Train waiting for level crossing

[ivorytoast28]

Today I was on the 13:25 Woking to Portsmouth harbour service and as we approached Milford we slowed down and stopped in the station (despite not being scheduled to call there)

I was in the front carriage and could see that cars were crossing the tracks in front of us. After about a minute the lights of the crossing came on and the barriers went down and we proceeded through. I can only think that due to the engineering works and amended timetable we were using the path of a service that usually stopped at Milford. But it felt very weird having the train stop for cars rather than the other way around

 

[DerekC]

Sounds much more like a level crossing fault to me. What sort of crossing is it? (I should know but I don't). There are some crossings and some faults where the signaller has to bring the train to a stand at the preceding signal, operate the crossing and then allow the train to proceed under caution.

 

[Lurcheroo]

Sounds much more like a level crossing fault to me. What sort of crossing is it? (I should know but I don't). There are some crossings and some faults where the signaller has to bring the train to a stand at the preceding signal, operate the crossing and then allow the train to proceed under caution.

I was thinking the same but it’s an automatic half barrier crossing so doesn’t usually have signal protection, but I suppose the could have been contacted by the signaller previously and told to stop before the crossing.

 

[ivorytoast28]

Sounds much more like a level crossing fault to me. What sort of crossing is it? (I should know but I don't). There are some crossings and some faults where the signaller has to bring the train to a stand at the preceding signal, operate the crossing and then allow the train to proceed under caution.

I've no idea about crossing types so I can't say but my initial thought was that it must've failed but then immediately I saw lights start flashing on it and then the barrier come down, it wasn't a long delay or anything and it's a pretty remote location

 

[Railsigns]

There's a signal at the end of the Down platform, a few metres away from the level crossing.

 

[Lurcheroo]

I've no idea about crossing types so I can't say but my initial thought was that it must've failed but then immediately I saw lights start flashing on it and then the barrier come down, it wasn't a long delay or anything and it's a pretty remote location

I had wondered if perhaps because it’s right at a station, in the direction where the train goes through the station first, if it’s setup to only start the crossing sequence once the train is stopped at the station or needs to be activated by the driver and has some sort of signal protection in just one direction, hence why your non stopping train had to stop, that would be quite an unusual setup for an AHBC (the AHBC at Harlech in Wales only operates automatically in one direction and requires the driver to activate it with a button in the other direction), BUT looking at some YouTube videos the barriers seem to be down in plenty of time before the trains arrive at the station. So this one does seem quite peculiar.

 

[DerekC]

The most plausible explanation I can come up with is that the crossing had a fault so was under local manual control, in which case the train would be stopped and cautioned at the preceding signal, I think. I wonder if the OP can say whether anyone in a hi-vi was standing at the crossing control unit.

 

[VP185]

The most plausible explanation I can come up with is that the crossing had a fault so was under local manual control, in which case the train would be stopped and cautioned at the preceding signal, I think. I wonder if the OP can say whether anyone in a hi-vi was standing at the crossing control unit.

Or just a signal fault with the signaller unable to clear it.

 

[TSG]

Could be a number of things including a fault, but possibly not. A basic AHB isn't linked to the signalling but I think Milford has a couple of features that muddy the waters. Firstly it has two sets of strike ins, stopping and non stopping. A stopping selection will delay operation of the crossing until a later strike in but keep the starting signal on for a time. If stopping was incorrectly selected the the non stop train would be brought to a stand. Related is signal regulation, which is used where a signal is within the strike in of the crossing. When the signal is ready to clear the crossing sequence begins and the actual display of a proceed aspect is delayed so that the train can't reach the crossing ahead of the sequence e.g. 15s for the fastest accelerating train from a stand to the crossing would delay clearance by 22s for a 37s sequence. Could be either of these that happened.

 

[satisnek]

Could be a number of things including a fault, but possibly not. A basic AHB isn't linked to the signalling but I think Milford has a couple of features that muddy the waters. Firstly it has two sets of strike ins, stopping and non stopping. A stopping selection will delay operation of the crossing until a later strike in but keep the starting signal on for a time. If stopping was incorrectly selected the the non stop train would be brought to a stand. Related is signal regulation, which is used where a signal is within the strike in of the crossing. When the signal is ready to clear the crossing sequence begins and the actual display of a proceed aspect is delayed so that the train can't reach the crossing ahead of the sequence e.g. 15s for the fastest accelerating train from a stand to the crossing would delay clearance by 22s for a 37s sequence. Could be either of these that happened.

Yes, certainly as installed, Milford AHB had stopping/non-stopping selection on the Down line. There was a fatal accident in the early years and the report describes the processes involved.

Accident at Milford Level Crossing on 22nd December 1978 :: The Railways Archive

Archive of UK railways documents

www.railwaysarchive.co.uk

 

[DerekC]

Yes, certainly as installed, Milford AHB had stopping/non-stopping selection on the Down line. There was a fatal accident in the early years and the report describes the processes involved.

Accident at Milford Level Crossing on 22nd December 1978 :: The Railways Archive

Archive of UK railways documents

www.railwaysarchive.co.uk

An interesting accident report - thanks. I agree that the crossing control being set to stopping mode for a non-stop train is the most likely explanation for the OP's observation.

If the arrangement has never been modified (and the report doesn't recommend that it should be), it seems that whenever the crossing control is set to "stopping" mode, an approaching train faces a crossing with barriers open to road traffic well within the overlap of the down starting signal. I guess that the signal is now probably fitted with TPWS, but that wouldn't prevent a train in wheel slide (which is what the report suggests may have happened in the 1978 accident) from entering the crossing before the barrier sequence starts to operate.

 

[Tomnick]

On the topic of stopping/non-stopping controls...

How are these typically selected on computer-based control systems? Is there a degree of automation even where full ARS isn't provided? Nowadays, it seems quite normal for us to be brought down at each and every location with such controls when, for example, being diverted (non-stop) away from booked route, or running to STP or even VSTP schedules. Upon grumbling to the signaller on one occasion, having been brought to a stand from 80mph on a slippery rail at a notorious location for sliding, I was just told "the system did it".

One wonders if that's what happened here – control has been transferred from Farncombe to the ROC now, hasn't it?

 

[TSG]

Farncombe is still in use. It will go to BROC in the next couple of years. It’s just a manual selection in a lot of places, including there . ARS can do it. I’ve also seen systems that use an output from the Train Describer and a lookup table to automate it, where ARS isn’t available

 

[CarrotPie]

Nowadays, it seems quite normal for us to be brought down at each and every location with such controls when, for example, being diverted (non-stop) away from booked route, or running to STP or even VSTP schedules.

When you're not running to a "standard" timetable, it's easier for the signaller to just make you stop rather than having to pull up your timetable somewhere, especially if they're busy.

 

[DelW]

Having read this discussion, I took a walk to Milford this morning.
I've just watched the 10:55 down stopper make its call, and the barriers closed about 15 seconds before the train stopped at the platform, and therefore a minute or two before it actually crossed the road.
It wasn't affected by the late running up train, the barriers reopened before that arrived.

(Edited to add) I waited for the next down fast at 11:20, which went through at speed as expected. The barriers were down before the train was in sight.

 

[Tomnick]

When you're not running to a "standard" timetable, it's easier for the signaller to just make you stop rather than having to pull up your timetable somewhere, especially if they're busy.

Easier for the signaller, maybe. At least they pick up the delay minutes and not me (if it's an STP problem rather than an ad-hoc diversion at least).

Wonder how much money we waste on fuel and brake pads across the industry as a result of this, let alone the increased risk from routinely letting trains approach signals at danger for no reason whatsoever?

Farncombe is still in use. It will go to BROC in the next couple of years. It’s just a manual selection in a lot of places, including there . ARS can do it. I’ve also seen systems that use an output from the Train Describer and a lookup table to automate it, where ARS isn’t available

Thanks. I suspect we have the latter on at least a couple of our routes. Works well in normal operation, with stop/non-stop controls at numerous different stations and a huge variety of stooping patterns. Awful when you're not meant to stop everywhere but get brought right down at every single one! Presumably it *can* be overridden manually, if the signaller chooses?

Is the lookup table able to be edited? Presumably it has to be, at each timetable change as a minimum? Does it distinguish between different days of the week? Sorry for all the questions – I am genuinely interested, not just here for a rant!

 

[CarrotPie]

Wonder how much money we waste on fuel and brake pads across the industry as a result of this, let alone the increased risk from routinely letting trains approach signals at danger for no reason whatsoever?

I'd argue an open level crossing is a perfectly valid reason to hold a signal at danger!

Is the lookup table able to be edited? Presumably it has to be, at each timetable change as a minimum? Does it distinguish between different days of the week? Sorry for all the questions – I am genuinely interested, not just here for a rant!

At more modern locations like TBROC it'll be a TRUST enquiry (so realtime).

 

[dk1]

Wonder how much money we waste on fuel and brake pads across the industry as a result of this, let alone the increased risk from routinely letting trains approach signals at danger for no reason whatsoever?

Considering all trains have been braking/accelerating down to 50mph from upto the 100mph line speed at Gypsy Lane between Ipswich & Stowmarket for the last 13 years, I don’t think that’s a priority

 

[TSG]

Thanks. I suspect we have the latter on at least a couple of our routes. Works well in normal operation, with stop/non-stop controls at numerous different stations and a huge variety of stooping patterns. Awful when you're not meant to stop everywhere but get brought right down at every single one! Presumably it *can* be overridden manually, if the signaller chooses?

Is the lookup table able to be edited? Presumably it has to be, at each timetable change as a minimum? Does it distinguish between different days of the week? Sorry for all the questions – I am genuinely interested, not just here for a rant!

For the lookup table system I saw you could certainly look at and edit the timetable and it distinguished between each day of the week. However, it wasn't updated with a timetable every 12 hours like an ARS system would be. I think it would be manual entry at timetable change. I don't remember the details but I'm certain there was a means of switching it out and the signaller still had the manual selection switches on the panel for use as above.

ARS can be switched out by turning off the relevant subarea. I'd assume that would apply to stopping/non-stopping selection as much as it would to route setting. The signaller would still have manual controls as above.

For manual selection you are reliant on the signaller transmitting a non-stopping selection before the train gets near the strike ins and it defaults to stopping selection. From memory I think the signaller can re-transmit a selection until the strike in if they want to change it. I think the ability to 'override' a selection made by an automatic system depends on how much time you have to switch off the automatic system and re-transmit a selection before the train hits the strike in.

At more modern locations like TBROC it'll be a TRUST enquiry (so realtime).

Is the selection not inferred from the timetable file sent to the ARS TTP? Is this to do TBROC being Traffic Management System ready?

 

[DerekC]

I am hoping somebody is going to tell me I am wrong, but the way this thread has developed, it seems that at Milford LX, trains are routinely brought to a stand at a red signal with an open level crossing only 25 yards beyond it. That seems to me like a repeat of the 1978 accident waiting to happen. For those who haven't read the report, a stopping train SPADed at the signal (probably due to wheel slide) and hit a car which was quite legitimately on the crossing, killing the driver. There is no way signalling engineers would allow this if it was another train which could be in that position. OK, we now have better wheel slide protection. OK, we now have defensive driving, but it just seems wrong in principle. And by stopping trains unnecessarily we are provoking the situation.

 

[theironroad]

As mentioned above Milford AHB has fast & slow settings so trains not calling can usually do line speed ( other factors permitting) as crossing will be closed to road traffice in time. Stopping services are checked down and signal clears just as train comes to a stand usually depending on driver approach speed and railhead conditions.

Believe that there was a overnight track circuit failure between milford and witley last night which probably impacted this and resulted in out of course working but I don't have details.

I am hoping somebody is going to tell me I am wrong, but the way this thread has developed, it seems that at Milford LX, trains are routinely brought to a stand at a red signal with an open level crossing only 25 yards beyond it. That seems to me like a repeat of the 1978 accident waiting to happen. For those who haven't read the report, a stopping train SPADed at the signal (probably due to wheel slide) and hit a car which was quite legitimately on the crossing, killing the driver. There is no way signalling engineers would allow this if it was another train which could be in that position. OK, we now have better wheel slide protection. OK, we now have defensive driving, but it just seems wrong in principle. And by stopping trains unnecessarily we are provoking the situation.

Probably wouldn't be allowed today as it is a tight. That said, Farncombe to Petersfield is in process of being resignalled so will have to see what happens...

From Network Rail:​

"Level crossing upgrades​

12 level crossings along the route will also be either renewed or upgraded as part of the F2P scheme. We’ll be working on Hammer, Liss, Liss Common, Peasmarsh, Petersfield, Farncombe East, Farncombe West, Milford, Sheet, Shutters, Princes Bridge and Kingsfernsden level crossings.

We’ll change Automatic Half-Barrier (AHB) crossings to either Manually Controlled Barrier CCTV crossings (MCBs) or Obstacle Detection (OD) level crossings; upgrade lighting at existing MCB crossings; and install Miniature Stop Lights (MSLs) on selected footpath crossings."

 

[DelW]

I am hoping somebody is going to tell me I am wrong, but the way this thread has developed, it seems that at Milford LX, trains are routinely brought to a stand at a red signal with an open level crossing only 25 yards beyond it. That seems to me like a repeat of the 1978 accident waiting to happen. For those who haven't read the report, a stopping train SPADed at the signal (probably due to wheel slide) and hit a car which was quite legitimately on the crossing, killing the driver. There is no way signalling engineers would allow this if it was another train which could be in that position. OK, we now have better wheel slide protection. OK, we now have defensive driving, but it just seems wrong in principle. And by stopping trains unnecessarily we are provoking the situation.

I can't be definitive, but I drive across the crossing pretty regularly, and I don't remember *ever* driving over it and seeing a down train stopped in the platform. The norm is as I described in post 15.

 

[GordonT]

Interesting that Chris Green makes a cameo appearance in Paragraph 17 of the Accident Report which was written almost 44 years ago.

 

[satisnek]

I am hoping somebody is going to tell me I am wrong, but the way this thread has developed, it seems that at Milford LX, trains are routinely brought to a stand at a red signal with an open level crossing only 25 yards beyond it. That seems to me like a repeat of the 1978 accident waiting to happen. For those who haven't read the report, a stopping train SPADed at the signal (probably due to wheel slide) and hit a car which was quite legitimately on the crossing, killing the driver. There is no way signalling engineers would allow this if it was another train which could be in that position. OK, we now have better wheel slide protection. OK, we now have defensive driving, but it just seems wrong in principle. And by stopping trains unnecessarily we are provoking the situation.

A repeat of the 1978 accident is waiting to happen at numerous locations all over the country - and probably has if you delve into it. As the report mentions, the factors are not unique to this location or even this type of crossing, so no recommendations could be made. Trains have been stopping at signals immediately in front of level crossings ever since the year dot, for some obscure historical reason level crossings are not treated in the same way as converging junctions.

Having said that, the resignalling here in these parts some years back saw the signals protecting the crossings at Hartlebury and Blakedown stations moved back to some distance before the platform, which would suggest that attitudes are changing.

 

[Falcon1200]

for some obscure historical reason level crossings are not treated in the same way as converging junctions.

Not really an obscure historical reason, more because to have level crossing barriers down in time to give all trains clear signals throughout would close the road for far longer, causing a different risk at half-barrier crossings of motorists becoming impatient and driving round the lowered barriers. And while there is of course still a risk at level crossings it is far lower than that of trains colliding at a junction.

 

[GordonT]

Not really an obscure historical reason, more because to have level crossing barriers down in time to give all trains clear signals throughout would close the road for far longer, causing a different risk at half-barrier crossings of motorists becoming impatient and driving round the lowered barriers. And while there is of course still a risk at level crossings it is far lower than that of trains colliding at a junction.

Still seems an uncharacteristically exposed potential for a collision with an innocent road user especially in areas where wheelslip is not exactly a rare phenonmenon. Giving the weight of protection to someone deliberately driving round a lowered barrier doesn't seem reasonable.

 

[3RDGEN]

Still seems an uncharacteristically exposed potential for a collision with an innocent road user especially in areas where wheelslip is not exactly a rare phenonmenon. Giving the weight of protection to someone deliberately driving round a lowered barrier doesn't seem reasonable.

AHB's have been operating like this for 50+ years and beyond the quoted 1978 incident how many other cases have they been? The principle for AHB's was/is to limit the time the barriers are down as road users will get impatient or think the crossing has failed and then misuse the crossing which is far more dangerous.

If a driver/pedestrian gets stopped today for several minutes at an AHB then next week they think twice about stopping for one they have just seen the barriers lower at and weave around as mentioned above, it's that behaviour the design of the crossing operation is trying to mitigate. Install full barriers with long waits and drivers will risk driving under lowering barriers and pedestrians will jump the barriers and walk across, you can't win unless you close the crossings and build bridges instead.

 

[satisnek]

Not really an obscure historical reason, more because to have level crossing barriers down in time to give all trains clear signals throughout would close the road for far longer, causing a different risk at half-barrier crossings of motorists becoming impatient and driving round the lowered barriers. And while there is of course still a risk at level crossings it is far lower than that of trains colliding at a junction.

I was referring to the distance (or lack of) beyond the signal, at any level crossing. I believe it's because a level crossing was not historically regarded as an 'obstruction'. There were numerous instances of trains demolishing level crossing gates back in steam days, but with lower levels of motorised road traffic the risk of death or injury was smaller than it is today.

 

[TSG]

A repeat of the 1978 accident is waiting to happen at numerous locations all over the country - and probably has if you delve into it. As the report mentions, the factors are not unique to this location or even this type of crossing, so no recommendations could be made. Trains have been stopping at signals immediately in front of level crossings ever since the year dot, for some obscure historical reason level crossings are not treated in the same way as converging junctions.

Having said that, the resignalling here in these parts some years back saw the signals protecting the crossings at Hartlebury and Blakedown stations moved back to some distance before the platform, which would suggest that attitudes are changing.

Indeed, standards do change. Practice now tends to pushing the protecting signal further from the crossing than was previously allowed. There needs to be either enough distance for the crossing sequence to give sufficient warning to road users (straight to red lights where a SPAD is detected), or a system of timing treadles or a speed measuring axle counter head for the purpose of predicting an imminent SPAD.

 

[DelW]

I don't know whether it's routine maintenance, or part of changes related to the resignalling, but Milford level crossing has been completely dug out this morning, with ballast being dug from between the sleepers as well.

Photo from a little distance away as I didn't want to offend anyone working there.