The article cites AvE's youtube video and brings up some skepticism about his hypothesis. His explanation is one of several and it's not yet conclusive.
Watching that video he talks about the column and cables in the middle of the bridge supporting it after construction - The design doesn't include a pier under that tower so can someone who's more mechanically inclined describe how it even helps the strength of the bridge? From my perspective, the more you tension those support cables, the more you're going to bow the center of the span towards the roadway.
To be fair AvE is not a structural engineer he’s a very entertaining tinkerer but I would hope that there are actually qualified people investigating it and I’m not entirely sure why the media is even picking up on folks like AvE.
It's a slow motion photogrammetry of the dashcam video of the collapse showing the bridge clearly failing first at the left side. At least one of the two is probably coincidence, but the point of failure on the lower deck seems to be at the same location that the transporter truck was modified to use, and on the upper canopy at the location where they were doing the tensioning.
This is a really thorough article that explains the reasoning behind the design of the bridge and looks at multiple reasons why the bridge collapsed. I found the animated concept video of what the completed bridge would look like very illuminating [0]. I've never seen a bridge made of concrete/rebar designed like this before. Novel designs are more likely to fail than tried and tested ones; if they built the bridge in a more conventional style I doubt it would have collapsed. Innovation is not always free.
Per-AvE, it's not that the design was necessarily bad. AvE thinks the proximate cause was that the installation plan was changed on-site due to the sort of changed circumstance that the designers didn't know about. See the videos (I posted links already below/above).
Right. I think you could say that the unusual design contributed to the failure, if only because the on-site changes were made using assumptions that would have worked on a more conventional design.
It's kind of like batteries. It used to be that you'd want to discharge battery cells completely before recharging them. This was true for Ni-Cad cells. With newer Li-ion cells, doing this is actually dangerous. It's not necessarily the chemistry's fault when an over-discharged cell starts a fire later on, but if you substitute a new design, you'd better cover the new rules with the implementation people.
"What we can say is that when complex technological systems fail, usually no single factor is to blame. In her study of the Challenger Space Shuttle disaster, the sociologist Diane Vaughan noted a phenomenon she called the “normalization of deviance,” an acceptance of assumptions and shortcuts that over time incrementally piles on risk until, like compounding interest on debt, a kind of technological bill suddenly becomes due." [1]
I used to study mechanics during my mechanical engineering studies and one of my teacher kept saying "the bridge will collapse" as a warning to all of us to take this very seriously. Usually these bridge designs use a factor of safety 5-7. I am really curious what went wrong with this one though. Some of the videos linked in this post are pretty good explaining the basics of reinforced concrete structures.
Factor of safety isn’t really uesd any more, it’s really a combination of load factors and resistance factors. However, the ‘factor of safety’ would have generally been in the range of 2, not 5-7.
Now, there’s dead load, live load, wind, etc, and they’ve all got different uncertainty factors. Resistance factors are assigned based on the variability of the material, concrete more than steel, and fracture critical higher than redundant.
I don't have my code books any more, when I was doing this it was 20 years ago and in a different profession. My bridge is still standing, but it was a lot less risky than that one.
LRFD puts a 1.2-1.6 multiplier on loads (and at time less than 1 for cases where you're relying on it for balance), and resistances are in the .8-.9 range.
I never did a whole lot with the Factor of Safety era codes, but they were generally in the same range once you multiply out all the factors. (As you'd expect, the newer codes tended to hit about the same design point, with some deviations.)
There are two main things to keep in mind here:
* A factor of safety is for normal variability in loads and normal variability in material/connection strengths. It doesn't cover you for blunders.
* The loads that are calculated are the extreme loads on the system, not the normal loads. 1.4 DL + 1.6 LL + 1.2 WL is often the design envelope, and that's... a lot. Add a huge factor of safety on that and you're into physically impossible cases. Like, people are being crushed to death if you get more than 300psf of human load over a significant area, but the typical Live Load is in the range of 100psf. (Now books on a reference library moving shelf system? That's 300psf)
From what I've seen of this bridge, there was no live load at the time, so it's unlikely that it was overloaded as designed. If I was doing failure analysis, I'd be looking at a combination of factors, and the ones I'd start with are the shop drawings vs. the engineer's plans and the quality of the concrete materials, placement, and curing.
AvE's video went through what might have gone wrong, if you haven't seen it.
IIRC, AvE thought it was basically negligence by the installers. One of the mobile supports was kind of near a curb, which was going to make the bridge a pain to install. The engineers installing it decided to move the support, and moved it to a bad spot. Putting the weight of the bridge there caused a bunch of damage. The engineers should have caught it, but they went ahead anyway.
It's like saying all hard drives fail. Kinda missing the point in this situation (HD's fail inside a lifetime, bridges like this should last multiple).
We may not know the cause of the collapse, however given that it was a new construction method being tested, surely the cars should never have been allowed to be there until the span had finished being surveyed. Is a new thing, plan for it failing until you are very sure it won't.
Generally civil engineering projects fail for three reasons, incorrect or substituted materials, or incorrect or poor installation practices, or poor or fraudulent inspections of those. Rarely does something fail due to poor engineering these days. Sometimes it is hard to identify the exact cause after a failure since the failure may obliterate the cause.
The bridge was meant as a symbolic new portal to the university, not just a way to safely cross the dangerous roadway. It would have open views, benches, planters, glass-enclosed elevators, Wi-Fi and a wide 30-foot deck for pedestrians and cyclists.
As a pedestrian and a cyclist, the last thing I want to do on your new bridge over a busy six-lane road is stick around on a bench perusing the wifi.
Having to quibble about why it failed means the margins were _way_ too low. It's a semi-permanent structure, if it costs double (and it wont) more to make it redundant in the worst case (loss of multiple main supports for example) then that's a good thing. Skyscrapers for are built with extensive redundancy, hence the pre-weakening needed before bringing one (or 3) down. I bet when the NTSB is done we find out that this bridge wasn't actually up to code.
