Larissa, Greece: Freight train collides with passenger train (01/03/2023)

[MarkyT]

This is quite an old article from 2006, detailing a project to install ETCS level 1 across the Greek network. Some background is given about the existing signalling

ERTMS implementation in Greece

Though the Greek Railway Network is located far off Central Europe, ERGOSE S.A.1 is speeding up the efforts to comply with the interoperability requirements set by the European Community to facilitate unhindered, at least by technical barriers, cross-border operations at a European level and to...

www.globalrailwayreview.com

The current signalling systems on the sections of line which comprise part of the new project (i.e. red highlighted section) are delivered by two suppliers; Bombardier (former ADtranz, ABB) and Alstom (former SASIB).

In the Greek Railway Network, both relay and computer-based interlocking systems exist. Most of the relay interlocking systems are supplied by Bombardier and are in compliance with British relay interlocking standards. On the Thessaloniki-Idomeni line, relay interlocking systems are supplied by SASIB.

Computer-based Interlocking systems are of the type ASCV (Alstom) and exist on the upgraded line sections Platy-Thessaloniki and Thessaloniki-Promahonas.

On the open line sections between Larissa and Platy, as well as within the TX1 area in Thessaloniki, blocks with 3-aspect block signals for capacity reasons exist. All other open line sections are not equipped with block signals.

Two basic track occupancy detection systems are used in the Greek Railway Network, Axle Counters on the line section Athens-Oinoi and Track Circuits on all other line sections (Onoi-Thessaloniki-Idomeni/Promahonas).

Currently, railway traffic in Greece is controlled by five Traffic Control Centres (TCCs). These are located in the railway stations of Korinthos, Athens, Lianokladi, Larissa and Thessaloniki. The TCC of Athens will soon be moved to the SKA railway complex area after the completion and opening of it.

No national implementation of an Automatic Train Protection System currently exists, therefore, no legacy system burdens the ETCS deployment.

 

[SmokeAndJoe]

Absolutely, but I'm a mainline driver and have had a signaller say to me before "it's likely I'll call you back and you'll be passing it at danger, but we need to run some checks first"

I'd assume those checks are for things like conflicting routes, train in section etc.

In my mind, declaring the signal failed should be the bottom thing on the list once all other reasons for it not clearing are exhausted.

It absolutely is usually the last resort.

I say usually because that depends on the temperament/experience/knowledge/capability etc of the signaller.

Frankly this rather parcohial view of "hobby signalmen" only shows you don't really know what you are talking about.

I am a signalman on a major heritage line and we have the following procedures:

• All signalmen are formally re-examined on their rules knowledge every 2 years
• All signalman must work at least 10 turns per year, if not they have to re-do their practical exam
• All signalmen will be observed every year in each signalbox they work by an examiner
• There are additional requirements for more extensive recent experience in a box that must be satisifed before a signalman can work a non-standard timetable.

A couple of times I was able to observe the large box at Kidderminster SVR with 55 working levers controlling two platforms with loops, a short section of double track and a large yard, on a busy day the signalman there has to combine continous shunting moves with running the train service and has a job vastly more complex that plenty of Network Rail signalmen. Even working at a smaller box I regularly have 3 seperate signalled moves underway simultaneously. So... don't trust me I guess.

I agree I have no idea about the requirements for a signaller on a heritage line.

This isn't meant to offend, but no, I wouldn't trust you. Not for the reasons you probably think though.

What I do know is the mindset required, and that's to be inherently distrusting of what anyone tells you, including other fully competent signallers.

You're the last line of defence against danger as a signaller, you don't have the luxury of taking people at their word. People don't truly learn this until they've had their fingers burned a few times though.

Always take your time, always check, recheck and get someone else if possible to check that.

 

[68000]

This is quite an old article from 2006, detailing a project to install ETCS level 1 across the Greek network. Some background is given about the existing signalling

ERTMS implementation in Greece

Though the Greek Railway Network is located far off Central Europe, ERGOSE S.A.1 is speeding up the efforts to comply with the interoperability requirements set by the European Community to facilitate unhindered, at least by technical barriers, cross-border operations at a European level and to...

www.globalrailwayreview.com

Quite informative, I wonder how much got binned during their recession. In any event, they needed to have a much better handling on the risks associated with Single Line Working if the signalling system is not available. From what I can see, it is very blase and it is a managerial / government failure rather than the poor chap who is going to be scapegoated

 

[Gag Halfrunt]

Currently, railway traffic in Greece is controlled by five Traffic Control Centres (TCCs). These are located in the railway stations of Korinthos, Athens, Lianokladi, Larissa and Thessaloniki.

If the Larissa Traffic Control Centre is "located in" Larissa station, perhaps the Larissa "stationmaster" who has been arrested is really the head of the Traffic Control Centre.

Yesterday demonstrators marched from the head office of Hellenic Train to the parliament. The Guardian's reporter spoke to a demonstrator who said that his son worked on signalling upgrading nine years ago.

Greek train crash: anger grows as officials admit rail network problems

Government says rail projects beset by ‘chronic public sector ills’, as death toll from crash rises to 57

www.theguardian.com

“They’ll try to cover it up but we’re not going to let them,” said Stelios Dormarazoglou, pulling his hood tight around his head and shouting “the dead will become one voice, their blood will pursue you”.

He said: “Everyone knows that if the Greek state had wanted, this accident could have been prevented. My own son worked on upgrading the signalling system – nine years ago. Ever since it’s been stalled because companies are only ever interested in profits.”

 

[22pilgrim]

Below is a link to a CNN article containing an audio clip of the station master and one of the drivers, moments before the disaster:


https://edition.cnn.com/videos/worl...manager-audio-bashir-pkg-ovn-intl-hnk-vpx.cnn

 

[w0033944]

So after reading the translated text, it appears that those controlling train movements at Larrisa were the stationmaster and also a 'locksmith' (which I guess is basically a signalling equipment operator) who was under the direction of the stationmaster. The signalling equipment looks like it was faulty or disabled at the time, so trains were being verbally authorized to pass red 'stop' signals. A crossover had been set to route a local train into a platform, was left in that position afterwards and was then forgotten/not checked so one of the trains involved in the collision was mis-routed onto the same track as the other train (after being verbally authorized to leave).

Basically a complete shambles... a combination of the UK Norwich (GE) and Foxcote (S&D) Victorian-era problems/failings...

If you're going to have a 'human interlocking', at least that must be just one person (no divided responsibility or communication issues) dedicated to the task and who is responsible enough to do basic stuff like checking the state of track switches before authorizing train movements.


From what I've seen and read about over the years, the equivalent of a stationmaster also controlling train movements/signalling is pretty common all over the world at smaller wayside stations. From what I've read, we used to have staff grades like 'porter-signalman' in the UK in quiet places.

Yes, there's something of Foxcote in there as well. Utterly horrifying that such basic mistakes have been made on a mainline route.

 

[43096]

It is said the passenger train was running at 140-160 km/h, the freight train likely 100-120 km/h - has there ever been a frontal collision between two trains at these speeds before?
(Compare, for example, the destruction of the two locos from Colwich, where one train was heavily braking and moving at close to a walking pace, with the other close to line speed...)

The only accident of comparable severity of that type I can think of is Ladbroke Grove where the HST was travelling at around 80mph and the Turbo at approx 50mph and the impact was almost head-on (although not quite). In this disaster the closing speed would seem likely to have been at 150mph or more (passenger train at 90mph or more; freight around 60mph).

 

[w0033944]

The only accident of comparable severity of that type I can think of is Ladbroke Grove where the HST was travelling at around 80mph and the Turbo at approx 50mph and the impact was almost head-on (although not quite). In this disaster the closing speed would seem likely to have been at 150mph or more (passenger train at 90mph or more; freight around 60mph).

Maybe Hull Paragon 1912?

 

[midland1]

Maybe Hull Paragon 1912?

Hull Paragon was in 1927 both trains were not moving that fast.

 

[Taunton]

It does appear that things were different to initial supposition that two trains had been dispatched onto a single line in operation. Reading elements of that report, a facing crossover does indeed appear to have been left reversed, forgotten about, and the passenger train crossed it in the dark at slow speed without realising it and then ran wrong line, just alongside the correct track of course, through to the collision. Thus commentary about standard of radio messages etc looks a bit of a red herring. It seems that the signalling was out of action but the power points were still in operation.

Apart from getting back to my initial suggestion of a similarity to the Amtrak South Carolina accident of a short while ago, where reversed points were overlooked during a signal suspension, there have been (very) occasional accidents in the UK with similar factors, and it does have some parallels to the one just outside Taunton, at Norton Fitzwarren, in WW2, where the driver in the dark thought they were running routed onto one track as normal but in fact were on the next parallel track for a couple of miles.

One consideration is that in degraded working you don't both run trains on a double track just by passed messages, AND start crossing trains from tracks in one direction onto the other through points no longer interlocked, as seems to have been done for the local train some time beforehand. One wonders why.

 

[stuu]

It does appear that things were different to initial supposition that two trains had been dispatched onto a single line in operation. Reading elements of that report, a facing crossover does indeed appear to have been left reversed, forgotten about, and the passenger train crossed it in the dark at slow speed without realising it and then ran wrong line, just alongside the correct track of course, through to the collision. Thus commentary about standard of radio messages etc looks a bit of a red herring. It seems that the signalling was out of action but the power points were still in operation.

Apart from getting back to my initial suggestion of a similarity to the Amtrak South Carolina accident of a short while ago, where reversed points were overlooked during a signal suspension, there have been (very) occasional accidents in the UK with similar factors, and it does have some parallels to the one just outside Taunton, at Norton Fitzwarren, in WW2, where the driver in the dark thought they were running routed onto one track as normal but in fact were on the next parallel track for a couple of miles.

One consideration is that in degraded working you don't both run trains on a double track just by passed messages, AND start crossing trains from tracks in one direction onto the other through points no longer interlocked, as seems to have been done for the local train some time beforehand. One wonders why.

Because that is how it had been run for years. I saw a cab ride video years ago on another forum and asked about the signalling - the replies were that it had stopped working and the cabling had been stolen in lots of places. That has been widely reported now as well

It must have been entirely normal practice for the trains to run wrong line, as the driver of the passenger train must have known he was on the wrong line, and yet accelerated away as normal.

This was more or less inevitable one day in those conditions

 

[Class15]

Not really related to the incident, but north of the crash site there’s a weird single-track line which runs parallel to the main line for some time, does anyone know what that is?

 

[stuu]

Not really related to the incident, but north of the crash site there’s a weird single-track line which runs parallel to the main line for some time, does anyone know what that is?

A lot of the line between Athens and Thessaloniki has been rebuilt onto new alignments suitable for higher speed and double tracked, including this section. The original route has been abandoned

 

[Taunton]

Because that is how it had been run for years. I saw a cab ride video years ago on another forum and asked about the signalling - the replies were that it had stopped working and the cabling had been stolen in lots of places. That has been widely reported now as well

It must have been entirely normal practice for the trains to run wrong line, as the driver of the passenger train must have known he was on the wrong line, and yet accelerated away as normal.

This was more or less inevitable one day in those conditions

The (automated) translation of the Greek is somewhat difficult to follow (can someone knowledgeable of Greek railway terms help?), but my understanding is that everyone expected the two trains to be running on parallel lines until the moment of collision. The oncoming freight would have also heard the radio messages and would have picked up if the passenger train was being intentionally sent wrong line. It was dark, and the passenger train went over the crossover from a standing start.

 

[Class15]

A lot of the line between Athens and Thessaloniki has been rebuilt onto new alignments suitable for higher speed and double tracked, including this section. The original route has been abandoned

Ah, thanks for clarifying that.

 

[deaftech]

That looks a really unbiased source!!

Why do you say that?

Maybe you are the one who is biased.

 

[Annetts key]

So, if @Taunton theory is representative of what happened, that’s rather shocking.

It’s a rather fundamental principle that before allowing a train to proceed, that the position of all points in the route are checked and confirmed to be in the correct position for the intended movement.

 

[AlterEgo]

The (automated) translation of the Greek is somewhat difficult to follow (can someone knowledgeable of Greek railway terms help?), but my understanding is that everyone expected the two trains to be running on parallel lines until the moment of collision. The oncoming freight would have also heard the radio messages and would have picked up if the passenger train was being intentionally sent wrong line. It was dark, and the passenger train went over the crossover from a standing start.

It would be interesting to know how gentle that crossover is. I think I would notice going over a crossover, blindfolded, even slowly, from a standing start, unless it was a high speed crossover of the type we have on the WCML South.

 

[Meerkat]

Why do you say that?

Maybe you are the one who is biased.

From knowing the title and reading the article.
Its the equivalent of the RMT press release for any incident!
The incident will be a political football. Hopefully someone coughs up to get the signalling sorted

 

[Annetts key]

From knowing the title and reading the article.
Its the equivalent of the RMT press release for any incident!
The incident will be a political football. Hopefully someone coughs up to get the signalling sorted

A union is there to protect its members. Some of which may have been killed or injured in this dreadful disaster. So the unions would obviously be calling out any unsafe practices and wanting safety to be properly considered.

Is the safety of workers something you disagree with?

 

[stuu]

The (automated) translation of the Greek is somewhat difficult to follow (can someone knowledgeable of Greek railway terms help?), but my understanding is that everyone expected the two trains to be running on parallel lines until the moment of collision. The oncoming freight would have also heard the radio messages and would have picked up if the passenger train was being intentionally sent wrong line. It was dark, and the passenger train went over the crossover from a standing start.

But the driver of the passenger train would have been able to see that he was on the wrong line. It was a modern electric loco so should have good enough headlights to tell which track it was on .The crash site isn't just outside the station, it was 10+ miles down the line

 

[Meerkat]

A union is there to protect its members. Some of which may have been killed or injured in this dreadful disaster. So the unions would obviously be calling out any unsafe practices and wanting safety to be properly considered.

Is the safety of workers something you disagree with?

Different subject from whether they are an unbiased news source. The Socialist Worker is about as credible as an unbiased news source as the Mail or Sun.
But this illustrates the problem - blame getting thrown about for political reasons clouding finding and fixing the real issue.
The poor stationmaster is in the middle of an ethical mess on the lines of 'following orders'. Bit surprised the Greek unions were letting this happen - I can't imagine the RMT accepting their members being forced into such a risky method of work as business as usual.

 

[Annetts key]

Different subject from whether they are an unbiased news source.

Point of order, there is no such thing as an unbiased news source. Even when an organisation tries to be as unbiased as possible, others will call them out for being biased.

Similarly, without very credible evidence from non-human sources, there is no such thing as “the truth”.

I don’t buy or routinely read The Socialist Worker, The Mail or The Sun or indeed any paid for newspaper. So, I don’t know how credible or not their articles are.

In the U.K. at the moment, the RMT is currently fighting proposed changes that will make the mainland U.K. rail network less safe.

 

[John Palmer]

There's a cab ride video here:

that covers the route between Thessoloniki and Larissa. It is at least five years old, so may not be representative of the current track layout. However, it shows one facing crossover only on the northern approach to Larissa station. It is possible that Intercity 62 began wrong road travel by being routed onto the Athens-bound track before it reached the Larissa platforms, but in that case it is difficult to understand how the stationmaster could not have realised it was moving against the normal traffic flow. For that reason I think it more likely that the diversion onto the Athens-bound line occurred at the crossover to be seen in the video at about 1 hr 26 minutes.

Since it appears that the Larissa stationmaster was labouring under the mistaken belief that 62 was running on the right-hand road in the normal way, his knowledge of the freight train's whereabouts is probably unimportant as he would have expected any opposing trains to pass each other on different tracks. If that's correct then his knowledge of the lie of the relevant crossover at the time of authorising 62 to proceed through the Larissa-Poros section may prove to be of much greater significance.

I'm increasingly disinclined to place much reliance upon the translated reports on the Protothema website. The report I linked includes what seems to be the recording of a radio or telephone conversation between the stationmaster and the pointsman who operated the critical crossover, but there's no indication of whether anything from the conversation has been omitted (e.g. any response to the pointsman's “Done, well, I'll leave it as it is diagonally” remark), and the translation may well not accurately represent in English what was said. There's also the curiosity that the stationmaster is variously referred to as 'Basil' and 'Santa' in the reported conversations. Neither of these appear to form part of the name of the man who has been charged under Greek criminal law.

The only conclusion about the conversation between the pointsman and the stationmaster on which it seems that much reliance can be placed is that the participants must have been remote from each other, implying that the stationmaster was not in a position to satisfy himself by personal observation that the lie of the crossover was such as to warrant issue of authority for 62 to pass a signal displaying a 'stop' aspect and proceed into the Larissa-Poros section. A system such as that at Larissa that relies instead on perfect communications between points operators and those authorising train movements strikes me as one carrying unacceptable levels of risk.​

There is a further report on the Protothema website that the direction in which 62 was travelling was displayed on a console the stationmaster was observing. The report doesn't say whether the display also indicated the track on which the train was running, merely that it only covers the immediate area of Larissa. If 62 entered the Athens-bound line at the crossover to be seen in the video, and the stationmaster's display did show the track on which it was running then it would have appeared to be on the correct track up to the moment it traversed the crossover, and may well have passed out of the area covered by the display shortly thereafter.

 

[Lawrence18uk]

YouTube video post separately (thanks @John Palmer)

Crash location 1:15:22
Several comments posted 3, 4, 5 years ago saying how come the driver is passing so many signals at Red? yes, the red signals have X under them which means that they not working, and someone else saying "yeah they're being dispatched by radio and this has been going on for years"

 

[theageofthetra]

YouTube video post separately (thanks @John Palmer)

Crash location 1:15:22
Several comments posted 3, 4, 5 years ago saying how come the driver is passing so many signals at Red? yes, the red signals have X under them which means that they not working, and someone else saying "yeah they're being dispatched by radio and this has been going on for years"

The question has to be asked how did the Greek rail unions, Greek rail regulator, and the EU regulator allow this to carry on for so long?

Surely that kind of emergency working should only be for as short a time as possible until the fault is fixed?

 

[w0033944]

Hull Paragon was in 1927 both trains were not moving that fast.

Of course!

 

[the sniper]

But the driver of the passenger train would have been able to see that he was on the wrong line. It was a modern electric loco so should have good enough headlights to tell which track it was on .The crash site isn't just outside the station, it was 10+ miles down the line

I don't buy into the idea that the Driver didn't realise they'd been crossed over or notice they were on the opposite line. Rather than discounting the failure of communication, it seems fundamental to me that a lack of a clear communication of what was being permitted is at the core of what went wrong here, beyond an initial mistake of leaving the points incorrectly set. In this country there's a clear distinction in the processes between passing a signal at Danger, passing multiple signals at danger and single line working for a reason, which makes a situation such as this playing out so casually far more unlikely.

The oncoming freight would have also heard the radio messages and would have picked up if the passenger train was being intentionally sent wrong line.

Does their system work like the US, like air traffic control where everyone can hear, or like here, where communication is usually direct only between the signaller and the Driver concerned?

EDIT: Seems from the video above it's the former, which given the standard of communication, opens up a bunch of other risks...

Frankly, the whole thing looks like a basket case operation.

 

[MarkyT]

The question has to be asked how did the Greek rail unions, Greek rail regulator, and the EU regulator allow this to carry on for so long?

Surely that kind of emergency working should only be for as short a time as possible until the fault is fixed?

Maybe the financial difficulties Greece and its railways faced caused maintenance and spares supply contracts to unravel leading to an insurmountable backlog of failed signalling equipment, hence so many signals OOU. Technicians with the skills neccessary to fix the scale of problems may also have been laid off. Operations staff may thus have become accustomed to using backup manual blocking methods routinely, without full protection of interlocking. Maybe the freight train was able to be signalled normally through the control and interlocking equipment remotely from the TCC, while a fault prevented normal signal clearance for the passenger train and then there was some confusion over who had authority to issue a verbal order to proceed.

 

[Lucan]

It is possible that Intercity 62 began wrong road travel by being routed onto the Athens-bound track before it reached the Larissa platforms, but in that case it is difficult to understand how the stationmaster could not have realised it was moving against the normal traffic flow. For that reason I think it more likely that the diversion onto the Athens-bound line occurred at the crossover to be seen in the video at about 1 hr 26 minutes.

There is also a facing crossover at 1:22:20 in the video - ie at a point further north from Larissa and still before the crash site. That does not change what you mean of course.

The (automated) translation of the Greek is somewhat difficult to follow (can someone knowledgeable of Greek railway terms help?), but my understanding is that everyone expected the two trains to be running on parallel lines until the moment of collision.

Everyone, maybe, except the driver of the passenger train. There is no way he could be unaware he was on the "wrong" side, so presumably he took it to be the signalman's intention.