Discontinuous electrification to Penzance - Worked example for Cross-Country 80x Hitachi Trimode

[zwk500]

As far as the grid or frequency converter is concerned it remains a single phase load, it is just that the traction side of the transformer has a centre tapped winding rated for 50kV end-to-end. The rail return is coupled to the centre tap and the two contact wires are connected to the end windings.

OK, interesting.

Within an island, no.
But if you use AT feeding you can use an inexpensive distribution line to interconnect all the islands and feed them either with two SFCs at different sites, or with a single SFC and a single phase feed from an existing 25kV installation.

I still don't see why this is going to be worth doing above having fast-charge points with a battery supply trickle-fed from the local grid. 20 Miles just feels like too short a length of OLE to be worthwhile with. The 'discontinuous' electrification ideas should be much more about skipping over gaps or between actually electrified areas, or avoiding the need to wire a low-frequency branch, rather than providing random gaps in an otherwise electrified line. IMO the electrification network should aim to remain contiguous as far as possible.

Liverpool-Norwich, Birmingham-Stanstead or Oxford-Cambridge (although this should be fully electrified at it's proposed frequencies) would be examples of the former. Matlock and Windermere would be examples of the latter (although again, Windermere is arguably suitable for actual electrification).

 

[HSTEd]

It's probably not related to your straw man.

Electrically the circuit would work with the up line at +25 and the down at -25, but that would probably create a lot of complications in the real world. Double length neutral sections on crossovers and the like, which would get very complicated in station areas.

I don't think the neutrla sections would have to be double length, as don't regular neutral sections already have to resist ~43.5kV because they have to be able to withstand a situation where the ends of the neutral section are fed from different phases?

However I agree it would introduce complexity in station areas, so in that case you'd probably install a single neutral section in the station approach and put the station area in one phase.
But in a discontinuous scheme it might be able to select charging areas with minimal trackwork.

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

OK, interesting.

I still don't see why this is going to be worth doing above having fast-charge points with a battery supply trickle-fed from the local grid.

Well I guess the primary concern is that a fast charging point would require the train to be stopped to use it.
On these sorts of routes, even a six mile segment will provide several minutes of charging with the train still in motion, and may include a station stop.

The cornish main line (as in this example!) is slow, but I think forcing waits of several minutes mid journey is still likely to be unacceptable.

20 Miles just feels like too short a length of OLE to be worthwhile with. The 'discontinuous' electrification ideas should be much more about skipping over gaps or between actually electrified areas, or avoiding the need to wire a low-frequency branch, rather than providing random gaps in an otherwise electrified line. IMO the electrification network should aim to remain contiguous as far as possible.

Liverpool-Norwich, Birmingham-Stanstead or Oxford-Cambridge (although this should be fully electrified at it's proposed frequencies) would be examples of the former. Matlock and Windermere would be examples of the latter (although again, Windermere is arguably suitable for actual electrification).

I guess the fundamental question is how much we are willing to pay for continuous electrification.
Even optimistic estimates put the cost at £1.5m/track-km, if not more.

The cost for a distribution pole line is something like a tenth of that, and obviously you only need one for a route, not per track.

 

[Nottingham59]

I understand the St. Ives and Looe branches are not suitable (yet) for battery operation as the turnaround times are too tight.

I don't know about Looe, but the St Ives branch is only 4 miles long with a maximum line speed of 30mph. If a 577kWh battery gave 802207 a 43 miles range at mainline speeds, then a round trip to St Ives won't need more than a tenth of that. Recharging at 4MW will take less than 60 seconds.

I reckon a simple 25kV conductor bar at plaform 3 would be just fine. And if that's not enough, then just extend the conductor bar a few hundred metres along the branch.

 

[zwk500]

Well I guess the primary concern is that a fast charging point would require the train to be stopped to use it.
On these sorts of routes, even a six mile segment will provide several minutes of charging with the train still in motion, and may include a station stop.

The cornish main line (as in this example!) is slow, but I think forcing waits of several minutes mid journey is still likely to be unacceptable.

Assuming decent layovers at Penzance, you shouldn't need much more than a 1-2 minute hit on the way through which most trains will be doing anyway. The branches are more problematic with short turnrounds to keep the unit requirement down.

I guess the fundamental question is how much we are willing to pay for continuous electrification.
Even optimistic estimates put the cost at £1.5m/track-km, if not more.

The cost for a distribution pole line is something like a tenth of that, and obviously you only need one for a route, not per track.

I get the arguments, and maybe for Cornwall with some internal local services as well it might stretch a case?

 

[Zomboid]

. The branches are more problematic with short turnrounds to keep the unit requirement down

How would the (presumable) improvement in acceleration of a BEMU affect that? Might buy an extra minute or two at the terminal stations which could be enough to make a dent.

 

[zwk500]

How would the (presumable) improvement in acceleration of a BEMU affect that? Might buy an extra minute or two at the terminal stations which could be enough to make a dent.

Potentially, as well as general performance improvement of a new train. Although with low linespeeds there's not much acceleration to be done. However, the branch train doesn't have to run at 100% full every time it departs, so there's ways to manage it.

 

[HSTEd]

Assuming decent layovers at Penzance, you shouldn't need much more than a 1-2 minute hit on the way through which most trains will be doing anyway. The branches are more problematic with short turnrounds to keep the unit requirement down.

Penzance does allow trains to hide in the layovers, although obviously it limits the use of layover time for service recovery, which may matter on such a long route.

St Erth to Penzance electrification provides ~20 minutes of charging even with a near immediate turnaround.
If yuo need 20 minutes to charge, you won't be able to violate that for service recovery.

And what happens at intermediate charging points? It's a long way from Bristol Parkway or Newbury to Penzance at the end of the day.

I get the arguments, and maybe for Cornwall with some internal local services as well it might stretch a case?

COrnwall is probably the extreme case of discontinuous electrification, given its probably the largest expanse of rail system in England with no electrification in it.

Of course, in most of England 25kV supplies are never that far away to power a charging island!

 

[Nottingham59]

And what happens at intermediate charging points? It's a long way from Bristol Parkway or Newbury to Penzance at the end of the day.

That's why the case study spec includes a diesel engine. If the train manangement system predicts you're going to run out of charge before the next OHLE island, because of delays or diversions or just cold weather, then it should fire up the engine to extend the battery life.

Diesel has a much higher energy density than any battery, so it's the best way to carry an emergency store of energy around. Provided it is used only occasionally, it doesn't add much to the carbon footprint.

 

[Trainbike46]

That's why the case study spec includes a diesel engine. If the train manangement system predicts you're going to run out of charge before the next OHLE island, because of delays or diversions or just cold weather, then it should fire up the engine to extend the battery life.

Diesel has a much higher energy density than any battery, so it's the best way to carry an emergency store of energy around. Provided it is used only occasionally, it doesn't add much to the carbon footprint.

The disadvantage of carrying a diesel engine that is only used rarely, is that it involves a complex machine which requires maintenance, and adds costs. It also needs to be considered against the alternative of using that space and weight for a larger battery instead. I suspect in most cases, it would simply be easier and safer to go for the larger battery option, and do away with diesel entirely.

 

[HSTEd]

That's why the case study spec includes a diesel engine. If the train manangement system predicts you're going to run out of charge before the next OHLE island, because of delays or diversions or just cold weather, then it should fire up the engine to extend the battery life.

Diesel has a much higher energy density than any battery, so it's the best way to carry an emergency store of energy around. Provided it is used only occasionally, it doesn't add much to the carbon footprint.

Yes, but I was pointing out that static charging is not really suitable for these distances.

Sure, the train could make the jump with the diesel generator, but if there is no en route charging you've not really electrified the line have you!

 

[Zomboid]

Yes, in the automotive world, PHEVs and Range Extenders are really worst of both worlds because you've basically got two powertrains, and the same really applies to trains. Just have the 1.7MWh battery and you can go further between chargers or survive more disruption without introducing the complexity and maintenance needs of a diesel engine.

 

[MatthewHutton]

Yes, in the automotive world, PHEVs and Range Extenders are really worst of both worlds because you've basically got two powertrains, and the same really applies to trains. Just have the 1.7MWh battery and you can go further between chargers or survive more disruption without introducing the complexity and maintenance needs of a diesel engine.

The BMW i3 range extender (which is more similar to a train) had 1 power train and a motorbike engine to recharge the battery.

 

[Zomboid]

The BMW i3 range extender (which is more similar to a train) had 1 power train and a motorbike engine to recharge the battery.

Yes, so it's an electric car, but you've still got to carry and maintain an internal combustion engine. Electric traction needs much less maintenance than internal combustion, but now you've got to do both.

 

[zwk500]

And what happens at intermediate charging points? It's a long way from Bristol Parkway or Newbury to Penzance at the end of the day.

As I mentioned, there'd be OLE to Exeter in this scenario.

 

[HSTEd]

As I mentioned, there'd be OLE to Exeter in this scenario.

I guess the fundamental issue is that much electrification (190 route km for Newbury-Exeter, ignoring the via Bristol route for a moment) is probably enough electrification to allow virtually all remaining diesel services to go to battery operations, at least in England. Maybe even in wales too.
The scheme in this thread proposes ~51 route miles / ~82 route km

 

[Nottingham59]

that much electrification (190 route km for Newbury-Exeter, ignoring the via Bristol route for a moment) is probably enough electrification to allow virtually all remaining diesel services to go to battery operations

That's the other way to do it. But in the original post with BEMUs able to cross 64km gaps, you only need to wire the 18 miles from to Castle Cary to Fairwood Junction. Expresses transit this OHLE island in only 12 minutes, but the relatively short gaps each side mean they will not need a full recharge:

• Newbury - Fairwood Jn: 44 miles
• Castle Cary - Whiteball: 38 miles

I haven't checked the mileages, but that island also looks enough to batterify Weymouth-Bristol too.

 

[Zomboid]

I would really work backwards and look at where you get the biggest benefits of electrification, and I think the Devon banks is one answer to that. So what is around 64km London direction of Newton Abbott?

 

[ac6000cw]

Prompted by discussions upthread about the Devon banks (which are double track up-down-Totnes-up-down) and 25kV pole lines...

A slightly off-the-wall idea occurs that (in principle) you could just string catenary over the uphill direction sections of track and then swap it over to the other track at the summits and at the low point of Totnes (maybe with a bit of both-tracks overlap to allow trains to crest the summits fully on OHLE power). As the trains will be BEMU they could be charging to some extent on the climbs (depending on available power limits over what is needed for traction) and getting some more charging from the regenerative braking going downhill. It reduces the cost of the wiring e.g. like BR singled parts of the Ely - Kings Lynn line when it was electrified to reduce the cost (and considered singling parts of the Diss to Norwich route for AFAIK the same reason).

 

[Zomboid]

So having had a look at the mileages, I'd propose this for wiring to get XC trains to Penzance (from Bristol):
Bristol Parkway - WSM
Cogload Jn - Tiverton
Newton Abbott - Plymouth
Par - Truro
Penzance - St Ives (do the branch as well, using neutral sections under bridges to guide the pan but running off the battery if necessary).

Add Cogload to Castle Cary and Newbury to Bedwyn to allow the London trains to go over to battery.

And specify the trains to have enough capacity to do those jumps without going below about 35% charge.

 

[Nottingham59]

I would really work backwards and look at where you get the biggest benefits of electrification, and I think the Devon banks is one answer to that.

I wasn't so worried about the Devon Banks. The highest elevation near Ivybridge is 150m. Raising a train to that altitude takes the same energy as accelerating it to 120mph, so well within the capabilty of the case study 80x Trimode. And you get most of that back with regenerative braking on the down gradient.

(It makes much more difference to freight, of course. Getting over a 150m hill takes four times the energy as accelerating to 60mph, and freight trains are much heavier than the 300t 80x).

 

[Zomboid]

There's other trains in the area too, and helping those out is a consideration. With my suggestion then all the hills that seem to have well known names (Dainton, Rattery, Hemerdon, Whiteball/ Wellington) are wired and so there should be no issues with getting up and down them.

Freight in the area isn't really heavy of course, but there's also the probable local Devon services that'll be running on batteries as well as the long distance trains, which may not be as capable as an 80x.

I know very little of Cornish railways, but Par - Truro gives the Newquay to Falmouth service group a nice 20 mile (so at least 20 minute long) section in the middle of the route to charge, and helps the main line services as well.

What I'm aware of having missed out are
* The Looe branch, where I'm not sure what the solution is since there's nowhere that the train stops for any length of time - maybe a tram-train is actually the solution there since it's fully self contained, and 750V overheads will be a simpler prospect than anything high voltage,
* Gunnislake: Plymouth station has wires, and put a static charger at Bere Alston if necessary as the dwells will be a couple of minutes in each direction, but I suspect that won't actually be needed.
* The Exeter area: I suspect St David's to Honiton or Axminster will need something to help with probable Waterloo line BEMUs, so that could help in that area. I suppose that might be DC BEMUs, in which case the con rail would probably stop at Central, and local trains would want to be dual voltage BEMUs to take advantage of what's available.

I'm going beyond the initial question of getting an XC train to Penzance, but that's because the train service goes a long way beyond that.

I think that ideally if we're saying that XC and GWR are going to use 80x style BEMUs then continuous electrification from Bristol Parkway to Plymouth would be the target to aim for, and any interim islands should be based around the feeding necessary to achieve that. Which might mean an ATFS near Taunton (where there's a 400kV substation) and Totnes (another 400kV substation). Exeter also has a 400kV substation, which might be useful, but that's actually out near Cranbrook station on the Waterloo line.

 

[MarkyT]

I wasn't so worried about the Devon Banks. The highest elevation near Ivybridge is 150m. Raising a train to that altitude takes the same energy as accelerating it to 120mph, so well within the capabilty of the case study 80x Trimode. And you get most of that back with regenerative braking on the down gradient.

(It makes much more difference to freight, of course. Getting over a 150m hill takes four times the energy as accelerating to 60mph, and freight trains are much heavier than the 300t 80x).

Most modern passenger trains seem to handle the Devon Banks quite easily. 2+8 HSTs were pretty good too, as long as both power cars were working well. Engines scream at near full output, but they all make it without slowing to a crawl. An isolated bit of wiring in the area might be able to improve section times marginally, but even today, Voyagers in particular really seem to fly. If you didn't know that Dainton is one of the steepest gradients in the UK you likely wouldn't notice it from performance. Voyagers and 80xs seem to be typically timed for 3 1/2 minutes for the circa 5km from Newton Abbot West to Dainton.

Signalling between Newton Abbot and Totnes is designed to prevent any stop awaiting forward block clearance going uphill on the steepest sections. For example, in the down direction, after E80, the 2-aspect on the level next to the A380 dual carriageway just past Newton Abbot West, the next signal capable of showing a red, DM217, is 4.4km away at Dainton tunnel's portal, the summit. A similar arrangement applies ascending from Totnes to the summit tunnel in the up direction. The near-constant tight curvature of the line between Newton Abbot and Plymouth is the main constraint on speed, though there are some straighter sections, Hemerdon bank, for example.

There's other trains in the area too, and helping those out is a consideration. With my suggestion then all the hills that seem to have well known names (Dainton, Rattery, Hemerdon, Whiteball/ Wellington) are wired and so there should be no issues with getting up and down them.

Freight in the area isn't really heavy of course

Not in terms of overall tonnage, but some heavy aggregates trains have been scheduled to be split into two portions between Plymouth and Exeter in the loaded direction.

but there's also the probable local Devon services that'll be running on batteries as well as the long distance trains, which may not be as capable as an 80x.

New regional trains are likely to have similar off wire range requirements to suit the same gaps on the mainline segments.

The banks, while requiring a brief burst of high output going up, will also return some of that energy through regenerative braking on the way down the other side to control speed.

 

[anthony263]

So having had a look at the mileages, I'd propose this for wiring to get XC trains to Penzance (from Bristol):
Bristol Parkway - WSM
Cogload Jn - Tiverton
Newton Abbott - Plymouth
Par - Truro
Penzance - St Ives (do the branch as well, using neutral sections under bridges to guide the pan but running off the battery if necessary).

Add Cogload to Castle Cary and Newbury to Bedwyn to allow the London trains to go over to battery.

And specify the trains to have enough capacity to do those jumps without going below about 35% charge.

I'd wiire to Westbury that might allow some.freight to use electric locos so pathing them might be easier.

Wire Cardiff to Swansea too

 

[HSTEd]

I'd wiire to Westbury that might allow some.freight to use electric locos so pathing them might be easier.

Wire Cardiff to Swansea too

Those two schemes, if done conventionally, add up to something like 90 route miles.

That would get you enough electrification to end a large portion of passenger diesel operations in the UK.

I am very skeptical electrification to Westbury would lead to a meaningful conversion to freight operations to electric.
The 70 route kilometres from Newbury to Westbury alone could easily cost £350-650m at the prices being obtained today.

 

[Zomboid]

In the spirit of discontinuous electrification, I'll cut back the Cogload to Castle Cary bit and extend the Newbury to Bedwyn section to Westbury. The Westbury end can take a supply from Melksham.

If locos like 99s are used on the stone trains then actually a good chunk of them could be electrified, with a static charger installed at the quarries.

 

[ac6000cw]

If locos like 99s are used on the stone trains then actually a good chunk of them could be electrified, with a static charger installed at the quarries.

The 99s don't have batteries, they are OHLE plus diesel.

 

[zwk500]

Those two schemes, if done conventionally, add up to something like 90 route miles.

That would get you enough electrification to end a large portion of passenger diesel operations in the UK.

Which 90 route miles would lead to a 'large' proportion of passenger diesel operations ending in the UK?

 

[MarkyT]

I am very skeptical electrification to Westbury would lead to a meaningful conversion to freight operations to electric.

If new multimode locos are built with the ability to run on OHLE that is installed along significant parts of their route, why would they not use it?

The 99s don't have batteries, they are OHLE plus diesel.

It's plausible versions with a small battery as well as the diesel engine may be produced in the future, building on experience from the same manufacturer's cl.93, as that could avoid starting and using the engine for short off-wire movements, shunting and running cleanly through sensitive and urban areas with no OHLE, as defined by geofencing.

 

[Zomboid]

The 99s don't have batteries, they are OHLE plus diesel.

Ok, so no need for the charger at the quarry then! They've still got full Tractive Effort on diesel though, so could run to Westbury and go pan up.

 

[Nottingham59]

[Coventry - Didcot is] 62 miles. That's a pleasing coincidence.

Looking around the network at other gaps for possible batterification, it looks like 64 miles off the wires would be good specification for a national standard BEMU. For those gaps I have checked:

• Newark - Cleethorpes and Doncaster - Cleethorpes are both around 62 miles. So wiring Grimsby to Cleethorpes would cover much of Lincolnshire. A booster at Lincoln station would also cover the Joint Line Doncaster-Spalding-Peterborough
• Basingstoke - Exeter is 124 miles, so a mid-way recharge between Yeovil Junction and Templecombe would cover that route. And a booster at Salisbury would also enable Portsmouth - Cardiff.
• Ely - Norwich is 50 miles
• Willington - Stoke is only 30 miles
• Dore - Hazel Grove is about 40 miles
• CLC line (Hunts Cross - Manchester) is 25 miles
• Man Victoria to Leeds via Bradford will be around 50 miles
• Newcastle - Carlisle is 60 miles
• Carnforth-Whitehaven-Carlisle is 114 miles. Booster at Whitehaven
• York - Scarborough is 42 miles and Hull to Scarborough is 55. Wiring Seamer jn- Scarborough would cover both those.
• Skipton - Carnforth is 38 miles
• Wiring at Shrewsbury and Hereford would cover all the the Marches: Hereford-Bromsgrove is around 40 miles.

So as far as I can see, electifictions islands at Exeter, Plymouth, Penzance, Westbury, Salisbury, Yeovil, Hereford, Shrewsbury, Scarborough, Whitehaven, would enable the batterification of almost the entire network in England.

What lines have I missed?

EDIT: Chiltern mainline. Cotswold Line.