Lack of train crew on Great Northern?

[bramling]

The two track section Hitchin to Cambridge compounds the issues too. Generally my observations commuting are that it's the timekeeping on the stopper which causes a lot of problems, in both directions, especially as the services interwork at King's Cross too.

The December recast should help by splitting up the stopping service at peak times - that thread across Welwyn slow to fast to slow doesn't work reliably enough. It's too common for the stopper to be 5 late at Stevenage, held for a fast at Knebworth to then fall behind a Moorgate stopper and lose more time.

I’m not sure turnaround times are that much if an issue - the stopping services get over 20 minutes at the London end, and considerably more at the country end. It’s only really at Kings Lynn on the fast services where things are tight, and I guess the constraints of the infrastructure are a substantial cause of this.

Where the up stopping services seem to lose time is being stuck behind Cambridge to Brighton services, whose Cambridge turnarounds are short. This will actually get worse in due course, when Cambridge South opens.

From a signalling point of view, I find there’s insufficient urgency in pushing up services through the Letchworth to Welwyn section, and once this happens it doesn’t take much for the service to then be stuck behind the 717 starting at Welwyn. Again, in the past use would have been made of the infrastructure to try and correct this (crossovers at Marshmoor, Potters Bar to use the up fast, or use of the up slow no.2 inwards from Alexandra Palace), but this simply doesn’t happen now - and this isn’t always a case of insufficient space on the fast lines, as it can happen at the same extremes of the day as well.

I guess one quick win would be to make use of both platforms at Kings Lynn, which would essentially add half an hour to the turnround time there - albeit at a cost of two extra units having to be out on the railway, albeit without incurring any extra mileage. For this to be effective this also relies on driver turnarounds being similarly lengthy, though with Kings Lynn being a crew depot presumably there’s a fair few services which have a fresh driver take the train south.

Are the staff at York overloaded with work so just unable to be across everything or is it a loss of familiarity and experience with a route to know the best way of resolving an issue on a real world basis rather than a headcode basis?

Shouldn’t really be lack of familiarity, as control has been there for some time now that the signallers should know how to work efficiently.

It does seem that there was some loss of expertise when the move happened though. The signallers now either don’t want to deviate, or are discouraged from doing so.

 

[43066]

"Lots of drivers being trained" is something we've been hearing since FCC days. If they are numerically not replacing the numbers of drivers retiring or moving TOCs, then, at best, it's retaining the status quo.

Whatever the debate, there are, have been and remain insufficient numbers of drivers for the timetabled service

This is the key point. Recruit as many as you want but if the same number leave, you will stand still. Plus there are limits to training school capacity, how many trainers and DIs you have, so you can’t easily ramp up the numbers you recruit.

 

[Spartacus]

Are the staff at York overloaded with work so just unable to be across everything or is it a loss of familiarity and experience with a route to know the best way of resolving an issue on a real world basis rather than a headcode basis?

Maybe a little of both, but in general one of the troubles with ARS and it's variants is that if you don't want it to do something you've got to tell it not to do it, and take the train out of ARS to do that. While you're doing that you're just doing that, not doing something else. So if you choose to change how something's signalled you might also be choosing not to signal something that's not ARS controlled at all, and while you're taking it out of ARS you might well be stopping it and causing delay to it, and maybe something else anyway. Alright if it's quiet and you're on top, knowing what screw ups ARS is prone to in your area, but on a busy panel.... Then there's the trouble that on a busy area with lots of movements sometimes trying to get rid of a small delay on one train can end up causing a delay to something else that might not otherwise occur, and there's a definite move towards a policy of 'if it's on time, keep it on time'. I've seen it enough times where someone tries 'ironing delays out', and just ends up creating more wrinkles instead. But that's really get further from the crew issues.

 

[bramling]

Maybe a little of both, but in general one of the troubles with ARS and it's variants is that if you don't want it to do something you've got to tell it not to do it, and take the train out of ARS to do that. While you're doing that you're just doing that, not doing something else. So if you choose to change how something's signalled you might also be choosing not to signal something that's not ARS controlled at all, and while you're taking it out of ARS you might well be stopping it and causing delay to it, and maybe something else anyway. Alright if it's quiet and you're on top, knowing what screw ups ARS is prone to in your area, but on a busy panel.... Then there's the trouble that on a busy area with lots of movements sometimes trying to get rid of a small delay on one train can end up causing a delay to something else that might not otherwise occur, and there's a definite move towards a policy of 'if it's on time, keep it on time'. I've seen it enough times where someone tries 'ironing delays out', and just ends up creating more wrinkles instead. But that's really get further from the crew issues.

I think the “if on time keep it on time” idea doesn’t really work on the KX lines. It’s debatable whether it ever did, but the problem now is the Thameslink services tend to be prioritised, which then essentially means the rest of the GN service is very prone to picking up small delays, which then snowball as the pack of cards tumbles.

Surely the whole point of ironing out stuff is so that you create a few small wrinkles across the board, but don’t completely castigate one service. Or, to put it another way, put a couple of minutes on a handful of up LNER services, and in the process avoid a GN service arriving 20 mins late into King’s Cross, which is way more undesirable. The argument that a load of slightly delayed services will cause bigger problems elsewhere across the network doesn’t really stand up to scrutiny for up services all heading to King’s Cross, and with core services already prioritised. I realise in the down direction things are more complex.

Where this all perhaps has some relevance to lack of drivers, is it’s all part of a wider problem, which seems to be there hasn’t been enough focus on performance on the GN side, especially with the advent of Thameslink. Whether that’s insufficient drivers to reliably deliver the timetable, infrastructure failures, or general fragility of the timetable, the end result is a service that’s just not particularly reliable any more. Surprising no one here predicted this all those years ago?!

 

[Bikeman78]

You would expect better service recovery and control by GTR for something like that otherwise you run late all throughout the peak.

If the timetable needs daily intervention by Control, then it doesn't work. Going by what has been written above, the mid afternoon departures from King's Lynn would hardly ever run if they kept terminating short to protect the peak.

From a signalling point of view, I find there’s insufficient urgency in pushing up services through the Letchworth to Welwyn section, and once this happens it doesn’t take much for the service to then be stuck behind the 717 starting at Welwyn. Again, in the past use would have been made of the infrastructure to try and correct this (crossovers at Marshmoor, Potters Bar to use the up fast, or use of the up slow no.2 inwards from Alexandra Palace), but this simply doesn’t happen now - and this isn’t always a case of insufficient space on the fast lines, as it can happen at the same extremes of the day as well.

Back in class 317 days, I recall that late running trains would switch to the up fast at Marshmoor or Potters Bar to dodge the class 313 stoppers. There aren't that many more fast trains now. As you say, there are plenty of big gaps.

 

[Horizon22]

If the timetable needs daily intervention by Control, then it doesn't work. Going by what has been written above, the mid afternoon departures from King's Lynn would hardly ever run if they kept terminating short to protect the peak.

Oh of course, but at the same if interventions are required all the time, and the complaints start flooding in about trains being turned short / run fast and eventually that goes back to NR & TOC about the train plan.

Where this all perhaps has some relevance to lack of drivers, is it’s all part of a wider problem, which seems to be there hasn’t been enough focus on performance on the GN side, especially with the advent of Thameslink. Whether that’s insufficient drivers to reliably deliver the timetable, infrastructure failures, or general fragility of the timetable, the end result is a service that’s just not particularly reliable any more. Surprising no one here predicted this all those years ago?!

Great Northern certainly seems to have become the awkward cousin within GTR

 

[choochoochoo]

I hear many 'inner' drivers complain how they come from Welwyn to Alexandra palace get moved onto at platform 1 (up slow 2) at Wood Green north junction when platform 2 (up slow 1) is free and then get moved back right back on to up slow 1 right after so they can call at Hornsey next. Adding a few minutes on to their journey.

 

[whoosh]

It's a continuous process, but you can see on this very forum the depots that have been hiring and still are.

If it takes 16-17 months to become a driver, but an existing driver can leave with about a month notice (4 weeks?) then there are always going to be ups and downs

Great Northern and Thameslink are three months notice.

 

[MikeWM]

The two track section Hitchin to Cambridge compounds the issues too. Generally my observations commuting are that it's the timekeeping on the stopper which causes a lot of problems, in both directions, especially as the services interwork at King's Cross too.

I've suspected as much for a while, indeed I wrote back in 2018:

I've long thought it is rather crazy that the 'final' plan is 6tph on the two-track section betwewn Hitchin and Cambridge - two stoppers, two semi-fast and two fasts, when the time difference between a fast and a stopper on this section is almost 20 minutes (particularly in the up direction, where there is no crossover until after Hitchin), and there is nowhere to overtake. Once - if ever - all these trains are running, I think it is all too obvious this is going to be a major cause of delays. At the very least passing loops should have been put in (near Royston, probably) to alleviate problems.

At least the plan to put the stoppers through the Thameslink core didn't happen, or who knows how bad things would be.

The December recast should help by splitting up the stopping service at peak times - that thread across Welwyn slow to fast to slow doesn't work reliably enough. It's too common for the stopper to be 5 late at Stevenage, held for a fast at Knebworth to then fall behind a Moorgate stopper and lose more time.

I'm somewhat concerned the attitude is going to be 'the recast will sort this' and so we're in for a thoroughly miserable year until then.

--

I guess one quick win would be to make use of both platforms at Kings Lynn, which would essentially add half an hour to the turnround time there - albeit at a cost of two extra units having to be out on the railway, albeit without incurring any extra mileage. For this to be effective this also relies on driver turnarounds being similarly lengthy, though with Kings Lynn being a crew depot presumably there’s a fair few services which have a fresh driver take the train south.

I was rather thinking this too.

I'm also not sure that the turnaround times are the main issue, but the time at KLN does appear inadequate given the collection of major constraints on the route (Kings Cross throat, very busy south end of ECML, Welwyn Viaduct, the very busy Hitchin-Cambridge section with no passing loops, the approach to Cambridge from the south, Ely North junction, the very long single track section between Littleport and Downham Market, the shorter single track section after Watlington - plus splitting at Cambridge, and often changing driver there too).

A longer layover at KLN would at least make recovery more likely when things start to go wrong.

Where this all perhaps has some relevance to lack of drivers, is it’s all part of a wider problem, which seems to be there hasn’t been enough focus on performance on the GN side, especially with the advent of Thameslink. Whether that’s insufficient drivers to reliably deliver the timetable, infrastructure failures, or general fragility of the timetable, the end result is a service that’s just not particularly reliable any more. Surprising no one here predicted this all those years ago?!

Well, indeed so :-/ All that said, while I don't have figures to prove it, the weekday evening service seems to have markedly deteriorated (further) relatively recently, it would be good to be able to work out why, though it may just be a case of there are so many things that can go wrong, that some of them do. My initial thought, as above, was the new Letchworth terminators added in December, but looking at the data it seems that while they can cause delays, they don't appear to be making a major difference most evenings.

 

[jon0844]

When you have diagrams that are designed to use as few trains as possible, there's always going to be a high risk. Take a train from Sevenoaks having to get to WGC, run empty to King's Cross to form a Cambridge service that might then form a fast back to London and another service to Kings Lynn or whatever it does.

If there's any delay from Kent, it has a knock on effect for multiple trains into the evening some 80-odd miles away.

(I believe there's also a service from Cambridge that might form a Sevenoaks in the morning too, so a similar risk).

 

[Bald Rick]

Re the stopper regualrly causing delays beyond Hitchin - are any regular users able to advise what is the cause of this?

 

[bramling]

Re the stopper regualrly causing delays beyond Hitchin - are any regular users able to advise what is the cause of this?

My own observation is I don’t think it’s the stopping train that’s the root cause of the problem. I’d say the problem is the Cambridge/Brighton services, which in the up direction often seem to delay the stopping service (including those which nowadays start at Letchworth). The stopping service then loses time through the Hitchin to Welwyn section, and if it loses too much time can then end up stuck behind the 717 from Welwyn, invariably with no attempt being made to recover any time.

 

[Magdalia]

When you have diagrams that are designed to use as few trains as possible, there's always going to be a high risk. Take a train from Sevenoaks having to get to WGC, run empty to King's Cross to form a Cambridge service that might then form a fast back to London and another service to Kings Lynn or whatever it does.

If there's any delay from Kent, it has a knock on effect for multiple trains into the evening some 80-odd miles away.

(I believe there's also a service from Cambridge that might form a Sevenoaks in the morning too, so a similar risk).

The morning and evening risks are not the same.

In the morning the 0526 Cambridge-Kings Cross goes empty for 0802 Welwyn GC-Sevenoaks. It has no impact on the GN.

In the evening the 1652 Sevenoaks-Welwyn GC goes empty for 1927 Kings Cross-Cambridge. This has more potential impact because the same unit then does 2124 Cambridge-Kings Cross and 2327 Kings Cross-Cambridge.

These are all stoppers, there is no direct impact on fast trains.

 

[Bald Rick]

My own observation is I don’t think it’s the stopping train that’s the root cause of the problem. I’d say the problem is the Cambridge/Brighton services, which in the up direction often seem to delay the stopping service (including those which nowadays start at Letchworth). The stopping service then loses time through the Hitchin to Welwyn section, and if it loses too much time can then end up stuck behind the 717 from Welwyn, invariably with no attempt being made to recover any time.

Interesting. 30 9Sxx have run so far today from Cambridge to London (and beyond), 19 of them were on time at Hitchin, another 5 were 1 minute late, and 4 more 2 mins late. The others were late because they left Cambridge late. So it doesn’t seem there is anythign wrong with the running times or the timetable itself on that stretch.

 

[bramling]

Interesting. 30 9Sxx have run so far today from Cambridge to London (and beyond), 19 of them were on time at Hitchin, another 5 were 1 minute late, and 4 more 2 mins late. The others were late because they left Cambridge late. So it doesn’t seem there is anythign wrong with the running times or the timetable itself on that stretch.

I suspect the issue is Brighton trains not getting away from Cambridge on time. Either because they arrive late on the inward journey, or they aren’t getting away on time for some other reason.

With the two services being timed to leave Cambridge very close together it isn’t going to take much of a delay on the Brighton services to result in a late start for the stopping service. Same effect for the services starting at Letchworth - Brighton train a few minutes late, and it’s a late start out of the sidings.

The train starting from the sidings can also clash with down services if any of those are late (either the 9S or the 2C) - doesn’t take the down stopping service to be many minutes late before there’s a clash, and generally priority seems to be given to the down service.

 

[James90012]

Realtime Trains | 2C55 1756 Cambridge to London Kings Cross | 20/02/2025

Real-time train running information for 2C55 1756 departure from Cambridge to London Kings Cross on 20/02/2025. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

https://www.realtimetrains.co.uk/service/gb-nr:W23527/2025-02-20/detailed

Here is an example of what I notice, losing time without any obvious reason. I did spot check a few others today and found some were OK but found this sub threshold time loss on a few services.

Edit: I'm not saying this is the beginning and end of reliability on the GN but perhaps suggests something is off.

 

[Failed Unit]

Re the stopper regualrly causing delays beyond Hitchin - are any regular users able to advise what is the cause of this?

Welwyn Garden City is the big problem. Too many trains with little margin for error.

I dont know what is happening at Digswell. But stow trains waiting for fast has gone on for years. The service recovered.

in the peak at Welwyn Garden City you have.
0821 London Kings Cross (platform 2)
0823 Welwyn GC - Moorgate (platform 3)
0828 London - Cambridge (platform 3)
0832 Welwyn - Sevenoaks (platform 3) - coming from carriage sidings
0835 Welwyn - Moorgate (platform 4) - stood for a while

if the.0821 is just 2 minutes late which is common they dispatch the 0823 and hold it on the bridge in the best case. Send it off to London ahead of the 0821 on a bad day.
the 0828 is often delaying the 0832 as it has a slow exit out the sidings
by Finsbury Park the 0821, 0823 and 0832 are often all late because of lack of track capacity.

not sure what can be done but basically no room for error and a 2 minute delay to any service snowballs rapidly. Even at Finsbury Park the Horsham service gets stuck in a queue behind the Sevenoaks.

going north you often need to wait outside WGC for a platform.

Realtime Trains | 2C55 1756 Cambridge to London Kings Cross | 20/02/2025

Real-time train running information for 2C55 1756 departure from Cambridge to London Kings Cross on 20/02/2025. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

https://www.realtimetrains.co.uk/service/gb-nr:W23527/2025-02-20/detailed

Here is an example of what I notice, losing time without any obvious reason. I did spot check a few others today and found some were OK but found this sub threshold time loss on a few services.

Edit: I'm not saying this is the beginning and end of reliability on the GN but perhaps suggests something is off.

Perfect example to my post. In the rush hour it can be stuck behind the Moorgate and 17 down as a result, pushing back the turnaround at Kings Cross to head back to Cambridge fast. The snowball grows.

 

[OxtedL]

Realtime Trains | 2C55 1756 Cambridge to London Kings Cross | 20/02/2025

Real-time train running information for 2C55 1756 departure from Cambridge to London Kings Cross on 20/02/2025. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

https://www.realtimetrains.co.uk/service/gb-nr:W23527/2025-02-20/detailed

Here is an example of what I notice, losing time without any obvious reason. I did spot check a few others today and found some were OK but found this sub threshold time loss on a few services.

Edit: I'm not saying this is the beginning and end of reliability on the GN but perhaps suggests something is off.

Yep this is really typical - especially bad in the autumn but happens all year round. No obvious reason why they sometimes lose minutes and sometimes don't.
Right behind it at Letchworth will be a 1Txx ex-Cambridge fast which has to overtake before Woolmer Green Jn. Often this 1Txx has to wait before Hitchin for a 9Jxx Horsham to use the platform first and gets delayed further (although not in the example above). But inevitably the 2Cxx will then be held before the 2 track section to let the 1Txx overtake it to stay in booked order, passing on a minimum 5-6 minute delay. In rush hour this will then mean getting to WGC late enough that the signaller has had to let a Moorgate out first. There is no longer any attempt to get them back in front of the stoppers, so you just get to Kings Cross 15-20 minutes late. This is late enough to delay the next departure.

The 1721 and 1821 arrivals at Kings Cross regularly arrive 15 minutes late after following stoppers from WGC, and still have to attach to extra carriages before working the 12 car 1739 and 1839 Kings Lynn services.
The 1751 arrival also had this problem until the last timetable change - making platform 8 a pinch point for 3 consecutive trains all with terrible performance - but I think this now forms an 8 car Peterborough instead.

 

[bramling]

Yep this is really typical - especially bad in the autumn but happens all year round. No obvious reason why they sometimes lose minutes and sometimes don't.
Right behind it at Letchworth will be a 1Txx ex-Cambridge fast which has to overtake before Woolmer Green Jn. Often this 1Txx has to wait before Hitchin for a 9Jxx Horsham to use the platform first and gets delayed further (although not in the example above). But inevitably the 2Cxx will then be held before the 2 track section to let the 1Txx overtake it to stay in booked order, passing on a minimum 5-6 minute delay. In rush hour this will then mean getting to WGC late enough that the signaller has had to let a Moorgate out first. There is no longer any attempt to get them back in front of the stoppers, so you just get to Kings Cross 15-20 minutes late. This is late enough to delay the next departure.

One point to make. I agree with all of the above, however it isn't a case of the signaller *having* to let a Moorgate out first. It's quite possible to hold it on the flyover for some time and not getting in the way of anything. It's a *choice* to let the Moorgate run first, albeit one which the signallers may be under a policy decision to make. I don't think it's an acceptable decision given the implications it has - especially now using the up fast line allowing the 2C services to subsequently overtake essentially doesn't happen.

 

[Failed Unit]

One point to make. I agree with all of the above, however it isn't a case of the signaller *having* to let a Moorgate out first. It's quite possible to hold it on the flyover for some time and not getting in the way of anything. It's a *choice* to let the Moorgate run first, albeit one which the signallers may be under a policy decision to make. I don't think it's an acceptable decision given the implications it has - especially now using the up fast line allowing the 2C services to subsequently overtake essentially doesn't happen.

The dreaded Ally Pally stop. In the past they would have put the 2C onto the fast at Marshmoor.

Going north you often see the 2c held before the viaduct for the Ely / Kimgs Lynn to pass. Sensible regulation but that is back to the start of the snowball again. Ironically probably late going north because of the short turnaround at King’s Cross because it followed a Moorgate in.

The Welwyn - Moorgate often pick up small delays which are not clear why. leaves Welwyn on time but I assume it is on yellows as they are frequently 2-3 mins late by Finsbury Park. Insignificant of course, except when the turnaround at Moorgate isn’t that big.

The slows seem to be much more intensively used now. I guess the fasts are as well with 2tph extra on them with Thameslink.

 

[Horizon22]

Realtime Trains | 2C55 1756 Cambridge to London Kings Cross | 20/02/2025

Real-time train running information for 2C55 1756 departure from Cambridge to London Kings Cross on 20/02/2025. From Realtime Trains, an independent source of train running info for Great Britain.

www.realtimetrains.co.uk

https://www.realtimetrains.co.uk/service/gb-nr:W23527/2025-02-20/detailed

Here is an example of what I notice, losing time without any obvious reason. I did spot check a few others today and found some were OK but found this sub threshold time loss on a few services.

Edit: I'm not saying this is the beginning and end of reliability on the GN but perhaps suggests something is off.

Minimal dwell times for passenger loadings in peak? There’s almost no recovery time at all before Hatfield. Congested routes and a minute here or there and you lose a path and then minimal turnaround at terminus at it snowballs.

Whatever the root cause is, it seems like something performance teams on the route should be really on top of although it might be filed into ‘too hard to solve’ or ‘yeah fixed next timetable change’.

 

[Failed Unit]

I suspect it is in the fixed at the next timetable change. We shall see. Especially with the platform occupancy at Welwyn Garden City. They still look very bunched there.

Looking this morning at 2C07 0556 Cambridge- London. It has lost a few minutes along the way which is not clear why. It is now in the danger zone of getting stuck behind the Moorgate at Welwyn Garden City. This seems a standard performance of the stoppers. 2-3 late shouldn’t matter so much but it seems to really escalate.

Going the other way the 2C04 London - Cambridge is 1 whole minute down at Welwyn Garden City but both the 0702 Sevenoaks and 0705 Moorgate will probably be late starting.

You do often see trains needing to wait at WGC as a train needing the flyover on Platform 3 prevents an arrival from London on platform 4. Which one gets priority (normally the arrival)

 

[jon0844]

Going the other way the 2C04 London - Cambridge is 1 whole minute down at Welwyn Garden City but both the 0702 Sevenoaks and 0705 Moorgate will probably be late starting.

You do often see trains needing to wait at WGC as a train needing the flyover on Platform 3 prevents an arrival from London on platform 4. Which one gets priority (normally the arrival)

Yes, the train that forms the xx05 and xx35 Moorgate has to arrive into platform 4 before the xx53 and xx23 can leave platform 3. More often than not, that starts late as a result. If late enough, it delays the inbound Cambridge and that then delays the Sevenoaks departure and the xx05 and xx35 that arrived late.

As you say, it's very tight. Also the 0732 Sevenoaks starting from platform 1 isn't really helping anything as if it's late in or out of the reverser, the xx23 train still can't start from platform 3 on time even if the train from Moorgate arrived on time. Plus the Sevenoaks often gets into platform 1 and then there's around six minutes for changing ends and departing, so any train that did go is held on the flyover anyway.

 

[Bald Rick]

Yep this is really typical - especially bad in the autumn but happens all year round. No obvious reason why they sometimes lose minutes and sometimes don't.

Well, there’s only 5 reasons, and one of then can be discounted (running times) as it seems to work most of the time.

Minimal dwell times for passenger loadings in peak?

This is another of them. But it seems that it does work many times in the peak, and doesn’t on others.

So that leaves:

Unit low on power
Signals not being cleared by signaller (seems unlikely with progressive time loss)
Driver

 

[Failed Unit]

Well, there’s only 5 reasons, and one of then can be discounted (running times) as it seems to work most of the time.




This is another of them. But it seems that it does work many times in the peak, and doesn’t on others.

So that leaves:

Unit low on power
Signals not being cleared by signaller (seems unlikely with progressive time loss)
Driver

Which is something you often can’t see at the smaller stations. The dwell can be long, as a passenger you can’t see if other passengers are obstructing the driver’s vision to close the doors.

As a user I don’t see much improvement on the sections in the peak compared to the 313s. However if you accelerate hard do you just hit a yellow / red signal? I have always thought the trains were always running on Yellow.

As you say it works perfectly some days so I don’t think the timetable is flawed. Just very tight.

As you say, it's very tight. Also the 0732 Sevenoaks starting from platform 1 isn't really helping anything as if it's late in or out of the reverser, the xx23 train still can't start from platform 3 on time even if the train from Moorgate arrived on time. Plus the Sevenoaks often gets into platform 1 and then there's around six minutes for changing ends and departing, so any train that did go is held on the flyover anyway.

I use that service a lot and have never understood why it uses platform 1, unless it is to keep drivers knowledge up to date. If it arrives before the Cambridge to London everything is great (but as you say when it doesn’t it causes lots of delays)

Returning to topic, the later part of the week the cancellations due to driver shortage are much better, so for whatever reason it must have just being the co-incidence about the half term holiday.

 

[Magdalia]

Here is an example of what I notice, losing time without any obvious reason.

No obvious reason why they sometimes lose minutes and sometimes don't.

Two reasons are obvious to me as an experienced GN traveller, and neither of them are new.

Minimal dwell times for passenger loadings in peak? There’s almost no recovery time at all before Hatfield.

Look at the dwell times at the stations on the branch, that's how 3 minutes are lost by Hitchin. The 1756 from Cambridge will be a busy train for commuters going home from Cambridge, plus people heading to London for the evening, so there are lots of people both getting off and getting on. Having used these trains when they were crowded 4 cars losing 3 minutes by Hitchin is quite tame.

And the solution is quite simple, though most of you won't like it: these trains need to be class 700s to speed up dwell times.

Right behind it at Letchworth will be a 1Txx ex-Cambridge fast which has to overtake before Woolmer Green Jn. Often this 1Txx has to wait before Hitchin for a 9Jxx Horsham to use the platform first and gets delayed further (although not in the example above). But inevitably the 2Cxx will then be held before the 2 track section to let the 1Txx overtake it to stay in booked order, passing on a minimum 5-6 minute delay.

This issue has been around ever since the fast Cambridge trains were switched to the Kings Cross route more than 30 years ago, and it won't be going away. In ye olden dayes the last signal on the up slow before Woolmer Green was K632 (it has been renumbered now) and I reckon that I have waited at K632 for more time than any other signal on the whole network.

This particular example is a bit unusual in that a whole flight of up ECML trains were also 3-5 minutes late, but the GC up from Sunderland was on time and also let through the 2 track section before the stopper, slightly increasing the delay.

One thing that has changed is the Thameslink trains going through the core. The up fast now has more trains so there is less room than there used to be to get the up stoppers out onto the fast line at Marshmoor or Potters Bar. I remember times when an up stopper could get out onto the fast at Marshmoor then run Potters Bar-Finsbury Park, entirely on the fast line, in about 8 minutes, but that's long gone.

30 9Sxx have run so far today from Cambridge to London (and beyond), 19 of them were on time at Hitchin, another 5 were 1 minute late, and 4 more 2 mins late. The others were late because they left Cambridge late.

One thing I would look for is any variation in performance between the xx23 and xx53 departures from Cambridge, because the risks of late start are different.

The xx23s mainly use platform 1 at Cambridge, and usually follow Ely-Kings Cross and Cambridge North-Liverpool Street trains, neither of which come through Ely North Junction.

The xx53s mainly use platform 7 at Cambridge, and usually follow Kings Lynn-Kings Cross and Norwich-Stansted trains. both of which come through Ely North Junction.

For those who don't know Cambridge, the exit from platform 7 going south is both longer and slower than from platform 1.

 

[Bald Rick]

Looking this morning at 2C07 0556 Cambridge- London. It has lost a few minutes along the way which is not clear why.

Well it can‘t be passenger loadings, at that time, on a Friday, in half term. And it would have been on green signals. Down to two options!

 

[OxtedL]

The dreaded Ally Pally stop. In the past they would have put the 2C onto the fast at Marshmoor..

The fasts are now so busy that there often no good opportunity to put the 2C onto the fasts even if the signallers wanted to try.

 

[Horizon22]

Well it can‘t be passenger loadings, at that time, on a Friday, in half term. And it would have been on green signals. Down to two options!

Surely an infrastructure issue of some description may also explain it (outside of your other 5 options!)?

 

[Bald Rick]

Surely an infrastructure issue of some description may also explain it (outside of your other 5 options!)?

Not for progressive time loss that affects only some trains on the same line.