Derailment of empty passenger train at Paddington causes major disruption - 16 June 2016

[whole diary]

[ChrisB]

To answer (either in this thread or the LTV^▸ (London [and] Thames Valley) disruptions thread) a couple of queries - which train was it and other details - this from uk.rail newsgroup

3H52 ECS^▸ (Empty Coaching Stock) Reading to Paddington normally hangs over at Old Oak Reception
 Sidings for 20 minutes before proceeding to PAD^» (Paddington (London) - next trains) platform 1, due 1754.
 Yesterday it skipped the hangover and so approached platform 1 20 minutes
 early, i.e. well before 1G60, the 1742 HST^▸ (High Speed Train) to Cheltenham Spa was due to
 leave platform 1.
 
 Now how could that sequence of events happen?

[ChrisB]

Current alternative routes and ticket acceptances can be found on this link
http://www.nationalrail.co.uk/static/images/content/GWR_alternative_routes.pdf

Elsewhere, the London Fire Brigade have tweeted 'no one hurt last night' and NR^» (Network Rail - home page) tweeted that the train will not be moved until 2200 tonight to allow as much access to the station as possible this evening.

[chrisr_75]

A Network Rail spokesman said the driver of the derailed train had not been hurt and it was “too early” to speculate about why the train had jumpped [sic] the red light.

No sympathy for the City worker though. Whatever happened, it is extremely unlikely to be anything to do with whoever he 'spoke' to.

Indeed, good to see BTP^▸ (British Transport Police) taking some action with these idiots who take out their frustrations on members of staff who probably know little more than the passengers they're trying to help. Yes it's annoying but it is no reason to abuse a fellow human being.

Equally encouraging that most quoted people's attitude was not of consternation and anger, but resignation Along the lines of 'these things happen, nothing I can do about it so I will just go with the flow'.

[SandTEngineer]

The link line only leads to platform 1 in any case, which looks very limited in its usefulness. I'm sure it could contribute more to operations if its connections were changed. Is there any plan for that (I've not seen any)?

Crank up the Way-back machine...
There used to be a set of points giving access from the Royal Oak sidings to line 1 at signal SN26, which meant that the Royal Oak sidings could access platforms 2-5 via line 1, as well as platform 1 via the link line.  After the post-Ladbroke-Grove remodeling, when all sorts of bells and whistles (TPWS^▸ (Train Protection and Warning System), SPAD^▸ (Signal Passed At Danger) detection and mitigation etc) were added to the interlocking, some "space" was needed in the computer logic; so these points and their associated routes were removed.

Now that the SSI^▸ (Solid State Interlocking) (solid-state-interlocking) has been replaced by a more-processing-power Smartlock, there's "room" to re-instate these routes and points.  The points are shown as being re-instated (although clipped & padlocked) at Christmas 2016; I don't know when the routes over them will be re-commissioned.

From what I have been hearing through another related project I'm working on at present its likely to be commissioned at Easter or Xmas 2017. There is a big Solid State Interlocking data rewrite update at Xmas 2016 and thats taxing enough to commission without adding other work to it.

[TaplowGreen]

A Network Rail spokesman said the driver of the derailed train had not been hurt and it was “too early” to speculate about why the train had jumpped [sic] the red light.

No sympathy for the City worker though. Whatever happened, it is extremely unlikely to be anything to do with whoever he 'spoke' to.

Couldn't agree more - his behaviour is as despicable and inexcusable as the situation in which he and tens of thousands of others found themselves in through no fault of their own - doesn't excuse threatening violence however.

[SandTEngineer]

...err no its not.  The trap points did the job they were designed to do which is to divert the train away from a running line.  The track under the train is the connection from the sidings to Platform No.1 and the line and points immediately to the right are the running connection from Line 1 to Platform No.1 and the derailed train is not foul of that.  In any case when a train is derailed on trap points a special detector mechanism called a 'Track Circuit Interrupter' breaks a circuit to place to danger all signals for train movements that could potentially conflict.
...

You may be right, in the sense that trains aren't wide enough to actually touch it. There is the usual difficulty of judging perspective at oblique view angles. But I still think it's closer the the protected line than the "design" of the catch points would call for. And yes, the distance it ran is clearly the reason for its ending up there, but that too is part of that "design".

In fact, on reflection, it could have been far worse. That stanchion took a lot of energy out of the train, so if it (the train) had been just a fraction of a metre less off line and hit a glancing blow it could have got a lot further and been pushed back outwards. And, given that the obvious reason for the signal to be red is that Platform 1 is still occupied, or its train is just coming out, a collision is quite possible. Given how busy Paddington is, automatically setting signals to danger leaves a significant residual risk. 

I made my statement based upon the evidence in the photograph of where the wheels of the first bogie ended up.  The train is not foul of the Line 1 to Platform No.1 connection.  However, I can only give my personal opinion here as I do live some 250 miles from the incident site and can only express that statement from what I can see in the photographs that have been published so far.

I also think that train speed at time of collision with the OLE^▸ (Overhead Line Equipment, more often "OHLE") mast was quite low as there is only limited damage to the fibreglass front end.  That design of OLE mast seems to be quite lightweight so probably quite flexible when hit.  Some 40 years ago I remember attending a high speed derailment that hit two such masts and there wasn't much left of them afterwards (now where is Electric Train when you need him ).

[LiskeardRich]

The points to derail are reportedly some 100m back from the gantry. No obstacles permitted within 50 m of such points.

[ChrisB]

Seems as if that rule needs updating?...

[SandTEngineer]

Seems as if that rule needs updating?...

Maybe, but then you will never be able to protect against every eventuality, but one way to achieve it would be not to run any trains, thats why these things are subject to Risk Assessment and subject to the well practised ALARP (As Low As Reasonably Practicible) principle applied by all safety critical industries.

Before I put my 'Approval' signature against new or altered signalling layout plans I need to ensure myself first that due process has been demonstrated. Checking to ensure the output of the layout risk assessment process has been thouroughly carried out and ALARP has been achieved is a key component of that.

[NickB]

Still a mess at Paddington this evening with a lot of cancellations. BTP^▸ (British Transport Police) on the concourse along with increased GWR^▸ (Great Western Railway) staff.

[TaplowGreen]

Still a mess at Paddington this evening with a lot of cancellations. BTP^▸ (British Transport Police) on the concourse along with increased GWR^▸ (Great Western Railway) staff.

I bailed out of work at 4pm in anticipation of that being the case - got back to Paddington to see most trains delayed or cancelled but managed to get on an Oxford stopper at about 1630 - only 3 carriages so sardines and left about 15 mins late but got to Slough and the connection for Taplow was right behind us so I was lucky - not so lucky are people going longer distance I fear - Friday evening is bad enough already in terms of overcrowding without loads of services being cancelled - good luck all!

[grahame]

I remain fully aware of two thread - and aware of repeated notifications to that effect. Somehow the two thread, same topic stuff feels trivial tonight - see http://www.firstgreatwestern.info/coffeeshop/index.php?topic=17162.msg197342#msg197342 . I'll sort it in the morning.  In the meantime, please feel free to carry on posting in this thread.

[tomL]

As a 'casual' traveller I'm not used to being caught up in major disruption like this (guess I've been lucky) but the mood did seem relatively calm travelling from Oxford to Swindon (with an extensive wait at Didcot for a train west). Really couldn't fault GWR^▸ (Great Western Railway) people on the ground.

- When an 'express' to Reading arrived the DM at Oxford endorsed my ticket to go to Reading, didn't need it in the end though as a Didcot call was added last minute.

- Staff at Oxford, Didcot and Swindon were out and about checking up on us all personally in waiting rooms and on platforms, as opposed to sticking to over air announcements.

Managed to help a few people out with plans using the real time signalling maps on sites like OpenTrainTimes too. People were happy to know "train has started its journey, so its worth waiting for now" and the like.  It's times like these knowing the extra information really helps.. 

An interesting experience none the less, hope it all clears up soon though. 

// end of two cents.

[SandTEngineer]

The points to derail are reportedly some 100m back from the gantry. No obstacles permitted within 50 m of such points.

I've done a check on the latest Signalling Scheme Plan I have access to and the distance from the Trap Points to the OLE^▸ (Overhead Line Equipment, more often "OHLE") support is about 60m.

[Electric train]

Is that some OHLE wires resting on the roof of the train? 

Those are the Return Conductors (RC) and Earth Wire (EW).  The RC is connected to the running rails at specific points and insulated to ensure the current stays in that wire, the EW is the main bonding conductor.

As a general note the protection system (trip) acts very quickly it has to operate in less than 200 mS but typically the system used on NR^» (Network Rail - home page) acts in less than 100 mS