Experiences of service recovery (or lack of) on GTR Great Northern

[bramling]

I still think some strategically placed passing loops on the Cambridge line (or, better still, additional platforms at rural stations) to allow flyers to overtake, with more overall recovery time built in for stoppers near these loops, would help. This is probably the best you can do short of adding a third rail on what is clearly the most congested part of the route with no overtaking capability.

In addition, procuring more stock and having massive recovery times at King’s Lynn (i.e. minimum 70 min) with a regularly used third platform so there’s essentially always an available unit to send southbound may reduce the fragility of the Fen line and, so I hear, is being explored.

I think, if the constraints are with infrastructure, no amount of tweaking the timetable will improve this situation if the entire thing fails in the event of even mild disruption which is what we’re seeing now.

It’s never been a route with high PPMs, especially since Thameslink, but it just so happens that the current timetable has made it considerably worse. Is ‘comparatively poor PPMs like before but still better than now’ what we should be aiming for?

The service was pretty reliable in the period say 2005-2015, under the later incarnation of WAGN and the entire FCC tenure.

They only operated 4tph, albeit with the same mix of services, but the big thing was northbound services had a much greater chance of being more-or-less right time at Hitchin.

Also worth mentioning that in those days they had to work with less infrastructure - no Hitchin flyover, no extra island platform at Cambridge.

 

[Magdalia]

If you’re going to try and run that sort of service, significant resilience work was required, and not just in the core. So many locations on the GN still essentially have a 1970s or 1980s layout, yet are trying to squeeze the whole Thameslink thing into them.

Lots of things were done in preparation for Thameslink:

• up slow line 2 between Ally Pally and Finsbury Park, including new platforms at both
• new bay platform at Stevenage
• Hitchin flyover
• platform extensions at various stations
• the new island platform at Cambridge

Prior to that, that until 1999 the maximum speed between Royston and Shepreth Branch Junction was 60mph, and that since then we have also had Cambridge South and a remodelled Shepreth Branch Junction.

Of course its not this simple as there's (a lot!) more to consider, but surely the best solution would be to have the fast, semi-fast (Thameslink) and stopper enter that stretch in that order in fairly quick succession.

On the up road that's what is in the current timetable, but what happens if the fast train is late because of delays on the single line sections or because of congestion at Ely? Do the Thameslink and stopping trains start on time, or do they wait for the late running fast service? Bear in mind that delaying Thameslink will have an impact all the way to Brighton, and delaying the stopper will mean that it misses its booked slot through the 2 track section at Woolmer Green. The stoppers running through the 2 track section on time is the most crucial aspect of maintaining the whole ECML timetable.

On the down road this requires the Thameslink to overtake the stopper on the ECML, and for the fast to overtake both on the ECML, to get them in the right order at Hitchin. It is just as flaky as what we have had since December 2025, except that the knock on delays hit the ECML trains instead, with potential consequences all the way to Scotland.

I still think some strategically placed passing loops on the Cambridge line (or, better still, additional platforms at rural stations) to allow flyers to overtake, with more overall recovery time built in for stoppers near these loops, would help.

I have long experience of the days when the bidirectional layout at Royston was used for overtaking, It is not a magic bullet. Overtaking puts lots of time into the stopping trains, and usually the fast trains still don't get a clear run either, so little time is gained compared to just following the stopper. New loops would be very expensive because of the signalling and OHL implications, and would take a long time to plan and install. Besides, there is already new and relatively new overtaking capacity at Hitchin northbound and at Cambridge South in both directions.

It’s never been a route with high PPMs, especially since Thameslink, but it just so happens that the current timetable has made it considerably worse. Is ‘comparatively poor PPMs like before but still better than now’ what we should be aiming for?

Action is needed to get some "quick wins" on punctuality, but that has to be within the overall constraints of the existing timetable and infrastructure. "Quick wins" are not going to come through new timetables or new infrastructure, because both of those take many years, but that doesn't mean that nothing can be done.

In the short term my priorities would be:

• sorting out turnrounds of trains and drivers at Kings Lynn, making permanent the arrangements introduced in the current temporary timetable
• getting better at splitting and joining at Cambridge
• more resource at Letchworth to clear terminating trains from the platform more quickly
• rejigging the evening ECS moves from Hornsey into Kings Cross to eliminate joins in the platforms

In the medium term, if I was looking for infrastructure interventions that were either relatively low cost or with potential for third party funding, they would be:

• shortening the single track sections south of Kings Lynn and north of Littleport (recognising that redoubling over the river bridges is probably going to be too difficult)
• doing a Marston Vale style station consolidation at Shepreth and Foxton, funded by new houses and being part of a transport hub for Cambridge commuting, which would speed up the stopping trains by about 3 minutes

 

[Failed Unit]

That is the problem with a loop, you could perhaps structure the timetable so the Kings Lynn service always overtook the stopper a certain station. However you are back to the assumption that either / both are on time.

If you look at other places in the UK, such as Retford when faster services are scheduled to overtake a train stopping there. If you hold the stopper for a late express it gets later. If you let the express go it may get back on time, but the stopper is delaying other things.

The Kings Cross - Cambridge was supposed to be half-hourly (it does operate for large parts of the day like this) but imagine how bad 6th would be. However as the thameslink is normally killed off in any disruption, the stopper running every 30 minutes is also welcome. Things are worse since December. Is it bedding in or just impossible. Even at WGC clearing empty stock and platform congestion is a big issue. How long does shunting stock onto Platform 1 take? I think this could help rather then sending it to the sidings as other trains heading north or Moorgate are impacted my the move to the sidings.

 

[sharpener]

On the down road this requires the Thameslink to overtake the stopper on the ECML, and for the fast to overtake both on the ECML, to get them in the right order at Hitchin. It is just as flaky as what we have had since December 2025, except that the knock on delays hit the ECML trains instead, with potential consequences all the way to Scotland.

Been looking at this a bit, hadn't realised how tight it is.

Here is the regular pattern at Welwyn North (for last Wednesday)



and here are the same trains by the time they leave Hitchin for Cambridge, taking 7, 14 and 13 mins



so no overtaking going on. The intermediate stops mean that the 9S slips back towards the path of 1T40 hence will eventually be only 4 mins ahead by the time it arrives at CBG. So this leaves no margin for lateness at all in a train that has come all the way from Brighton.

Who dreamt that up?

 

[bramling]

Been looking at this a bit, hadn't realised how tight it is.

Here is the regular pattern at Welwyn North (for last Wednesday)

View attachment 209104

and here are the same trains by the time they leave Hitchin for Cambridge, taking 7, 14 and 13 mins

View attachment 209107

so no overtaking going on. The intermediate stops mean that the 9S slips back towards the path of 1T40 hence will eventually be only 4 mins ahead by the time it arrives at CBG. So this leaves no margin for lateness at all in a train that has come all the way from Brighton.

Who dreamt that up?

I wonder if one quick win could be altering Cambridge-Brighton to Cambridge-Three Bridges?

Obviously this would require the return of 4tph Bedford-Brighton, which would at least be a simpler way of doing things.

I wonder if this would then mean it would be possible for drivers to do two complete end-to-end trips per duty, which would simplify things a lot.

This isn’t going to solve the issue of delays, but it means at least issues south of Three Bridges shouldn’t affect the Cambridge route, so reducing the likelihood of issues a bit.

 

[sharpener]

I wonder if one quick win could be altering Cambridge-Brighton to Cambridge-Three Bridges?

Obviously this would require the return of 4tph Bedford-Brighton, which would at least be a simpler way of doing things.

I wonder if this would then mean it would be possible for drivers to do two complete end-to-end trips per duty, which would simplify things a lot.

This isn’t going to solve the issue of delays, but it means at least issues south of Three Bridges shouldn’t affect the Cambridge route, so reducing the likelihood of issues a bit.

That would would have a lot of consequences. though I see they sometimes terminate short at Three Bridges anyway.

I am thinking that reverting to the relationships in the previous timetable might have fewer ramifications. Don't know where I can find it though.

Seems to me the down Cambridge stoppers should be timed not far ahead of the ex Brightons at Welwyn, and the fasts close enough behind to overtake both before Hitchin as @Magdalia suggests. This would give the Brightons the longest possible float in the stretch to CBG. Maybe that's how it was.

Meantime I am playing with other RTT pages as OH is en route from INV to CBG and seems to have encountered various problems along the way.

 

[Failed Unit]

Been looking at this a bit, hadn't realised how tight it is.

Here is the regular pattern at Welwyn North (for last Wednesday)

View attachment 209104

and here are the same trains by the time they leave Hitchin for Cambridge, taking 7, 14 and 13 mins

View attachment 209107

so no overtaking going on. The intermediate stops mean that the 9S slips back towards the path of 1T40 hence will eventually be only 4 mins ahead by the time it arrives at CBG. So this leaves no margin for lateness at all in a train that has come all the way from Brighton.

Who dreamt that up?

Heading North the Kings Lynn overtakes the stopper at Welwyn Garden City. The dwell is a little longer as a result. Although I think that was the same in the previous timetable. The Thameslink however does not over take. It stops Ashwell and Morden, I assume to keep a bit of a gap but as others have highlighted it is only just behind at Cambridge.

Looking at the 2 Services.
Great Northern 1342 Finsbury Park - Cambridge 1505
Thameslink. 1408 Finsbury Park - Cambridge 1509
Great Northern 1424 Kings Cross - Cambridge 1517

So the fast does have a bit more wriggle room. The Stopper is frequently held either at the station or the just before the viaduct for the fast to pass it. However both of these have started from KX so that normally happens.

 

[tatitiliti]

How about relocating Welwyn North slightly further south to the very northern extent of the four tracks so all units can run at 100 through Digswell, then adding some extra sidings to reverse London to Welwyn stoppers there instead? This may reduce instances of minor delays accumulating on the ECML stretch.

 

[tatitiliti]

In the medium term, if I was looking for infrastructure interventions that were either relatively low cost or with potential for third party funding, they would be:

• shortening the single track sections south of Kings Lynn and north of Littleport (recognising that redoubling over the river bridges is probably going to be too difficult)
• doing a Marston Vale style station consolidation at Shepreth and Foxton, funded by new houses and being part of a transport hub for Cambridge commuting, which would speed up the stopping trains by about 3 minutes

I like the idea of consolidating Shepreth and Foxton and think it would make a considerable improvement.

However, I think DfT is going to be very reluctant to add any more track to the Fen line given the spiralling costs of the current level of track and trackbed maintenance.

I wonder if it would be cheaper and equally effective to leave Littleport to Downham as is and instead dual between Watlington and just before the old South Lynn/A47 bridge where there are far fewer river bridges and the ground is considerably more stable? It might prevent turnarounds at Lynn getting so badly delayed.

 

[Magdalia]

I wonder if one quick win could be altering Cambridge-Brighton to Cambridge-Three Bridges?

One of the useful features of the origin destination matrix dataset is the regional tag, which makes it fairly easy to pick out by eye the biggest long distance Thameslink flows. Cambridge-Brighton was the 3rd biggest South East region-East of England region flow in 2024/25.
38775 Gatwick-Cambridge
35426 Gatwick-St Albans
26884 Brighton-Cambridge

Furthermore, Cambridge flows to both Gatwick and Brighton in 2024/25 were suppressed significantly by the large number of engineering blocks associated with Cambridge South construction.

I don't see through running Cambridge-Brighton being dropped.

The Thameslink however does not over take. It stops Ashwell and Morden

There are parts of the day where Ashwell and Morden is only served by Thameslink. Various stoppers miss Ashwell and Morden to reduce overall journey time between Hitchin and Shepreth Branch Junction.

How about relocating Welwyn North slightly further south to the very northern extent of the four tracks so all units can run at 100 through Digswell, then adding some extra sidings to reverse London to Welwyn stoppers there instead? This may reduce instances of minor delays accumulating on the ECML stretch.

Moving Welwyn North has been discussed here many times. What you propose involves moving a station from one side of a 30 metre deep valley to the other, with no decent roads in between. Furthermore, it doesn't achieve your objective, and actually makes things worse. Stopping trains would be entering and leaving the two track section at Digswell at much lower speeds than now.

However, I think DfT is going to be very reluctant to add any more track to the Fen line given the spiralling costs of the current level of track and trackbed maintenance.

I wonder if it would be cheaper and equally effective to leave Littleport to Downham as is and instead dual between Watlington and just before the old South Lynn/A47 bridge where there are far fewer river bridges and the ground is considerably more stable? It might prevent turnarounds at Lynn getting so badly delayed.

One of the advantages of shortening the Watlington-Kings Lynn single line at the Kings Lynn end is that Watlington is in the fen but Kings Lynn is on terra firma. Another is that, combined with the proposed unit/driver changes at Kings Lynn, it becomes possible for up trains to depart Kings Lynn before late running down trains have arrived.

From the perspective of limiting the impact of ground shrinkage related speed restrictions, shortening the Littleport-Downham single line section is more important because it is longer. Once the speed restrictions get so severe that it is impossible to run Littleport-Downham in less than 15 minutes, then the timetable falls apart. Paying particular attention to level crossing locations, I think it would be relatively straightforward to add about 2 miles of double track at the Littleport end, because it is already double track at the Littleport bypass, whereas, at the Downham end, the Downham bypass and Denver level crossings are still single track and would have to be upgraded.

 

[Failed Unit]

Is there anywhere you could build a terminating platform on the route between Hitchin and Cambridge to reduce the risk of trains terminating at Letchworth slowing everything down? They can get a Moorgate terminator turned around in about 5 minutes at Welwyn Garden City and Moorgate (Which is slower then the 313s of course). However if a terminating train a Letchworth did have its own platform and no need to shunt then it would mean that no-one would need to go through the train making sure everyone had got off.

I don’t know either Letchworth, Baldock or Royston well enough to know if space exists. It sounds like Letchworth is a no for a 12 car train judging by other posts.

 

[sharpener]

Is there anywhere you could build a terminating platform on the route between Hitchin and Cambridge to reduce the risk of trains terminating at Letchworth slowing everything down?

The yard and stabling to the east of the station at Letchworth is quite big and might have scope for some re-arrangement, but it is a long way from the station proper.

I think the cure is slicker operation (as with splitting and joining at CBG), a production engineer or T&M specialist of old would quickly come up with an optimal procedure and then thorough training would ensure staff do not deviate from it. In your dreams.

The other (theoretical) possiblilty would have been the originally intended 4 platform layout but as someone (?@Magdalia) pointed out, that ship well and truly sailed with the introduction of 12 car trains.

Personally I quite like the idea of adding OHLE to the down loop at Royston but others have not been keen. It would delay the train being overtaken quite a lot as it would have to recess into the loop and then separately draw forward into the platfom (assuming a station stop, as most make). Less useful on the up.

Other ideas have been quite extensively discussed upthread. They used to reverse trains at Baldock but don't any more IIRC. I never understood the thinking for it. It is on an embankment so the civils for any alterations would be very difficult. There was a recent post about re-siting toward Cambridge in conjunction with housing development with a link to the development plan.

 

[bramling]

The yard and stabling to the east of the station at Letchworth is quite big and might have scope for some re-arrangement, but it is a long way from the station proper.

I think the cure is slicker operation (as with splitting and joining at CBG), a production engineer or T&M specialist of old would quickly come up with an optimal procedure and then thorough training would ensure staff do not deviate from it. In your dreams.

The other (theoretical) possiblilty would have been the originally intended 4 platform layout but as someone (?@Magdalia) pointed out, that ship well and truly sailed with the introduction of 12 car trains.

Personally I quite like the idea of adding OHLE to the down loop at Royston but others have not been keen. It would delay the train being overtaken quite a lot as it would have to recess into the loop and then separately draw forward into the platfom (assuming a station stop, as most make). Less useful on the up.

Other ideas have been quite extensively discussed upthread. They used to reverse trains at Baldock but don't any more IIRC. I never understood the thinking for it. It is on an embankment so the civils for any alterations would be very difficult. There was a recent post about re-siting toward Cambridge in conjunction with housing development with a link to the development plan.

The Baldock reversing didn’t really work well, and probably was never expected to, it was something of a sop to the significant housebuilding that Baldock was/is planned to have.

In terms of could a terminating platform be provided somewhere, maybe.

Letchworth is a problem now because the platforms are now 12 cars, with constraints at both ends, plus also the lift towers. I wouldn’t say it’s impossible, but very expensive and probably disruptive.

Baldock I’m not sure there’s space nowadays, might have been in the past.

That essentially leaves Royston, there probably is space where the down loop is.

It is quite ridiculous that we’re even having the discussion about having to spend a significant sum to provide a bay platform just because GTR can’t tip out 8-car trains in less than about 7 minutes!

 

[sharpener]

In terms of could a terminating platform be provided somewhere, maybe.
...

That essentially leaves Royston, there probably is space where the down loop is.

Yes from this satellite view you could do it at the expense of some of the car park. Conveniently the crossover is right up against the existing platform so could be left as a loop killing two birds etc. (What is the shiny thing/?structure at the London end of the pic?)

 

[Magdalia]

That essentially leaves Royston, there probably is space where the down loop is.

Yes from this satellite view you could do it at the expense of some of the car park. Conveniently the crossover is right up against the existing platform so could be left as a loop killing two birds etc. (What is the shiny thing/?structure at the London end of the pic?)

The prospects for using Royston Down Siding for passenger traffic are zero. It isn't passed for passenger traffic, it isn't long enough, it doesn't have main line signalling and it isn't wired. If you look at the Sectional Appendix you will see that is is classified as a siding, not a loop, even though it can be accessed from both ends.

Royston Down Siding (to give it the name used in the Sectional Appendix) is heavily used during the RHTT season. Trains have to stop before entering the siding, then given access with shunt signals. Exits from the siding are controlled by shunt signals too, K244 to go towards Litlington and K253 to go into Royston platform 2. Entrance from the Litlington end is a shunt signal on K977 signal. When 8 and 12 car trains stop at K977, they block the Green Drift foot crossing.

I can't find the speed limit on the points at either end of the siding, but I think it is 10mph. I'm fairly sure the siding would not take an 8 car train, never mind 12, the longest train that I can remember using it was a class 47 dragging a 5 car class 442 to Ely for storage, and that didn't have much room to spare.

Turning the siding into a passenger loop would require at least the following:

• new points at either end, including moving the Litlington end to a new location
• passenger standard track in the loop
• overhead line in the loop
• moving K977 signal to the other side of Green Drift crossing
• a new signal for exiting the loop at the station end
• probably some mitigations at Green Drift crossing
• somewhere else for the RHTTs to use

How much will all that cost (tens of £millions?), who is going to finance it, and how long would it take to design, consult and build?

 

[Failed Unit]

When you look at it like that, you might as well spend the money on extra staff to make sure that empty coaching stock actually clear the sidings quickly at Letchworth. I assume if the fast is right up the tail of the slow, holding the slow at Hitchin and letting the fast use the ladder probably isn’t likely to happen because of the crossing of the London bound lines.

 

[Magdalia]

I assume if the fast is right up the tail of the slow, holding the slow at Hitchin and letting the fast use the ladder probably isn’t likely to happen because of the crossing of the London bound lines.

It can be done, if there is a clear path across the up fast line.

An example from last December shows that it isn't without other risks though:

Overspeed at Cambridge Junction, Hertfordshire, 11 December 2025

www.gov.uk

2. Summary of the incident​

At 10:39 on 11 December 2025, train reporting number 1T20, the 10:12 London King’s Cross to Ely service, operated by Great Northern, travelled at too high a speed across Cambridge Junction. The junction is located immediately north of Hitchin station in Hertfordshire. At this junction, trains travelling north on the East Coast Main Line (ECML) can diverge onto the line towards Cambridge.

Train 1T20 was routed from the Down Slow line to take the diverging route over the junction towards Cambridge by passing over six sets of points and a switch diamond. For this route, the first two sets of points have a permanent speed restriction of 25 mph (40 km/h). The permanent speed restriction for the remainder of this route, over the other points and the switch diamond, is 40 mph (64 km/h). However, when the train began to traverse the first two sets of points it was travelling at 56 mph (90 km/h) instead of 25 mph (40 km/h).

This caused the train to lurch sideways as it traversed the junction, with several passengers falling from their seats onto the floor and the few passengers who were standing being thrown off balance. No injuries were reported following the incident and there was no damage caused to the train or to railway infrastructure.

The incident has its own discussion so it doesn't need to be discussed again here.

Overspeeding at Cambridge Junction (Hitchin) -- 11/12/25

Saw this earlier: https://www.gov.uk/government/news/overspeeding-at-cambridge-junction At 10:39 on 11 December 2025, train reporting number 1T20, the 10:12 London Kings Cross to Ely service, operated by Great Northern (a trading name of Govia Thameslink Railway Ltd), was involved in an...

www.railforums.co.uk

 

[jon0844]

When you look at it like that, you might as well spend the money on extra staff to make sure that empty coaching stock actually clear the sidings quickly at Letchworth. I assume if the fast is right up the tail of the slow, holding the slow at Hitchin and letting the fast use the ladder probably isn’t likely to happen because of the crossing of the London bound lines.

Shutting down a 700 is basically the time to walk through the train. Shutting down a 387 or 379 is probably twice as long given the need to exit and reboard, plus lots more areas to check, if one person. Perhaps 20-30% longer if there's two.

The issue isn't going to be simply shutting a unit down but things like having to do assists on the same unit. If there's a VIP or wheelchair user, they need to be taken off first (and possibly handed over to gateline for further help) and then the shut down starts. If it's a 387 or 379, there's a chance of assists in both coach 3 and 7.

But more than that, a lot of homeless people travel that refuse to get off and some may be in a toilet.

Sometimes a driver will assist and throw them off, but people refusing to alight can't normally be manhandled off and it can take some time to wake or persuade someone to get off.

If they've taken their shoes off and spread all their possessions around, the shut down isn't going to be quick.

I believe Letchworth has two people during the day for platform but I really do think there's always going to be a risk of delay when shutting down a train on a 'live' line. Stop shutting down trains as much as possible, especially during the peaks, and the issues disappear there.

Do the problems move up the line? Perhaps, assuming the train won't just turn around at the destination.

Another solution might be having staff able to remove passengers and their belongings with a bit more 'encouragement', perhaps Travel Safe Officers, but there's then a high chance there's going to be an assault on staff at some point.

 

[philjo]

in 2019 or possibly early in 2020 prior to Covid restrictions, there was a public meeting held in Letchworth for the various rail user groups on the Cambridge branch and GTR staff and local MP present.
The possibility of converting the down goods loop at Royston into a new bay platform was mentioned as a way to enable additional peak services starting/terminating at Royston without blocking the through lines. It would have to be limited to 8 coaches and would require removal of some of the adjacent car park.
I haven’t heard anything more about this since that meeting.

An alternative option which would require extra land is to convert Meldreth to have an island platform on the up side (currently next to a field) with the new central line being a peak flow bidirectional platform to allow fast trains to overtake stoppers in either direction, or some stoppers starting fro there instead of Royston.

 

[Magdalia]

An alternative option which would require extra land is to convert Meldreth to have an island platform on the up side (currently next to a field) with the new central line being a peak flow bidirectional platform to allow fast trains to overtake stoppers in either direction, or some stoppers starting fro there instead of Royston.

This isn't going to happen, at least not without tens of thousands of new houses. This is similar in magnitude to Beaulieu Park, and would also require replacement of the bridge that carries Station Road over the railway at the south end and/or replacement of the bridge that carries the railway over the River Mel at the north end.

As with Royston, how much will all that cost (tens of £millions?), who is going to finance it, and how long would it take to design, consult and build?

 

[Bald Rick]

An alternative option which would require extra land is to convert Meldreth to have an island platform on the up side (currently next to a field) with the new central line being a peak flow bidirectional platform to allow fast trains to overtake stoppers in either direction, or some stoppers starting fro there instead of Royston.

In short, you don’t want a simple overtaking facility on the branch as it a) hardwires the timetable around that very point and b) extends journey times for the stoppers.

 

[London Trains]

Would the Cambridge stoppers be able to cope with loadings in the peaks if they weren't split into two separate services in the peak? Maybe with the addition of the KGX-Peterborough peak extras calling at Welwyn North, Knebworth and Hitchin (and potentially the Kings Lynn services calling at Baldock if necessary)?

The reason I ask this is that my suggestion would just be to not have anything terminating at Letchworth, with the off-peak terminators re-extended to Cambridge, but that still leaves the peak extras.

 

[bramling]

Would the Cambridge stoppers be able to cope with loadings in the peaks if they weren't split into two separate services in the peak? Maybe with the addition of the KGX-Peterborough peak extras calling at Welwyn North, Knebworth and Hitchin (and potentially the Kings Lynn services calling at Baldock if necessary)?

The reason I ask this is that my suggestion would just be to not have anything terminating at Letchworth, with the off-peak terminators re-extended to Cambridge, but that still leaves the peak extras.

Probably not.

There’s the scope to extend the Peterborough extras to 12 cars (since Covid all but one of these are 8 car), however the 2tph extra is pretty important especially for the stations between Baldock and Knebworth).

If you don’t run them, what tends to happen is people end up taking the slower services from further out (essentially what happened before 2018, and even more so before FCC’s “seats for you” timetable), which would then cause severe overcrowding further in.

All this because of Thameslink. It’s very yes prime minister like. You can’t have 2tph extra extra capacity without Thameslink because King’s Cross is full, but because that can’t run workably we will now have to take 2tph away. Guess a few people get to make their couple of times a year journey to Gatwick without having to change…

 

[Simon_DM]

In the medium term, if I was looking for infrastructure interventions that were either relatively low cost or with potential for third party funding, they would be:

• shortening the single track sections south of Kings Lynn and north of Littleport (recognising that redoubling over the river bridges is probably going to be too difficult)
• doing a Marston Vale style station consolidation at Shepreth and Foxton, funded by new houses and being part of a transport hub for Cambridge commuting, which would speed up the stopping trains by about 3 minutes

The bridges over the River Great Ouse and the Ouse Relief Channel were constructed to accommodate two tracks, so redoubling across them may not be quite as difficult as first appears (the current single line runs on the eastern side of both bridges). The greater complication may be the overhead line equipment and other infrastructure between Littleport and Downham Market, much of which has been installed on or close to the former track formation.

Nevertheless, redoubling the Littleport–Downham Market section could deliver significant improvements in operational resilience and timekeeping. At present, signallers have to decide whether to hold an up service when a down train is running late at Littleport. That delay can then propagate through the service: if the late-running down train has a short turnaround at King’s Lynn, the subsequent up working may also depart late (others have suggested improvements at KLN to help mitigate these issues).

There is no straightforward answer. Although the December 2025 timetable appears to have reduced the likelihood of conflicting up and down movements at the single-line section, conflicts still occur when services are disrupted. The capacity constraints around Ely North Junction add another, rather more substantial, complication.

More generally, the reliability of Great Northern’s Fen Line service is currently woeful. From a passenger’s perspective, there appear to be recurring maintenance problems across several parts of the railway: overhead line equipment — including the failure that reportedly damaged more than half of the Class 379 fleet this month — signalling, track and rolling stock. The unreliable air conditioning on some of the 379s is another increasingly familiar problem.

The condition of the track formation across the peatland is a separate and particularly difficult issue that is unlikely ever to have a simple or permanent solution. Even so, more effective maintenance of the infrastructure and rolling stock ought to be capable of delivering worthwhile improvements while the larger and more expensive capacity schemes are considered.

 

[Mattplans]

Any news on disruption on the MML this evening? As there are OHLE issues, I am wondering if it is related to those issues with pantographs on 700s

 

[philjo]

Lots of cancellations today. It appears that there was a broken down train at Farringdon earlier. Several Cambridge-Brighton services are cancelled. The screens at Hitchin were running in disruption mode when I got there late afternoon. As it is Saturday the stoppers only go to Cambridge hourly and some of those were also cancelled/delayed. The stoppers appeared to be providing the only service to stations on the Cambridge branch for a time this afternoon.

 

[jon0844]

Lots of cancellations today. It appears that there was a broken down train at Farringdon earlier. Several Cambridge-Brighton services are cancelled. The screens at Hitchin were running in disruption mode when I got there late afternoon. As it is Saturday the stoppers only go to Cambridge hourly and some of those were also cancelled/delayed. The stoppers appeared to be providing the only service to stations on the Cambridge branch for a time this afternoon.

An incident at Oakleigh Park too.

 

[Intercity_225]

A rare case of a Letchworth service overtaking a Welwyn stopper!

https://www.realtimetrains.co.uk/service/gb-nr:C18624/2026-08-17/detailed

 

[bramling]

A rare case of a Letchworth service overtaking a Welwyn stopper!

https://www.realtimetrains.co.uk/service/gb-nr:C18624/2026-08-17/detailed

Interestingly, this service is actually booked to use the down fast between Alexandra Palace and Potters Bar, presumably to overtake a freight path. I use this service semi often, and it’s about 40/60 whether it actually uses the fast. One wonders if they actually managed to get something right by mistake, given how poor the regulating from York often is.

2R01 ran non-stop Letchworth to KX this morning (I’m not sure if it actually took on any passengers at Letchworth). They did manage to run that on the up fast from Digswell right through to KX, but then managed to get it held at signals both coming in and out of KX, so the next service still ran a few minutes late.

 

[TSGN Spotter]

OHLE issues between Downham Market and Kings Lynn. Signaller at Watlington reported significant continuous arch from rear pan on 1T55