Major Incident at Huntingdon (01/11/2025)

[sharpener]

whilst braking for the turnout at Huntingdon North junction for which the signalling was not yet set

Vital part of the plan! He/she was clearly aiming to maximise chance of success by braking (if necessary to a halt) but also buying the signaller some precious extra seconds while waiting for the interlock to time out*. Only downside is lost speed if for some reason Huntingdon not a viable option and has to overshoot to St Neots.

In the event the new route was proved 2'15' seconds after the old one was canx so signaller well on the ball also. Does the hardware enable him to queue requests or have a status indicator showing progress of timeout, or did it just involve him watching the clock and pressing the button at exactly 2'01"?

* I understand how this to avoid setting routes with too little time to brake, or confronting drivers with a sudden major change of signal aspect, but it seems a tad inflexible if there is no override available in any circumstances.

 

[SpannerWeilder]

Long time lurker here, no connection with the industry other than travelling thousands of miles every year.

I just wanted to add my voice to those wishing your colleague well, and to praise the actions of the train crew and everyone else connected with this horrific episode. From what I can see as an outsider, everyone performed impeccably, and doubtless saved the incident from being even worse.

It's also been a pleasure to witness your members self-police the conversation, and to discuss the event with respect, calm authority, and without hyperbole.

Your industry should be very proud of itself this weekend.

 

[Unixman]

Or indeed - as I saw online recently - that drivers only need six weeks of training!

100% this ^^

All of the people involved will of dealt with many situations, but nothing will ever prepare you for something like last night.

Thoughts with the member of LNER staff currently in hospital. You are a hero.

Telegraph is naming the driver involved. I won't post the name here though.

 

[bicbasher]

The railway came together last night. The safety of the passengers from all involved was their first priority and many lives were saved because of their actions. Thoughts with the LNER employee and hope they recover.

 

[flitwickbeds]

Anons on the internet know and that is why there were 1000 watchers last night.

Because we are a self-selecting group of people interested in and knowledgeable about the train system in this country and how it operates. We are aware of, and regularly use, third party sites presenting industry data, and know how to interpret them. Now try to do the same kind of thing for a subject you have no knowledge of. Large parts of the Internet are failing because of a 500 error in Google. Write a good enough article in the next 5 minutes which explains exactly what is going wrong and what is being done to put it right, but in friendly, non-technical language - relying on a journalist on shift at 10pm on a quiet Saturday night.

Never forget that no one can be expert in all topics ready to interpret what can be quite complex information at the drop of a hat.

Can we raise a glass to @ainsworth74 for his watchful eye on the forum yesterday evening when there were at one point over 1000 people viewing this thread?

Hopefully those visitors lurking will have seen that Rail Forums is the antithesis to the cesspool that is Twitter/X.

Agreed.

 

[Banham7]

The railway came together last night. The safety of the passengers from all involved was their first priority and many lives were saved because of their actions. Thoughts with the LNER employee and hope they recover.

The on-train crew (plus the team at York ROC and LNER Control) were absolutely heroes last night. I cannot believe there are tweets circulating criticising the on-train crew's actions. I'm wishing the best for the team member who is critical and really hope they pull through.

 

[Unixman]

Because we are a self-selecting group of people interested in and knowledgeable about the train system in this country and how it operates. We are aware of, and regularly use, third party sites presenting industry data, and know how to interpret them. Now try to do the same kind of thing for a subject you have no knowledge of. Large parts of the Internet are failing because of a 500 error in Google. Write a good enough article in the next 5 minutes which explains exactly what is going wrong and what is being done to put it right, but in friendly, non-technical language - relying on a journalist on shift at 10pm on a quiet Saturday night.

Never forget that no one can be expert in all topics ready to interpret what can be quite complex information at the drop of a hat.


Agreed.

400 errors: your problem.
500 errors: their problem

 

[Steve Harris]

They BBC was happy to quote - without fact checking - an MP who misidentified it! This is for the sole reason they are an MP.

And we all know that MP's always tell the truth

If I worked in the media, they would be the last source I'd trust.

 

[WillC]

See: Man, 32, treated as only suspect in train stabbing as staff member in life-threatening condition, police say - latest - BBC News https://www.bbc.co.uk/news/live/cm2zvjx1z14t?post=asset%3A833aada0-be5d-4dba-b0cd-3c611fa9cca0#post

Long time lurker here, no connection with the industry other than travelling thousands of miles every year.

I just wanted to add my voice to those wishing your colleague well, and to praise the actions of the train crew and everyone else connected with this horrific episode. From what I can see as an outsider, everyone performed impeccably, and doubtless saved the incident from being even worse.

It's also been a pleasure to witness your members self-police the conversation, and to discuss the event with respect, calm authority, and without hyperbole.

Your industry should be very proud of itself this weekend.

100% this ^ , from another forum lurker and regular LNER traveller. If anyone involved is reading this take heart because Joe Public is with you and impressed beyond words.

 

[TT-ONR-NRN]

Prayers for the brave LNER staff member in a critical condition. I imagine they were attacked while trying to protect their passengers and colleagues. 800111 should well be named after them, with their permission of course.

 

[godfreycomplex]

Can we raise a glass to @ainsworth74 for his watchful eye on the forum yesterday evening when there were at one point over 1000 people viewing this thread?

Seconded

 

[Steve Harris]

The BBC on the rolling feed have posted about how the driver and signaller worked together to stop at a station.

Guessing that’s been written from comments here, as there’s been no official source!

There was an interview with Simon Calder this morning on the BBC News channel and that is what he said.

 

[D6130]

Seconded

Thirded!

 

[Steve Harris]

Seconded

Thirded.

Although I would also praise the forum members last night who used some good old common sense and didn’t start to speculate, we just stuck to the facts.

 

[gray1404]

So we now have a dreadful situation where somebody who was not involved in the incident the 35 year old, was wrongly arrested.

 

[800001]

So we now have a dreadful situation where somebody who was not involved in the incident the 35 year old, was wrongly arrested.

I would rather someone be arrested and released, than have no one arrested at all.

Yes, it's not nice, horrible for the person involved, but at the end of the day they are now free to go.

 

[Steve Harris]

National Rail now saying Huntingdon station will remain closed all day Monday.

That was posted on the Thameslink website early this afternoon. See post #337

 

[AlterEgo]

So we now have a dreadful situation where somebody who was not involved in the incident the 35 year old, was wrongly arrested.

The public are said to have identified this person, in fairness. This person was similar in age and appearance to the perpetrator.

Police have said the following about the hospitalised member of staff: https://www.bbc.co.uk/news/live/cm2...set:833aada0-be5d-4dba-b0cd-3c611fa9cca0#post

This casualty is a member of LNER rail staff who was on the train at the time and tried to stop the attacker.

"Detectives have reviewed the CCTV from the train and it is clear his actions were nothing short of heroic and undoubtedly saved many people's lives."

 

[YorkshireBear]

I would rather someone be arrested and released, than have no one arrested at all.

Yes, it's not nice, horrible for the person involved, but at the end of the day they are now free to go.

Agreed. And hopefully that person understands that in what was a very chaotic situation.

 

[Morgsie]

My thoughts are to those injnured including the LNER staff mnember who I hope makes a full recovery. Kudos to the emergency services, LNER, Network Rail and all the other agencies involved. I also would like to commend @ainsworth74 for last night regarding here

 

[Steve Harris]

I would rather someone be arrested and released, than have no one arrested at all.

Yes, it's not nice, horrible for the person involved, but at the end of the day they are now free to go.

Indeed this ^

 

[flitwickbeds]

Telegraph is naming the driver involved. I won't post the name here though.

GB News has had the same name published for a few hours now.

400 errors: your problem.
500 errors: their problem

Not necessarily. 401, 405, 408, 413, 414, 421, 423, 424, 425, and 431 could easily be the web developer's/server administrator's mistake. And 511 could be a client side problem.

</ot>

 

[AlterEgo]

Any news on the grade of the injured staff member? (Please don't identify them of course)

If the message has got through to the driver without a guard (if it is them who was incapacitated) then another staff member has acted very quickly.

It transpires it is the guard who is in hospital. Hopefully they make a full recovery.

I assume another member of staff or a passenger has alerted the driver and spoken to them verbally. Is it possible to do this using the passcomm on the Hitachis?

 

[Richardr]

The BBC have the name of the train driver with a picture of him on their website, so I assume it has been released.

https://www.bbc.co.uk/news/live/cm2zvjx1z14t

 

[Steve Harris]

Vital part of the plan! He/she was clearly aiming to maximise chance of success by braking (if necessary to a halt) but also buying the signaller some precious extra seconds while waiting for the interlock to time out*. Only downside is lost speed if for some reason Huntingdon not a viable option and has to overshoot to St Neots.

In the event the new route was proved 2'15' seconds after the old one was canx so signaller well on the ball also. Does the hardware enable him to queue requests or have a status indicator showing progress of timeout, or did it just involve him watching the clock and pressing the button at exactly 2'01"?

* I understand how this to avoid setting routes with too little time to brake, or confronting drivers with a sudden major change of signal aspect, but it seems a tad inflexible if there is no override available in any circumstances.

I don't work on the railway's but I do take an avid interest in watching signalmaps and I have noticed thatbin the Peterborough area it does seem that a route can be preset while a preceding train is going through a junction, so to answer your question.. I would say it's a yes (based solely on my observations), if I'm wrong I'm happy to be corrected.

 

[edwin_m]

In the event the new route was proved 2'15' seconds after the old one was canx so signaller well on the ball also. Does the hardware enable him to queue requests or have a status indicator showing progress of timeout, or did it just involve him watching the clock and pressing the button at exactly 2'01"?

Queuing of requests, known as pre-selection, is not permitted. Something might have changed between the signaller requesting the route and it becoming available, or it might have become available incorrectly due to a momentary wrong side track circuit failure or similar. The signaller is expected to check that everything looks correct before setting another route.

ARS systems (and presumably the DRS system used here) can set a route immediately a conflicting route has cleared, but also include checks that track circuits are occupying and clearing in the normal sequence and will not set the route if they aren't.

When the approach locking timer is active after cancelling a route, the route itself remains set and the signal shows flashing red on the panel (obviously just normal red on the tracksider). After the timeout period the signal indication reverts to steady red and the white route lights go out.

I don't work on the railway's but I do take an avid interest in watching signalmaps and I have noticed thatbin the Peterborough area it does seem that a route can be preset while a preceding train is going through a junction, so to answer your question.. I would say it's a yes (based solely on my observations), if I'm wrong I'm happy to be corrected.

I can't speak for Peterborough specifically but for modern interlockings in general, a following route can be set as soon as a preceding train has passed over all the points that need to be changed to set the new route. If there are no points to change then the following route can be set immediately, but if cancelled before the signal clears it will release immediately with no timeout. That situation didn't arise in this incident.

 

[Annetts key]

Vital part of the plan! He/she was clearly aiming to maximise chance of success by braking (if necessary to a halt) but also buying the signaller some precious extra seconds while waiting for the interlock to time out*. Only downside is lost speed if for some reason Huntingdon not a viable option and has to overshoot to St Neots.

In the event the new route was proved 2'15' seconds after the old one was canx so signaller well on the ball also. Does the hardware enable him to queue requests or have a status indicator showing progress of timeout, or did it just involve him watching the clock and pressing the button at exactly 2'01"?

* I understand how this to avoid setting routes with too little time to brake, or confronting drivers with a sudden major change of signal aspect, but it seems a tad inflexible if there is no override available in any circumstances.

With conventional interlocking (relay interlocking) the signaller is not provided with any provision to "queue requests". Indeed, the circuitry I worked on (which is different to the signalling being discussed here) is designed to prevent "preselection". However, a route can be set before a train has cleared the route that you want to call.

The signaller only has the "route lights" as an indication that a route is either set or is locked (the route is locked on approach and during the passage of a train once the signal has given a proceed aspect, or the signal should have given a proceed aspect). When these extinguish, they the signaller knows that they can set a new / different route. No timeout indication is given. Although on more modern computer based interlocking, the signallers signal indication will flash red instead of being a steady red during a time out.

Not all timers (which are associated with the signal, not the route, the time applies to all routes for that signal) always have the same time. It depends on a number of factors, but the "default" for a main aspect signal is two minutes (+/-10%).

Approach locking was introduced to prevent points from being moved as a train approached (maybe at line speed) after accidents caused by the signaller changing the position of the points after giving a train a proceed aspect. Hence no override is provided.

 

[800001]

It transpires it is the guard who is in hospital. Hopefully they make a full recovery.

I assume another member of staff or a passenger has alerted the driver and spoken to them verbally. Is it possible to do this using the passcomm on the Hitachis?

Is that from the BBC website? As that differs from what I'm aware of.

Think again, best to wait for official announcements (nothing that Btp or David Horne has said states who the injured member of staff is ) , either way, best wishes to them and hope they pull through.

 

[Tester]

My somewhat buggy website gets a mention... I don't currently display route data in that area. Here is a link if anyone wants to look themselves.

Taking a look manually at that stored data which I don't currently display on there I can confirm the following from my record of signalling data. Anything in italics I've tried to explain the meaning of after the timeline.

19:33:29 Route proves from the Up Main to Up Slow at Huntingdon North Junction for 9J61 to enter Platform 2 as booked (pink route below)
19:33:32 Signal 352 displays a proceed aspect
19:36:50 9J61 passes signal 352 and it displays a danger aspect
19:37:13 9J61 clears the pointwork at Huntingdon North Junction and the route releases
19:37:37 Route proves from the Up Main to Up Fast at Huntingdon North Junction for 1Y90 to continue on the fast lines as booked (red route below)
19:37:41 Signal 352 displays a proceed aspect
19:40:19 Route set earlier for 1Y90 to stay on the Up Fast cancelled (also known as replacing to danger / pull the signal/road)
19:40:21 Signal 352 reverts to a danger aspect. 1Y90 is just one signal away at this point approaching signal 356 - if the signal was visible, the driver would have seen it go from green to a single yellow (they would have obviously been warned by the signaller this would happen / requested it as mentioned up thread)
19:41:11 9J61 clears the signalling section in Huntingdon Plat 2. It might have been a few more seconds before a route could be set due to local interlocking rules. Regardless an approach locking timer would prevent a route being set into Plat 2 for 1Y90 for usually 2-3 minutes at 125mph (clearly 2 here based on the next timestamp) after the route was cancelled - so possible to set a route from 19:42:19 ish onwards
19:42:34 Route proves from the Up Main to Up Slow at Huntingdon North Junction for 1Y90 to enter Platform 2 (pink route below)
19:42:37 Signal 352 displays a proceed aspect
19:42:51 1Y90 passes signal 352 and it displays a danger aspect, and then enters Platform 2 at Huntingdon

I think this shows some incredible quick thinking. Hats off to the driver, signaller, and anyone else involved in getting the train into a platform as soon as possible.

Annotated screen grab of the state at 19:40:19 when signal 352 when the red route was cancelled and signal 352 reverted to danger.
View attachment 191590

Some further explanations:
For a route to "prove" this means that the points have swung to the required direction and it's detected that they've locked - you don't want them not going all the way or locking and having points move under a moving train! The time it takes for this to happen once the signaller requests a route to be set depends on the pointwork installed but I'd say its usually within the general ballpark of 5 to 10 seconds for an important junction like this. Maybe slightly quicker. From this the signaller acted almost immediatly once the timeout on the cancelled route from signal 352 ended.
Proceed aspect - green/yellow/double yellow. Danger aspect - red. The signalling feed's don't differentiate between different possible proceed aspects.
Route release - unless set as an auto route, the route will "release" and the points will usually/sometimes* automatically move to their normal position once a train has cleared the points. No route is automatically set over them so the signal protecting the points stays red, until the signaller or ARS/DRS sets a route.
The interlocking rules prevent conflicting movements happening. Track is split into signalling sections and if a train is detected in a section or a route set over it the interlocking system prevents any conflicting moves being made.

I have no formal relation to the industry and am just sharing what I've learnt on here, simulating various signalling areas on SimSig, and creating the Vail website linked above. If anyone who knows the area better and has any corrections to make please let me know!

*Some points will auto revert to normal position (Up Main -> Up Fast in this case) but I have no idea how widespread this is. It might be standard on most points in the country everywhere or not, I don't know and it's very niche stuff so I haven't been able to figure it out from a quick google.


It's near Stilton Fen / Holme on the two track section just south of Peterborough. Huntingdon was previously controlled by Kings Cross PSB before moving to York ROC.

This is a seriously impressive post.

As someone who took up railway signal engineering as a career 53 years ago, I am humbled that someone from outside the industry has put this together so well.

Re your asterisked note, points generally do not auto-normalise. There are occasions where this is done for specific reasons, but it is rare, and there is no reason for it to apply here. (Your observation of niche is spot on!)

In addition, what this timeline shows is that comprehensive approach locking (which I commented on in an earlier post) was irrelevant here; the timeline would have been identical either way.

Well done to you, and well done to the driver and signaller, who appear to have managed this masterfully.

 

[43096]

It transpires it is the guard who is in hospital.

Do you have a source for that?