Driver in Spain Rail Crash Reportedly Was On Phone With Renfe

Jul 30th, 2013 | Posted by

Reports from Spain indicate that the driver of the train that derailed in Santiago de Compostela last week was on the phone at the time of the crash. But before you equate this to the Chatsworth crash (as I did earlier today), read the details:

According to the investigation so far, train driver Francisco Jose Garzon Amo received a call from an official of national rail company Renfe on his work phone in the cabin, not his personal cellphone, to tell him what approach to take toward his final destination.

The Renfe employee on the telephone “appears to be a controller,” a person who organizes train traffic across the rail network, said a statement from a court in Santiago de Compostela, where the investigation is based.

“From the contents of the conversation and from the background noise it seems that the driver (was) consulting a plan or similar paper document.”

That’s a completely different situation than the 2008 Chatsworth crash, blamed in part on a driver who ignored a red light because he was texting on his personal cell phone. The driver in last week’s crash appears to have been doing his job, taking a call on the phone in the cabin from his bosses regarding the train’s route. Further investigation will be needed to determine whether he should have taken the call, whether Renfe should have made the call, what he was doing during the call, and what role if any it had in causing the derailment.

Many reports prior to today had been blaming the driver for this crash, and the headline of today’s story would appear to deepen his responsibility. But the truth may be more complex. A full investigation will shed light on exactly what happened and what lessons we should draw from it.

Tags: ,
  1. John Nachtigall
    Jul 30th, 2013 at 18:55
    #1

    He braked 2.5 miles late?

    It’s not the call, it’s the fact he lost situational awareness of where he was and where his train was

    Engineering Student Reply:

    Correct me if I’m wrong, but say the train was moving at the 50mph speed limit. That’d be a mere 3-second delay (2.5mi divided by 50mph).

    That’s even less considering the speeds observed – which would be 1.3 seconds.

    Alon Levy Reply:

    It’s minutes, not seconds.

    John Nachtigall Reply:

    Please tell me you’re still an engineering student. If so go to your professors and demands remedial course in math.

    It’s 50 miles per 60 min or 5 miles per 6 min or 2.5 miles per 3 min (or 180 seconds).

    Now start a timer and sit quietly not looking at the timer (no phone, radio, tv, etc.) for 3 min to see if you can estimate how long that is. I bet you, that you underestimate by at least 1 min. That is a long time.

    He messed up, but the system messed up more by relying only on people. People fail eventually, even 30 year veterans.

    Marc Reply:

    The train wasn’t at 50 MPH when he received the phone call, it was at 120 MPH, so he had more like 75 seconds.

    Y’all should try putting yourself in his shoes (I know, you’re all too smart to screw up like this, you should be the ones driving trains). You’re at the end of an hour or more uninterrupted run at 200 km/h, you receive a phone call from your boss who asks you to look up something in the paperwork. You’re passing through a rapid succession of tunnels, open straight stretches, overpasses, etc. Somewhere in there one or more signs tell you to slow down to 80 km/h, but you’re still looking for stuff on paper. Then the phone call ends, you look up, and your next thought is “Oh sh*t, where the f*ck am I?”…

    John Nachtigall Reply:

    We are all in his shoes every day we drive cars. There are many distractions and it is safe to say that many accidents occur because people lose situational awareness. Happens to pilots and boat captains also.

    Its understandable, its common, and unless there is other compelling information (like he did it all the time) I would classify this an accident not a criminal act.

    What is not acceptable, however, is the decision to allow this track to exist without automatic speed controls for the train for exactly the reason given above. The use conditions on the track changed and they (train management) made a conscious decision not to install common and proper controls. The shame is on the management, not the conductor

    Paul Dyson Reply:

    The conductor does not drive the train. Driver is best term to use IMHO, as engineer has other (higher) connotations. In Spanish it’s mechanista.

    Alon Levy Reply:

    In French it is confusingly conducteur. For years I thought that the RER had two employees per train because someone had mistranslated conducteur as conductor.

    Paul Dyson Reply:

    In French it is mechanicien!

    Clem Reply:

    sans ‘h’

    Paul Dyson Reply:

    And with an accent over the first e! Quite right.

    John Nachtigall Reply:

    Your right I meant driver

    jonathan Reply:

    Paul, Agreed. Only North Americans ever call a train-driver an “engineer” Train drivers drive trains. Engineers do engineering, which includes things like designing trains, and rail lines, and bridges.

    In French, “Chauffeur” means.. fireman, of a steam locomotive (think: “heater”!); or second crewman in a diesel or electric. The engine-driver is.. see above. Both nouns are high-school level material, for Anglophones learning La Franc,ais Fondamental (deuxie`me degre’).

    ericmarseille Reply:

    In French “chauffeur” (“heater”) is first understood as “car driver servicing his passengers” (taxi driver or private driver) ; a less known meaning is second crewman of a locomotive (esp. steam locomotive). Fireman in french is “pompier” (“pump serviceman”).

    Richard Mlynarik Reply:

    “Heater” in the sense of “somebody whose job is primarily keeping a seat warm” is a pretty good match over here.

    PS ““Engineers”

    Emmanuel Reply:

    I agree with Richard. Honestly that’s what his job should have been. All he should be able to do is lower the speed if necessary. High speed trains should never be programmed to go faster than what the track was designed for. Of course that would require that Renfe gets a programs that keeps data for designed track speeds. Nope, let’s rather take the risk and add a human being into the equation.

    isgota Reply:

    “In Spanish it’s mechanista.”

    “maquinista” how is exactly written.

    Best regards from Spain.

    John Nachtigall Reply:

    one other thing…that said he owed his first duty to those passengers and he knows it. He is a professional, he should have not allowed himself to get distracted. I understand how and why it happened (we are human) but he has to take some of the blame because he is a professional with 30 years of experience, he knew better.

    Clem Reply:

    The driver is just one link in the accident chain. There were a lot of factors that had to align just right for the accident to happen. One can feel bad for the victims of the wreck and also empathize with the driver, who is only human.

    One thing that will be extremely interesting is to see if Renfe or ADIF have records of previous close calls at this location. I imagine emergency braking events get logged in records somewhere.

    jonathan Reply:

    Dear Clem, just how long do you thin such emergency-braking events would be logged for?
    How long do you think, for example, that CBOSS would log them?

    I’m not being sarcastic, I’m genuinely curious what you think. Hanging onto data for long enough to look for patterns like those you suggest, requires.. .foresight and planning.

    (Is William listening?)

    Clem Reply:

    It’s not the signaling system that would log them, but whatever anomaly reporting system Renfe uses. I’m sure emergency stops are not taken lightly, and would have to be accounted for by the crew.

    jonathan Reply:

    Clem, I tell you three times: you know not whereof you speak.

    Have you ever operated a national IP backbone? (These days a regional, like the former BARRnet, would dwarf a national backbone like the one I built and operated.)

    Have you ever deployed MRTG? Or even a software continuous-integration tool like .. Tinderbox, or Buildbot, or CruiseControl, or Hudson?

    Disk space is finite. It’s always getting fulll. More disk space comes out of *capital budget*, and it;s hard, it’s very very VER hard, to persuade Management that there’s a ROI on keeping logs going back…. years. Decades.

    Admittedly, the payloads of systems like ZUB or ETCS are _small_. But managmenet alwasy looks at capital costs, and wants savings.

    Engineering is about quantitative trade-offs. (And absolute performance; sane people take that as given, though in the Bay Area Transport-Industrial Complex, it’s apparently at best an after-thought).

    I mean, really. Do you think the California DMV can do milepost-by-milepost breakdowns of speeding tickets over the last 10 years? Hmm?

    Maths, and _thinking_ apparently _are_ genuinely hard.

    Clem Reply:

    I was thinking of a simple paper form for the driver to report what happened and why. I am not sure what to make of IP backbones in this context, though I concede you must be an eminent expert in this field.

    Richard Mlynarik Reply:

    I have no answers to any of your questions, but I know this sort of data exists, and at much greater detail than shown in that image, and extending back more than a day. I’m pretty sure the answers to your questions are already known.

    Note that there is never any need to delete any old logs. Disks are free. Ask the NSA! And the quantity of data is trivial. (The Swiss have decades of data of pretty much every time a signal changes or a block is occupied.)

    Note also in our Special Local American Conditions with PTG and allied criminals porking out at the CBOSS scam trough, we can expect several thousand dollar per megabyte annual storage charges for CBOSS data, a custom proprietary commerically secret data format, and tens of million dollar ECOs to upgrade to double-sided, high-density 1.44 MB floppy disks.

    jonathan Reply:

    Dear Technically Incompetent,

    logs going back 24 hours are not at issue. The issue isl, logs going back weeks.. months… years.
    One has to actually plan in advance to have that; and (having done a similar exercise in a different domain), when you _start_, you don’t know what the “interesting” events will be, a year from now, or a decade from now.

    So, if you’re wlrld-class, you log all the “primary” data. So that you can, a decade later, data-mane what was (at the time of capture) irrelevant.

    As for the preposterous n onsense that “Disks are free”….. Is this the same Richard “Djugashvili” M.
    who assures us that PB=CHSRA=PJBW are totally incompetent and will screw anything up? Or is this another case of: the Swiss do it, therefore anyone who doesn’t, is corrupt and incompetent and should be put up against a wall and shot?

    The sad, the depressingly sad thing, is that I’m not stretching what Richard M. actually says. Not an iota.

    Max Wyss Reply:

    Such information would be logged in the vehicle data recorder. I don’t, however, know for how long the data is retained; it could be a year, it could be 5 years… depends on the legal base, or operator-specific information. With older speedometers, the continously recording devices paper strips have to be replaced at regular intervals, but again, I don’t know for how long they keep them.

    Emergency braking would be recorded. On the other hand, it would be quite a bit of work to extract such data, as, particularly with the paper based recorders, the geographical information is missing, which would have to be retrieved from the locomotive log (where the driver enters the personal data plus the diagram.

    In any case, it would not be the signalling system recording that information; the signalling system would keep records (maybe) for passing a red signal (or there could be manual records of such events, due to the regulations of the operator).

  2. Derek
    Jul 30th, 2013 at 19:47
    #2

    Shouldn’t a co-driver or navigator be the one to consult the “plan or similar paper document” so the driver can keep his attention focused on the road?

    Michael Reply:

    I assume you’re kidding about the co-driver or navigator. If not, please take a look at:

    http://www.youtube.com/watch?v=FOt4o9BoR-8

    where you will see the one person who operates a typical TGV. 4:46 in the YouTube clip shows the operator looking at his printout. Around 5:15, you can also hear the audible alert that lets the operator know things like reduction in permitted line speed.

    Derek Reply:

    He wasn’t looking at the printout. He didn’t even have his glasses on.

  3. Paul Dyson
    Jul 30th, 2013 at 20:50
    #3

    Extraordinary comments. How do people think a train works, does the driver have a steering wheel? Does he select his own route? So if he is about to be directed into a different platform track, so what? There are signals for diverging routes, and if the speed limit is different for that route it would be signed, and/or there should be approach control. I would hazard that there would be no need for such a call, unless it were an emergency situation, and nobody has claimed that. It IS similar to Chatsworth in that the ‘phone was a distraction from concentrating on the job to be done. As someone has previously posted, in and out of tunnels, signals flashing past, you have to be focused. We’ll see if those concerned disclose the true content of the conversation and its operational importance, or lack thereof.

    Ted Judah Reply:

    I’m more curious as to why this story refuses to die. I understand posting updates every couple of days, but four posts in a row is not reinforcing much of a meme.

    VBobier Reply:

    The Boss ought to be the one charged, since it appears the driver was doing His job and was called by His boss about the line, a bit too late though to keep the crash from happening.

    D. P. Lubic Reply:

    You think these responses are bad? Take a look at the responses on the oil question, and on Yonah Freemark’s opinion that HSR (from here, just a few days back) is still a good proposal for America:

    http://news.yahoo.com/saudi-arabia-fear-north-dakota-one-man-says-132802213.html

    http://www.cnn.com/2013/07/26/opinion/freemark-high-speed-trains/index.html

  4. Peter Baldo
    Jul 30th, 2013 at 22:13
    #4

    Does the high speed line end right before the curve? Or are there several miles of curvy track before the accident site? Google maps (Viaducto de Angrois, Santiago de Compostela, Spain) shows the high speed line under construction right before the curve. Is that section now finished?

    VBobier Reply:

    I’ve no idea what ends where, but the line in question is now back in service according to the TV News…

    thatbruce Reply:

    @Peter Baldo:

    The curve is the first one encountered after a long distance of newly constructed (finished in 2011) mostly straight track that alternates tunnels and viaducts. The mid-section of the miles and miles of straight track has ETCS on it, but the end sections have an older PTC system called ASFA.

    There were a number of contributing factors, but the short version is that design compromises in the handover between ETCS and ASFA meant that drivers had to manually ensure that the train’s speed did not exceed that recommended for the curve, AND that the driver had lost his situational awareness of where he was in part due to an official phone call from his employer. Look through the last weeks worth of posts from Clem and Richard for longer ones that give more technical details.

    Richard Mlynarik Reply:

    I’ve twice posted a link to a full cab-ride video (in the same direction, even, though on a different train type) and to a full line schematic showing every bridge, tunnel, signal, grade change, and signal system change. Do yourself a favour. Knock yourself out. Foam away!

    It turns out the train that crashed was running on tracks that exist.

    Perhaps you were thinking of this scenario?

    Not that any of this is remotely on-topic, blog owner postings notwithstanding.

    PS ETCS/ASFA handover had no effect in this particular tragic instance, again because the train wasn’t running under ETCS but solely ASFA, the latter system having no speed enforcement. It could have been a factor in some alternate crash, but it wasn’t here.)

    Max Wyss Reply:

    I just hook up my message here; it has not a strict connection to the previous one.

    FWIW, if the legacy system can make a differenciation between “warning” and “stop”, it is possible to implement a crude but simple and not that expensive speed check. By putting up a signal A, and a signal B at a given distance, where the magnets (or balises) are set up to transmit these two states, it is possible to set a timer which switches from “stop” to “warning” (or even simply off) after a given time, which corrsponds to the time for the train to brake down to the scheduled speed. It is crude but quite efficient. Such a set up has for example been built in at the northern end of the Simplon tunnel where a curve follows shortly after the tunnel (speed in tunnel: 140 km/h; speed for the curve: 80 km/h).

  5. D. P. Lubic
    Jul 31st, 2013 at 04:45
    #5
  6. trentbridge
    Jul 31st, 2013 at 07:08
    #6

    What happened to the CAHSR log? Is nothing happening in California regarding this project?

    Richard Mlynarik Reply:

    In today’s episode, heroic PB FiscalEngineer front-line forces selflessly battle the swarthy CEQANIMBY horde.

    Reedman Reply:

    CAHSR has a board meeting tomorrow (Aug 1). I’m sure there will be hot train news emanating from Sacramento as part of it.

    StevieB Reply:

    The California High-Speed Rail Authority’s board members, at their meeting in Sacramento, will formally begin their search for companies to survey and map parcel lines, utilities and easements. Those activities will set the stage for property appraisals and acquisition after the agency finalizes a route through the southern San Joaquin Valley for the statewide bullet-train system.

    Read more here: http://www.fresnobee.com/2013/07/30/3415558/rail-agency-to-weigh-contracts.html#storylink=cpy

    Not as exciting as a train wreck.

    John Nachtigall Reply:

    I dont think he wants to point out that they are going to have to move groundbreaking back again because they still have not bought the land for even the 1st section yet.

  7. jimsf
    Jul 31st, 2013 at 13:34
    #7

    I am excited to announce that I will finally experience a high speed train ( since cali is taking its sweet time to get ours built) I’m going to France for my honeymoon next year and plan several trips via TGV. At first I had planned a Paris Lyon Strasbourg triangle. But with visits to friends near Nevers, and a disdain for packing and unpacking and shlepping with lugagge, we will keep Paris as home, and use TGV in a hub and spoke. This will work because it is how France is designed anyway. And the amount of France available in 2.5hr or less makes for serveral easy day trips. Alons y!

    jimsf Reply:

    I am still far more worried about the plane crashing than any of the trains crashing

    Keith Saggers Reply:

    enjoy

    VBobier Reply:

    Enjoy!

    Max Wyss Reply:

    Enjoy!

    Alon Levy Reply:

    Have fun!

    jimsf Reply:

    In doing travel research I am amazed at how simple it will be to visit France without using a car, or even a bus or bus tour, how walkable the cities are. I intend to walk mainly in paris and avoid the metro as the distances are similar to familiar walks in SF. YOu can do a trip to Strasbourg, get whisked there in 2 hours, get dropped in the center, walk it, see the gingerbread buildings and xmas markets, and zip back to paris for late dinner. Its brilliant.

    why we didn’t build this in cali 30 years ago Ill never understand.Its a no brainer.

    Max Wyss Reply:

    The cities are mostly walkable, and you get around relatively easily. However, when you get to the countryside, things differ considerably, and in few places, the schedules are coordinated, which means you may see surprises. But if you are mainly using TGVs, you can be assured that there are car rentals at the stations…

    swing hanger Reply:

    As Max said. There seems to be a divide between France and the German-speaking portions of Europe with regards to transport policy, especially in regards to schedule coordination and comprehensive services in less populated areas.

    Max Wyss Reply:

    Well, France (outside of the Ile-de-France) is just a bit behind. The transfer of the regional services to the Régions is slowly showing results, and you find fixed-interval schedules in more and more places. But the coffers of the Régions are not unlimited, which means that it will take some more time to get a coordinated system. But there is slow progress; the bus schedule from my home town to the next bigger town with a station shows connections to/from trains…

    But then, the buses are really not well filled during the day, which makes the whole thing a bit difficult (also considering the cheap fares (€1 for short distances; €2 for longer ones, but no transfers).

    ericmarseille Reply:

    If you plan your honeymoon in late spring or summer, a short hop to Marseille is a must…Nothing beats telling your Paris-based friends who saw you last dinner, still all white, that you got your sunburn at Marseille.

    And if you make it, don’t miss the area between Borély and Pointe Rouge, the Marseille part that no media ever tells about, because it doesn’t fit the sinister reputation of the town, or try a picnic in the calanques, the best thing to do in my opinion being taking the highway to Saint Cyr (30 miles away through breathtaking scenery) and having a good beach day in a holiday atmosphere.

  8. jimsf
    Jul 31st, 2013 at 15:46
    #8

    Human error aside, shouldn’t there never be a situation where a transition between lines and signaling is allowed to take place at a high speed?

    Joey Reply:

    Define “high speed.” The line in question is currently limited to 200 km/h (125 mph). And how would it be any different if you had an 80 mph signal transition right before a 40 mph curve?

    Paul Dyson Reply:

    Train drivers have been coping with speed reductions from higher speed sections to curves since 1825 at least. That’s not to say that with modern technology we can’t do better, but just like with flying it can become dangerous to rely too much on the instruments.
    Train driving requires particular skills, especially concentration, plus a good memory. Unlike road vehicles, trains can move for long periods without any action on the part of the driver unless a warning device has to be cancelled. Even that can become “mechanical”. Focus is essential. Incoming calls to the cab must be limited to absolute emergencies unless there is a second driver.

    jimsf Reply:

    too high a speed for that section. There should not have been a way for the train to be going that fast into that curve.

    Max Wyss Reply:

    The changeover between the signalling system (and also in some cases the power) occurs at 140 to 160 km/h in several places at the SNCF. The changeover is essentially automatic, and there is a “stop dead” balise which stops the train if the signalling system for the HSL somehow did not get activated properly.

    Changeover between Signum/ZUB and ETCS L2 occurs on the SBB network at speeds between 120 and 160 km/h (Neubaustrecke, Lötschberg Base tunnel).

    Therefore, such changeovers are done routinely at such speed.

  9. Keith Saggers
    Aug 1st, 2013 at 07:44
    #9
  10. synonymouse
    Aug 1st, 2013 at 10:44
    #10

    CD.STREETS blog article refers to this blog in some detail:

    http://dc.streetsblog.org/2013/07/31/will-technology-save-us-from-another-train-crash-like-spains/

    I like one of the comments:

    “What “progress” are you referring to? The project still doesn’t have nearly enough money to get it built, and a Sacramento judge’s decision due next month may order the state to put the issue back on the ballot, since the project has changed so much since 2008 that it amounts to bait-and-switch.”

    HSR opponents continue to err on the side of optimism and some residual faith in the system. Big mistake. Rubberstamp house judge will highball JerryRail. On the other hand the collapse of Deserted Xprss reveals the achilles heel of the hsr juggernaut: private interests and private operation. The PB-CHSRA scheme is totally incompatible with a private operation like LV Rail. Government owned and operated DogLegRail will spawn a bloated payroll and bleed red ink. The State won’t be able to afford the larger subsidyand will be forced to divest it. Unlike BART and its greedy unions and management the DogLeg is not indispensable nor enjoys an artificial monopoly.

  11. Derek
    Aug 1st, 2013 at 13:05
    #11

    Calif. high-speed rail seeks to reassure on safety
    by Juliet Williams, Associated Press via sfgate.com, 2013-08-01

    …a train that crashed in Spain last week, killing 79 people, was not operating on a system like the one planned for California. Chairman Dan Richard says the segment of track where the accident occurred did not have automated train controls that would have overridden the driver and slowed down the train.

    synonymouse Reply:

    This means the orphan ARRA-IOS will have “automated trains controls”?

    Richard Mlynarik Reply:

    Calif. high-speed rail seeks to reassure on safety

    We have America’s Finest Transportation Planning Professionals on the case. Now sign this blank check. You’re welcome.

  12. synonymouse
    Aug 1st, 2013 at 13:54
    #12

    Gadgetbahn du jour:

    http://www.foxnews.com/tech/2013/08/01/magnetic-levitating-sky-trains-may-be-coming-to-city-near/?intcmp=obnetwork

    Go for it, PB.

    Joey Reply:

    Oooh, another PRT scheme. I wonder how that will turn out…

    Alon Levy Reply:

    Tel Aviv is studying a maglev hochbahn PRT since its subway-surface line is running over budget.

    D. P. Lubic Reply:

    Why do these PRT and maglev projects use such cheesy photo illustrations? Take a look at the shot used on the Fox piece; it shows what is supposed to be a “pod” in front of a building, the building looking like some large in-town structure–but in reality, it’s a modest apartment building, note the handrails and porches! That thing would be more along the scale of a model in the photo; maybe you could put a dog or a cat in it, certainly not a person!

    D. P. Lubic Reply:

    I have a weird brain. Someone says something, or I think of something, and then I think of something else. In this case, it is recalling Kim Pederson’s personal monorail at his house in Niles, Ca. Kim (masculine here, by the way) is a strong monorail proponent.

    http://monorails.org/tMspages/Niles.html

    Another fellow built a non-operating “model” monorail as a prop for puppies:

    http://www.monorails.org/tMspages/PuppyMover.html

    Good for fun . . .

    Reality Check Reply:

    Found a couple YouTube videos of this nut’s home-brew ride-on garden monorail:
    Niles Monorail – Mike Rowe (2002)
    Niles Monorail – Believe it!

  13. Keith Saggers
    Aug 1st, 2013 at 15:18
    #13

    PRT scheme?, America’s Finest Transportation Planning Professionals?, Gadgetbahn du jour?

Comments are closed.