S- Bahn collision Munich

[Fragezeichnen]

Unlike things like derailments, shouldn't head on collisions through human error be something that can be made near enough impossible in the 21st century in a wealthy country like Germany.

You can of course run the railway on that basis if you want. However, the passengers won't be too pleased if you tell them "Sorry, a microchip has failed and we can't change the signal to green, no trains for the next two weeks until we fix it". So virtually every rulebook contains provisions for degraded working. The greater the danger, the more elaborate the procedures.

Is it that in Germany, different expectations exist for the fallibility/competence of the individual and there are therefore fewer protections against it?

I think it's true that different railway administrations have evolved different views on the risk of different activities, as a result of the culture, and of what kind of incidents have occurred in their history.

It amuses me to see the massive arguments in the UK about DOO safety even though there are elaborate CCTV and mirror systems, and dispatchers at large stations. In Germany DOO means door sensors and the driver sticking his head out the window and it seems to work fine. Call me cynical, but I suspect that might have something to do with the lack of ticket gates meaning there are limited labour savings to be made from driver dispatch in Germany.

On the other hand, no train is allowed to exceed 100mph in Germany without full continuously-updated in-cab signalling monitoring the speed, wheras in the UK they planned to go up to 140mph with only lineside signals. This goes back to a number of nasty accidents in Germany at the time high speed running was first introduced, in which drivers inexplicably failed to reduce speed and derailed at high speed. Even on lines without that drivers are given enough assistance in the form of written route guides, additional signage, speed signalling, and the strict braking curves enforced by Indusi that it should always be very clear what speed is permitted.

Overall I think signallers are trusted more, and drivers are trusted less under the German system, but you could probably find counterexamples for both.

 

[Bletchleyite]

You can of course run the railway on that basis if you want. However, the passengers won't be too pleased if you tell them "Sorry, a microchip has failed and we can't change the signal to green, no trains for the next two weeks until we fix it". So virtually every rulebook contains provisions for degraded working. The greater the danger, the more elaborate the procedures.

And there's little more dangerous than a head-on collision, hence why the UK doesn't have anything like Zs1 for lines where that could occur. Once it had been properly established that the line actually was clear (not just one signaller thinking it was, though I'm not sure what the actual procedure is), pilotman working would generally be used, where one designated person is required to be on board any train travelling on the single line (and can be moved back and forth by road if necessary).

 

[edwin_m]

And there's little more dangerous than a head-on collision, hence why the UK doesn't have anything like Zs1 for lines where that could occur. Once it had been properly established that the line actually was clear (not just one signaller thinking it was, though I'm not sure what the actual procedure is), pilotman working would generally be used, where one designated person is required to be on board any train travelling on the single line (and can be moved back and forth by road if necessary).

If two trains are to follow in the same direction, I believe the pilotman is still allowed to instruct the first driver to proceed and accompany the second. Though a road transfer might still be needed if for some reason the second train was cancelled at short notice.

 

[Bletchleyite]

If two trains are to follow in the same direction, I believe the pilotman is still allowed to instruct the first driver to proceed and accompany the second. Though a road transfer might still be needed if for some reason the second train was cancelled at short notice.

Yes, forgot about that, basically human "train staff and ticket".

 

[Spoorslag '70]

And there's little more dangerous than a head-on collision, hence why the UK doesn't have anything like Zs1 for lines where that could occur. Once it had been properly established that the line actually was clear (not just one signaller thinking it was, though I'm not sure what the actual procedure is), pilotman working would generally be used, where one designated person is required to be on board any train travelling on the single line (and can be moved back and forth by road if necessary).

Well, if a driver would override the TPWS and depart against a signal at danger, you would have the very same effect that happened in Schäftlarn. Most likely the Zs1 did not have too much influence on the incident, if at all.

 

[Bletchleyite]

Well, if a driver would override the TPWS and depart against a signal at danger, you would have the very same effect that happened in Schäftlarn. Most likely the Zs1 did not have too much influence on the incident, if at all.

Yes, but a driver is required to consult the signaller before doing that. OK, he could maliciously fail to do so, but is anyone required to be consulted before triggering Zs1?

An improvement for TPWS would be to require the driver to obtain an override code from the signaller, or even better from TOC Control, I guess.

 

[duesselmartin]

With Zs1 the signaller has to consult the next block. Similarity to TPWS again.

The problem I see is the speed allowed under Zs1 and that the Signals in the other direction do not give a warning.

 

[MarkyT]

With Zs1 the signaller has to consult the next block. Similarity to TPWS again.

The problem I see is the speed allowed under Zs1 and that the Signals in the other direction do not give a warning.

Or to be more accurate, whatever the speed allowed after seeing the substitute signal, the big risk is in some cases that opposing route locking engaged by a normal full speed movement authorised by a signal and underway in the other direction is not tested and proved disengaged before Zs1 or indeed Zs7 (proceed on line of sight) is allowed to show proceed. This cannot happen with the similar UK POSA (proceed on sight authority) type of signal under any circumstance, although of course a verbal or written authority might still be given instead, which could result in the same outcome.

 

[158756]

I can get the benefits of Zs1 on unidirectional lines - if the rear train drives on sight including checking points positions before crossing them, all is good (bar points shifting under the train, I suppose). But it can't possibly work on a single line or a bidirectional line, because one train driving on sight is all very well, but the other one could be bearing down on it at 200km/h with no idea what is going on.

The mind boggles as to who thought this could even possibly be a good idea.

It can still cause problems on unidirectional lines, like this case in 2017 https://en.m.wikipedia.org/wiki/Meerbusch_train_crash

(In that case the driver apparently didn't or wasn't instructed to proceed at a reduced speed, after the signaller had lost track of the stationary freight train occupying the section)

 

[Bletchleyite]

It can still cause problems on unidirectional lines, like this case in 2017 https://en.m.wikipedia.org/wiki/Meerbusch_train_crash

(In that case the driver apparently didn't or wasn't instructed to proceed at a reduced speed, after the signaller had lost track of the stationary freight train occupying the section)

True. I would agree that anything that overrides the interlocking before the line has been physically checked and reported clear should be line-of-sight - if Zs1 allows full speed, then it needs banned.

 

[duesselmartin]

Meerbusch also showed that the lack of tail lights in freight trains are also a factor. The train might have been seen earlier.
Zs1 certainly needs reforming ie. Snail pace running only. I hope the EBA will finally tighten regulations here, even if that might not be the cause in Munich.

 

[43096]

Zs1 certainly needs reforming ie. Snail pace running only. I hope the EBA will finally tighten regulations here, even if that might not be the cause in Munich.

That's fine as far as it goes, but the issue is also that the train coming the other way is likely running at line speed.

 

[MarkyT]

That's fine as far as it goes, but the issue is also that the train coming the other way is likely running at line speed.

Exactly, just as at Bad Aibling. Even if the specially authorised train was going walking pace, the other may have had no idea what might be coming round the corner and has had a full-speed normally signalled authorisation.

 

[BRX]

Under what circumstances exactly do you end up with a train in a section that the signaller would have forgotten about or not noticed? Are these situations where there's not track circuits?

 

[MarkyT]

Under what circumstances exactly do you end up with a train in a section that the signaller would have forgotten about or not noticed? Are these situations where there's not track circuits?

In this case fully 'track circuited', although using axle counters, which have been standard for decades in Germany. If you play the cab video up-thread on slow you can see the sensors bolted to the rails, and the little square termination boxes for the associated cables next to the track.

 

[edwin_m]

Under what circumstances exactly do you end up with a train in a section that the signaller would have forgotten about or not noticed? Are these situations where there's not track circuits?

Presume a track circuit (or an axle counter section) shows occupied, but the signaller has come to the conclusion that it is a failed detector not a train.

 

[quentinc]

SZ Article today is saying one of the drivers, not the signalling/track, is the subject of the investigation.

 

[MarkyT]

SZ Article today is saying one of the drivers, not the signalling/track, is the subject of the investigation.

There is an override switch for the PZB in the cab, which would allow a train to pass the 2000Hz inductor, the trainstop, at the signal when it is at red. This must never be used lightly and rules state there must be authorisation for this from the signaller, either verbal/written or by use of the Zs1 or Zs7 substitute signal, where equipped. If the override is not used, then there will be an emergency brake application if the driver attempts to pass the red signal. The PZB trainstop is designed to be 'failsafe' and should intervene even if the local signalling equipment power supply has failed and signals are extinguished.