes, one of the first engineering studies was done by a Boing engineer regarding airframes. It was shown dramatically that even mild steel does that, when a 30 y/o ship slit i half. ~ in the 40s.(CRS) The process itself though is a solid state phenominon that was studied in physics, but for different reasons. I just gave the WWII date, because the subject occurs in 50s era metallurgy and engineering books. I realize that there are different quality and levels of texts and that more detailed knowledge doesn't spread far.

Yes, I have dealt with politicians. I can't say much now, except they are just people. It's like advising on critical car work. The car owner has to make a decision regarding the mechanic's claims. The mechanic may not be able to say when the wheel will fall off, but he can make the case as best he can that it will in fact do so. The pol must make the decision for the community regarding public property. Something like a bridge is simple. If the engineer sees somehting like a bridge containing fatigue cracks, or even unsafe stress levels, his duty is to inform both the exec and the public, because it's public property and safety involved. The public certainly understands bridges and "public" buildings collapsing, and anyone that is a pol certainly should.

[fox0566]

Kiran is a lib, but a few corrections to your post-

Kiran didn’t start working at FNC until March of 2001, and didn’t start as a fill-in co-host for Fox and Friends until about a year later. Yes, she did remark that Edwards was “dreamy”, but I’m lead to believe that she was saying it as a joke at the time.

I figured that Gretchen was a Republican, but didn’t have any proof. She is just who Roger Ailes is looking for in a co-host for his morning show. Chetry is way too far to the left for any kind of mainstream weekday program, at least for anything that would have been on FNC. I think she’s on cloud nine at CNN! You can tell she’s taking great pleasure in clobbering the heck out of the Bush Administration and his Iraq War policy! I wonder if she’s having fun hanging out with Christianne, Wolf and John Roberts? Somehow, they seem so intellectually above her!

[spunkets]

The lower chord is in tension. It’s unlikely anything hit the bridge. That’s a major incident that would have been reported and repaired if there was any damage.

[sodpoodle]

http://www.msnbc.msn.com/id/20099056
Just a sample of a chart of the 245 bridges in the District of Columbia - which has the worst rating in the US.

Keys to using the data below:
Status of “D” means the bridge has been rated structurally deficient; “O” means it has been rated functionally obsolete.

Sufficiency rating is a value obtained by calculating four separate factors (structural adequacy and safety, serviceability and functional obsolescence, essentiality for public use, and special reductions) that is indicative of a bridge’s sufficiency to remain in service. A value of 100 represents an entirely sufficient bridge and a zero represents a deficient bridge.

The raw dataset of the National Bridge Inventory, which is compiled by the U.S. Department of Transportation’s Federal Highway Administration from state reports, used for analysis was provided to MSNBC.com by the National Institute for Computer-Assisted Reporting at the University of Missouri. The most recent data includes inspections through December 2006.

Status Rating Road Place County Feature Location
O 64.0 DISTRICT OF COLUMBIA POTOMAC R. & C&O CANAL “OVER POTOMAC RIVER
O 63.0 29 DISTRICT OF COLUMBIA WHITEHURST FRWY WHITEHURST FRWY
O 74.0 DISTRICT OF COLUMBIA SOUTH CAPITOL STREET M STREET & S. CAPITOL ST.

[spunkets]

forgot... The parabola comes from the thickness squared dependence of the bending moment of a cantilever as the distance from the support increases. The beam will remain flat if the thickness decreases parabolically as the distance from the support increases.

[Abigail Adams]

LOL! Here’s a link to the bridge sufficiency list from MNDOT:

http://www.dot.state.mn.us/bridge/Misc/xyz/FHWA-DeficientBridgeList.pdf