Merseyrail semi-fast running antics.

[bluenoxid]

Am I correct in thinking that Delay Repay only kicks in at 30 minutes on Merseyrail. Bit underwhelming to be hit with a 29 minute delay and no financial recognition of that if the following train runs well.

 

[Kite159]

Am I correct in thinking that Delay Repay only kicks in at 30 minutes on Merseyrail. Bit underwhelming to be hit with a 29 minute delay and no financial recognition of that if the following train runs well.

And isn't merseyrail still on an older style delay scheme which only pays out for certain reasons under an hour?

 

[TheSel]

And that last statement is incorrect. Boxing Day services run semi-fast on all their services, including West Kirby line.

Only because most of the stations were shut!

 

[BenW390Fan]

I have noticed more recently that Merseyrail will send their trains non stop from Hooton - Birkenhead Central if 5 mins late, which would make more sense if they just ran it fast to Rock Ferry instead of Birkenhead Central - otherwise it leads to stuff like this:



(Screenshot shows a Chester - Liverpool service running non stop from Hooton - Birkenhead Central, arriving at Birkenhead Central nearly 5 minutes early, and the stop at Bromborough is an RTT error)

 

[TheSel]

I have noticed more recently that Merseyrail will send their trains non stop from Hooton - Birkenhead Central if 5 mins late, which would make more sense if they just ran it fast to Rock Ferry instead of Birkenhead Central - otherwise it leads to stuff like this:

View attachment 183259

(Screenshot shows a Chester - Liverpool service running non stop from Hooton - Birkenhead Central, arriving at Birkenhead Central nearly 5 minutes early, and the stop at Bromborough is an RTT error)

Pedant alert: Not non-stop. Called at Bromborough.

 

[185]

Another one ending up 5 early. No-one in our railway management sees a problem with this.

Colleague's comment the other day was... "in thirty years, I have never known our railway to have become so stupid".

 

[Edgeley]

Last week I planned a trip to Hooton allowing myself two hours* to walk along the Wirral Way to Hadlow Road preserved station, have a sandwich lunch break and then walk back [*to connect with a fast train back at Lime St]. Do-able if things ran reasonably close to schedule.

My train into Lime Street was 10 minutes late, but on getting down to the low level platform a Chester service was just coming in, also running about ten late.

At Birkenhead Central it was announced that the service would run fast to Hooton - where we arrived about 1-2 minutes earlier than my planned arrival time. Just the job when you have a tight schedule.

 

[jamesst]

Pedant alert: Not non-stop. Called at Bromborough.

It wouldn't of stopped at Bromborough, that seems to be a mistake on real train times on some semi fast services

 

[Northerngirl]

The other good one the do is on the Ormskirk line, when there is disruption they turn back at Maghull, but how the timing works, the following service ends up held up for 5 minutes at the points, and then will run empty from Ormskirk to Kirkdale, usually after letting everyone on and then kicking them off

 

[jamesst]

The other good one the do is on the Ormskirk line, when there is disruption they turn back at Maghull, but how the timing works, the following service ends up held up for 5 minutes at the points, and then will run empty from Ormskirk to Kirkdale, usually after letting everyone on and then kicking them off

Rarely happens unless its particularly severe disruption, Ormskirks- particularly with currently running through to Hunts Cross- are a nightmare to get back on schedule if theres been issues

 

[Northerngirl]

Rarely happens unless its particularly severe disruption, Ormskirks- particularly with currently running through to Hunts Cross- are a nightmare to get back on schedule if theres been issues

I must be really unlucky then, had it happen to me probably about every 6 weeks this year

 

[TUC]

Surely with this train, it makes sense to go non stop from Liverpool to Hooton, then all stops to Chester, as passengers for stations between Liverpool and Hooton can get the Ellesmere Port train, whereas passengers for Capenhurst, Bache, and Chester have no alternative other than waiting the half hour.

All well and good if that intent is announced at the start of the journey (where the train has arrived late on its inward journey). It's skipping stops when the journey has started where problems arise.

 

[Vexed]

and the stop at Bromborough is an RTT error)

Pedant alert: Not non-stop. Called at Bromborough

It wouldn't of stopped at Bromborough, that seems to be a mistake on real train times on some semi fast services

Indeed, Darwin data shows it didn't stop



Because of data limitations, RTT has to guess when trains run though non-stop so it can sometimes miss skipped stops if train go through them slowly, or the underlying timing data used in the model is wrong.

 

[BenW390Fan]

I understand the reasoning behind the semi fast running but what confused me is why they don't run it fast from HOO - RFY like they do on the outward journey, from RFY - HOO. Here's another one that happened yesterday - funnily enough the same service!



(Screenshot shows a Chester - Liverpool service running non stop from Hooton - Birkenhead Central, arriving at Birkenhead Central nearly 5 minutes early, and the stop at Bromborough is an RTT errror)

 

[jamesst]

I understand the reasoning behind the semi fast running but what confused me is why they don't run it fast from HOO - RFY like they do on the outward journey, from RFY - HOO. Here's another one that happened yesterday - funnily enough the same service!

View attachment 183303

(Screenshot shows a Chester - Liverpool service running non stop from Hooton - Birkenhead Central, arriving at Birkenhead Central nearly 5 minutes early, and the stop at Bromborough is an RTT errror)

It comes down to when the decision was made and the delay at the time.

Unfortunately once the decisions been made and the train has been sent semi fast if it then makes up time quicker than expected there are very limited options to get in touch with the driver to resume all stations meaning the train will go to the original station and stand time there.

Basically the Chester line timetable is not fit for purpose and badly needs a recast, a 4/5 min turnaround with no recovery time is asking for trouble.

 

[BenW390Fan]

It comes down to when the decision was made and the delay at the time.

Unfortunately once the decisions been made and the train has been sent semi fast if it then makes up time quicker than expected there are very limited options to get in touch with the driver to resume all stations meaning the train will go to the original station and stand time there.

Basically the Chester line timetable is not fit for purpose and badly needs a recast, a 4/5 min turnaround with no recovery time is asking for trouble.

I appreciate the reply, but I still don't understand why with the starting delay at Chester being 5 minutes, even if makes up no time, it doesn't add up with the delay threshold of about 7 minutes for running fast between RFY and HOO on the outward journey. They've done it that many times now they surely must have a general idea of how much time a service can gain back when running fast between 2 stations. Sorry if I've over complicated my question, but basically why can't they just send the inbound journey fast from HOO - RFY (like the outward journey) instead of HOO - BKC? I recently noticed there's a TSR outside Rock Ferry going towards Liverpool so maybe that has something to do with it.

EDIT: I was silly and misread your reply, my apologies. I guess it makes sense if they decided to run it fast before the train had even turned around at Chester.

 

[jamesst]

I appreciate the reply, but I still don't understand why with the starting delay at Chester being 5 minutes, even if makes up no time, it doesn't add up with the delay threshold of about 7 minutes for running fast between RFY and HOO on the outward journey. They've done it that many times now they surely must have a general idea of how much time a service can gain back when running fast between 2 stations. Sorry if I've over complicated my question, but basically why can't they just send the inbound journey fast from HOO - RFY (like the outward journey) instead of HOO - BKC? I recently noticed there's a TSR outside Rock Ferry going towards Liverpool so maybe that has something to do with it.

EDIT: I was silly and misread your reply, my apologies. I guess it makes sense if they decided to run it fast before the train had even turned around at Chester.

No apologies needed!

The semi fast patterns in the event of disruption towards Liverpool can be Hooton- Rock Ferry/ Hooton- Bhead Central/ Hooton- Moorfields or if things are really screwed up Chester- Moorfields.
Towards Chester itll be any of Lpool Central- Hooton/ Lpool Central to Chester/ Hamilton Square or Birkenhead Central to either Hooton or Chester or Rock Ferry to Hooton or Chester.

This of course depends on a wide range of factors and the ultimate aim is to tell traincrew as soon as possible.
Like I said the issue is getting in touch with traincrew once they have started a semi fast.
Hope all of that makes sense.

 

[BenW390Fan]

No apologies needed!

The semi fast patterns in the event of disruption towards Liverpool can be Hooton- Rock Ferry/ Hooton- Bhead Central/ Hooton- Moorfields or if things are really screwed up Chester- Moorfields.
Towards Chester itll be any of Lpool Central- Hooton/ Lpool Central to Chester/ Hamilton Square or Birkenhead Central to either Hooton or Chester or Rock Ferry to Hooton or Chester.

This of course depends on a wide range of factors and the ultimate aim is to tell traincrew as soon as possible.
Like I said the issue is getting in touch with traincrew once they have started a semi fast.
Hope all of that makes sense.

Ah I see, thanks! Maybe they just sent it fast all the way to BKC just incase they didn't turn around as quick or it lost more time, like you say. But yes I agree the Chester timetable needs more recovery time, but as has been said the lack of space on the network is showing and limits service

 

[driverd]

Lengthening the turnaround times would result in less capacity. Presumably you'd be the first person to complain about that if they did this?

I think thats a tad unfair/an uncharitable interperitation of the OPs sentiment. Looking at the timetable, adding a single additional unit diagram (and associated crew diagrams) and interworking the Ellesmere Port services, with either leg picking up the others service at Hooton (eg: the xx:00 off Chester becomes the xx:15, meanwhile the xx:19 ex Ellesmere becomes the ~xx:34 (with departure adjusted for reoccupancy by the xx:37 arrival or additional pathing time added as required) etc), we have a pretty simple solution that adds a heck of a lot of resilience.

Is there any reason for not doing this that I've missed? Surely 1 additional unit on a Sunday won't break the network, assuming this is the only constraint?

While the leaf-fall timetable is better, the big difficulty with the Wirral Line is that you can only have layover at one end, and the big difficulty with Chester is that you can only have a maximum of about 13 minutes' layover because there's only one platform used for the service.

Where do you get the 13 minute figure from sorry? The usual situation would be circa 27 minutes maximum for a terminus with a single platform.

If a Chester service comes out the loop 7+ mins late it'll generally be sent semi fast Birkenhead Central to Hooton.
Once you get to around the 12/13min late mark it'll generally be either Liverpool Central to Hooton or Birkenhead Central to Chester.
20mins + could be Liverpool Central to Chester direct or it'll be turned back at Hooton and 30+ Rock Ferry.

This all depending of course for the reason behind the delay and potential knock on effects on other services, there's a lot to take into account.

Sounds like a solid plan if dealing with late running isolated to a single unit. Of course, if the whole network is delayed (eg: a single signal failure adding, say, 10 mins to each service), skipping stops on every service unfairly impacts users of a certain handful of stations. Under such circumstances, surely you're entering the territory of just allowing late running to occur until such a point as either a. Trains end up 30 mins late and you cancel 1 circuit and step trains up or b. the problem gets fixed and at that point you skip stops as a 1 off event to recover the service?

It's an interesting discussion really, Merseyrail is one of these suburban networks that borders on a metro system, but on Sundays, operates more to a conventional heavy rail frequency, making it somewhat of a hybrid.

 

[Starmill]

I think thats a tad unfair/an uncharitable interperitation of the OPs sentiment. Looking at the timetable, adding a single additional unit diagram (and associated crew diagrams) and interworking the Ellesmere Port services (eg: the xx:00 off Chester becomes the xx:15, meanwhile the xx:19 ex Ellesmere becomes the ~xx:34 (with departure adjusted for reoccupancy by the xx:37 arrival or additional pathing time added as required), we have a pretty simple solution that adds a heck of a lot of resilience.

Is there any reason for not doing this that I've missed? Surely 1 additional unit on a Sunday won't break the network?

It wouldn't have helped in their situation which was a 14 minute late departure. That would still be breaking the turnaround time and causing conflicting moves putting delays into other trains. Several extra crew diagrams and an additional unit have been justified at times in the past, if I remember rightly leaf fall Sundays had them? But at the end of the day more units to run the same service brings a definite and large cost, and a small and nebulous performance benefit.

 

[BenW390Fan]

It's all well and good putting out extra diagrams but with a lot of the terminuses having only one platform there really isn't that much space for extra trains

 

[driverd]

It wouldn't have helped in their situation which was a 14 minute late departure. That would still be breaking the turnaround time and causing conflicting moves putting delays into other trains. Several extra crew diagrams and an additional unit have been justified at times in the past, if I remember rightly leaf fall Sundays had them? But at the end of the day more units to run the same service brings a definite and large cost, and a small and nebulous performance benefit.

Sure, but the turn around at Chester with interworking would be 19 minutes, so assuming the delay didnt get any worse, this is the sort of resilience you'd gain, where by up to 15 minutes of delay can be absorbed at Chester and nearer 20 at Ellesmere Port. I wouldn't call that a small/nebulous improvement, given 5 minute delays result in cancellations as things stand.

Cost wise, a quick bit of rough maths suggests the unit would be out for around 17 hours, with a circa 40 mile trip performed 11 times, and very approx EMU running cost of £5/mile. Staff wise, drivers at approx £50/hour and guards at £40/hour (pension/NI taken into account), you're looking at £2200 for the unit, £900 for drivers and £720 for your TM (Plus potential additional lease costs if charged on mileage basis). To add significant resilience, £4000 per Sunday doesn't seem horrendous - especially if a more resilient service equates to more footfall.

There's also less quantifiable benefits from longer turn arounds, such as less fatiguing diagrams for crew and increased ability to use facilities, which helps improve morale and reduces incident risk.

It's all well and good putting out extra diagrams but with a lot of the terminuses having only one platform there really isn't that much space for extra trains

Again, I'm making this observation as someone who was once a train planner. From what I can see, as explained above, a single extra unit diagram and interworking the Chester/Ellesmere Port Sunday service would work at both termini. I'm not overly familiar with the network however, so if I'm missing something, I'll happily stand corrected.

 

[Starmill]

Sure, but the turn around at Chester with interworking would be 19 minutes, so assuming the delay didnt get any worse, this is the sort of resilience you'd gain, where by up to 15 minutes of delay can be absorbed at Chester and nearer 20 at Ellesmere Port. I wouldn't call that a small/nebulous improvement, given 5 minute delays result in cancellations as things stand.

Sure but realistically I don't think Merseyrail control would allow a train to run 14 minutes late from one end or the other, even with a 19 minute turnaround, because it's so disruptive. That might be your and my preference but it's very unlikely.

== Doublepost prevention - post automatically merged: 9 Jul 2025 ==

Cost wise, a quick bit of rough maths suggests the unit would be out for around 17 hours, with a circa 40 mile trip performed 11 times, and very approx EMU running cost of £5/mile. Staff wise, drivers at approx £50/hour and guards at £40/hour (pension/NI taken into account), you're looking at £2200 for the unit, £900 for drivers and £720 for your TM (Plus potential additional lease costs if charged on mileage basis). To add significant resilience, £4000 per Sunday doesn't seem horrendous - especially if a more resilient service equates to more footfall.

Indeed the additional cost could be justified, which is strongly hinted at by an extra diagram having run some years during leaf fall timetables. But if you did it year-round you would have to reorganise part of the maintenance schedule to do slightly less on weekends, which would involve fitting it in on weekdays instead. It is quite unlikely that a weekday diagram would be withdraw to allow a unit to layover for so long or Sundays.

 

[BenW390Fan]

An extra unit on the Chester line has been done but it did occasionally lead to a wait outside Chester station waiting for the platform. IIRC Wirral line services did used to interwork but when disruption occurred instead of delaying one line it delayed numerous. Personally I think there just needs to be more allowances on the Chester services (easier said than done) instead of having a 30 second scheduled station stop on most stations, increase it to a minute at more than just BKC. You can see with the service below how easily trains can lose time and with near on zero recovery time can easily lead to stations being skipped. Whilst a 2 minute delay may seem insignifcant, with the threshold for skipping stations being so low it doesn't take much.



(Screenshot shows a Chester - Liverpool service gradually losing time at stations)

 

[driverd]

An extra unit on the Chester line has been done but it did occasionally lead to a wait outside Chester station waiting for the platform. IIRC Wirral line services did used to interwork but when disruption occurred instead of delaying one line it delayed numerous. Personally I think there just needs to be more allowances on the Chester services (easier said than done) instead of having a 30 second scheduled station stop on most stations, increase it to a minute at more than just BKC. You can see with the service below how easily trains can lose time and with near on zero recovery time can easily lead to stations being skipped. Whilst a 2 minute delay may seem insignifcant, with the threshold for skipping stations being so low it doesn't take much.

View attachment 183366

(Screenshot shows a Chester - Liverpool service gradually losing time at stations)

And adding an extra unit diagram is precisely how you achieve that. As you say, with the current number of units required to operate the timetable, you cant stretch the timings any further. With an additional unit you can add time, be that at termini or through additional dwell time en-route.

 

[jamesst]

An extra unit would be the easiest option, however it can lead to standing time outside Chester, increased journey time and a through platform being occupied at Chester for longer which with it also being used by Tfw isn't ideal.

As for Sundays a 30min service with no recovery time between 2 busy cities with minimal turnaround is always going to lose. There are rumblings about a 15min Chester service on a Sunday so hopefully that could make a difference.

 

[driverd]

An extra unit would be the easiest option, however it can lead to standing time outside Chester, increased journey time and a through platform being occupied at Chester for longer which with it also being used by Tfw isn't ideal.

As for Sundays a 30min service with no recovery time between 2 busy cities with minimal turnaround is always going to lose. There are rumblings about a 15min Chester service on a Sunday so hopefully that could make a difference.

If you have a look at my post #109, which is a relatively straight forward solution, you can see that the xx:56 arrival could form an xx:15 departure, giving an 11 minute margin for platform re-occupation at xx:26. That shouldn't result in standing time at Chester unless a unit is severely delayed. I'm not sure where you've drawn that conclusion?

I did a quick scan of RTT for the Sunday spread, and couldnt find any instances where 15 minutes longer occupancy of platform 7 would impact any TfW services. Again, I'm not sure where you're drawing that conclusion from?

Longer journey times would only occur if you added additional dwell time, rather than additional turn around time. But fair comment.

I should add that I do understand that we're broadly in agreement here, I just think there's no need to make excuses for the operator when a simple solution is relatively workable, should they have the financial incentive to do so.

 

[Bikeman78]

It comes down to when the decision was made and the delay at the time.

Unfortunately once the decisions been made and the train has been sent semi fast if it then makes up time quicker than expected there are very limited options to get in touch with the driver to resume all stations meaning the train will go to the original station and stand time there.

Basically the Chester line timetable is not fit for purpose and badly needs a recast, a 4/5 min turnaround with no recovery time is asking for trouble.

We've all talked about the Chester line for years. Nothing seems to change. Cancelling a train at eight stations because it's five minutes late is not a good look. Same thing happened with the 1530 from Chester.

== Doublepost prevention - post automatically merged: 9 Jul 2025 ==

It's all well and good putting out extra diagrams, but with a lot of the terminuses having only one platform there really isn't that much space for extra trains

Chester is the only route that has a problem. All the others have recovery time somewhere in each round trip. I'd definitely prefer the three minute wait outside Chester to apply all year round, as opposed to random skip stopping.

== Doublepost prevention - post automatically merged: 9 Jul 2025 ==

Cost wise, a quick bit of rough maths suggests the unit would be out for around 17 hours, with a circa 40 mile trip performed 11 times, and very approx EMU running cost of £5/mile. Staff wise, drivers at approx £50/hour and guards at £40/hour (pension/NI taken into account), you're looking at £2200 for the unit, £900 for drivers and £720 for your TM (Plus potential additional lease costs if charged on mileage basis). To add significant resilience, £4000 per Sunday doesn't seem horrendous - especially if a more resilient service equates to more footfall.

Would the unit cost actually increase that much? Assuming you don't need to build another unit, then an extra train with longer turnaround times would reduce the mileage of every unit on the route. The only extra would be a few miles to and from the depot.

== Doublepost prevention - post automatically merged: 9 Jul 2025 ==

Sure but realistically I don't think Merseyrail control would allow a train to run 14 minutes late from one end or the other, even with a 19 minute turnaround, because it's so disruptive. That might be your and my preference but it's very unlikely.

I've seen it happen on Chester bound trains during the leaf fall timetable. The worst that can happen is a slight delay to an Ellesmere Port train but that's fine because they have plenty of dwell time at the terminus.

 

[Starmill]

I've seen it happen on Chester bound trains during the leaf fall timetable. The worst that can happen is a slight delay to an Ellesmere Port train but that's fine because they have plenty of dwell time at the terminus.

It also happens when the controllers are under too heavy a workload to make any changes. Doesn't mean it's going to be policy though. In any case the leaf fall example being extended year round is about the only solution, but this would have to be paid for somehow, and potentially with capacity taken out of the weekday timetable in compensation for it. Unless there's something forthcoming to sign that off, I think the current situation will continue.

 

[driverd]

Would the unit cost actually increase that much? Assuming you don't need to build another unit, then an extra train with longer turnaround times would reduce the mileage of every unit on the route. The only extra would be a few miles to and from the depot.

Yes, absolutely correct (you can tell I've never been a finance person) - the actual mileage will be the same as it's the same timetable, the only additional cost will be the electric for units stood on increased dwell times, plus the transit to the depot.

So really you are only looking at a bit extra on the crew bill.

My guess would be that there's probably a contractual element on unit availability and the timetable being built around an expected level of performance from the units, that for whatever reason, isnt being achieved. The fix to this problem is both cheap and simple enough to implement (and is the case in Autumn anyway), so there must be a reason not to do it.