Would Electrification islands be better starting points than extensions?

[Meerkat]

The rise of main line bi-modes makes me wonder if, rather than extending electrification end on from current extremities it would be better to do islands around the cities/termini? Obviously you would have to do from nearest grid connection to the city first.
Means the bi-modes can accelerate on electric and regen brake into stops, and they aren’t running or idling in stations and built up areas.
Then you concentrate on that city’s metro services before filling in the gaps that aren’t so essential due to bi-modes.
An example would be going North from Kegworth grid connection into Nottingham, then Derby, plus extensions to enable 25kv/battery locals in that area.
Maybe Exeter. Electrifying up the hill to central would speed up, quieten down, and de-fume the city Centre (Waterloo-Exeter will get Express tri-modes soon right???), then electrify Exmouth-Paignton, and enough of Barnstaple-Axminster for battery bi-modes.
More obscure I notice there is a pylon line crossing the railway between Penzance and St Erth, so you could electrify from Hayle down taking diesel idling out of Penzance and the stabling, and convert the St Ives branch (got to be ideal for a 769 shuttle!)

The general aim is to get best value out of Intercity bi-modes and enable enough range for battery bi-mode metros/locals if full electric not practical, in order to clear out the pure diesels as soon as possible without needing too many new diesel bi-modes.
It would good for PR to get diesels out of the cities - dirty diesels running across the open countryside won’t be so obvious as a noisy rattler next to a crowded platform.

 

[Bald Rick]

The rise of main line bi-modes makes me wonder if, rather than extending electrification end on from current extremities it would be better to do islands around the cities/termini? Obviously you would have to do from nearest grid connection to the city first.
Means the bi-modes can accelerate on electric and regen brake into stops, and they aren’t running or idling in stations and built up areas.
Then you concentrate on that city’s metro services before filling in the gaps that aren’t so essential due to bi-modes.
An example would be going North from Kegworth grid connection into Nottingham, then Derby, plus extensions to enable 25kv/battery locals in that area.
Maybe Exeter. Electrifying up the hill to central would speed up, quieten down, and de-fume the city Centre (Waterloo-Exeter will get Express tri-modes soon right???), then electrify Exmouth-Paignton, and enough of Barnstaple-Axminster for battery bi-modes.
More obscure I notice there is a pylon line crossing the railway between Penzance and St Erth, so you could electrify from Hayle down taking diesel idling out of Penzance and the stabling, and convert the St Ives branch (got to be ideal for a 769 shuttle!)

The general aim is to get best value out of Intercity bi-modes and enable enough range for battery bi-mode metros/locals if full electric not practical, in order to clear out the pure diesels as soon as possible without needing too many new diesel bi-modes.
It would good for PR to get diesels out of the cities - dirty diesels running across the open countryside won’t be so obvious as a noisy rattler next to a crowded platform.

Islands, even tiny ones, still need a power supply, and that power supply needs to be future proofed. It would be disproportionally expensive.

 

[Meerkat]

Islands, even tiny ones, still need a power supply, and that power supply needs to be future proofed. It would be disproportionally expensive.

That is just an image issue though isn’t it - ie first section looks expensive? The supply will be needed eventually anyway, it just makes the later fill ins cheaper.

 

[edwin_m]

That is just an image issue though isn’t it - ie first section looks expensive? The supply will be needed eventually anyway, it just makes the later fill ins cheaper.

Not necessarily. You might have to provide a duplicated supply for the island to ensure that the power stays on if one feed is lost. That second feed might become unnecessary once the island is connected to the rest of the electrified network, which (depending on configuration) could provide the necessary backup on its own. Arguably you could avoid doing this if all the trains were bi-modes and just run on diesel if the power went off, but not if the island was big enough to warrant some electric-only units.

There's also the general point of not providing facilities earlier than necessary. Money and using resources used to provide a feeder that isn't really needed until later can't be spent on providing a possibly more valuable piece of electrification somewhere else.

 

[GRALISTAIR]

The rise of main line bi-modes makes me wonder if, rather than extending electrification end on from current extremities it would be better to do islands around the cities/termini? Obviously you would have to do from nearest grid connection to the city first.

I have thought long and hard about this and almost posted a thread about 3 months ago myself. I like the idea of islands then infill rather than just extensions. We actually need both. The long lead time items as well as costly are Grid Feeder points. Start with those!!!!

So on MML there is/will be a feeder at Braybrooke and then Kegworth and close by Chesterfield. Start with these!!
Corby - Kegworth and then Kegworth to Nottingham.
Chesterfield north to Sheffield then do rolling program to Doncaster, Leeds, York etc.
Chesterfield south to Nottingham.
THEN start all the infill

With Braybrooke in place you could even do Wigston South Glen Parva and to Nuneaton first if you so chose rather than Leicester. It might be less disruptive doing it this way too.

 

[tbtc]

I like the idea, in theory (though, obviously, I'm just an uninformed outsider with no knowledge of the costs or the practicalities of actually wiring up independent bits of railway).

There are a number of bits of railway with a high frequency where air pollution is a problem and some wiring could remove several diesel engines each hour, but the "branched" nature of the service pattern means it'll be complicated/expensive to do the whole lot.

Take the Moor Street/ Snow Hill line through Birmingham. Lots of DMUs polluting central Birmingham but (if the option is "pure DMU" vs "pure EMU") then you'd have to wire all the way from Worcester to Stratford on Avon to remove the local DMUs (ignoring Chiltern for a moment).

You'd avoid the need to tackle some "difficult" bits of electrification (bridges/ tunnels/ scenery), e.g. Haymarket to Dalmeney could remove half a dozen diesels through Edinburgh each hour without the complication of dealing with the Forth Bridge.

The air quality at Sheffield Midland is certainly an issue, locally, but little scope to replace diesels with electric any time soon. If anything, we are going in the wrong direction - e.g. wiring the GWML through the countryside but keeping diesel trains chugging away through Bristol/ Oxford!

 

[GRALISTAIR]

Agreed that Oxford Bristol and Bath need doing

 

[Philip Phlopp]

Electrification islands are a complex solution to design and implement. They're not for the faint hearted.

The first thing to remember is that battery EMUs are more expensive to procure and to operate. They're currently 25% more expensive in procurement costs and 10-15% less efficient to operate with penalties on track access charges, energy consumption, general maintenance. The battery pack has a life expectancy of around 7 years, someone, whether it's the TOC or the ROSCO will need to budget for their replacement around 5 times during the life of a unit.

The second thing to remember is that bi-mode diesel units are not particularly suited to running with electrification islands. Diesel engines like running at temperature and at steady revs, they do not like shutting down and starting up sequences, that will dramatically cut engine life, massively increase emissions and significantly increase costs.

With that in mind, the actual complication of intermittent electrification...

OLE does two things, not just one thing. This is a feature which most enthusiasts of electrification islands and battery trains forget. I know everybody knows it's the way of delivering 25kV AC to each electric train. The forgotten feature is that OLE is its own 25kV AC distribution grid. This feature means that the OLE can be fed from a diverse range of grid feeding stations, so should one feeding station have to be taken offline, the OLE can remain energised at 25kV and a train service can still operate (though it may not be the full timetabled service). That's equally true whether it's conventional or auto-transformer supply. It's normally thought of as two adjacent feeders covering for their next door neighbour but it can be quite a bit more complex than that.

Complex resilience feeding arrangements can see 25kV cable alongside the track in a trough (or indeed away from the track alignment) from one feeding station to another remote feeding station. This can be necessary in areas where isolation of the OLE would prevent cross-feeding from neighbouring grid feeding stations, or where the failure of one grid feeder can't be covered by neighbour feeders.

This resilience is ejected out of the window when you try and build electrification islands. There is no way at all, none, that we will run extensive lengths of 25kV cabling in troughs to connect the isolated feeding stations, so they can cross-feed to each other. Which raises the questions - how exactly do you specify battery EMU units with sufficient resilience to cope with feeder unavailability, and how do you design your service so the unavailability of a feeder doesn't stop the job. How do you deal with having to recharge your unit for twice as long at one point than intended ? How do you cope with additional demand on other feeders when one feeder is out of service, given we already have to reduce services in some circumstances when a feeder is offline on conventional 25kV OLE.

The next problem comes from the distances between feeding stations - where do you place them, and can you place them in the correct places for your service pattern. Heyrod is a good example, you would want to charge your battery units at Victoria, if the grid connection is remote to a termini, how exactly do you operate your service in an efficient and passenger friendly way that still permits charging en-route ? How many miles of OLE do you install and how does the length of OLE that you install match with battery capacity, charging requirements, line speed and electrified route mileage - basically, for the OLE that's available, can you charge the battery pack up sufficiently to get to the next OLE island ?

There are more problems, one that I like to remind people about is that battery units effectively take three bites of the cherry from a feeder - they're going to take some power to move the train when it's under OLE, they're going to take some power to charge the battery when it's under OLE, and they're not going to return any regenerated power back into the system balancing out demand. Mass is also against them and they'll use that extra bit of electricity to move the battery pack around too. The location of feeders and the way in which they will need to be specified to cope with extremely heavy loads in the event the maximum number of trains need the maximum amount of power all at the same time will almost certainly result in larger than necessary numbers of highly specified feeders. I would worry that the calculations will determine that we need 5 or 6 x 10MVA feeders for a given route comprised of small islands when normal OLE on the same route would be fed with 3 x 15MVA feeders (and I'd say - good luck finding grid connection points for more feeders than are really needed).

I know someone will post some nonsense about trickle charging batteries or some such nonsense - fine in practice, maybe even viable on a short branch with one train an hour. They don't come close to working out in the real world of disruptive, trespass incidents and a constant procession of trains all backed up in adjacent signal sections coming to the same station or sections of OLE and all needing to recharge their battery packs.

There's a very real risk that to provide a resilient electrification island system, there will be extremely high costs and it will not provide any level of value for money.

Oh, and how many times do I need to tell people, the Forth Bridge will be challenging but not particularly expensive to electrify. There are the tunnels in Fife that are much more concerning than the Forth and Tay Bridges. And there's semaphores between Dunblane and Perth to deal with before wiring can really commence there now. Focus on the bread and butter stuff, not the most complex and fanciful schemes that create a dozen problems for every problem they solve.

 

[GRALISTAIR]

@Philip Phlopp - god I have missed your input over the last 4 years. In my defense I think outside of the box hence my MML suggestions.

 

[Meerkat]

Electrification islands are a complex solution to design and implement. They're not for the faint hearted.

There’s always someone turning up with facts and boring reality to spoil the fun!
Thanks, very interesting. Electrification islands and 769s til the gaps are filled then!

 

[HSTEd]

Are there really any places left that would be islands?

And there's semaphores between Dunblane and Perth to deal with before wiring can really commence there now. Focus on the bread and butter stuff, not the most complex and fanciful schemes that create a dozen problems for every problem they solve.

I am as pro modernisation of signalling as anyone I can think of this forum......
But do the semaphores need replacing before electrification?

Haven't there been several examples of semaphores on 25kV lines in the UK in the past?

 

[London Trains]

Are there really any places left that would be islands?

Nottingham, Derby and Sheffield could all be depending on which lines are electrified first.

 

[Philip Phlopp]

@Philip Phlopp - god I have missed your input over the last 4 years. In my defense I think outside of the box hence my MML suggestions.

Electrification islands built around grid feeders which are designed for continuous electrification are the least worst (and only viable) option. The one thing that I never really understand when it's brought up, whether it's industry or enthusiasts, is what sort of obstacles one expects to exist that makes specifying expensive battery units financially prudent (I'm excluding the use of battery units to run beyond the extent of electrification, such as on Stranraer services).

The usual answers are bridges and tunnels, but looking at major obstacles on the Fife Circle, beyond the usual overnight and weekend engineering works for route clearance (bridges and stations) and the usual piling, mast erection, signalling works and such, the big job will be the track lowering and electrification of the tunnels which 10-15 weeks of closures (as a guess) clears. It's a major inconvenience for commuters but does a 15 week route closure justify the cost of a battery unit ?

 

[HSTEd]

Nottingham, Derby and Sheffield could all be depending on which lines are electrified first.

Nottingham and Derby don't really support high intensity metro operations.
And if Nottingham and Sheffield want electrified local services, I'm afraid the answer is probably 750Vdc overhead, not 25kV. (ie. tram extensions/tram trains)

 

[yorksrob]

I don't necessarily disagree with a concentration on city suburban services first. In reality, most of these wouldn't be "islands" anyway as most urban areas have a toehold on the electrified network already. For somewhere like Leeds, where you already have the beginings of an electrified suburban network, there could be a lot to be said for extending this to places like York and Harrogate etc.

 

[edwin_m]

Take the Moor Street/ Snow Hill line through Birmingham. Lots of DMUs polluting central Birmingham but (if the option is "pure DMU" vs "pure EMU") then you'd have to wire all the way from Worcester to Stratford on Avon to remove the local DMUs (ignoring Chiltern for a moment).

This one avoids the feeder issue to some extent because it can take a feed from the New Street routes to provide redundancy (assuming those aren't maxed out, and if they are then something probably needs doing about that in any case).

However it would need a large number of bi-mode units, on what may be the busiest remaining diesel-worked suburban service, frequently stopping and starting that extra weight.

Are there really any places left that would be islands?

Any that are would probably not be for long, assuming the government actually bites the bullet on decarbonization which will involve electrifying nearly every main line. So the extra cost of feeding islands would be doubly wasted as it would become abortive after a few years.

I am as pro modernisation of signalling as anyone I can think of this forum......
But do the semaphores need replacing before electrification?

Haven't there been several examples of semaphores on 25kV lines in the UK in the past?

There were a few, but very few. By the time you've worried about all the issues of sighting and electrical safety it's probably cheaper to put in a colour light instead. Current policy is not to mix the types in the same area so if you need to do one you have to do quite a few more. If you are referring to retention of absolute block with semaphores replaced by colour lights, this was done on quite a few routes when the money started to run out and still exists at Stockport and perhaps a few other places. But there aren't many routes with AB these days, and most of them would be well down the queue for electrification.

 

[MarkyT]

There were a few, but very few. By the time you've worried about all the issues of sighting and electrical safety it's probably cheaper to put in a colour light instead. Current policy is not to mix the types in the same area so if you need to do one you have to do quite a few more. If you are referring to retention of absolute block with semaphores replaced by colour lights, this was done on quite a few routes when the money started to run out and still exists at Stockport and perhaps a few other places. But there aren't many routes with AB these days, and most of them would be well down the queue for electrification.

Making existing semaphores electrically safe for maintenance staff to work on could be a big challenge. Theoretically possible but probably not worth it. Stockport is and until fairly recently Stafford was controlled by mechanical levers but in both cases under TCB regulations and with complex electrical controls that fully mimic those of 60s era relay interlockings, including sectional route locking, approach locking, timed approach releases for junctions, etc. These features are not always present at mechanical installations operated under AB, whether using semaphores or colour lights.

 

[Philip Phlopp]

Haven't there been several examples of semaphores on 25kV lines in the UK in the past?

Not with the current clearance regulations for new schemes.

 

[Meerkat]

Electrification islands built around grid feeders which are designed for continuous electrification are the least worst (and only viable) option. The one thing that I never really understand when it's brought up, whether it's industry or enthusiasts, is what sort of obstacles one expects to exist that makes specifying expensive battery units financially prudent (I'm excluding the use of battery units to run beyond the extent of electrification, such as on Stranraer services).

The usual answers are bridges and tunnels, but looking at major obstacles on the Fife Circle, beyond the usual overnight and weekend engineering works for route clearance (bridges and stations) and the usual piling, mast erection, signalling works and such, the big job will be the track lowering and electrification of the tunnels which 10-15 weeks of closures (as a guess) clears. It's a major inconvenience for commuters but does a 15 week route closure justify the cost of a battery unit ?

I don’t really like the battery trains but as part of the islands concept it means you can change services off diesel earlier (and then cascade the battery trains elsewhere when can be fully electric) and covers the services that are unlikely to be viable electrification schemes until much later, if at all.

 

[edwin_m]

I don’t really like the battery trains but as part of the islands concept it means you can change services off diesel earlier (and then cascade the battery trains elsewhere when can be fully electric) and covers the services that are unlikely to be viable electrification schemes until much later, if at all.

I think there may be rather few places where battery trains can be used now, as the gaps in electrification are often too long. So it may be a case of buying small batches as electrification opens up more routes that have unelectrified sections within the range of a battery train. This has the advantage that new batches can incorporate improvements in battery technology, but the problem that small batches are expensive and turn into microfleets. So best if the battery is an option to be fitted to a standard EMU platform, and even better if it can be retro-fitted and retro-removed (is that a word?) as needs change.

 

[HSTEd]

There were a few, but very few. By the time you've worried about all the issues of sighting and electrical safety it's probably cheaper to put in a colour light instead. Current policy is not to mix the types in the same area so if you need to do one you have to do quite a few more. If you are referring to retention of absolute block with semaphores replaced by colour lights, this was done on quite a few routes when the money started to run out and still exists at Stockport and perhaps a few other places. But there aren't many routes with AB these days, and most of them would be well down the queue for electrification.

I'd say it would be cheaper to wait for ETCS myself......

 

[MarkyT]

I think there may be rather few places where battery trains can be used now, as the gaps in electrification are often too long. So it may be a case of buying small batches as electrification opens up more routes that have unelectrified sections within the range of a battery train. This has the advantage that new batches can incorporate improvements in battery technology, but the problem that small batches are expensive and turn into microfleets. So best if the battery is an option to be fitted to a standard EMU platform, and even better if it can be retro-fitted and retro-removed (is that a word?) as needs change.

I believe a number of recent stock orders have a small traction battery capability for emergencies and depot shunting. The new T&W metro vehicles will have this as well as additional space allocated for a larger battery capable of making longer off-wire excursions if required in the future.

 

[The Planner]

This one avoids the feeder issue to some extent because it can take a feed from the New Street routes to provide redundancy (assuming those aren't maxed out, and if they are then something probably needs doing about that in any case).

I would doubt you could get much more on to the West Mids, Bromsgrove was effectively an extension lead.

 

[Philip Phlopp]

I don’t really like the battery trains but as part of the islands concept it means you can change services off diesel earlier (and then cascade the battery trains elsewhere when can be fully electric) and covers the services that are unlikely to be viable electrification schemes until much later, if at all.

I'm well aware (unfortunately) of the straw grasping going on to try and make battery trains work, but it's mainly incoherent nonsense being pushed by a variety of interested parties who stand to make quite some significant sums, primarily battery suppliers, rolling stock manufacturers and the 'old three' rolling stock companies who have been pushed out of the market by the new pension funds and investment group backed owning companies.

It's nothing more than a giant con to hide the costs of decarbonisation (and the cost of failing to electrify route mileage) off the state owned balance sheet at Network Rail. Which is so stupid given how cheap it is to borrow all of the money needed to electrify every last inch of the network right now (and has been for the last 12 years or so). It's particularly stupid given that the new entrants in the rolling stock market exist to generate far greater interest rate returns on deployed capital for their owners than is available by buying government debt (the very same debt that could be used to pay for network wide electrification). Typically pension funds and large investment bodies, and in purchasing battery electric units at a 25% premium, their deployed capital increases by 25%.

The idea of cascading rolling stock is fanciful too - to operate battery islands you will need hundreds of battery electric units, to operate beyond the wires services on route that will never be electrified, you'll need dozens of units. It does mean there will be procurement of battery electric units which will need to be converted to electric only operation, not the end of the world as long as this is done when the battery is life expired and due for replacement, but it does mean the structure of the unit will retain the modifications for battery packs, and it does mean some expensive electrical equipment needed to charge and control the battery pack will have to be skipped. That money isn't free and will need to come from somewhere. It's all going to be baked into the leasing cost of the unit long after the unit has been converted from battery electric to pure electric operation.

The additional risk of cascading is that for electric islands which will eventually be converted to full electric route mileage, train lengths will be considerably greater than those needed for beyond the wires services, think four car battery electric units for some of the electric islands (running in 8 car formations) versus 2 car units running beyond the wires.

 

[edwin_m]

I'd say it would be cheaper to wait for ETCS myself......

As someone pointed out a few weeks ago, there are all sorts of interactions between the large signaling program and the large electrification program that are both needed. This may result in some inefficiencies such as premature replacement of equipment and being able to wait for ETCS because the existing signaling isn't immunised.

 

[Chester1]

I know someone will post some nonsense about trickle charging batteries or some such nonsense - fine in practice, maybe even viable on a short branch with one train an hour. They don't come close to working out in the real world of disruptive, trespass incidents and a constant procession of trains all backed up in adjacent signal sections coming to the same station or sections of OLE and all needing to recharge their battery packs.

Are there any serious proposals for trickle charging batteries that don't fit into that category? Would it work for Windermere branch for instance?

 

[Energy]

Would it work for Windermere branch for instance?

Not the Windermere branch again. Battery 331s are planned for it but the best option is still just to electrify it, its only 10-15 miles long.

 

[Philip Phlopp]

As someone pointed out a few weeks ago, there are all sorts of interactions between the large signaling program and the large electrification program that are both needed. This may result in some inefficiencies such as premature replacement of equipment and being able to wait for ETCS because the existing signaling isn't immunised.

Probably me.

There's no easy way to do it, re-signal then electrify is optimum but won't always work out so the next best option is re-signal with ETCS compatible components on a minimal expenditure basis and then fully re-signal with ETCS when budget/workforce permits.

Are there any serious proposals for trickle charging batteries that don't fit into that category? Would it work for Windermere branch for instance?

None that I'm aware of, outside of this forum and a handful of others. It's very expensive because you've got two sets of battery packs and to provide the necessary resilience the trickle charged battery pack at, say, Windermere will be double or treble the size of the battery EMU battery pack so any disruption can be managed.

The direction of travel, as I sense it, is towards slightly more powerful grid connections using the static frequency converters ABB et al have developed in recent years, which can get a few hundred kWh of charge into a battery relatively quickly during a layover, but even that carries risk of disruption and insufficient layover times impacting then on return paths from places like Windermere.

Timetabling, pathing and flighting of service isn't my brand of witchcraft but there seems to be a very real risk that if a train is delayed by 10 minutes, needs a 15 minute layover to recharge, and has to depart 10 minutes late rather than on time, it may then accumulate further delays on the return journey and end up, after two or three trips, running in the path of a later service. And with the need to get a unit to a specific point to recharge, swapping diagrams won't necessarily be straightforward.

It's basically like putting in enough petrol in your car to get to the next petrol station and betting on being able to get there, no diversions, no major delays, no need to turn back more than half way through the journey. I dread to think how anybody will handle a broken rail 50 feet from the end of the platform at somewhere like Windermere, with a battery unit lacking enough electric to get back to the OLE or another charging point. Resilience then kicks in during design and battery packs end up getting massive, battery EMUs end up being four and five carriages long to carry all the batteries, they end up heavier than DMUs, track access charges skyrocket and someone on an internet forum complains why you point out stringing up some wire on some metal posts is so much easier.

 

[MarkyT]

There's no easy way to do it, re-signal then electrify is optimum but won't always work out so the next best option is re-signal with ETCS compatible components on a minimal expenditure basis and then fully re-signal with ETCS when budget/workforce permits.

All modern signalling schemes are specified to be 'ETCS ready'. Not difficult for processor based systems, and any signalling can be made compatible by various means and resourceful engineers could notionally even overlay some form of ETCS in semaphore areas if somebody really wanted to do it! That's not a recommendation by the way. The big issues with signalling electrification compatibility are train detection, point detection, lineside circuits and signal structures. In most simpler single and dual-track areas, signal structures are not usually a big issue, electrification or not, as simple straight post mountings can be employed. The Integrated Lightweight Signal product from Unipart Dorman doesn't even need any special screening for maintenance work near the wires, as the GRP post folds down to ground level whenever they need any attention. They require no routine maintenance anyway with a self-cleaning system incorporated in the lens and the lighting unit has a very long life (>10 years).

https://www.unipartrail.com/wp-content/uploads/2020/04/lightweight-signalling-handbook.pdf

Unipart Dorman’s range of Lightweight Signals provides a new generation of signalling by delivering increased safety through the reduction of staff exposure to risk both on track & working at height and by significantly reducing both installation and lifetime ownership costs.

Note: similar products are available from other manufacturers.