• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Derailment near Watford Junction (16 September 2016)

Status
Not open for further replies.

Clip

Established Member
Joined
28 Jun 2010
Messages
10,822
As previously suggested. Better Mapping technology is needed for railways to avoid these situations.

Good advertising but would any sort of laser mapping have been able to forecast a large month of mainly dry hot weather followed by a very large downpour of rain which may have disturbed the ground and made this landslide?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Zoidberg

Established Member
Joined
27 Aug 2010
Messages
1,270
Location
West Midlands
From the RAIB announcement of it investigating at

https://www.gov.uk/government/news/derailment-and-collision-watford-tunnel

At about 06:55 hrs on Friday 16 September 2016, a southbound passenger train was travelling at about 70 mph (112 km/h) on the up slow line of the West Coast Main Line, when the leading carriage was derailed by a landslip just before entering the northern portal of Watford tunnel. The train stopped in the tunnel, partly blocking the adjacent line. About two minutes later, it was struck a glancing blow by a northbound passenger train which did not derail. Two passengers suffered minor injuries. Both trains suffered significant damage, including the cab to saloon door on the derailed train becoming jammed.

The southbound train was the 06:19 hrs service from Milton Keynes to London Euston, comprising two four-coach class 350 electric multiple units; a total of eight coaches. The northbound service was the 06:34 hrs service from London Euston to Birmingham and was formed by one four-coach class 350 unit. The two trains were carrying more than 150 passengers who were transferred onto another train which took them from the tunnel at about 10:40 hrs.

The driver of the southbound train used an emergency facility on the train’s in-cab radio to alert other train drivers. The driver of the northbound train received the emergency stop message via his in-cab radio and applied the brake, slowing the train from 80 mph (129 km/h) to 32 mph (51 km/h) before the collision.

The landslip material which caused the derailment had slid from the steep-sided, deep cutting on the tunnel approach after very heavy rainfall in the two hours immediately before the accident. The preceding southbound train had passed the site without incident at 06:43 hrs. The sides of this cutting were being strengthened at the time of the accident, but this work had not yet reached the location where the landslip occurred.

Could have been so much worse without the emergency comms facility.
 

Peter Mugridge

Veteran Member
Joined
8 Apr 2010
Messages
16,334
Location
Epsom
It has been published on a Yahoogroup this morning by a member of LM staff quoting official internal sources that:

* 350 117 is now back in service.

* 350 233 will be removed from Northampton tomorrow on four low loaders to go back to the Siemens factory for assessment and possible repair.

* 350 264 will be removed from Watford in the next few days by four low loaders, also for the Siemens factory, and is most likely a write off to be withdrawn.
 

D365

Veteran Member
Joined
29 Jun 2012
Messages
12,249
* 350 264 will be removed from Watford in the next few days by four low loaders, also for the Siemens factory, and is most likely a write off to be withdrawn.

Withdrawn and scrapped, or withdrawn and rebuilt?
 

XCTurbostar

Established Member
Joined
13 Sep 2014
Messages
2,134
I think its highly unlikely that a 350 will be scrapped. The need for it is very high.
Thanks,
Ross
 

sng7

Member
Joined
22 Oct 2013
Messages
167
Location
Edinburgh
Even if one is written off it doesn't mean it still can't return see 156478's imminent return as proof of that.
 

Peter Mugridge

Veteran Member
Joined
8 Apr 2010
Messages
16,334
Location
Epsom
Withdrawn and scrapped, or withdrawn and rebuilt?

Even if one is written off it doesn't mean it still can't return see 156478's imminent return as proof of that.


I do suspect it may be rebuilt eventually, but the posting specifically stated that the cost of repair would be greater than the value of the unit so it could go either way. Apparently while it has been parked up at Watford yard, it's not just the exterior that has been covered in graffiti but the inside as well, plus a lot of the cab fittings have been stolen.

Makes you wonder about the security and CCTV given that the yard is literally opposite the platforms...
 

Andrew1395

Member
Joined
30 Sep 2014
Messages
627
Location
Bushey
The exterior will be a vinyl that would be replaced anyway? Stick it on the Abbey branch. You only need one cab with controls and the guard can control the reverse.
 

LNW-GW Joint

Veteran Member
Joined
22 Feb 2011
Messages
21,127
Location
Mold, Clwyd
The article says the collision between the two trains happened approx. 2 minutes after the derailment of the first.
It then mentions the emergency driver communications that allowed the northbound train to slow significantly before the impact.
What I don't quite understand is whether the signalling system (brand new at Watford) was able to detect the derailment and turn signals to danger automatically, irrespective of driver action.
The northbound train must have been near Watford Jn station when the derailment happened.
I'm just interested to know how "intelligent" the modern signalling system is with an incident of this kind.
Otherwise everything depends on driver actions, which in this case appear to have prevented a much more serious collision.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
7,003
Location
Torbay
I'm not clear how a signalling system could detect a derailment where not all axles were off the road.

The signalling system itself couldn't help, not using normal train detection methods. I'd like to know if any sensors on the fairly modern train could detect a derailment through unusual forces being experienced however. It can't be beyond the realms of possibility, if such an event could be detected on board, that the alarm could be relayed automatically through GSM-R, and could initiate emergency stops for other traffic in the vicinity.
 

LNW-GW Joint

Veteran Member
Joined
22 Feb 2011
Messages
21,127
Location
Mold, Clwyd
I'm not clear how a signalling system could detect a derailment where not all axles were off the road.

No, but the train itself would know all was not well and can report problems remotely.
Does the driver emergency activation also inform the signaller, so he/she can take action?
Just clutching at straws really.
The idea of an inevitable collision taking 2 minutes to unfold is quite chilling, like the Thorpe accident in 1874.
 

Bletchleyite

Veteran Member
Joined
20 Oct 2014
Messages
105,095
Location
"Marston Vale mafia"
No, but the train itself would know all was not well and can report problems remotely.

Yes, but the problem is false positives grinding the network to a halt on a daily basis.

Does the driver emergency activation also inform the signaller, so he/she can take action?

The GSM-R red button automatically informs all other trains directly, AIUI - no need for the signaller to do anything at all.

The idea of an inevitable collision taking 2 minutes to unfold is quite chilling, like the Thorpe accident in 1874.

The thing is, it can take a long time to stop a train from linespeed. The GSM-R red button was activated, and the train was already slowing. This is why the collision's severity was reduced substantially.

Even at very low speed a train has a lot of kinetic energy.
--- old post above --- --- new post below ---
The signalling system itself couldn't help, not using normal train detection methods. I'd like to know if any sensors on the fairly modern train could detect a derailment through unusual forces being experienced however. It can't be beyond the realms of possibility, if such an event could be detected on board, that the alarm could be relayed automatically through GSM-R, and could initiate emergency stops for other traffic in the vicinity.

Possibly - but the driver did do exactly this using the "red button", hence the reduced impact of the collision. At full linespeed the collision could have been far worse.

Could the train have detected it more quickly? Possibly, but you're talking a couple of seconds I expect. The driver and guard (remember, these are double manned) will notice a problem quite quickly.
 
Last edited:

SpacePhoenix

Established Member
Joined
18 Mar 2014
Messages
5,491
Could tilt sensors be added to trains so that if a tilt of more then a certain amount is detected, an emergency GSM-R call gets made? Probably won't catch every derailment but would hopefully catch the worst ones
 

mikestone1952

Member
Joined
12 Feb 2011
Messages
285
The report on the 350s did state that they are to receive four more 319s.
;
My thoughts are that may be two replacements plus two to cover the 350s ordered for Bromsgrove but already diagrammed elsewhere?
 

theageofthetra

On Moderation
Joined
27 May 2012
Messages
3,565
Location
Beckenham
The speed of reaction of the driver in hitting the GSMR Stop should be commended. Every second of delay would have increased the speed and severity of the subsequent collision.
 

Bletchleyite

Veteran Member
Joined
20 Oct 2014
Messages
105,095
Location
"Marston Vale mafia"
The report on the 350s did state that they are to receive four more 319s.
;
My thoughts are that may be two replacements plus two to cover the 350s ordered for Bromsgrove but already diagrammed elsewhere?

Don't know, but it's to LM's credit that they are doing this rather than just short-forming and saying "suck it up, we're using all our trains" as I think some TOCs might, naming no names :)
--- old post above --- --- new post below ---
The speed of reaction of the driver in hitting the GSMR Stop should be commended. Every second of delay would have increased the speed and severity of the subsequent collision.

Agreed.
--- old post above --- --- new post below ---
Could tilt sensors be added to trains so that if a tilt of more then a certain amount is detected, an emergency GSM-R call gets made? Probably won't catch every derailment but would hopefully catch the worst ones

That kind of thing could work; another thing that could would be to have some kind of "trip cable" between coaches that would break easily and trigger an alarm if broken, possibly well before the actual inter-coach cables fail, and some kind of shock sensors.
 

najaB

Veteran Member
Joined
28 Aug 2011
Messages
32,331
Location
Scotland
That kind of thing could work; another thing that could would be to have some kind of "trip cable" between coaches that would break easily and trigger an alarm if broken, possibly well before the actual inter-coach cables fail, and some kind of shock sensors.
That's adding a lot of complexity - read cost to install and maintain - to deal with a very low probability event. If anything, spend the money on monitoring the infrastructure to stop the event happening in the first place.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
7,003
Location
Torbay
That's adding a lot of complexity - read cost to install and maintain - to deal with a very low probability event. If anything, spend the money on monitoring the infrastructure to stop the event happening in the first place.

That would be a good investment. Too much complexity could lead to unreliability if the normal running of trains requires so many safeguards and so much real time sensor monitoring on board, and the default response is not just to stop this train as in the the traditional behaviour of continuous brakes but, by extending the automatic response, to stop all trains in the vicinity. The system would have to be ultra-reliable and finely adjusted to avoid false positives. Unlike the highly dynamic on board environment, the infrastructure (hopefully) moves around to a lesser extent so technically it should be easier to monitor. If the infrastructure starts to move in unexpected ways I'd say alarms should be relayed automatically to signalling where that's possible, changing signals to red and sending emergency stop messages to trains in the area.
 

carriageline

Established Member
Joined
11 Jan 2012
Messages
1,897
The GSM-R red button automatically informs all other trains directly, AIUI - no need for the signaller to do anything at all.

Wow, Of course there is!! The REC call is primarily for the signallers benefit (in the sense that the driver pushes the red button, a very high priority call is made to the signaller to inform THEM of the circumstances. All other trains get some sort of alarm(?) in the cab telling them to stop, and can hear and partake in the call if need be)

It's absolutely no different from a member of staff ringing on the SPT to report a derailment. Signals MUST go back to danger to protect it and any attempts to stop trains must be made. GSMR is a good system, but it doesn't always work like it should. You can't just sit there and hope trains come to a stand with green signals leading into a derailed train!!!

The signaller may also need to get the juice off in an emergency etc. The signallers are still the guardians of the railway, I think people underestimate what we do. As technology moves on, the job doesn't change. Ok, ARS may be moving trains for us, but the potential areas of control are now much, much larger than before. ARS moves trains, it isn't programmed to stop them. As soon as something adverse happens, it gets turned off and back to "hands on"
 

Bletchleyite

Veteran Member
Joined
20 Oct 2014
Messages
105,095
Location
"Marston Vale mafia"
Sorry, I didn't mean to imply the signaller had nothing to do, quite the opposite as you say. The point I was making was that the signal to stop goes to all trains automatically rather than just to the signaller who then has to relay it.

I must admit to being surprised it doesn't automatically throw all signals to danger as well, though.
 

tsr

Established Member
Joined
15 Nov 2011
Messages
7,400
Location
Between the parallel lines
Wow, Of course there is!! The REC call is primarily for the signallers benefit (in the sense that the driver pushes the red button, a very high priority call is made to the signaller to inform THEM of the circumstances. All other trains get some sort of alarm(?) in the cab telling them to stop, and can hear and partake in the call if need be)

Yes, it most definitely is an alarm! In my experience it will also cut out audio from cab-to-cab / radio conversations and default to the emergency call, often at a heightened volume. Not sure how that may vary by traction or GSM-R equipment, including software.

Although it seems like it, it's not so much communication "between trains" as "between incident train and signaller, also created as a radio call in active cabs everywhere in a given radius". My understanding is that the equipment on the train involved in the incident doesn't really know or care if it's broadcasting between itself and other trains, though the radio call would involve any if need be.
 
Status
Not open for further replies.

Top