• Our booking engine at tickets.railforums.co.uk (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!

Carmarthenshire and Pembrokeshire frequency and electrification

A S Leib

Member
Joined
9 Sep 2018
Messages
785
With the current thread on reopening Carmarthen–Aberystwyth, how viable would it be to improve the service on existing lines? By my count the lines west of Carmarthen and the Cambrian line are amongst only around a dozen lines in the UK with less than an hourly service; how viable would it be to ensure at least an hourly service for Pembroke and Milford Haven (and Aberystwyth, although I think that's just missing three or four trains per day for an hourly service) and at least one train per two hours to Fishguard? The fact that they're bi-modes on a route which isn't really on the way to anywhere else means it's less pressing than, say, Market Harborough to Sheffield, but how viable would electrification from Cardiff to Carmarthen - now that there's a bi-hourly Paddington service - be? I'm not saying that if Aberystwyth–Carmarthen does - somehow - ever go ahead other improvements should be ignored, but it seems odd to ignore what I'd guess is by far the lower-hanging fruit.
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
.., how viable would it be to improve the service on existing lines?
This bit, easy. We had a thread on South Pembs and bumping to hourly a year or so ago, is almost feasible without needing any more rolling stock or crew, due to how bad the current service is for utilization.
.. but how viable would electrification from Cardiff to Carmarthen - now that there's a bi-hourly Paddington service - be?
It wouldn't.
 

30907

Veteran Member
Joined
30 Sep 2012
Messages
18,114
Location
Airedale
This bit, easy. We had a thread on South Pembs and bumping to hourly a year or so ago, is almost feasible without needing any more rolling stock or crew, due to how bad the current service is for utilization.
Without repeating the whole thread, the problem with Pembroke is that you would need to cut the journey time by several minutes either side of Tenby to make an hourly service feasible.
 

Dr Day

Member
Joined
16 Oct 2018
Messages
545
Location
Bristol
how viable would it be
Depends on your definition of 'viable'. Generally the further away from major centres of population, the less financially viable a service will be. As others have said, with some journey time improvements it may be possible to create a theoretically operationally feasible timetable, but it will still need a government commitment to fund the ongoing subsidy on top of any capital costs required to generate the journey time improvements before anything could actually happen.
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
Without repeating the whole thread, the problem with Pembroke is that you would need to cut the journey time by several minutes either side of Tenby to make an hourly service feasible.
TL;DR - you can run current service frequency with one fewer units by shifting to pass at Whitland instead of Tenby.

They have the daftest timetable at the moment:

The up service - having sat at Pembroke Dock (PMD) for 55 mins, sets off. It waits at Tenby (11.37a) for the down to arrive (11.39a) and release the token manually (~90 seconds). The up leaves at 11.42 having collected that token (driver walks back, sits down etc). There's almost no spare time at all if the Down is on time.

If the Down is late - then the Up has to be late by almost the same amount.. or at best, it's only 2 minutes fewer late than the incoming Down. Hence, they re-import the Down's delay to the system beyond Whitland (WTL) - where pathing at WTL and CMN matter.

A better arrangement would be for the Up to leave PMD +5 minutes later, so that the Down might get delayed in the unlikely event that the Up (which is just having had a 55 minute layover). At least that would help reduce re-imported delays.

Or, with passing at WTL instead of TEN - they can skip 3 mins on down and 5 mins on the up of timetabled wait (assuming they can be given both tokens at WTL..) - and do WTL->PMD->WTL in 1h56 mins - one unit achieving 2-hourly frequency, with the same (poor) reliability as the current arrangement!

I did read they intend to shut Llanion Crossing (stop and blow) - that should save 30 seconds at the PMD end - planned for ~2020, it's gone quiet..
 

anthony263

Established Member
Joined
19 Aug 2008
Messages
6,550
Location
South Wales
TL;DR - you can run current service frequency with one fewer units by shifting to pass at Whitland instead of Tenby.

They have the daftest timetable at the moment:

The up service - having sat at Pembroke Dock (PMD) for 55 mins, sets off. It waits at Tenby (11.37a) for the down to arrive (11.39a) and release the token manually (~90 seconds). The up leaves at 11.42 having collected that token (driver walks back, sits down etc). There's almost no spare time at all if the Down is on time.

If the Down is late - then the Up has to be late by almost the same amount.. or at best, it's only 2 minutes fewer late than the incoming Down. Hence, they re-import the Down's delay to the system beyond Whitland (WTL) - where pathing at WTL and CMN matter.

A better arrangement would be for the Up to leave PMD +5 minutes later, so that the Down might get delayed in the unlikely event that the Up (which is just having had a 55 minute layover). At least that would help reduce re-imported delays.

Or, with passing at WTL instead of TEN - they can skip 3 mins on down and 5 mins on the up of timetabled wait (assuming they can be given both tokens at WTL..) - and do WTL->PMD->WTL in 1h56 mins - one unit achieving 2-hourly frequency, with the same (poor) reliability as the current arrangement!

I did read they intend to shut Llanion Crossing (stop and blow) - that should save 30 seconds at the PMD end - planned for ~2020, it's gone quiet..
Takes 25 minutes Pembroke dock to Tenby but 30 minutes to 35 minutes in the other. Having to stop at level crossings won't help. But definitely need a extra passing loop on the branch. Say Templeton
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
Takes 25 minutes Pembroke dock to Tenby but 30 minutes to 35 minutes in the other.
Timetabled at 27 mins for Tenby - Whitland in WTT.
Having to stop at level crossings won't help. But definitely need a extra passing loop on the branch. Say Templeton
Just sayin' - the current service is actually worst of both worlds - a two hour frequency and delays re-imported every time.. for zero cost, they could fix the timetable to stop re-importing delay.
 

Western 52

Member
Joined
19 Jun 2020
Messages
1,127
Location
Burry Port
Hourly Swansea to Tenby and Haverfordwest, 2 hourly on to Pembroke Dock and Milford Haven, plus 2 hourly Swansea to Fishguard. This gives 2 or 3 per hour Whitland to Swansea with TfW, plus the GWR services to Carmarthen. The Pembroke line does need to be speeded up though and would benefit from another loop between Tenby and Whitland. This would be a decent service, and would also allow Kidwelly and Ferryside to be served hourly.

As regards electrification, that would be great, but probably not viable.
 

Snow1964

Established Member
Joined
7 Oct 2019
Messages
6,293
Location
West Wiltshire
Wales news outlets reporting that Swansea electrification might be being revived

DISCUSSIONS about electrifying the mainline between Cardiff and Swansea are reportedly “back on the table”.

Swansea Council leader Cllr Rob Stewart made the remark at a green economy conference in the city.

The UK’s Department for Transport has not commented directly on what the Labour politician said. But it said a group called the Wales Rail Board was identifying priority proposals for new infrastructure.

The Coalition Government led by Conservative prime minister David Cameron promised in 2012 to electrify the mainline between London Paddington and Swansea - but five years later the Cardiff to Swansea section was shelved. At the time the UK Government said this was because project costs had overrun significantly.

UK ministers said the introduction of “bi-mode” trains which switch from electric to diesel power would run just as quickly between Cardiff and Swansea as electric-powered ones. Electrifying the section between the two cities was based on a cost of around £430 million, in 2012-23 prices.

Transport consultant and academic Mark Barry told the Local Democracy Reporting Service that the case for electrification had never really gone away. He said lighter electric trains accelerated faster and had a better environmental footprint than bi-mode and diesel trains.
The Local Democracy Reporting Service asked the Department for Transport if the Cardiff to Swansea electrification proposal was being looked at again, and if so what stage discussions were at and whether any preliminary costings had been done.

It didn’t answer the questions directly, but said: “The Wales Rail Board has begun work to identify priority proposals for infrastructure in Wales, with more information on specific projects in due course.”

The Wales Rail Board, which was was set up in 2021, has senior representation from the Welsh Government, Department for Transport, Office of the Secretary of State for Wales, Transport for Wales and Network Rail, and is responsible for deciding and developing priority schemes. Its meetings are held behind closed doors and no agendas or minutes are published. The board doesn’t hold the purse strings for new rail investment – that remains under existing arrangements, with the UK Government largely responsible.

 

Western 52

Member
Joined
19 Jun 2020
Messages
1,127
Location
Burry Port
Great to see electrification is under discussion again, and hopefully we'll get positive news in due course. Funding will be the sticking point of course: if the UK government won't pay, who will? Unlikely Welsh Government could afford it.
 

anthony263

Established Member
Joined
19 Aug 2008
Messages
6,550
Location
South Wales
Great to see electrification is under discussion again, and hopefully we'll get positive news in due course. Funding will be the sticking point of course: if the UK government won't pay, who will? Unlikely Welsh Government could afford it.
Theres a lot of pressure from the tories own mps abd business supporters who were against electrification being cut back to Cardiff. Euro 2028 opening ceremony in Cardiff also helping push the case.

Great to see electrification is under discussion again, and hopefully we'll get positive news in due course. Funding will be the sticking point of course: if the UK government won't pay, who will? Unlikely Welsh Government could afford it.
Welsh government might be able to fi d some money. Frees up class 197s too especially if swanline goes to EMUs
 

Rhydgaled

Established Member
Joined
25 Nov 2010
Messages
4,568
Replying to some comments by @nigelsporne @Jez and @anthony263 in the class 197 topic here...

Although I know very little about TfW services I find this thread interesting. So with the increase in the frequency from Bidston using 197s is there also a move to increase services west of Carmarthen? The services here seem very poor in this day and age. Are services restricted because of the amount of single line and the spacing or lack of crossing points? Is there a demand? Are 197s suitable?
197s will run all TFW services West of Carmarthen eventually. As for the frequency there isnt much they can do with the single line sections i dont think aside from the planned Fishguard increase to 1 train every 2 hours.
All three Pembrokeshire branches I think could handle a more-frequent service than that currently offered, despite the single-line sections.

Given the cost of fuel/staff etc. I don't think it is realistic to expect much more than a train every two hours on the Fishguard branch, which can easily accomodate that frequency without infrustructure enhancement.

Milford Haven and Haverfordwest should justify an hourly service in my view. Haverfordwest has two platforms (although only one is normally used by passenger services, presumably because the only step-free access is a barrow crossing) so I assume trains could pass there if the timetable required in which case an hourly service is probably achievable without additional loops. Some time savings between Haverfordwest and Milford Haven might be desirable though to improve robustness to disruption.

The Pembroke Dock branch is where the current infrustructure falls short of that which would be required to efficiently run the level of service that would be desirable. Ideally there would be at least an hourly service between Swansea and Tenby (at least in the summer) on memorable clockface timings. However, west of Tenby rail is at a huge geographical disadvantage to the A477, which now seems to have bypassed almost all the settlements on route. As such I think a better approach would be to run an express bus or coach non-stop between Carmarthen and Pembroke (which would then proceed to Pembroke Dock) to complement the train service (where the minimum service between Tenby and Pembroke Dock should be the current every two hours). Unfortunately I think the best that can be done without building additional loops, or (to play devil's advocate) closing the line west of Tenby, is a shuttle every 90 minutes between Pembroke Dock and the bay platform at Whitland. This would be a more-frequent service than the current every two hours, but would mean passengers having to change at Whitland (yuck) and prevent the introduction of memorable clockface timings. You might even be able to push that to a shuttle every 75 minutes at the cost of being even less-memorable.

Certainly with the Pembroke dock branch you'd need to reinstate the loops at Manorbier and Templeton as well as raise linespeeds especially west of Tenby
Actually, as above I'm pretty sure you can actually improve the frequency (though not to hourly) without additional loops at the cost of terminating at Whitland. You would only need to shave a few minutes off the running time* to get from Whitland to Tenby and back in just under an hour, so if you did close the line west of Tenby an hourly service would be possible in theory**. However, doing that would mean the hourly trains towards Tenby would always be departing Whitland just a few minutes after they arrive from Tenby - at a similar point of the hour. They could in theory continue to/from Carmarthen and Swansea, but would be passing at Whitland would therefore not be able to fit into a nice, memorable, regular-interval clockface timetable between Swansea/Cardiff and Milford Haven / Fishguard Harbour since those would ideally not be passing at Whitland. Either something would have to have a long wait somewhere to space the trains out to the ideal of 30 minute invervals between Llanelli and Swansea, or the Tenby trains in one direction would be just a few minutes behind/in-front-of the Milford/Fishguard workings then nothing for nearly an hour. To get the Tenby, Fishguard and Milford trains to fit nicely together into a regular-interval timetable (the ideal) then you do need an additional passing loop, probably around Templeton. However, an additional loop would probably also be required west of Tenby (possibly that would be Manorbier - I've not really looked into it since I don't see there being sufficient demand for hourly beyond Tenby so what I was trying to do was hourly Swansea-Tenby (passing at Templeton) with one every two hours continuing to Pembroke Dock***)

* You probably could do that by raising linespeeds and closing Saundersfoot station - it's nowhere near the town centre and outside the built-up area, replace it with much-improved bus-links to/from Kilgetty and Tenby.

** getting to hourly without closing the line west of Tenby or building additional loops would require some fairly heroic journey-time reductions west of Tenby**** and still wouldn't solve the issue of terminating at windswept Whitland, not the nicest place to have to change trains due to the exposed wait

*** either I couldn't get this to work without an additional loop, or it would only work if the trains that terminate at Tenby sat there for two hours (being passed by a Pembroke Dock service in at least one direction while waiting there) before heading back

**** there are some level crossings between Tenby and Pembroke Dock which require the train to stop and sound the horn before proceeding, but would upgrading these to allow the train to cross at linespeed (50mph, although I think one of them is also the start/end of a 20mph section into/out of Pembroke Dock station) save the required 5+ minutes in each direction?
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
The Pembroke Dock branch is where the current infrustructure falls short [..]. However, west of Tenby rail is at a huge geographical disadvantage to the A477, which now seems to have bypassed almost all the settlements on route. As such I think a better approach would be to run an express bus or coach non-stop between Carmarthen and Pembroke [..].
The most popular destination from Pembroke Dock is Tenby (followed by Carmarthen (1/3rd the size of the Tenby flow) and Swansea) according to the recent ticket data shared in another thread. The low numbers going to Carmarthen and beyond probably are lower than could suppressed demand due to journey time - I know some people drive from Pembroke/Pembroke Dock to Kilgetty (20 mins) to catch the train (40 mins) when they've missed it at Pembroke. The Tenby-Pembroke/PemDock bit does fairly well on a Friday and Saturday night when the buses in the area have finished!

* You probably could do that by raising linespeeds and closing Saundersfoot station - it's nowhere near the town centre and outside the built-up area, replace it with much-improved bus-links to/from Kilgetty and Tenby.
That'd not be popular. Even if it's 2 miles from Saundersfoot centre, it's a quick run down in the car or a taxi. On a Saturday in summer the road to Kilgetty is often clogged so that'd be a loss for Saundersfoot. Not going to happen.
**** there are some level crossings between Tenby and Pembroke Dock which require the train to stop and sound the horn before proceeding, but would upgrading these to allow the train to cross at linespeed (50mph, although I think one of them is also the start/end of a 20mph section into/out of Pembroke Dock station) save the required 5+ minutes in each direction?
Those could be done.

There was a plan to close Llanion Crossing (https://www.networkrail.co.uk/wp-co...-routes-east-and-west-of-Llanion-Cottages.pdf is the map/plan) but that has gone quiet.

That A477 speed advantage could be reduced if services didn't have 5-6 minutes for token exchange/passing at Tenby, and I guess 4 minutes total messing around with level crossings (3 stop and blow, and one at Manorbier the driver gets out to operate). That'd cut the Kilgetty-Pembroke Dock 40mins vs 20 mins difference to 30 mins vs 20 mins.
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
The three stop, toot and go level crossings are being replaced by 5 mph passing.

A video demo on Network Rail went up today:


The line's currently closed for 10 days (bus replaced):


says:
Network Rail engineers will be replacing more than 200m of track and installing more than 300 tonnes of new ballast (track stone) close to Pembroke station. In addition, essential drainage work will be carried out between Narberth and Kilgetty stations.
 
Last edited:

AlastairFraser

Established Member
Joined
12 Aug 2018
Messages
2,165
The three stop, toot and go level crossings are being replaced by 5 mph passing.

A video demo on Network Rail went up today:


The line's currently closed for 10 days (bus replaced):


says:
How much time will this save?
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
How much time will this save?
Hard to say. I don't think the current arrangement has a speed limit once the stop and blow is complete, whereas the new method sounds like 5 mph throughout. For a 1-car 153, must be faster in the new arrangement by maybe 30 seconds - but if a 5-car IET - that's a long time to do 5mph if it's for the whole length of train passing. There is no alteration in the summer timetable to take advantage of any speed up.
 

Bald Rick

Veteran Member
Joined
28 Sep 2010
Messages
29,225
Hard to say. I don't think the current arrangement has a speed limit once the stop and blow is complete, whereas the new method sounds like 5 mph throughout. For a 1-car 153, must be faster in the new arrangement by maybe 30 seconds - but if a 5-car IET - that's a long time to do 5mph if it's for the whole length of train passing. There is no alteration in the summer timetable to take advantage of any speed up.

Note it is 5mph on approach. Presumably there will be “A” boards that enable the train to accelerate immediately after the front cab is over the crossing.

These crossings have had numerous incidents at them in recent years.
 

Djgr

Established Member
Joined
30 Jul 2018
Messages
1,684
As a slight aside at the weekend as we came into Crewe, the conductor on the Birmingham to Liverpool service said "Change here for Shrewsbury, Aberystwyth and Carmarthen"
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
Note it is 5mph on approach. Presumably there will be “A” boards that enable the train to accelerate immediately after the front cab is over the crossing.
I do hope that that's the plan, no reason why it shouldn't be but A boards aren't that common, or are they? First I heard of them was Birmingham New St.
These crossings have had numerous incidents at them in recent years.
They have - incidents more than accidents - it takes a degree of skill to have an accident with a train that's blown its horn and is accelerating from stop. One incident saw a bin lorry reversed whilst the train was stopped, with its banksman waving the bin lorry back (Llanion Crossing, 200 yds from Pembroke Dock) as if the train was letting it go.

An actual accident had a taxi slam in to the rear bogie of a 153 at the Beavers Hill crossing (for points not prison, nor lost licence). The taxi driver said he had misjudged the speed of the train (presumably thought it was going faster than it actually was) - https://www.bbc.co.uk/news/uk-wales-45260451

They had a rather expensive plan to close Llanion Crossing, diverting paths, building footpaths, acquiring land (see plan at https://www.networkrail.co.uk/wp-co...-routes-east-and-west-of-Llanion-Cottages.pdf) in 2020. The new idea sounds much more cost effective, and presumably the other idea dropped.
 

fishwomp

Member
Joined
5 Jan 2020
Messages
550
Location
milton keynes
Unlikely to make an hourly interval service feasible, even if it were justified - not least because Tenby to Whitland is also too long a journey time.
Is 27 minutes on some diagrams. Also, that'd be timed (hasn't changed) for a 153, which is a bit of a lard-butt - 213kW for 41 tonnes (5.2 kW/t, vs 375kW for the upcoming 197s over ~50 tonnes (~7 kW/t). The incline from Tenby, IIRC is around 1 in 50 so power is needed - and would make a difference. At 27 mins it'd work, but you'd need to get rid of exchanging the token at the box (beyond the station, in Carmarthen direction), and have automatic signalling from where the branch diverges at Whitland.
 

AlastairFraser

Established Member
Joined
12 Aug 2018
Messages
2,165
Hard to say. I don't think the current arrangement has a speed limit once the stop and blow is complete, whereas the new method sounds like 5 mph throughout. For a 1-car 153, must be faster in the new arrangement by maybe 30 seconds - but if a 5-car IET - that's a long time to do 5mph if it's for the whole length of train passing. There is no alteration in the summer timetable to take advantage of any speed up.
30 seconds for 3 crossings? 1.5 mins then. Not too bad
Unlikely to make an hourly interval service feasible, even if it were justified - not least because Tenby to Whitland is also too long a journey time.
Perhaps some of the curvature could be eased - this seems to hold back the journey time significantly.
 

HST43257

Established Member
Joined
10 Apr 2020
Messages
1,446
Location
York
IMO

Electrify to Swansea, with the existing 1tph from the marches and 1tph from London.
Maybe extend electrification to Carmarthen and maybe extend the marches service.
Crucially, Pembroke and Milford Haven on shuttles from Carmarthen (or maybe Swansea?) unless I can be told that there is realistic demand to/beyond Swansea using rail.

Where do year-round passengers from the Pembroke or Milford Haven branches go?
 

Topological

Member
Joined
20 Feb 2023
Messages
772
Location
Swansea
IMO

Electrify to Swansea, with the existing 1tph from the marches and 1tph from London.
Maybe extend electrification to Carmarthen and maybe extend the marches service.
Crucially, Pembroke and Milford Haven on shuttles from Carmarthen (or maybe Swansea?) unless I can be told that there is realistic demand to/beyond Swansea using rail.

Where do year-round passengers from the Pembroke or Milford Haven branches go?

Electrifying to Swansea would mean potentially splitting services, but does make sense.

Swanline services would benefit if an EMU can outperform the 197 for acceleration and braking (presumably possible) because that removes a common delay factor for GWR. Swanlines go beyond Swansea, but could presumably be split.

Marches trains are unlikely to be electric, so are unlikely to benefit from the wires. Highly unlikely there would even be bi-modes on the Marches so I cannot see those trains using the wires. Thanks to the TfW vanity project there are not many Marches services to Swansea any more anyway.

IF Bristol Temple Meads gets wired then a Swansea to Bristol train is possible, though that would mean a third "fast" train between Swansea and Cardiff and I am not sure if the demand is there. The other fast train candidate is the present Cardiff terminator from Paddington, but likewise it would be demand dependent.

So really Swansea to Cardiff just allows GWR to have their services fully electric and gives potential resilience to delays created by Swanline services pathed in front of the GWR.
 

Rhydgaled

Established Member
Joined
25 Nov 2010
Messages
4,568
They had a rather expensive plan to close Llanion Crossing, diverting paths, building footpaths, acquiring land (see plan at https://www.networkrail.co.uk/wp-co...-routes-east-and-west-of-Llanion-Cottages.pdf) in 2020. The new idea sounds much more cost effective, and presumably the other idea dropped.
While the new proposal sounds like it could be cheaper, but it isn't clear what the relative benefits could be. Presumably there still won't be wig-wag lights for the open crossings and, if there will be, why will the trains be limited to just 5mph?

Unlikely to make an hourly interval service feasible, even if it were justified - not least because Tenby to Whitland is also too long a journey time.
Is 27 minutes on some diagrams. Also, that'd be timed (hasn't changed) for a 153, which is a bit of a lard-butt - 213kW for 41 tonnes (5.2 kW/t, vs 375kW for the upcoming 197s over ~50 tonnes (~7 kW/t). The incline from Tenby, IIRC is around 1 in 50 so power is needed - and would make a difference. At 27 mins it'd work, but you'd need to get rid of exchanging the token at the box (beyond the station, in Carmarthen direction), and have automatic signalling from where the branch diverges at Whitland.
I expect you would need automatic signalling (and possibily a longer (dynamic) loop) at Tenby too. At around 28 minutes either side of Tenby, I doubt there is sufficient time to be operating token exchange equipment anywhere if you want to run a punctual hourly service between Whitland and Tenby. Would you actually need to remove the token exchange at Whitland though since, as noted above, if Tenby is to remain the only passing loop between Whitland and Pembroke Dock the service would probably turn back in the bay at Whitland rather than passing the next incoming service there.

Where do year-round passengers from the Pembroke or Milford Haven branches go?
Mostly local travel west of Carmarthen with smaller numbers to Swansea, Cardiff, Bristol and London if Pembroke's representative.
https://railalefan.co.uk/labs/flowstats/PMB/
Current usage of any route does not necessarily reflect potential demand - if the current service does not provide for it. For example, if there was a massive flow of 9-to-5 commuters between Cardigan and Haverfordwest* they would currently have to go by car since I think the last bus back to Cardigan leaves at 5pm on the dot (so no time to get from office to bus stop - and even if there was the bus takes twice as long as driving (due to a massive detour) so most would probably drive anyway).

Years ago, the North Pembrokeshire Transport Forum did a survey regarding Fishguard rail services. One of the questions asked related to the destinations to which locals currently travel or would like to travel by train from Fishguard. The most-popular response was London, followed by Cardiff and Swansea (in that order), and then Carmarthen and Bristol (both attracting only 31 responses, along way behind Swansea at 79). More details can be found on the NPTF website:
* I don't know whether there is, this is purely to illustrate my point.

The demand pattern may be similar for Pembroke Dock, or it may not. Even if there is similar strong demand for Cardiff, Swansea and London, I expect many drive to at least Whitland (and probably much further) rather than using the slow, infrequent, service from Pembroke Dock. Unless you can do something to considerably speed up the railway (or slow dow the road) I expect the majority of passengers on trains between Pembroke Dock and Tenby will be making short trips (mostly within Pembrokeshire), regardless of how many people there are in Pembroke Dock wanting to go further afield. Hence my earlier suggestion of an express bus non-stop between Pembroke or Pembroke Dock and Carmarthen.

That A477 speed advantage could be reduced if services didn't have 5-6 minutes for token exchange/passing at Tenby, and I guess 4 minutes total messing around with level crossings (3 stop and blow, and one at Manorbier the driver gets out to operate). That'd cut the Kilgetty-Pembroke Dock 40mins vs 20 mins difference to 30 mins vs 20 mins.
It'd still be about 7 miles further by rail than road from Whitland to Pembroke Dock. A saving of 4 minutes each way between TEN and PMD by closing the level crossings or upgrading them to automatic with wig-wag lights** would in theory allow the train to get from TEN to PMD and back in just under an hour - do something about the token exchange in Tenby and you could potentially improve the frequency side of the equasion by enabling an hourly service subject to all the above.

** since they are on narrow roads even an 'Automatic Half Barrier' would presumably block the whole road, avoiding the AHB problem of dangerous driving by motorists zig-zagging through the 'open' half of the crossing

Electrify to Swansea, with the existing 1tph from the marches and 1tph from London.
Maybe extend electrification to Carmarthen and maybe extend the marches service.
Electrifying to Swansea would mean potentially splitting services, but does make sense.

Swanline services would benefit if an EMU can outperform the 197 for acceleration and braking (presumably possible) because that removes a common delay factor for GWR. Swanlines go beyond Swansea, but could presumably be split.

Marches trains are unlikely to be electric, so are unlikely to benefit from the wires. Highly unlikely there would even be bi-modes on the Marches so I cannot see those trains using the wires. Thanks to the TfW vanity project there are not many Marches services to Swansea any more anyway.

IF Bristol Temple Meads gets wired then a Swansea to Bristol train is possible, though that would mean a third "fast" train between Swansea and Cardiff and I am not sure if the demand is there. The other fast train candidate is the present Cardiff terminator from Paddington, but likewise it would be demand dependent.

So really Swansea to Cardiff just allows GWR to have their services fully electric and gives potential resilience to delays created by Swanline services pathed in front of the GWR.
I think the more-urgent issue between Cardiff and Swansea is the lack of overtaking opportunities for fast trains to overtake Swanline stoppers. Not sure if full four-tracking is necessary but Swanline stations with loop platforms off the up and down mains is needed at a minimum. While that is dealt with, England needs to be getting the wires from Didcot to Oxford (and beyond). Wiring Swansea to Cardiff doesn't really allow GWR to take the diesel generators off (m)any 800s/802s anyway since they also run through to Carmarthen (quite often) and Pembroke Dock (very occasionally).

I do think the wires should be extended to at least Swansea though - whether to prioritise Carmarthen or not rather depends on what happens with the GWR services. As far as these are concerned the priorities in my view should be:
  1. have DfT/GBR/??? launch a competition to design a new long-distance train seat that acheives full marks under the seat comfort scoring system that was developed by (I think) the RSSB a few years ago for standard class 2+2 interiors
  2. lengthen the platforms / adjust the signalling / extend loops etc. at Carmarthen (and elsewhere eg. Newquay / Tenby / Pembroke Dock if neccessary) to safely accomodate a 9-car 800/802 set
  3. order a number of additional class 800/801/802 TS and/or MS vehicles, without diesel engines/gensets and fitted with the newly designed comfortable seats
  4. bin the existing Fainsa Sophia seats in the class 800, 801 and 802 units and replace them with the newly designed comfortable seats
  5. insert the new TS and/or MS vehicles into a large number of GWR's class 800/802 units, and move either genset-fitted vehicles (or the gensets themselves) between units as appropriate so that GWR ends up with significantly more 9-car sets and either no 5-car sets at all or only a small number of them that would be kept to things like off-peak London-Worcester runs (personally I would be surprised if even these can happily accomodate the demand all day long without needing at least a 7-car set at some point in the diagram - even if the diagram avoids London in the peak I expect peak travel in and out of cities like Oxford will fill a 5-car full-and-standing but I don't know this for sure) and never run in multiple. Since eliminating portion working like this would mean GWR has far too many InterCity units I would cascade some (my original idea was 8-car sets, but some say even-numbers aren't possible for the design) to XC (possibily could still be maintained by at the Bristol Parkway depot).
4 and 5 should be done in parallel while the trains are in works anyway for the other job

I'm guessing that 9-car sets would mean insufficient demand to continue the current frequency of GWR Carmarthen services, in which case leave it unwired west of Swansea for some time yet.
 

Topological

Member
Joined
20 Feb 2023
Messages
772
Location
Swansea
It is drifting a little, but the obvious sections that require an overtaking option for the GWR are between Bridgend and Cardiff, where loops do exist, and between Port Talbot and Swansea, where I do not think there is any option. For the latter Britton Ferry could be given a loop (I think).

I agree that the wires are unlikely to go any further west and that if GWR continue to offer the Carmarthen service then that will need the diesel capabilities of the 80x. It makes sense to continue to have some Carmarthen, so I agree that 9-car could be looked at. However, whilst GWR have the option to run 2x5 I suspect that splitting and joining at Swansea makes sense.

Whether going to the extreme of having GWR serve Carmarthen hourly would strengthen the case for other services to be split (such as Swanline, or West Wales to Cardiff) I do not know. Turning at somewhere like Whitland is still not Carmarthen at that end.

A lot to think about, but a solution to make the Swansea to Cardiff section more robust would help soften any splits.
 

anthony263

Established Member
Joined
19 Aug 2008
Messages
6,550
Location
South Wales
I've been on swanlines looped between Taibach and Margam to allow a GWR train to overtake. December will see Swanline retimed to leave Swansea 5 minutes behind the GWR service
 

Topological

Member
Joined
20 Feb 2023
Messages
772
Location
Swansea
I've been on swanlines looped between Taibach and Margam to allow a GWR train to overtake. December will see Swanline retimed to leave Swansea 5 minutes behind the GWR service
That will be a big improvement.

The other problem is the Maestegs that stop between Bridgend and Cardiff.

Whenever I have cause to track a GWR into Cardiff it is the Swanline getting held up by an ex-Maesteg in front of that one. Letting the GWR catch up with the Maesteg will be a bit better though.

I was also looped at Margam on a train to Manchester pre-Covid. No problem for me as I was going to Manchester, less good for all those going to Cardiff/Newport/Bristol who crowded onto the 175/1 when the GWR they really wanted sailed past.
 

Top