• 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!

Wrong-side signalling failure at Wingfield

Status
Not open for further replies.

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,792
Location
West is best
Would I be right in thinking that automatic signals would be tested by manually activating the corresponding track circuits or axle counters?
It depends on the local circumstances. In the past, for Western Region designs, if the signal is a three aspect and normally (with no trains around) ‘sits at’ green, the normal practice is to slip a link to simulate the next track circuit going occupied, or pull a fuse to cause the next signal to go out (black/unlit). That gets you a red aspect.
To get a yellow, you do the above, but to the next signal ahead. That signal then goes red, and the signal that you are testing then goes to yellow.
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Chris M

Member
Joined
4 Feb 2012
Messages
1,060
Location
London E14
The RAIB have just tweeted: https://twitter.com/raibgovuk/status/1605503965457571842
As a result of the work we’ve undertaken following the incident on 26 October at Wingfield, North Derbyshire, this is now a full investigation as there is a wider reach in safety learning than with our Safety Digests.

The tweet links to the page about the incident on their website, https://www.gov.uk/government/news/wrong-side-signalling-failure-at-wingfield which has been updated and is a lot more detailed than previously:
At about 07:02 hrs on 26 October 2022, an empty coaching stock train was travelling on the down main line between Derby and Chesterfield on signals showing green proceed aspects, when it unexpectedly encountered a signal in the South Wingfield area displaying a red danger aspect. As the train was approaching the signal at 110 mph (177 km/h), it was unable to stop before the signal and passed it by about 760 metres. The driver called the signaller straight away to report the incident.

The signaller contacted the driver of the following passenger train on the down main line at 07:07 hrs, not long after this train had departed from Belper. The signaller instructed the driver of this train to proceed at a reduced speed due to the train ahead of it passing a signal at danger. At about 07:19 hrs, the train approached the signal that the previous train had passed at red. The signal was now displaying a yellow caution aspect. After passing the signal at a speed of about 20 mph (32 km/h), the driver of this following train saw the taillights of the first train, which was stationary ahead of it. He brought his train to a controlled stop, with the trains coming to stand about 75 metres apart in the same signal section.

The signal which was passed at red had been disconnected and reconnected the previous night as part of planned track maintenance work. Evidence available to RAIB is that it had been placed back into service after the work with a fault. This caused a wrong side failure, with the signal’s red and yellow aspects being displayed incorrectly.
 

Taunton

Veteran Member
Joined
1 Aug 2013
Messages
12,226
As before, an issue happening to the first train (or combination) to come through after works. Suggested above that the signaller might have thought the initial red a "normal" trackside failure. One would have thought it would be known that the signal had been worked on overnight, now it's reported by the first train as unexpectedly a different aspect to what it should be ... join the dots.

As we are describing prior similar events, I will add the Athelney level crossing accident, first train through after overnight works in the section which had upset the crossing control sequence. Which some knew about the next morning, and some didn't.
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,792
Location
West is best
Unofficially I’ve heard further details that indicate that it is definitely an infrastructure problem. And yes, its due to the work that was carried out before the line was reopened to traffic.

As there will now be an RIAB report, I think it best we wait for this rather than speculating.
 

Falcon1200

Established Member
Joined
14 Jun 2021
Messages
6,514
Location
Neilston, East Renfrewshire
Unofficially I’ve heard further details that indicate that it is definitely an infrastructure problem. And yes, its due to the work that was carried out before the line was reopened to traffic.

That seems to be what the RAIB statement in Post #92 says; The signal had been wrongly reconnected causing Yellow to be displayed vice Red, and vice versa, a very serious wrong side failure. The investigation now is presumably how and why this occurred, and measures to prevent it happening again.
 

Foggycorner

Member
Joined
16 Nov 2016
Messages
202
Location
bolton
That seems to be what the RAIB statement in Post #92 says; The signal had been wrongly reconnected causing Yellow to be displayed vice Red, and vice versa, a very serious wrong side failure. The investigation now is presumably how and why this occurred, and measures to prevent it happening again.
If the work had been done and tested to the SMTH then it would not have happened as was said in post 17
 

Chris M

Member
Joined
4 Feb 2012
Messages
1,060
Location
London E14
The investigation needs to go deeper than that, they need to established exactly what happened. There are (at least) two aspects to the investigation: A. what was done (or not done) that caused the signal to display incorrect aspects; and B. Why was this not detected?
Part A will want to establish why the relevant thing was (not) done and what can be done to prevent that happening again.
As I see it there are several theoretical possibilities (not all of equal likelihood) regarding part B:
1. The proper procedures were followed but they failed to prevent the wrong-side failure.
2. The proper procedures were partially followed
3. The proper procedures were not followed at all.
In the case of 1, they will need to establish why they failed to prevent it and what needs to change to that it does in future
In the case of 2 and 3 they will also need to establish why the procedures were not followed (completely), whether following them completely would have detected the problem and what needs to happen to ensure that procedures which would detect the problem are followed in the future.
 

Foggycorner

Member
Joined
16 Nov 2016
Messages
202
Location
bolton
The investigation needs to go deeper than that, they need to established exactly what happened. There are (at least) two aspects to the investigation: A. what was done (or not done) that caused the signal to display incorrect aspects; and B. Why was this not detected?
Part A will want to establish why the relevant thing was (not) done and what can be done to prevent that happening again.
As I see it there are several theoretical possibilities (not all of equal likelihood) regarding part B:
1. The proper procedures were followed but they failed to prevent the wrong-side failure.
2. The proper procedures were partially followed
3. The proper procedures were not followed at all.
In the case of 1, they will need to establish why they failed to prevent it and what needs to change to that it does in future
In the case of 2 and 3 they will also need to establish why the procedures were not followed (completely), whether following them completely would have detected the problem and what needs to happen to ensure that procedures which would detect the problem are followed in the future.
how long is your piece of string and can you push it up hill
as I said following SMTH there would not have been a problem and as someone who did this for 30 years I might just know what is involved!!
 

alxndr

Established Member
Joined
3 Apr 2015
Messages
1,678
how long is your piece of string and can you push it up hill
as I said following SMTH there would not have been a problem and as someone who did this for 30 years I might just know what is involved!!
I think the point that @Chris M was trying to make is that it’s not enough to simply say that they should have SMTH tested it. The RAIB will need to investigate if SMTH was carried out, if the fault would have been identified by a correctly performed SMTH test, and why SMTH wasn’t carried out fully if that was the case (and look into any other failings that contributed that might).

I assume from your post that you’re of an S&T background, so to use an analogy, what do you do if a fuse blows? You don’t just fix the glaringly obvious failure and stick a new one in and hope for the best because it might just keep blowing. You investigate, take current readings, replace other failing equipment to address the root cause, etc. The RAIB is doing that investigating and while they can’t fix the root cause themselves they can recommend that something’s done about it.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,610
Location
Nottingham
how long is your piece of string and can you push it up hill
as I said following SMTH there would not have been a problem and as someone who did this for 30 years I might just know what is involved!!
The widely-circulated Network Rail Powerpoint addresses the immediate issue that SMTH almost certainly wasn't followed, and acts a a general briefing on the importance of doing so. That's clearly an essential step to reduce the risk of similar incidents in the short term.

The root causes are more for the RAIB investigation: https://www.gov.uk/government/news/wrong-side-signalling-failure-at-wingfield
It will include consideration of the roles, responsibilities and competence management of the staff involved, the factors that influenced their actions and the testing processes followed for the planned overnight work to the signalling equipment. It will also consider any underlying management factors, including the rail industry response to RAIB’s recommendations made its Cardiff East Junction (RAIB report 15/2017) and Waterloo (RAIB report 19/2018) investigations.
 
Last edited:

Foggycorner

Member
Joined
16 Nov 2016
Messages
202
Location
bolton
I think the point that @Chris M was trying to make is that it’s not enough to simply say that they should have SMTH tested it. The RAIB will need to investigate if SMTH was carried out, if the fault would have been identified by a correctly performed SMTH test, and why SMTH wasn’t carried out fully if that was the case (and look into any other failings that contributed that might).

I assume from your post that you’re of an S&T background, so to use an analogy, what do you do if a fuse blows? You don’t just fix the glaringly obvious failure and stick a new one in and hope for the best because it might just keep blowing. You investigate, take current readings, replace other failing equipment to address the root cause, etc. The RAIB is doing that investigating and while they can’t fix the root cause themselves they can recommend that something’s done about it.
The point I was trying to make was that if SMTH had been followed correctly there would not have been a failure after hand back of the equipment and yes I agree that RAIB will have to investigate to find who was at fault
At the end of the day you can only make things fool proof not idiot proof
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
28,610
Location
Nottingham
The point I was trying to make was that if SMTH had been followed correctly there would not have been a failure after hand back of the equipment and yes I agree that RAIB will have to investigate to find who was at fault
At the end of the day you can only make things fool proof not idiot proof
RAIB does not attribute blame, although it's often pretty obvious who made the mistake. However, if a mistake was made, there's potential for someone else to make the same mistake in future. That's what they are trying to prevent.

There's a risk though, that the original mistake might have been due to pressure of work, and the suggested solution might just increase the workload.
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,792
Location
West is best
I believe from what I have unofficially heard, that part of the problem was the railway trying to make the system ‘idiot proof’.

It’s currently not possible to completely remove human beings from the specification of, design of, construction of and installation of signalling equipment. And only people can do the required safety and functional testing of said equipment (although in some cases they may be assisted by technology).

As in a lot of engineering systems, reducing the risks of human error in one part, may not remove the possibility of human error, but may transfer it to somewhere else in the system.

Hence, by now, it’s almost certain that the equipment involved in the ‘failure’ and what actually happened is very well understood. But that the whole system including management, design and specifications and the systems of testing are still being looked at.

There are wider lessons that need to be learned.
 

Nicholas Lewis

Established Member
Joined
9 Aug 2019
Messages
7,991
Location
Surrey
Curious that six weeks after the incident RAIB decides to upgrade to a full investigation suggests that they, or NR, have uncovered something more significant as on the information so far provided is its a simple failure to follow due process. Not sure the relevance of Cardiff and Waterloo as they were both Works Testing driven. The more recent incident at Dalwhinnie station would seem relevant as well.

This report will be interesting but will be a while before we see it now.
 

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,494
Location
Moorpark, CA
I believe from what I have unofficially heard, that part of the problem was the railway trying to make the system ‘idiot proof’.

It’s currently not possible to completely remove human beings from the specification of, design of, construction of and installation of signalling equipment. And only people can do the required safety and functional testing of said equipment (although in some cases they may be assisted by technology).

As in a lot of engineering systems, reducing the risks of human error in one part, may not remove the possibility of human error, but may transfer it to somewhere else in the system.

Hence, by now, it’s almost certain that the equipment involved in the ‘failure’ and what actually happened is very well understood. But that the whole system including management, design and specifications and the systems of testing are still being looked at.

There are wider lessons that need to be learned.
Recently-retired (after nearly 44 years) Signal Engineer here. I'm minded of something someone said in a meeting some years ago that "idiot-proof" systems/procedures are vulnerable to the World's uncanny ability to continue producing better idiots: but, I digress.

Part of the problem with "procedures" is that it gets hammered into people that they must "follow the procedure" which can lead to them slavishly doing so to the extent that common sense goes out the window and a shortcoming in the "procedure" gets overlooked by the people following it. Having said that, as others have said, this shouldn't have happened - SMTH has been around now for what, 30 years? I'd have thought that if there was a loophole in the relevant Test Plan, it would have been discovered by now. We'll (officially) find out the full story in a year or so.

Coincidentally, something like this happened on LA Metro a number of years ago where two of the aspects were crossed when reconnected: the signal was signed back into use, but what was of more concern to me was that over a dozen trains passed the wrong aspect for the way they were routed before a driver reported/challenged it.....
 

snowball

Established Member
Joined
4 Mar 2013
Messages
8,144
Location
Leeds
Not just on the railways ...


A hospital is investigating how a pair of metal surgical forceps were left inside a patient after they had been stitched up after abdominal surgery.

Worcestershire Acute Hospitals NHS trust has apologised unreservedly and said the incident at Redditch’s Alexandra hospital was “exceptionally rare”.

The medical blunder only became apparent after a seven-hour abdominal procedure last month, according to BBC Midlands, when the forceps were reported to be missing.

The worst fears of medics were confirmed when the missing 15cm arterial clamp was found by an X-ray while the patient was still under anaesthetic.

The surgical instrument could not be immediately removed and the patient was moved to intensive care overnight before another operation was performed the next day to retrieve the clamp.
 

Taunton

Veteran Member
Joined
1 Aug 2013
Messages
12,226
At the end of the day you can only make things fool proof not idiot proof
There can be more to it than that, design can make mismatch mistakes not possible. Don't believe you can't improve. To take examples from cars, which might be familiar, it was long possible to connect up the spark plug leads incorrectly, so the engine didn't work. A longstanding procedure was thus to label them before disconnecting, but you still had an issue if you didn't do that. Only in more recent times did manufacturers build wiring harnesses that were impossible to connect up in other than the correct sequence. Likewise putting the wrong fuel, diesel or petrol, into cars was long seen as something only an idiot would do, and was treated with derision, not thinking that there were families that had one car of each type, and certain familiarities developed, or people rented a car different to their own, etc, etc. It took a long while to develop nozzles and fillers that didn't fit the wrong type each way

So that's what can be considered. Are the wire connectors different for each aspect so you just can't connect them incorrectly. Are both ends of the connector colour coded? If you take the red and yellow colour filters out to clean them is it impossible to put them back reversed? That sort of thing. The aviation industry has long been experts at engineering design avoiding such issues.
 

alxndr

Established Member
Joined
3 Apr 2015
Messages
1,678
Having said that, as others have said, this shouldn't have happened - SMTH has been around now for what, 30 years? I'd have thought that if there was a loophole in the relevant Test Plan, it would have been discovered by now. We'll (officially) find out the full story in a year or so.
You'd think so but there was a major overhaul of a lot of cable testing test plans earlier this year that resolved some confusion over selection of suitable test plans and how to implement them. I'm lead to believe its something they try to do on a semiregular basis.
Likewise putting the wrong fuel, diesel or petrol, into cars was long seen as something only an idiot would do, and was treated with derision, not thinking that there were families that had one car of each type, and certain familiarities developed, or people rented a car different to their own, etc, etc. It took a long while to develop nozzles and fillers that didn't fit the wrong type each way
It's still possible to put petrol in a diesel. I know people who have done so and only realised when they broke down a mile or two away. Their excuse was that they were distracted. Usually they were perfectly competent and not idiots!
Are the wire connectors different for each aspect so you just can't connect them incorrectly. Are both ends of the connector colour coded? If you take the red and yellow colour filters out to clean them is it impossible to put them back reversed? That sort of thing. The aviation industry has long been experts at engineering design avoiding such issues.
No such safe guards unfortunately. Plug couplers mitigate these sorts of risks somewhat, but doesn't appear to be applicable in this case.

Cable cores should have ident beads on, and these should be checked to ensure they're correct before carrying out work, but as we already know, sometimes the holes in the Swiss cheese line up. The physical cable cores should be labelled within the cable as well, but often these are very hard to read, especially if the cable isn't brand new.
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,792
Location
West is best
Well, although not necessarily relevant to this incident, just to point out that idiots can and do find ways.

As an example, with conventional filament lamp signal heads, the lamps used have three locating pins in the lamp base and corresponding ‘holes’ in the fitting (rather than two in a normal consumer lamp). The three pins are deliberately spaced differently relatively to each other to prevent the lamp being put in the holder with the incorrect orientation (the lamp is a focused type with two filaments, each filament being connected to one of the base contacts, the outer metal base being the common) so the orientation is important).

But of course, one day a ‘technician’ did manage to fit one of these lamps in the wrong way round. And not only did the lamp not work correctly, the ‘team leader’ could not understand what was going on…

Like I said earlier, the design of the equipment and the procedures also need to be included.

== Doublepost prevention - post automatically merged: ==

Believe me, plug couplers, and pinned receptacles etc. are not always idiot proof.

SMTH is still riddled with inconsistencies unfortunately. Those with a good understanding of the signalling engineering principles, the equipment and the procedures know what is supposed to be done. But if you just read some of the sections literally, some parts either don’t make sense, or contradict other sections that cover the same topic. The fundamental principles are okay, it’s mainly the document being badly worded, or steps not being in a logical order.
 
Last edited:

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,494
Location
Moorpark, CA
You'd think so but there was a major overhaul of a lot of cable testing test plans earlier this year that resolved some confusion over selection of suitable test plans and how to implement them. I'm lead to believe its something they try to do on a semiregular basis.

It's still possible to put petrol in a diesel. I know people who have done so and only realised when they broke down a mile or two away. Their excuse was that they were distracted. Usually they were perfectly competent and not idiots!

No such safe guards unfortunately. Plug couplers mitigate these sorts of risks somewhat, but doesn't appear to be applicable in this case.

Cable cores should have ident beads on, and these should be checked to ensure they're correct before carrying out work, but as we already know, sometimes the holes in the Swiss cheese line up. The physical cable cores should be labelled within the cable as well, but often these are very hard to read, especially if the cable isn't brand new.
50 year old installation with (possibly) the original cable where the core numbers are all but unreadable. Not identified due to time constraints? Holes lined up indeed.
 

Class 170101

On Moderation
Joined
1 Mar 2014
Messages
8,661
50 year old installation with (possibly) the original cable where the core numbers are all but unreadable. Not identified due to time constraints? Holes lined up indeed.
They might be bad to read however how about putting red tape on the red wire and yellow tape on the yellow wire before you remove anything at the start of the job? (or some other method that does not require reading 50 year old handwriting etc)
 

MadMac

Established Member
Joined
13 Jun 2008
Messages
1,494
Location
Moorpark, CA
They might be bad to read however how about putting red tape on the red wire and yellow tape on the yellow wire before you remove anything at the start of the job? (or some other method that does not require reading 50 year old handwriting etc)
Precisely. If the existing identification is difficult to read, identify them yourself.
 

Annetts key

Established Member
Joined
13 Feb 2021
Messages
3,792
Location
West is best
They might be bad to read however how about putting red tape on the red wire and yellow tape on the yellow wire before you remove anything at the start of the job? (or some other method that does not require reading 50 year old handwriting etc)
The relevant standards are already clear, if a wire is not already correctly and uniquely identified, then a temporary label (or labels) should be used before any disconnection takes place.

Normally wires and cable cores are labelled with a number, sometimes prefixed with one or two letters.

However, as I keep saying, speculation in this case is not helpful, as the problem (as I’m lead to believe) is related to design issues.
 

Signal Head

Member
Joined
26 May 2013
Messages
460
The relevant standards are already clear, if a wire is not already correctly and uniquely identified, then a temporary label (or labels) should be used before any disconnection takes place.

Normally wires and cable cores are labelled with a number, sometimes prefixed with one or two letters.

However, as I keep saying, speculation in this case is not helpful, as the problem (as I’m lead to believe) is related to design issues.
That's an interesting development, since I wouldn't expect any design to be issued for an SMTH relaying job. Unless at the same the signal head was being changed to LED?

I'm aware of at least one instance of design being incorrect in a similar case, where the designer had failed to take into account the different terminal numbering between filament and LED heads, resulting in (I think), the Red feed being applied to the Green, and the Yellow feed going to the (non-existent) top Yellow. Fortunately found before commissioning.

Another possibility - the head had already been renewed as LED but the drawings hadn't been updated to reflect that. This is quite a common occurrence in my experience. Similar effect - terminal, and probably cable core numbers, will be different from the drawings, quite easy to get confused, especially if under pressure but, as stated earlier, the final aspect test should have found that mistake.

Another technically interesting RAIB report when it comes out.
 

Gaelan

Member
Joined
3 Apr 2023
Messages
997
Location
Edinburgh
RAIB report is now out: https://www.gov.uk/government/news/...in-the-same-signal-section-at-south-wingfield

At about 07:02 hrs on 26 October 2022, a train travelling between Derby and Chesterfield unexpectedly encountered a signal displaying a red (stop) aspect. The previous signal had displayed a green (proceed) aspect. As the train was travelling at 100 mph (161 km/h), it was unable to stop before the red signal and passed it by about 760 metres. The train’s driver called the signaller straight away to report the incident. About 17 minutes later, the following train approached the signal which was now displaying a yellow (caution) aspect. After passing the signal, while travelling at about 20 mph (32 km/h), the driver of the second train saw the taillights of the first train stationary ahead of it and braked to a controlled stop. The second train stopped about 75 metres from the rear of the first train, with both trains now in the same signal section. There were no significant consequences and both trains were able to continue their journeys after obtaining permission from the signaller.

The signal had displayed incorrect aspects to the drivers of the two trains as the wiring controlling its red and yellow aspects was crossed on two terminals in a nearby equipment cabinet, which was where a cable running to the signal was connected to the rest of the signalling system. This cable had been disconnected and reconnected during track engineering work the previous night and this work had introduced the wiring cross, which was not identified when the signal was tested afterwards. The testing was affected by a combination of time pressure, tester workload and possibly by unfamiliarity with the configuration of the signalling equipment. An underlying factor was that Network Rail had taken steps to assure the signal maintenance testing carried out by its own staff but had not yet included testers employed by contractors. A second underlying factor was that no one was carrying out any signalling related assurance activities when this type of track engineering work was taking place.

RAIB noted that the potential for a collision between the two trains was reduced by the actions taken by the signallers at East Midlands Control Centre and the second train driver. RAIB observed four issues with the testing work covering test records, tester licensing, deficiencies with drawings, and the omission of tests the previous night. RAIB also observed that while initial welfare checks were carried out for both drivers, follow-up post-incident welfare checks were only carried out for one of the drivers involved.

Recommendations
As a result of the investigation, RAIB has made five recommendations. The first two are addressed to Bridgeway Consulting and Randstad Solutions and seek to enhance the non-technical skills among the staff working for them, with specific emphasis on effective communication, safe decision‑making, and safe behaviours when placed under time pressure. The third is for Network Rail to better manage the workload on lead testers. The fourth is for Network Rail to implement measures to better assure itself that signal maintenance testing by contractors on this type of track engineering work is to the required standard. The fifth, also addressed to Network Rail, is to provide testers with a means of recording the test steps when a signal’s aspects are tested.

RAIB has also identified six learning points. The first highlights the importance of signal maintenance testers following the required testing process and the second is about staff working in management or supervisory roles not placing testers under undue time pressure to complete their work, even when they are under time pressure themselves due to work overrunning. The third and fourth highlight the importance of communication between signallers and train drivers when an incident occurs. The fifth is a reminder about testing signalling equipment, particularly track circuits, after engineering work has taken place on the track, and the sixth is a reminder to carry out follow-up post-incident checks with all drivers involved in a signalling irregularity.
 

Gaelan

Member
Joined
3 Apr 2023
Messages
997
Location
Edinburgh
It seems like an obvious test would be to place a track circuit clip after the signal, and see if it turns red - is there a reason this isn't part of the procedure?
 

ainsworth74

Forum Staff
Staff Member
Global Moderator
Joined
16 Nov 2009
Messages
31,072
Location
Redcar
How close can you get to recreating Clapham Junction without recreating Clapham Junction? Good grief that's a slightly chilling read in the summary.

It seems like an obvious test would be to place a track circuit clip after the signal, and see if it turns red - is there a reason this isn't part of the procedure?
Wouldn't work in axle counter areas of course.
 
Status
Not open for further replies.

Top