Dramatic Video of Spain Train Crash

[fecrails]

http://www.youtube.com/watch?v=o32nxuHshv4

[imnew]

Amazing footage. I wonder why the onboard security system allowed him to run that fast It should have "denied" him to run at that speed on that segment of the line. Tragic event.

[mrennie]

Amazing footage. I wonder why the onboard security system allowed him to run that fast It should have "denied" him to run at that speed on that segment of the line. Tragic event.

There's an explanation of this in El Pais today. The General Secretary of the driver's union seems to be pointing his finger at the decision not to equip the final section of 4kms leading into the station, with ERTMS - the truly high-speed part of the line is equipped with Level 1 ERTMS which would have stopped the train. The last 4kms have a different system called ASFA which also has an automatic braking system, but the differences are that it doesn't do a continuous monitoring of the speed, but rather a "spot check" when passing fixed monitoring positions, and applies the brakes automatically only if the train is going at over 200km/h. Below that speed, there are just the visual and audible warnings to the driver. ADIF, responsible for the track and infrastructure (RENFE runs the trains), say that ASFA was sufficient for those 4kms because it's the entrance to a station and urban area and trains wouldn't be expected to be going fast there anyway. There's also a comment from the union that in a matter of seconds, the driver has to brake hard to reduce speed from 250km/h on the high-speed section, to 80km/h for the urban section, which uses conventional (non-high speed) track and has that very sharp curve.

it sounds to me like this was an accident waiting to happen, but we'll have to wait for the results of the investigation. Whatever the cause, it's a tragedy, like all such accidents.

[imnew]

Thanks mrennie

[Bananarama]

There was also mention that Spain pinched a penny when finishing off the line. Instead of re-grading to accommodate higher speed, they opted to keep the original ROW intact, replacing track and increasing super-elevation only. Most will point the finger at the crew, but as it was mentioned above, this was an accident waiting to happen.

[philmoberg]

There was also mention that Spain pinched a penny when finishing off the line. Instead of re-grading to accommodate higher speed, they opted to keep the original ROW intact, replacing track and increasing super-elevation only. ... , but as it was mentioned above, this was an accident waiting to happen.

I've spent the better part of the day thinking about this before responding; and having thought about it a while, I'd like to offer a few comments.

While I take your point, the wreck site was, after all, 4Km (2.5 miles) outside the station. Most of the high speed railroads I'm familiar with would have begun to slow their trains substantially by that time, if for nothing else than for the sake of passenger comfort. The French TGV system was a pioneer in tying high speed lines into conventional infrastructure - originally, to avoid the excessive cost of running new rights of way through heavily developed areas, which financially crippled the early Shinkansen program, for example - and they've handled multiple instances of this with an impeccable safety record. A good many civil engineers I know would comment that substantial infrastructure improvement - particularly realignment - might be uneconomical because the rail movements would be (or at least should be) operating at a diminishing rate of speed in order to make the station stop. Whether the officials in charge of the (re)construction of the junction with the high speed line in fact ignored the judgement of professional engineers and opted to forgo additional improvements to the junction because of the cost is a very crucial question - to my mind, at least - and may have a substantial bearing on the culpability for this incident.

Something as simple as stepping down the speed limit over the course of 3Km or so, such as Amtrak does on the NEC, would have been a simple and easily-implemented response to the situation, given that the rail union is quoted as having seen the danger and having complained about it repeatedly. The old New Haven guys that taught me the business, over four decades ago, probably would have regarded such as step as unnecessary, since they expected you to know the railroad well enough to know where the speed restricted curves were, and to begin slowing down for them a safe distance out (among a broad range of other things) before you was ever considered qualified for promotion to engineer, or, for that matter, to operate on a new division, regardless of your seniority. Of course, these were old school railroad men teaching old school railroading, and we live in a different age (NOTE: there is no sarcasm intended here): the perception of responsibility and liability have changed substantially since then, as has the way incidents such as these are reported and discussed. That said, the very existence of a security camera within the right of way at that particular location suggests that all parties were aware of a potential problem of overspeed operation, and that all parties agreed there was a need to monitor compliance with established procedures.

Another point that has been discussed very little, and only in a very small audience, is the question of equipment failure. These are machines, and machines can break. We are fortunate to live in a time in which our machines, for the most part, are far more reliable than they ever have been. However, any failure of the lineside signal system and/or any of the control and safety systems in the power car could have initiated the cascading series of failures so horrifically in evidence in the video. The more sophisticated a system gets, the more it becomes vulnerable to what might be an otherwise insignificant component failure. This why most commercial aircraft systems - particularly the most crucial of them - tend to be triple redundant these days. Yes, it is possible that this was operator error; but it is also possible that the operator was rendered as helpless and terrified as the passengers because of the extremely unlikely failure of a simple component.

It is - or at least should be - safe to say that it is too early to draw conclusions about what actually happened because there is to much tragedy and to little hard evidence. It's tempting to chalk it up to operator error because that tends to figure heavily in wrecks that aren't caused by track faults or other external causes, but as I recently read on another blog, "... reality is often more interesting and more complex than such stereotypes." It is at least possible (IMHO probable) that a number of things contributed to this. I'm sure it will be a while before we see or here anything definitive.

[mrennie]

Just a few observations on the last post.

Renfe said the train had passed a technical inspection that very morning.

Passengers had complained about the drastic speed reduction (from 250km/h to 80km/h) in the past, saying they thought it had to be a very difficult thing for the driver to do. I haven't read anything that explains why the trains don't slow down over a longer time and distance. Of course, if that was the procedure, you'd expect that it was because it was considered safe by those responsible for defining it and carrying it out in practice.

It was mentioned in the same article that the old line was kept for the 4km approach into the urban area because a) trains wouldn't be going at high-speed on the approach and b) to avoid the high cost of expropriations to clear the way for a new line.

[1225fan5358]

The engineer has been arrested. He posted pictures on his Facebook page of his speedometer when speeding on an earlier train. The press says (get your popcorn now)

The engineer has been arrested. He is known to have a love of speed. One man, an American inboard recalled swing a monitor showing the train's speed around 30 seconds before the crash. The passenger said it was way to fast.

[philmoberg]

Just a few observations on the last post. ... .

A technical inspection that morning would suggest that, while still not impossible, and onboard technical failure would be far less likely. I'll be interested to hear what they say about the signal infrastructure.

Given that the train had to shed about two thirds of its maximum allowable velocity, the engineer should have begun braking about 4Km (2.5 miles) in advance of the lower speed limit. I don't wonder that the passengers were uncomfortable: 0.25G has long been the industry standard for the maximum change of velocity while maintaining passenger comfort, and this clearly was well in excess of that, even if he'd managed to stay on the rails. I can't argue with your conclusion about the procedure - if in fact that was the specified procedure - but I am forced to wonder what they were thinking, given the precedent in the rest of the EU, not to mention Japan, which has been in the business longer than all of them.

Your third point would tend to confirm that RENFE had learned the same lesson form the Japanese that the rest of the HSR operators in the EU have learned; only something doesn't quite add up given the tragic outcome. I would still be curious to know whether the design of the junction was somehow compromised - it could be as simple as an unanticipated consequence of an unanticipated situation - or whether this incident was at least largely due to negligent or reckless operation.

The engineer has been arrested. He posted pictures on his Facebook page of his speedometer when speeding on an earlier train. ...

The engineer seems to have given them plenty of probable cause for an arrest warrant, at least under the rules in force in Common Law countries. I'm aware that Spanish jurisprudence has its differences from Common Law, but I'm not familiar enough with them to comment on them. The fact that he would have posted something like this on social media leaves me shaking my head. I'll say no more about it than that, partly because I'm in complete disbelief that somebody with this sort of perspective (or lack of it) would ever qualified to operate railroad equipment, and partly because I'd be tempted to use language I've committed not to use in this forum. Suffice it to say that perhaps he should have been a ditch digger, instead, because he's dug himself quite a hole in very short order.

[Chacal]

I think there is nothing in the press that says the speed he posted on Facebook was over the limit.

[GSkid]

I think there is nothing in the press that says the speed he posted on Facebook was over the limit.

CORRECT! CBS news reported that in the Facebook posting he claimed that if he went ANY FASTER he would be fined. That says he was at the maximum speed allowed when he took that picture.

The cars went off the rails first because the locomotive is so much heavier and resists derailing more. In the video it looks like the locomotive might have stayed on the tracks if it weren't for the derailing cars behind it giving it a big LATERAL tug on it's rear coupler.... and causing it's real wheels to come off and of course the rest followed. Look closely. I honestly think had that just been the loco going around the curve at that speed, I think it would have made it.

[philmoberg]

... In the video it looks like the locomotive might have stayed on the tracks if it weren't for the derailing cars behind it giving it a big LATERAL tug on it's rear coupler... .

It appears that, if he had been going a little slower - and I don't think it would have taken much - the lead power car might indeed have stayed on the track. I stopped the video at 0.27, at which point the forwardmost cars had already broken free of the lead power car and were going their own way. That would suggest the coupling was broken by then - perhaps a second or two earlier - and a cascading series of failures started. At that point, two things happened to to forward power car: 1.) it was relieved of the braking force of the train behind it, and 2.) the lateral component of the of the mass of the train behind it (there was a lateral component because the train was on the curve) was no longer counteracting the centripetal reaction of the power car to lean toward the outside of the curve. What appears to have happened at that point, given that the larger proportion of the mass of the power car is toward the rear, is that the outside wheels of the rear truck climbed the rail. The power car didn't start to roll over until after this happened.

This leaves me wondering about the extent to which the outside rail on this track was curve worn. I recently dug up an an accident report for a friend of mine who was on troop train that derailed back in the early'-50s: the train went in the ditch precisely because the outside rail was so badly curve worn that the flanges of the locomotive and the first few cars climbed it with ease, even though they were well below the limit for this speed restricted curve. The margins for error are smaller for HSR; and while not an excuse for speed that was clearly excessive for track conditions, it may have materially contributed to the tragedy. It may also suggest that - apart from any flaws in the metallurgy - it was at least a reasonably common practice to take this curve at a faster speed than would be prudent, thus causing unreasonable wear on the outside rail. That might explain the why the security camera was where it was.

As to the fact checking on the Facebook photo, the very notion that he thought he had the time to take his eyes off his job and photograph the speedometer when the train was at its maximum permissible speed is a demonstration of irresponsibility sufficient to be held on suspicion in this wreck. Personally, I can't reconcile this with the professional standards that were expected of me and my colleagues; but as I noted earlier, perhaps that was because it was a different age.

[Bananarama]

This leaves me wondering about the extent to which the outside rail on this track was curve worn.

Probably little, as the new express tracks were only completed within the last year or so.

[philmoberg]

Apropos of this discussion, I cam across an interesting article in from an unexpected source:

http://www.theatlanticcities.com/commut ... iver/6337/

The last sentence pretty well explains why the authorities arrested the engineer.

[hertsbob]

http://www.theguardian.com/world/2013/jul/30/spanish-train-crash-black-boxes

In an official statement, the court handling the case said that "minutes before the derailment, [Garzón] received a call on his professional telephone to signal to him the route he had to take on arriving in Ferrol. It appears, from the content of the conversation and the background noise, that the driver consulted a plan or some similar paper document."