• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Mexico City Metro Bridge Collapse

Status
Not open for further replies.

Ashley Hill

Established Member
Joined
8 Dec 2019
Messages
4,058
Location
The West Country
The BBC are reporting that an overpass has collapsed whilst a metro train passed over it. It was on line 12 but no location has yet been given. It collapsed onto the road below adding to the disaster. Combined deaths announced so far is 20. How awful and sympathies to all concerned.
Mexico City metro: Fears structural failure behind deadly crash https://www.bbc.co.uk/news/world-latin-america-56985164

Edit :- link added.
 
Last edited:
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

nlogax

Established Member
Joined
29 May 2011
Messages
5,684
Location
Mostly Glasgow-ish. Mostly.
The video of the collapse that's doing the rounds on social media is horrific. No warning at all, one second traffic is flowing quite happy under the structure, next second the road is piled with concrete and metal and a giant debris cloud. From the angle of the footage it looks like a deck section between two pillars completely gave way.
 

185

Established Member
Joined
29 Aug 2010
Messages
5,475
Re the collapsed section.

I assumed a straightforward concrete failure, however at the point the line narrows from 3-track to double, the triangular concrete trackbed (with the points on) seems to be quite badly resting on the pillar support. All the other rectangular sections without points on look fine, this is the only place where this angled interface in box girders occurs. Additionally, I notice by zooming in on Google maps some random rocks lying at the base of the pillar that have been swept neatly next to the column. https://www.google.com/maps/@19.305...gTmo0AcTRd5EeMOzTjgQ!2e0!7i16384!8i8192?hl=en

Seems to also be a lot of play in the next column at the start of the double line - if the section has shifted a few inches due to a train braking heavily, it could explain why it has just dropped.
 
Last edited:

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
26,610
Location
Nottingham
View attachment 95491


Here's the collapsed section.

I assumed a straightforward concrete failure, however note the red box shows a joint which *may* be nothing, but at the point the line goes from double track to single line, this triangular concrete trackbed (with the points on) seems to be quite badly resting on the pillar support. All the other rectangular sections for double and single line sections look fine, this is the only place where this angled interface in box girders occurs. Additionally, I notice by zooming in on Google maps some random rocks lying at the base of the pillar that have been swept neatly next to the column. https://www.google.com/maps/@19.305...gTmo0AcTRd5EeMOzTjgQ!2e0!7i16384!8i8192?hl=en

Seems to also be a lot of play in the next column at the start of the single line - if the section has shifted a few inches due to a train braking heavily, it could explain why it has just dropped.

The video still on the BBC homepage (screenshotted below) seems to suggest that the section you show is still standing, and it's the next section to the left that has collapsed. Also it seems to be three tracks to two not two to one, and the train was on the track with no points.

1620130342518.png
Can't tell for certain, but I think it may have failed at mid-span, near where the taper actually finishes. See fifth photo in the detailed story:
Mexico City metro overpass collapse kills 23 - BBC News
Yet the incident did not occur on one of the older lines, which have been through at least two major earthquakes in the past 35 years. Rather it happened on Line 12, completed as recently as October 2012.
There will be difficult questions for the mayor's office to come about the construction of the line, including for several former mayors.
 

185

Established Member
Joined
29 Aug 2010
Messages
5,475
Apols I had edited mine before you had posted it, from underneath it looked like two to one. Seems there is a stabling loop.

You're right, it does look mid span - from underneath, the colours of concrete slabs are of different shade to the ones on the other side of the failure, perhaps some repair had taken place.
 

nlogax

Established Member
Joined
29 May 2011
Messages
5,684
Location
Mostly Glasgow-ish. Mostly.
Yes, going by the sleeper lengths in the BBC photos it looks like this is where it may have failed. Curious to see what structural engineers will make of this particular section.

Screenshot 2021-05-04 at 15.19.40.png
 

Ediswan

Established Member
Joined
15 Nov 2012
Messages
3,256
Location
Stevenage
Yes, going by the sleeper lengths in the BBC photos it looks like this is where it may have failed. Curious to see what structural engineers will make of this particular section.
It is a complicated feature which only exists at the ends of the loop. As such, it provides additional opportunites for design, construction, or inspection/maintenance errors. For example, if somebody got the metalurgy wrong, the welding may have resulted in a weaker structure than was planned.

On the other hand, it may be entirely innocent.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
26,610
Location
Nottingham
It is a complicated feature which only exists at the ends of the loop. As such, it provides additional opportunites for design, construction, or inspection/maintenance errors. For example, if somebody got the metalurgy wrong, the welding may have resulted in a weaker structure than was planned.

On the other hand, it may be entirely innocent.
Indeed. Also worth noting that the train was on the track that isn't above this part of structure. If it was going to fail here it was probably more likely to as a train passed directly above, so that more of its weight was being borne in the critical area.
 

ac6000cw

Established Member
Joined
10 May 2014
Messages
3,432
Location
Cambridge, UK
Yes, going by the sleeper lengths in the BBC photos it looks like this is where it may have failed. Curious to see what structural engineers will make of this particular section.
I'm definitely not a structural engineer (so this is speculation), but with the extra plates and welds in the area where the diagonal beam joins the main one, I wonder if water might have got into the joint area and started unseen corrosion. Plus with the points above that bridge section, it's likely to have extra vibration and lateral stress at times compared to 'plain line' so stress cracking maybe?
 

Grumpy Git

On Moderation
Joined
13 Oct 2019
Messages
2,221
Location
Liverpool
Also worth noting that the beam on the angle in this section only, is half the depth of the others! Also note that the large beam to which it is welded is essentially carrying all the additional load at this end of the beam.
 

DanielB

Member
Joined
27 Feb 2020
Messages
1,193
Location
Amersfoort, NL
Yes, going by the sleeper lengths in the BBC photos it looks like this is where it may have failed. Curious to see what structural engineers will make of this particular section.
I'm not a structural engineer, but looking from the other side (directly underneath where the train fell down, https://goo.gl/maps/X1gmDdP7bSDZ4joe7 ) the concrete appears to be in worse shape around where the bridge failed than in other sections.
It looks like water has been leaking trough there, potentially compromising the structure in a way that might be poorly visible during inspections. What also stands out: there's a weld in the metal beams supporting the span exactly in the spot where the bridge failed.
 
Joined
7 Jan 2009
Messages
946
From the Google shots, the structure doesn't look in bad condition. The photo of the incident seems to imply that the end of the beam closest to Olivos station failed first, possibly at the end being supported by the pylon: so corrosion of the reinforcing bars in the beam might be an issue leading to catastrophic collapse. But that's a pretty extreme scenario. Fatigue fracture in the steel box girder is also possible, particularly if the metal was subsstandard when installed (pace Bouch and the Tay Bridge...)
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
26,610
Location
Nottingham
From the Google shots, the structure doesn't look in bad condition. The photo of the incident seems to imply that the end of the beam closest to Olivos station failed first, possibly at the end being supported by the pylon: so corrosion of the reinforcing bars in the beam might be an issue leading to catastrophic collapse. But that's a pretty extreme scenario. Fatigue fracture in the steel box girder is also possible, particularly if the metal was subsstandard when installed (pace Bouch and the Tay Bridge...)
If I have my geography right, the end closest to Olivos is still resting on its pier, so I don't think that's the first point of failure. It's hard to see how the first failure could have been falling off the pier at the other end unless the pier itself moved, which would have displaced or distorted the beams on the next span. I don't see any evidence of either. In either of these scenarios the beam must also have had some major weakness at its midpoint to account for it splitting in two. So I'm still favouring the idea that this even has been caused by a fracture at or near the joint of the tapering beam with the main one, though if this joint is the point of failure then I'm still puzzled by the fact that the train initiating the collapse was on the other track.
 

DerekC

Established Member
Joined
26 Oct 2015
Messages
2,287
Location
Hampshire (nearly a Hog)
New Civil Engineer has a news item which doesn't provide much specific information but is interesting as background:

https://www.newcivilengineer.com/la...apse-05-05-2021/?eea=*EEA*&deliveryName=DM410

Litany of construction faults flagged years before Mexico metro bridge collapse​


05 May, 2021 By Catherine Kennedy


Poor materials, sleeper defects and inadequate supervision during construction are among faults previously identified on Mexico City’s metro line 12 which collapsed onto a busy road on Sunday night.
At least 23 people were killed and dozens more injured after an overpass on the line collapsed onto traffic near Olivos station in the southeast of the city.
Faults in the $1.7bn (£1.2bn) line had been highlighted six years ago, in a 2015 report by the metro line’s investigative committee, which was set up by the Mexican Government after construction faults were discovered at 11 of the line's stations.
It is unclear whether the faults have been rectified since the 2015 report.
In the line's underground section, it was found that material used below the rail track was made up of gravel and sand instead of concrete. This can cause the tracks to become unstable.
According to the report, “hidden defects” were also detected in “various elements of the fixed installations”, such as the sleepers, while the ballast was “not optimal to absorb the stresses in tight curves”.



The report adds: “Atypical vibrations and vertical movements were detected in all curves with radii less than 300m.”
In addition, construction was poorly supervised and work began without a comprehensive final design. The report says there were “important flaws in the conception and planning of the project” and “poor, hasty and incomplete certification on the functionality and safety of the line”.
The investigation was ordered after Mexico City's government was forced to suspend service at 11 of the 20 stations on line 12 in March 2014 to repair structural defects. These were due to rail unevenness, high levels of vibration and wear and tear.
At the time of the repairs, the line had been operational for less than two years. It opened in October 2012, with the project's tunnel boring machine having begun work in May 2010.
Independent bridge consultant Simon Bourne told NCE that “sadly, throughout the world, the lack of construction supervision has become commonplace over the last 20 years”.



He added: “Generally, in most countries, detailed designs are always checked or reviewed carefully, although not everywhere. But construction supervision has effectively disappeared as a tool everywhere."
Following the bridge failures in Florida and Genoa, in 2019 Bourne published a paper in The Structural Engineer which highlighted this issue.
It says: “The role of site supervision has declined enormously over the last 20 years, as contractors operate quality assurance systems. However, this process of self-certification may leave projects exposed to unscrupulous practices, to hasten programme or reduce costs. This is certainly a major issue (and a factor in failures) in the developing world and there is growing evidence that it is also becoming a serious issue in developed countries.”
The paper adds that a balance needs to be found between “achieving best value while not compromising integrity or safety”, and calls for all bridges to undergo independent design checks and independent construction checks through a "competent site supervision process".
Such site supervision has become “increasingly rare worldwide”, the paper says, “even though its cost is small compared to that of the overall project”.
In the case of the Mexico City collapse, Bourne said that given the combination of poor construction materials and this global trend for a lack of construction supervision his “main suggestion” is that the collapse was due to the failure of the concrete top slab at midspan, “presumably due to poor quality materials and/or poor workmanship (resulting in poorly compacted concrete)”.
He added: “How it survived eight years is an issue, as there seems to have been only one passenger train on the deck at the time, whereas the span would have been designed for two, heavier goods or service trains.
"One can only assume that some final higher level of impact loading and/or some further deterioration of the poor concrete finally pushed it over the edge. Clearly, the issue of sub-standard materials and/or poor workmanship looks like an issue on this metro line.”
The 25km line has 20 stations - eight of them below ground - and provides service in seven Mexico City boroughs. The contractor on the scheme was Mexican firm Ingenieros Civiles Asociados (ICA).

Bourne's key points

  • Bridges generally fail short-term during construction (eg Florida) due to poor design, poor details or poor materials/workmanship - limited design checks and lack of proper independent construction supervision are usually contributory.
  • The second major cause of bridge failures is longer-term (eg Genoa), primarily due to poor maintenance regimes over many years.
  • Here, in the medium-term, poor design is probably less of an issue, but residual construction defects combined with the increasing effect of little, or no, maintenance could combine to cause a failure of this nature.
 

ac6000cw

Established Member
Joined
10 May 2014
Messages
3,432
Location
Cambridge, UK
Some quotes from the story on the BBC website:

Fernando Espino, secretary-general of the union representing metro workers, said the incident was related to a construction problem.
"This is a structural failure... We don't know if it's related to the material with which this [section] was built. We need to see what happened," he said, according to El Universal newspaper.

Elevated parts of the line, including the collapsed overpass, had to be closed for multiple repairs in 2014. Residents also reported cracks in the structure after an earthquake in 2017, and transport authorities made repairs following the reports.

In 2017, the then director of the city's metro, Jorge Gaviño, said the line "was born with endemic problems that would never be solved in its life" and that it would require "permanent" maintenance.

...Mmm
 

Wolfie

Established Member
Joined
17 Aug 2010
Messages
6,983
Hate to say this but corruption seems to figure large in many such issues in developing countries.
 
Last edited:

ac6000cw

Established Member
Joined
10 May 2014
Messages
3,432
Location
Cambridge, UK
Hate to say this but corruption seems to figure large in many such issues in developing countries.
Mexico might be officially a developing country but it's the 15th largest economy in the world - around $1.15 trillion GDP in 2018 - and has been one of the three NAFTA countries since 1994.

So (IMHO) it's got no excuses for poor construction and maintenance standards on an important part of the infrastructure in its largest city & urban area.
 
Status
Not open for further replies.

Top