Spanish track was fractured before high-speed train disaster, report finds | Hacker News Reader

Spanish track was fractured before high-speed train disaster, report finds

bbc.com

220 pointsRygian222 comments

While these events are statistically very rare, it is worth remembering that there have been two separate events in the past twenty years in Spain where high-speed trains have derailed leading to multiple fatalities [1][2]. In contrast, the Japanese Shinkansen has a spotless record since its introduction in the 1960s [3]. Not a single fatality due to a crash or derailment. And that's in a country with a much larger population and much higher passenger count per year.

What do they do differently?

[1] https://en.wikipedia.org/wiki/Santiago_de_Compostela_derailm...

[2] https://en.wikipedia.org/wiki/2026_Adamuz_train_derailments

[3] https://en.wikipedia.org/wiki/Shinkansen#Safety_record

Track maintenance?

Yep.

Which is the secret of preventing 99%+ of sudden mechanical failures of pretty much any type of infrastructure.

Higher passenger count could imply ability to pass higher maintenance budgets?

Could weather or some other geographic/similar aspect be a factor?

The geographic aspect of russian agents being in vincinity of the traintracks. Week before supply trains in Germany also derailed, as they do once per month.

I'm not saying it couldn't have been the Russians, but it would be strange for them to target Spain, since it's the only NATO country that doesn't want to increase its defence spending.

Spain basically does not do the required maintenance:

https://www.reuters.com/world/spains-deadly-rail-accidents-p...

From the linked article:

> [The] stretch of track that was renovated last May and inspected on January 7.

The track had been inspected very recently. Maybe the inspection standards are inadequate?

The linked article also shows figures that are quite meaningless without context.

> [The] vast majority [of Spain's high-speed rail budget] went to new infrastructure with only some 16% earmarked for maintenance, renewal and upgrades. That compares with between 34% to 39% spent by France, Germany and Italy,

They simply can't compare those numbers as-is. Of course Spain will be spending less in maintenance as a percentage of the total budget if it's still mainly building new tracks. It's not a useful figure.

Yep, plus their network is pretty new anyways. Which generally needs less maintenance than older infrastructure.

Just because something is new, doesn't mean it's full of faults.

> The track had been inspected very recently. Maybe the inspection standards are inadequate?

Spanish officials are very good at deflecting blame and playing politics. Nobody wants to be held accountable for a catastrophe. Also see the 2024 floods in Valencia; a partially preventable tragedy, followed by a whole lot of mud slinging, but zero accountability.

So while inspection standards might be inadequate, I would take anything a senior official says with a pound of salt.

But he is correct. If you have a large enough budget for new construction it can make any maintenance expenditure look tiny. The right figures to compare are normalized by length and age of track, not percentages of the total budget.

> 2024 floods in Valencia; a partially preventable tragedy, followed by a whole lot of mud slinging

sigh

Of course you're right

Specifically the fractured track was a soldered joint that joined a track from 1989 with a new one from a few weeks ago.

Soldered eh? No wonder then that it broke.

English is unusual in that we have both Germanic "weld" and Latinate "solder" and they've acquired different meanings. Spanish (and other Romance languages) use the term "solder" (soldado) for both.

As an aside: Chinese also uses the same term for both (焊接), and the standard English translation is "welding". This can lead to some confusion when Chinese manufacturers start talking about e.g. "surface-mount welding". :)

Heh, that would be a funny misunderstanding to have as well as the opposite, when you get back something soldered when you expected it to be welded.

Interesting. In dutch we use 'solderen' vs 'lassen', in German they use 'schweizen' and 'loten'.

English has a third term like that as well called 'brazing', then there is silver solder (a high temperature version of soldering), in dutch we'd call that 'hardsolderen', whereas what the English call brazing we call oxy-acetyleen lassen (which is more of a process name by virtue of naming the ingredients).

Soldadura autogeno and Soldadura en el arco (sp?) are what I think the modifiers used in Spanish to indicate brazing and (arc) welding.

Schweissen und löten. Has nothing to do with Switzerland (Schweiz) ;)

Ah yes, you are right! I was going by ear, rather than by the written version, in fact I can't recall seeing it written. German is a language that I will happily use but don't ask me to write a letter in it, you'll probably need exponential notation to represent the number of errors.

As a matter of fact schweissen is only correct spelling in Switzerland.

In Germany it would be schweißen.

Czech uses "Pájení" (derived from "joining") vs "Svařování" (derived from "boiling".

So, also different with different etymology in a language from a different group (although these things were probably influenced by German)

Yeah - the Czech wording is quite clever:

* the first one makes it clear a something (a different material) is used to join things together

* the second one implies you melt/boil the things to join them together

This was a track laid a few weeks ago? I think that's the problem.

> Spain spent an average of about 1.5 billion euros ($1.76 billion) a year from 2018 to 2022 on its high-speed network, more than any other country. However, the vast majority went to new infrastructure with only some 16% earmarked for maintenance, renewal and upgrades. That compares with between 34% to 39% spent by France, Germany and Italy, whose networks are far less extensive, according to the Commission data.

Conflating the maintenance budget with the money invested in new infrastructure in this way is not very useful IMHO. How much inspection/maintenance money was spent per km of (high-speed and overall) railway track would be much more informative...

I think even more important is the seismic activity in Japan asa risk factor here

Are you suggesting this leads to more inspections or better inspections or better build quality or what?

That despite seismic activity they managed to avoid a catastrophe like ones in Spain

> Santiago de Compostela derailment

Hey that infrastructure looks perfectly fine and new, ahhh ok... they were going 180kmh where the speed limit was 80kmh..

Which is also exactly how the most deadly rail incident in the past half century in japan happened.

https://en.wikipedia.org/wiki/Amagasaki_derailment

We oftentimes take ridiculous risks to save only 1-5 minutes of our time. Although reading about the Spanish disaster, the driver was rather reckless.

Which is a problem that would have been prevented had they not purposefully disabled the ERTMS signaling system to avoid delays.

I am not sure what conclusion can we draw from, as you said, two very rare incidents over a long period of time.

Reminds me of when Malaysian airlines crashed two planes in a short period of time. It was a good time to get cheap flights from Europe to south east Asia as long as you can withstand relatives thinking you are literally going to die in their third crash.

Bit of an odd comparison, given that one of those flights (MH17) was shot down by a Russian Buk squad. That was not an issue attributable to the carrier in any way, and after the incident the likelihood of it happening again to Malaysia Airlines specifically was negligible.

It could be prevented by simply not flying over an active war zone, something airlines do all the times to prevent the exact same thing from happening.

Airlines started being more sensitive to this after the 2014 crash

Or Girkin not ordering the civilian plane full of people to be shot down. It was a civilian plane at 10km altitude with a transponder on. Really doesn't look like a jet on a radar.

And up to that point Russia wasn't known to supply the separatists with an anti air system and the crew to run it.

It would seem the air defense systems used could not reliably determine what you imply they should [1][2]. I’m not sure where you’re coming from, or why it would matter what one country was known or not known to do.

[1]: https://www.technologyreview.com/2014/07/18/12951/how-can-a-...

[2]: https://www.washingtonpost.com/news/the-switch/wp/2014/07/18...

> why it would matter what one country was known or not known to do.

It absolutely matters.

Flying over a war zone with known anti aircraft missiles is quite different to flying over a low level conflict that is using small arms only.

There was no war zone at that time.

> Flying over a war zone with known anti aircraft missiles is quite different to flying over a low level conflict that is using small arms only.

What was in the news at the time, and the news are still linked from https://en.wikipedia.org/wiki/War_in_Donbas#Escalation_in_Ma...

2 June 2014: "Luhansk airstrike"

14 June 2014: "A Ukrainian Air Force Ilyushin Il-76MD was shot down"

20 June 2014: "The insurgents [...] shot down a Su-25 bomber."

14 July 2014: "Ukrainian Air Force launched air strikes targeting insurgent positions across Donetsk and Luhansk oblasts. The Ukrainian government said that 500 insurgents were killed"

17 July 2014: "DPR forces shot down a civilian passenger jet, Malaysia Airlines Flight 17"

Doesn’t look like a F14 either but a US warship, rather than some guys in a field, still managed to pull that off and send 290 people to their graves.

But it did look like an F-14. There really was an F-14, just on the ground at an Iranian airbase. And the Vincennes was under armed attack at the time--Iran let a civilian jetliner overfly their own attack. Plenty of blame for them, also.

> But it did look like an F-14

It absolutely did not. The RCS of an F-14 v/s an Airbus A300 is an order of magnitude different (probably 2 or 3 orders).

> There really was an F-14, just on the ground at an Iranian airbase

There was, but that’s a red herring for the root cause. Each ship’s radar independently and correctly identified and tracked the Airbus separate from the Mode 2 targets, but when communicating the track information between ships, the tracks were mixed up.

Source: The US Navy’s own account: https://www.history.navy.mil/content/history/nhhc/about-us/l...

> There was a combat camera team aboard the Vincennes, and the footage depicts considerable confusion and even ill-discipline amongst the crew (cheering, shouting, football game atmosphere) that contributed to one of the most tragic events in U.S. Navy history

The URL you linked to results in a 503 error (Service unavailable) and the Wayback Machine returns "Error code: 403 Forbidden" with "Looks like there’s a problem with this site", for all timestamps I tried, in 2025 or 2024.

I'm outside the US so that's probably the cause. Is such information available elsewhere?

I’m outside the US too and the link works for me

But this also works: https://archive.md/XsxT8

And also this: https://web.archive.org/web/20251208110440/https://www.histo...

The point is this is a fog of war situation. Mixing up who is who in combat is a very real issue where we have gotten better over time but have never truly solved.

I put the primary blame on Iran because they cleared a civilian plane to overfly combat they initiated. They set the situation up, a mistake happened.

Fog of war isn't like in a video game where it's just whether you can see something--in the real world by far the biggest factor is identifying what you see rather than simply seeing it.

That’s not the point, though.

Don’t fly a commercial passenger jet over an active known war zone. Then you don’t even really have to think about whether the separatists below you know whether your signature looks like a fighter jet or not lol.

Never leave your safety to the vagaries of Russian incompetence or malice, surely.

Russia is active war zone. Russians are flying commercial passenger jets over active war zone and then shooting them. Embraer E190 was the latest victim of Russians. Russia is the problem.

I never said Russia isn’t the problem. But intentionally flying your 777 over an active war zone is asinine regardless, is it not?

Yep, Russians does that intentionally, while Malaysian and other airlines are just following rules and air corridors.

I remember seeing video of the guys behind it seeing the wreckage and saying something like 'shit it was an airliner'. I think they shot thinking it was a military aircraft.

I'm not sure why Girkin would want to shot down an airliner?

And the other one was, as far as I remember, likely deliberate based on the pilot’s flight simulation data.

That one doesn't reflect well on the airline IMO. There should be systems in place to help employees cope with mental health issues so that they don't end up hijacking their own plane.

Also odd considering the other crash (MH370) was almost certainly a pilot suicide. Take my opinion with a grain of salt, but I’m not sure what they could have possibly done to prevent it.

Thankfully the aviation industry and related agencies are not so quick to dismiss human factors as unmitigateable.

imagine thinking the same way after the first crash, just

as long as you can withstand relatives thinking you are literally going to die in their second crash,

and then you die in their second crash.

The incidents were 4 months apart, so considering the number of flights the odds were still pretty good on that bet.

They are two very different accidents: The second was insufficient/poor maintenance: Supposedly the train that checks for this had passed 2 months before, and someone will have to wonder whether it's just not passing often enough, or if the inspections are just poor in general.

The first was purely a matter of not upgrading the signaling in a very low speed section: The crash could have happened with regional trains too. Every engineer knew that it was unsafe and one distraction was enough to get someone killed, but Spain is still well in the middle of track expansion, so it's all the horrors of politicking. Unless you have a crash, not upgrading those signals costs nothing, but, say, the very expensive connection to Asturias was worth a lot, so iffy tradeoffs were made.

Hopefully better engineering-driven tradeoffs are made regarding track maintenance, but hey, this is Spain, not a place where we are good at efficient, reliable safety processes: See the failures in Valencia for the DANA, where the chain between the meteorologists seeing a risk that led to recommending evacuation, and the actual order of evacuation was so slow, so we ended up with 229 deaths.

For comparison, the Japanese high speed rail track inspection trains run three times every month. A lot more frequent.

They run at full speed between regular train operations.

I saw one of them running on my last trip, which is said to be good luck.

Perhaps there are less FSB agents blowing up sections of track with shaped charges in Japan.

Source?

A component here is the highly unfortunate timing of two trains crossing one another as one of the trains derailed. Both trains look like rigid HSRs, and usually when these derails they stay very stable and rarely have fatalities.

Yeah. Japan really has better quality standards here overall.

Now - Japanese mentality is strange to me, but the quality standards and thought process, are convincing.

After reading Shogun, Cryptonomicon and watching plenty anime and documents about Japan (including Japanese rail system - still using the "pointing and naming" method I've learned from them) I would risk saying that Japanese do literally everything differently.

A list consisting entirely of fictional works (one by an American who has never lived in Japan even) is not a good basis for claiming to understand a culture.

Seriously, Cryptonomicon is a bizarre thing to put on this list. I like the it a lot, but none of that book takes place in Japan and the closest intersection is Japanese soldiers during World War II, with a brief participation of a single fictional Japanese company in the modern section of the book.

Well I have watched the show adaptation of Shogun, which features authentic Japanese language, and enjoy the occasional Omakase (in Brooklyn), so I’d say I’m pretty qualified to comment on Japanese rail over the past sixty years.

I've managed to draw the Japan flag in middle school one time. Add me to the list of reputable sources.

I’ve read the Wikipedia article about Japan and had a friend living there. Beat that!

I grew up playing all the Mario games and wrote a dissertation on an Internet forum, so now I have a PhD in both Japanese and Italian culture!

Regardless of Cryptonomicon's utility in understanding Japan, the statement that "none of that book takes place in Japan" is not true.

You seem to be ignoring the important point: "The Pointing and Naming System".

While this protocol is not oriented toward maintaining equipment like tracks and wheels -- it does seem to be a good indicator that the Japanese deal with these systems and the safety concerns around them differently.

And their track record (pun intended) shows the result of this focus.

https://en.wikipedia.org/wiki/Pointing_and_calling

> Railways in Japan use a safety system called “pointing and calling.” This method of physically pointing toward an item to be checked while vocalizing its name was invented in Japan about 100 years ago. The combination of looking, acting, speaking and hearing reduces errors by as much as about 85%.

There are probably better sources than those two. What's next, citations from Enoch Root?

Japanese people are just people. They have a unique culture... Like literally every other identifiable culture on earth.

I love Cryptonomicon but it engaged in that distinctly American brand of orientalism when it got into Japanese soldiers killing themselves and whatnot.

I don't watch anime or really follow anything specifically Japanese, but I read Shogun as a teenager and then decades later (lately) I read about the Mishima Incident which attempted to restore the Emperor to power in 1970. Quite frankly the way the article was written and the events that transpired were extremely reminiscent of Shogun. The latter was written in 1975 but I am skeptical how much non-Japanese information was available about it leading up to 1975 when Shogun was published, considering this Wikipedia article has an obviously rough translation. Just the way the people involved relate to each other is quite unexpected from a Western perspective.

My tentative conclusion is that there is something really unique about Japanese culture and there is certainly nothing wrong with that.

https://en.wikipedia.org/wiki/Mishima_Incident

Japan has a culture of perfection.

But every culture has its exceptions. 2 words: Tataka airbags.

> And that's in a country with a much larger population and much higher passenger count per year.

These are actually points making the Japanese system easier to maintain. Because of smaller surface area it’s much denser.

earthquakes, tho? Maybe the constant state of necessary vigilance has something to do with it here.

Different soil? Different climate/weather patterns.

Japan having to build to earthquake standards, so being more robust overall? Or to specific failure modes, at least.

Doctor Yellow.[1] Full rail inspection every ten days.

[1] https://en.wikipedia.org/wiki/Doctor_Yellow

At full speed!

Which matters, because Doctor Yellow inspection trains can be put into the schedule with the regular trains. There's no need to shut down traffic while a slow inspection car chugs along.

BART recently got a full-speed inspection car.[1] They needed a specially built one because BART has a non-standard track gauge.

The Federal Railroad Administration has track inspection cars, but only six of them for the whole country. One was seen on CALTRAIN track.[2]

[1] https://www.youtube.com/watch?v=GHa1Si7CW8Q

[2] https://www.youtube.com/watch?v=RQcRHn9YiPg

> What do they do differently?

Accountability.

My understanding is that Shinkansen that is high speed rail in Japan is grade separated system. That is tracks are only used by high speed rail. In Europe generally tracks are shared outside few specific links.

This means that Shinkansen tracks are designed and build to much higher standard.

In Spain the high speed network is separate from the traditional network too. There is some inter connectivity to allow for high speed trains to call at traditional stations, but the high speed network is for high speed trains only.

Yeah, the planned Czech high speed trains (VRT) have the same gauge but are expected to be used by the high speed trains almost exclusively, with a limited number of normal-speed passenger trains and AFAIK no cargo traffic at all.

Japanese rails are all built on commuter style architectures and the tracks are generally owned by its users. So train operators are strongly incentivized to keep them in good shape.

Also, Far East right now is also massively cash poor yet labor rich relative to the rest of the world. Everything is crazy undervalued and there are clear gaps between amounts of money changing hands vs work being done. Skilled-labor-intensive tasks are going to be much easier when cheap skilled labor is just perpetually available.

2 questions one rail related, one societal.

1) Can you expand on your first sentence? When you say user owned what does that mean exactly?

2) If skilled labor is undervalued does that mean those with those positions live kind of meager lives? Or what is that like?

1) There are type 1/2/3 train operators that owns everything/borrows rails/owns just rails as Japanese laws classify them, but type 3 lines are mostly rural low-traffic branch lines. Most high-traffic lines are owned(+accesses leased to type 2s) by train operators. Some of regional Shinkansen lines are technically owned by government run JRTT agency and leased to local JR company, which are probably as unrelated as UK government and its Royal Mail are.

2) I mean, like, it's the place where the English loanword for "death by overwork" came from. Also, undervalued means things costing less than they are worth. Trash costing little isn't undervalued, that's more adequately valued.

Thank you, haha on answer 2 !!

Answer 1 is still not clear to me. Can you contrast it with how Spanish track is managed?

Shinkansen is built like a giant subway. Fixed transponders everywhere, mission control monitoring and coordinating everything, stations right on the main line etc etc. They even use the same callouts as intra-city trains and they've long been at almost GoA 2 levels in train people terms?

I admit that I was a bit uninformed about specifics of Spanish train system in that, the rails were in fact laid by then-Spanish national rail and the operator was then-national company, but still, they don't seem like built and maintained like the BART or the NYC subway that happens to go 200mph in straight sections. That Shinkansen architecture is unique, and that is also guaranteed to be more labor intensive than how everything in most HSRs are.

Japan GDP PPP per capita is about USAs in 2012, so they aren't exactly impoverished.

High public competency and government capacity allows a lot to get done.

> While these events are statistically very rare

These events happening 4 times in 3 days are statistically nonexistent. Even less existent is them starting to happen right on the day before a major politician in Spain visits Israel to talk about buying Israeli security and monitoring systems.

I wonder how common it is for train tracks to fracture? And what systems are in place to actually detect this. There was recently a post on a German subreddit where the OP found a fracture in the German rail[0], albeit much smaller.

0. https://old.reddit.com/r/drehscheibe/comments/1qe9ko2/ich_gl...

In November, a bigger missing part of a train track was due to sabotage: https://www.bbc.com/news/articles/cp85g86x0zgo

It happened near Polish-Ukrainian border and officials were vocal about sabotage.

That’s pretty far from Spain

> I wonder how common it is for train tracks to fracture?

That entirely depends on which class of tracks we're talking about. And on top of that, remember that Europe is at war with Russia, railway sabotage has been attributed to Russia already in Poland [1] - and if you ask me, I don't believe for a single goddamn second that "cable thieves" were the cause behind the infamous 2022 attack on Germany's railways [2] either.

> And what systems are in place to actually detect this.

In Germany, dedicated railway cars called "RAILab" [3] that can measure track performance at up to 200 km/h perform the bulk of the work. In addition, each piece of infrastructure has something called an "Anlagenverantwortlicher", a person responsible for it - and that person has to walk each piece of infrastructure every two years at the very least, sections that have shown to be problematic get walked sometimes weekly.

[1] https://www.bbc.com/news/articles/c4gknv8nxlzo

[2] https://en.wikipedia.org/wiki/October_2022_German_railway_at...

[3] https://de.wikipedia.org/wiki/RAILab

Copper theft has been a recurring problem in multiple European countries for well over a decade. Railway outages caused by damaged cabling increased as the copper price rose, and decreased as police cracked down on scrap dealers accepting railway cabling without proper provenance. Damaged aluminum and fiber cables are also common, as thieves usually aren't exactly the smartest people.

The German Federal Police has nothing to gain by lying about this incident. They explicitly investigated the possibility of foreign sabotage, found the perpetrators, and concluded that it was just regular theft. Sometimes a horse is just a horse, even when there are zebras running around.

> The German Federal Police has nothing to gain by lying about this incident.

And yet, there are multiple different theories floating around on who bombed North Stream. The police and DA assume that Ukrainians were behind this mess, possibly under orders of back-then UA army chief Zaluzhnyi according to leaks and rumors, but official communication on that has been ... lacking and that's putting it mildly. It doesn't help that there are credible suspicions of Russia being behind it either, the only theory I'd move to the "conspiracy bin" is that it was a CIA operation.

When it comes to anything involving this war, there really is no reason to trust anyone.

> They explicitly investigated the possibility of foreign sabotage, found the perpetrators, and concluded that it was just regular theft. Sometimes a horse is just a horse, even when there are zebras running around.

I agree... but still, the timing is so incredibly close that it's just as possible that for once the cable thieves were capable of good OPSEC practices.

> I wonder how common it is for train tracks to fracture?

very

> And what systems are in place to actually detect this.

track circuit detection would pick up most cases I would have thought

Yes, but it would be shown as a false "block occupied" signal. Which could also have plenty of other causes, such as broken cabling or defective track circuit equipment, and as the Weyauwega disaster taught us it wouldn't detect a partial break.

Provided track circuit detection is even used, of course. I vaguely recall it not being compatible with either 25kV electrification or high-speed rail in general, and most modern tracks therefore using axle counters instead.

> I vaguely recall it not being compatible with either 25kV electrification or high-speed rail in general, and most modern tracks therefore using axle counters instead.

Track circuits aren't incompatible with that per se, but axle counters are simply easier to install and much more maintenance-friendly – no longer having to worry about

- mixing track circuit currents and traction return currents together

- having to keep the rails sufficiently isolated from the ground and each other to prevent the track circuits from falsely showing occupied

- insulated block joints

- having to use each bit of track once every twenty-four hours to prevent rust from falsely showing a track as clear

- extreme leaf fall and/or sanding potentially causing false clears, too

- length restrictions on the maximum length of a single track circuit, although that's only really a problem on more sparse trafficked lines with long block sections

In return, axle counters have the drawback that they

- don't detect broken rails (although it needs to be said that track circuits very much aren't perfect broken rail detectors, either)

- can be falsely reset (with more or less protections, depending on local operating practices)

- don't detect maintenance vehicles freshly placed upon a track until they enter the next axle counter section

but since most to almost all new installations seem to use axle counters, the trade-offs are apparently worth it to infrastructure operators.

Nice find. The gap in the Spanish track is massive. I don’t know enough to speculate on technical reasons but it seems quite odd.

Rails expand and contract according to the temperature (11mm per degree C per km). They are continuously welded together and installed under tension and heated to a neutral / median temperature for the location. It was around 0C that night in an area that gets up to 47C (and rails might get hotter under the sun) so there was at least 300mm of contraction per kilometre of rail.

The rail fractured into pieces during the derailment. You can see some of those pieces lying around in the photographs.

As the article notes: the initial break left marks on the wheels of several previous trains. The final gap is big enough that no train could possibly make it past it, so it is pretty clear that the gap got larger as the incident progressed.

Fractures could happen with ground shifting - perhaps recent flooding could have contributed

Rare, but not unheard of. See for example the Hatfield disaster in 2000, or the 2021 Ghotki rail crash.

Most railways regularly inspect their tracks to detect issues before they turn into a disaster. The big question here will be: why wasn't this caught earlier?

AFAIK continuously welded tracks (like those used in high speed rail) are also slightly tensioned, so a break in a single point could make it look like a whole section of track is missing, as tension is released.

CWT is laid in such a way that it has net zero stress in a "neutral" temperature, which naturally depends on the climate. Both extreme heat and extreme cold can cause damage, buckling and fracturing/embrittlement respectively, and choosing the neutral temperature is balancing act. But even if completely cut, track cannot shrink longitudinally much at all, it's the job of the sleepers and the ballast to keep it anchored in place. And if the track is laid on a concrete slab rather than ballast, it isn't moving anywhere.

Fun fact: the reason modern concrete or composite sleepers (e.g. [1]) have a slightly concave profile is to better resist lateral forces (i.e. buckling) than traditional straight-profile wooden sleepers.

[1] https://www.romicgroup.com/permanent-way/concrete-railway-sl...

Thanks for the clarification!

An article published in Saturday's edition of the Mexican newspaper La Jornada provides more details about the cause of the crash. The article is in Spanish; here are some of the main points, translated into English:

1. According to the CIAF, the break in the track was "practically undetectable." The fracture on the track was not noticed by the trains that passed over it, or by the technicians responsible for the maintenance of the infrastructure.

2. The damaged train, which belongs to the Italian company Iryo, is heavier than other trains running on the track; the additional weight of the Iryo train may have been a factor, or possibly even one of the causes, of the derailment.

3. The CIAF said that the notches registered in the wheels and the deformation in the rail are "compatible" with the fact that the track was broken before the Iryo train passed over it.

4. Spanish Transportation Minister Óscar Puente rejected criticism of the delay of the rescuers; according to the Minister, rescuers arrived within "18 minutes."

The full article is available here: https://www.jornada.com.mx/2026/01/24/mundo/020n3mun

“…not only did Iryo train's front carriages which stayed on the track have "notches" in their wheels, but three earlier trains that went over the track earlier did too.”

This sounds like something a camera mounted on a sample of trains watching a wheel could catch.

It would require a very high speed camera, and a floodlight, which may be impractical.

As a train wouldn't have the space??

Not necessarily, no. Train underframes can be quite crowded and this equipment is very industrial.

Japan, Germany, France, China, and the UK check their tracks at high speed. I don't know if Spain does, but there are news articles about them ordering such an inspection train in 2019: https://www.railjournal.com/infrastructure/adif-orders-high-...

https://de.wikipedia.org/wiki/RAILab

https://en.wikipedia.org/wiki/Doctor_Yellow

https://en.wikipedia.org/wiki/SNCF_TGV_Iris_320

https://en.wikipedia.org/wiki/China_railways_CIT_trains

https://en.wikipedia.org/wiki/New_Measurement_Train

The camera would probably only need to look at a very small section at high speed. They could be specifically made to film the tracks or the wheels of the train. Such cameras exist. Not cheap, but even some YouTubers have similar ones, to film high speed impact videos of things going much faster than trains. Might be worth it for trains.

More simply, you measure the impact for dangerous forces. No need to overcook it.

Too much processing. Accelerometer or even a microphone would do the job.

We actually have had 4 train accidents and incidents in a week.

https://people.com/train-collides-with-crane-arm-in-4th-rail...

It's clear some of them are probably caused by neglect in maintenance, others are freak accidents.

It's pretty crazy the statistical probabilities involved for something like this.

An Asturias Circanías train collided with debris from a collapsed tunnel wall on Thursday afternoon in Olloniego. No injured though

On Spain’s conventional and high-speed rail network, inspection frequency is defined by ADIF rules and EU railway safety standards.

High-speed lines (AVE): Visual and geometry inspections are performed daily to weekly using inspection trains and onboard measurement systems. Ultrasonic rail flaw detection is typically done every 1 to 3 months, depending on traffic and tonnage.

Source: ADIF high-speed maintenance programs and EU interoperability maintenance requirements.

Could electrical resistance be measured in train tracks to monitor sudden drops, such as fractures, before they cause loss of life?

it is possible, track signals can be triggered by shorting between two rails for example.

A new mode for fluke meters is born: the train conductor

What are the some of the ways that tracks are monitored for fractures like this? It must have been pretty substantial in order to be described as "complete lack of continuity". Makes me think of literally electronic continuity tests -- are those ever used in this context? Or how about cameras mounted on trains using image processing? Or drones?

It seems a shame that a few other trains passed beforehand with this anomaly in place and yet it went undetected.

Measurement trains filled with cameras and LIDAR

For example, in the U.K.:

https://en.wikipedia.org/wiki/New_Measurement_Train

LIDAR is good, but as another commenter pointed out, Ultrasonic Flaw Detection (USFD) is the gold standard for crack/flaw detection.

https://en.wikipedia.org/wiki/Doctor_Yellow

There are special trains with measurement equipment on board, but yes, it sounds to me like every train should be equipped with some basic sensors for anomaly detection.

The measurement trains drive slowly in the night.

Not necessarily, the measurement train my company develops can go up to 100 km/h and measure certain rail features every 5mm at that speed.

100 km/h is slow compared to passenger train (even non-high-speed ones). Depending on how packed the schedule is, it might not be possible to analyse track during the day without causing backups.

They can go at high speed:

Germany: https://de.wikipedia.org/wiki/RAILab

Japan: https://en.wikipedia.org/wiki/Doctor_Yellow

France: https://en.wikipedia.org/wiki/SNCF_TGV_Iris_320

China: https://en.wikipedia.org/wiki/China_railways_CIT_trains

UK: https://en.wikipedia.org/wiki/New_Measurement_Train

Indeed!

> Line inspection is carried out at full speed, up to 270 km/h or 168 mph on the Tōkaidō Shinkansen and 285 km/h or 177 mph on the Sanyō Shinkansen

Italy: https://decode39.com/3045/diamond-fs-diagnostics-train-luigi...

AFAIK, one technique for monitoring cracks uses ultrasonic sensors. They send sound waves through the rails and detect cracks by analyzing reflected waves.

TFA indicates a 40cm gap — huge!

I suppose that counts/was caused by a fracture but almost a half meter of gap in the track is nuts. Like describing a limb that’s totally removed as a bone fracture.

Though conceivably the break was very small and a train impacting the slightly lifted rail just caused a good chunk of it to explode.

> Though conceivably the break was very small and a train impacting the slightly lifted rail just caused a good chunk of it to explode.

The crown (top) of the rail seems to be missing after the gap. The crown-less section then continues ~3 meters before it disappears behind the investigator on the left. IDK what that might indicate.

ref pic: https://ichef.bbci.co.uk/news/1024/cpsprodpb/ecb4/live/53924...

The rail is laying on its side in that picture, so what is visible is the foot not the web.

edit: other angles of the same location here: https://youtu.be/DIQ4SrGSua0?t=1174

> The rail is laying on its side in that picture

Ah, I see it now. The marks from contact with the ties should have clued me in earlier.

Yes, the “fracture” (the problem was actually at a joint) was there for a while. The missing segment of rail was still there when the train arrived - the derailment affected only the last cars.

The "fracture" being referred to is a weld that somehow failed. The gap you are seeing is because an enormous, heavy train travelling at 200km/h hit that fracture and the rear half of the train derailed, tearing up sleepers and kicking all manner of debris around including ballast and, in this case, parts of newly-fractured (and therefore weakened) track.

No, that gap was created after the rail broke and the train derailed as a result.

The crack was in the weld, causing one side to sink and the wheel to hit the start of the next section of rail which was no longer welded to it, causing stress fractures to form in the rail which later caused that 40cm piece to break off.

Next to the weld, if we're being pedantic. The weld itself is stronger than regular rail, but the welding process weakens the rail right next to it.

Wheel Impact Load Detector.

It measures vertical forces in kips - (kilo-pounds-force, 1 KIP = 1,000 lbs)

They have these in the USA.

Those are for the opposite problem – detecting defective trains (overweight respectively otherwise faulty weight distribution as well as wheel flats).

You can look at the Wikipedia page on railway defect dectectors [0].

Under "rail break monitors" it mentions both electrical continuity and time-domain reflectometry can be used, and are most frequently used on high-speed tracks.

In addition, there are vast array of other detectors using acoustic sensors, strain gauges, accelerometers, cameras in the visible and infrared spectrum or laser measurement, that potentially could have detected an anomaly (i.e. damage to the wheels of other trains before the incident).

[0]: https://en.wikipedia.org/w/index.php?title=Defect_detector

Wheel impacts are the main way. But hardware can be bulky and trains can be surprisingly cramped.

We squeezed some track condition monitoring hardware into some locos but it was single-driver operations locations and we cannibalised some of the room that would have otherwise been occupied by the second driver.

~4 'derailing' accidents within 3 days, starting right before the president of the province of Madrid visits Israel to talk about buying Israeli security monitoring systems. Coincidence indeed.

Some more info from Spanish media. The track that broke was from 1989 and had not been maintained properly.

Got a link?

And how does it accord with the many statements made early on about the track being renewed recently?

I have one in Spanish. Seems the latest info is that it broke where the new rails meet the old rail.

https://www.elmundo.es/economia/2026/01/25/697635e8fc6c83c42...

Apparently the weld that broke joined an old segment with a new one installed last year as the tracks are renovated piecemeal.

Still the media in question, "El Mundo", is a mouthpiece for the opposition parties, seeking to create indignation against the government and scoring the head of the Transport Minister in particular.

They also want to make a parallel with the situation of the former President of the Valencian Community, from their party, who had to finally resign one year after being unreachable for hours on a date while hundreds of valencians drowned as his administration waffled aimlessly.

Of course the government is ultimately responsible for the state of the infrastructure, so the Minister well might have to resign after all is said and done, but the innuendo in that piece is pure politicking, not serious journalism.

No, the claim is that the broken rail was the new one but it happened at the transition from old to new.

Jupp you are right I had not read up on the news today.

Wow, that's a really big gap. No wonder it derailed

A stupid journalist, opposed to the current government, read a date in YY-MM-DD format as DD-MM-YY

My gut feeling says a lot of fatalities could have been prevented with a physical barrier between both tracks. Shouldn't this be mandatory with high speed trains?

I think the physics of the situation don't make a barrier feasible: a derailed train going >100 mph is going to transfer a lot of energy to any kind of barrier it impacts, which in turn might exacerbate the situation (by spreading debris).

I think these kinds of accidents are largely mitigated by rail defect monitoring. I know rails in the US are equipped with defect detectors for passing trains; I'm surprised that a similar system doesn't exist for the rails themselves. Or more likely, one does exist and the outcome of this tragedy will be a lesson about operational failures.

In principle only, if a barrier could keep a train on its side of the barrier, scraping along the barrier for a long distance instead of smashing headfirst into it, the energy could be dissipated over a long period of time, preventing fatalities. But what kind of barrier can withstand a train?

This collision happened precisely because of unfortunate circumstance that break in the rail and derailment happened just before the switch leading to the opposite track. Without the "help" of the switch, carriages of the first train likely wouldn't have invaded the second track.

The tracks are less than 3m from each other, a derailed car doesn’t need to get very far to be a risk to incoming traffic.

if they are already doing a poor job maintaining their tracks, what gives you such confidence that they would maintain the barrier properly?

the more you build the more maintenance costs rise.

I’d rather they spent the money ensuring no trains ever left their tracks rather than halving the destruction if they do.

There was a switchover which made the derailed cars of the first train move into the track of the second one, you can't have a wall there anyway.

The 20-ton bogie was flung 300m. What do you expect the weight of a whole car to do to such a wall?

More practical but still probably unnecessary is having the planned “passes” be where the tracks are separated by some distance.

But that requires the trains mostly always being on schedule.

You happened to have an opposing train at exactly the point where the train derailed.

That's simply really, really rare bad luck.

Practically anything you can think of is going to be a more effective use of safety resources than trying to contain a derailing high-speed train.

Thank about the change in airflow. The train would use more energy because of having to push air that is trapped by the barrier

Also the issues other comments described, including that any fault in the barrier means a new safety hazard

Pure economics. In Minneapolis the railroad demanded a crash wall to separate the light rail trains from their trains. It runs 1 mile and somehow cost nearly $100 million. This is a 5x increase from the original estimate but still $20 million for a 1 mile wall is a heck of a lot of money

Quite a tragedy.

Spain needs to rethink the way it operates trains. I think Switzerland handles this better, overall, though they probably also don't have as many fast trains because there are so many mountains. But I refer more to the intrinsic quality control and assumption made. If I recall correctly in Spain, there was the other train also coming in. I am sure they could have built the tracks differently. Granted, the issue here is cost, and an attempt to keep the cost down, but if you then accept disasters like that, it seems really awkward to me to want to save money here. And now that we know the track was already damaged, that just adds more validity to questioning whether the quality control systems were overall proper.

I mean maybe something of merit in that, but Spain has nearly 4000km of hitherto excellent and safe high speed rail and Switzerland around 200 km. Who should be giving lessons to whom? ;) Totally different scale of operations

Your comparison is nonsense and using nonsense metric.

Spain having built 20x more HSR than Switzerland in absolute terms, and much more HSR in terms proportional to country size, does actually does give them the right not to be lectured on HSR by a tiny country with a well-known superiority complex - especially when it's a cheap, incoherent shot soon after a tragedy.

I could make a cheap shot about fires in bars...but I won't.

You seem to have some serious anger issue unrelated to the fact that you tried to use distance as the only proxy metric for relevance of experience of rail operators.

Distance isn't even remotely relevant to complexity of running safe operations.

Data like fatalities per 1 million kilometers driven is significantly more relelvant - all metrics at which Swiss railways significantly exceed any European rail despite operating at significantly larger density of traffic.

Spanish railways have lately killed a horrifying amount of passengers - perhaps instead of insulting more competent operators, it's time to look at what keep going wrong there? Any responsible person would call for experts and advice here - e.g. we commonly see thet in air safety where foreign safety experts are invited to investigate and provide their own reports and recommendations.

What you seem to be doing here is something directly opposed to it and very very toxic - nationallistically trying to insults others and save face, which is very contraproductive for future safety.

> You seem to have some serious anger issue

Ad hominem so I'll discard most of your post. You lose when you try to insult my character.

> Spanish railways have lately killed a horrifying amount of passengers

Swiss bars have lately burned a horrifying amount of customers. Get your own house in order :)

Reply on news.ycombinator.com