Changes needed to mitigate against the effects of Thameslink core failures

[johncrossley]

Moderator note: split from

Thameslink Core Major Power Failure 28/8/25

I understand why this is the most sensible service to take out first as its completely duplicated North of St Pancras by the 9Txx (Brighton-Bedford) it causes very little angst on that side. It's not completely duplicated - the 9Rxx trains don't stop at West Hampstead Thameslink, and I agree...

www.railforums.co.uk

It is definitely time to split the service. No service beyond Three Bridges yet again today.

 

[Gerald Fiennes]

Or rejig the crewing strategy, ie. code for employ more drivers and locate some of them at better places (Farringdon? FPK? ECR?).

Oh, and just get to the bottom of the power supply issues at Farringdon/Blackfriars (been going for c. a decade) and fix them?

It is better to sort the problems that cause problems than have better ways of mitigating the consequences of the problems (as no one ever said) but can recall being taught in the days of 'Organising for Quality'....

 

[johncrossley]

Given the severe disruption on a daily basis (the overnight services almost never run these days), it probably should be just Peterborough to Stevenage and Horsham to East Croydon for the time being.

 

[Tester]

Or rejig the crewing strategy, ie. code for employ more drivers and locate some of them at better places (Farringdon? FPK? ECR?). Oh, and just get to the bottom of the power supply issues at Farringdon/Blackfriars (been going for c. a decade) and fix them? It is better to sort the problems that cause problems than have better ways of mitigating the consequences of the problems (as no one ever said) but can recall being taught in the days of 'Organising for Quality'....

Ah yes, I remember it well - O for Q

(Don't say it too quickly !)

 

[zwk500]

ie. employ more drivers

That's an excellent joke in the current financial climate.

 

[Bald Rick]

It is definitely time to split the service. No service beyond Three Bridges yet again today.

How would splitting the service help mitigate a points failure at Preston Park? Saying there was ‘no service’ is something of an exaggeration don‘t you think?

Or rejig the crewing strategy, ie. code for employ more drivers and locate some of them at better places (Farringdon? FPK? ECR?). Oh, and just get to the bottom of the power supply issues at Farringdon/Blackfriars (been going for c. a decade) and fix them? It is better to sort the problems that cause problems than have better ways of mitigating the consequences of the problems (as no one ever said) but can recall being taught in the days of 'Organising for Quality'....

As I have written before, there has been a sudden and fundamental shift in some traincrew matters, and Thameslink / Great Northern have responded by recruiting additional drivers. However this will take 2 years to get the full benefit. Fortunately, it started well over a year ago.

Basing crew at Farringdon is not the answer.

And what poeer supply idsues at Farringdon / Blackfriars? The issues there are almost entirely signalling (and that is being sorted, very expensively).

 

[johncrossley]

How would splitting the service help mitigate a points failure at Preston Park? Saying there was ‘no service’ is something of an exaggeration don‘t you think?

Splitting the service in the way I suggested means fewer trains required.

A points failure shouldn't be happening anyway (see https://www.railforums.co.uk/threads/are-swiss-trains-genuinely-reliable.293543/)

There were no Thameslink trains from Horsham between 1727 and 1838. That's effectively 1.5 cancellations - in the evening peak!

You can't seriously expect people to wait for nearly an hour in the dark and let Southern trains pass without stopping.

This is not an isolated incident. The overnight trains are regularly cancelled as well.

 

[GraphitProjekt]

I didn't think power supply was a significant issue in the core. The original incident in this thread was a fault on a train that caused the third rail to trip. I hear a lot about broken down trains on the core causing huge problems

 

[Bald Rick]

A points failure shouldn't be happening anyway

Agreed, but they do - on Swiss railways too.

There were no Thameslink trains from Horsham between 1727 and 1838. That's effectively 1.5 cancellations - in the evening peak! You can't seriously expect people to wait for nearly an hour in the dark and let Southern trains pass without stopping. This is not an isolated incident.

No, I’d expect them to catch a Southern train and change where necessary (the 1750 and 1820 to Victoria both ran).

Incidentally it’s one cancellation.

 

[johncrossley]

Agreed, but they do - on Swiss railways too.

But very rarely, compared to UK railways.

No, I’d expect them to catch a Southern train and change where necessary (the 1750 and 1820 to Victoria both ran).

You can't catch them if they go through the station without stopping. There was no ticket acceptance on local buses advertised for Ifield and Littlehaven as far as I could tell.

 

[Bald Rick]

But very rarely, compared to UK railways.

Haveyou got some data about that?

You can't catch them if they go through the station without stopping

Theyboth stopped at Horsham

 

[yorkie]

Haveyou got some data about that?

Let's discuss Switzerland comparisons in the dedicated thread. (I've posted about my experiences in that thread and stand by them, but there is no need to go over it all again in other threads)

 

[johncrossley]

They both stopped at Horsham

What about Ifield and Littlehaven?

 

[Bald Rick]

What about Ifield and Littlehaven?

What about them? you said Horsham.

Neverthelss each of them average about 8 people per train. Littlehaven passengers could have caught the 1756 to Horsham and doubled back, or dare I say walked the mile or so to Horsham.

More difficult for Ifield, clearly, but nevertheless, the number of people affected would have been relatively few.

Let's discuss Switzerland comparisons in the dedicated thread.

Oops, yes, apols.

 

[johncrossley]

Littlehaven passengers could have caught the 1756 to Horsham and doubled back

Is that allowed, without being given explicit permission?

 

[yorkie]

Is that allowed, without being given explicit permission?

During disruption, ticket acceptance normally applies. Also there is a right to be re-routed if a delay is anticipated to be over 60 minutes.

What guidance was given? When posting about disruption, it's best to link to and quote from a source, as well as staying the details of the journey you are making.

 

[johncrossley]

During disruption, ticket acceptance normally applies. Also there is a right to be re-routed if a delay is anticipated to be over 60 minutes.

What guidance was given? When posting about disruption, it's best to link to and quote from a source, as well as staying the details of the journey you are making.

There was no guidance given as far as I can tell. Sometimes they allow you to use local buses but I didn't see such an announcement in this case. I didn't know, however, there is actually a right to rerouting. I don't see why the delay needs to be one hour, though.

 

[Mattplans]

Services to/from Rainham need to be given back to Southeastern. The service is too complex and there's too many variables and factors. Perhaps the services to Orpington could be restored instead. These are easier to recover during disruption and cause less conflicts in NKE Junction area

== Doublepost prevention - post automatically merged: 1 Nov 2025 ==

Or rejig the crewing strategy, ie. code for employ more drivers and locate some of them at better places (Farringdon? FPK? ECR?).

There is a driver depot at Blackfriars

== Doublepost prevention - post automatically merged: 1 Nov 2025 ==

Have we considered that the Thameslink route is simply too long for a single service?

Also, it's been brought up previously that Thameslink don't employ enough drivers for their services, and hope that some will want overtime. Is this standard in the railway industry?

 

[JonathanH]

Services to/from Rainham need to be given back to Southeastern.

How does that impact a passenger needing to get to stations between Crawley and Horsham. It is easy to single out this service as being out of place but the fact it runs doesn't of itself cause disruption.

Have we considered that the Thameslink route is simply too long for a single service?

What do you mean by this? What is your remedy?

 

[Minstral25]

How does that impact a passenger needing to get to stations between Crawley and Horsham. It is easy to single out this service as being out of place but the fact it runs doesn't of itself cause disruption.


What do you mean by this? What is your remedy

I think the service from Peterborough to Horsham is only too long as the crew and train bases are mostly at the extremes of the route, it would make sense if the crew and train depots were closer to London - like maybe Welwyn and Redhill. It would give space for trains to be turned early rather than being rushed past intermediate stations and that crew are likely to be nearer issues.

All 9J trains run from either Peterborough or Horsham, excluding one early train from Three Bridges depot - I think there is some swapping with late night services via Kings Cross to Hornsey. I feel that is a key part of what causes so many issues on the service. Sadly, it would probably never change due to the infrastructure and crew moves required.

 

[Bald Rick]

There is a driver depot at Blackfriars

No there isn’t, and hasn‘t been for 7 years.

 

[Horizon22]

Splitting the service in the way I suggested means fewer trains required.

That is in no way guaranteed, and not always true - it depends how the fleet diagrams are built, and how 'efficient' you want your turnarounds to be at the terminus. It is also not good for wider network capacity as you need 4 terminals with sufficient platform capacity, not 2.

== Doublepost prevention - post automatically merged: 1 Nov 2025 ==

Have we considered that the Thameslink route is simply too long for a single service?

Yes in great detail, multiple times, over multiple threads. A return to the "old ways" simply isn't going to happen though.

Also, it's been brought up previously that Thameslink don't employ enough drivers for their services, and hope that some will want overtime. Is this standard in the railway industry?

Fairly standard yes, and not just for drivers! It is changing a little though as previous fairly stable rates about the % of drivers / crew willing to do overtime have changed since Covid, and as @Bald Rick says they are hiring more drivers, but this will take some time to filter through into productive drivers.

 

[johncrossley]

but this will take some time to filter through into productive drivers.

So until then we need interim measures to avoid the extreme ends of the route having no service for long periods of time.

 

[Horizon22]

So until then we need interim measures to avoid the extreme ends of the route having no service for long periods of time.

I would hardly consider a complete timetable rewrite (which will take just as long, if not longer), an "interim measure"!

 

[johncrossley]

I would hardly consider a complete timetable rewrite (which will take just as long, if not longer), an "interim measure"!

So what are they supposed to do at Littlehaven and Ifield at the moment? The service is not reliable enough. Either run the service properly or cancel it altogether. Maybe run a shuttle between Three Bridges and Horsham every hour and have local buses accepting train tickets at all times. Or just have the Southern service calling at all stops on alternate trains.

 

[Horizon22]

So what are they supposed to do at Littlehaven and Ifield at the moment? The service is not reliable enough. Either run the service properly or cancel it altogether. Maybe run a shuttle between Three Bridges and Horsham every hour and have local buses accepting train tickets at all times.

Well on the matter of reliability, these are the averages between Blackfriars and Ifield. So the 1654/1754 (as the worst performer) is on average 50% of the time, 5 minutes late or better. Of course that means it is also 50% of times over 5 minutes late (or cancelled)

Going back a month (mid September), it was actually only cancelled twice with an average delay of 9 minutes. So obviously not great, but not awfully bad either.

If an official service disruption (i.e CSL2) has been declared then yes, bus ticket acceptance would often be agreed. A shuttle would be reasonable (using P1), during significant service disruption and a contingency plan, but you'd be ripping up all the driver diagrams. Might be slightly easier if crewed solely by Horsham drivers.

 

[bramling]

How would splitting the service help mitigate a points failure at Preston Park?

Massively. A points failure at Preston Park would no longer impact journeys between London and Cambridge, which is the case now - both directly in the form of the unreliable Cambridge-Brighton service, and indirectly in the form of that unreliable service delaying other, more reliable, GN-side services like King’s Cross-Cambridge and King’s Cross-Kings Lynn.

Saying there was ‘no service’ is something of an exaggeration don‘t you think

There seem to have been a lot of days recently where Cambridge-Brighton has been pulled at some point in the day. What with it reducing to hourly in the evening, it’s more of a bonus service if it actually runs.

Whatever the reasons, dependability on the GN side is abysmal at the moment. Out of nine journeys I’ve made this week, I can only think one one that arrived at my destination less than 10 late. I guess I should consider myself fortunate that I actually seemed to go through a week without a cancellation somewhere, which is actually highly unusual.

 

[Magdalia]

Around 10 years ago I wasted a lot of breath telling people how bad Thameslink would be, especially for GN passengers.

The start in 2018 was spectacularly bad, exceeding my worst fears, but longer term Thameslink has saved itself from my two biggest nightmares. One was never introducing the completely bonkers Cambridge-Maidstone service and the other was never getting to 24tph in the peak.

The title of this discussion distracts from the two big issues that remain: at 20tph the core is not the problem. Both big outstanding issues are not core related.

Have we considered that the Thameslink route is simply too long for a single service?

One is that some limbs of Thameslink are too long. Here time taken to get between City Thameslink and the extremities is a good metric. Time is more important than distance.

Lots of end points are 65-75 minutes from City Thameslink, including Bedford, Brighton, Sevenoaks, Horsham and Cambridge. This should be regarded as a sensible maximum for reasons of resilience.

Cambridge stopping trains would have been about 95 minutes to/from City Thameslink, Maidstone about 85 minutes. Of the existing services Rainham is 96 minutes and Peterborough is 87 minutes. Both of these are too long and need to be truncated or amputated.

Or rejig the crewing strategy,

The other is the strategy of crewing routes predominantly or entirely from the extremities, and with little or no knowledge of alternative routes. The Peterborough-Horsham service shows what a disaster this has been, with any issue on one side of the river inevitably leading to loss of service on both sides of the river.

I am only familiar with old style crewing on the GN. FCC's crew diagrams were a thing of beauty, with 5 depots and 5 unit types, but everyone signed everything except that Peterborough crew didn't do Cambridge-Kings Lynn (yes, Kings Lynn drivers had a turn with class 313s to Moorgate). Some of that sort of flexibility would give controllers more options for service recovery. Less "end to end" crewing would help too, though please no crew changes in the core.

 

[Mattplans]

Cambridge-Maidstone service

I have to agree with you. What a ridiculous idea.

the other was never getting to 24tph in the peak.

Why is this so bad? I'm of the opinion they should run extra trains via Bromley South to reach their promised frequency. Why would this be so disastrous?

 

[NCT]

So until then we need interim measures to avoid the extreme ends of the route having no service for long periods of time.

Or just 'grin and bear it'. It'll take more time to design and implement an interim measure than for the problem to pass.