Engineers puzzle over bridge collapse

http://news.bbc.co.uk/2/hi/americas/6927526.stm ^

[traumer]

Engineers are trying to understand what caused the catastrophic collapse of the bridge over the Mississippi river in Minnesota.

Resurfacing work was taking place, but the bridge was last inspected in 2006 and no significant structural problems were found.

Such complete bridge collapses are a very rare occurrence.

If they happen, it is either because the load is too heavy, or the connections between the bridge's structural elements are too weak, Keith Eaton, chief executive of the UK's Institution of Structural Engineers, told the BBC.

"The engineers will have to see where the collapse started. Clearly a failure occurred somewhere which imbalanced the whole thing," he said.

Speculation that hot weather contributed to the accident by weakening the concrete or expanding the steel framework was not a likely explanation, he added, as modern bridges are built to cope with extremes.

A crack in the steel making up the bridge's structure was the most likely explanation for the disaster, he said.

Corrosion

The I-35W highway bridge (Bridge 9340) was built using a framework of rafters, posts and struts - a structure known as a truss bridge.

In 2005, it was one of thousands across the US rated as "structurally deficient" on the federal National Bridge Inventory database.

It rated 50 on a scale of 100 for structural stability in that study, White House press secretary Tony Snow said.

About 140,000 cars are thought to have used the bridge every day, but a 2001 report by University of Minnesota's civil engineering department found traffic levels were below those the bridge was designed for.

See graphic of the bridge collapse

The report went on to express concerns that a single crack in the main truss could "theoretically" lead to the entire bridge's collapse.

However, it also said that even if there was a crack, the load could "theoretically" be redistributed along the steel trusses or the concrete deck of the bridge, keeping the bridge aloft.

It added that no fatigue cracking had occurred, and that the bridge "should not have any problems with fatigue cracking in the foreseeable future".

File photograph of the Minnesota bridge The bridge crossed the Mississippi River near downtown Minneapolis

The state need not "prematurely replace this bridge because of fatigue cracking, avoiding the high costs associated with such a large project".

The truss bridge was built in 1967, with eight lanes over a span of 581 meters (1,900ft). It had no piers in the water, allowing easy passage for river traffic.

While no longer the cutting edge of bridge design, truss bridges are relatively cheap to build, and were a very popular structural choice in the US in the 1960s and 1970s, Mr Eaton said.

They have a downside, however.

"They are made of lots of complex pieces of metal, interconnected bolts or rivets," Mr Eaton told the BBC.

"They have little corners between two pieces of steel where water can collect and cause corrosion."

Nesting pigeons could also be an issue.

"Their droppings are very corrosive, which can be a problem," he said.

[cherry]

somebody on the tv said that they drove over that bridge before the collapse and there was a gaping hole that the workers were working in.....you have to wonder....

[cherry]

I am anxious for “seconds to disaster” on this one....

[Cheburashka]

Did Al Gore make any comments ?

None to date that make any sense.
Come to think of it...

---

Yeah, but you’re talking about his entire political career, not just comments about this bridge.

[Cheburashka]

I blame gravity.

[SuziQ]

Little spray heads popped up on the road and sprayed corrosive chemicals all winter long.

Yep, that oughta do it.

[TXnMA]

As Don W said, the term is "scouring".

~~~~~~~~~~~

I have spent several hours doing frame-by-frame failure analysis of the collapse video -- using "pixel-matched 'difference-mode-transparency' analysis" of minute differences between frames. Basically, this technique subtracts one frame from its predecessor.

Where nothing has changed, the result (of subtracting a number from itself) is zero -- which in RGB is BLACK. Any slight (non-zero) change stands out brightly against the black background.

Here is what I have concluded:

~~~~~~~~~~~

On the bank on the side with the the navigation lock, the concrete footings and piers are set back several feet from the edge, and were shielded by a concrete wall at the water's edge. Scouring obviously was not a problem there.

On the other bank, the concrete footing/piers were very near the water's edge, and were surrounded/protected only by "loose" rock "rip-rap".

That footing/pier structure did shift and tilt slightly toward the water -- but it did so only well after the main span had fallen into the water and the "splash wave" had subsided.

(Thanks to Danette!)

In the surveillance cam movie, the last visible collapse is of the section whose end rested on those piers. It appears that when those piers finally shifted/tilted river-ward, the landward end of that roadway section was pulled off its support. The fall of that section "yanked" the section of roadway that rested on the "kingpost" supported by those river-side piers -- back toward the shore.

That "tug" from the collapsing and tilting roadway then tilted (rotated) the "kingpost" that rested on those piers back toward the land -- and, finally, off of the tops of the piers. That is the final, sudden "collapse" that many have commented on (and some ridiculuously implied was caused by "demolition charges").

Again, the failure of the footing/piers that may have been affected by scouring occurred at the very end of the collapse, and clearly did not appear to be the initiating POF (which occurred on the opposite side where the piers were well up on the shore).

~~~~~~~~~~~~~~~~~

FWIW, I followed the live thread, and observed that jeffers did a credible job of analyzing the collapse -- as the images became available (thanks, in many cases, to Danette). I expect that jeffers and I are in in pretty fair agreement as to how the failure progressed...)

[evad]

Good comments guys...thx for the info.

[Eye of Unk]

I suspect a resonance was being built into the structure of the bridge with concrete mixers on one side of the 4 lane while the other side was gridlocked in traffic, I drive concrete mixers and they are just like a side type of washing machine that has say a huge wet sleeping bag in it and makes a thump thump thump except we are turning over 20 tons ( the mixer I drive is actually rated at 12 cubic yards @ about 4,000 lbs per yard)and I know for a fact on a bridge improvement project I was on this spring in Alaska my sole mixer induced a harmonic wave that set up in the span and made it start bouncing several inches enough to be distinctly noticeable much more than normal, the roadwork on the I35 bridge had several from what I saw on video much larger front discharge concrete mixers. I hope enough people with more background in wave resonance and such look at my theory. Perhaps the structure started to have side to side stresses never engineered at the time it was built couple with wear at key points.

[4yearlurker]

It was global warming and the tax cuts that made the bridge collapse. See.....it is Bush’s fault. (S)

[Leftism is Mentally Deranged]

Just blame Bush. The media will support you on it.

[Bean Counter]

Excellent points! I never thought of a concrete truck in that way before, but the washing machine analogy makes perfect sense. I’m with you on the vibration...

The center span (the portion actually over the water) looks like it could easily have a resonance introduced into it from any number of sources, because it looks to be mechanically isolated and damped by the shore structures and the piers.

If the construction workers inadvertantly hit just the right frequesncy of vibration, from whatever source or combination of sources, or even if the shock waves induced into the structure reflected back onto each other (and multiplied the forces at whatever point they met) the results would eventually be devastating. Do it right, and you can in fact turn that bridge into a giant, self-destructing, 1100 foot long tuning fork using only construction equipment.

I’ve seen what big chunks of metal can do when you hit them exactly hard enough, with exactly the right force, in exactly the right spot. Metal will fail in the most spectacular ways if you fatigue and stress it properly...

I’m standing by my thoughts on “inferior steel” as well, at least until we hear from the NTSB. I know they will do a metallurgical analysis of the steel, and if they rule it out, I will stand corrected. Ultimately that may be a contributing factor to this as well; there likely will not be a single cause to this, rather a combination of things.

[Popman]

I know this is probably a minority opinion, but I still think it is possible that the bridge come down by terrorism via small charges at key points on the bridge structure

My thinking it was a trial run to see if it could be done.

Tinfoil hat shined and firmly placed on head

[Eye of Unk]

I am speculating that the loaded concrete mixers which are extremely heavy with a full load of concrete may have contributed to inducing an unknown variable in wave harmonics or inducing a wave.
I have been posting this because I really do have some serious background on that theory.

Firstoff I am not an armchair general or looking for popularity.

Its BECAUSE I am a supervisor at a concrete batching facility in central Alaska with 12 years experience of driving concrete mixers. This spring I was on a bridge of two lane that is approximately the same length and design as the I35 near Talkeetna, Alaska, the supervisor of the earthquake improvement project notified me my concrete was too “dry” or stiff and asked me to reslump it by adding water and re-mixing it at high speed for several minutes.

The result immediately felt was like being on a suspension pedestrian bridge and jumping up and down, the whole bridge was actually bouncing, everyone thought it was an earthquake which we have a lot of but it was too coincidental with what I was doing at the same time.

I saw on some video what appeared to be several even larger concrete mixers on the bridge, could this have contributed to making the structure to start a destructive “bounce’?

Anybody seen the Mythbusters episode where they tested Teslas theory that a small simple oscillating device can shake a huge steel structure like the Carquinez Straits Bridge in the Bay Area of California? A weak link added with corrosion, weight on the bridge like slow moving traffic on one side with construction equipment on the other with jackhammers and all this was probably never figured into the original design specs 40 years ago especially without computers or simulation models.

Other than sabotage and terrorist I believe I have a working theory, there could be still another variable like a weak base at the kingpost and it shifted.

( I just posted this in another forum but it fits better here so its being copied and pasted)

[Bean Counter]

This image is taken from a Bathymetric Survey of the foundations of the Interstate Bridges that carry Interstate 5 across the Columbia River, between Portland Oregon (Left side) and Vancouver, Washington (on the right). the top of the picture is West.

The dark blue color represents the deepest water, and red represents the shallowest. The dark blue area toward the top is the main navigation channel in the Columbia River, which follows the Washington side in this area.

The piers for the bridges are visible across the river. You can clearly see how deep the scour is behind the Oregon side piers for these bridges.

There are two sets of piers, built side by side. The oldest set were built in 1917, and were constructed on top of wooden pilings that were pounded into the riverbed. The second set were constructed in 1965, and are on similar pilings.
The severe scour, and the substandard pilings add up to a severe seismic vulnerability on these bridges, and in an earthquake of the right magnitude, soil liquifaction could cause the spans to roll right off of their piers.

So what are Oregon and Washington doing about this??

We are spending $65 Million Dollars of our Transportation money to figure out how to leave those spans in place, and build a Light Rail system instead. (I swear I am not making that up, there is a proposal floating around to do exactly that.)

In any case, the funding for anything to do with this project is years away, and construction even further out.

So it goes...

[SauronOfMordor]

I’m not saying that it was low quality steel per se, but these structures were never meant to last much beyond fifty years or so, and that was under the engineering standards that existed at the time that particular bridge was built in that particular location. The steel that we produced was high grade for the time, but due to process improvements over the years, the material used in 1960 likely would not meet the building standards of today. There are simply much better formulations for structural steel now, than then.

We have the bills coming due on the Baby Boomer generation retiring onto Social Security. It looks like we have an infrastructure bill also coming due, from all the things built during the 60's highway boom coming to the end of their service life

We can not afford the welfare state any more

[jeffers]

I found your post fascinating, but want to clarify before commenting.

Are you saying that after mainspan collapse was complete and the splash had subsided, the next event in the sequence was lateral movement of both the pier and sidespan superstructure, off the sidespan’s northern supports and towards the south?

That the pier tilted, dragging the sidespan with it, BEFORE the sidespan superstructure dropped?

I’m not arguing with you, because I have not done the kind of analysis you have, but if this is what you’re seeing we’re talking about a near inconcievable energy budget at work, and I don’t want to go digging in that can of worms till I’m sure that’s what your analysis shows.

[evad]

FReepers amaze me with their insights.

Excellent points!

[RightWhale]

Haven’t seen a good view of the structure. Have seen a so-so view of what they have now. Was there a steel structure for a span support? It looks like the steel box collapsed, which would let the entire structure down. A single failed rivet would do that if the steel box were built without redundancy.

[Gideon7]

[Doe Eyes]

The bridge had an active de-icing system that ran in the winter time. Little spray heads popped up on the road and sprayed corrosive chemicals all winter long.

I would be very surprised if the engineers that oversaw the installation of the de-icing system didn't consider the corrosive affect it would have on the bridge. My guess would be that it was minimal.