Which junctions would you grade-separate?

[mike57]

How frequent is the service on the Yarm line? As far as the level crossings go we're probably not talking about Poole or Egham style down times here, are we? Especially if the only trains to usually use them are northbound.

There is a double track line which crosses various roads on the level. Freight headed to or from Yarm currently uses them in both directions as there then a burrowing junction to the south that gets freight on to the ECML southbound slow line. I think the whole layout is a leftover from the days when there was also the diverging line towards Ripon and Harrogate. Its a while since I was there, I used to live over that way years ago, and if there were a couple of freights going through barriers could be down for a while (memory is up to 10 mins). If passenger services also used the route then its one TPE train per hour each way and a few GC services spread throughout the day to add in to the mix.

 

[Zomboid]

If passenger trains run that way, the southbound trains can still use the existing station without conflicting with northbound trains, so then you'd be looking at 1tph extra (plus GC) over the level crossings.

 

[lachlan]

I don't think anyone has mentioned
a) Wootton Bassett
b) Ordsall Lane - if road over- and under-bridges allow without excessive gradients.

Is Wootton Bassett a problem / likely to become a problem in the future?

There is talk of hourly Bristol-Oxford trains plus if Melksham ever went hourly this would add to trains using this junction. But there are other constraints like the single-track Melksham line and capacity through Bath.

 

[positron]

Is Wootton Bassett a problem / likely to become a problem in the future?

There is talk of hourly Bristol-Oxford trains plus if Melksham ever went hourly this would add to trains using this junction. But there are other constraints like the single-track Melksham line and capacity through Bath.

I think it limits how many stopping services you could have which limits the BCR for re-opening stations on the nearby lines. One of them being wootton Bassett station itself.

Though I think a new through platform and quad tracking from Swindon to the junction would actually help capacity more than grade separation.

 

[swt_passenger]

The only thing I think you can do with Newark Flat Crossing is to put the east-west route (NOB1) onto a viaduct/embankment with the curve junction (NSE) starting further east to ease the gradient. You may also have to change the A46 Newark Bypass so that it would go under the new viaduct instead of over it.

It'll be expensive either way!

In the 2020 thread about the Newark Crossing, @Bald Rick explained that a conventional flyover at 1:100 gradient does fit in the roughly 800m distance between going under the A46 bridge and over the ECML. No need for another redesign of the bypass.

 

[lachlan]

I think it limits how many stopping services you could have which limits the BCR for re-opening stations on the nearby lines. One of them being wootton Bassett station itself.

Though I think a new through platform and quad tracking from Swindon to the junction would actually help capacity more than grade separation.

Good point. Corsham station is another one though I don't know whether the line through Bath can support many more trains.

 

[mike57]

If passenger trains run that way, the southbound trains can still use the existing station without conflicting with northbound trains, so then you'd be looking at 1tph extra (plus GC) over the level crossings.

It would make sense, you'd have to reinstate the platform(s) and make a route from the main station, from memory its about 100yds to the west. Doable, but in the current climate probably wont happen. If you are doing it you may as well reinstate the Southbound as well.

 

[Zomboid]

In this fantasy land then yeah, put a platform on the southbound line too if the budget stretches that far. But if the crossings are a real issue (and it's not like there are loads of freight paths per hour though there) then use the existing platform wherever possible.

 

[zwk500]

My suggestion for the grade separation at Newark: Reroute the Nottingham-Lincoln line to run via new stations at Farndon and Newark South then a flyover over the ECML to run into Newark Northgate on the East side of the WCML tracks, from where trains would carry on towards Lincoln. Newark Castle station could then be closed, and people would instead have direct interchange between the two lines at Northgate. Picture shows a map of this.

How are you fitting in a flyover there with Beacon Hill Road overbridge just south of the existing station?

A more practical option to me would be to divert the line north of the existing, crosing the Trent, ECML, and A1 near South Muskham, and then build a chord down to the old alignment, which would then allow Nottingham-Lincoln trains to run directly into Northgate, then reverse in the Platform and continue on their way in both directions, while freight would bypass the town. Future Provision for a station at Kelham could be included.

Something like this:

It's outrageously expensive and will have horrendous ground conditions but does at least have the room to build it.

 

[Zomboid]

If there's not space for the ramps between the A46 and crossing, you'd gain some length by deviating a little too the north and crossing the ECML just above where it crosses the river. If that's long enough, there's then enough length to get back to the existing line before the A1.

 

[Bald Rick]

My suggestion for the grade separation at Newark: Reroute the Nottingham-Lincoln line to run via new stations at Farndon and Newark South then a flyover over the ECML to run into Newark Northgate on the East side of the WCML tracks, from where trains would carry on towards Lincoln. Newark Castle station could then be closed, and people would instead have direct interchange between the two lines at Northgate. Picture shows a map of this.


View attachment 178496

Good Lord! that would be about 5 times the price of a straight flyover next to the exsiting crossing. And a lot more difficult to get consent for. For minimal benefit!

 

[InTheEastMids]

In the 2020 thread about the Newark Crossing, @Bald Rick explained that a conventional flyover at 1:100 gradient does fit in the roughly 800m distance between going under the A46 bridge and over the ECML. No need for another redesign of the bypass.

If there's not space for the ramps between the A46 and crossing, you'd gain some length by deviating a little too the north and crossing the ECML just above where it crosses the river. If that's long enough, there's then enough length to get back to the existing line before the A1.

A more practical option to me would be to divert the line north of the existing, crosing the Trent, ECML, and A1 near South Muskham,

The A46 Scheme is being designed to not prevent Network Rail's preferred option for grade separation. The screenshot below is from the DCO documentation, dating from as recently as February 2025:


Key text extracted below for anyone with difficulty reading the text in the image:

Network Rail: NR request that the dualling of the A46 considers options which do not adversely impact the ability of the rail industry to remove Newark Flat Crossing by not impinging on a corridor to the south of the Nottingham to Lincoln Railway

National Highways: The Applicant has consulted with NR, the Department for Transport and their design teams to review feasibility proposals for the grade separation scheme and can confirm that the design proposed within the submitted application does not preclude such a scheme from being constructed

 

[brad465]

I think it limits how many stopping services you could have which limits the BCR for re-opening stations on the nearby lines. One of them being wootton Bassett station itself.

Though I think a new through platform and quad tracking from Swindon to the junction would actually help capacity more than grade separation.

Good point. Corsham station is another one though I don't know whether the line through Bath can support many more trains.

I wonder if Bathampton junction would be one to consider for grade separation for this reason, which I can see being a bottleneck should more services be approved through Bath.

Bristol-Bath-Chippenham-Swindon could do with more services that do a local calling pattern, so that passengers flows between these stations don't rely so much on London services which have very high demand for London (from Bath, Chippenham is the third highest flow, Swindon 5th, both with over 100k a year). The Oxford service being approved permanently would certainly help.

 

[Helvellyn]

A flyover at Basingstoke doesn't deal with the conflicting moves there in that Up stopping services from there having to cross all lines to access the Up Slow from platform 1 (Down Slow).

You could deal with that by building a station in Oakley and extending the London-Basingstoke stoppers to Oakley, in effect taking advantage of the existing grade separation where the Salisbury and Southampton lines separate, to get those trains on the the correct line for the return journey to London. And with the bonus of Oakley - which is quite a reasonable sized community - gaining access to the rail network.

That might be viable if a station could be sited between Battledown Flyover and Oakley on the WoE to serve the two Manydown developments - North will be North of the railway and South, well South! www.manydownbasingstoke.co.uk/south/

I suppose an alternate option would be to run through to Micheldever. The A303 is close by to make it potentially attractive as a rail head (if parking can be increased), but as a turnback location you could reopen platform 1 by moving the Up Line across into the old Up loop and use the current platform 1 as a turnback line between the Up and Down lines.

Either option shouldn't need more Drivers because many of the terminators have to be shunted into a siding to free up platform 1, but it would need additional Guards because those ECS moves will be DOO.

 

[brad465]

That might be viable if a station could be sited between Battledown Flyover and Oakley on the WoE to serve the two Manydown developments - North will be North of the railway and South, well South! www.manydownbasingstoke.co.uk/south/

I suppose an alternate option would be to run through to Micheldever. The A303 is close by to make it potentially attractive as a rail head (if parking can be increased), but as a turnback location you could reopen platform 1 by moving the Up Line across into the old Up loop and use the current platform 1 as a turnback line between the Up and Down lines.

Either option shouldn't need more Drivers because many of the terminators have to be shunted into a siding to free up platform 1, but it would need additional Guards because those ECS moves will be DOO.

I'd be surprised if the paths to Micheldever exist, even if terminating off the running lines. By this point the 4-track railway goes down to two and gains freight paths.

 

[The Ham]

Would special permission be needed for an extra mile or so of new third rail?

Almost why bother for such a short distance? Just put a battery in some trains.

 

[Zomboid]

Almost why bother for such a short distance? Just put a battery in some trains.

If they had a fleet of suitable battery trains for other routes (which isn't impossible) that might be a winner, but it seems silly to create a batch for a journey of about a mile.

 

[BrianW]

I wonder whether the paucity of candidates for flying junctions, and the readiness of contributors to comment thereon, says something? The shortage of provision of flying junctions or dive-unders over recent years also might indicate a lack of obvious benefits, esp in relation to capital costs and disruptions. Observations on other threads seem to suggest a certain 'out-of-touchness' of rail 'enthusiasts' ?

 

[HSTEd]

I wonder whether the paucity of candidates for flying junctions, and the readiness of contributors to comment thereon, says something? The shortage of provision of flying junctions or dive-unders over recent years also might indicate a lack of obvious benefits, esp in relation to capital costs and disruptions. Observations on other threads seem to suggest a certain 'out-of-touchness' of rail 'enthusiasts' ?

I think it says more about cost escalation in recent decades.

When a dive under costs £200m or so (see Werrington), the number of plausible candidates is quite small.

 

[Zomboid]

I think it says more about cost escalation in recent decades.

When a dive under costs £200m or so (see Werrington), the number of plausible candidates is quite small.

That doesn't usually affect the suggestions of enthusiasts armed with crayons.

It probably says that most of the obvious and vaguely possible ones have already been done.

 

[Bald Rick]

That doesn't usually affect the suggestions of enthusiasts armed with crayons.

It probably says that most of the obvious and vaguely possible ones have already been done.

Indeed. I had a long think about this last night, and with current service patterns I couldn’t think of many likely candidates at all.

Widespread roll out of ETCS (preferably with ATO) may well change the equation though

 

[zwk500]

Indeed. I had a long think about this last night, and with current service patterns I couldn’t think of many likely candidates at all.

Would there be any stations where services terminate where grade-separation of the terminating services would release any paths?

 

[Bald Rick]

Would there be any stations where services terminate where grade-separation of the terminating services would release any paths?

That was exactly my thought process, and I couldn’t think of any that would be helpful.

 

[DynamicSpirit]

Good Lord! that would be about 5 times the price of a straight flyover next to the exsiting crossing. And a lot more difficult to get consent for. For minimal benefit!

Well yes it would be a lot more expensive to build. But I think 'minimal benefit' is rather understating things. If I lived in Lincoln and wanted to get to London, I'd probably view going from one train every 2 hours to 3 trains every 2 hours (with a change at Newark Northgate for the additional ones) as quite a huge benefit. Likewise if I wanted to get from Newark to Lincoln, I'd regard it as quite a big benefit that every train left from the same station (and the one with much better facilities too) rather than having to decide which of two stations to go to depending on what time I'll reach the station. And if I lived anywhere between Nottingham and Newark, I'd probably regard it as quite a big benefit that I could now, with a simple change at Northgate, get to any of numerous important destinations along the ECML.

 

[Western Sunset]

1. Woking
2. Woking
3. Woking

 

[Bald Rick]

But I think 'minimal benefit' is rather understating things. If I lived in Lincoln and wanted to get to London, I'd probably view going from one train every 2 hours to 3 trains every 2 hours (with a change at Newark Northgate for the additional ones) as quite a huge benefit. Likewise if I wanted to get from Newark to Lincoln, I'd regard it as quite a big benefit that every train left from the same station (and the one with much better facilities too) rather than having to decide which of two stations to go to depending on what time I'll reach the station. And if I lived anywhere between Nottingham and Newark, I'd probably regard it as quite a big benefit that I could now, with a simple change at Northgate, get to any of numerous important destinations along the ECML.

A billion quid extra - which is roughly what it would be - is rather a lot of money to satisfy those relatively minor objectives for very few people.

 

[DynamicSpirit]

Indeed. I had a long think about this last night, and with current service patterns I couldn’t think of many likely candidates at all.

That's arguably not at all surprising when you consider that current service patterns will have been designed around what is achievable with the current infrastructure! If there were many candidates for future grade separation that were needed to run current service patterns, we'd probably be questioning how well the timetable planners who write current service patterns were doing their job!

I'd suggest the more relevant test for grade separation is, would the grade separation allow us to reliably run better service patterns in the future that are likely to be needed. (Or even: Better service patterns that are needed today, but can't be achieved because of the lack of grade separation)

 

[MrPosh]

I see a number of 'votes' for Newark. Although trains could be held at Newark Castle station in order to 'fit in' to a gap in the ECML service, it still needs there to be such a gap- into which perhaps a busier and better paying London train could be fitted. It must be nightmare to timetable and manage, so that has my vote. Did I not see plans being worked up for it in relation to parallel A46 improvements?

.

 

[DynamicSpirit]

I'm surprised to be only the second person suggesting Colwich Junction. Especially now HS2 isn't happening that far north.

The other thing with Colwich is that the LNWR London-Crewe trains are (in my experience) very heavily used these days, and really could do with becoming half-hourly. I would imagine Colwich is one the biggest barriers to that happening (although capacity nearer London is also an issue).

 

[zwk500]

The other thing with Colwich is that the LNWR London-Crewe trains are (in my experience) very heavily used these days, and really could do with becoming half-hourly. I would imagine Colwich is one the biggest barriers to that happening (although capacity nearer London is also an issue).

Paths into London is a far, far bigger issue than Colwich.