RAIB: Collision at Stainforth Road level crossing

[LAX54]

Im sure a signalling fratenity bod will know, but I'm sure some if not all AHBs have a timeout relay so that they automatically raise after a certain amount of time if a train hasn't struck out. ISTR that it will then show as failed in the box.
So if this is a long freight that's been checked at a signal for a while whilst overhanging the crossing that may be the cause.

No, they should not raise, if a train fails within the crossing strike in / strike out tracks/treadles, then yes, there will be a failed indication in the box, with an alarm, and the booms will stay down, and all trains on the other line (if there is one) cautioned and not to pass over crossing until they can see its clear, just in case you get the fraternity that like to zig zag, be interesting to see what happened here ! and the car that went over previously that called the Police, who called Control, who called the Signaller...why not stop and use the emergency phone on the crossing direct to Signalman ?

 

[krus_aragon]

No, they should not raise, if a train fails within the crossing strike in / strike out tracks/treadles, then yes, there will be a failed indication in the box, with an alarm, and the booms will stay down, and all trains on the other line (if there is one) cautioned and not to pass over crossing until they can see its clear, just in case you get the fraternity that like to zig zag, be interesting to see what happened here ! and the car that went over previously that called the Police, who called Control, who called the Signaller...why not stop and use the emergency phone on the crossing direct to Signalman ?

In the dark, with your headlights facing away from the crossing (because you've just crossed the line yourself), the emergency phone may not be particularly visible. Unless you've had occasion to use one before, you might not know that there's even a phone there. The layman may not be aware of how much quicker an emergency phone would be, and if you know you've got a mobile phone in your pocket anyway...

 

[Bromley boy]

No, they should not raise, if a train fails within the crossing strike in / strike out tracks/treadles, then yes, there will be a failed indication in the box, with an alarm, and the booms will stay down, and all trains on the other line (if there is one) cautioned and not to pass over crossing until they can see its clear, just in case you get the fraternity that like to zig zag, be interesting to see what happened here ! and the car that went over previously that called the Police, who called Control, who called the Signaller...why not stop and use the emergency phone on the crossing direct to Signalman ?

I don’t mean to doubt you but you 100% sure that’s correct for all AHB crossings?

I also seem to remember being told during my rules course that they would automatically time out if a train failed across them, and that there’s absolutely no means of controlling them from any signal box.

 

[DelW]

You have to bear in mind that 50% of drivers are, by definition, below average ability.

To be both pedantic and maybe off-topic, 50% of drivers are, by definition, below median ability. Anything up to one fewer than the sample size may be below average ability, in the commonest sense of the mean value. Meanwhile back to the proper discussion....

 

[twpsaesneg]

No, they should not raise, if a train fails within the crossing strike in / strike out tracks/treadles, then yes, there will be a failed indication in the box, with an alarm, and the booms will stay down, and all trains on the other line (if there is one) cautioned and not to pass over crossing until they can see its clear, just in case you get the fraternity that like to zig zag, be interesting to see what happened here ! and the car that went over previously that called the Police, who called Control, who called the Signaller...why not stop and use the emergency phone on the crossing direct to Signalman ?

Ahh! Thanks - for some reason I thought they raised too.

 

[GB]

Looks to me the crossing is designed so as that when the train hits the strike out treadle, a timer starts, the timer reaches zero then the reopening sequence activates. With the train having struck out but come to a stand there was nothing else to prevent the crossing from opening. ie no independent TC to hold them down.

Incidentally I remember working a manual gated crossing. Up trains you had to wait for the train to be 20 meters clear before getting the release to throw the gates. Down trains you got the release with half the train still on the crossing. Not and issue with the train at speed as the train was clear by time you got out. But slow trains you would find the rear of the train still to go over.

 

[Dai Corner]

Looks to me the crossing is designed so as that when the train hits the strike out treadle, a timer starts, the timer reaches zero then the reopening sequence activates. With the train having struck out but come to a stand there was nothing else to prevent the crossing from opening. ie no independent TC to hold them down.

Incidentally I remember working a manual gated crossing. Up trains you had to wait for the train to be 20 meters clear before getting the release to throw the gates. Down trains you got the release with half the train still on the crossing. Not and issue with the train at speed as the train was clear by time you got out. But slow trains you would find the rear of the train still to go over.

I assume it's a compromise between the barriers staying down until the longest, slowest train has cleared the crossing and rising as soon as the shortest, fastest train has.

 

[Chris M]

why not stop and use the emergency phone on the crossing direct to Signalman ?

Is there a phone at every level crossing?
I describe myself as a rail enthusiast and I don't know the answer to this question, so I don't know why you would expect the average member of public to know that (a) there is a phone, (b) that it's there for public use in appropriate circumstances, (c) that this is an appropriate circumstance, and (d) that this will directly contact the signaller (not a call centre or something).
Also, are you sure that even if they did know all that that, stopping, going back to the crossing, finding the telephone, working out how to use it*, waiting for the signaller to answer and explaining to them signaller is going to get a quicker response than calling the police from your bluetooth headset immediately?

*I've never had cause to use a level crossing telephone, but I remember reading an accident report (one of the early RAIB ones I think) where someone failed to get in contact with the signaller because they assumed the phone worked like the motorway emergency phones they were familiar with. The motorway phones connect automatically when you lift the receiver, the railway ones - at least at the crossing concerned, you have to press a button - but all the illumination of the instructions telling you this went out when you lifted the receiver (the theory being that this highlighted the button needing to be pressed). I have had to use the motorway phones on I think three occasions, and had I not read the report in question I strongly suspect I would have assumed they worked in the same way.

 

[tsr]

Is there a phone at every level crossing?
I describe myself as a rail enthusiast and I don't know the answer to this question, so I don't know why you would expect the average member of public to know that (a) there is a phone, (b) that it's there for public use in appropriate circumstances, (c) that this is an appropriate circumstance, and (d) that this will directly contact the signaller (not a call centre or something).
Also, are you sure that even if they did know all that that, stopping, going back to the crossing, finding the telephone, working out how to use it*, waiting for the signaller to answer and explaining to them signaller is going to get a quicker response than calling the police from your bluetooth headset immediately?

*I've never had cause to use a level crossing telephone, but I remember reading an accident report (one of the early RAIB ones I think) where someone failed to get in contact with the signaller because they assumed the phone worked like the motorway emergency phones they were familiar with. The motorway phones connect automatically when you lift the receiver, the railway ones - at least at the crossing concerned, you have to press a button - but all the illumination of the instructions telling you this went out when you lifted the receiver (the theory being that this highlighted the button needing to be pressed). I have had to use the motorway phones on I think three occasions, and had I not read the report in question I strongly suspect I would have assumed they worked in the same way.

No, there isn't a phone at every level crossing. However, there will usually be a way to directly contact the signaller, whether that's a level crossing phone or an adjacent signalbox, for example. (Foot crossings are different, but unlikely to be relevant in this specific scenario.) AHBs are usually in areas remote from signalboxes - in fact, I can't think of any which are immediately next to a box - and therefore tend to have a phone contained in one of the distinctive yellow boxes. This is for the use of the general public and train crew, if need be, and can sometimes be used by road vehicle drivers for non-emergency matters. At various crossings - not AHBs, but usually "occupational" crossings with manual gates - the phone will sometimes be used to confirm the line is clear for any vehicle to cross. At others, including AHBs, signs will either direct the user to use the level crossing phone or to dial a specific phone number for permission to cross with animals or cumbersome vehicles, which might take longer to cross than the automatic warning sequence allows.

There are also various different types of phone around the network which contact the signaller in different ways. Some may have one or more buttons for different lines, others (like SPTs) may connect automatically once you lift the handset. Away from level crossings, there are other types of phones, including Emergency Telephones, which feature a keypad for dialling within the private railway phone network.

When it comes to level crossings, I should imagine there are at least one or two types of crossings which each individual road vehicle driver will have never seen. This will vary around the country. In my area, most drivers will never have seen an occupational crossing with manual gates or even manual barriers, but 20 miles away, they might have several in their area, but no AHB crossings. There seems to be little emphasis for learner drivers in being assessed on all the different types. Even train crew may not readily remember the precise nuances of every single variation, even if they know the crossings in their patch by name, location, whether or not it's monitored, and an approximate level of risk. Outside technical Network Rail circles or this forum, I also doubt there are many people who know what each acronym means!

And in terms of emergency phones, it's hard to know what people think they should be used for. So much emphasis is placed on getting people away from crossings if there's an incident and a train approaches - but relatively little on what the appropriate contact methods are.

 

[Taunton]

No, they should not raise, if a train fails within the crossing strike in / strike out tracks/treadles, then yes, there will be a failed indication in the box, with an alarm, and the booms will stay down, and all trains on the other line (if there is one) cautioned and not to pass over crossing until they can see its clear

Unless you are at Athelney, Somerset, where the AHB sequence was known to have been disrupted by a maintenance unit running wrong line over the treadles - but the next service was then allowed to run at 100mph over the crossing, with unfortunate results. Oh dear! I seem to have cast doubt on the conclusion of the RAIB accident report - just like the rest of Somerset has done.

the car that went over previously that called the Police, who called Control, who called the Signaller...why not stop and use the emergency phone on the crossing direct to Signalman ?

Isn't telling the Police about any concerning issue what the incessant "See It, Say It, Sorted" autobabble is all about ...

 

[Signal Head]

Drawings I have (for a different AHB) suggest that the crossing will reset after 2 minutes if the strike-out treadle and track circuit aren't operated. This caters for the false activation of the crossing by right side failure of the strike-in track circuit, and is intended to limit the road closed time to avoid encouraging road users to zig zag round the barriers.

The train seems to have stopped out of course, not at a signal. Having a signal within the max train length after the crossing would be an obvious cause of potential blocking back by trains, onto the crossing, and would be avoided by design.

I don't know what happens if the strike-out is operated but the strike-in remains active, but I suspect it will time out in the same manner, which could be what has happened here.

I'm not immediately sure if there is a way to protect against this, eg disabling the timeout once the strike-out has occurred might help (ie it is then known that a train has passed over the crossing) but there may be other problems presented by having that control.

 

[Tomnick]

Unless you are at Athelney, Somerset, where the AHB sequence was known to have been disrupted by a maintenance unit running wrong line over the treadles - but the next service was then allowed to run at 100mph over the crossing, with unfortunate results. Oh dear! I seem to have cast doubt on the conclusion of the RAIB accident report - just like the rest of Somerset has done.

In that case, the 'failed' indication would have been displayed in the box in due course - but the barriers had been down for such a short period of time (albeit longer than normal) at the time of the collision that I don't think it would even have got that far! The panel were well aware of the 'issue' and kept the protecting signal at danger for as long as possible to keep the 'barriers down' time as short as possible.

 

[Chris M]

When it comes to level crossings, I should imagine there are at least one or two types of crossings which each individual road vehicle driver will have never seen. This will vary around the country. In my area, most drivers will never have seen an occupational crossing with manual gates or even manual barriers, but 20 miles away, they might have several in their area, but no AHB crossings. There seems to be little emphasis for learner drivers in being assessed on all the different types.

I was living in Cheddar in Somerset when I learned to drive, the test centre was in Weston-super-Mare so that was where most of my lessons were. I don't know of any level crossings in the area, let alone multiple types. How do you propose that I could have been assessed on them?

 

[tsr]

I was living in Cheddar in Somerset when I learned to drive, the test centre was in Weston-super-Mare so that was where most of my lessons were. I don't know of any level crossings in the area, let alone multiple types. How do you propose that I could have been assessed on them?

I don't propose such a thing... at least, not with practical testing.

That's my point - it's hard to figure all that stuff out!

Perhaps there could be clearer testing at the Theory / Hazard Perception stages. Goodness knows really.

 

[alxndr]

I don't know what happens if the strike-out is operated but the strike-in remains active, but I suspect it will time out in the same manner, which could be what has happened here.

Google street view shows the end of a track circuit on the approach to the crossing, so if the wagons were only partially covering the crossing it seems unlikely that the strike-in would still be active, although I don't have intimate knowledge of that crossing to say for certain.

 

[Dai Corner]

I was living in Cheddar in Somerset when I learned to drive, the test centre was in Weston-super-Mare so that was where most of my lessons were. I don't know of any level crossings in the area, let alone multiple types. How do you propose that I could have been assessed on them?

There are a couple - Puxton & World and Hewish (Huish). I also learned to drive in Weston-super-Mare and my instructor never took me there or, as far as I remember, ever mention level crossings. He would have known that they weren't on the test routes so there was no need for me to be familiar with them to get through my test. I just needed to know what the Highway Code said in case I was asked a question about them which I wasn't.

 

[LAX54]

Unless you are at Athelney, Somerset, where the AHB sequence was known to have been disrupted by a maintenance unit running wrong line over the treadles - but the next service was then allowed to run at 100mph over the crossing, with unfortunate results. Oh dear! I seem to have cast doubt on the conclusion of the RAIB accident report - just like the rest of Somerset has done.


Isn't telling the Police about any concerning issue what the incessant "See It, Say It, Sorted" autobabble is all about ...

I'd argue that not stopping his car and reporting the incident via the phones provided, could be construed as a partial contribution to the accident ! saw he 'Saw It' but didn't 'Say it' ! .... not sorted

 

[Dai Corner]

I'd argue that not stopping his car and reporting the incident via the phones provided, could be construed as a partial contribution to the accident ! saw he 'Saw It' but didn't 'Say it' ! .... not sorted

I doubt if the average person would recognise the significance of the wrong-line movement.

 

[Taunton]

In that case, the 'failed' indication would have been displayed in the box in due course - but the barriers had been down for such a short period of time (albeit longer than normal) at the time of the collision that I don't think it would even have got that far! The panel were well aware of the 'issue' and kept the protecting signal at danger for as long as possible to keep the 'barriers down' time as short as possible.

But (Athelney accident) AHBs do not have "protecting signals". In this case, because of the upset of the treadles by the wrong-line works train, the barrier sequence started as soon as the route was set for the first train of the day (I don't think this mode of their operation was generally known outside the inner technical circle). And the fact that the train was running at full 100mph and had never even seen any cautionary aspect in the far distance on a line with very widely spaced signals gives a bit of a clue about how far in advance that was. It would have been good if the RAIB inspector had been a bit more penetrating with their questioning, or understood what didn't quite add up.

Accidents with the first train of the day in a particular situation after overnight works have disrupted the normal signalling sequence are of course not unknown elsewhere. Think Clapham, and last year at Waterloo.

 

[Tomnick]

But (Athelney accident) AHBs do not have "protecting signals". In this case, because of the upset of the treadles by the wrong-line works train, the barrier sequence started as soon as the route was set for the first train of the day (I don't think this mode of their operation was generally known outside the inner technical circle). And the fact that the train was running at full 100mph and had never even seen any cautionary aspect in the far distance on a line with very widely spaced signals gives a bit of a clue about how far in advance that was. It would have been good if the RAIB inspector had been a bit more penetrating with their questioning, or understood what didn't quite add up.

Accidents with the first train of the day in a particular situation after overnight works have disrupted the normal signalling sequence are of course not unknown elsewhere. Think Clapham, and last year at Waterloo.

They do have "protecting signals" in as much as a signal in rear can be replaced to danger to protect them, and any stop signal inside the controls will be interlinked with the crossing's operation as was the case at Athelney (to prevent the barriers being down for an excessively long time when that signal remains at danger). I seem to remember the RAIB report stating that the barriers had been down for 90-odd seconds when the collision occurred (plenty of time for the protecting signal to be cleared in time for the train to receive a green aspect at the signal in rear), which is less than the time that the barriers would have to be down for to present a 'failed' indication. I'd suggest that the method of operation employed, if not an official practice, was a sensible means of passing the first train without keeping the barriers down for an excessive amount of time but also whilst avoiding a delay to that first train - whereas the motorist's decision to weave around the barriers having waited less than two minutes is harder to understand.

 

[Taunton]

The Athelney report only gives little hints, scattered around, of the situation. The distant UA136 behind the "protecting" signal is 3km in the rear, back at Cogload Junction, where the train would still be accelerating from the Taunton stop, and is visible well before that. The driver only ever saw greens. Some of us have our own thoughts on when the route was called, and the crossing sequence thus started. And why the signalling centre thought the crossing being out of sync was already resolved. Read the report closely several times and you will spot it.

The fact that this cavalier way of working unchecked after a crossing got out of sync was not followed by other signalboxes in the Region, and was banned forthwith at Exeter box after the accident (described in report) is a bit of a giveaway.

 

[GB]

It is my understanding there were different understandings on what “approach to” meant in regards to clearing the signal after being out of sync. Half the signallers thought it meant when the train was approaching and close to that signal, others thought it was clearing so the train doesn’t see a restrictive aspect. It’s the former that should have been done and how it is currently, the latter was what happened.

Sure this should have been picked up on, but I’m not so sure it’s s cavailer attitude...though admittedly I haven’t read the full report or have inner knowledge.

The car driver showed such attitude imo though.

 

[eman_resu]

Drawings I have (for a different AHB) suggest that the crossing will reset after 2 minutes if the strike-out treadle and track circuit aren't operated. This caters for the false activation of the crossing by right side failure of the strike-in track circuit, and is intended to limit the road closed time to avoid encouraging road users to zig zag round the barriers.

The train seems to have stopped out of course, not at a signal. Having a signal within the max train length after the crossing would be an obvious cause of potential blocking back by trains, onto the crossing, and would be avoided by design.

I don't know what happens if the strike-out is operated but the strike-in remains active, but I suspect it will time out in the same manner, which could be what has happened here.

I'm not immediately sure if there is a way to protect against this, eg disabling the timeout once the strike-out has occurred might help (ie it is then known that a train has passed over the crossing) but there may be other problems presented by having that control.

Any chance the strike out treadle has been operated by the train as usual, except that the train has stopped before the last wheel flange has passed through the treadle?

Thus treadle arm rises slowly to normal, and the crossing clears... IIRC the treadle arm had an inbuilt delay so that it never had chance to return to normal before the next flange would depress it again.

 

[2HAP]

RAIB report published.

https://assets.publishing.service.g...ile/727288/R082018_180719_Stainforth_Road.pdf

At around 05:50 hrs on 11 January 2018, a car collided with the rear-most wagon of a stationary freight train at Stainforth Road Automatic Half-Barrier level crossing, near Doncaster. The crossing’s warning equipment was not operating and its half- barrierswere raised when the car approached and entered the crossing. As a result ofthe accident, the driver of the car suffered scratches and bruises but their car was damaged beyond economic repair.

The train was at a stand because its brakes had been applied by the locomotive’s vigilance device. This occurred because the driver of the train did not respond to the device’s audible alarm in the time period permitted, probably due to the high level of ambient noise in the locomotive’s cab. The car driver was not alerted to the presence of the train by the crossing’s warning devices because the design of the level crossing’s control circuits had permitted it to re-open to road traffic while it was still occupied by the train. The car driver did not see the wagon with enough time to take effective avoiding action, given her speed of approach. This was because the train was unlit and unreflective and also because there was no ambient light near the crossing.

 

[Harpers Tate]

Regardless of the rights and wrongs of the workings of the crossing, and the responsibility or otherwise of the Railway to mitigate such events - all I can say is - it's as well the obstruction wasn't a child. Or any human being. etc. One must always drive according to visibility and not assume the road ahead is clear of obstruction. There is a difference between being "in the right" and being right.

 

[AlterEgo]

Regardless of the rights and wrongs of the workings of the crossing, and the responsibility or otherwise of the Railway to mitigate such events - all I can say is - it's as well the obstruction wasn't a child. Or any human being. etc. One must always drive according to visibility and not assume the road ahead is clear of obstruction. There is a difference between being "in the right" and being right.

I get your point, but how fast do you drive on unlit roads at night, with a national speed restriction then?

 

[Dai Corner]

I get your point, but how fast do you drive on unlit roads at night, with a national speed restriction then?

http://www.highwaycode.info/rule/126

Drive at a speed that will allow you to stop well within the distance you can see to be clear.

 

[AlterEgo]

http://www.highwaycode.info/rule/126

This means, in practice, on an unlit road at night, that traffic should drive at around 20 miles per hour, because you don't drive looking into the middle distance as you do in the day but rather use other road users and the white centre line as immediate references.

The road user in this case as doing 40-50mph, quite conservative for a road with a speed limit of 60.

It's very easy not to see an obstruction like the one that caused the accident until you are almost upon it.

 

[Dai Corner]

This means, in practice, on an unlit road at night, that traffic should drive at around 20 miles per hour, because you don't drive looking into the middle distance as you do in the day but rather use other road users and the white centre line as immediate references.

The road user in this case as doing 40-50mph, quite conservative for a road with a speed limit of 60.

It's very easy not to see an obstruction like the one that caused the accident until you are almost upon it.

If you can only see 12m ahead, then yes you should only drive at 20mph according to the Highway Code.
https://www.highwaycodeuk.co.uk/answers/what-is-the-stopping-and-braking-distance-of-a-car

I don't think the report attaches any blame to the driver in this case though.

 

[Spartacus]

Regardless of the rights and wrongs of the workings of the crossing, and the responsibility or otherwise of the Railway to mitigate such events - all I can say is - it's as well the obstruction wasn't a child. Or any human being. etc. One must always drive according to visibility and not assume the road ahead is clear of obstruction. There is a difference between being "in the right" and being right.

Sounds like only the very rear was over the crossing, so the driver would have had a clear view through to the other side and may have mistaken the buffer beam etc for part of the crossing.