Kolkata Overpass Collapses At least 24 killed

I've driven by bridges that look like they're going to do that. That is some scary shit. wow

An executive for the construction company building the 2-kilometer overpass, IVRCL, said the collapse was not the company's fault, CNN IBN reported.
"It's nothing but a god's act. So far in 27 years, we have constructed several number of bridges ... it never happened," K Panduranga Rao of the Hyderabad-based company told reporters.

Click to expand...

Gods plan is fucking horrible.

Not an expert on the subject, but I very recently read Why Buildings Stand Up and Why Buildings Fall Down, two books by Mario Salvadori a structural engineer.

A number of things could have gone wrong. If I have to make a guess, I would say more than cutting corners or using material of a lower quality than what the calculations where made for (the security margins are very very conservative for this type of things and the bridge was not in use), the fact that it collapsed while it was being built makes me think a boneheaded move was done. It could be something as simple as using part of the currently unused bridge as a storage area. Very convenient, but if you store concrete slabs or iron beams you can rapidly reach a weight that is well above the expected operational weight that the bridge is supposed to withstand, leading the vertical structure elements to buckle. And buckling is very much something that has slight forewarning before a brutal and sudden collapse.

In the pictures (EDIT: but not in the one posted here), one can see what seems to be a vertical metallic element that is totally bent just like it would be after buckling, but it's not possible to tell if it was the first to buckle or if it buckled after some other support elements gave way.

In India engineering is done relative to units Vishnu of force and mass. It's likely the engineer was thinking seven Krishnas load on those supports and the construction workers took it to mean seven Ramas instead.

Let's just say that the huge majority of building don't fall down when being built or when being used, so for it to happen something exceptional must happen, which includes the exceptionally stupid and/or unlucky! You usually need both because you need a bad decision and then no one noticing it in time. The storage thing is one of the example in Salvadori's book, but it was in a building and they noticed it before it was too late (easier to notice also because it was the horizontal elements that were bending more than expected).


Remember thatbuilding that collapsed in Bangladesh in 2013? Look what it took for it to fall down (notes by me):

- Building built first without authorization on a pond (note: it's like building on a sponge, the ground move as you add weight to the building. Very bad unless you have foundation deep enough to find solid ground).
- Conversion from commercial use to industrial use (note: the main impact here is the operational weight expected on a given floor. An industrial building is a lot more sturdy than a commercial one as heavy machinery is expected)
- Addition of 4 floors above the original permit (note: unclear if the structure was designed with these 4 additional floors in mind or not)
- The use of substandard construction material which led to an overload of the building structure aggravated by vibrations due to the generators.

It's both horrifying (especially when you considered that cracks were seen in the building the day before it collapsed and people were sent to work anyway - 1000+ deads) and reassuring that a total shit show is required for a building to collapse instead of just a single of these issues.

All steel structures look "kind" of like that (not the part that buckled and collapsed). That's why in our connections we use a 7/8" x 1" horizontal slot instead of a round hole for 3/4" bolts. So they can be trued up and squared after all the steel is erected. An 1/8" per connection multipled by however many connections you have adds up to a pretty decent amount when it's all first hung. AISC has actual numbers to use in erection you aren't supposed to exceed as far as out of true and square but everyone eyeballs it and goes by experience. I've seen some pretty wonky structures in the last 25 years that get trued up and square and you would never notice the difference when they are finished.

It's also nice in the petro chemical industry if you have large diameter piping or large duct work you can kind of "lean" the structure to make it all line up if needed with a good field crew.

Think about that next time you see a steel building going up and you enter it later lol. You can also have a crew knock off at 3pm when shit won't fit at all someone is panicking that shit isn't fitting 60' off the ground. Then have them come in before daylight and it fit like a glove because of the temperature difference of not being in the sun all day. Ahh the fun I have at times lol.

I will say I'm kind of suprised they didn't have a temporary support (unless it's buried under the ruble) while they were still working on it.

Also in that first pic in the OP I will say those splice plates on the web and flange of the girders where they come together would NEVER fly in the US and the AISC (American Institue of Steel Construction). That's not what failed but that looks incredibly not thought out and I'm suprised they didn't have a tear section with the bolt holes that close to the end of the girder. Those bolts are extremely close to the edge in a high stress connection, not to mention it doesn't go down the full depth of the web.



Here's a normal splice plate design for just a simple beam, not a girder and notice the difference. Their edge distance in the first set of holes to the edge is nearly the same as their next set of holes.



This is a larger girder, but you get the idea.

The walkway collapse is the gold standard of the shop making a decision they weren't qualified to make. I see it a lot. I suggest lots of design changes all the time. For ease of building, ease of erection, based on experience from what I have seen in the past. But always got to get the engineer of record with a stamp to sign off on it. I see lots of times where people/companies skip that part.

Also I'm guessing they lack any quality control or standards organizations to ensure the the bolts, steel. Concrete, etc are correct such as astm, aci etc. While the US has been doing this a looooong time.

I if any of you are at all familiar with India, the real shocker is that this doesn't happen everyday. That right there is the real power of God. In a country where bloated regulatory policy is a 50$ bribe away from irrelevance, it's not just the construction industry where you will see horrifying craftsmanship.

As an aside , I know someone who was fairly high up in water/waste management engineering in a large city in India. He was blown away by the ability we have to flush toilet paper. He wasn't able to fathom how it didn't end back up in the drinking water. I didn't want to ask about the obvious implication but just know that next time u get anything with water in it(even a drink) in a 3rd world country