What are you suggesting here? Are you saying that the titanium dioxide is somehow to blame? Or are you arguing that any project that uses concrete mixed on site is suspect?
Concrete cures more slowly when cool, but the longer cure time doesn’t affect its final strength.
Definitely just WAGs from my gut, don’t read too much into it. I have little to no relevant knowledge about building bridges or much of anything else. It just seems to me like it might have been tough getting the mix right with this new special concrete, out there on the side of the road.
I have been involved in many concrete jobs. On site mixing is perfectly fine. The addition of titanium dioxide was probably a very small part of the mix, like coloring. At any rate, sample cylinders were filled from the actual concrete that was mixed, from the middle of the batch. Some are cured in perfect conditions to test for the highest possible strength. Others are cured on site in the prevailing conditions. Comparison of the two results shows how it actually cured.
The bridge sections were cast on site. Concrete samples would have been tested at various stages of cure to determine (f0r example) the 7 day strength, the 21 day strength and the final, 28 day strength. A curing rate chart will be developed and it must meet design.
Dennis
Titanium dioxide is pretty unremarkable…its the colorant that makes almost every white thing white. It’s in your toothpaste, for instance.
And as Mixdenny said, on-site mixing is totally a thing. I have no idea why you think it was “tough getting the mix right,” and evidently, neither do you.
It seems strange to throw around techical theories for a bridge collapse when you know you aren’t equipped to evaluate those theories. It’s harmless enough, I guess…I just find it baffling.
I guess… you must be baffled by a lot of posts in a lot of similar MPSIMS threads then.
And it’s not “strange” at all to want to hear from people who are equipped to evaluate those theories.
The heaviest load on the Golden Gate bridge was when it was closed to traffic for the 50th anniversary celebration and hundreds of thousands of pedestrians swarmed shoulder to shoulder on it. The weight was enough to flatten the slight arch in the roadway and the CEs were kicking themselves for not thinking of applying some strain gauges before the day.
Remember that fire that destroyed an elevated section of I-85 in Atlanta last year? And how quickly they got it restored?
Guess what bridge making method they used to do that. Guess.
Drive carefully Atlanta commuters.
I, for one, learned some new things from the replies to those speculations. So there’s that.
Just put things out there and see if they hold up?
Even so, having a tolerance of a few tons would seem to be cutting it too close.
I think it’s strange that you find it strange. Have you met many humans?
Although I cannot find the article I read, the bridge was supposedly designed for a 90 lb/sf loading with a 20,000 single point load (vehicle) traversing at the same time. If someone has the full dimensions of the complete bridge we can see how many people it was designed for.
And bear in mind, the design load is a small fraction of the failure load, probably 5 to 1 for a man-rated structure. That’s what we use for cranes. For a 20,000 lb crane they hoist a giant weight weighing 100,000 lbs and travel the entire length and width of the tracks while measuring the center point sag.
When we certified the two cranes in a new building when I was part of the team they used some fixed concrete and steel weights then tuned it with a huge water balloon to the exact specs.
Dennis
Wait, are we still sniffing at what was clearly presented as a WAG, or are you referring to the method the engineers used to erect the bridge over a very active and busy road?
Former civil/structural engineer here. There is just not enough reliable information yet even to speculate. I can say that based on my experiences, there is very little chance that the overall structure was underdesigned for the temporary load case. It could possibly be a connection design error though. Connection design can be tricky and if there is a design error causing a failure it is typically a connection design error.
It is not uncommon for a signature structure to be overdesigned to make it look good. One example of this is the Boston Central Artery Zakim Bridge. It is one of the world’s shortest (if not THE shortest) cable stay vehicle bridges. Why? Because structurally it did not need to be a cable stay bridge. It could have been steel box girders or similar. I heard that it was a cable stay bridge because that was the only proposal that got all the powers-that-be to sign off on the Big Dig (This is while I was a structural engineer in Boston in the late '90s working on the Big Dig. Like every other structural engineer in Boston in the late '90s).
An axiom of structural engineering is that concrete cracks. The investigators will have to figure out if the cracks were related to the collapse.
I owe Voltaire an apology. I was tired, cranky and sick, and I let that creep into my second post in this thread. I’m sorry. The last thing I’d want to do (especially on these particular boards) is discourage anyone from asking questions.
I’m still tired and sick, but hopefully less cranky. I’d like to clarify what I found baffling, with the caveat that I understand that my bafflement is no one’s problem but my own:
Voltaire, you cast a gimlet eye toward both titanium dioxide and on-site concrete mixing. I read these as very specific technical criticisms, not questions. I didn’t understand why you would put these out there as theories rather than questions. (E.g., “My guess would be that the titanium dioxide is the culprit” rather than, “Could it have been the titanium dioxide?”)
In contrast, your query about whether cooler-than-usual temperatures could have been a factor didn’t raise my hackles at all–it was clearly a question.
I realize it’s absurd to demand that anyone ask their questions in one particular way. Sorry about that!
Apology accepted, but hardly necessary. I knew I was pretty clearly * not* stating anything authoritatively, so I didn’t see the need to phrase everything as a question (to who?), but hoped for and expected people to treat them as such, as is usual in these threads.
And not to beat the horse (it’s not dead!), but even after considering mixdenny’s excellent reply to my post, I’m no less suspicious of the concrete. Everything I’ve read has noted how large, heavy and overbuilt it was for a pedestrian bridge. That’s a hell of a lot of special concrete to produce at once on-site out on the side of the road, isn’t it? (Note the question mark! 
) You can’t just make lots of small batches when casting a large component, right? Which batch were the testing samples from, all of them?
The concrete wasn’t mixed right there. It was moved into place in one piece, there’s a video of that happening in one of the links earlier.
I’ll have to look for a cite if you don’t believe me, but I did read somewhere that the concrete was indeed mixed right there. Yes, the almost fully assembled bridge was moved into place very far ahead of schedule, but it was fully built there on the side of the road not long before.
For a large pour the minimum requirement is a sample taken every 100 cubic yards. 950 tons of concrete is about 500 cubic yards. I would expect they would have taken more samples then required. They must be taken by a certified tech and involve getting concrete 3 times from the middle of each truck load, mixed together, prodded, settled, etc. For standard cure rate concrete the sample cylinders must remain undisturbed for 8 hours, then taken to a lab and fully cured under specified conditions including 100% humidity and controlled temperatures.
For a structure like this they must avoid cold joints, it should have been poured as fast as possible unless the joints were designed for (like in a large dam where it is poured in blocks).
I have been studying the self cleaning concrete field. It has been around at least 10 years and numerous structures have been built, but there are still seminars to discuss it going on.
One thing is for sure, the titanium dioxide only needs to be in the outer layer or perhaps even parged on. I can’t find a document that states how much they use.
Dennis