North London improvement ideas?

[leytongabriel]

Like a sycophantic Royal name would be, by over 60s who don't use PT? Or pay for it.

I think Northern City is a bit mundane too, and of course overlaps with Northern line (by design) and Waterloo & City line too.

If it was to get a Tube like singular name, I might suggest the Islington Line, given that it runs through a majority amount of the borough, and of course stops at H&I - and services Arsenal, Essex Road, Upper Street and Finsbury Park itself. Plus parts of Old St and Moorgate areas are in the LB as well.

Then we'd have the Izzy line as well as the Lizzie line!

 

[Class15]

I agree generally, but the missing link on the north side of Thameslink is the Hertford Loop, and Crews Hill New Town is going to need Thameslink more than the Northern City Line. A Thameslink Welwyn/Crews Hill-Sevenoaks/Orpington wouldn't be everywhere to everywhere, it is a 4tph Metro Alexandra Palace-Bickley, and similar to the St Albans-Sutton loop trains.

The everywhere to everywhere problems for Thameslink are elsewhere, mainly Horsham-Peterborough and Luton-Rainham.


This can't be done because there are no paths on the fast lines for the Kings Cross-Letchworth/Cambridge trains. They still have to run on the slow lines, and, the more intermediate calls they have, the less different they are from the all stations service.

For Crews Hill New Town to have maximum chance of success, it needs trains to the Thameslink core. And the stations between Bowes Park and Gordon Hill will benefit from Thameslink too. The Hertford loop already shares with Thameslink between Finsbury Park and Alexandra Palace so it isn't a homogeneous operation now.

The Northern City Line service means that the Hertford loop would not be completely reliant on Thameslink, like Hitchin-Peterborough. Adding 2tph Thameslink trains isn't going to break it.

Right. Let me establish all the drawbacks of this plan.

1: (already asked without answer): what is being done to provide the southern MML with an appropriate level of service during the peaks, if the 2tph Luton-Orpington is being lost?
2: what is the great problem with a cross-platform interchange at Finsbury Park?
3: if the Hertford Loop is turned over to Overground, wouldn’t it make more sense to run a frequent overground service?
4: why are we adding yet more branches to Thameslink, giving it more chance of falling over?

etc etc

 

[Magdalia]

Letchworth/Cambridge enough by removing some calls, or is it a service eternally destined to be slow due to the Digswell bottleneck and fast line capacity constraints?

Yes. Sometimes I wonder whether the solution is to tun all the way Moorgate-Letchworth with class 717s. This needs another way to serve the villages between Royston and Cambridge, but given that most passengers at the villages want to go to/from Cambridge, I'm not convinced that has to be through trains to/from London.

1: (already asked without answer): what is being done to provide the southern MML with an appropriate level of service during the peaks, if the 2tph Luton-Orpington is being lost?

In the evening peak there is one Orpington-Luton and no Luton-Orpingtons. There is (almost) mothing to do.

The morning peak is different, there are Luton-Orpingtons and vice versa because they fill in for Rainham trains that don't run through to/from Luton, and terminate/start at West Hampstead instead.

Luton to/from the core need not lose any trains. If there are any gaps to be filled this can be done by running to/from Elephant and either using the Herne Hill turnback or the sidings at Bellingham. That might be the Luton trains, or it might be the Crews Hill trains, depending on how the timetable fitted together.

2: what is the great problem with a cross-platform interchange at Finsbury Park?

It makes lots of journeys 2 changes instead of 1. The final destination isn't necessarily a Thameslink station with direct trains at Finsbury Park.

3: if the Hertford Loop is turned over to Overground, wouldn’t it make more sense to run a frequent overground service?

Running 2tph to/from Thameslink does not prevent that.

4: why are we adding yet more branches to Thameslink, giving it more chance of falling over?

Thameslink is already impacted by what happens on the Hertford Loop because it shares tracks between Finsbury Park and Alexandra Palace, especially platform capacity at Finsbury Park.

 

[Farigiraf]

Yes. Sometimes I wonder whether the solution is to tun all the way Moorgate-Letchworth with class 717s. This needs another way to serve the villages between Royston and Cambridge, but given that most passengers at the villages want to go to/from Cambridge, I'm not convinced that has to be through trains to/from London.

Do any of the stations between Stevenage and Royston have turnback facilities for trains coming from the north end? There may be a case for doing away completely with the Letchworth/Cambridge slows, which typically don't fill more than a 4-car off-peak anyway, and instead extending Sevenoaks-Welwyn to Letchworth, taking over the paths through Digswell of the GN Letchworth/Cambridge stopper. Then the village stations between Royston and Cambridge would have a shuttle service to the nearest station with turnback facilities, hence my question at the start, with good connections onto a 1st stop Finsbury Park Thameslink service to London. Then the problem shifts to there not being direct trains from Cambridge to Welwyn, making e.g. Cambridge - Brookmans Park a 2-change journey, but this is less impactful than other options especially if you ensure good connections.

 

[Stephen42]

In the evening peak there is one Orpington-Luton and no Luton-Orpingtons. There is (almost) mothing to do.

The morning peak is different, there are Luton-Orpingtons and vice versa because they fill in for Rainham trains that don't run through to/from Luton, and terminate/start at West Hampstead instead.

Luton to/from the core need not lose any trains. If there are any gaps to be filled this can be done by running to/from Elephant and either using the Herne Hill turnback or the sidings at Bellingham. That might be the Luton trains, or it might be the Crews Hill trains, depending on how the timetable fitted together.

At the start of service both Orpington and Gillingham/Rainham send all trains to Luton. The Orpington ones get there first with two Orpingtons before the first Gillingham arrives. There are three Orpington services south before the first Rainham as one unit is stabled on the north side.

There are 4tph starting from Luton between 7am and 8:45am. This is 2tph Rainham and 2tph in the Orpington path with half finishing at Elephant & Castle instead of running through to Orpington. 5 trains from the south in the morning peak finish at West Hampstead when they would arrive at Luton after the 4tph period. This is a mix of Orpington and Rainham likely for stock rotation purposes. If the Orpingtons ran all day most of these would remain in service for Orpington to Kentish Town though a couple would stable with the reduced running time.

If the Orpingtons became all day and ran to Finsbury Park instead those early morning 2tph would need to come from somewhere else. Also would need to keep running those paths for the 4tph period. With 1 hour run time on the existing Luton to Elephant & Castle plus 15 min turnaround time at Luton, even a bounce back at Herne Hill would need 5 units on the circuit. To Crews Hill or Hertford North might be one unit less than the current Orpington to luton. That's quite a few extra 700s to be found and stabled, when there are several 717s unused at peak times which could run additional Crews Hill/Hertford North services if the drivers and funding existed for them.

Running 2tph to/from Thameslink does not prevent that.

The Moorgate timetable would need to be built around the Thameslink paths though. The core is constrained and if not wanting Luton to lose out now need to find 10tph from Blackfriars to Elephant instead of 8tph today. None of the existing Hertford Loop bays are long enough for a class 700 and would want a significant layover that would need to be on the running lines. The platforms generally not being long enough to make full use of the class 700 may lead to worse distribution than unusual as the rear coach will rarely open.

The Moorgate trains wouldn't be able to skip stops to shorten the circuits to reduce unit requirements and may have pathing time to wait for the Thameslink train to return south. Some of these things could be improved with infrastructure interventions, but that's going to be far more costly than employing some more drivers (which would need to do anyway if switching) and running extra 717s to increase Hertford Loop service.

Thameslink is already impacted by what happens on the Hertford Loop because it shares tracks between Finsbury Park and Alexandra Palace, especially platform capacity at Finsbury Park.

At Finsbury Park typically the Moorgates will use 1 and 8 with the Thameslinks on 2 and 7. It doesn't always work and there are some impacts between the two but it is relatively minimal as shared use goes. If something happens on the Hertford Loop a Thameslink train destined for it may block other Thameslink trains in a way that doesn't happen today.

 

[321362]

Do any of the stations between Stevenage and Royston have turnback facilities for trains coming from the north end?

Royston does, and possibly Letchworth but Traksy isn't being clear

 

[NCT]

Those Orpington paths belong to the MML. To suggest a permanent 6tph cap for Brent Cross is ok is ridiculous.

 

[Class15]

At the start of service both Orpington and Gillingham/Rainham send all trains to Luton. The Orpington ones get there first with two Orpingtons before the first Gillingham arrives. There are three Orpington services south before the first Rainham as one unit is stabled on the north side.

There are 4tph starting from Luton between 7am and 8:45am. This is 2tph Rainham and 2tph in the Orpington path with half finishing at Elephant & Castle instead of running through to Orpington. 5 trains from the south in the morning peak finish at West Hampstead when they would arrive at Luton after the 4tph period. This is a mix of Orpington and Rainham likely for stock rotation purposes. If the Orpingtons ran all day most of these would remain in service for Orpington to Kentish Town though a couple would stable with the reduced running time.

If the Orpingtons became all day and ran to Finsbury Park instead those early morning 2tph would need to come from somewhere else. Also would need to keep running those paths for the 4tph period. With 1 hour run time on the existing Luton to Elephant & Castle plus 15 min turnaround time at Luton, even a bounce back at Herne Hill would need 5 units on the circuit. To Crews Hill or Hertford North might be one unit less than the current Orpington to luton. That's quite a few extra 700s to be found and stabled, when there are several 717s unused at peak times which could run additional Crews Hill/Hertford North services if the drivers and funding existed for them.

The Moorgate timetable would need to be built around the Thameslink paths though. The core is constrained and if not wanting Luton to lose out now need to find 10tph from Blackfriars to Elephant instead of 8tph today. None of the existing Hertford Loop bays are long enough for a class 700 and would want a significant layover that would need to be on the running lines. The platforms generally not being long enough to make full use of the class 700 may lead to worse distribution than unusual as the rear coach will rarely open.

The Moorgate trains wouldn't be able to skip stops to shorten the circuits to reduce unit requirements and may have pathing time to wait for the Thameslink train to return south. Some of these things could be improved with infrastructure interventions, but that's going to be far more costly than employing some more drivers (which would need to do anyway if switching) and running extra 717s to increase Hertford Loop service.

At Finsbury Park typically the Moorgates will use 1 and 8 with the Thameslinks on 2 and 7. It doesn't always work and there are some impacts between the two but it is relatively minimal as shared use goes. If something happens on the Hertford Loop a Thameslink train destined for it may block other Thameslink trains in a way that doesn't happen today.

Those Orpington paths belong to the MML. To suggest a permanent 6tph cap for Brent Cross is ok is ridiculous.

I agree wholeheartedly with both these posts. Stephen42, thanks so much for expressing and analysing these ideas in a way I don’t have the knowledge to do. I think this proves that the idea of sending the Orpington service to Crews Hill is not a good one and should not happen. And agreed with NCT as well, I think those Orpington paths absolutely need to go onto the MML.

 

[Farigiraf]

Those Orpington paths belong to the MML. To suggest a permanent 6tph cap for Brent Cross is ok is ridiculous.

Provided there is capacity in the core (which I do think there is, as tph in each direction is currently 10-12tph less than the maximum potential frequency) there's no reason why you couldn't run both, with a separate ECML/Crews Hill - Orpington and Luton - Caterham for example. But the Hertford Loop can also do perfectly well with up to 8tph self-contained with cross-platform transfers at Finsbury and Highbury.

 

[cle]

The Hertford Loop does not need two operators. If Crews Hill indeed becomes a big source of demand, then I would suggest to redirect WGC-Moorgate services, and simplify further.
And then you'd have say 12tph Moorgate-Gordon Hill, with 2 turning there, 4 at Crews Hill, 4 at Hertford North, 2 at Stevenage - or so. All day. It's a metro.

Those diverted WGC services could go to Thameslink if possible (swap from Peterborough-Horsham, I think Londoners use core services more and more days per week, P'boro goes all Kings Cross) - or to Kings Cross.

The Hertford Loop service gives an awful lot of options:
Finsbury Park (TL services, KX and Vic/Picc downstairs)
Highbury (cross to Vic - winner - and ELL/NLL for east and west)
Old St (many jobs, Northern line)
Moorgate (SSL, Northern line and Elizabeth)

it's much more useful than Chiltern or even Euston services - gives far more options for onward travel as well as walkable work. Under-rated.

 

[Dave W]

If Crews Hill indeed becomes a big source of demand

This being the big question. LB Enfield (and partners) will have to go some way to convince new towners that they don't need a car up there, especially so close to two M25 junctions. I don't think the success of Crews Hill Town will rest on its connectivity to the Thameslink Core, either.

Re success of existing Hertford Loop - I have to say I've never seen it not full, save around Covid. I pick up Mrs DW from Winchmore Hill as Mini DW's nursery is round the corner - usually the 1757. The numbers piling out just at the one station are eye opening, especially considering neither Palmers Green nor Grange Park are all that far away.

How many tph could the Loop serve? 12 seems high but I don't know the technical ins and outs.

 

[NCT]

Considering at the moment it's only 4tph in the shoulder peaks and 6tph in one high peak hour, there's room to grow to peak 8tph (possibly 10) and that should be adequate.

 

[Brubulus]

This being the big question. LB Enfield (and partners) will have to go some way to convince new towners that they don't need a car up there, especially so close to two M25 junctions. I don't think the success of Crews Hill Town will rest on its connectivity to the Thameslink Core, either.

Re success of existing Hertford Loop - I have to say I've never seen it not full, save around Covid. I pick up Mrs DW from Winchmore Hill as Mini DW's nursery is round the corner - usually the 1757. The numbers piling out just at the one station are eye opening, especially considering neither Palmers Green nor Grange Park are all that far away.

How many tph could the Loop serve? 12 seems high but I don't know the technical ins and outs.

8tph is the reasonable limit for the route, and would give it a similar capacity to the outer branches of most tube lines.

Remember there's only 25 717s and 4tph is needed on the Welwyn route, so that's the absolute maximum the fleet would stretch to.

 

[MarkyT]

Delete please

 

[Dave W]

Great Northern - Northern City Line to TFL?

Just found this article via Reddit, regarding TFL submitting a business case to take over the GN Northern City line, haven't seen a recent thread anywhere else but apologies if there is! Link to RAIL article What are our thoughts on this? How likely is it to happen? Would it be incorporated...

www.railforums.co.uk

I note that there's an actual, non-speculative, discussion of the Hertford Loop to TfL bit in the main forum now.

So shall we get back to our regular programming of crayon-eating and pipe dreaming

 

[Instinct04]

What's the least-bad compromise? Having the Letchworth/Cambridge crawl behind all-stops services on the slow lines (meaning Foxton/Shepreth/Meldreth passengers should change at Stevenage, and Knebworth/Welwyn North get a really slow service unless they double-back at Stevenage - due to no fast line calls at all from WGC), or a less frequent Welwyn 'metro' (i.e. just 2tph GN and 2tph Thameslink) giving more capacity to route a semifast service through?
Even with 4tph to Welwyn, would you be able to increase the speed of the Letchworth/Cambridge enough by removing some calls, or is it a service eternally destined to be slow due to the Digswell bottleneck and fast line capacity constraints?
Either way, there will be losers - lower frequency to suburban ECML stations, or slower services for Knebworth, Welwyn North etc.
The tremendous cost of quad-tracking through Welwyn North and over Digswell Viaduct, after which you'd probably still have the fast lines at capacity, shows the need for things like the HS2 eastern branch or a new-build high-speed line paralleling the ECML in sections.

The 2CXX services need to DROP Alexandra palace ASAP.

It‘s been said before, and I’ll say it again.

Then maybe if there’s room, they could sneak onto the fasts at Potters Bar? Probs not feasible.

Thameslink should absolutely go to Hertford North.

Would it run as a stopper or semi fast, skipping the likes of bowes park, Grange Park, bayford etc. ?

 

[Instinct04]

Great Northern - Northern City Line to TFL?

Just found this article via Reddit, regarding TFL submitting a business case to take over the GN Northern City line, haven't seen a recent thread anywhere else but apologies if there is! Link to RAIL article What are our thoughts on this? How likely is it to happen? Would it be incorporated...

www.railforums.co.uk

I note that there's an actual, non-speculative, discussion of the Hertford Loop to TfL bit in the main forum now.

So shall we get back to our regular programming of crayon-eating and pipe dreaming

Ok. My turn

- 2tph Stansted to Peterborough, 1tph Norwich and 1tph XC Birmingham. 4tph Liverpool Street. Achieved by doubling Stansted tunnel

- Remove Roydon on the semi fast Stratford to Bishops Stortford service, and give it to Brimsdown, as that station recieves just over 1.1 million passengers a year. (Northumberland Park gets 3tph but it has less passengers than Brimsdown)

- 4tph Hertford/WGC all day. Remove Ally Pally stop on Cambridge to Kings Cross slow trains.

- Platforms on Willesden Junction WCML slow lines and WLL. Would allow for better interchange opportunities

- Increase weaver line frequencies to 3tph Cheshunt and Enfield off peak, and 4tph on both branches during the peak.

- West Hampsted Interchange with Chiltern, Thameslink, EMR connect, Metropolitan and Jubilee line trains all calling

- New station at Picketts Lock in Edmonton

- New station near Carterhatch Lane in Enfield

- Stop a few semi fast services at Hayes and Harlington for interchange purposes to Heathrow Airport. (Would be for people coming from the West)

 

[A S Leib]

- Platforms on Willesden Junction WCML slow lines and WLL. Would allow for better interchange opportunities

- Stop a few semi fast services at Hayes and Harlington for interchange purposes to Heathrow Airport. (Would be for people coming from the West)

Better interchange opportunities could be provided by stopping everything on the GWML at Old Oak Common / Willesden Junction once opened.

- West Hampsted Interchange with Chiltern, Thameslink, EMR connect, Metropolitan and Jubilee line trains all calling

I agree in theory, but I think it would be too expensive to build now. Does the Chiltern Main Line (and Aylesbury via Amersham) have, or would it have with the frequency it had in December 2019, the capacity for everything to call at West Hampstead without improvements elsewhere?

 

[Instinct04]

I agree in theory, but I think it would be too expensive to build now. Does the Chiltern Main Line (and Aylesbury via Amersham) have, or would it have with the frequency it had in December 2019, the capacity for everything to call at West Hampstead without improvements elsewhere?

Unless you stopped everything there, then probably not.

The Chiltern lines go off in a weird direction and don’t really serve anywhere important (bar birmingham, Oxford and Aylesbury)

 

[Recessio]

Electrify the Greenford branch with 4-rail DC and convert it to be a Central line branch (or even just the existing West Ealing-Greenford shuttle service)

 

[321362]

Electrify the Greenford branch with 4-rail DC and convert it to be a Central line branch (or even just the existing West Ealing-Greenford shuttle service)

I think the opposite, I think OHLE and run it as either an Elizabeth Line branch, or a new Overground line like the Liberty Line (which I think should also be Elizabeth Line to be honest)

 

[Instinct04]

I think the opposite, I think OHLE and run it as either an Elizabeth Line branch, or a new Overground line like the Liberty Line (which I think should also be Elizabeth Line to be honest)

If it was part of the Lizzy, you’d have to spend money to extend all the platforms to accommodate 345s (unless the branch would have its own trains) I don’t know if the usage of the line would justify that.

 

[321362]

If it was part of the Lizzy, you’d have to spend money to extend all the platforms to accommodate 345s (unless the branch would have its own trains) I don’t know if the usage of the line would justify that.

Get a 323 or two once Northern get rid of theirs?

 

[Recessio]

I think the opposite, I think OHLE and run it as either an Elizabeth Line branch, or a new Overground line like the Liberty Line (which I think should also be Elizabeth Line to be honest)

Elizabeth line would be a nightmare getting paths into/out of Paddington, which I believe is why the line terminates at West Ealing now. I guess you could keep it as a West Ealing-Greenford shuttle Overground shuttle, you're right, perhaps in the same style as the Liberty line between Romford-Upminster. But I feel like not having to change at Greenford would be nice too.

== Doublepost prevention - post automatically merged: 14 Jan 2026 ==

If it was part of the Lizzy, you’d have to spend money to extend all the platforms to accommodate 345s (unless the branch would have its own trains) I don’t know if the usage of the line would justify that.

This is why I was thinking make it a branch of the Central line (or a Central line shuttle) - you can get away with shorter platforms

 

[321362]

If it was 345s you could close Castle Bar Park since if the rear of the train's at Drayton Green the front will already be halfway to South Greenford!

 

[MarkyT]

Elizabeth line would be a nightmare getting paths into/out of Paddington, which I believe is why the line terminates at West Ealing now. I guess you could keep it as a West Ealing-Greenford shuttle Overground shuttle, you're right, perhaps in the same style as the Liberty line between Romford-Upminster. But I feel like not having to change at Greenford would be nice too.

== Doublepost prevention - post automatically merged: 14 Jan 2026 ==


This is why I was thinking make it a branch of the Central line (or a Central line shuttle) - you can get away with shorter platforms

Or a shuttle branch branded in EL purple but using a short train of suitable stock. Reflects the majority use of the line as a feeder for the EL. Which TfL division and contractor actually runs and maintains the trains is not really important from a passenger perspective as long as quality is maintained. Maybe all the Thames Valley branches might be run this way, with GWR or its GBR successor as the operator.

 

[NCT]

Nobody is going to fund running 9-car trains on the Greenford branch.

The Elizabeth line operation will not absorb any non-standard odds and sods.

 

[Instinct04]

Nobody is going to fund running 9-car trains on the Greenford branch.

The Elizabeth line operation will not absorb any non-standard odds and sods.

Exactly.

Elizabeth line would be a nightmare getting paths into/out of Paddington, which I believe is why the line terminates at West Ealing now. I guess you could keep it as a West Ealing-Greenford shuttle Overground shuttle, you're right, perhaps in the same style as the Liberty line between Romford-Upminster. But I feel like not having to change at Greenford would be nice too.

I think that if it was a London Overground shuttle, perhaps at a more attractive 4tph frequency only from West Ealing to Greenford it could work. Direct trains aren't everything.

Just look at the flow of passengers between the likes of Cambridge/Audley End/Stansted Airport to Stratford (They all change at Tottenham Hale)

If you have short shuttle routes at high frequencies that connect to longer distance routes at high(ish) frequencies, then it makes changing trains less of a hassle.

 

[cle]

I think that if it was a London Overground shuttle, perhaps at a more attractive 4tph frequency only from West Ealing to Greenford it could work. Direct trains aren't everything.

There just isn't the local population to support 4tph. The only thing I've seen (and not for a while, since the OOC plan) - that might see more uptake was threading in services to Chiltern, e.g. sending 2tph West Ealing to High Wycombe. But would require works at Greenford. And those local calls would then feel very jarring.

 

[NCT]

We should count ourselves lucky the Greenford branch exists in this glorified parliamentary form of 2tph. If it didn't exist nobody would think of building it - the E number buses are already adequate.