Ickenham overrun/SPAD, 12th March 2022

[Tangent]

I am surprised that this didn't end up as a RAIB investigation:

The driver fell asleep between Ruislip and Ickenham, failing to initiate the brakes in time to bring the Tube to a stop. This resulted in the train overrunning the platform at Ickenham station and going through a red signal without permission.

Evening Standard article today:

Tube driver overran platform by 50m after falling asleep at controls

Unions raise concerns about ‘punishing’ shift patterns for Tube drivers

www.standard.co.uk

TfL response to FOI request, November 2022

The correct operational procedures were implemented following the incident and welfare checks undertaken with the train operator who was relieved from duty whilst further fact finding was undertaken. There was a delay associated with the incident but no injury or damage occurred as a result.
In accordance with our obligations under Data Protection legislation, information on what action was taken against the driver has been withheld as required by section 40(2) of the FOI Act.

FOI request detail

FOI request detail

tfl.gov.uk

 

[bramling]

I am surprised that this didn't end up as a RAIB investigation:

Tube driver overran platform by 50m after falling asleep at controls

Unions raise concerns about ‘punishing’ shift patterns for Tube drivers

www.standard.co.uk

FOI request detail

FOI request detail

tfl.gov.uk

There is actually some potential for collision on LU if a driver disregards a starting signal at full speed, as in some places overlaps on starting signals are of nominal length only, though the way things are designed extra protection is provided if there is potential for a train ahead to *routinely* stop in a location protected only by a nominal overlap.

Fatigue remains the big elephant in the room. It is very difficult to reconcile an industry which has always worked with extreme and erratic shift patterns with natural human fatigue. It’s something to remember when people moan about how highly train drivers are paid, many people might fancy the salary but either wouldn’t want or couldn’t cope with the shift patterns, and the self-discipline they require including when not at work.

 

[12LDA28C]

Fatigue remains the big elephant in the room. It is very difficult to reconcile an industry which has always worked with extreme and erratic shift patterns with natural human fatigue. It’s something to remember when people moan about how highly train drivers are paid, many people might fancy the salary but either wouldn’t want or couldn’t cope with the shift patterns, and the self-discipline they require including when not at work.

Indeed, and yet the industry (or rather the Government via the DfT and RDG) wants to move towards even more punishing / fatiguing rosters with even greater movement off spare, late notice movement of Rest Days etc, all in the name of 'efficiency'.

 

[endicomark69@g]

Indeed, and yet the industry (or rather the Government via the DfT and RDG) wants to move towards even more punishing / fatiguing rosters with even greater movement off spare, late notice movement of Rest Days etc, all in the name of 'efficiency'.

If it were possible to "Make" Government and the DfT exist on similar shift patterns they'd be a lot less keen on inflicting increases in misery on others. Or spouting nonsense in press releases about "efficiency" when they have literally no idea what the word means.

 

[TomCrame]

There is actually some potential for collision on LU if a driver disregards a starting signal at full speed, as in some places overlaps on starting signals are of nominal length only, though the way things are designed extra protection is provided if there is potential for a train ahead to *routinely* stop in a location protected only by a nominal overlap.

There's very little potential in such a circumstance. The normal design process is to calculate the full speed overlap of a starting signal, and if there is no credible collision risk within that calculated overlap (i.e. no berth or confliction within it) then a short overlap for 35km/h may be provided. If there is a confliction then things are reverse engineered, the longest possible overlap is provided and then speed control is provided which validates the use of that shorter overlap.

For example, if there isn't enough space for a full speed overlap of 50mph but there is enough for 38mph, then a 38mph overlap is provided on the starter and speed control is provided on approach. That speed controlled signal's overlap must be good for 50mph, and the signal can only clear if the train is approaching at 38mph (or less, depending on whether or not you take reacceleration into account.)

Most if not all of the very short starter overlaps (which truly were for a nominal speed of <15mph) have now been removed.

 

[Surreytraveller]

I am surprised that this didn't end up as a RAIB investigation:


Evening Standard article today:

Tube driver overran platform by 50m after falling asleep at controls

Unions raise concerns about ‘punishing’ shift patterns for Tube drivers

www.standard.co.uk

TfL response to FOI request, November 2022

FOI request detail

FOI request detail

tfl.gov.uk

A SPaD wouldn't ordinarily result in a RAIB investigation. They aren't an uncommon occurrence

 

[philthetube]

Metropolitan line trains are designed in a way that makes it possible to stand while driving.

 

[Mawkie]

Most if not all of the very short starter overlaps (which truly were for a nominal speed of <15mph) have now been removed.

There is a good example I think on the Picadilly Line at Leicester Sq (eastbound) with a speed controlled train stop ¾ of the way down the platform. Obviously Covent Garden is very close. If I recall, it's set to drop at 15 or 16mph?

 

[Towers]

Silly question time then; are they not ATO?!

 

[rebmcr]

Silly question time then; are they not ATO?!

Metropolitan line trains are not currently provided with ATO-compatible signalling as far north as Ickenham.

Piccadilly line trains have no ATO capability at all.

 

[Towers]

Metropolitan line trains are not currently provided with ATO-compatible signalling as far north as Ickenham.

Piccadilly line trains have no ATO capability at all.

Well, didn't know that! My assumption was that all the S Stock was ATO!

 

[Lewlew]

Well, didn't know that! My assumption was that all the S Stock was ATO!

Still being rolled out. Met line is only Finchley Road - Aldgate. Uxbridge branch pencilled in for early 2025.

 

[rebmcr]

Well, didn't know that! My assumption was that all the S Stock was ATO!

Much more information is available in this thread: https://www.railforums.co.uk/threads/ssr-resignalling.166182/

 

[Strathclyder]

There is actually some potential for collision on LU if a driver disregards a starting signal at full speed, as in some places overlaps on starting signals are of nominal length only, though the way things are designed extra protection is provided if there is potential for a train ahead to *routinely* stop in a location protected only by a nominal overlap.

Fatigue remains the big elephant in the room. It is very difficult to reconcile an industry which has always worked with extreme and erratic shift patterns with natural human fatigue. It’s something to remember when people moan about how highly train drivers are paid, many people might fancy the salary but either wouldn’t want or couldn’t cope with the shift patterns, and the self-discipline they require including when not at work.

OT, but this is what scuppered my childhood aspirations of becoming a train driver once they came into contact with that cold, unflinching object known as reality. I know full well I'd find it extremely difficult, if not outright impossible, to adapt/adjust my sleeping patterns (which are all too easily disrupted) to the shift patterns and to maintain the self-discipline needed, on and off-duty. I just ain't built for it. It's something I've long since made peace with and it has only heightened my respect & admiration for those behind the controls.

 

[bramling]

There's very little potential in such a circumstance. The normal design process is to calculate the full speed overlap of a starting signal, and if there is no credible collision risk within that calculated overlap (i.e. no berth or confliction within it) then a short overlap for 35km/h may be provided. If there is a confliction then things are reverse engineered, the longest possible overlap is provided and then speed control is provided which validates the use of that shorter overlap.

For example, if there isn't enough space for a full speed overlap of 50mph but there is enough for 38mph, then a 38mph overlap is provided on the starter and speed control is provided on approach. That speed controlled signal's overlap must be good for 50mph, and the signal can only clear if the train is approaching at 38mph (or less, depending on whether or not you take reacceleration into account.)

Most if not all of the very short starter overlaps (which truly were for a nominal speed of <15mph) have now been removed.

Presumably there's still plenty of 35 km/h overlaps on starting signals in the tunnel section of the Picc Line, which in many cases seem unlikely to be sufficient for a train running at full speed and making no attempt to slow down for the platform. I know a collision would be the result of a number of things coinciding, however I wouldn't fancy the chances of something running through somewhere like Wood Green westbound at full speed (bottom of a 1 in 60 downhill gradient) and stopping within 70 metres!