×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Contact US

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Mexico City subway overpass collapses, killing at least 23 and injuring dozens
2

Mexico City subway overpass collapses, killing at least 23 and injuring dozens

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

This forum was my first stop for information (after news articles). I watched the security camera video a number of times. My first fleeting impression is that the failure occurred at mid-span of a simply supported beam, rather than at the end connections. I'll be interested in following this thread.

Edit to add: I found the best copy of the security cam video of the collapse on Twitter, from Voice of America, https://twitter.com/VOANews. No distractions of news heads, just the security cam.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

This seems to be the location:

https://goo.gl/maps/BzFDuS1Av8RHAsbh9

Maybe the connection area to that kicked out stringer is the source of the problem?

Construction PE (KY)
Bridge Rehab, Coatings, Structural Repair

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Based on the poor video of the collapse it looks to me that the overpass failed mid span rather than falling off a column, the spalling on the underside of the span seems to be close to where it failed.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Good spot.

Right where that beam connects in at mid span - Is that a crack in the flange?

Picture from one side



Picture from the other side



That connection does look more than a little odd.

Looking at other joints it seems there is a construction weld in the middle of the span.

Something failed hideously fast though.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Not the best looking weld is it?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

I would have guessed that the tracks were on top of the beams, but it appears from the overhead news picture that there are 2 sets of tracks where there are 2 beams so the tracks must straddle the beams. The third beam starting to split off to the right is for the start of a 3rd set of tracks.

I count at least 5 welds in each bottom flange on each beam in the section that failed. So, that's a welded construction beam. Is the flange in that type of construction commonly made from pieces butt welded together? I would have expected the flanges to be continuous pieces.

Is it also possible the train punched through the deck and that shock loaded the beams enough they failed. Maybe it de-railed? In the news picture the right end of the train looks to be about a 1/2 car width out of line with the track.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Maybe it is just my eyes, but the top flanges of those beams appears to be not as wide as the bottom flanges. That would only be logical if the beams are composite with the deck. But the deck sections appear to be precast, so how does that work?

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Fatigue took out one of those butt welds?

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

From the photos the train appears centred over this girder:



This is roughly midspan. There is a large crack opened up joint in the deck, and associated corrosion, and a fairly prominent bottom flange splice in the same location. You can see pits etc in the weld.





If deck joints are cracking, and rusty, the beam may have been working harder.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Speculation. Comments and contradictions welcome.
1. The train jumped the tracks.
2. The train broke through the rotten concrete.
3. The train broke the beam butt welds on the way down.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

The failure was sudden, so definitely it was fatigue cracking of the butt weld across the bottom flange - category E' by AASHTO ( 2.5 ksi constant amplitude fatigue threshold). Anyway, although shop welding of the elements of the railway bridge is permitted in controlled environment and with the welds checked with X-ray, this was definitely not the case, as the welds looks bad, and were likely done in the field, which is strictly forbidden by the AREMA. If the rest of this elevated structure was done the same way more such tragic accidents are to be expected.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

I don’t think the train jumped the tracks. It looks like the span simply broke in two.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

you beat me to the punch wiktor. I agree with your assessment; my gut feeling after seeing the photos was fatigue. In the photo from Tomfh, there appears to be a floorbeam at midspan but the there's a large gap on the web between the bottom of the connection plate and the bottom flange; not a good detail. I would be interested to know the fracture toughness of the steel; may have played a factor as well.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

That 'crack' in the deck is actually a joint between two precast deck planks. I agree that the butt welds look rough, but doubt they were made in the field.

Any ideas as to why the top flange is smaller than the bottom?

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Quote (hokie66)

That 'crack' in the deck is actually a joint between two precast deck planks

Yeah.

Looks like they've been patching up joints in other areas? This is several spans further along.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

The shop welds in a railway bridge shall be grinded smooth and x-rayed - so much for good workmanship. Any irregularity in the tension elements acts as a stress riser
The smaller top flange was due to assumed composite action of precasted deck. With the deteriorated joint that was not the case, and increased stresses in the critical section.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

I agree that there was poor workmanship in the welds. That is not surprising in Mexico, but still doubt the welds were made in the field.

As to composite action, I suggested that above as the reason for the smaller top flange. But if that was assumed, how could it have actually been accomplished, with those precast deck sections? If there was not actually composite action, the top flange may have failed before the butt welds in the bottom flange. Anybody have photos of that failed area?

Just supposition, but perhaps the deck was to be cast in place, but the contractor decided to use precast, without anyone giving due consideration to the implications.

The articles said this was a "subway overpass collapse", so actually part of the Metro system, with this part being overhead trestle. Not sure how extensive the overhead part is, but the entire bridge system needs a rethink now.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

The composite action of precast decks is actually quite simple - holes in the precast planks where the shear studs are located, filled with HS grout. The quality of the fill in between planks is critical, as these shall be done extremely well. To avoid locking DL in the steel the span shall be uplifted during installation of the planks.
I did not notice initially the big difference in flange sizes which could be critical, as the local buckling of the top flange could lead to lost of plane bending. The diaphragms holding the bottom flanges doesn't look too strong, and the one just below the crack appears slightly bowing. So, it would be either fatigue crack at the tension flange, or buckling of the entire beam.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

That concept may be simple, but it would require well planned and supervised execution. I've never seen it done, but that means little. Anybody else used that system?

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

This article says the line was closed for structural repairs in 2014, including the collapsed span, and that there was cracking after 2017 earthquake.

https://www.bbc.com/news/world-latin-america-56985...

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Quote (wiktor)

So, it would be either fatigue crack at the tension flange, or buckling of the entire beam.

Beams seem well aligned after collapse, so presumably the former?

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

To help aggrivate matters, from the photo, it appears that the spandrel was discontinuous in the vicinity of the failure to accommodate a light standard of some kind.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

I think that is just a non structural fascia, not a spandrel beam as such.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Quite likely, hokie... just added dead load... it looked substantial enough that it would provide some stiffness and a marked discontinuity. At about 0.6 span...



Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

That connection of the divergent beam at the mid span I don't think is coincidental.

It looks quite strange, the main beam is the same size, it's a point load in the middle of the span and appears to rest on the bottom flange only.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

What is striking to me is how cleanly both of the girders are fractured, suggesting fatigue failures of both girders? Is it possible that both girder bottom flanges were substantially cracked through without being noticed at any point?

If it were only one girder which had a crack through the bottom flange, I would have expected one of the girders to maybe fail in fracture, and the other to fail in a more ductile manner. But this does not appear to have happened.

I wonder how often these bridges were inspected?

Agree with the previous poster that the precast decks were probably intended to be made composite by grouting a shear stud into a recess and also grouting between the deck units. This would explain the narrower top flanges. But there appears to be cracks between the adjacent panels, which might suggest that there was less (or no) composite action than anticipated and the bottom flange was working much harder than designed for (leading to fatigue cracks through the flanges at the butt welds).

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Is there a chance the steel may not have been weldable? or maybe poor low hydrogen welds?

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Since hydrogen contamination of welds only requires storing the rods in humid conditions and not in a rod oven, that could explain both the poor appearance of the welds, and cause hydrogen embrittlement.

This is easy to prove either way with relatively simple forensics.

Photos published in the Washington Post

As often occurs this disaster is already impacting the politics of Mexico.
https://www.nytimes.com/2021/05/04/world/americas/...

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Yes, a material failure, either in the base metal or the weld metal, could be at issue.

But I keep going back to the small top flange. While the concept of developing composite action sounds attractive, the reality of introducing enough shear connection between a narrow flange and those precast deck units is questionable. That would require a lot of connectors, I think in a single line, so the deck units over the beam would have to be like Swiss cheese. And then the joints between the panels have to be grouted, leaving no gaps, at least over the beams.

And as gusmurr said, if composite action is not achieved, the bottom flanges would have to work too hard.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Same picture as tomfh, but there's something odd about that truss. It's the only one which is not in line with the lines of the slabs above. It actually looks slightly bent but I thought that was google street view being not precise.



Did it get "shifted" so they could repair the gap in the concrete above it?

The column set up also changes at this point to move over for a three track layout vs 2 track leading to a bit of instability in the column forces.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

That flared beam coming in does appear suspicious. It would add a mid-span reaction to the exterior girder, and also the load is eccentric to the web centre-line, adding a torsion to the main girder as well. At the same time, the same simple cross frame is located on the back-side as exists along the span.

The "butt" weld is likely (hopefully) a complete joint penetration weld (CJP), also called Full-Penetration weld. In the highway design codes, if properly tested and ground in direction of stress, the stress category is "B", which is pretty good and better than the weld of the vertical stiffeners to the bottom flange ("C'").

My guess is that with 1 day of looking at the drawings, they already know the causes - but have to properly document it before it is released to the public.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

This may be indicative of the workmanship or lack thereof.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

sometimes composite members have smaller top flanges, in particular in areas where material costs are high and labour costs are low... I have no idea of how it was designed or misdesigned.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Waross, your derailment first idea does have a bit of truth to it. When the story first broke the first reports mentioned the operators complaining about overweight trains. They complained that the overweight trains were wearing out the tracks for over a year. A worn out track could lead to a derailment. Operators usually aren't keen on observing structural condition so track likely does relate to the rails, sleepers, or ballast.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Yes, rail wear, or perhaps a switch problem, could have led to derailment and an impact caused collapse. In some way, that switch location is probably involved.

Dik, yes composite members can have smaller top flanges, but the question is how composite this steel/precast connection was in practice.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

I had the good fortune of visiting Mexico city in 2017 right after the Puebla earthquake. there was a lot of odd damage around mexico city, mostly settlement related. The ground that the city is built upon is so soft, all the buildings are leaning one way or another.

Mexico city is such a lovely place to visit. beautiful thing about their metro train system, its only 5 pesos (~25 cents US) to ride that train anywhere along the vast array of lines.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

One possibility may be a split switch.
That is, the switch hangs up part way through its travel.
A split switch will cause a derailment.

Bill
--------------------
Ohm's law
Not just a good idea;
It's the LAW!

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Except it sure looks like the train was on the other track, the one that doesn't split.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

There is a cross over from one track to the other.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

LittleInch,
The truss you refer to is called a cross-fame, used for lateral stability during construction and to account for lateral load transfer such as wind between the girders. Also, if the top flange was unbraced (non-composite), it would provide top flange bracing in the final condition. It does not look like that was the intent though. I noticed it earlier too, it is puny in my experience, especially for a light rail bridge. It does give the appearance that it is buckling.

IC

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

Forensics Firm DNV Tapped to Probe Mexico City Subway Collapse (ENR) Norwegian risk management firm DNV will lead an independent forensic investigation into Mexico City’s subway overpass collapse.
Using a foreign firm is likely an effort to ensure the appearance of a politics free investigation.

RE: Mexico City subway overpass collapses, killing at least 23 and injuring dozens

==> Follow-up 06/13/2021:

NY Times is reporting the following, that the steel studs that make the steel beams & concrete slab composite were poorly welded:



Link

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members! Already a Member? Login


Resources

Low-Volume Rapid Injection Molding With 3D Printed Molds
Learn methods and guidelines for using stereolithography (SLA) 3D printed molds in the injection molding process to lower costs and lead time. Discover how this hybrid manufacturing process enables on-demand mold fabrication to quickly produce small batches of thermoplastic parts. Download Now
Design for Additive Manufacturing (DfAM)
Examine how the principles of DfAM upend many of the long-standing rules around manufacturability - allowing engineers and designers to place a part’s function at the center of their design considerations. Download Now
Taking Control of Engineering Documents
This ebook covers tips for creating and managing workflows, security best practices and protection of intellectual property, Cloud vs. on-premise software solutions, CAD file management, compliance, and more. Download Now

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close