Upcoming Games

No games to display

Add a game

Upcoming Events

17/11/2018 12:00:00
01/06/2019 12:00:00

Who's Online

clive, GeoffM, jem771, JohnMiller, MikeBR152, pedroathome (6 users seen recently)

All RAIB Reports

You are here: Home > Forum > Miscellaneous > The real thing (anything else rail-oriented) > All RAIB Reports

Page 16 of 16

All RAIB Reports 10/10/2018 at 11:01 #112564
BarryM
Avatar
2030 posts
Gee, what a report, 50 pages long!
Barry, Sydney, New South Wales, Australia
Log in to reply
All RAIB Reports 11/10/2018 at 00:26 #112569
RainbowNines
Avatar
253 posts
BarryM in post 112564 said:
Gee, what a report, 50 pages long!
About par for the course for serious incidents. Sandilands ran to 175 pages.

As for this one, it was an absolute catalogue of errors and misunderstandings by all involved that to all intents and purposes seemed doomed to result in fatalities. That it didn’t was almost entirely down to luck.

Log in to reply
All RAIB Reports 11/10/2018 at 16:18 #112578
bugsy
Avatar
320 posts
RainbowNines in post 112569 said:
BarryM in post 112564 said:
Gee, what a report, 50 pages long!
About par for the course for serious incidents. Sandilands ran to 175 pages.

As for this one, it was an absolute catalogue of errors and misunderstandings by all involved that to all intents and purposes seemed doomed to result in fatalities. That it didn’t was almost entirely down to luck.
Yes. It’s unbelievable that not one of those in authority actually requested that the current be turned off!
When you look at how close the live rail is to the vertical steps that some of the passengers use to detrain, it's a miracle that none of them stepped on it and I don’t recall the report mentioning that anybody told them not to. Not only that, it was wet and dark and there's hardly any room between the live rail and the nearby bushes.
Very lucky, as RainbowNines said.


Post has attachments. Log in to view them.
My goodness, that was difficult, but I managed it.
Log in to reply
All RAIB Reports 11/10/2018 at 16:56 #112581
Danny252
Avatar
1439 posts
bugsy in post 112578 said:
Yes. It’s unbelievable that not one of those in authority actually requested that the current be turned off!
From my reading of the report, I thought the issue was simply that none of those in authority (above the level of the driver and platform staff) were aware that a trackside evacuation was occurring, due to a series of communication failures and incorrect assumptions. The driver's control had convinced themselves he was in the platform, while the signalling staff were pretty much in the dark about what was happening throughout the incident (and were expecting further communication before any evacuation).

Last edited: 11/10/2018 at 16:59 by Danny252
Reason: None given

Log in to reply
All RAIB Reports 11/10/2018 at 17:11 #112582
bugsy
Avatar
320 posts
Danny252 in post 112581 said:
bugsy in post 112578 said:
Yes. It’s unbelievable that not one of those in authority actually requested that the current be turned off!
From my reading of the report, I thought the issue was simply that none of those in authority (above the level of the driver and platform staff) were aware that a trackside evacuation was occurring, due to a series of communication failures and incorrect assumptions. The driver's control had convinced themselves he was in the platform, while the signalling staff were pretty much in the dark about what was happening throughout the incident (and were expecting further communication before any evacuation).
The issue of miscommunication and misunderstanding seems to crop up in a number of the RAIB reports unfortunately. It seems to be down to the fact that the correct procedures are not always followed.
On this particular ocassion, it's a pity that the driver or a member of the platform staff who did know what was happening didn't check.

My goodness, that was difficult, but I managed it.
Log in to reply
All RAIB Reports 11/10/2018 at 22:45 #112587
RainbowNines
Avatar
253 posts
Think it’s all a bit of an unfair reflection on the driver who had clearly been led down the garden path and then hung out to dry if the report is to be believed (fair enough, the signaller SHOULD have been told quicker). By the end he had been subject to abuse from just about every party involved and he probably wasn’t complimentary about himself in his own head either - and that pressure led to a pretty big error which was compounded by total failure of the safety framework, which should protect against errors under pressure (impossible to totally eliminate in safety critical environments). I found the report quite compelling as the investigator just seemed to unveil another cock up in every paragraph!

Fact is that at least one, and by looking at the social media fall out a couple more, of the evacuated passengers decided to stumble into the four foot for an obligatory “disgusted of Tunbridge Wells” photo - in the dark on an icy permanent way *over* the live rail... terrifying.

Log in to reply
All RAIB Reports 12/10/2018 at 07:57 #112590
kbarber
Avatar
1343 posts
Something I was told within days of joining was that to assume anything makes an ASS of U and ME. And so it proves. This would appear to have been a complete mess from start to finish, all based on people thinking they knew what was going on and without thinking to check.

Of course, fragmentation and lack of local knowledge played a part as well. In BR days, even after implementation of the big powerboxes, signalmen would be expected to have some knowledge of the area they controlled. And signalbox diagrams usually made it clear where track circuits extended beyond platforms, and therefore a TC covering a platform being occupied wouldn't always mean the train was wholly int he platform. (What isn't clarified at all is whether the TC in rear was also occupied - something the signaller should have been able to see, but which wouldn't have been visible to a CCF user, leading to that particular bit of confusion.) Also, in BR days everything would have happened via the signalman; the idea that a driver might be receiving instructions from any other source would have been anathema. One major issue seems to have been that there were at least 3 controllers (the TSC & incident controller in South East Control and the controller at ARL), none of whom seemed to be quite clear what any of the others were doing. If ever I saw a recipe for disaster...

Log in to reply
The following user said thank you: BarryM
All RAIB Reports 13/10/2018 at 19:44 #112610
KymriskaDraken
Avatar
619 posts
Danny252 in post 112581 said:
The driver's control had convinced themselves he was in the platform, while the signalling staff were pretty much in the dark about what was happening throughout the incident (and were expecting further communication before any evacuation).
I would have expected the Signalman to go through the "Train an unusually long time in section" procedure and stop another train to find out what was going on. I certainly wouldn't have waited more than five or six minutes with a train stopped for no reason.

Kev

Log in to reply
All RAIB Reports 14/10/2018 at 18:24 #112625
Giantray
Avatar
180 posts
Before privatisation, before mobile phones, before in-cab radios.... a Driver of a Train could only converse with the Signalman. All information was passed to a Driver by the Signalman in nearly all cases. (exception being by notice from station staff or use of Handsignalmen). The advent of Cab-secure radio enabled the Signalman to patch other parties through to Drivers whilst the Signalman monitored the call. Today, all Drivers have company Mobile Phones. Despite constantly being told the Signaller is the first point of call before doing anything else, Drivers are still under pressure from their Companies to avoid delays and thus think it saves delay time contacting their Company directly before ringing in to the Signaller. And what do we see in several RAIB reports, conversations from Control being misinterpreted by Drivers as authority to do something, forgetting to ask the Signaller first! Never happened before, only since Drivers can contact their Control direct leaving the Signaller in the dark!

As for the track circuit in rear of the station being occupied, it wasn't. A six car train can easily fit in between the end of the overlap at the signal in rear of Peckham Rye and the london end of Peckham Rye Platform ( the train involved was a five car 378 unit).

Professional Railwayman since 1981. Railway Historian (SER, LCDR)
Log in to reply
All RAIB Reports 14/10/2018 at 20:53 #112630
JamesN
Avatar
1000 posts
When drivers contact me, at a stand with a fault or whatever the first question I’ll ask them is “have you informed the controlling signaller” - if the answer is no I tell them to terminate the call; speak to the signaller to let them know what’s going on and then call me back.

The only time that sequence of events doesn’t happen is if they’ve contacted me on the move (which I’m not 100% comfortable with in the first place; but it is permitted for them to use GSM-R on the move while running under clear signals) - those instances are usually questions about what they’re doing next or similar during disruption; or their train is full and standing and they don’t want to stop at X.

Log in to reply
All RAIB Reports 14/10/2018 at 22:59 #112632
RainbowNines
Avatar
253 posts
I’m speculating a bit, but I wonder if the largely self contained London Overground network has led to a state where the prevailing approach from drivers is to go to their control first.

I’ve read a lot of accident reports from the days of the Board of Trade through to now (guess we all have our Railway vice...) and taking an affirmative response from the wrong person as permission (to move, detrain, etc) pops up fairly regularly, but maybe it’s more common now. It’s also quite common in air investigations.

I know a couple of the RAIB inspectors are human factor experts, and they often get these cases to investigate. Per Giantray, the multiple management lines must be of concern to them. Safety is everyone’s responsibility, blah blah, but if, as a driver, you can contact your fitter, your control, the signaller, every man and his dog - which one is ultimately responsible for that safe environment? In the cold light of day that’s probably still the signaller, of course - but if the other parties all add tasks to the driver’s workload it’s not that surprising the signaller was forgotten.

Log in to reply
All RAIB Reports 15/10/2018 at 12:53 #112644
Splodge
Avatar
337 posts
At my TOC we don't have company phones - GSMR can be used to contact both control and maintenance if needed, but the signaller will always be our first port of call. Usually he will then liase with control whilst I talk to maintenance to resolve any problems. I have spoken to control during disruption, but that is mostly to keep them informed of a situation whilst the guard is engaged rather than a requirement to.
There's the right way, the wrong way and the railway.
Log in to reply
All RAIB Reports 15/10/2018 at 14:00 #112645
KymriskaDraken
Avatar
619 posts
How does the Rule Book phrase the requirement of a Driver to contact the Signalman if the train stops out of course? If it doesn't say "The Driver must immediately inform the Signalman..." then it should.


Kev

Log in to reply
All RAIB Reports 15/10/2018 at 18:45 #112653
Splodge
Avatar
337 posts
Quote:
If your train is stopped by failure, you must immediately tell the signaller about the circumstances and whether you need an assisting train.
From module M2

There's the right way, the wrong way and the railway.
Log in to reply
The following user said thank you: KymriskaDraken
All RAIB Reports 16/10/2018 at 12:11 #112669
AndyG
Avatar
1771 posts
RAIB has today released its report into extensive track damage between Ferryside and Llangennech, Carmarthenshire, 30 October 2017.

R172018_Ferryside.pdf

I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either.
Log in to reply
All RAIB Reports 18/10/2018 at 17:37 #112731
AndyG
Avatar
1771 posts
RAIB has today released its interim report following the loss of speed restriction data to trains on the Cambrian line, 20 October 2017.

IR012018_Cambrian_TSRs.pdf

{Note:This is only an Interim Report, but an interesting read on safety critical data within computer based signalling systems.}

I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either.
Log in to reply
All RAIB Reports 18/10/2018 at 21:27 #112734
RainbowNines
Avatar
253 posts
A really intriguing one this - seems like the reason for the delay is that it took them the majority of the year to reliably reproduce the state that caused the incident.

The “interim” nature of this report is a result of the strategy adopted by the Branch a couple of years ago to get reports out within 12 months. Where this isn’t possible they’ll release an interim update, which aligns them with the approach of the “senior” AAIB and MAIB.

Log in to reply
All RAIB Reports 19/10/2018 at 09:21 #112739
Danny252
Avatar
1439 posts
I wonder whether there have been similar incidents with such "computerised" train control systems in Europe? It sounds like it will be difficult to guarantee that all components are working correctly, especially if the one that fails (as in this case) isn't deemed safety-critical.

One feature I found intriguing was that the signalling system performs a reset when a movement authority is requested but cannot be granted, followed by (what I took to be) noticeable disruption to the service. It seems strange that a signalling system would consider that to be such an exceptional circumstance - I can't think of any other signalling system where the result of requesting a route that cannot be given isn't simply that the interlocking prevents it, without any wider impact on the system. Given that it happens approximately once a month on the Cambrian, how would it scale on busier parts of the network? Will the system reset itself several times a week during the rush hour if applied to a busy urban area?

Last edited: 19/10/2018 at 09:23 by Danny252
Reason: None given

Log in to reply
All RAIB Reports 19/10/2018 at 10:13 #112741
headshot119
Avatar
2859 posts
I believe the system only rollsovers when a train requests the same movement authority as already issued to another train.

Train A is given MA 1 for a route from S1 to S2.

Oversetting the route for train B doesn't cause the rollover, but train B saying oh can I use MA1 does.

"Platform 1 your Maesteg service. Platform 1A your Aberdare. Platform 2 your London!"
Log in to reply
All RAIB Reports 19/10/2018 at 10:33 #112742
Danny252
Avatar
1439 posts
In what situations would a train request an MA? In RETB that would be the normal procedure, but it sounds like ERTMS is mainly the signaller setting routes for trains. Is it situations where (to use Simsig terminology!) a "new" train appears, so e.g. a train that divides would potentially result in both parts of the train requesting the same MA?
Last edited: 19/10/2018 at 10:34 by Danny252
Reason: None given

Log in to reply
All RAIB Reports 19/10/2018 at 11:29 #112743
headshot119
Avatar
2859 posts
Danny252 in post 112742 said:
In what situations would a train request an MA? In RETB that would be the normal procedure, but it sounds like ERTMS is mainly the signaller setting routes for trains. Is it situations where (to use Simsig terminology!) a "new" train appears, so e.g. a train that divides would potentially result in both parts of the train requesting the same MA?
You've hit the nail on the head, they've obviously done something to improve things as it used to be a daily occurrence with the splits at Mach.

"Platform 1 your Maesteg service. Platform 1A your Aberdare. Platform 2 your London!"
Log in to reply
All RAIB Reports 01/11/2018 at 13:07 #113037
Edgemaster
Avatar
22 posts
Today RAIB announced that they will not produce a full report into the derailment at Stonehaven, Aberdeenshire.

https://www.gov.uk/government/news/derailment-at-stonehaven-aberdeenshire

In particular:
Quote:
The evidence which they collected shows that the points had been set, and the relevant signal had been cleared, for the train to move from one line to the other. After the cab of the train had passed it, the signal reverted to danger and the points moved under the train, causing two carriages to derail.
...
The crossover is only used by empty trains travelling at low speed and, in common with many similar crossovers that are only used for shunt moves, the points are not locked by the presence of the train. For this reason, and given that that all other railway equipment appears to have operated as designed, the RAIB has concluded that the potential for safety learning is insufficient to justify further investigation.

Log in to reply
All RAIB Reports 01/11/2018 at 19:17 #113047
Hap
Avatar
190 posts
Edgemaster in post 113037 said:
Today RAIB announced that they will not produce a full report into the derailment at Stonehaven, Aberdeenshire.

https://www.gov.uk/government/news/derailment-at-stonehaven-aberdeenshire

In particular:
Quote:
The evidence which they collected shows that the points had been set, and the relevant signal had been cleared, for the train to move from one line to the other. After the cab of the train had passed it, the signal reverted to danger and the points moved under the train, causing two carriages to derail.
...
The crossover is only used by empty trains travelling at low speed and, in common with many similar crossovers that are only used for shunt moves, the points are not locked by the presence of the train. For this reason, and given that that all other railway equipment appears to have operated as designed, the RAIB has concluded that the potential for safety learning is insufficient to justify further investigation.
All I will say on this is, watch the media closely. It is perhaps about to kick off.

HAP
Log in to reply