Salisbury - Exeter punctuality improvements?

[pompeyfan]

Hi all,

I have a couple of questions regarding the Salisbury - Exeter route and its poor punctuality.

There are regularly delays on this route, usually ranging from 5 - 15 minutes with seemingly nothing major ongoing. This led me to question whether the 1tph since the Axminster loop was commissioned had ever been delivered punctually, or whether there’s slowly been in decline in timekeeping.

I’m aware the route has several single line sections, but from what I can see, trains are timed to pass to the minute which leads to delays “snowballing” or not being recovered throughout the day.

If the 2008 timetable was delivered better, what has changed that has made the situation worse? Defensive driving, the age of the units affecting their acceleration, increased passengers leading to longer dwell times? None of the above? All of the above? If so, what would the silver bullet be to fix these issues?

 

[The Ham]

Hi all,

I have a couple of questions regarding the Salisbury - Exeter route and its poor punctuality.

There are regularly delays on this route, usually ranging from 5 - 15 minutes with seemingly nothing major ongoing. This led me to question whether the 1tph since the Axminster loop was commissioned had ever been delivered punctually, or whether there’s slowly been in decline in timekeeping.

I’m aware the route has several single line sections, but from what I can see, trains are timed to pass to the minute which leads to delays “snowballing” or not being recovered throughout the day.

If the 2008 timetable was delivered better, what has changed that has made the situation worse? Defensive driving, the age of the units affecting their acceleration, increased passengers leading to longer dwell times? None of the above? All of the above? If so, what would the silver bullet be to fix these issues?

In answer to how to fix it, the largest silver bullet surely would be to redouble and electrify the whole route.

 

[Brush 4]

Yes, more track, simple as that. Redoubling some of it, where exactly is the key point.

 

[Kite159]

If the 2008 timetable was delivered better, what has changed that has made the situation worse? Defensive driving, the age of the units affecting their acceleration, increased passengers leading to longer dwell times? None of the above? All of the above? If so, what would the silver bullet be to fix these issues?

Cranbrook opening hasn't helped the timekeeping. As well as increased dwells at times of day due to volume of passengers.

The timetable is very fragile with trains timed to pass at Pinhoe & at Tisbury loop. If a London train gets delayed by 5 minutes at Pinhoe, then it might also delay the following Exeter at Axminster and will delay the Exeter 2 hours later at Tisbury.

Ideally the double track would extend from Pinhoe further east (which would allow 2tph between Exeter & Honiton/Axminster), maybe also rebuilding Tisbury station to be double track considering the industrial area on the former platform 2 land is wasteland)

 

[pompeyfan]

When I say silver bullet I would really appreciate it if people could keep their responses realistic to prevent this thread becoming classed as speculative. I know there’s ongoing plans for a new loop at Cranbrook that could reduce the reactionary delays.

 

[The exile]

You are never going to eliminate all primary delays. If (as is described here) there is a significant problem with delays snowballing on a single line there is basically one answer - allow more time for trains to cross each other. This can be done in 2 ways: a) lengthen passing loops into dynamic loops so a small delay to one train merely alters exactly where in the loop it passes its “oppo” [the gold-plated version of this being to double throughout] or b) pad the timetable so that trains sit for much longer at crossing points. Option B is unlikely to go down well with passengers (and may require extra trains /crew), while option A is unlikely to go down well with the Treasury.

 

[Zomboid]

I feel like it would help to provide two platforms at both Tisbury and Templecombe, so potentially delayed trains don't have to block the single line whilst making their stops. Making the loops looking enough that most acceleration and deceleration is done on the double section ought to help with keeping occupancy time of the single line to a minimum, too.

 

[pompeyfan]

Cranbrook opening hasn't helped the timekeeping. As well as increased dwells at times of day due to volume of passengers.

The timetable is very fragile with trains timed to pass at Pinhoe & at Tisbury loop. If a London train gets delayed by 5 minutes at Pinhoe, then it might also delay the following Exeter at Axminster and will delay the Exeter 2 hours later at Tisbury.

Ideally the double track would extend from Pinhoe further east (which would allow 2tph between Exeter & Honiton/Axminster), maybe also rebuilding Tisbury station to be double track considering the industrial area on the former platform 2 land is wasteland)

Do you know whether the timetable was reasonable resilient before Cranbrook opened? I have noticed that if a train is formed of 3 coaches then it will slowly lose time as well, backing up your theory regarding passenger numbers.

I also think I read somewhere (possibly on here) that the 159s have had their engine output reduced slightly, I think the rev governor was reduced to save on wear and tear. I would imagine this could affect units, especially on the steep gradients on that route.

 

[devon_belle]

Extending the loops would presumably allow for the timetable to absorb more delay. In an ideal world, redoubling nearly all the line (good luck getting under the M5) would be the best option.

Double track as far as Tisbury (inc. restored down platform), using the old Up line which is still in situ (albeit unused since the 90s) for part of the way near Dinton would seem sensible. Even just extending Tisbury loop into the platforms would help. Connecting Axminster and Chard loops as a long dynamic loop - as originally envisioned - would help around there. The long-touted '6 mile dynamic loop' around Whimple would probably be good for regulation as well as the extra tphs for Devon Metro.

Beyond all that, some better traction (electric?) to get over the banks at more than 55 mph would surely be a game changer for service recovery, provided they didn't just squeeze the timetable more...

 

[brad465]

Cranbrook opening hasn't helped the timekeeping. As well as increased dwells at times of day due to volume of passengers.

If Chard Parkway opens in the absence of any other improvements that will worsen things still. They'd at a minimum have to make Chard Junction-Axminster one long dynamic loop to extend where trains can pass in this area.

 

[AlastairFraser]

I feel like it would help to provide two platforms at both Tisbury and Templecombe, so potentially delayed trains don't have to block the single line whilst making their stops. Making the loops looking enough that most acceleration and deceleration is done on the double section ought to help with keeping occupancy time of the single line to a minimum, too.

If you assume around 800m average new track + platform to redouble the line at every station between Salisbury and Exeter (stations like Templecombe would probably need closer to 500m of new track, given their proximity to existing loops) and bear in mind that you might need staggered platforms at some of them to avoid encroaching on housing, there are 6 stations that fit the criteria - therefore we're looking at approximately 3 miles of new track plus platforms.
That seems like a reasonable investment to support more services and the reliability of the existing services.

 

[pompeyfan]

If you assume around 800m average new track + platform to redouble the line at every station between Salisbury and Exeter (stations like Templecombe would probably need closer to 500m of new track, given their proximity to existing loops) and bear in mind that you might need staggered platforms at some of them to avoid encroaching on housing, there are 6 stations that fit the criteria - therefore we're looking at approximately 3 miles of new track plus platforms.
That seems like a reasonable investment to support more services and the reliability of the existing services.

Stations like Templecombe would need a lift etc installed. The current platform stands in the previous down line, with the old up platform now a volunteer garden but out of operational use.

 

[Zomboid]

At Templecombe specifically it may be possible to build a path to the old up platform from adjacent to the high street bridge, but in general that could be an issue.

I wouldn't necessarily suggest putting loops at every station (though I won't say that it's a wholly bad idea!), but the single platform adjacent to a loop arrangement isn't the optimum for slick operations and could be improved upon.

 

[brad465]

At Templecombe specifically it may be possible to build a path to the old up platform from adjacent to the high street bridge, but in general that could be an issue.

I wouldn't necessarily suggest putting loops at every station (though I won't say that it's a wholly bad idea!), but the single platform adjacent to a loop arrangement isn't the optimum for slick operations and could be improved upon.

If trying to cut costs for two-platform accessibility, Crewkerne would be a good station to redouble, as a road bridge goes directly over the site, so maybe a ramp could be provided, and certainly no new footbridge.

For reliability improvements on the current service, I think the minimum redoubling requirements would be extending the Tisbury loop, Templecombe station and connecting Chard Junction and Axminster into one long dynamic loop. To enhance the service though, especially at the Exeter end, redoubling almost the entire line west of Axminster will be necessary, which is an ambition of the action group whose name I can't immediately recall.

 

[Sir Felix Pole]

I would start redoubling sections which can be done easily without the need for expensive second platforms / lifts and complex signalling / pointwork. So Tisbury Loop - Wilton South and Honiton - Axminster - Chard Junction. As the traffic builds up then the case for the more difficult sections can be made.

Between Pinhoe and Broad Clyst clearly the M5 overbridge is a big problem - the embankment is not substantial enough to tunnel through for a second track. It could be done by cut and cover in two halves - with traffic diverted to the other carriageway. Recent resurfacing / central reservation work nearby has seen this arrangement in place so it is not impossible.

 

[Zomboid]

I would start redoubling sections which can be done easily without the need for expensive second platforms / lifts and complex signalling / pointwork. So Tisbury Loop - Wilton South and Honiton - Axminster - Chard Junction. As the traffic builds up then the case for the more difficult sections can be made.

Between Pinhoe and Broad Clyst clearly the M5 overbridge is a big problem - the embankment is not substantial enough to tunnel through for a second track. It could be done by cut and cover in two halves - with traffic diverted to the other carriageway. Recent resurfacing / central reservation work nearby has seen this arrangement in place so it is not impossible.

Are those sections in the right place to help make the service reliable? Do down trains typically get held waiting for an up at Tisbury?

If they do, then the loop needs to go further to the west, and doubling towards Salisbury won't help. OTOH, if it's up trains getting delayed waiting for a down then extending towards Salisbury will help.

 

[30907]

Are those sections in the right place to help make the service reliable? Do down trains typically get held waiting for an up at Tisbury?

If they do, then the loop needs to go further to the west, and doubling towards Salisbury won't help. OTOH, if it's up trains getting delayed waiting for a down then extending towards Salisbury will help.

Looking at yesterday's handful of delays (which can mostly be traced back to a delay to the 0725 from Okehampton!) and at sectional times:
1. the margins at Axminster West and Pinhoe are minimal, Tisbury-Gillingham is also tight and Yeovil better.
2. the margins at Wilton and Templecombe are comfortable (+/- 10min)
3. Chard Jn is no longer much use and Honiton is awkward.

So redoubling East of Tisbury and at Templecombe are of little use - Tisbury station, unfortunately, is what's needed at that end.
With that done, an hourly down train could reach Axminster earlier, so extending the loop towards Honiton brings a benefit as it in turn gets trains to Pinhoe earlier. Pinhoe-M5 would further improve matters (to Broad Clyst even better but expensive).
Of course, for the Devon Metro you need a further loop, and to allow half-hourly throughout a loop at Hewish; you could consider extending Honiton loop westwards as well.

 

[Zomboid]

Pinhoe to the M5 is a colossal 300m, so there's nothing to be done there without fixing the bridge, at which point you may as well carry on to Cranbrook (only 2 miles).

Honiton to Axminster seems sensible, as I think there are also problems with using Honiton to pass GWR trains when they're diverted.

Then do a second platform at Tisbury and it should be a fair bit more resilient.

 

[devon_belle]

Anecdotally, someone told me SWR management has spent over £15m on their discontinuous electrification with 450s scheme. Money which could have been spent on redoubling several miles of the line! I have no source or way to verify this, though

Ditto Zomboid I think Tisbury platform should be the first intervention. At least trains won't be held in the countryside any more in that case.

 

[AlastairFraser]

Stations like Templecombe would need a lift etc installed. The current platform stands in the previous down line, with the old up platform now a volunteer garden but out of operational use.

Fair point. That shouldn't be a major cost in the scheme of things.

Anecdotally, someone told me SWR management has spent over £15m on their discontinuous electrification with 450s scheme. Money which could have been spent on redoubling several miles of the line! I have no source or way to verify this, though

The medium to long term savings from going 100% electric (training/unit maintenance/unit leasing) will probably save more than that tbh.

 

[PTR 444]

I would start redoubling sections which can be done easily without the need for expensive second platforms / lifts and complex signalling / pointwork. So Tisbury Loop - Wilton South and Honiton - Axminster - Chard Junction. As the traffic builds up then the case for the more difficult sections can be made.

Between Pinhoe and Broad Clyst clearly the M5 overbridge is a big problem - the embankment is not substantial enough to tunnel through for a second track. It could be done by cut and cover in two halves - with traffic diverted to the other carriageway. Recent resurfacing / central reservation work nearby has seen this arrangement in place so it is not impossible.

In all honesty, it was incredibly short sighted that the M5 overbridge was built without any provision for redoubling in the first place.

How feasible would it be to demolish the entire bridge and rebuild for double track?

 

[Zomboid]

Anecdotally, someone told me SWR management has spent over £15m on their discontinuous electrification with 450s scheme. Money which could have been spent on redoubling several miles of the line! I have no source or way to verify this, though

That may or may not be the case, but 159 replacement is a different problem to the provision of more resilient passing loops.

 

[Brubulus]

In all honesty, it was incredibly short sighted that the M5 overbridge was built without any provision for redoubling in the first place.

How feasible would it be to demolish the entire bridge and rebuild for double track?

Not particularly, it would be much cheaper to just leave it single track and have a loop after the bridge than to rebuild it.

 

[30907]

Pinhoe to the M5 is a colossal 300m, so there's nothing to be done there without fixing the bridge, at which point you may as well carry on to Cranbrook (only 2 miles).

Sorry, it looked further on the map - I agree there's no point in 300m

Honiton to Axminster seems sensible, as I think there are also problems with using Honiton to pass GWR trains when they're diverted.

I was only suggesting "towards" Honiton, not the whole 11 miles, though if the tunnel isn't a problem it could be cost-effective.

Then do a second platform at Tisbury and it should be a fair bit more resilient.

Looks to be the first step.

 

[Zomboid]

I was only suggesting "towards" Honiton, not the whole 11 miles, though if the tunnel isn't a problem it could be cost-effective

It might turn out to be, but if I'm remembering right about not being able to pass GWR trains at Honiton then solving that as well has extra benefits.

 

[30907]

It might turn out to be, but if I'm remembering right about not being able to pass GWR trains at Honiton then solving that as well has extra benefits.

Possibly.

Where you put the loops depends on the timetable option chosen - you could construct one around Gillingham and Honiton as crossing points(they are an hour apart, after all), with a new loop east of Crewkerne rather than at Hewish....

 

[Zomboid]

Though a loop east of Crewekerne would probably amount to an extension of the double track from Yeovil.

 

[Sir Felix Pole]

In all honesty, it was incredibly short sighted that the M5 overbridge was built without any provision for redoubling in the first place.

How feasible would it be to demolish the entire bridge and rebuild for double track?

The big problem with rebuilding the bridge is that the next junction north on the M5 is miles away at Cullompton - the old A38 just wouldn't be able to cope if you closed the motorway entirely. It would have to be done by by alternately closing each carriageway.

There is an overbridge over a lightly used lane a little to the north which would take a second track and has been looked at, but there are houses on the Exeter side which would require demolition.

The existing bridge would take gauntleted track which would save on pointwork if not signalling.

 

[30907]

Though a loop east of Crewekerne would probably amount to an extension of the double track from Yeovil.

Yes, probably.

 

[devon_belle]

The existing bridge would take gauntleted track which would save on pointwork if not signalling.

That's interesting - I hadn't heard of that option having been considered. Is there a precedent for that in the UK? I can't think of any examples besides on tram lines.