Level crossings that would bring the greatest benefits if closed

[The Planner]

The several crossings on the Windsor Lines (North Sheen, Egham, etc) are an absolute cap on capacity because although you could get a few more per hour on the rail side, the road congestion on extremely busy roads (including pedestrians), plus the current high frequency, leads to traffic being backed up all day long. Development long ago right up to the four corners of the crossing prevents any bridge/underpass solution. This is what did in the Heathrow Airtrack scheme, when it was realised how that would make all the crossings untenable local political campaigns reached right up into Parliament.

One issue, from me knowing at least North Sheen for a very long time, is each generation of resignalling has increased road closure times. Before Feltham power box opened (1970?) the local gate box staff were past masters at minimising any delay, there were significant complaints back then when that resignalling took place. The more recent change has again led to further increased closure time per train, it is as if the signalling designers just give this aspect insufficient consideration because it is not a "railway side" issue.

I can't find any specific rules that specify any LCs limit the capacity, there are upper limits on trains per hour on some lines in SW London, but nothing specific on the LCs being the reason.

 

[Mr. SW]

The problem with Vine Road in Barnes is the presence of the Beverley Brook so dipping either railway or road would have to be carefully considered.

 

[Mcr Warrior]

First thing that came to my mind was the crossings on the line north of Altrincham, I think it's Navigation Road and one other I've forgotten. But whatever the future of Metrolink or the Mid-Cheshire route, the single track section for both lines will always be a constraint and so removing the crossings would surely allow the maximum use of the bottleneck.

Not totally convinced by that argument. In my experience, heavy rail and Metrolink trams aren't all that often delayed by the level crossing operation(s) at Navigation Road and Deansgate Lane, even when the trams are operating every six minutes (each way) over their single track section. If the road traffic is slightly delayed, then so be it. Also, there is a small business estate and housing on the North West side of the Deansgate Lane level crossing which would be completely inaccessible by road traffic if the Deansgate Lane crossing were to be permanently closed.

Interchanging from the Mid Cheshire Line onto Metrolink can often still be a pain (say for a journey from Stockport to Timperley) with there being no overbridge at Navigation Road. However, is there actually room at the location to build one?

 

[Halifaxlad]

The crossing at Crosshills & Kildwick between Steeton & Silsden and Cononley on the Aire Valley line from Leeds to Skipton. I would close Connoley too so there isn't a single crossing on this stretch!

Starbeck crossing after Harrogate but before Knaresborough on the Harrogate line.

Crosshills & Kildwick and Starbeck would reduce congestion massively! The first could be done with a new road bridge and relaligned road the second would be much more complicated as the line would have to be sunk.

 

[TheGrew]

Mount Pleasant Road in Southampton I have often thought, would spend more time down than up. It fortunately isn't super high traffic, though.

 

[Big Jumby 74]

Mount Pleasant Road in Southampton I have often thought, would spend more time down than up

It did generate road user complaints from time to time back in BR days as I recall.

 

[steamybrian]

Helpston level crossing must be difficult to cross as a motorist or pedestrian.
It crosses 5 tracks (or is it 6?) including the busy East Coast Main Line.

 

[30907]

The crossing at Crosshills & Kildwick between Steeton & Silsden and Cononley on the Aire Valley line from Leeds to Skipton. I would close Connoley too so there isn't a single crossing on this stretch!

Starbeck crossing after Harrogate but before Knaresborough on the Harrogate line.

Crosshills & Kildwick and Starbeck would reduce congestion massively! The first could be done with a new road bridge and relaligned road the second would be much more complicated as the line would have to be sunk.

You beat me to it - Cononley, interestingly, now has an alternative pedestrian route through a new housing development, but closing the main village access to vehicles would be impossible without a vast amount of expense. (It was closed 30-odd years ago when the old gates were taken out - and the night it snowed heavily caused chaos....)

 

[Northerngirl]

While the line speeds aren't very high, it amazes me that there is so many crossings on merseyrail, right through the towns on the Southport line, 4tph in both directions must be up there for busyness

 

[positron]

The one in the entrance to Gloucester always seems to cause delays (even if timetabled they still add to journey times unnecessarily).

Failing that the ones either side of Llanelli station always make me think it must be a bottle neck. Iirc it's level crossings such as these that prevent 10 carriage trains from running west of Swansea.

 

[Nottingham59]

I can't find any specific rules that specify any LCs limit the capacity, there are upper limits on trains per hour on some lines in SW London, but nothing specific on the LCs being the reason.

I understood LCs are limiting because of the rail industry's self-imposed rule that for any given level crossing the risk should not increase over time.

Which means that any increase in train frequency requires mitigation work or replacements on all level crossings on that line because risk is related to frequency. That factor is constraining upgrades like Ely North and the Marston Vale line.

 

[GraphitProjekt]

Indeed there is a level crossing somewhere on the Breckland Line whose risk assessment increased (due to more car traffic?) and as such it's had a TSR of something like 40mph for at least 5 or 6 years

 

[The Planner]

I understood LCs are limiting because of the rail industry's self-imposed rule that for any given level crossing the risk should not increase over time.

Which means that any increase in train frequency requires mitigation work or replacements on all level crossings on that line because risk is related to frequency. That factor is constraining upgrades like Ely North and the Marston Vale line.

Yes, but there is currently nothing specific I can see that limits the timetable based on a level crossing. I would expect to see an entry in the planning rules that says "no more than x trains per hour over y level crossing"

 

[Krokodil]

Yes, but there is currently nothing specific I can see that limits the timetable based on a level crossing. I would expect to see an entry in the planning rules that says "no more than x trains per hour over y level crossing"

No, but presumably any track access application for extra services will be considered against a risk assessment for level crossings on the route. Longer or more frequent closures, more chance of misuse.

 

[Halifaxlad]

I would also go for a lot of the crossings between Harrogate and York, it's a quite line but it has a lot of crossings that could be replaced with bridges.
It would save a lot of money if staff were redeployed to other areas!

 

[bluenoxid]

The crossing at Crosshills & Kildwick between Steeton & Silsden and Cononley on the Aire Valley line from Leeds to Skipton. I would close Connoley too so there isn't a single crossing on this stretch!

Starbeck crossing after Harrogate but before Knaresborough on the Harrogate line.

Crosshills & Kildwick and Starbeck would reduce congestion massively! The first could be done with a new road bridge and relaligned road the second would be much more complicated as the line would have to be sunk.

I am not sure that closing Starbeck High Street crossing and replacing it with a nearby alternative would reduce congestion that much. I feel that it would shift the traffic jam to the next choke point, which in my mind would be the Queen Empress roundabout in Harrogate (sat in the Stray so no easy ways to expand the junction) and the mini roundabouts at the northern end of Knaresborough High Street.

There is also a second level crossing on that section. Belmont Crossing on Forest Lane.

 

[Halifaxlad]

I am not sure that closing Starbeck High Street crossing and replacing it with a nearby alternative would reduce congestion that much. I feel that it would shift the traffic jam to the next choke point, which in my mind would be the Queen Empress roundabout in Harrogate (sat in the Stray so no easy ways to expand the junction) and the mini roundabouts at the northern end of Knaresborough High Street.

There is also a second level crossing on that section. Belmont Crossing on Forest Lane.

I really don't know where you get that idea from as I certainly wasn't suggesting closing the road!

The only way it could be done would be to sink the line and station.

 

[zwk500]

Yes, but there is currently nothing specific I can see that limits the timetable based on a level crossing. I would expect to see an entry in the planning rules that says "no more than x trains per hour over y level crossing"

There are, however, Level Crossings where trains are restricted in standing time, or required to be timed within a certain time of each other, due to level crossings.

Littleport & Watlington have restrictions mentioned, March requires -1 public offset in Down XC trains to counter cautionary aspects at Norwood Road Level Crossing, and Down freight to use the Loop if stopping to avoid blocking the LC.
Whittlesea has maximum dwells for MCG crossing. And that's just Anglia...

 

[fishwomp]

Failing that the ones either side of Llanelli station always make me think it must be a bottle neck. Iirc it's level crossings such as these that prevent 10 carriage trains from running west of Swansea.

That'd surprise me. The problem would be that you couldn't fit a 10 car (260m) between the two crossings, so you also couldn't fit a long enough platform. One of the platforms is only 170m - so only 6 cars of a Hitachi - rather awkward!

So, you could use selective opening - but need to have Llanelli passengers quite well trained... Or skip Llanelli!

Would the crossing or signalling care about the length? It's way shorter than the oil trains.. but obviously one of the barriers has to be down for the whole duration of a station stop.

The other problem would be that Carmarthen's longer platform is only 217m long, so really there's no fitting for 10 cars properly anywhere -- so you might as well not drag two 5 cars given the complications. You'd probably lock one out if you did take it..

You might get away with a 9 car though - there have been plenty railtours that were longer.

 

[positron]

That'd surprise me. The problem would be that you couldn't fit a 10 car (260m) between the two crossings, so you also couldn't fit a long enough platform. One of the platforms is only 170m - so only 6 cars of a Hitachi - rather awkward!

So, you could use selective opening - but need to have Llanelli passengers quite well trained... Or skip Llanelli!

Would the crossing or signalling care about the length? It's way shorter than the oil trains.. but obviously one of the barriers has to be down for the whole duration of a station stop.

The other problem would be that Carmarthen's longer platform is only 217m long, so really there's no fitting for 10 cars properly anywhere -- so you might as well not drag two 5 cars given the complications. You'd probably lock one out if you did take it..

You might get away with a 9 car though - there have been plenty railtours that were longer.

Yeah 9 car where the sdo works better would probably be a better idea. Though tbh there's absolutely no need for two level crossings so close together they should just close one.

 

[zwk500]

So, you could use selective opening - but need to have Llanelli passengers quite well trained... Or skip Llanelli!

SDO is used plenty of other places so no reason it couldn't work.

Would the crossing or signalling care about the length? It's way shorter than the oil trains.. but obviously one of the barriers has to be down for the whole duration of a station stop.

The crossing or signalling won't but the local traffic might be impacted by extended barrier closures.

The other problem would be that Carmarthen's longer platform is only 217m long, so really there's no fitting for 10 cars properly anywhere -- so you might as well not drag two 5 cars given the complications. You'd probably lock one out if you did take it..

You might get away with a 9 car though - there have been plenty railtours that were longer.

Railtours pull right down into the headshunt, IIRC. That may not be allowable for ordinary passenger services.

 

[Krokodil]

The crossing or signalling won't but the local traffic might be impacted by extended barrier closures.

ORR tend to take a dim view of it these days too.

 

[fishwomp]

`SDO is used plenty of other places so no reason it couldn't work.

The reason I had in mind is the platform length 170m - or 6.5 cars of Hitachi.

If you're in a 2x5 Hitachi - that's only 3 doors of the whole second unit at most available (- possibly less if the driver has to be short of the platform end/?)?!
SDO's fine - but unusual to be _that_ restricted!

 

[The Foghorn]

Helpston is six tracks. I would suggest that nearby Tallington is prone to bigger traffic queues though.

 

[Rail Ranger]

The use of tram train vehicles on the Metrolink Altrincham line would enable the reinstatement of the previous (pre-July 1991) conventional double track layout through Navigation Road station. This would eliminate delays to Metrolink services caused by the single line section. A significant downside would be that freight and Northern passenger services would potentially be delayed by Metrolink services. At the moment they have their own dedicated tracks.

 

[daodao]

The use of tram train vehicles on the Metrolink Altrincham line would enable the reinstatement of the previous (pre-July 1991) conventional double track layout through Navigation Road station. This would eliminate delays to Metrolink services caused by the single line section. A significant downside would be that freight and Northern passenger services would potentially be delayed by Metrolink services. At the moment they have their own dedicated tracks.

How would your proposal facilitate closure of the 2 level crossings on this section of line just north of Altrincham?

 

[Paul AC]

I agree that the wire gradient is the issue, but Network Rail did want to replace this bridge at the time of the Great Western electrification. I am sure that most people remember all the objections by the locals and in the end they got their way because it was a grade 2 listed structure. If the bridge was rebuilt, then the 110mph restriction for electric traction would be lifted to 125mph.

Wire gradient was the issue, though the electrification team considered that the final design and installation at Steventon was good for 125 mph - in other words the specification (for max line speed versus wire gradient) was over-cautious.

Thus, when the Colas 125 mph electrification test train was in the area for other work in May 2025, doing some checks on the wiring through the Severn Tunnel, the opportunity was taken to test the wiring through Steventon at 125 mph. The test train consisted of Colas class 43 power cars and four Mk.3 carriages, with 91117 in the centre of the train; 91117 carried the same HS-X pantograph that had performed the Midland Main Line wiring checks during 2024, the pan being specially instrumented to measure the performance of the wiring. 91117 was powering during these tests, both to draw current from the overhead and to ensure that the required speeds were obtained. Special authorisation was obtained for this train to pass through the Steventon area at 125 mph rather than the posted 110 mph limit, which is only there for electric trains but applies to all trains for simplicity.

This test train has been mentioned on this site under "Mystery train on GWML" https://www.railforums.co.uk/threads/mystery-train-on-gwml.287036/

On the night of 25/05/2025 (into 26/05/2025) the test train passed through Steventon both ways at 110 mph to get a baseline and check that the wiring was as expected. Next night (26/05/2025 into 27/05/2025) the train made two runs each way through Steventon at 125 mph. I was on 91117 throughout these runs in a technical support role. The team monitoring the pantograph reported no problems, confirmed (I understand) by later detailed analysis of the results taken.

This appeared to clear the way for the line speed at Steventon to be officially increased to 125 mph, but the current (March 2026) Sectional Appendix still shows it at 110 mph.

 

[Rail Ranger]

How would your proposal facilitate closure of the 2 level crossings on this section of line just north of Altrincham?

It wouldn't. Someone said that the parallel single line sections through Navigation Road would always be a constraint. Someone else asked about providing a footbridge at Navigation Road. There has never been one since the station opened in 1931. One could probably be provided but it would have to be very high to provide clearance over the overhead line (which is very high due to the level crossing) unless the footbridge was at the north end of the station and it would probably need to be fully accessible (with either lifts or ramps) making it expensive.

 

[Instinct04]

Brimsdown, Enfield Lock and Cheshunt level crossings.

Would make WAML four tracking easier and would end long traffic jams in the area, especially around Green street in Brimsdown as it gets terrible over there.

 

[Paul AC]

Not totally convinced by that argument. In my experience, heavy rail and Metrolink trams aren't all that often delayed by the level crossing operation(s) at Navigation Road and Deansgate Lane, even when the trams are operating every six minutes (each way) over their single track section. If the road traffic is slightly delayed, then so be it. Also, there is a small business estate and housing on the North West side of the Deansgate Lane level crossing which would be completely inaccessible by road traffic if the Deansgate Lane crossing were to be permanently closed.

Interchanging from the Mid Cheshire Line onto Metrolink can often still be a pain (say for a journey from Stockport to Timperley) with there being no overbridge at Navigation Road. However, is there actually room at the location to build one?

Speaking as a very local resident, closure of either Navigation Road or Deansgate Lane level crossings is an absolute no-no: as well as the bus routes mentioned they are very important local routes, though a deliberate "narrowing" of Deansgate Lane near the crossing stops larger vehicles from using it. The lack of even a pedestrian footbridge at Navigation Road station+level crossing is very much "a pain": I live close to Timperley tram stop, and if I am heading home by train and passing through Stockport, if a mid-Cheshire line train is "soon" I will change at Stockport and Navigation Road onto a tram for the one stop, as it is usually quicker than going via central Manchester. However as has been said there is no room for a footbridge there. If I decide to travel by train from Navigation Road and not use the tram to get there, I know to use a slightly longer walking route to get to the correct platform rather than risk losing many minutes at either level crossing.

I will say that these are 'lightweight' complaints, I've lived here for over 30 years and see no reason to move....