Signals

THE SIGNAL BOX


Railway signalling discussion

Collision in Germany

Signalling outside the UK (but including Northern Ireland), past, present and future

Re: Collision in Germany

Unread postby JanH » Tue Feb 16, 2016 10:25 pm

(somehow I managed to double post this, see post below for full content)
Last edited by JanH on Wed Feb 17, 2016 10:05 am, edited 1 time in total.
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby Frank » Tue Feb 16, 2016 10:40 pm

Hello,


1) What conditions are required to be satisfied before such a subsidiary signal can be cleared ?


a) technical
On Relay Boxes like that one in Bad Aibling you must first lock the automatic movement of points (Weichenlaufkette sperren ,lock Point driving chain)
then you could turn on the Zs1 Subsidiary Signal.This Signal is illuminated then for 90 seconds.

b)Rules
The Signal man must check that the Line is clear.
This is done by the Track Circuit if he is corect working (Einzel-Räumungsprüfung / Line Clear Ceck).



2) When a driver proceeds on the authority of such a subsidiary signal, what restrictions apply ?


If you as a Train Driver override the active PZB-Magnet of a Signal at Danger, the PZB-Eelectronic restricts
the speed of the Train to 40 Km/h, In the Cab a Horn sound come up and on the PZB Display the Befehl 40 is
illuminated.After a distance (depending on what kind of Train it is (Freight/Passenger)) the supervising ends.
On Passenger Trains the distance is 1.250 Meter after passing the Signal.


3) How does the signaller clearing the subsidiary signal know that the single line is unoccupied ?


see above at 1)
User avatar
Frank
Rest-day relief
Rest-day relief
 
Posts: 503
Joined: Mon Nov 26, 2007 10:18 pm
Location: Dortmund-Rhine Westphalia- Germany

Re: Collision in Germany

Unread postby Wilkinstown » Wed Feb 17, 2016 12:45 am

Thanks Frank.

Two further questions

Is this section of line controlled from one or two locations ?

If co-operatively controlled from two locations does the signaller at the opposite end of the section have any control of the Zs1 subsidiary signal at the opposite end ?
Wilkinstown
Branch line box
Branch line box
 
Posts: 169
Joined: Sun Feb 10, 2008 11:43 pm

Re: Collision in Germany

Unread postby JanH » Wed Feb 17, 2016 9:59 am

Frank wrote:Maybe......but there is something that didn`t match with this theory....with Ersatzsignal the Train Driver receive the Orders

1.Pass the Signal at Danger
2.go ahead with medium speed and on sight

Actually no, you're confusing Zs 1 (Ersatzsignal) and Zs 7 (Vorsichtssignal).
Zs1 requires a reduced speed of 40 km/h only while actually passing the signal (this is also enforced by the train protection system - you can't bypass the train stop function of the PZB when going faster than 40 (or at most 45 ?) km/h), within station or junction areas and the adjoining point area ("anschließender Weichenbereich", Zs 1 means no speed signalling information, so a blanket speed restriction applies) or if the signal is a combined main/distant signal (red main aspect means you get no info about the state of the upcoming signal, so again a blanket speed restriction until you've either reached the next signal or 2000 m have elapsed, i.e. the maximum possible distance between distant and main signal plus some contingency). In all other cases, e.g. for simple plain line running, no speed restriction applies. This is because, just like a written order to the same effect, Zs 1 is supposed to be given only if the signaller can make sure that the line is really clear. In all other cases, running on sight needs to be instituted.

Zs 7 on the other hand means running on sight at 40 km/h max or slower depending on sighting conditions, a bit like the new PosA aspect in GB. It is usually provided for signals reading into/within stations for new installations controlled from (area) signalling centres - the classic way of making sure the line is clear within the station area is looking out of the signalbox window, which is no longer possible if the station is remotely controlled from possibly quite a distance away. Hence the easiest way to deal with failures within a station area is to simply institute running on sight with the help of the Zs 7 subsidiary signal.

However for signals reading onto the plain line, most of the time even for new/renewed installations a Zs 1 is provided (a signal can only have one of them, either a Zs 1 or a Zs 7, but not both), because not all failure conditions warrant running on sight, e.g. a burnt filament, or a points failure in a neighbouring line leading to loss of flank protection wouldn't. Because plain line sections can be relatively long, running on sight looses a proportionally larger amount of time, hence a classic Zs 1 is provided.
If there's some failure with the train detection systems and the signallers can't ascertain that the line is clear (because with the advent of large scale signalling centres, there's nobody at the next station who can verify that the last train to traverse the affected section arrived complete), the Zs 1 mustn't be used and a written order including running on sight needs to be used instead.

Frank wrote:Hello,
If you as a Train Driver override the active PZB-Magnet of a Signal at Danger, the PZB-Eelectronic restricts
the speed of the Train to 40 Km/h, In the Cab a Horn sound come up and on the PZB Display the Befehl 40 is
illuminated.After a distance (depending on what kind of Train it is (Freight/Passenger)) the supervising ends.
On Passenger Trains the distance is 1.250 Meter after passing the Signal.

Sorry to correct you again, but not quite. The speed limit of 40 km/h (although technically, the PZB only trips you past 45 km/h) while passing a signal at danger is enforced only while the "Befehl" override switch is being operated. After replacing the "Befehl" switch, you might still be subject to some residual 1000 Hz/500 Hz restrictions from approaching the signal you've just passed, but those will elapse after a few 100 m - the 1250 m you cite apply for the 1000 Hz speed control after passing a distant signal showing a restrictive aspect, so the 1000 Hz speed control will terminate 1250 m after the distant signal.
And as I said above, for a Zs 1 or the equivalent written order the rules requires a speed restriction only within station/junction areas (where speed restricted points are quite likely) or if the signal you had to pass was a combined main/distant signal anyway.

Wilkinstown wrote:Two further questions

Is this section of line controlled from one or two locations ?

It's one signal box controlling both stations.
Wilkinstown wrote:If co-operatively controlled from two locations does the signaller at the opposite end of the section have any control of the Zs1 subsidiary signal at the opposite end ?
As far as the interlocking is concerned, no.
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby Mike Hodgson » Wed Feb 17, 2016 11:07 am

Thanks Jan. That sounds more sensible.

So you're saying that if the line was indicated as occupied and the Fdl mistakenly thought it was a fault, he should have issued a written order and ought not to have cleared Zs1? This might have reduced the closing speed but would not necessarily have avoided the collision.

Is he allowed to authorise this without getting some independent confirmation of his intended action? I realise this might be organisationally difficult to achieve and doing so may well cause further delay in genuine fault situations, but delay might well allow the first train to clear the section in situations like this, limiting damage to the delay and a great deal of embarrassment.

How would his written order be transmitted ? If it's a large area I assume he can't normally hand it to the driver - is it read out privately by signal post telephone/secure radio or over open radio that the other train should overhear? Does the process of issuing it cause enough delay to allow the first train to clear a long section?
User avatar
Mike Hodgson
Double-manned box
Double-manned box
 
Posts: 2508
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Collision in Germany

Unread postby Frank » Wed Feb 17, 2016 12:14 pm

Hello Mike,

no the Zs1 is the replace of a written Order but they still there.
Train orders are transmitted via Train Radio (GSM-R) directly to the Train Driver if nessarary or if there is Track works etc. a special Man do this at the Track.

Here photos of the Box:
http://www.rosenheim24.de/rosenheim/man ... 11501.html

The view is in the direction of the Accident.

The Sation is in Kilometer (Milepost) 27,8 of the Line (Holzkirchen) Bad Aibling - Kolbermoor - (Rosenheim).

At MP 28,7 follows the automatic Block Signal SBk 313 and in MP 33 the next (Unmanned and from Bad Aibling remote controlled) Station Kolbermoor.

In the opposite direction there is the SBk 314.

Both SBk are for the Protection of the following Crossings.So that means the SBk remains in Danger until the Train has triggered the Activation Point for the Crossing and the Crossing is then closed.

So if you for example set the Route Bad Aibling to Kolbermoor, the Track Circuit shows ocupied between Bad Aibling and SBk 313.
The next Track Circuit shows free until the Level Crossing is working and vice versa.
But the Starter Signals in Kolbermoor ware locked.

In this Trap the unlucky Signmal man get.




@Jan
Actually no, you're confusing Zs 1 (Ersatzsignal) and Zs 7 (Vorsichtssignal).

No i am not confused, because both Signals are for different Purpose developed from the former State Railway.

and

(although technically, the PZB only trips you past 45 km/h


drive on the technical tolerance Limit is not the right way.
User avatar
Frank
Rest-day relief
Rest-day relief
 
Posts: 503
Joined: Mon Nov 26, 2007 10:18 pm
Location: Dortmund-Rhine Westphalia- Germany

Re: Collision in Germany

Unread postby JanH » Wed Feb 17, 2016 1:20 pm

Frank wrote:@Jan
Actually no, you're confusing Zs 1 (Ersatzsignal) and Zs 7 (Vorsichtssignal).

No i am not confused, because both Signals are for different Purpose developed from the former State Railway.
Sure, but I was specifically referring to this statement of yours (highlighting mine):
Frank wrote:Maybe......but there is something that didn`t match with this theory....with Ersatzsignal the Train Driver receive the Orders

1.Pass the Signal at Danger
2.go ahead with medium speed and on sight
As I said, a Zs 1 by itself confers no further restrictions after actually passing the signal, except where required because of distant signals or potentially diverging points within a station/junction area.

Frank wrote:
(although technically, the PZB only trips you past 45 km/h

drive on the technical tolerance Limit is not the right way.

You're right of course.

One thing to keep in mind is that it is quite possible that at the time the eastbound train was authorised to pass those signals at danger, the single line section was still unoccupied because the westbound train was still waiting for its scheduled departure time at Kolbermoor, where the trains were originally scheduled to cross. It might be possible that the signalman got confused about the fact that he'd already cleared the route from Kolbermoor to Bad Aibling and then thought the route lights for the westbound train [1] were coming from a failed attempt to set the route for the eastbound train from Bad Aibling, which for some reason [2] wouldn't clear. But I admit that's getting rather into speculative territory...

[1] As far as I've heard, the westbound train departed on a green signal, so barring a very weird interlocking failure, at the very least the route from Kolbermoor towards the block signal protecting the level crossing had been set. It also means that, as is often the case with such accidents, the timing was very unlucky. If the westbound train had departed a bit sooner, maybe the block occupation would have alerted the signaller a bit earlier, or the westbound train would have come to a stand at the block signal, reducing the collision speed at the very least. As it was, the westbound train had passed the distant signal, but, if I remember correctly, was still travelling at around 50 - 60 km/h when the collision occurred. Or if the westbound train had waited a bit longer, the starter at Kolbermoor would have gone back to red once the eastbound train had passed the block signal in the middle of the section.
[2] Some reason in this case of course being the interlocking refusing to set two opposing routes onto a single line.
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby michiel » Wed Feb 17, 2016 2:01 pm

It seems very strange to me, that there would not be some specialised indication about the single track section being occupied, i.e. the direction of that line should be shown, barring any movement in the other direction. Think the Fahrtrichtung-feld in the good old Siemens Block. To have to trust track circuits which can happily show 'clear' because of a level crossing not being closed, seems to be an accident waiting to happen ...
Regards,
Michiel Rademakers
User avatar
michiel
Crossing box
Crossing box
 
Posts: 61
Joined: Sun Nov 11, 2007 3:20 pm
Location: Tilburg, NL

Re: Collision in Germany

Unread postby JanH » Wed Feb 17, 2016 2:56 pm

I think Frank's description must have become somewhat lost in translation.
To clarify, the basic panel indications for a German relay interlocking are very similar to their British counterparts:
  • Dark, unlighted track means no route set. It also means that that piece of track is probably clear, but you're not allowed to rely on this, because a blown bulb in the panel lighting is equally possible.
  • White/yellowish route lights mean a route has been set and the track is positively proved clear.
  • Red means occupied
Level crossing or not makes no difference there as far as the track indications are concerned.

The only effect the level crossing will have is that if the crossing is of the type interlocked with the main signals [1], it will keep those protecting block signals in the middle of the single line section at danger even with a route set until the crossing has successfully operated.

[1] Similar to how MCB/CCTV crossings are interlocked in GB, except in Germany half-barrier crossings can be fully interlocked as well. The only difference is that for half barriers the "crossing clear" is effectively automatically operated once the barriers are down.
Last edited by JanH on Wed Feb 17, 2016 4:13 pm, edited 1 time in total.
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby Frank » Wed Feb 17, 2016 3:36 pm

Hello,


The only effect the level crossing will have is that if the crossing is of the type interlocked with the main signals [1], it will keep those protecting block signals in the middle of the single line section at danger even with a route set until the crossing has successfully operated.


And the second effect is that the Track on the Panel is dark (Free) until the Protecting Block Signal go to procced.


But that only one fact.

For understanding:
Set Route from Kolbermoor to Bad Aibling, the Erlaubnis-Block (Permission Block) locks all Starter Signals in Bad Aibling to direction Kolbermoor and also all
Block Signals in that direction (in this case it was only one).This is done automatically by the Relais system.
On the Panel there is the Route shown by Yellow Light from Starter Signal to the Border of the Station (Einfahrsignal), the Block section is still stark from Kolbermoor.
Also there is a little Arrow iluminated yellow in the direction of Bad Aibling on the Panel Track after the Starter Signal , pointing at Bad Aibling.
In Bad Aibling another Arrow pointed at Kolbermoor and shows the Signal man that the Block (Erlaubnis/Permission) is set Kolbermoor to Bad Aibling.

So it is mystery why the Signal man set the Zs1
User avatar
Frank
Rest-day relief
Rest-day relief
 
Posts: 503
Joined: Mon Nov 26, 2007 10:18 pm
Location: Dortmund-Rhine Westphalia- Germany

Re: Collision in Germany

Unread postby JanH » Wed Feb 17, 2016 7:35 pm

Ah, I think I see what you mean about the block section, although I'm still a bit puzzled about the references to the level crossing as far as the track indications are concerned.
To expand, with modern-ish German interlockings, there are two ways to signal the sections of track between two stations. One possibility is using some sort of automatic block system of whatever variety - like the normal automatic block, or centralised block ("Zentralblock"), or whatever. In that case you set a route from the station into the automatic block section and another route out of the block section into the next station, but everything in between is handled by the block system. This also means that by default, no route lights are displayed within the automatic block section (although unless I'm mistaken, you can temporarily turn them on manually one block section at a time to get a positive indication that the track is clear if the rules call for it under certain failure conditions), so automatic block sections normally are either dark (default) or red (occupied). The state of any level crossings shouldn't have any influence there.
For bi-directional sections of track, as Frank has described an arrow-shaped direction indicator is provided to indicate for which direction of travel routes can be set.

The other possibility occurs where multiple neighbouring stations are controlled from the same interlocking. In that case, the block sections on the plain line between stations can actually be signalled to the same interlocking principles as a station area, i.e. via setting of routes. In that case, a direction of flow arrow and explicit directional controls are not necessarily provided. This should also mean that you'll get route lights throughout the whole area between the two stations, including the plain line. The interlocking will of course still prevent the setting of opposing routes through the same piece of track, but there's no need to explicitly switch the permitted direction of the track (the "Erlaubnis").

From what I've read elsewhere, it's quite probable that the single line between Bad Aibling and Kolbermoor was signalled according to the latter principle, i.e. as far as the interlocking was concerned, the single line was worked as if it was part of a station area.

But yes, the big mystery remains - how did the signaller manage to overlook the fact that he'd already set at least part of the route for the westbound train...
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby Wilkinstown » Thu Feb 18, 2016 12:47 am

So if I understand correctly

1) Both stations are on the same panel.

2)Setting the route from Kolbermoor to Bad Aibling sets the traffic direction but only illuminates a yellow light on the panel from Kolbermoor to the einfahrsignal for Kolbermoor. The einfahrsignal is in British terms the station home signal. The block section remains dark on the panel until occupied by the train.

3)The Zs 1 signal at Bad Aibling towards Kolbermoor can be cleared despite the traffic direction being set in the opposite direction.
Wilkinstown
Branch line box
Branch line box
 
Posts: 169
Joined: Sun Feb 10, 2008 11:43 pm

Re: Collision in Germany

Unread postby JanH » Thu Feb 18, 2016 2:10 am

1) Yes.
2) Not quite. Your description sounds correct for normal automatic block signalling, but if the sources I've found are correct, from an interlocking perspective, the single line between those two stations is signalled as if part of the station area, i.e. no automatic block signalling with directional controls, but simply normal route setting. So in that case there would be no direction indicator, but the route lights should light up for the full extent of the route.
3) Yes. As Frank said, the automatic calling of points needs to be suspended and later interlocking generations might place some additional requirements, but basically it can be cleared pretty much at will.
Compared to British signal boxes, German NX-style signalling generally seems to have traditionally included a greater number of manual overrides - if necessary, signallers can also pretty much release any route at will, even directly in front of a train and also e.g. operate occupied points. This gives a greater flexibility when dealing with train detection failures (no more track circuit failures that lock up half the station until the technician arrives), but has of course the drawback of placing a much greater amount of responsibility on the shoulders of the signaller during such situations.
JanH
Trainee
Trainee
 
Posts: 9
Joined: Mon Nov 19, 2012 6:21 pm

Re: Collision in Germany

Unread postby hmmueller » Thu Feb 18, 2016 7:56 am

JanH wrote:... if the sources I've found are correct, from an interlocking perspective, the single line between those two stations is signalled as if part of the station area, i.e. no automatic block signalling with directional controls, but simply normal route setting. So in that case there would be no direction indicator, but the route lights should light up for the full extent of the route.


Photographs of such a situation can be seen in this posting of the SpDrS60 at Grafing near Munich in (the German version of) my interlocking blog. I had the chance to take these photos in my home town a few years ago - something normally not allowed in German signal boxes. A click on each picture shows a somewhat larger picture (I can also provide the full-format pictures with 2560x1920px).

The interlocking was built around 1971 and, due to various extensions (especially quadruple tracking from Munich to Grafing for the S-Bahn around 1999), became quite crammed.

Here, the single track line between Grafing Bahnhof and Ebersberg is technically a track inside the "interlocking limits" Grafing+Ebersberg, i.e., a "simple station track," which therefore is not secured by a line block system, but by in-station routes like any other station track. The third picture shows a route setup for train 8400 from Ebersberg's stub track (the two tracks at the bottom) to Grafing Bahnhof (the single line directly above is the continuation - it shows white lights indicating that there is a route set up and, in the third matrix field from the left, the barely visible cleared home signal of Grafing Bahnhof). And in theory, the train director could clear here the infamous Zs1 on any starter signal in Grafing Bahnhof in the direction towards Ebersberg ...

H.M.
==== Austrian signal boxes and interlockings (German and English), with some looks beyond ====
User avatar
hmmueller
Crossing box
Crossing box
 
Posts: 99
Joined: Sat Feb 23, 2013 8:24 pm
Location: Bavaria, Germany

Re: Collision in Germany

Unread postby Wilkinstown » Thu Feb 18, 2016 1:27 pm

Again thanks for the clarifications.

The stand out point appears to be that Zs 1 can be cleared to admit a train onto a single line section of track to which another train has already been admitted, in the opposite direction, on the authority of a normal running signal.

The use of Zs 1 clearly needs to be very tightly controlled (as I am sure it is by regulation) as misuse has the potential for catastrophe.
Wilkinstown
Branch line box
Branch line box
 
Posts: 169
Joined: Sun Feb 10, 2008 11:43 pm

PreviousNext

Return to Signalling - overseas

Who is online

Users browsing this forum: No registered users and 2 guests