I-35 Bridge Collapse

[Rodbuckler]

I have reviewed the 2001 inspection/testing/analysis document for this bridge and find that they are accurate in their accessment as far as the obvious is concerned. This is a non-redundant structure with multiple possible modes of failure in the primary structural system (Eight lanes of Interstate traffic on two lines of trusses). The secondary structural system was junk, and the primary system of twin trusses was highly dependent on composite action with the deck as to forces within the trusses. Final determination for cause of collapse will be enlightening.

I would say that we(they) are victims of the Great Depression as far as this bridge collapse is concerned. The design philosophy is consistent of that with senior design engineers of that period. We just need to systematically recyle these bridges into new structures, and maintain the poorer but safe steel structures of that era as our "strategic steel reserve". Modern concrete and steel bridges are much different as far as internal redundancy.

 

[Gary Allan]

Quote:

---

I have reviewed the 2001 inspection/testing/analysis document for this bridge and find that they are accurate in their assessment as far as the obvious is concerned.

---

So, 6 years ago their assessment recommended leaving it go until it collapsed as a course of action toward renewal

As many have said, there's no shortage of politicians jumping on the opportunity to "swing into action".

States Across the Country Race to Inspect Bridges
Keith Bedford for The New York Times


Workers continued recovery operations on Friday at the Interstate 35W bridge collapse. An uncertain number of people were still missing and presumed dead beneath the waters.

One bridge, in Missouri, was closed indefinitely as a precaution. Elsewhere, bridge inspectors were told to cancel their vacations to conduct the emergency inspections and, in some cases, also re-examine bridges with designs unlike the fallen Interstate 35W bridge.

In Washington, the inspector general for the federal Transportation Department was ordered to review the National Bridge Inspection Program, which oversees bridges across the country, including more than 70,000 that have been found to be structurally deficient. Among other things, investigators are to examine whether the necessary repairs are being made to the deficient bridges.



NY Times - registration required

 

[Al]

Quote:

---

If that's the case, the assumptions regarding fatigue life are out wildly.

Heads should roll.

---

Call me crazy..but I will be amazed if Fatigue in and of itself will be found to be responsible (IMHO). Something else failed and I am guessing it involved corrosion and maybe a design flaw that allowed very high local forces. Also note that the collapse occurred at 6 O'Clock where the heat of the day produced close to the greatest expansion. Those joints near the supports look mighty nasty. Note the one closest to you. Thats a lot of joints in a small area.

It will be most interesting to find out exactly what did fail.

 

[Gary Allan]

I'm sure that there's some explanation like corrosion, Al. I would expect that to show up on some inspection report.

Just what is used to determine worthiness?? Strain gauges? Radiography? Sonic tests? A drive-by with a glance and pencil whipping the report?

 

[Al]

Quote:

---

I have reviewed the 2001 inspection/testing/analysis document for this bridge and find that they are accurate in their accessment as far as the obvious is concerned. This is a non-redundant structure with multiple possible modes of failure in the primary structural system

---

Rodbuckler: What I don't understand/like about this bridge is that the arch is a totally compressive member. A normal multi section truss bridge has each section dependent on the others. This is bc the top compressive forces are balanced by the bottom tension members.

I thought for a minute that the arch might be a tension member but this is not the case bc the box beam gets heavier at the ends (hence it is in compression)

All this is terrible bc the arch is actually stabilized by the next arch. So when this thing fails probably near the bottom of one arch its comin' down really fast. Maybe I'm missing something. If not..this a terrible, terrible design (understatement)

 

[Rodbuckler]

Al, this multi-span truss design is a totally different animal than an arch design. This truss design is similar to a multi-span I-girder design with selected pieces cut out of the web. The curved profile of the bottom chord allows the structural depth of the truss to change in proportion to the magnitude of the bending moment diagram.

Thus at midspan the top chord is in compression and the bottom chord is in tension. Near the channel span piers the top chord is in tension and the bottom chords are in compression. The horizontal component of force in the diagonals allows the chords to transition from tension to compression.

No arch action is intended, but you would get some if the bearings locked up. The arch thrust force plus thermal forces would put high demand on those tiny piers.

 

[oilyriser]

If the traffic had been driving quickly, I wonder how long it would have taken for people to notice the bridge was out and slow down. Would wrecks have to pile up and fill the gap before people stopped?

 

[Al]

Quote:

---

to a multi-span I-girder design with selected pieces cut out of the web.

---

OK I get it. But you said:

Quote:

---

Near the channel span piers the top chord is in tension and the bottom chords are in compression.

---

So at the pier, is each section free standing? (more or less)..that is: Is there zero compression at the bottom and zero tension at the top. Does one section physically need the next section to transmit forces all the way to the end? of the entire bridge. Obviously the bridge is tied together at the piers but is tension and compression very low at the supports? I'm guessing its low but does exist.

Thanks a lot..I appreciate it.

 

[Rodbuckler]

Force in the top and bottom chords at the channel span piers is high. The bending moment in the superstructure is maximum over these pier reactions.

 

[1sttruck]

There was a problem with the design limits though, indicated by the observation that some of the steel members were out of plane per a previous inspection. A guess is that it was originally designed for lower loads / use, and over the years traffic volume has picked up, including a lot more heavy trucks. Higher ice and snow loads than originally assumed may have contributed too.

Oregon had some bridges on I5 off limits to trucks over a certain weight for awhile, and they've been replacing the bridges.

 

[Al]

Quote:

---

Force in the top and bottom chords at the channel span piers is high. The bending moment in the superstructure is maximum over these pier reactions.

---

Thanks a lot

 

[Shannow]

I guess that if the sliding feet were starting to drag/stick, you'd like both of them to stick rather than just one.

One sticky foot could provide some pretty impressive loads (once per day) at those "messy" joints that Al pointed out, plus supplying some of the forces that would be inclined to buckle members out of plane.

Speculation.

 

[Rodbuckler]

If you watch the video you will see that the bottom chord in panels 2 and 3 buckles out of and in plane in compression. Also, the top chord fails in tension in panel 1 over the pier. Bottom chord in panel 1 and the tension diagonal in panel 1 remain intact. Truss remains intact from top chord 3 on, and from bottom chord 4 on.

Vertical members that are 2nd (from the pier), 4th and 6th (midspan) are zero-force members that are there to prevent in-plane buckling of the compression chord. All tension diagonals (leaning towards the pier) or all compression diagonals (leaning towards midspan) would have allowed a more natural flow of force through the truss elements without the zero-force verticals.

 

[Mystic]

The last I heard Shannow the reported useage was some 141,000 vehicles a day on this bridge. It was apparently the most heavily used bridge in the entire state and if you look at that bridge it was not designed for that kind of useage. In fact there were figures of 170,000 some vehicles a day on some days. Some of the old bridges that were built very, very well have lasted 70, even 100 years. This bridge was supposed to last 70 years and it only lasted 40 years.

There are already signs of fishy business. Some guy who was on the state safety board got a job after the government work as a consultant for a firm that did consulting work for this bridge. I can already smell a 'Sweatheart Deal.' People who have worked in construction know about those.

In the end of course it will wind up costing more to replace these junkpile bridges with superior bridges than if the bridges had been built properly 40 years ago in the first place.

Heck, in the photograph you can see corrosion! Pretty obvious. Looks like an old ship ended for the breaker's yard.

The way these structures are pounded nowadays (and probably worse in the future) they maybe should have bolts and nuts used in the construction. And a few good old Roman Arches would not hurt any either.

 

[Hooch]

I'm not an engineer, but I won't let that stop me from speculating on the internet.

The abutment at the north end of the bridge could be the cause of the collapse. If you check the pictures, you'll notice it is leaning quite a lot.

 

[Al]

Quote:

---

There are already signs of fishy business. Some guy who was on the state safety board got a job after the government work as a consultant for a firm that did consulting work for this bridge.

---

The all-knowing-after-the-fact-I-told-you-so drones always come out of the woodwork after the action.

Where was he last month.

 

[Turk]

I live in the area there and have traveled across that Brige a lot over the decades...

I believe the structure began giving way just to the side of the King Posts.

We see from -35 to 100 Degrees, TONS of road salt, the new de-icing chemicals and an incredible amount of trucks across that bridge.

It probably just couldn't take it anymore.

Being in Engineering, development and knowing quite a bit about maintenance, Preventative Maintenance (PM's), maintaining something just ain't sexy. There's no Oooo's & Ahhhh's about PM'ing things, so things get neglected.

Also, there's no Repeatability or Reproducibility (Gauge R&R's) in inspections - I've run them. It's hard to get the same outcome or score from different inspectors on different days. Maybe a different inspector would've recommended a shutdown!

 

[Merkava_4]

Quote:

---

The welder was good, but couldn't pass NDT on these beams.

So we NDTed the beams themselves, and found them to be a barely functional collection of laminations that never would have left a steel mill these days.

---

Shannow, does that mean then that the welder could have passed the NDT test had he had some good sound steel to work with?

 

[Mystic]

I worked in the engineering department of a steel fabrication firm. What about you?

 

[Mystic]

Yes, with the very cold temperatures in that area and the snow and all the road salt there could have been a lot of corrosion. And 40 years ago people probably had no idea of the weight (especially overloaded weight) of some of these massive trucks in use today. A typical car, truck, van or SUV is like a fly on a bridge like that. But every massive truck has an impact.

Personally I don't like the design of this bridge. I think a more robust design was needed for the amount of traffic that this bridge had to handle, and the environmental conditions (extreme heat and cold). And there are some 700 of these bridges like this one still out there. Steel truss design like this is okay for a giant barn or a warehouse. But I like support cables and Roman Arches for a bridge that is going to see a lot of traffic.

Inspite of all the welding we did I don't like welding by itself. I don't know if this particular bridge was welded design or not but there have been cases of defective welds in pipes used in nuclear power plants inspite of major inspection and supervision. I still like massive bolts and nuts for use in stuff like bridges. If a bridge is going to see massive traffic and heavy truck use in a fairly hard environment and be expected to last 70 years or so that bridge needs to be overbuilt.