Bucknell Level Crossing - why locally monitored?

[Meerkat]

Bucknell LC on the Heart of Wales line is Auto Half Barrier (though not sure if train crew operated northbound as right next to the platform) after previously being AOCL.
Why is it locally monitored with a 10mph speed limit since it had barriers installed? AHBs manage elsewhere.
Also if it is monitored by the train crew why are there 4 CCTV camera pointing at it, and who is watching them?

 

[haddock1000]

According to the Wales and Western Sectional Appendix (see Western and Wales Sectional Appendix December 2024.pdf [warning, 117 MB file!] page 665 in Dec '24), Bucknell LC is not an AHB (Automatic Half Barriers), but an AOCL+B (Automatic Open Crossing, Locally Monitored with Barriers).



This type of crossing is an AOCL which has had barriers retrofitted to give an increase in safety. There is also a local instruction associated with this crossing (page 1060 in Dec '24):

Bucknell, Dolau, Llangadog, Ffairfach, Cilyrychen, Llandybie, Brynmarlais and Tirydail level crossings. The instructions for ABCL/ AOCL level crossings in Rule Book, Module TW8, Section 4 apply at these crossings with the following modifications:- The crossings are operated by approaching trains or the operation of the Driver's plunger. In the event of the crossing sequence not being initiated by the approach of the train or should the white light stop flashing before the train reaches the crossing, the Driver must operate the plunger provided in a locked cabinet (Driver's No.1 key) on the white light post to activate the crossing. When the white light is flashing the Driver may proceed as normal. If after the operation of the plunger the white light still does not flash the Driver must treat the crossing as failed. A telephone to Pantyffynnon signalbox is provided.

From an operational perspective, the operation of an AOCL+B is the same as an AOCL, its just the barriers also lower/raise. The driver still checks that the crossing is clear, and checks for the DCI (Driver's Crossing Indicator, referred to above as "the white light") is flashing before crossing.

For a crossing to have been upgraded from an AOCL to an AHB would be a more extensive upgrade. An AHB's operation is monitored (from a failures point of view) by a signaller, rather than the approaching driver. Additionally, the presence of the adjacent station would probably require stopping/non-stopping controls installing (or their absence justified/mitigated by e.g. a speed restriction). These controls would be controlled from the Signaller in Pantyffynnon Signal Box, and to add in the communications link, controls in the box, etc would be very difficult. There may be other areas which would make an AHB a lot more expensive than an AOCL+B. This is why the AOCL+B is invented - it provides a lot of the benefits of AHBs, but at a vastly lower cost.

I suspect that the CCTV cameras are just for monitoring for user compliance, and for handing over to relevant parties for enforcement action/insurance claims in the event of misuse. The mere presence of things that look like CCTV cameras can also act as a deterrent.

 

[Meerkat]

Thanks. Is the driver responsible for checking the crossing is clear, or is it treated like an AHB for that?
Is there still a 10mph limit?
It alls adds up to the HoW being cripplingly slow

 

[Annetts key]

Yes the driver is responsible for checking the crossing is clear, that’s the reason for the 10 MPH approach speed limit.

 

[Harpo]

It alls adds up to the HoW being cripplingly slow

It also aids the CWL’s survival by keeping its costs down.

 

[Legolash2o]

It also aids the CWL’s survival by keeping its costs down.

Is it just the initial cost or do they have much higher maintenance cost?

 

[Meerkat]

Yes the driver is responsible for checking the crossing is clear, that’s the reason for the 10 MPH approach speed limit.

Why when they aren’t responsible at an AHB?

It also aids the CWL’s survival by keeping its costs down.

Yet helps kill it by making it too slow to be particularly useful

 

[Harpo]

Is it just the initial cost or do they have much higher maintenance cost?

The more tech there is, the higher the cost of maintaining it. Although I never came across a maintenance section coughing up a saving when infrastructure was renewed, rationalised or simplified!

Yet helps kill it by making it too slow to be particularly useful

The CWL has long been a commercial basket case so it’s not as if any acceleration is going to see a farebox payback.

 

[Belperpete]

Why when they aren’t responsible at an AHB?

I think that is best answered by asking why aren't drivers responsible for checking AHBs. With AHBs the signaller is provided with failure indications, and is responsible for stopping trains if the crossing fails. This requires circuits between the crossing and the signalbox, and a signal not too far from the crossing at which an approaching train can be stopped. For a level crossing in the middle of nowhere, this can cost a lot, and adds to the signallers workload.

Wth an AOCL or ABCL, the driver monitors the crossing operation through the Drivers White Light Indicator. There is no need for control and indication circuits, or a protecting signal. However, it does mean that line speeds at AOCLs and ABCLs have to be limited, so that the driver can stop the train if the DWL doesn't indicate. Having provided the DWL, with the requirement for the driver to check the DWL, then it makes sense for the driver to also check that the crossing isn't obstructed.

Whereas at AHBs, with their higher line speeds, by the time the crossing has operated, there is usually little chance of the driver stopping the train before the crossing if he sees that it is obstructed. Because of this, there are quite strict limitations on where AHBs can be used, compared to AOCLs and ABCLs. And extra safeguards against the crossing failing to operate, making AHBs more complex and expensive.

Even with their extra safeguards, AHBs are still fundamentally less safe. I think that it is intended to phase out AHBs.

There was an open crossing version of the AHB, the AOCR, but very few were ever installed before the concept was dropped through safety concerns, and I think most have now been converted to something else.

 

[Lewisham2221]

Yet helps kill it by making it too slow to be particularly useful

I could perhaps understand this argument if it were for a crossing on a section of line between stations, where the line speed would otherwise be significantly higher.

However - in this case - the crossing is immediately adjacent to the station, where the train would be slowing anyway, so ultimately any time saving would be so minimal as to make absolutely no meaningful distance to end-to-end running time.

On the other hand, the full barrier crossing at Llandrindod requires all trains to come to a complete stop whilst the driver operates the crossing controls, waits for the barriers to lower and confirms that the crossing is clear...

 

[Meerkat]

However, it does mean that line speeds at AOCLs and ABCLs have to be limited, so that the driver can stop the train if the DWL doesn't indicate. Having provided the DWL, with the requirement for the driver to check the DWL, then it makes sense for the driver to also check that the crossing isn't obstructed.

You could move the DWL further away and let them accelerate as soon as they see it.

However - in this case - the crossing is immediately adjacent to the station, where the train would be slowing anyway, so ultimately any time saving would be so minimal as to make absolutely no meaningful distance to end-to-end running time.

Request stop though, and if they get the same system as the Far North Line they wouldn’t even need to slow for that.

 

[Lewisham2221]

Request stop though, and if they get the same system as the Far North Line they wouldn’t even need to slow for that.

How does the timetable accommodate passing through request stops at line speed? Does it just fall apart when the train has to stop at a couple of the stations? Or is the end-to-end time still slower to provide some allowance for station stops?

As mentioned above, the full barrier crossing at Llandrindod requires all trains to stop to operate the crossing, so that increases the time far more than slowing a little bit sooner for Bucknell.

The biggest barrier to improving journey times on the line is the single line, passing loops and token exchanges - speeding up trains over a handful of level crossings will make next to no difference in the grand scheme of things.

 

[edwin_m]

You could move the DWL further away and let them accelerate as soon as they see it.

The lights and barriers would then have to operate that much sooner before the train arrived, increasing the risk that a road driver would decide the crossing had failed and dodge round the barriers.

How does the timetable accommodate passing through request stops at line speed? Does it just fall apart when the train has to stop at a couple of the stations? Or is the end-to-end time still slower to provide some allowance for station stops?

When I ran the timings of an actual journey against a traction simulation a few years ago it looked like each request stop was timed for a stop of approximately 30s, but the dwell time was about 45s if a stop actually took place. If there was no stop the train would still have to brake to a low speed, in case somebody was waiting to board. However there was enough time at each passing station to recover if several stops were actually needed when approaching it. If that's actually the intent then the request stops are essentially a performance benefit to recover lateness if they are not needed.

 

[Belperpete]

Where there is a station very close to an automatic crossing, and the crossing is expecting the train to stop, then the stop has to be enforced. Otherwise the train will arrive too soon, without adequate warning to road users. With AOCL and ABCL, this can be enforced by the Drivers White Light. However, with an AHB that doesn't have a DWL, a signal has to be provided to enforce the stop.

Some automatic crossings assume that all trains stop. However, in some cases, a stopping/non stopping control is provided, either operated by the signaller, or from the train description.

Full barrier crossings inherently have a protecting signal (or stop board). They are usually closed in sufficient time that the train driver doesn't get a restrictive aspect on the protecting signal, or signals in rear. However, this means that such crossings are closed to road users for significantly longer. In some cases, however, the crossing may be operated by train crew, and in such cases the train inherently has to stop.

 

[Railsigns]

Full barrier crossings inherently have a protecting signal (or stop board).

Apart from AFBCL crossings.

 

[Belperpete]

Apart from AFBCL crossings.

You are correct. I should have said manually controlled crossings (including MCBOD),

 

[Annetts key]

Let’s be frank. If you wish to reduce the time taken for a train to travel along this line, it means significant changes to the signalling system including significant changes to the level crossing systems. It may also require more sections of double line or longer loops.

Unless an external (to Network Rail) source of funding comes along this just isn’t going to happen. Network Rail as it is currently set up doesn’t have the money for major improvements unless it’s part of a large renewal scheme replacing a life expired system.

GBR may have more flexibility. But even then, on what grounds can the cost be justified? Will an improved system result in a large increase in passengers?

 

[Harpo]

GBR may have more flexibility. But even then, on what grounds can the cost be justified? Will an improved system result in a large increase in passengers?

Precisely. The boom decades for rail travel did nothing for the CWL. It’s remained a social railway and possibly the worst financial performer of TfW’s routes at a guess?

Unless a West Wales to Manchester corridor suddenly looked hugely generative, nursing the route and keeping its costs low has to be the way.

 

[edwin_m]

Let’s be frank. If you wish to reduce the time taken for a train to travel along this line, it means significant changes to the signalling system including significant changes to the level crossing systems. It may also require more sections of double line or longer loops.

Unless an external (to Network Rail) source of funding comes along this just isn’t going to happen. Network Rail as it is currently set up doesn’t have the money for major improvements unless it’s part of a large renewal scheme replacing a life expired system.

GBR may have more flexibility. But even then, on what grounds can the cost be justified? Will an improved system result in a large increase in passengers?

Also, if trains pass more than once along the single line, they have to be timed so they meet at passing loops. So there's a risk that any acceleration just results in spending the time saved sitting at the next loop, especially if service frequencies are increased so more trains need to pass each other.

 

[Harpo]

Also, if trains pass more than once along the single line, they have to be timed so they meet at passing loops. So there's a risk that any acceleration just results in spending the time saved sitting at the next loop, especially if service frequencies are increased so more trains need to pass each other.

Shouldn’t be an issue if running times loop-to-loop in either direction remain equal.

 

[edwin_m]

Shouldn’t be an issue if running times loop-to-loop in either direction remain equal.

If the times in both directions are shorter but remain equal, then the interval between trains also has to reduce (to twice the loop to loop distance minus a bit for passing time) so they still meet at the loops. This may have other impacts on the service or timetable.

I've tried to timetable more frequent and/or faster services on the CWL and it's not easy.

 

[Meerkat]

Let’s be frank. If you wish to reduce the time taken for a train to travel along this line, it means significant changes to the signalling system including significant changes to the level crossing systems. It may also require more sections of double line or longer loops.

Unless an external (to Network Rail) source of funding comes along this just isn’t going to happen. Network Rail as it is currently set up doesn’t have the money for major improvements unless it’s part of a large renewal scheme replacing a life expired system.

GBR may have more flexibility. But even then, on what grounds can the cost be justified? Will an improved system result in a large increase in passengers?

I do agree, but think that if it’s not worth doing properly then don’t bother and close it with the savings spent on buses. Converting the line to a bike route would probably attract more tourists than the railway!

BtW what is the ballpark cost of putting in the red/green lights on occupation crossings? Question triggered because I saw an old cab ride on the HoW and there were some crawls over farm crossings on bends that must each cost a minute or two when on otherwise linespeed sections.

 

[Taunton]

I think that is best answered by asking why aren't drivers responsible for checking AHBs. With AHBs the signaller is provided with failure indications ... and adds to the signallers workload.

Really? How often does the signaller have to attend to an AHB failure indication.

And even where there is a significant issue, such as the Athelney AHB accident some years ago, where the crossing sequence had been disrupted by a tamper running wrong line over the operating treadles shortly before the accident train arrived at full speed, it was apparent from the report that the signallers had never been instructed in the detail of how the crossing operated.

 

[Annetts key]

BtW what is the ballpark cost of putting in the red/green lights on occupation crossings?

Substantially less if a VaMoS system is used instead of conventional designs.

VaMoS is an MSL (Miniature Stop Light) system typically used for occupation or accommodation (e.g. user worked gates, open vehicle access or footpath) crossings to indicate to the user if it is safe to cross the infrastructure. It’s a prebuilt independent modular overlay system that is assembled on site. It only needs a mains supply, it doesn’t need or have any connections to any existing signalling system.

I don’t know the cost, but an indication may be in one of the documents on the ORR site. As the ORR asked Network Rail some time ago to investigate how to improve level crossings by using alternatives to conventional designs.

 

[The Planner]

Substantially less if a VaMoS system is used instead of conventional designs.

VaMoS is an MSL (Miniature Stop Light) system typically used for occupation or accommodation (e.g. user worked gates, open vehicle access or footpath) crossings to indicate to the user if it is safe to cross the infrastructure. It’s a prebuilt independent modular overlay system that is assembled on site. It only needs a mains supply, it doesn’t need or have any connections to any existing signalling system.

I don’t know the cost, but an indication may be in one of the documents on the ORR site. As the ORR asked Network Rail some time ago to investigate how to improve level crossings by using alternatives to conventional designs.

£150k.

 

[Taunton]

£150k.

Probably Heart of Wales ticket takings for a year ...