• 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!

RAIB investigation launched into a dangerous occurrence at Todmorden, West Yorkshire, 7 May 2026.

Whisky Papa

Member
Joined
8 Aug 2019
Messages
722
I saw this announcement yesterday evening, just before the Bedford incident started unfolding. I don't believe the Todmorden incident was discussed at the time, I was certainly unaware of it despite being mere yards from home.

https://www.gov.uk/government/news/dangerous-occurrence-at-todmorden

At around 22:28 on 7 May 2026, the driver of an out-of-service passenger train, operated by Northern, was given permission by the signaller to move their train closer to a signal on Todmorden West Curve, in West Yorkshire. However, the driver was not aware that their train was already standing with its driving cab beyond this signal. This resulted in the train making an uncontrolled movement towards Todmorden Viaduct Junction. The driver realised what had happened and brought their train to a stop close to the junction.


Around 12 seconds after the out-of-service train came to a stop, the 21:57 Manchester Victoria to Leeds passenger service, also operated by Northern, passed through the junction at about 45 mph (72 km/h). This service had just left Todmorden station.


There were no injuries or damage caused as a result of the incident, which occurred during a period of revised working arrangements that had temporarily increased the number of shunting movements involving Todmorden West Curve.


Our investigation will seek to identify the sequence of events which led to the near miss. It will also consider:


  • any factors that may have influenced the actions of those involved
  • the status and function of the signalling systems in the area
  • the management, training and competence of staff
  • how the risk of revised working arrangements was being assessed and controlled
  • any underlying management factors.

Our investigation is independent of any investigation by the railway industry or by the industry’s regulator, the Office of Rail and Road.


We will publish our findings, including any recommendations to improve safety, at the conclusion of our investigation. This report will be available on our website.
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Efini92

Established Member
Joined
14 Dec 2016
Messages
2,137

Peter749

Member
Joined
27 Apr 2016
Messages
262
Location
Salford
Seems strange the driver did not know the position of the train in relation to the signal
 

GadgetMan

Member
Joined
9 Jan 2012
Messages
980
Seems strange the driver did not know the position of the train in relation to the signal
Don't know the area or the incident. But as the post mentions shunt moves it is possible the train was doing a wrong direction move to go behind a signal facing the other way and driver had unknowingly already passed it. Just one of many possible scenarios.
 

Efini92

Established Member
Joined
14 Dec 2016
Messages
2,137
Don't know the area or the incident. But as the post mentions shunt moves it is possible the train was doing a wrong direction move to go behind a signal facing the other way and driver had unknowingly already passed it. Just one of many possible scenarios.
It was a signalled move onto a single line. Was 2 158’s, when he changed ends the cab was ahead of the signal. The northern report mentions reading through to the single on the up L&Y.
 

rower40

Member
Joined
1 Jan 2008
Messages
577
Change ends THROUGH the train, or OUTSIDE on the cess? Discuss…
(The point I’m trying to make is that if the driver had been walking outside the train while changing ends, they’d’ve seen the signal.)
 

AlterEgo

Verified Rep - Wingin' It! Paul Lucas
Joined
30 Dec 2008
Messages
29,187
Location
LBK
Change ends THROUGH the train, or OUTSIDE on the cess? Discuss…
(The point I’m trying to make is that if the driver had been walking outside the train while changing ends, they’d’ve seen the signal.)
They might not have seen it, depending on the route taken.
 

Efini92

Established Member
Joined
14 Dec 2016
Messages
2,137
Change ends THROUGH the train, or OUTSIDE on the cess? Discuss…
(The point I’m trying to make is that if the driver had been walking outside the train while changing ends, they’d’ve seen the signal.)
Through the train, there’s no reason to go outside there.
 

Shaw S Hunter

Established Member
Joined
21 Apr 2016
Messages
3,485
Location
Over The Hill
The curve is indeed a single line.

When it was first restored to traffic Northern had insufficient units to enable the planned service using it to operate. But trains did run from the Manchester direction, terminating at Todmorden and using the curve as a reversing siding.

When I was a guard there was an unofficial procedure when making reversing moves of this sort for the guard to remain in the back cab as the train entered such a track and give the driver a short single buzz to confirm the train was inside the signal. But only if such an arrangement was fully agreed by both parties.

I wonder if the traincrew in this case had any previous experience of reversing at Todmorden or had ever used the procedure I have described.
 
Last edited by a moderator:

AverageJoe

On Moderation
Joined
23 Dec 2021
Messages
671
Location
United Kingdom
I sign that route but not the curve.

It’s a 65mph line speed and not all services stop at Todmorden, glad the driver caught it.

I’m sure this will be talked about in the northern depots and everyone will be extra careful to make sure they are actually behind that signal.

I’m assuming this will be classed as a SPAD as one of the reasons for passing a signal at danger with authority is when starting beyond a signal.

Feel bad for the driver but glad it didn’t end up in being another serious incident.
 

CHAPS2034

Member
Joined
13 Mar 2018
Messages
644
The curve is indeed a single line.

When it was first restored to traffic Northern had insufficient units to enable the planned service using it to operate. But trains did run from the Manchester direction, terminating at Todmorden and using the curve as a reversing siding.

When I was a guard there was an unofficial procedure when making reversing moves of this sort for the guard to remain in the back cab as the train entered such a track and give the driver a short single buzz to confirm the train was inside the signal. But only if such an arrangement was fully agreed by both parties.

I wonder if the traincrew in this case had any previous experience of reversing at Todmorden or had ever used the procedure I have described.
Would the correct stopping point issue in this location be solved by putting up stop boards at the lineside to indicate the correct stopping position to be inside the signal, eg 2-car stop, 3-car stop, 4-car stop. Seen it in use elsewhere eg in the headshunt south of Wilmslow on the down side.
 

AverageJoe

On Moderation
Joined
23 Dec 2021
Messages
671
Location
United Kingdom
Would the correct stopping point issue in this location be solved by putting up stop boards at the lineside to indicate the correct stopping position to be inside the signal, eg 2-car stop, 3-car stop, 4-car stop. Seen it in use elsewhere eg in the headshunt south of Wilmslow on the down side.
That would of course help. But if this does turn out to be driver error either just having one of those days or his route knowledge needs improving then a driver is also capable of forgetting his train length, stopping at the wrong board and making this same mistake.

It doesn’t completely solve the issue.
 

Whisky Papa

Member
Joined
8 Aug 2019
Messages
722
With the train not fully behind the signal, shouldn't the interlocking have prevented the route being set for the train to Leeds?

That had been my first thought too. I had always assumed trains making this reversing move would actually pull up to PN337 (ie the signal protecting Stansfield Hall Junction), change ends then either wait there for PN318 to clear, or move forward to PN318 to wait until it cleared. As I've seen trains stood at PN318 I assume the latter, or is that actually where the change of ends has taken place and I've been wrong about my first assumption?

The curve is indeed a single line.

When it was first restored to traffic Northern had insufficient units to enable the planned service using it to operate. But trains did run from the Manchester direction, terminating at Todmorden and using the curve as a reversing siding.

When I was a guard there was an unofficial procedure when making reversing moves of this sort for the guard to remain in the back cab as the train entered such a track and give the driver a short single buzz to confirm the train was inside the signal. But only if such an arrangement was fully agreed by both parties.

I wonder if the traincrew in this case had any previous experience of reversing at Todmorden or had ever used the procedure I have described.

I was working in the ticket office at Todmorden one of the first Sundays the terminating service you refer to started, in fact I think it was the first day on May 18th 2014.

Slightly OT and in no way meant to detract from the seriousness of this thread, there was a misunderstanding by the signaller at Preston that day that resulted in the train moving from the curve into P1 at Todmorden well before its departure time, when the train from Leeds was still due to call ahead of it.

The terminating train from Manchester Victoria was due into Todmorden at xx30, and all morning had run onto the curve, quickly reversed and then stood on P1 until departure time back to Victoria at xx56, clear of the then-hourly train from Leeds to Victoria which called at xx09.

What the signaller had overlooked was that after the 1209 departure, the trains from Leeds changed to xx51 from Todmorden for the remainder of the day. At around 1240, they mistakenly allowed the 1256 departure into P1 from the curve, so I popped out to chat to the train crew and was told in no uncertain terms their train was staying put! I can't now recall if I spoke to Northern Control, Preston box or both, but the upshot was it soon returned to the curve for the service from Leeds to pass with little delay. Good job PN317 at Todmorden platform end was included in the revised layout to allow for this!
 

Signal Head

Member
Joined
26 May 2013
Messages
458
With the train not fully behind the signal, shouldn't the interlocking have prevented the route being set for the train to Leeds?


Correct
There's an overlap track circuit between the signal and junction (60m if I remember correctly), once the train has cleared the track circuit over the junction points, with the rear end standing in the overlap and therefore still beyond the 'outlet' signal, it has passed beyond the clearance point with the main line, so the interlocking can call those points Normal for a conflicting move.
 

Harpo

Established Member
Joined
21 Aug 2024
Messages
3,736
Location
Newport
I’m trying to understand this without a schematic in front of me but wonder if this is an odd example of an incident that wouldn’t have happened if the move had been discussed from an SPT instead of by radio?
 

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,492
Location
Moorpark, CA
I’m trying to understand this without a schematic in front of me but wonder if this is an odd example of an incident that wouldn’t have happened if the move had been discussed from an SPT instead of by radio?
I’ve remarked on this before in regard to “near misses“ where the protection hasn’t matched the worksite. People now call in from cellphones, the signaller has no idea where they are and has to take what they say at face value. I’d agree that the same would apply in this instance.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,597
Location
Nottingham
There's an overlap track circuit between the signal and junction (60m if I remember correctly), once the train has cleared the track circuit over the junction points, with the rear end standing in the overlap and therefore still beyond the 'outlet' signal, it has passed beyond the clearance point with the main line, so the interlocking can call those points Normal for a conflicting move.
However, if the train stopped with that overlap still occupied then the signaller would know the train hadn't fully passed the signal. I wonder if the actually intended for the train to continue a bit further towards Stansfield Hall to clear the overlap and get inside the signal that would clear for their reverse move back to Todmorden, but there was some miscommunication and the driver believed they were being told to move in the other direction. I wonder also if, due to the out-of-course working, they were reversing a longer train than usual.
I’ve remarked on this before in regard to “near misses“ where the protection hasn’t matched the worksite. People now call in from cellphones, the signaller has no idea where they are and has to take what they say at face value. I’d agree that the same would apply in this instance.
Wouldn't a GSM-R call identify the berth the train was in?
 

saismee

Established Member
Joined
20 Oct 2023
Messages
1,685
Location
UK
Wouldn't a GSM-R call identify the berth the train was in?
Which would be the one the signaller expects if they haven't gone far enough to occupy the track circuit.

Does anyone have a diagram of the layout?
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,597
Location
Nottingham
Which would be the one the signaller expects if they haven't gone far enough to occupy the track circuit.

Does anyone have a diagram of the layout?
The berth steps forward when the train first occupies a track circuit beyond the previous signal. In this case there would be a separate overlap track circuit so as the train passed the junction signal leaving Todmorden station the description would step into the berth in between the two signals on the curve. This probably means that the signaller and the GSM-R can't tell which of the two signals it is actually standing at. However the signaller will be able to see by occupation of track circuits whether the whole train is between the two signals.

Expanding on my earlier suggestion, one possibility is:
  1. Train signalled from Todmorden onto the curve (Down direction).
  2. Driver stops where they think they are fully past the Up direction signal but the tail of the train has not passed that signal, possibly because this is a longer train than the driver is accustomed to reversing here.
  3. Signaller sees track circuit occupied between Up direction signal and junction, at a time when the train should be at a stand on the curve.
  4. Signaller calls driver, assuming they have not changed ends, and thinks they are instructing them to draw forward towards the next Down signal (protecting Stansfield Hall).
  5. Driver has in fact changed ends or assumes the instruction is intended to request an Up direction move after they have changed ends.
  6. Driver sees no Up signal ahead of them, possibly disoriented in the dark and assumes it must be out of sight round the bend.
  7. Driver moves train incorrectly back towards Todmorden.
 
Joined
8 Feb 2021
Messages
1,033
Location
Manchester
Arriving from Manchester at Todmorden, the junction signal is before the platform and has a feather if you are going on the West Curve. Thus for the signaller in Preston PSB, if you are in the platform at Todmorden, you are in the same berth as if you are clear of the overlap and ‘on’ the curve in the correct shunt position
 

SuspectUsual

Established Member
Joined
11 Jul 2018
Messages
6,678
The thing I don’t understand is the phrase “was given permission by the signaller”, ie there had been a request by the driver.

If the driver believed he was between the two signals on the curve, why would he need permission to move a few yards within that area?
 

Signal Head

Member
Joined
26 May 2013
Messages
458
The berth steps forward when the train first occupies a track circuit beyond the previous signal. In this case there would be a separate overlap track circuit so as the train passed the junction signal leaving Todmorden station the description would step into the berth in between the two signals on the curve. This probably means that the signaller and the GSM-R can't tell which of the two signals it is actually standing at. However the signaller will be able to see by occupation of track circuits whether the whole train is between the two signals.

Expanding on my earlier suggestion, one possibility is:
  1. Train signalled from Todmorden onto the curve (Down direction).
  2. Driver stops where they think they are fully past the Up direction signal but the tail of the train has not passed that signal, possibly because this is a longer train than the driver is accustomed to reversing here.
  3. Signaller sees track circuit occupied between Up direction signal and junction, at a time when the train should be at a stand on the curve.
  4. Signaller calls driver, assuming they have not changed ends, and thinks they are instructing them to draw forward towards the next Down signal (protecting Stansfield Hall).
  5. Driver has in fact changed ends or assumes the instruction is intended to request an Up direction move after they have changed ends.
  6. Driver sees no Up signal ahead of them, possibly disoriented in the dark and assumes it must be out of sight round the bend.
  7. Driver moves train incorrectly back towards Todmorden.
A small point - the junction signal (PN320) is on the Manchester side of Todmorden station, so the TD step has already taken place before the train arrives at it.

OTT shows separate berths for PN318 & 337, however I think on the panel there is only one display shared between them.
 
Last edited:

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,597
Location
Nottingham
They would, but if the following train was approaching they could have set the
I think some words have gone astray here, I presume you intended something like "...route for the following train towards Hebden Bridge". Which I agree with, but the signaller would also want to get the terminating train inside the signal so it could return towards Manchester without causing delay.
 

Signal Head

Member
Joined
26 May 2013
Messages
458
I think some words have gone astray here, I presume you intended something like "...route for the following train towards Hebden Bridge". Which I agree with, but the signaller would also want to get the terminating train inside the signal so it could return towards Manchester without causing delay.
That post was abandoned when I realised I was answering a point different to the one you were actually making. I think the editor has hung on to the text and included it with my subsequent post without me noticing.

Now edited to remove.
 

saltywalrus

Member
Joined
21 May 2022
Messages
29
Location
Lancaster

That post was abandoned when I realised I was answering a point different to the one you were actually making. I think the editor has hung on to the text and included it with my subsequent post without me noticing.

Now edited to remove.
Is Preston PSB still working? I thought it had been moved to Ashburys ROC
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,597
Location
Nottingham
Is Preston PSB still working? I thought it had been moved to Ashburys ROC
Yes, the WCML North power boxes are still operational, and for reasons that aren't clear to me someone decided back in the 70s that Preston should cover the Copy Pit line and a chunk of the Calder Valley line around Todmorden. The photo posted upthread shows that the signalling of the reinstated curve is from a panel not a screen.
 

SuspectUsual

Established Member
Joined
11 Jul 2018
Messages
6,678
Yes, the WCML North power boxes are still operational, and for reasons that aren't clear to me someone decided back in the 70s that Preston should cover the Copy Pit line and a chunk of the Calder Valley line around Todmorden. The photo posted upthread shows that the signalling of the reinstated curve is from a panel not a screen.

I suppose it may have been the only logical choice - at that stage the rest of the Calder Valley was still local boxes wasn't it?
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,597
Location
Nottingham
I suppose it may have been the only logical choice - at that stage the rest of the Calder Valley was still local boxes wasn't it?
Yes, but I don't know why that part was replaced when the rest of the route carried on with similar mechanical signalling for several more decades. Especially peculiar in the 80s when the Copy Pit line was under threat of closure, and the Calder Valley section would have had no direct connection back to its controlling box.

The curve closed in 1972, about the time the power box was being brought in, but I don't know if the original curve was ever on the panel. There was however a stub at Stansfield Hall that was used as a refuge for a banking engine up to Copy Pit.
 

Top