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 Frank » Wed Mar 9, 2016 6:58 pm

Hello,

It seems counter intuitive for the facing Switch at Bad Aibling not to be set and locked for the Kolbermoor train.


That is not practice in Germany on Relay Boxes,Route is set when the Train is entering the Block section (you here a bell) or without Block when the Time Table arivve is there.


Do my figures agree with the black box timings?

They are at the Federal Investigation Board (EBA) and we have to wait for the Report from there.


if he reacted at any time within 1:30 seconds after seeing the Kolbermoor train illuminate his board then the crash may have been averted presuming both trains could come to a standstill in 27 and 21 seconds respectively.


That is Time enought, because on every Signal Box on the Telephone there is a short Dial-Button for the GSM-R Emergency Call.
You just have to press the Button and automatically the GSM-R System activate a Roundspeak to all active Train Radios or Handheld-Phones with GSM-R in the Area.This is trained every Year with every Signal man on a Box.
The Message is very Simpel
Betriebsgefahr, alle Züge zwischen Bad Aibling und Kolbermoor sofort anhalten, ich wiederhole Betriebsgefahr, alle Züge zwischen Bad Aibling und Kolbermoor sofort anhalten
(Emergency All Trains between Bad Aibling and Kolbermoor Stop immediatly , I Repeat Emergency All Trains between Bad Aibling and Kolbermoor Stop immediatly )
For the Train Driver the Rules say, that the shown Flashlight off the Emergency Call without a Voice (or with not understanding Words) restrict him to
40 Km/h and proceed on sight.Also to call the Signal Man on the next Box.
If he hear the Voice Message, he must apply Emergency Break on the Train.On the modern EMU this this brings up a Retarding of 2m/s ore more depending on the Equipment

That is the second Mistery, why he do that not and use instead a "normal" call.
User avatar
Frank
Rest-day relief
Rest-day relief
 
Posts: 501
Joined: Mon Nov 26, 2007 10:18 pm
Location: Dortmund-Rhine Westphalia- Germany

Re: Collision in Germany

Unread postby Culburrarail » Thu Mar 10, 2016 12:14 am

I have been investigating the situation where Human Error became a factor. As you say the mystery of why the signal man did what he did will be the subject of the official enquiry. My enquiry is entirely focused upon how the PZB90 systems allowed a green light at Kolbermoor to remain active after the Bad Aibling train had entered the Block after the signal man used the Zs 1 red signal over-ride. The train from Bad Aibling had been en-route for about 1:30 before the Kolbermoor train left the station with that Green light still showing. It seems to me that the system as used allows independent use of the ends of a block. This could never happen here and I suspect is another accident just waiting to happen in Germany. The system as used seems totally open to malicious use.
Culburrarail
Crossing box
Crossing box
 
Posts: 11
Joined: Tue Mar 8, 2016 8:34 am
Location: Culburra Beach N.S.W. Australia

Re: Collision in Germany

Unread postby hmmueller » Thu Mar 10, 2016 10:25 am

Well - here we are again: Someone who does not have the slightest idea about a system ... (the following sentence shows that you never looked up what PZB90 is and does
Culburrarail wrote:... how the PZB90 systems allowed a green light at Kolbermoor to remain active...
) ... knows with absolute certainty that "this could never happen here":
Culburrarail wrote:This could never happen here and I suspect is another accident just waiting to happen in Germany. The system as used seems totally open to malicious use.

The question, of course, remains why nevertheless there occurred are all those incidents and accidents "here" - see the reports of your local safety authority (according to Directive 2004/49/EC, Art. 1(d) if your country is a member of the EU; or the corresponding authority elsewhere) ... that must be a real mystery to you. The reason is of course that all safety systems are heavily interlocked and cross-referencing systems of rules and technology with explicit and implicit compromises regarding human behaviour, technical feasibility, and efficiency of the rail system. Thus, just picking one rule or feature and then declaring that that system "seems totally open to malicious use" is absolutely useless and pointless.

Still, it's nice to have such people around so that I can continue tell stories of how laymen try to get a slight grasp of this sort of system problems ...

H.M.
==== Austrian signal boxes and interlockings (German and English), with some looks beyond ====
User avatar
hmmueller
Crossing box
Crossing box
 
Posts: 96
Joined: Sat Feb 23, 2013 8:24 pm
Location: Bavaria, Germany

Re: Collision in Germany

Unread postby Mike Hodgson » Thu Mar 10, 2016 10:30 am

Culburrarail wrote:The system as used seems totally open to malicious use.


I don't think there is the slightest suggestion that the Fahrdienstleiter deliberately caused this crash.
Signalmen get tired, make honest mistakes, etc, just like everybody else and although some have been blamed for accidents in the past for acts or omissions seen as negligent, cases in which a signalman has intentionally caused a derailment or collision are extremely few and far between. The only such case that comes to mind was Conington South, and the man in question may have had mental health problems.

In any case, I doubt it would be possible to design a transport system that could not be abused maliciously by somebody, whether that be an operator, system designer/builder or maintenance technician, quite apart from what passengers or the general public might do maliciously.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2485
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Collision in Germany

Unread postby Richard Lemon » Thu Mar 10, 2016 10:54 am

"This could never happen here" - presumably meaning the UK.

Signalmen always have the authority to authorise a driver to pass a signal at danger - it is the signalman's task to physically check that all points etc., are in the correct position and that the line ahead is not occupied (unless specifying the fact).

Interlocking is fallible and mistakes happen (Moreton-on-Lugg comes to mind!).

Richard
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 378
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: Collision in Germany

Unread postby John Hinson » Thu Mar 10, 2016 11:07 am

Richard Lemon wrote:Signalmen always have the authority to authorise a driver to pass a signal at danger - it is the signalman's task to physically check that all points etc., are in the correct position and that the line ahead is not occupied (unless specifying the fact).

and, it should be added, drivers should then proceed cautiously at a speed able to stop short of an obstruction.

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6859
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Collision in Germany

Unread postby Frank » Thu Mar 10, 2016 11:49 am

Hello Mike,

In any case, I doubt it would be possible to design a transport system that could not be abused maliciously by somebody, whether that be an operator, system designer/builder or maintenance technician, quite apart from what passengers or the general public might do maliciously.


you pointed it out.

For Example the PZB90 (Indusi) here.....Signal at danger or any failure remains the Track side Magnet ative.
But how do you in a failure get a Train passing by without the apply of the Brake systen and more problematic the Time it costs (specially Freight trains)
to establish the Presure on the Brake system again.
Also it is very unconfortable to passengers to get a emergency Brake.
So for that the override was built in the PZB90.


What a Mistery for me (and all) is that the Signal Man didn`t use the emergency Call on the Train Radio.
Two Trains facing on a Single Line gives you audience at the Big Boss but that is much better then a accident.


In 1975 on that Line there where also a great Accident (Warngau 1975) , but at that Time there was no Line Block between Schaftlach and Warngau and also no Train Radio.The Line is curved and so like Bad Aibling the Train Drivers had no Chance, to see the other Train early enought.
User avatar
Frank
Rest-day relief
Rest-day relief
 
Posts: 501
Joined: Mon Nov 26, 2007 10:18 pm
Location: Dortmund-Rhine Westphalia- Germany

Re: Collision in Germany

Unread postby Culburrarail » Thu Mar 10, 2016 9:26 pm

I am in Australia and live close to a single track system with quite heavy traffic

"
In any case, I doubt it would be possible to design a transport system that could not be abused maliciously by somebody, whether that be an operator, system designer/builder or maintenance technician, quite apart from what passengers or the general public might do maliciously.

I am a Logician and could never resist a challenge I will post here an interlocked system that will be foolproof.
Culburrarail
Crossing box
Crossing box
 
Posts: 11
Joined: Tue Mar 8, 2016 8:34 am
Location: Culburra Beach N.S.W. Australia

Re: Collision in Germany

Unread postby davidwoodcock » Thu Mar 10, 2016 9:39 pm

The real challenge is to design an interlocked system that is not only foolproof in normal function mode but which remains foolproof when it has to be overridden to enable trains to pass in failure mode.
davidwoodcock
Rest-day relief
Rest-day relief
 
Posts: 950
Joined: Tue Jul 12, 2011 7:58 am
Location: Champlon, Belgique

Re: Collision in Germany

Unread postby Culburrarail » Thu Mar 10, 2016 11:51 pm

Railway Block Operating System
This document explains how a system of levers, interlocks and signals control a section (Block) of single track and how it is operated. The single track connects to two stations (A & B) with two tracks (Up and Down).

1/. A signal man at Station A has a train ready to depart for Station B. He operates the Route selection button. This sends a message (Bells) to the operator at station B requesting the Block.

2/. The signal man at Station B can accept or reject this request. I assume here that he accepts and operates his Route Lever which causes two things to happen.
Firstly the Switch (Points) at station B operate setting the rails to route the single track into the correct track (UP) within Station B.
Secondly it engages interlocks which have three functions.
i/. The Switch (Point) is locked, mechanically or electrically.
ii/. The Signal protecting the Block entry point (Down) from the Station B is set to Red.
iii/. The Route (UP) for the Switch and Signals at Station A are enabled.

At this point the Block is owned by the Station A operator.

3/. The signal man at Station A now operates the lever on his controls to set the departure track (Up) on to the single track Block. This causes two things to happen.
Firstly the Switch (Point) operates to route the departure track within the Station A Block to the Single Track Block (Up)
Secondly an interlock system operates which also has three functions.
i/. The Switch (Point) is mechanically or electrically locked.
ii/. The Signal controlling station exit is enabled. NB not set to green.
iii/. The Switch (Point) at station B has it’s operating lever disabled.

The route is now set in a fail safe configuration waiting for the signal man at Station A to operate his Signal controls and release the train.

4/. The signal man operates the button that sets the departure signal to green and the train can depart. This button has two other functions.
Firstly the lever controlling the Switch (Points) at Station A is disabled.
Secondly on-board indicators are sent to the train driver.

The train driver now Owns the Block and can now proceed under normal movement controls on to and along the single track sure in the knowledge that there cannot be another train coming at him in the opposite direction.
Axel counters and rear end signal procedures process the train as it proceeds along the single track.

5/. The end of the train arriving at Station B Up platform releases all the interlocks applied to the Block at both stations. The train will now be subject to Station B Block control.

The Single track Block is now in a quiescent state and can accept a train from either direction.

Interlock over-rides will be available for track maintenance etc and procedures for Emergencies will be in place to stop the train.

Stations where Locomotive Run-Around is used will have a different procedure of operation and interlocks.
Culburrarail
Crossing box
Crossing box
 
Posts: 11
Joined: Tue Mar 8, 2016 8:34 am
Location: Culburra Beach N.S.W. Australia

Re: Collision in Germany

Unread postby JRB » Thu Mar 10, 2016 11:52 pm

Exactly, a truly foolproof system would stop the entire line sometimes.
JRB
Double-manned box
Double-manned box
 
Posts: 3216
Joined: Sun Dec 9, 2007 10:12 pm
Location: GWR

Re: Collision in Germany

Unread postby Culburrarail » Fri Mar 11, 2016 1:56 am

http://www.railway-technology.com/proje ... ight-line/


Trackside signals are now extinct on the Pilbara Rail system of Hamersley and Robe River Railway’s following their replacement with in-cab signaling. It relays the limit of the driver's authority using a display in the cab.
Culburrarail
Crossing box
Crossing box
 
Posts: 11
Joined: Tue Mar 8, 2016 8:34 am
Location: Culburra Beach N.S.W. Australia

Re: Collision in Germany

Unread postby Richard Lemon » Fri Mar 11, 2016 7:51 am

Culburrarail wrote "Trackside signals are now extinct on the Pilbara Rail system of Hamersley and Robe River Railway’s following their replacement with in-cab signaling. It relays the limit of the driver's authority using a display in the cab."

This is as it is here in Wales on the Cambrian - ETCS!
User avatar
Richard Lemon
Branch line box
Branch line box
 
Posts: 378
Joined: Mon Feb 4, 2008 5:43 pm
Location: Chepstow

Re: Collision in Germany

Unread postby kbarber » Fri Mar 11, 2016 9:13 am

I can only imagine a logician would have no objection if a failure resulted in their sitting on a train that was completely unable to move for several hours because of the 'foolproof' block system; presumably their manager (if they were going to work) would also be a logician and likewise have no objection.

Such systems already exist and, with varying degrees of human involvement (i.e. provision or otherwise of trainstop/ATP facilities) safely and successfully control many thousands of train movements each week.

If multi-hour delays because of trivial failures are unacceptable - the real-world situation, I would contend - override facilities of some kind have to be provided, together with procedures for their initiation. Such facilities and procedures inevitably rely more heavily on the human element, which of course we know to be the most fallible part of the entire system. As it was a failure in this part of the system that resulted in the collision, I'm afraid we have to accept it is pretty much impossible to achieve absolute safety without also accepting absolute paralysis. This is the calculation every one of us makes when deciding to get out of bed in the morning (will we trip over our own slippers in our half-asleep state) or go to the shops (will we, in a moment of distraction, step out in front of a bus).
User avatar
kbarber
Rest-day relief
Rest-day relief
 
Posts: 550
Joined: Tue Nov 13, 2007 7:12 pm
Location: London

Re: Collision in Germany

Unread postby Culburrarail » Fri Mar 11, 2016 10:14 am

The real quesriom that needs to be asked is why the Signal blocking system did not cause the trains to be halted. The two trains started their final journey at almost the same time both travelling approximately 2.5km. The train from Bad Aibling after passing the Red signal under the Zs 1 Call on had made a stop at Bad Aibling Kurpark. The signal man had had 3 minutes to recognize his mistake before the trans started their last movement. Yes question why he did not use the emergency stop process. Yes question how he decided to use the Zs 1.
BUT
the Bavarian system relies upon the Signal Block System to prevent head on collisions. This did not work. Had the system as used caused the Switch(Points) at Bad Aibling to be locked in favour of the Kolbermoor train then when the signalman went to release his late train onto the single track he would have had to use TWO over-ride actions and this would most probably made him cancel the Kolbermoor to Bad Aibling route he had previously set.
The crash site is about 2.5KM from Kolbermoor and a similar distance from Bad Aibling Kurpark.
No one has suggested that there was any trivial equipment failure.
How many other trains was the Signal man controlling at this start of the morning rush hour?
Culburrarail
Crossing box
Crossing box
 
Posts: 11
Joined: Tue Mar 8, 2016 8:34 am
Location: Culburra Beach N.S.W. Australia

PreviousNext

Return to Signalling - overseas

Who is online

Users browsing this forum: No registered users and 2 guests