-
Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
Experiences of service recovery (or lack of) on GTR Great Northern
- Thread starter Failed Unit
- Start date
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R
RailUK Forums
I saw the Stevenage MP announce this and wonder how LNER will share any revenue from those using contactless. Of course at that time of night, many people won't bother to buy a ticket anyway!
I think that the allocation of such revenue is generally based on survey data, but doubt that such use would be measured due to its rarity. Moreover, as both operators are now nationalised, I hope that nobody loses any sleep over a few wooden dollars for the sake of making the weary passenger happy! We might even increase overall industry revenue by doing so.
Certainly hope so. I expect ticketing will change massively if they introduce designated intercity, regional, metro and other such services.
For now, a couple of late night trains nobody will be checking isn't going to make much difference.
For now, a couple of late night trains nobody will be checking isn't going to make much difference.
Back from holiday - and the 0946 off Ely ran this morning, at least.
However, I note Littleport, Downham Market, Watlington *and* Waterbeach are now all in the bottom 10 stations *in the country* at ontimetrains, which is pretty incredible really.
However, I note Littleport, Downham Market, Watlington *and* Waterbeach are now all in the bottom 10 stations *in the country* at ontimetrains, which is pretty incredible really.
Magdalen Road
Member
My 2026 commute stats are now 25 hours of commuting delays and 59.62% of commutes delayed.
September 2026 must be one of the poorest performing months for years.
Well, that lasted a whole 1 day. 0946 off Ely cancelled yet again this morning, this time apparently due to a 'failure of the electricity supply earlier' which, oddly enough, didn't prevent the incoming service getting as far as Cambridge...
It's now available on the FLUA website at https://www.flua.org.uk/Fen line action plan Sep 2026.pdf so I think that clears up any questions about its legitimacy.
tatitiliti
Member
As usual, understating the severe impact of the crappy, unprepared procurement of the 379s and links to pantograph/OHLE damage…
OneOfThe48
Member
Yes - they should’ve said ‘the 379s fell apart before we could get them out the field they had been sat in for a few years and Network Rail’s continuing overhead wire problems are ruining our pantographs’
Do be serious
RailwayRookie
Member
Except the overhead wire problems only affected the 379 stock and nothing else that ran between Ely and Cambridge under the wires.
But if you're going to put cheaper carbons on the Pans then what else can you expect.
OneOfThe48
Member
Well that is what happens when the fault is between Ely and King’s Lynn
RailwayRookie
Member
Sorry let me clarify, by nothing else I meant other types of stock.
The near four days the Fen Line was suspended because of pan damage there were 3 other pans damaged South of Ely on 379 stock, identical damage.
It stands to reason that if the drivers damage their pans before Ely, the first they will see it going North is when they walk back down the train at Lynn to take it south again.
Also of note is that the various 720 runs to and from Lynn reported zero damage. (Although adding a different variable isnt the best way to nail down the source of a problem I'll admit).
Yesterday evening we nearly got a unicorn - 1T48 was on-time at Letchworth, and the stopper (1C80) and semifast TL (9S42) ahead were both on-time too.
But by Cambridge everything was between 6 and 9 minutes late. Why? Because the Barrington freight train *that reverses at Foxton in the middle of the evening peak* was late. I'm not sure I'd fully appreciated before that they try to fit a freight *reversal* in the middle of this really busy line, that already is the cause of many delays, right in the middle of the evening peak. No doubt this is brought to us by the same people who think it a good idea to push freight services through 13/14 at Manchester Piccadilly right in the middle of the evening peak.
As to why the freight was late, looks like that was due to waiting for 1T49 (1641 KLN-KGX) which passed Foxton 6 minutes late after being delayed entering the single section at Downham Market because 1T40 (the 1524 KGX-KLN) was late getting to Littleport. And *that* was late because it was stuck between Letchworth and Cambridge behind the late-running 9S34 (1344 BTN-CBG) which lost a few minutes between Gatwick and East Croydon, apparently because it was stuck behind 1B35 (the 1256 Bognor to VIC) which waited to depart from Gatwick Airport for a couple of minutes longer than timetabled, for whatever reason.
On other words, a Southern train from Bognor, dwelling at Gatwick Airport for a couple of minutes longer than timetabled, directly results in non-trivial delays for hundreds of commuters on three consecutive trains trying to get home to Cambridge/Ely/etc., over 4 hours later. You can see how easy it is for GN services to run into problems, and while a lot of issues are directly down to them, the total fragility of this timetable is ludicrous.
But by Cambridge everything was between 6 and 9 minutes late. Why? Because the Barrington freight train *that reverses at Foxton in the middle of the evening peak* was late. I'm not sure I'd fully appreciated before that they try to fit a freight *reversal* in the middle of this really busy line, that already is the cause of many delays, right in the middle of the evening peak. No doubt this is brought to us by the same people who think it a good idea to push freight services through 13/14 at Manchester Piccadilly right in the middle of the evening peak.
As to why the freight was late, looks like that was due to waiting for 1T49 (1641 KLN-KGX) which passed Foxton 6 minutes late after being delayed entering the single section at Downham Market because 1T40 (the 1524 KGX-KLN) was late getting to Littleport. And *that* was late because it was stuck between Letchworth and Cambridge behind the late-running 9S34 (1344 BTN-CBG) which lost a few minutes between Gatwick and East Croydon, apparently because it was stuck behind 1B35 (the 1256 Bognor to VIC) which waited to depart from Gatwick Airport for a couple of minutes longer than timetabled, for whatever reason.
On other words, a Southern train from Bognor, dwelling at Gatwick Airport for a couple of minutes longer than timetabled, directly results in non-trivial delays for hundreds of commuters on three consecutive trains trying to get home to Cambridge/Ely/etc., over 4 hours later. You can see how easy it is for GN services to run into problems, and while a lot of issues are directly down to them, the total fragility of this timetable is ludicrous.
OneOfThe48
Member
Well, they are all on the same line served by the same services so you would expect them to be performing about the same?
bramling
Veteran Member
I definitely think it worked better when the stopping services ran in their 2018 paths. Is there any reason why they can’t? It would still be possible to run off one platform at King’s Cross, just require a couple of minutes to be added to the up service to allow time for the down service to leave.
A return trip from Cambridge to Bournemouth at the weekend provided two perfect examples of how the King's Lynn trains are disadvantaged at every turn.
Last Friday the up 1T29 (already running 17 mins late) was within sight of P8 at CBG when the 9S33 to Brighton was let out just in front of it. No chance of making up any time after that and it was 21 down at Hitchin. Driver was obvious not best pleased and made a heartfelt Delay Repay announcement on arrival at KGX 20 late. 9S33 in contrast was nowhere more than 2 late (and arrived on time) so at least that did not suffer.
On return journey on Monday 1T34 was on time as far as Letchworth until it caught up with the late running 9S28, and then had to crawl along arriving CBG 9 mins down. Further delays (dried-out peat?) meant it eventually arrived KLN 17 late which will no doubt have had further knock-on effects.
The timetabled headway between 9S28 and 1T34 arriving in Cambridge is only 4 mins which does not make any realistic allowance for a train coming all the way from Brighton to be the slightest late. What were they smoking when this crackpot timetable was devised?
bramling
Veteran Member
On the latter point, this was discussed in considerable volume here before 2018, with the prediction made by many of us that the combination of Thameslink through services, Welwyn, Hitchin-Cambridge variation of stopping patterns, and the single-line sections north of Ely, would produce exactly this outcome. And here we are.
We were even told that the GN network is “simple”, and “like the Northern Line”. Well here we are, perhaps the individual concerned would like to come back and tell us all how he thinks it has all panned out.
If it’s all so “simple”, how is it stations like Littleport and Downham Market now have some of the worst performance in the country?
I’m sure he would also be welcome to offer his thoughts to the various rail user groups and associated social media groups along the length of the GN route, who are in various stages of despair at the dependability of the service, just in case anyone feels that the views here are not typical of regular users.
Failed Unit
Established Member
Poor EMR. On the list because XC don’t manage any stations.
Of course in the peak it is even worse, as not only does every KLN train have an ex-Brighton 4 minutes ahead of it, but that has a stopping KGX-CBG service ending up just 4 minutes ahead of that by Cambridge South! The train coming all the way from Brighton has to slot in precisely between these two, and if either that or the stopper are late, the KLN gets delayed.
Look at 1T48 tonight for example - leaves on time but then gets stuck behind the 4-late 9S42, in this case that eventually itself gets stuck behind the 7-late stopper 1C80. 1C80 then actually lost a couple more minutes at Shepreth due to the same late freight train reversing at Foxton (!) as yesterday (again that being late because 1T49 was a bit late, but at least tonight that wasn't due to something happening hours earlier at Gatwick...).
Result : 1T48 10 late at Cambridge, yet again.
We could see this coming as soon as the timetable was published, eg. here's myself in June 2025 earlier in this thread:
and a number of others made similar points, so it is truly baffling that the people who actually do this for a living thought this could possibly work.
--
Well, indeed so. Add this disastrous 2025 timetable - at least the 2018 timetable spaced out the down services between Hitchin and Cambridge about as well as possible given the constraints, not something that can be said now - and mix in the continuing issues GN have with the 379s, and it's a right mess.
I think the data speaks for itself. The ontimetrains rankings show this is probably currently the worst-performing line in the entire country, so the despair and the regular complaints we make here seem entirely justified.
I really think the only way to fix the problem is have Kings Lynn trains shuttling to/from Cambridge and then try and make it easy to connect to other services south - including the original Cambridge Cruiser, the semi-fast train to Brighton, or the stopper to KGX.
No splits and joins, and no drivers out of position because of issues miles away. Get the reliable stock to run, with a backup unit or two, a spare driver or two, and then see how that fares.
I get that people may not like the need to change train, but surely that would be less inconvenient than the current setup which clearly can't work when there's so much to go wrong even before it heads north of Cambridge.
If there are speed restrictions due to the quality of the trackbed, at least the issues are contained to a degree. Things are more easy to predict.
No splits and joins, and no drivers out of position because of issues miles away. Get the reliable stock to run, with a backup unit or two, a spare driver or two, and then see how that fares.
I get that people may not like the need to change train, but surely that would be less inconvenient than the current setup which clearly can't work when there's so much to go wrong even before it heads north of Cambridge.
If there are speed restrictions due to the quality of the trackbed, at least the issues are contained to a degree. Things are more easy to predict.
A shuttle might work for ppl going to London who might not care exactly which onward connection they get at Cambridge with 6 an hour. But not for ppl going further afield, they would surely want a journey with a more or less predictable arrival time. And ideally a non-stop run after Cambridge South.
In the down direction the shuttles cannot wait at CBG for trains from London, as the whole point is to maintain the regularity of the service to KLN over the single track sections. This won't be any use if there is no-one on them bc the connection has been missed, and not popular if it involves a 29 min wait.
If every half hour there are three down trains that are timetabled to arrive within 8 mins the corollary is that ATM there is a 22 min gap without any - from KGX/STP at least. Surely someone amongst "the people who actually do this for a living" can work out how to space them out a bit better even given all the constraints about crews and layover times. Or some AI tool.
bramling
Veteran Member
I’m not sure what would really be a good outcome, though you might be right it’s the only way of making the current service work reliably.
However I’d absolutely be ditching the through Thameslink service before that, since it is essentially that which is the root cause of all the trouble.
Really LNER should be able to run their service using fewer platforms at King’s Cross than they do. Doesn’t help that someone thought it was a good idea to have five different train types.
I wonder if it could theoretically be possible to run the KX service through to Ely, supplemented by an hourly Cambridge to Kings Lynn shuttle in the way you suggest. Would be a load of issues to work through, like where does the shuttle terminate, and can Ely actually turn round 2tph (bearing in mind they won’t be arriving on time). The only reason for suggesting this is it avoids the single-line sections.
Magdalen Road
Member
One of the reasons for delays is the relief driver not being available at CBG for the changeover. As this happens in both KGX and KLN directions, a shuttle service wouldn’t resolve the driver issue.
As a passenger, I suspect that a shuttle service might introduce more issues, not least a longer journey if you have to swap trains at CBG.
As a passenger, I suspect that a shuttle service might introduce more issues, not least a longer journey if you have to swap trains at CBG.
Last edited:
Yesterday, it felt like essentially every train since the morning was delayed. The 0901 from CMB got in at 0913; the 2248 from KGX arrived at 2255 thereabouts and only departed just before 2300, and called at neither Letchworth nor Royston.
I reckon there are many things GTR can do to keep the reliability up at least for Cambridge passengers, who are disproportionately affected by the abysmal reliability of the Kings Lynn trains.
GTR need to recognise delays are inevitable, and work within that using extra train capacity, maybe a changeover at Ely, better signalling, and dwell time. Thameslink is particularly susceptible, but I don't think getting rid of it is the right move. It provides a hugely useful alternate service especially late at night when the fast trains stop running, a direct service to Gatwick for Cambridge commuters, and many more.
Firstly, they're running incoming trains as the next ones immediately. For such a high-frequency line this is mental, since delays can and do accumulate extremely easily. They really ought to have spare capacity which can take off at scheduled times at either terminus and at key intermediate stations—Letchworth sidings, Cambridge, Ely.
Next, it looks like trains tend to depart Kings Lynn on time, but are delayed at the single-line section as they conflict with delayed down trains. I'd say hold the latter at Watlington until the on-time up train passes. Combined with the above, it would mean this one train that's delayed won't knock on to others.
This signalling applies to other trains as well—if a delayed fast train will be stuck behind an all-stations stopping or semi-fast then they should hold the stopper to let the delayed fast train depart first, even if it means the stopping train would be delayed by a few minutes—naturally this is at Cambridge
Then, the Cambridge fast needs to terminate either at Cambridge central, or Ely at the latest, and Kings Lynn travellers can wait at the platform—it wouldn't be an exceptionally long wait because the shuttle would keep up its 30-minute frequency, and delays far along the line wouldn't affect it. In fact I believe this shuttle would probably end up becoming more reliable, even with the track settlement issues.
Finally, the biggest step would be to bin the unreliable and old (and do I dare say it—absolutely filthy) 379s and replace them with Aventras or single-mode FLIRTs.
I reckon there are many things GTR can do to keep the reliability up at least for Cambridge passengers, who are disproportionately affected by the abysmal reliability of the Kings Lynn trains.
GTR need to recognise delays are inevitable, and work within that using extra train capacity, maybe a changeover at Ely, better signalling, and dwell time. Thameslink is particularly susceptible, but I don't think getting rid of it is the right move. It provides a hugely useful alternate service especially late at night when the fast trains stop running, a direct service to Gatwick for Cambridge commuters, and many more.
Firstly, they're running incoming trains as the next ones immediately. For such a high-frequency line this is mental, since delays can and do accumulate extremely easily. They really ought to have spare capacity which can take off at scheduled times at either terminus and at key intermediate stations—Letchworth sidings, Cambridge, Ely.
Next, it looks like trains tend to depart Kings Lynn on time, but are delayed at the single-line section as they conflict with delayed down trains. I'd say hold the latter at Watlington until the on-time up train passes. Combined with the above, it would mean this one train that's delayed won't knock on to others.
This signalling applies to other trains as well—if a delayed fast train will be stuck behind an all-stations stopping or semi-fast then they should hold the stopper to let the delayed fast train depart first, even if it means the stopping train would be delayed by a few minutes—naturally this is at Cambridge
Then, the Cambridge fast needs to terminate either at Cambridge central, or Ely at the latest, and Kings Lynn travellers can wait at the platform—it wouldn't be an exceptionally long wait because the shuttle would keep up its 30-minute frequency, and delays far along the line wouldn't affect it. In fact I believe this shuttle would probably end up becoming more reliable, even with the track settlement issues.
Finally, the biggest step would be to bin the unreliable and old (and do I dare say it—absolutely filthy) 379s and replace them with Aventras or single-mode FLIRTs.
Last edited:
RailwayRookie
Member
The issue with regulating the single lines north of Ely is that it is always a lose/lose situation.
It is not as simple as regulating late running trains over junctions to follow eachother.
The signallers at Kings Lynn and Magdalen Rd (Watlington) have to also consider the impact a decision would have on regulating the single line between Downham and Littleport.
While the 'preferred' state of regulating is to prioritise the UP road as it has further to travel, in reality there is far more to consider. Such as the Down road you hold for it is then late on its return Up journey and you do the whole dance agaim.
Similarly, with the littleport/Downham section being timetabled at 11 minutes. Is it better to let a 5 minute late Down train run its booked path and put 3 minutes into an Up road. Or hold it at littleport and make it over 15 late and risk its return journey and backing it into the on time service behind it?
Let's also not forget than some times the best intentions of the signallers are also overruled by control too.
That's where that extra capacity comes in. If you have at least one train waiting at the terminus or nearby sidings that can be pressed into service to cover for a delay, then the problem goes away; the late Down train can then come in when it does, and either work the next service, or replace the sidings train that just left.
One question I have is:
Where are the relief drivers? If a driver or conductor goes missing in my area then there is always a spare to take over. GN never seem to have this on a daily basis.
Granted, it is a different service type, but I don't think I've ever seen a Liverpool Street to Norwich train be cancelled due to lack of train crew.
Where are the relief drivers? If a driver or conductor goes missing in my area then there is always a spare to take over. GN never seem to have this on a daily basis.
Granted, it is a different service type, but I don't think I've ever seen a Liverpool Street to Norwich train be cancelled due to lack of train crew.
It worked much better before this setup started in the December 2025 timetable. Granted there were a whole host of other issues that were complained about in this thread and others, but this change has added an extra level of 'worse' from before.
If we look at the May 2025 timetable:
- The fast leaves Letchworth around x07 and gets to Cambridge around x31
- The semi-fast leaves Letchworth around x11 and gets to Cambridge around x41
- The stopper leaves Letchworth around x20 and gets to Cambridge around x53
That's about as good as you're going to get I think, and while not ideal, at least there we have a decent spread and a delay of a few minutes to any of those doesn't delay the one behind.
I think the main problem with this setup is that - if everything is on time - it requires the fast to overtake both the semi-fast and the stopper before Hitchin, and also for the semi-fast to overtake the stopper before Stevenage. So this way you have a bit of a delicate ballet going on in this congested part of the ECML and potentially getting in the way of longer-distance trains - you can see why, on paper at least, it might be desireable to untangle that. The problem is that the current timetable which removes this overtaking south of Hitchin (where at least you mostly have a 4-track railway and the ability to reorder trains) results in this utter mess once you get past Hitchin (which has no mechanism for reordering).