At least 56 people have been killed and more than 70 injured after a passenger train derailed outside the northern Spanish city of Santiago de Compostela.

All 13 train carriages carrying 218 passengers plus crew left the tracks and four carriages overturned completely. Images from the scene showed crumpled metal and smoke billowing from the wreckage.

View the full story

The loss of life and impacts involved in this crash is frightening.

A few things not explained in most of the news stories:
a. Its a RENFE Class 730 passenger train, which is a the standard 130 made into a hybrid by adding a trailing car with a generator behind the loco.
b. That generator car is what caught fire, one photos shows the entire engine and generator has been torn out by the crash and thrown some distance.
c. One of the passenger cars was thrown over a wall onto a roadway.
d. It's a train with active gauge conversion wheel sets.

Looking at the pictures on news websites, it looks like the train was one of those TALGO style trains where there is only one axle at the back of each car and the other end is pivoted off the back of the one in front.

http://www.youtube.com/watch?v=3wUGT_k87e4

That is a terrifying video.


From my casual observance I always thought the wheel arangement of those sets would make them ride rather poorly. I've never been on one though. I assume it's a weight/maintainace saving with less wheelsets...

According the news reports, the speed limit is only 80kmh.

MelbourneSparks, without speculating, do a search for Talgo or Talgo Train or similar. The pictures MAY suggest it is a Talgo type train, I can't be sure though.

According to their site, their are High Speed Talgo's and Very High Speed Talgos.

There is a good explanation here:
http://www.euronews.com/2013/07/25/spain-rail-crash-track-lacked-high-speed-protection-system/

There is a good explanation here:
http://www.euronews.com/2013/07/25/spain-rail-crash-track-lacked-high-speed-protection-system/

Yes they are Talgos and I hope this example puts to rest any calls for lightweight aluminium passenger cars to be used in Australia. I know the ones used in the USA have been extensively modified to meet USA standards which are similar to ours. Other than that I am very sorry for the terrible loss of life experienced in Spain. My thoughts are with you. Also lets not prematurely execute the driver without the benefit of an enquiry and appropriate legal proceedings.

The month of July has so far been a disaster for rail around the world. A freight train in Canada carrying 72 wagons of chemicals derailed and exploded killing 47. Not long ago, a French passenger train derailed killing 6 and injuring more than 200, and now this disaster. My condolences go out to the victims of these disasters. Terrible tragedy for rail.

Yes they are Talgos and I hope this example puts to rest any calls for lightweight aluminium passenger cars to be used in Australia. I know the ones used in the USA have been extensively modified to meet USA standards which are similar to ours. Other than that I am very sorry for the terrible loss of life experienced in Spain. My thoughts are with you. Also lets not prematurely execute the driver without the benefit of an enquiry and appropriate legal proceedings.

I'm sure the "crash-survivability" aspects of the construction of the sets would be looked at by the investigations that result.  That said, if the train was doing the speed reported (~180 km per hour) there's not much that construction is going to do for you in that sort of incident.  Parts of the train hit the concrete wall at fairly high angles and either stopped abruptly or tumbled, meaning the forces on the occupants would have been pretty extreme.  A more rigid vehicle may not have made any real difference to the outcome (there's a limit to what a human body will tolerate) - in some respects it could make the outcome worse.  I suspect restraining people in their seats would be a more effective way of reducing casualties - there's no point having a robustly constructed derailed car come to rest mostly intact if its occupants are still travelling at relatively high speed inside it.

There are undoubtedly many differences - but a vehicle whose construction could be classed as "lightweight aluminium", with similar number of people aboard, had an oblique impact with a wall at the start of the month, at similar speed.  The casualty rate for occupants, who were all restrained in this case, was much lower.

Fundamentally - you want to stop the train coming off the tracks in the first place, particularly at speed.  Given reports so far I suspect the causes and potential solutions to that will be more of a focus of the investigations than crash survivability.

Yes they are Talgos and I hope this example puts to rest any calls for lightweight aluminium passenger cars to be used in Australia. I know the ones used in the USA have been extensively modified to meet USA standards which are similar to ours. Other than that I am very sorry for the terrible loss of life experienced in Spain. My thoughts are with you. Also lets not prematurely execute the driver without the benefit of an enquiry and appropriate legal proceedings.

How about not executing the design, designers and standard setters "without the benefit of an enquiry and appropriate legal proceedings"?

How about not executing the design, designers and standard setters "without the benefit of an enquiry and appropriate legal proceedings"?

Not executing anyone except for stating a fact that these passenger cars do not meet USA or Australian standards and that we have these standards for good reasons and should not be tempted to depart from them.

Not executing anyone except for stating a fact that these passenger cars do not meet USA or Australian standards and that we have these standards for good reasons and should not be tempted to depart from them.

I do wonder how well an Australian car would stand up to an accident like this, reported to be 100km/h or more above the speed limit on the curve.  We'll never now, of course, for to be 100km/h above the posted speed for a curve on an Aussie main line would be above the maximum speed they can attain.

I do wonder how well an Australian car would stand up to an accident like this, reported to be 100km/h or more above the speed limit on the curve.  We'll never now, of course, for to be 100km/h above the posted speed for a curve on an Aussie main line would be above the maximum speed they can attain.

when the CTT fell off the rails going nearly 50% over speed (like this one, the driver mentioned same when recovered from train, however he was never arrested nor charged), it fell onto mostly flat soft earth with only a minor embankment near the loco. Hence there were no deaths and the train was righted and rebuilt with no new stock required. In NSW a Tangara fell off the rails on a cutting going too fast and the train was mostly destroyed, difference, the rock wall embankment.

The German ICE derailed and speed and hit a bridge which then collapsed on part of the train, NSW experienced a similar issue in the 70's, the outcome looked pretty much the same.

While I know there are things they can do to help the train protect its passengers should something happen, the huge variability in types of accidents that can occur and the different structures, mass of train and structures should they get ontop of each other and terrains they can happen, along with the size of the train especially some of those very long TGV and ICE sets in with speeds exceeding 200km/hr and for some 300km/hr. I just don't think we can expect too much more than this when it all goes pear shaped.

Meanwhile Boeing still makes planes out of aluminium and now carbon fibre and they move a lot faster but at least make there people sit down and be strapped in.

This horrible incident probably puts to rest the automation argument for the NSW NWRL. Lets hope humanity gets a break from more of these incidents for a while.

Not executing anyone except for stating a fact that these passenger cars do not meet USA or Australian standards and that we have these standards for good reasons and should not be tempted to depart from them.

It just seemed a bit inconsistent to tell us all not to jump to conclusions by blaming the driver while in the same breath you seemed to blame the vehicle. Could just as easily blame that this line is reported as not having continuous speed monitoring / overspeed emergency stop whereas all true European HSR lines do.

It just seemed a bit inconsistent to tell us all not to jump to conclusions by blaming the driver while in the same breath you seemed to blame the vehicle. Could just as easily blame that this line is reported as not having continuous speed monitoring / overspeed emergency stop whereas all true European HSR lines do.

I think he was blaming the driver for the cause and the car design for the high death toll

How about not executing the design, designers and standard setters "without the benefit of an enquiry and appropriate legal proceedings"?

I have to agree.  Having an accident investigation process focused on achieving a successful prosecution is probably not going to result in a huge improvement in passenger safety.

Putting on my amateur accident investigator’s hat on, and having a quick look at the (probably) crash site:
- Google Earth sat images show the high speed line still under construction:
https://maps.google.com.au/maps?q=Santiago+de+Compostela,+Spain&hl=en&ie=UTF8&ll=42.858145,-8.524425&spn=0.012175,0.032015&sll=-32.010396,135.119128&sspn=103.185426,262.265625&t=h&hnear=Santiago+de+Compostela,+A+Coru%C3%B1a,+Galicia,+Spain&z=16

View Larger Map

There are some immediate points of note:

- The cessation of high speed track and commencement of the curve in question and "normal" running occurs immediately after leaving a tunnel.  A driver has no visual cues as to the pending perway changes other than that provided by lineside and/or in-cab signalling, or the driver's memory of the route.

- Some of the media reports linked to in previous posts suggest there is full ATP protection on the high speed section of the line, but that stops at the end of the high speed section.   Clearly the ATP system didn't effectively reduce the train speed to within acceptable tolerances prior to leaving the high speed section.  

I know this is pre-judging thing enormously, but it seems highly likely the driver did not receive any in-cab signalling cues to slow the train.  Safe operation was entirely reliant on drivers either responding to line side signals zipping past at 200kph, or memory of (and recognition of - tricky at 200kph inside a tunnel) the route to apply braking.

- The cessation of high speed track and commencement of the curve in question and "normal" running occurs immediately after leaving a tunnel.  A driver has no visual cues as to the pending perway changes other than that provided by lineside and/or in-cab signalling, or the driver's memory of the route.

Dare I say the end of the tunnel might make a memorable landmark?

Route knowledge is the most important part of train driving, knowledge of speeds, curves and gradients is the only way you can drive a train properly. It's not like driving a car!

I don't believe the driver would have not known about the speed or the curve, and having signs to say "80km/h zone 2000m ahead" would be of no value - it's just telling you stuff you already know.

The very first thing you learn about any new route is the low speed boards and any tricky gradients.

M

Dare I say the end of the tunnel might make a memorable landmark?

Route knowledge is the most important part of train driving, knowledge of speeds, curves and gradients is the only way you can drive a train properly. It's not like driving a car!

I don't believe the driver would have not known about the speed or the curve, and having signs to say "80km/h zone 2000m ahead" would be of no value - it's just telling you stuff you already know.

Independent of this accident - a system that relies on route knowledge alone to provide the necessary advance warning is deficient in this day and age (well, for any system where a failure in route knowledge could have serious consequences).  There are enough examples of where reliance on route knowledge alone has caused problems.

Dare I say the end of the tunnel might make a memorable landmark?

Route knowledge is the most important part of train driving, knowledge of speeds, curves and gradients is the only way you can drive a train properly. It's not like driving a car!

I don't believe the driver would have not known about the speed or the curve, and having signs to say "80km/h zone 2000m ahead" would be of no value - it's just telling you stuff you already know.

The very first thing you learn about any new route is the low speed boards and any tricky gradients.

M

For the QR CTT incident the bame headed towards QR for not having ATP and advance warnings of speed reduction. QR's offset was to have a 2nd driver which is failure mode from the start. The CTT derailed doing just under 100 on a 60 curve, hardly high speed. The same could happen to any freighter or commuter service in many parts of Australia where there is a tight curve at the end of straight.

So in both cases I don't buy the lack of prewarding excuse and the drivers should ultimately be held accountable for their actions just like a bus driver would. However the cost for those extra signs is minimal and when there is a reduction of speed of more than say 40km/hr, perhaps its not a bad idea if it helps save the odd life and millions of dollars in costs.

Dare I say the end of the tunnel might make a memorable landmark?

Except there were only 10 other similar tunnels in the preceding 15km.

In this case it would seem safe operation requires the driver to actively do something in response to little or no visual cues.  I don't want to downplay the significance of route knowledge, and indeed that may not have been deficient in this case.  The question is what prompts are there - if any - for the driver to take the requisite action to  avoid killing half the people on the train.

It has been reported that the driver was driving at twice the speed limit posted for that section.
No wonder the tracks were veering left (in relation to the train direction) and the train wanted to go straight ahead.

So in both cases I don't buy the lack of prewarding excuse and the drivers should ultimately be held accountable for their actions just like a bus driver would. However the cost for those extra signs is minimal and when there is a reduction of speed of more than say 40km/hr, perhaps its not a bad idea if it helps save the odd life and millions of dollars in costs.

Just speculating, but seeing and obeying lineside boards at "normal" speeds is not an unreasonable demand.  But standard signalling and lineside boards are useless for HSR because the trains are travelling too fast for any normal person to see them, much less respond to them.  This is why in-cab signalling was developed for the TGV and used on all/most HSR systems (though this could be argues to be MSR I suppose).

It's a shame (AFAIK) there isn't a European wide ATSB/NTSB equivalent that publishes it's findings in English.

Independent of this accident - a system that relies on route knowledge alone to provide the necessary advance warning is deficient in this day and age (well, for any system where a failure in route knowledge could have serious consequences).  There are enough examples of where reliance on route knowledge alone has caused problems.

Trains are not like cars. If you think you can drive a train under moderately extreme condition with no route knowledge then you have no idea what you're talking about! (Sorry, can't sugar coat that!)

Any rail system with extreme conditions (whether it be high speed or steep grades) requires route knowledge. With it, nothing else is required, without it, nothing else is sufficient. I suppose that when the French got the TGV to 575km/h they had an alarm that gave a point of no return to crashing thru the buffer stops with an emergency appication? I don't think so!

If you want to drive a train over a demanding route with no route knowledge then you will kill people.

M

Except there were only 10 other similar tunnels in the preceding 15km.

In this case it would seem safe operation requires the driver to actively do something in response to little or no visual cues.  I don't want to downplay the significance of route knowledge, and indeed that may not have been deficient in this case.  The question is what prompts are there - if any - for the driver to take the requisite action to  avoid killing half the people on the train.

I drive trains through ten tunnels almost every day, if you can't tell them from each other then you need to be taken off the road.

M

Trains are not like cars. If you think you can drive a train under moderately extreme condition with no route knowledge then you have no idea what you're talking about! (Sorry, can't sugar coat that!)

Any rail system with extreme conditions (whether it be high speed or steep grades) requires route knowledge. With it, nothing else is required, without it, nothing else is sufficient. I suppose that when the French got the TGV to 575km/h they had an alarm that gave a point of no return to crashing thru the buffer stops with an emergency appication? I don't think so!

If you want to drive a train over a demanding route with no route knowledge then you will kill people.

M

Ok!

To sugar coat it a bit (I didn't write the reply harshly, it is factual) then a good example of route knowledge would be knowing every crossover on your journey through Sydney. Every set of points either has a limit of 25km/h or has a marked speed limit of a higher or lower speed. Before you can qualify to drive through Sydney over a particular route, you have to know the location of every single one of them, their speed limit, and the signal indications you require to take those points. No warnings about points of no return (there are plenty), no warning on the speed limit, but a certain knowledge of every single one. That doesn't just apply to drivers of high performance and good braking electric trains, it also applies to drivers of lumbering freights.

You can't competently control a massive steel train, coal train or container train without that knowledge! These trains get either two or three signal aspects as advice of a turnout.

Another example is the length of time to train a Cityrail driver, about a year. It's not as if they have to steer! Learning how to accelerate and brake doesn't take longer than to drive a car, learning the speed boards and points is best done on the job, learning the theory of red and green signals and electric power with air brakes is not that hard. The time is all spent learning the routes and their requirements.

None of those examples are extreme railroading, they are the normal day to day railway work that so many do. It would not be possible to have a partial high/partial low speed railway system where route knowledge wasn't paramount.

In the old days with steam engines, three things were important: Loco control, train handling and route knowledge. Then it's simpler with a diesel: loco control is really simple, but train handling and route knowledge are still paramount. of course, those high speed EMU trains are coupled tight, so train handling is really simple, leaving only route knowledge as the paramount requirement. It's much simpler now than it has ever been before, so really any warning of a low speed ahead should be totally unnecessary.

If you would like more explicit examples of route knowledge requirements, then ask. But also ask yourself how many sets of points you cross on a trip through Sydney on an electric train!

M