Posted on 08/01/2007 4:28:27 PM PDT by 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.......
That chord should be in compression in a cantilever; the top is the tensor on both sides of the support bay.
http://www.nytimes.com/2007/08/06/us/06collapse.html?_r=1&oref=slogin
tons of construction materials stored on bridge
I was wrong and you were both right about the end of the chord being in compression. In my haste I got stuck on simple support, rather than cantilever and never drew a diagram. I would have seen the torque about the top of the king post then. I'll be looking at the MNDOT site, but not right away.
I see that sheer facture in compression is consistent.
I was wrong and you were both right about the end of the chord being in compression. In my haste I got stuck on simple support, rather than cantilever and never drew a diagram. I would have seen the torque about the top of the king post then. I'll be looking at the MNDOT site, but not right away.
I see that sheer facture in compression is consistent.
Thanks. Yes, I was thinking simple support, not cantilever.
Terrible news...I was just in Minnesota a few days ago, going over the Mississippi bridge that goes through the U of M campus. Prayers for the families of the victims. Hopefully the MN DOT will make an effort to inspect bridges more carefully in the future.
The pier kingposts support almost the entire weight of the bridge. I find the design a little odd, actually. Had the pier had diagonals going down toward it, each of those diagonals would have had to support about half as much compressive load as the kingposts (there would be twice as many of them). If the pier kingpost failed, the bridge would start to drop. The force on the lower chord would increase toward infinity as the bridge did so, until the bridge had fallen enough for the lower chord to start being under tension.
Eh, it happens. I've been playing some with that bridge design page and its sorta fun, though it will occasionally accept bridge designs where part of the lattice is redundantly connected and part is flexible (as opposed to everything being non-redundantly rigid). Such cases can produce really bizarre nonsensical numbers.
It's also interesting, though, to observe the amount of mechanical advantage that's available even when one only puts objects on grid squares. Obviously if one puts objects at finer boundaries one can get really huge levels of mechanical advantage, but even when confined to grid squares the values are still pretty huge.
I clicked on the link to see what it looks like, but firefox gave me a blank page. LOL! I want to upgrade that, but I don’t want to lose my bookmarks.
>>So far, investigators say they have ruled out nothing and will consider everything from the expansion and contraction of the bridge in the extreme weather conditions of Minneapolis to the possible corrosive role of bird droppings.
You gotta be kidding me!<<
Look at the oval panels in the compression chords—the ovals used to be open, but now they’ve been retrofitted with a perforated screen. Why? Pigeons were roosting in them.
When I was a kid I climbed the Northern Pacific railroad truss bridge, about 3 miles upriver from the 35W bridge, with similar oval-holed compression chords. You could climb right up the diagonal beam like a ladder using those holes. All was well until I scared up a bunch of pigeons that just about hit me in the head as they were bailing out, good thing I was hanging on tight...
I’m thinking 30 years of pigeon dung and urea packed in around your welds and rivets is not a good thing for a bridge.
Right on the money with the cantilever diagnosis and diagram of compression and tension members.
You can confirm it in the photo—the chords along the bottom and the ones angled upwards from the piers are box girders with oval holes—a compression member. The ones angled down from the top of the piers are I beams—tension members.
It’s curious that people call it an “arch” or a “deck truss” when it’s so clearly a cantilever—a truss would have tension members along the bottom, and would be tallest midspan, not around the piers. And arch would need a beefier arch for such a massive bridge. (Like the 494 bridge over the Mississippi or the Cedar Ave Bridges over the Minnesota River elsewhere in town—these are 3 or 4 lane bridges and have much fatter tubular steel arch members.) I can see most people getting it wrong, but surely there must be a few people that the media has found that actually understand bridges. (But, when they keep saying cars fell 64 feet to the river and any dummy can tell it’s more like 100’, I guess I’m overoptimistic)
Note also the views from MS Earth, the on-shore cantilever arms are one segment longer than the arms making the central span. So unlike classic cantilevers that have anchored on-shore ends as a counterweight to a suspended truss in the middle, this cantilever has extended on-shore arms, which probably function as something of a “suspended truss” on shore.
Jeffers—the diagrams from p. 49 and 50 of the MnDOT report—can you tell if the “reversal members” are on the center span or the on-shore arms? I’m guessing the on-shore arms.
On that same note, has anyone found a diagram in the MnDOT report that identifies where all the locations (pier 6, L4, etc.) are on the bridge itself?
concrete
is my business wrote:
I have not had much time to watch the videos or see the news but heard some concrete guys here
speculate that the train might play a part in the final report, if the bridge failed at that point. The live load
of all the stopped traffic, the steep slope cut into the bank next to the tracks, combine with shaking from
the train, and all that silt on the bottom, no bedrock there, could have made for ‘a whole lot of shaking
going on’.
*********
I haven’t been able to determib\ne if that was a moving train or spotted cars. The railroad people I know are debating if the line under the southern approaches was active at all. Apparantly it used to extend further west, but was abandoned/torn up/not abandoned/not used/not abandoned/seldom used depending on who you ask.
A train at low speed or idle should be capable of setting up vibrations, potentially of destructive amplitude. More energy to work with at higher speeds, but the doppler component would have to remain within the structure’s resonant range to continue propagating the effect, and the dynamic loads would alter the bridge’s vibration characteristics to the point that post failure modelling may be difficult.
There is precedent for a wide range of resonant frequencies, obviously, as the Tacoma Narrows bridge tended to oscillate at widsspeeds from 25 mph winds and up before its collapse.
I see a bit more damping in this case though.
Advancing traffic can add or subtract to the resonance, and I’ve seen statements from concrete mixer operators that they have to be careful mixing at full speed on suspended structures due to vibration effects.
Mat or floating pile foundations, as opposed to foundations on bedrock could assist or impede resonance, together in concert with fixed and mobile bearing plates, and this bridge has a history of bearings that froze and had to be replaced. Tricky stuff, a lot to look into before drawing conclusions.
A bridge, traffic, cellphone, or security cam showing the reverse angle of the one we have now would be useful to watch.
That little truss builder is an interesting program. Too bad it’s not running on a Cray, and...and...and...
Getting deeper into this...so far I see:
1. SE Kingpost probably buckled under load. Eyeball says it didn’t just fall over sideways and bend when it hit the ground. It died ugly.
2. SE span 7 bottom chord separated near the kingpost, SW span 7 bottom chord did not. It failed further out.
3. Per the 2006 inspection report, the main trusses cantilever the approach spans at the crossbeams/truss ends. The crossbeams bear the riverward ends of the approach span deck beams. The crossbeams are supported on rocker bearings by the main trusses. In 1986 the SE rocker bearing froze and had to be replaced. The whole bridge was closed and the “span had to be jacked up” to replace the pin and repair related damage.
4. In one part of the 2006 report (page 12), it says all four rocker bearings now function normally with “obvious signs of movement. In another section (page 30), it says “SW rocker bearing has no movement”. The SW rocker bearing may have been frozen as of the 2006 report.
5. The SE crossbeam suffered significant damage when the rocker bearing froze in 1986. This was repaired by adding plates, drilling cracks, and adding braces.
6. There is significant evidence of similar damage at the NE crossbeam around the rocker bearing, which has also required repairs.
7. Further south, on span two, there is a hinge joint which is frozen in the maximum expansion position. This has rocked pier one to the north.
8. The “roller nest” under the SE kingpost showed “no obvious signs of movement” as of the 2006 report, and the bearings showed “rust and corrosion”. It may well have been frozen as of that time.
The interplay between all these points is still somewhat hazy, but it’s clear that there were expansion problems, especially on the south ends of both trusses and the north end of the east truss.
It’s also clear that somehow significant loads had to be transferred to the SE kingpost, if it indeed buckled under load.
Supercat discussed failure of the kingpost and subsequent loads imposed on the bottom chord trending near infinite.
Post collapse imagery shows a buckled SE kingpost and a bottom chord fracture consistent with extreme compression loads.
I’m not yet able to to connect the bridge’s expansion problems, expansion related damage, potential effects of jacking the structure up, potential cantilever dynamics in the event of crossbeam failure, and the SE kingpost and bottom chord failures, but I’m starting to disbelieve the idea that these are unrelated coincidences.
In the big picture view, we have a known frozen expansion mechanism near the south end of the approaches, a historical frozen rocker bearing at the SE corner of the truss system, a potentially frozen bearing now at the SW corner of the truss assembly, and a potentially frozen roller nest under the SE kingpost. Additionally we have significant damage at both the northern and southern truss end/crossbeams. Finally we have imagery of a mangled SE kingpost and possibly spectacular compression failure of the heavy bottom chord box beam.
Speculation:
If expansion issues failed the southern crossbeam, (or just the eastern end of it), the truss assemblies there would be relieved of the counterbalancing weight of the approach span. That could induce a moment around the base of the southern kingposts, and could fold up the truss assembly where the cantilever transitions to a simple truss on the mainspan just north of the southern piers.
However, I don’t see how that could translate to high compression loads on the kingposts, of the kinds which are required to buckle them and make the compression loads on the bottom chords go infinite.
I also don’t see any rotation towards centerspan of either the SE or SW kingposts in the post collapse imagery. I do see a lot of “missing” road deck in that area, as discussed earlier.
The best (tenuous) connection I can come up with so far has the east end of the southern crossbeam failing, and the moment, instead of rotating the SE kingpost about its (possibly frozen) base, created enoug instability and interaction with adjoining members to cause it to buckle, which then compression loaded the bottom chord beyond its design strength.
If the SE kingpost did buckle, it could put all the horizontals and diagonals between the SE and SW kingposts into excessive tension and pull the SW kingpost over sideways as indicated by the post collapse imagery. In the process, the south end of the mainspan seperates from the kingposts and cantilever group, drops, and the rest is as previously discussed.
There’s weak circumstantial support for this. From a birdseye view, the SW kingpost (and organic truss panel) is rotated clockwise from where it would be if it simply leaned over to the east. If the SE kingpost buckled beginning at it’s base, the lower crossmembers tying it to the SW kingpost would tear losse first, possibly leaving the upper crossmembers attached long enough to pull the SW kingpost over laterally.
At the same time, with the eastern truss collpasing with the SE kingpost, the western truss near the south pier is sagging towards the river, trying to pll the SW kingpost riverward off its base.
This is consistent with (but by no means the only explanation for) what we see in the pictures, the rotated western truss pier panel.
I know it’s thin, I don’t like trying to turn a rotational moment from cantilever counterbalance loss, about the base of the SE kingpost, into enough compression to buckle it, but right now, it’s the best mechanism I have to connect historical and current expansion problems to what we see in the pictures.
Thoughts?
I didn’t hear about the bridge collapse because I was on a road trip to Texas.
Bout four minutes before hitting the bridge in Memphis over the Mississippi, I said ‘lets switch to Fox News for a minute’ and heard the following;
“...the bridge collapsed, and we don’t know how many people are down in the water of the Mississippi......’
Took two commercial breaks before they mentioned in was way up north...made for a couple of interesting minutes approaching the bridge to say the least.
There are reversal members on all three truss spans.
Regarding shore-end cantilever counterweights, see my most recrent post above regarding “crossbeams”. In effect, the riverward ends of approach spans functioned as counterweights.
You are making far too much sense! And please keep me on your ping list. This is very interesting.
How did those birds know all those cement trucks were going to park there?
The NTSB is saying they didn’t find anything significant on the south side of the bridge, and are now focusing on the north side to find the cause of the collapse. How does that figure into your figurings? I’m not an engineer, so I only understand about half of your notes, LOL!
Yes there are photos of a train on tracks that run under the bridge, partially crushed by the bridge.
>>I haven’t been able to determib\ne if that was a moving train or spotted cars. The railroad people I know are debating if the line under the southern approaches was active at all.<<
In 1967, Northern Pacific ran under a south approach span, the Great Northern ran under the north side approach span (where the RR cars are today). Neither was a heavy freight line, used mainly for passenger trains to access the Great Northern Station (at Hennepin and the river).
The RR line clearance was the reason the bridge was so high. It had to have a fair bit of clearance over the river channel (30’, 40’?), and if it went under the RR tracks it wouldn’t have enough river clearance. So it ramped up over the RR tracks at both ends, making it extremely high over the river gorge.
Today, the bridges used by both RR lines are ped bridges, the tracks on the south have been abandoned, and the tracks on the north are just a spur line to the grain elevators. Used for storage. If there was a train in motion there at the time of collapse, it was probably switching cars at about 5 mph, with a 10 mph speed limit.
Since the collapse started on the south cantilever platform or in the middle of the span, it’s unlikely a train under a north approach span had anything to do with it.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.