Design for a universal Battery EMU for the GB rail network

[billio]

The trouble with a single new battery fleet is that, as with some of the first generation dmus, it is quite possible that this whole fleet could turn out to be duds and have a short life before scrapping as the technology is progressing fairly fast and things tend to improve with time and experience.

Then it is possibly a good idea to separate the power-storage unit(s) from the passenger accommodation ?

 

[Nottingham59]

it is 109 miles from Crewe to Holyhead, well the train will need to travel at least 300 miles on battery to do teh round trip from Holyhead to Crewe

For that line, would it be enough to recharge at the coastal terminus? So could be served by the 120-miler proposed above.

You'd also need recharging capability at Bangor and Llandudno Junction to provide a safety margin, even if you didn't use it routinely, and to cope with reversing trains early.

The cost of providing 2-3 span lengths of non tensioned OLE, and a link to the local supply, will be relatively small.

Yes, agreed, but you might it at several places on the route, so cheapness is important.

I'm suggesting 750V DC with a dual-voltage pantograph, at the platforms. Especially if the national spec BEMU already has third rail capability. This avoids the expense of a major grid connection to cope with the phase imbalance inherent in 25kV AC, but would need a pretty thick conductor bar to handle the current.

 

[HSTEd]

They’ll be no need for a plug in charger under any scenario. There is no point having another charging system on the train when you have one already on the roof. The cost of providing 2-3 span lengths of non tensioned OLE, and a link to the local supply, will be relatively small.

The cost may be "relatively small" in railway terms, but it will be very high compared to provision of plug in chargers and equipment on the train to connect to them.
Literally millions of 350kW range chargers are going to end up being installed, they are already commodity items that you can buy, and will only become moreso.

Given one of the advantages of electrical charging points is they are much lower maintenance than fueling points, I feel it would be foolish not to at least take on the capability to use standard cheap (by railway standards) chargers that could be put anywhere desired, even if most charging takes place using an OLE based solution. Even cheaper, lower power compatible chargers could be inexpensively placed even in rural stations and provide useful top up charging capability during disruption.

For one thing charging in depots would be significantly simplified if it could involve no exposed conductors at all!

Given the on boarding charging equipment will weigh tens of kilos and cost hundreds of pounds, not doing so just seems "penny wise and pound foolish"

 

[Bald Rick]

Yes, agreed, but you might it at several places on the route, so cheapness is important.

You really don’t want traincrew having to plug things in/out at random places. There’s a well established method for getting electricity to a train.

I'm suggesting 750V DC with a dual-voltage pantograph, at the platforms.

But why? It’s not that difficult to get the same power in 25kVAC to the train vs 750v DC.

Literally millions of 350kW range chargers are going to end up being installed, they are already commodity items that you can buy, and will only become moreso.

There’s literally millions of 6 cylinder Diesel engines around, but they don’t work very well in a multi-mode train!

 

[HSTEd]

There’s literally millions of 6 cylinder Diesel engines around, but they don’t work very well in a multi-mode train!

Is there a problem with the QSK-19 fitted 73s?!

I will get my coat

 

[Nottingham59]

You really don’t want traincrew having to plug things in/out at random places.

Agree. I'm (now) suggesting pantograph connection for any recharging needed during the normal operations.

Any plug-in connection capability would be for emergency use or in sidings. Same as now, wherever trains use hotel power. But it would be a mistake in my view to have a BEMU specification that does not at least have a connector able to take CCS charging.

But why? It’s not that difficult to get the same power in 25kVAC to the train vs 750v DC.

We're talking about end-of-the-line recharging in some places where the local grid may not be very robust.

Recharging a 500kWh BEMU in 10 minutes will need 3 MW of power. Using 25kV AC OHLE requires all of that current to be supplied in a single phase. This creates huge load imbalances in the three-phase grid supply, which normally need a substantial and very expensive 400kV or 275kV connection to avoid degrading the local network. Or a Static Freqency Converter, which AIUI internally converts each phase to DC and then converts that DC back into single phase AC. See https://www.witpress.com/Secure/elibrary/papers/CR16/CR16019FU1.pdf

But you don't need AC to charge batteries. And I think DC would be cheaper.

To charge the batteries on the train, the 25kV AC would have to be rectified back into DC anyway. And if the BEMU were designed to be able to run on third rail, then the batteries will necessarily be configured to be able to accept charge from 750V DC supplies. (Which is why I suggest 750V, rather than 1500V DC which some tram systems use.)

And DC connections would also be compatible with using a local powerbank to deliver a fast charge to the BEMU. The powerbank batteries could then be trickle charged over several hours before the next unit arrived, like Vivaral use on the Greenford Shuttle. So even places like Whitby and Fishguard with feeble local grids could recharge trains.

But if it were cheaper to install 25kV AC OHLE, at every location you might want to recharge a BEMU, then by all means take that route.

 

[Bald Rick]

Recharging a 500kWh BEMU in 10 minutes will need 3 MW of power. Using 25kV AC OHLE requires all of that current to be supplied in a single phase. This creates huge load imbalances in the three-phase grid supply, which normally need a substantial and very expensive 400kV or 275kV connection to avoid degrading the local network. Or a Static Freqency Converter, which AIUI internally converts each phase to DC and then converts that DC back into single phase AC.

3MW of power is 3MW of power, whether it is AC or DC, and therefore the strain on the local grid will be the same.

You don’t need a 400kV or 275kV connection for 3MW. Plenty of connections much bigger than that off 133kV, and 3MW is, frankly, small beer (even single phase). Worst case is a trickle charged battery, and the extra cost is an inverter.

 

[Energy]

and 3MW is, frankly, small beer (even single phase).

3MW of a relatively small 33kv DNO feed is only 91 amps, very much possible with only 1 phase.

This avoids the expense of a major grid connection to cope with the phase imbalance inherent in 25kV AC, but would need a pretty thick conductor bar to handle the current.

The big feeder stations with major grid connections will be supplying a larger number of trains. Though a train recharging will take a lot more than a train maintaining its current speed.

 

[RobShipway]

For that line, would it be enough to recharge at the coastal terminus? So could be served by the 120-miler proposed above.

You'd also need recharging capability at Bangor and Llandudno Junction to provide a safety margin, even if you didn't use it routinely, and to cope with reversing trains early.

Yes, agreed, but you might it at several places on the route, so cheapness is important.

I'm suggesting 750V DC with a dual-voltage pantograph, at the platforms. Especially if the national spec BEMU already has third rail capability. This avoids the expense of a major grid connection to cope with the phase imbalance inherent in 25kV AC, but would need a pretty thick conductor bar to handle the current.

Why would you need the recharging capability, if by the time a train has say got to Crewe from travelling from London it has the capability to be travelling the distance from Crewe to Holyhead or Crewe to Llandudno Junction and be able to travel back without the need to have the batteries recharged.

Also, why would you want recharge capability at Bangor? I could understand Holyhead and Llandudno or Llandudno Junction, but why have one at Bangor? The recharging capability needs to be in a place where the train can be static for a good hour possibly to recharge. This would need to be done either in sidings near to the end destination or at the platform at the end destination.

So for instance, if you have a service that was worked by an all electric train from Crewe to Holyhead. You would have recharge capability possibly with an OHLE on platform 10 at Crewe and say platform 2 at Holyhead. But if the train can cover say 300 miles on recharge, then you would only need to do the recharging at Crewe as it is 109 miles each way.

As both Energy and Bald Rick have commented it could be done whether it is OHLE power or 3rd rail power doing the recharging of the batteries.

 

[Bald Rick]

The big feeder stations with major grid connections will be supplying a larger number of trains. Though a train recharging will take a lot more than a train maintaining its current speed.

A train recharging will have its power limited by the traction system, which would be likely to be in the same region as full power for the train itself.

3MW is a little less than what an 8 car 700 draws.

 

[Nottingham59]

But if the train can cover say 300 miles on recharge, then you would only need to do the recharging at Crewe as it is 109 miles each way.

Sure. If BEMUs had a range of 300 miles, then you wouldn't need end-of-line charging. I'd be surprised if the economics of that worked out, but it would be great news if it did as we could roll out BEMUs throughout the GB rail network, starting immediately.

A train recharging will have its power limited by the traction system, which would be likely to be in the same region as full power for the train itself.

Wouldn't it be better for a BEMU to be able to take charge at a faster rate than that? Why limit it to the capacity of the traction motors? Otherwise, the train would have to spend as long under the wires at it did off the wires to be sure of getting a full recharge. That would limit where you could use them.

== == ==
Both these comments underscore to me the need for national BEMU planning. The design of the trains, the cascade plan, recharging method and capacity, and where you choose to locate recharging infrastructure are inextricably linked. If the railway introduces BEMUs line by line (as we are in Merseyside and Greenford), then a lot of flexibility will be lost and the eventual cost will be much higher than it needs to be.

 

[Bald Rick]

Wouldn't it be better for a BEMU to be able to take charge at a faster rate than that? Why limit it to the capacity of the traction motors?

Not the traction motors, the traction system - the current collection, the cabling, the inverters, all the power electronics, the battery’s capacity to take charge, etc etc.

 

[ac6000cw]

Wouldn't it be better for a BEMU to be able to take charge at a faster rate than that? Why limit it to the capacity of the traction motors? Otherwise, the train would have to spend as long under the wires at it did off the wires to be sure of getting a full recharge. That would limit where you could use them.

No it wouldn't - the average power consumption of a train is much lower than the full power it has available for traction. On level track it's only likely to be using full power during acceleration after stops or speed restrictions. When rolling along at a constant speed it will be using just enough to compensate for the losses due to friction, aerodynamics and within the traction system. When coasting or stopped it's not using power for traction (just for the 'hotel' power loads), and during braking it will be regenerating power back into the batteries.

 

[Bald Rick]

Otherwise, the train would have to spend as long under the wires at it did off the wires to be sure of getting a full recharge. That would limit where you could use them.

No it wouldn't - the average power consumption of a train is much lower than the full power it has available for traction. On level track it's only likely to be using full power during acceleration after stops or speed restrictions. When rolling along at a constant speed it will be using just enough to compensate for the losses due to friction, aerodynamics and within the traction system. When coasting or stopped it's not using power for traction (just for the 'hotel' power loads), and during braking it will be regenerating power back into the batteries.

Quite. Although as a very general rule of thumb, having trains ‘on the juice’ for as long as they are off it (in a diagram) is a good rule of thumb for battery trains.

 

[Trainbike46]

What should be specification be for a national fleet of Battery EMU's to decarbonise the GB rail network?

It seems to me there is a need for Battery EMU's now, which can be deployed the many routes where there are currently diesels running under the wires. As electrification progresses, this fleet would be cascaded across the network, as new electrification projects open up new routes for battery trains, and the pure EMU's are brought in on those routes where infill electrification has filled in the gaps.

That seems like a good plan! specifically to start replacing those diesel trains that are nearing their end-of-life

This need to cascade means that a single national design of BEMU would be much better than many different models of BEMU ordered for specific lines which would then form an inflexible collection of micro fleets, not be easily moved elsewhere.

This is one possible national BEMU specification:

• 100 miles guaranteed range off the wires
• Four-car (4x25m) units, with end-corridor connection, like class 350's, to allow longer trains up to 12-car. Reconfigurable into 3 and 5 car versions (or 2 and 6 car if necessary, depending on platform lengths)

Aren't there gauging issues on some routes for coaches longer than 20m? In any case 25m seems very long for a universal unit

• Top speed 110mph, which I think is the fastest allowed with connecting doors, and without losing space in leading cars.
• OHLE (of course, for traction and recharging batteries under the wires)
• Provision to recharge at remote termini using plug-in 350kW CCS superchargers, as used for rapid charging of EV cars

Is this actually useful?

Parts that I'm not sure about. Any ideas or thoughts on these?

• Carriage doors at 1/3-2/3 spacing for speed of access?

That would probably be best for a universal unit

• Convertible to 750V DC third rail supply?

That depends on how universal you want them to be; there's definitely a use-case for this though

• Recharging at the platform, rather than having to go off to a depot to refuel? Might need charging cables on swing gantries rather than having cables trailing across the platform.
• Is 100 miles off the wires right? Would 50 mile range be enough?
• Should BEMU's have a range-extender diesel engine to give them effectively unlimited range? HVO fuel would be low-carbon, and a 100kW car engine would be enough to top up batteries.

Probably not needed for the purpose you describe

• Dual voltage OHLE to allow recharging at 1500V DC via the pantograph at termini without high-voltage grid connection.

Wouldn't it be better to set up a section at 25kV in this situation, so that it can form part of future electrification projects?

• Is a 4-car base too long? Would a 3-car standard be better? Would allow reconfiguration into 2 and 4-car units.

you would probably want a mix of 2, 3 and 4 car units, depending on where they would be deployed first

• Would BEMUs need megawatt recharge capability, as in the emerging CharIn MCS standard for trucks?

Charging via existing OHLE / 3rd rail would seem best

reading further down the thread I see some of the points I raised were already covered, please ignore if so!

120-MILE RANGE
The 120-miler would be useful in the in the South-West and Scotland where there are longer distances between existing electrification.

Scotland already has a decarbonisation plan in place so wouldn't take part in this national standard BEMU suggestion in any case

• Exeter is 120-130 from the limits of wiring at Basingstoke and at Newbury, and also 120 miles from Penzance. So electrifying 10-15 miles each side of Exeter would allow BEMUs to run Penzance-Exeter-Waterloo and Penzance-Exeter-Paddington, effectively covering all of the South West.

So you are suggesting replacing the 80x fleet? I don't think the case for swapping out bimodes with BEMUs is that strong tbh

 

[Nottingham59]

That seems like a good plan!

Thank you

gaugin issues on some routes for coaches longer than 20m?

I don't know. That's why i asked the question! Maybe there is a case for a 158 replacement with long range and relatively low top speed, and an 80x/Voyager/HST replacement with high speed and relatively short range off the wires?

better to set up a section at 25kV in this situation

It looks like it.

Scotland already has a decarbonisation plan in place

So maybe England and Wales should adopt whatever standard the Scots come up with?

So you are suggesting replacing the 80x fleet?

I've not got that far in my thinking. But we should only buy more 80x units, if future models will alllow swapping out some of the diesel engines and replacing them with batteries.
== == ==
When I started this discussion, I had envisaged BEMUs being used at the outer margins of a rolling electrification programme. Which would require cascades as the programme evolved.

What I have come to realise is that we should be designing the final OHLE layout for GB now. If 220-mile BEMUs are economically feasible, then almost everywhere in the country is accessible from OHLE that is already committed, with only a few top-up locations to be added (eg around Exeter, Inverness, and Shrewsbury). And even if 220-milers are too much today, then they probably will be feasible in 15 years time, and we should be planning for that now.

Not the traction motors, the traction system - the current collection, the cabling, the inverters, all the power electronics, the battery’s capacity to take charge, etc etc.

I don't think I agree with you. The faster you can add charge, the shorter the length of OHLE you need to deliver it. There will have great economic benefits on routes that have no electrification at present.

At the extreme case, if the BEMUs could take charge fast enough, then you could electrify the entire Chiltern Mainline by putting overhead conductor bars up at just Marylebone and Moor St. That would save perhaps two billion in OHLE electrification costs.

 

[HSTEd]

To charge a battery train that fast would require enormous traction systems aboard the train, especially if you tried to charge at 25kVac.

The traction transformer alone would be enormous, you'd probably have to take a leaf out of the book of the late Soviet Union and go to 6kV or even 12kV DC.

 

[Nottingham59]

The traction transformer alone would be enormous,

That's why DC recharging is cheaper.
I did say that wiring just the termini was the extreme case. In reality, you'd probably wire Stourbridge to Dorridge and Bicester to Princes Risborough with 25kV AC, which would still save a billion or more compared to wiring the whole route.

 

[Bald Rick]

I don't think I agree with you.

You don’t have to agree with me!

But you do have to comply with the laws of physics.

 

[Energy]

A train recharging will have its power limited by the traction system, which would be likely to be in the same region as full power for the train itself.

Good point. What I was trying to say was that a train coasting at say 60mph isn't at full power so a feeder station for battery recharging where its going to have 1 train at full power isn't going to be entirely comparable to a more normal feeder station serving many more trains but they won't all be at full power at the same time.

I don't think I agree with you. The faster you can add charge, the shorter the length of OHLE you need to deliver it. There will have great economic benefits on routes that have no electrification at present.

When you already have a feeder station etc. the cost of installing wires over a bit more track is going to be comparatively small. Sending that amount if power over OHLE is going to be expensive.

 

[Bald Rick]

What I was trying to say was that a train coasting at say 60mph isn't at full power so a feeder station for battery recharging where its going to have 1 train at full power isn't going to be entirely comparable to a more normal feeder station serving many more trains but they won't all be at full power at the same time.

Quite correct. Essentially, with battery trains you need (at a network level) the same amount of power put into the system as with full OLE. That assumes that the power losses through the battery charging / discharging over a short distance are roughly equivalent to those from distributing the power to a full OLE system over a long distances.

 

[Greybeard33]

So, if the battery trains are to recharge on existing electrified lines, you probably have to uprate the power supplies to those lines substantially, to cope with the extra trains all drawing full power all the time, not just during acceleration. This would likely be particularly costly in 3rd rail land, where lots of extra substations would be needed.

No such thing as a free lunch!

 

[Energy]

So, if the battery trains are to recharge on existing electrified lines, you probably have to uprate the power supplies to those lines substantially, to cope with the extra trains all drawing full power all the time, not just during acceleration. This would likely be particularly costly in 3rd rail land, where lots of extra substations would be needed.

No such thing as a free lunch!

It would probably need looking at individually. Some feeder stations will be already sized enough for battery trains.

 

[Basil Jet]

I think that the best solution might be to have removable batteries that are charged offline and then replaced when the train arrives at a station. That way you can charge them as slowly or quickly as you like, and you don't have to start extending layovers, because any station stop would be long enough to replace an empty battery with a full one. But I cannot imagine a physical configuration that doesn't have loads of moving parts and potential reliability issues.

 

[Bald Rick]

So, if the battery trains are to recharge on existing electrified lines, you probably have to uprate the power supplies to those lines substantially, to cope with the extra trains all drawing full power all the time, not just during acceleration. This would likely be particularly costly in 3rd rail land, where lots of extra substations would be needed.

No such thing as a free lunch!

In many cases yes. In third rail land - perhaps not so much, as there is so little diesel working (except freight) that the marginal increase is relatively small. Getting extra freight electrified is a big issue however.

 

[Nottingham59]

So, if the battery trains are to recharge on existing electrified lines, you probably have to uprate the power supplies to those lines substantially, to cope with the extra trains all drawing full power all the time, not just during acceleration. This would likely be particularly costly in 3rd rail land, where lots of extra substations would be needed.

No such thing as a free lunch!

It would depend, but there are not many unelectrified lines in third-rail land, except for long-distance trains arriving from the West. It would be a problem if a 100-miler arrived at Southampton from Bristol with no charge left, with only the run to Portsmouth and back to recharge. Trains from Exeter would be less of a problem, as they would have the full length of Basingstoke to Waterloo to recharge. I guess the BEMU specification should include software to manage the recharging profile, so that it drew extra current to recharge at those places where supply was available.

If modelling shows that wouldn't be enough, you might need to beef up the DC supply at terminating stations like Weymouth and Southampton, or even install 25kV OHLE around Salisbury to catch trains from Cardiff and from Exeter before they hit the third rail.

EDIT: Elsewhere, I think it is highly likely that current plans for new OHLE electrification are not specifying the capacity that they will need to cover bimode freight traction and recharging BEMUs in addition to EMU services.

 

[Greybeard33]

It would depend, but there are not many unelectrified lines in third-rail land, except for long-distance trains arriving from the West. It would be a problem if a 100-miler arrived at Southampton from Bristol with no charge left, with only the run to Portsmouth and back to recharge. Trains from Exeter would be less of a problem, as they would have the full length of Basingstoke to Waterloo to recharge. I guess the BEMU specification should include software to manage the recharging profile, so that it drew extra current to recharge at those places where supply was available.

If modelling shows that wouldn't be enough, you might need to beef up the DC supply at terminating stations like Weymouth and Southampton, or even install 25kV OHLE around Salisbury to catch trains from Cardiff and from Exeter before they hit the third rail.

EDIT: Elsewhere, I think it is highly likely that current plans for new OHLE electrification are not specifying the capacity that they will need to cover bimode freight traction and recharging BEMUs in addition to EMU services.

Might this be an issue with the mooted Merseyrail battery extensions, e.g. Bidston to Wrexham, Kirkby to Wigan, Hunts Cross to Warrington?

Of course the elephant in the room is how to decarbonise all the diesel over third rail freight in the South....

 

[Trainbike46]

Might this be an issue with the mooted Merseyrail battery extensions, e.g. Bidston to Wrexham, Kirkby to Wigan, Hunts Cross to Warrington?

I thought merseyrail recently uprated the power supply system, ahead of the new trains and route extensions?

Of course the elephant in the room is how to decarbonise all the diesel over third rail freight in the South....

It would probably involve some kind of trimodal freight locomotive (OHLE, 3rd rail & diesel/battery). Freight at night/off peak time might be easiest as there's fewer passenger trains around and therefore more spare capacity in the 3rd rail system at those times. Freight at peak times may be harder, and would probably require improving the power supply system, or revisiting the electric spine from Southampton!

 

[HSTEd]

The third rail specification already includes controlled voltage-sag, you simply have to programme the converters to avoid drawing current for battery recharging if the voltage sag has reached problematic levels. And conversely to draw as much power as possible for charging if the voltage has climbed above nominal values.

It's a few lines of code in the software these days.

Of course the elephant in the room is how to decarbonise all the diesel over third rail freight in the South....

Use Class 92s and pay a significant but not enormous sum of money for additional supply improvements......

A lot of people seem to want to spend billions on 25kV electrification to avoid £100m in power supply improvements.....

 

[Trainbike46]

Use Class 92s and pay a significant but not enormous sum of money for additional supply improvements......

Forgot about those! I'm surprised they don't get used more, given that they can run under both 3rd rail and OHLE.

Though for some routes a new locomotive capable of handling (short) stretches without electrification is probably necessary or desirable