Bedford (Elstow) collision reporting by Daily Express: is this accurate or misleading?

KevinTurvey · 26 Jun 2026

Daily Express, say no more! For a newspaper specialising in Princess Diana, Madelaine McCann, or how much house prices have gone up, it cannot be taken seriously on any matter of technical reporting.
All of the Reach plc newspapers (and they own a lot of them) are a joke.

RailUK Forums

DerekC · 26 Jun 2026

Bedpan said:
The problem with the statement as I see it is that it could be interpreted that, if there hadn't been a fault with the AWS which turned the signal to red, the signal would have remained green after the 810 had passed it. So if, for instance, a train had made a routine station stop there would be nothing to prevent a following train crashing into it. Or at least, my wife interpreted it that way and said it had put her off travelling by train in future.

RAIB has reworded it and it's better, although you could still interpret it as a sequence of events, I suppose:

"RAIB’s preliminary examination has found that train 1B67 was standing next to WH152 signal when the accident occurred. The train had come to a stop unexpectedly because a fault had developed with the Automatic Warning System (AWS) equipment fitted to it, which caused the brakes to apply.

Data from railway systems recorded that the track circuit for the section of track on which train 1B67 was standing was shown as occupied. The signal interlocking would have therefore changed the signal behind the train, WH154, to display a red (danger) aspect, with the intention of protecting train 1B67 from other train movements."

https://www.gov.uk/government/news/...9ead-4e99-9f7d-be974a3f5e40&utm_content=daily

I am thinking this drip-feed of interim reports is a bad idea. It allows the rubbish media to come up with partially informed theories that get a lot of traction with the general public. Attempts to put them right are seen as coverup.

E27007 · 26 Jun 2026

Bedpan said:
The problem with the statement as I see it is that it could be interpreted that, if there hadn't been a fault with the AWS which turned the signal to red, the signal would have remained green after the 810 had passed it. So if, for instance, a train had made a routine station stop there would be nothing to prevent a following train crashing into it. Or at least, my wife interpreted it that way and said it had put her off travelling by train in future.

Yes, the misconception by the Express writer, " the AWS failure of the first train set the signal to display a danger aspect" , the writer has the misconception, the AWS system on the train has the functional capacity to control a signal. AWS is an indication system, and does not control signals!
We should all know, track circuits detect the presence of a train in a section and track circuits set the signal aspect to danger to protect the rear of the train.
The misconception is forgivable, as an ex S&T technician and then a driver, many railway enthusiasts too have misconceptions as to how signalling systems work

whoosh · 26 Jun 2026

Bedpan said:
The problem with the statement as I see it is that it could be interpreted that, if there hadn't been a fault with the AWS which turned the signal to red, the signal would have remained green after the 810 had passed it. So if, for instance, a train had made a routine station stop there would be nothing to prevent a following train crashing into it. Or at least, my wife interpreted it that way and said it had put her off travelling by train in future.

Real life isn't like this of course, but if you were to send a child to school either by walking cycling or even giving them a lift yourself every school day, OR putting them on a train on all seven days of the week and all year including the school holidays - the train would be safer.

D1537 · 26 Jun 2026

Harpo said:
I’d trust the writers of Viz over any newspaper on railway details.

Especially given that founder Chris Donald is a railfan. Some of his spoof rail posters are excellent (link below).

Jolly Days Out by Chris Donald

Harpo · 26 Jun 2026

D1537 said:
Especially given that founder Chris Donald is a railfan. Some of his spoof rail posters are excellent (link below).

Jolly Days Out by Chris Donald

Exactly!

John Webb · 26 Jun 2026

DerekC said:
......I am thinking this drip-feed of interim reports is a bad idea. It allows the rubbish media to come up with partially informed theories that get a lot of traction with the general public. Attempts to put them right are seen as coverup.

Thanks for the update on the RAIB wording.

But I have to disagree with you about the RAIB's interim reports being 'drip-feed'. I accept that the few railway accidents that occur (compared to what happens daily on the roads) tend to draw an excess of attention to themselves. RAIB are under considerable pressure, as indeed are investigators into any incident where life has been lost*, and to not release information once it is available would lead to even more accusations of 'a cover-up', I think.
*I speak from personal experience of investigating fatal fires....

sharpener · 26 Jun 2026

DerekC said:
RAIB has reworded it and it's better, although you could still interpret it as a sequence of events, I suppose:

https://www.gov.uk/government/news/...9ead-4e99-9f7d-be974a3f5e40&utm_content=daily

I am thinking this drip-feed of interim reports is a bad idea. It allows the rubbish media to come up with partially informed theories that get a lot of traction with the general public. Attempts to put them right are seen as coverup.

There is an element of misinformation in that AFAICS the (revised) report page just says "Published 24 June 2026" without mentioning the revision history or saying where previous version(s) can be found - if they can officially be found at all. I would have expected better.

bramling · 26 Jun 2026

DerekC said:
RAIB has reworded it and it's better, although you could still interpret it as a sequence of events, I suppose:

https://www.gov.uk/government/news/...9ead-4e99-9f7d-be974a3f5e40&utm_content=daily

I am thinking this drip-feed of interim reports is a bad idea. It allows the rubbish media to come up with partially informed theories that get a lot of traction with the general public. Attempts to put them right are seen as coverup.

I’m not really too bothered if things are seen as a coverup, once the coverage has got to that point then as far as I’m concerned it’s passed beyond the point where opinion matters.

There will always be people who have odd views, normally off the back of either their own political agenda, or simply because they have mental health problems. Unfortunately social media platforms like Facebook do give them a means to mouth off in a way they wouldn’t otherwise be able to do, and little to stop them being abusive when facts are presented.

I feel the industry could have done a bit better with their expectation management though. There was an opportunity to get some senior leadership figures out over the weekend and explain, even in general terms, what had happened, give a bit of basic explanation as to how signalling works, that sort of thing. The door has been left wide open for silly stuff now.

I even saw a post on Facebook where someone have an explanation about signalling and was accused of having a political agenda against Keir Starmer.

swj99 · 26 Jun 2026

I think the main takeaway from this is that newspapers cannot be trusted to present an accurate statement of the facts of a situation. This is nothing new, and comes as no surprise to me, although when I posted a similar comment on a well known social media website in 2020, I was accused of being a conspiraloon.

LAX54 · 26 Jun 2026

mmh said:
There's absolutely nothing wrong with that report, it's written and explains the situation to a layman well.

It says: This failure automatically turned the signal behind it to red, that is incorrect, the signal would hae been red to start with, the track occupation would see to that.
But this is the D.E !

chriseurope · 26 Jun 2026

Interesting in the context of previous accidents where, in the broadest sense of the phrase, a signal was passed at danger - I'm thinking specifically here of Ladbroke Grove 1999 - the RAIB doesn't mention anything about considering wider operational issues, such as 1H46 staying on the slow line to pass the stationary 1B67 and so maybe reducing delays as well ? I've no idea if there would have been sufficient time for this to be even considered, or what the consequences of that might have been, but it stands to reason that if a line is blocked for whatever reason, after a certain amount of time the "automatic" operation would have to be overridden. (I'm trying to chose my words carefully, given the debate about expert vs layman vs tabloid journalists).

If I remember correctly, at Paddington in 1999, SPADs at SN109 leaving Paddington had been fairly regular (i.e. far too often) and the Slough IECC signallers' response was not quick enough, mainly due to poor training and procedures, which was highlighted in the report ("The general picture which emerged was of a slack and complacent regime, which was not alive to the potentially dire consequences of a SPAD or of the way in which signallers could take action to deal with such situations.")

To be clear, as former BR engineer, I'm in no way criticising the signalers/control or driver. Systems are becoming increasing more sophisticated and inexpensive (not to mention AI - oops ...). But if the much used phase 'to make sure this never happens again' is to be more than just a slogan, I think the RAIB and Network Rail have to dig deeper, and politicians also have to wake up.

Trying not to be too political, I'm sure most people in the UK would rather get to their destination by trains on the existing extensive rail network safely, rather than getting from London/Old Oak to Birmingham at 320 km/hour. That's the reality of the situation, money can seem to grow on trees when Government wants it to.

Hopefully not too off-topic !

AlterEgo · 26 Jun 2026

The practice of giving updates and then an interim and final report is well founded.

The initial update from RAIB is welcome. There are about, hmm, a dozen different potential causes when dealing with a rear ending like this, that range from extreme concern (systems failure/runaway/major systemic fault in how train movements are actually conducted countrywide) to lower concern (SPAD, driver incapacity/medical etc). The RAIB investigate accidents not just to make the railway safer but also to maintain confidence in the railway, whether that be passenger confidence or the confidence of staff.

At the moment, RAIB have updated that the signals were, from initial enquiries, working correctly by displaying the correct aspects, and video/OTDR evidence showed the driver had accelerated beyond a cautionary speed and also passed a red aspect which was the proximal cause of the crash. Distal causes have yet to be examined, and this will take some time. This has sparked healthy debate in professional and what I'd call higher-functioning enthusiast circles who have an interest in how these incidents can happen. I am sure people who are currently driving on the mainline have chills about the incident, and have stopped to think if such an incident could happen to them or someone they work with. In my view, limited speculation is healthy and should not be suppressed; we have been fortunate on the forum that whatever speculation there has been has been informed, particularly by staff - some of whom sign the route, and some of who are S+T professionals who can give more granular explanation and what should and should not be possible. Speculation has also been respectful of the driver, who has sadly died.

RAIB have also communicated what they do not yet know. They do not know:

the condition of the AWS
whether it was functional
whether it gave an error
whether a warning was given, and then cancelled by the driver
whether a warning was given, not acknowledged, and a fault meant the train was not stopped automatically but the driver induced the braking, or
whether a warning was given, not acknowledged, and the braking experienced by the train was not driver induced but the emergency brakes responding automatically.

And they have told us why they do not know this yet.

I'm not sure what else they can do. The mainstream media does not show much interest in reporting things with surgical precision like RAIB do (every word RAIB put out is careful and deliberate). Was the signal red because the 810 had an AWS fault? Well, sort of, yes, in the sense that had it not had the fault it would have been doing 125mph and been 5 miles away when the 360 cleared the crossover...but also - no, signals don't work like that and they just communicate whether a train or obstruction is in the section. I am not sure the public really care and I doubt an incident of this nature will drastically affect public confidence unless some surprising, gravely serious fault with how signals work is eventually uncovered. And Facebook will always be Facebook, a lot of bored souls on there I'm afraid.

bramling · 26 Jun 2026

chriseurope said:
Interesting in the context of previous accidents where, in the broadest sense of the phrase, a signal was passed at danger - I'm thinking specifically here of Ladbroke Grove 1999 - the RAIB doesn't mention anything about considering wider operational issues, such as 1H46 staying on the slow line to pass the stationary 1B67 and so maybe reducing delays as well ? I've no idea if there would have been sufficient time for this to be even considered, or what the consequences of that might have been, but it stands to reason that if a line is blocked for whatever reason, after a certain amount of time the "automatic" operation would have to be overridden. (I'm trying to chose my words carefully, given the debate about expert vs layman vs tabloid journalists).

If I remember correctly, at Paddington in 1999, SPADs at SN109 leaving Paddington had been fairly regular (i.e. far too often) and the Slough IECC signallers' response was not quick enough, mainly due to poor training and procedures, which was highlighted in the report ("The general picture which emerged was of a slack and complacent regime, which was not alive to the potentially dire consequences of a SPAD or of the way in which signallers could take action to deal with such situations.")

To be clear, as former BR engineer, I'm in no way criticising the signalers/control or driver. Systems are becoming increasing more sophisticated and inexpensive (not to mention AI - oops ...). But if the much used phase 'to make sure this never happens again' is to be more than just a slogan, I think the RAIB and Network Rail have to dig deeper, and politicians also have to wake up.

Trying not to be too political, I'm sure most people in the UK would rather get to their destination by trains on the existing extensive rail network safely, rather than getting from London/Old Oak to Birmingham at 320 km/hour. That's the reality of the situation, money can seem to grow on trees when Government wants it to.

Hopefully not too off-topic !

The RAIB don’t really go down the road of being too descriptive, which is a contrast to what we saw from the likes of Cullen.

I always got the feeling that part of what seemed to trouble Cullen with the signallers at Ladbroke Grove was that he seems to have formed a view that they were a little untruthful as regards what was done, in particular whether and how soon a radio message was sent to the 165 driver. It wasn’t that the signallers did nothing, indeed ISTR they did manage to put a signal back. The whole thing from start to finish was about 33 seconds, that unfortunately isn’t much to be able to take meaningful action.

I suspect a lot of this won’t be relevant at Bedford as from what we know there wasn’t much time for anything to be done. It’s quite probable the 360 had been signalled over the junction before the 810 even came to a stand.

== Doublepost prevention - post automatically merged: 26 Jun 2026 ==

AlterEgo said:
The practice of giving updates and then an interim and final report is well founded.

The initial update from RAIB is welcome. There are about, hmm, a dozen different potential causes when dealing with a rear ending like this, that range from extreme concern (systems failure/runaway/major systemic fault in how train movements are actually conducted countrywide) to lower concern (SPAD, driver incapacity/medical etc). The RAIB investigate accidents not just to make the railway safer but also to maintain confidence in the railway, whether that be passenger confidence or the confidence of staff.

At the moment, RAIB have updated that the signals were, from initial enquiries, working correctly by displaying the correct aspects, and video/OTDR evidence showed the driver had accelerated beyond a cautionary speed and also passed a red aspect which was the proximal cause of the crash. Distal causes have yet to be examined, and this will take some time. This has sparked healthy debate in professional and what I'd call higher-functioning enthusiast circles who have an interest in how these incidents can happen. I am sure people who are currently driving on the mainline have chills about the incident, and have stopped to think if such an incident could happen to them or someone they work with. In my view, limited speculation is healthy and should not be suppressed; we have been fortunate on the forum that whatever speculation there has been has been informed, particularly by staff - some of whom sign the route, and some of who are S+T professionals who can give more granular explanation and what should and should not be possible. Speculation has also been respectful of the driver, who has sadly died.

RAIB have also communicated what they do not yet know. They do not know:

the condition of the AWS
whether it was functional
whether it gave an error
whether a warning was given, and then cancelled by the driver
whether a warning was given, not acknowledged, and a fault meant the train was not stopped automatically but the driver induced the braking, or
whether a warning was given, not acknowledged, and the braking experienced by the train was not driver induced but the emergency brakes responding automatically.

And they have told us why they do not know this yet.

I'm not sure what else they can do. The mainstream media does not show much interest in reporting things with surgical precision like RAIB do (every word RAIB put out is careful and deliberate). Was the signal red because the 810 had an AWS fault? Well, sort of, yes, in the sense that had it not had the fault it would have been doing 125mph and been 5 miles away when the 360 cleared the crossover...but also - no, signals don't work like that and they just communicate whether a train or obstruction is in the section. I am not sure the public really care and I doubt an incident of this nature will drastically affect public confidence unless some surprising, gravely serious fault with how signals work is eventually uncovered. And Facebook will always be Facebook, a lot of bored souls on there I'm afraid.

I agree with all this. One thing though, I think elements of the public *do* care to some extent. Thankfully people in this country aren’t nowadays used to seeing trains collide, it simply isn’t something they expect to see happen, so I think it will have shaken confidence a bit, especially for those people using trains regularly and all the more so for those using trains in the same area. Plus there’s the whole “nothing seems to function properly in Britain at the moment” feeling, which I’m not going to get into the rights or wrongs of here, however the incident will certainly feed into that feeling.

It’s very easy for people on here and elsewhere to say that the general public don’t really care or aren’t concerned, however especially in the social media age this isn’t the case. I know one person from my work who was on the Corby train, and one person from a rowing club that I’m a member of was also on there, and who now won’t be able to compete at Henley next week as a result. When there’s a paragraph about it in a rowing club newsletter of all places, it does show wider interest and concern. I’ve had many many messages from people with no interest in railways asking about this, with questions like “how could this happen?” and in particular “surely there is some automated system that stops trains going near each other?”

MartinC · 26 Jun 2026

Harpo said:
No national media outlet ever covers any technical issues well unless it is a trade specific title (e.g. Modern Railways) and at a parochial level the words ‘local’ and ‘crap’ are completely interchangeable where the media are concerned.

I’d trust the writers of Viz over any newspaper on railway details.

See my Post 26.

Harpo · 26 Jun 2026

chriseurope said:
the RAIB doesn't mention anything about considering wider operational issues, such as 1H46 staying on the slow line to pass the stationary 1B67

RTT on the day recorded them as 2.5 minutes apart at Bedford South if I remember correctly? If so there would have been no need to consider doing anything other than following the timetable with 1H46 following 1B67 as happened.

riceuten · 26 Jun 2026

I suspect that the Daily Express are desperate to hang this on the public sector in general, Keir Starmer in particular, and "lazy, overpaid" railways staff. They will inevitably be seeking the Transport Minister's head on a platter and an immediate general election.

To be honest, I would have said exactly the same if the Conservatives had been in power - that's just how this feeble rag operates.

bramling · 26 Jun 2026

riceuten said:
I suspect that the Daily Express are desperate to hang this on the public sector in general, Keir Starmer in particular, and "lazy, overpaid" railways staff. They will inevitably be seeking the Transport Minister's head on a platter and an immediate general election.

To be honest, I would have said exactly the same if the Conservatives had been in power - that's just how this feeble rag operates.

The Express is just a joke, it always has been, nothing that they say merits taking much notice of. The problem is this is actually quite dangerous - after hundreds of “Britain’s told to take cover as hurricane winds about to batter the country” front pages, on the one day when there actually is something serious people won’t take notice.

Sadly most of the rest of the media isn’t much better, and even the ones we’d expect to be more reputable like the BBC have slipped over time.

Having said all this, it could well yet turn out that there are political connections. It’s certainly the case, for example, that the Bedford area is handling 2020s levels of service on a layout essentially designed in the 1980s, it certainly wasn’t designed for today’s intense MML and TL service. *If* RAIB were, for example, to say that the signal layout was contributory in creating a SPAD trap then this would then raise some political questions, both in terms of the railway infrastructure, and successive governments policy of flooding the area with housebuilding. Likewise if the driver had the bulk of his training during Covid then is does raise interesting questions about the quality of that training and how it might have been affected by government-directed COVID restrictions.

Annetts key · 26 Jun 2026

bramling said:
I suspect a lot of this won’t be relevant at Bedford as from what we know there wasn’t much time for anything to be done. It’s quite probable the 360 had been signalled over the junction before the 810 even came to a stand.

Keep in mind that many of the complex route relay interlocking systems (and the computer based systems that followed on) are/were designed to allow the signaller to route a second train right behind the "first" train, not even having to wait until the "first" train clears the section. Note the word "route", the route may be able to be set but that does not always mean that the signal will immediately clear to a proceed aspect.

The only limitation is that if any points need to move, the "first" train has to have cleared the track circuit (or axle counter) section that directly locks the the relevant points, or the track circuit (or axle counter) section that locks the relevant part of the "route locking".

Once the signal has been routed by the signaller, they will very likely then move to their next task, dealing with other things.

Once the "first" train completely clears the signal section, the signal that has just been routed will automaticity clear up to the appropriate aspect (single yellow unless approach control/approach release or other factors/controls are required).

Now, I don't know what actually happened in this particular case and I am not going to speculate. I'm just pointing out how the design of the signalling system (to maximise the efficiently of signalling trains) can work.

To really understand, you really need to experience being on the operating floor of a busy PSB/panel while an experienced signaller goes about their work. How fast they can work a panel, setting different routes is something that may amaze people.

bramling · 26 Jun 2026

Annetts key said:
Keep in mind that many of the complex route relay interlocking systems (and the computer based systems that followed on) are/were designed to allow the signaller to route a second train right behind the "first" train, not even having to wait until the "first" train clears the section. Note the word "route", the route may be able to be set but that does not always mean that the signal will immediately clear to a proceed aspect.

The only limitation is that if any points need to move, the "first" train has to have cleared the track circuit (or axle counter) section that directly locks the the relevant points, or the track circuit (or axle counter) section that locks the relevant part of the "route locking".

Once the signal has been routed by the signaller, they will very likely then move to their next task, dealing with other things.

Once the "first" train completely clears the signal section, the signal that has just been routed will automaticity clear up to the appropriate aspect (single yellow unless approach control/approach release or other factors/controls are required).

Now, I don't know what actually happened in this particular case and I am not going to speculate. I'm just pointing out how the design of the signalling system (to maximise the efficiently of signalling trains) can work.

To really understand, you really need to experience being on the operating floor of a busy PSB/panel while an experienced signaller goes about their work. How fast they can work a panel, setting different routes is something that may amaze people.

Agree on the last point. A lot of people don’t realise that signallers have numerous tasks potentially on the go, and don’t / can’t always monitor every train movement. And in terms of tasks, as you say the mentality has to be “do it, implement it, assume it’s been carried out correctly, move on to the next matter”, otherwise at busy times they would drown in the workload.

I’ve always felt the criticism of the Ladbroke Grove signallers was unfair. That the signallers managed to look at the screen, work out what had happened, work out what was still happening, decide what to do, and implement any kind of response within 33 seconds was reasonable as far as I’m concerned. That isn’t to say that it wasn’t right that there needed to be a focus on sharpening up the response to SPADs in terms of having a set immediate response (remembering that radio was a lot less dependable and user-friendly than it is today), but I felt the rather scathing criticism of the signallers was unwarranted.

riceuten · 26 Jun 2026

bramling said:
I felt the rather scathing criticism of the signallers was unwarranted.

There has to be a scapegoat - preferably not senior management.

Was this the accident that was caused in part by glare on one of the signals making it impossible to read ?

bramling · 26 Jun 2026

riceuten said:
Was this the accident that was caused in part by glare on one of the signals making it impossible to read ?

This was never stated definitively as the cause IIRC, but it was certainly suspected as a possible factor.

E27007 · 26 Jun 2026

bramling said:
Agree on the last point. A lot of people don’t realise that signallers have numerous tasks potentially on the go, and don’t / can’t always monitor every train movement. And in terms of tasks, as you say the mentality has to be “do it, implement it, assume it’s been carried out correctly, move on to the next matter”, otherwise at busy times they would drown in the workload.

I’ve always felt the criticism of the Ladbroke Grove signallers was unfair. That the signallers managed to look at the screen, work out what had happened, work out what was still happening, decide what to do, and implement any kind of response within 33 seconds was reasonable as far as I’m concerned. That isn’t to say that it wasn’t right that there needed to be a focus on sharpening up the response to SPADs in terms of having a set immediate response (remembering that radio was a lot less dependable and user-friendly than it is today), but I felt the rather scathing criticism of the signallers was unwarranted.

Surely the finger of blame points in the direction of management who seemingly did not tackle the safety issue of problem signal SN109, the signal had 8 spad events, the fateful day of Ladbroke Grove the ninth

bramling · 26 Jun 2026

E27007 said:
Surely the finger of blame points in the direction of management who seemingly did not tackle the safety issue of problem signal SN109, the signal had 8 spad events, the fateful day of Ladbroke Grove the ninth

Exactly, which to be fair was heavily criticised by Cullen as well. RAIB don’t directly criticise in the same way, they tend to just report facts. It did feel at times that Cullen was out just to have a go at as many people as possible, perhaps to satisfy those who wanted to see blood drawn.

I felt the Cowden report also seemed to be quite pre-disposed to laying down blame as well, in this case towards the driver. To be fair, this was probably one of the most difficult reports to compose as there was so little hard evidence to go on, and the driver didn’t help himself by having broken rules in having the guard riding up front. But nonetheless the report seemed to weigh heavily on the driver and brush off other factors one by one.

TunbridgeWells · 26 Jun 2026

Not directly related to the Daily Express coverage but I had a query over AWS and this seemed like the most relevant place to ask.

It sounds like a driver receives the same AWS warning noise if the signal passed is amber, double-amber or red. Would it not be better for a "stronger" warning noise to be given for a red signal than for an amber or double-amber?

GB · 26 Jun 2026

TunbridgeWells said:
Not directly related to the Daily Express coverage but I had a query over AWS and this seemed like the most relevant place to ask.

It sounds like a driver receives the same AWS warning noise if the signal passed is amber, double-amber or red. Would it not be better for a "stronger" warning noise to be given for a red signal than for an amber or double-amber?

The issue with the same warnings for yellow and red was brought up and known about decades ago. The consensus is there is no quick or cheap fix for it.

Annetts key · 26 Jun 2026

TunbridgeWells said:
Not directly related to the Daily Express coverage but I had a query over AWS and this seemed like the most relevant place to ask.

It sounds like a driver receives the same AWS warning noise if the signal passed is amber, double-amber or red. Would it not be better for a "stronger" warning noise to be given for a red signal than for an amber or double-amber?

This has been answered in the other related topics, but I will give a summery:
AWS gives a warning in the form of a horn for a signal that is not lit, or is showing a red, single yellow or double yellow. If the driver does not acknowledge the horn within a short time, the bakes are automatically applied.
It gives a bell only for a signal that is lit and showing a green.

The question on having an advanced system has been debated for many, many years. But there are practical problems. Not every signal is wired up the same way, so there is no universal design.

After the Clapham railway accident, ATP was proposed. Two pilot systems were brought into use. One remains and is still operational (on the GWML between Bristol and Paddington). ATP solves many of the problems and is far superior to AWS (and also much better than TPWS). But now it's obsolete. It's no longer made.

The long term solution really is ETCS/ERTMS but that's expensive (there currently is a limited budget) and the railways in this country have had problems with implementation. Although it's in use elsewhere in the world and in the U.K. for fairly small sections of line.

The stop-gap (IMHO) that the railways brought in across the country after the two train crashes on the GWML near London some years ago is TPWS. But, part of the reason that TPWS costs less than ATP or ETCS is that it is only fitted at signals where its considered that there is a high risk, which in practice, most of the time, is signals protecting points or a junction.

The signal that was passed at red in the Bedford accident did not have TPWS fitted because it was not considered high risk.

The topic where this is discussed in detail is here.

Starmill · 26 Jun 2026

bramling said:
Exactly, which to be fair was heavily criticised by Cullen as well. RAIB don’t directly criticise in the same way, they tend to just report facts. It did feel at times that Cullen was out just to have a go at as many people as possible, perhaps to satisfy those who wanted to see blood drawn.

Indeed. Technically, RAIB are not able to apportion lawful liability for anything. Their drafters are great at ensuring they can't be cast as doing so by a prosecutor. Their statutory duties are to contribute to safety improvement and to accurately inform the public and industry by investigating the varied causes of accidents etc. Lord Cullen was of course partly responsible for this approach.

TunbridgeWells · 26 Jun 2026

Annetts key said:
This has been answered in the other related topics, but I will give a summery:
AWS gives a warning in the form of a horn for a signal that is not lit, or is showing a red, single yellow or double yellow. If the driver does not acknowledge the horn within a short time, the bakes are automatically applied.
It gives a bell only for a signal that is lit and showing a green.

The question on having an advanced system has been debated for many, many years. But there are practical problems. Not every signal is wired up the same way, so there is no universal design.

After the Clapham railway accident, ATP was proposed. Two pilot systems were brought into use. One remains and is still operational (on the GWML between Bristol and Paddington). ATP solves many of the problems and is far superior to AWS (and also much better than TPWS). But now it's obsolete. It's no longer made.

The long term solution really is ETCS/ERTMS but that's expensive (there currently is a limited budget) and the railways in this country have had problems with implementation. Although it's in use elsewhere in the world and in the U.K. for fairly small sections of line.

The stop-gap (IMHO) that the railways brought in across the country after the two train crashes on the GWML near London some years ago is TPWS. But, part of the reason that TPWS costs less than ATP or ETCS is that it is only fitted at signals where its considered that there is a high risk, which in practice, most of the time, is signals protecting points or a junction.

The signal that was passed at red in the Bedford accident did not have TPWS fitted because it was not considered high risk.

The topic where this is discussed in detail is here.

Thanks for the reply!

Would it be fair to say that the AWS doesn't know the actual aspect displayed on the signal - It only knows that either it is green (and lit) or that it is either yellow, double yellow, red or unlit? Therefore it isn't possible for it to sound a stronger warning for a red aspect?

norbitonflyer · 26 Jun 2026

TunbridgeWells said:
It sounds like a driver receives the same AWS warning noise if the signal passed is amber, double-amber or red. Would it not be better for a "stronger" warning noise to be given for a red signal than for an amber or double-amber?

Originally AWS (and the earlier GWR ATC) were set uo to give drivers an audible indication of the state of distant signals, which can only show "caution" or "clear", and not stop signals. At that time distant signals were generalky separate from their associated stop signals, rather than combined with the previous stop signal as in modern multiple aspect systems.
And both ATP and AWS were purely binary systems - energised/ not energised. Any more nuanced indication would have been beyond early to mid 20th century technology.

Bedford (Elstow) collision reporting by Daily Express: is this accurate or misleading?

Member

RailUK Forums

Established Member

Member

Established Member

Established Member

Established Member

Established Member

Member

Veteran Member

Member

Established Member

New Member

Verified Rep - Wingin' It! Paul Lucas

Veteran Member

Member

Established Member

Member

Veteran Member

Established Member

Veteran Member

Member

Veteran Member

Member

Veteran Member

New Member

Established Member

Established Member

Veteran Member

New Member

Established Member