Our two electrification systems

[swtandgw]

The problem with the 750V DC 3rd rail system is that it was designed for when trains didn't have much technology besides the obvious basics, but since the proliferation of Desiros, Junipers and Electrostars, the system is currently operating at its technical maximum, and despite their design speed, they are often limited in service to speeds no higher than 90mph. My ex-girlfriend's mother, who used to work for British Rail down in the Brighton region, told me that the system has long been overdue a replacement since these days, but Southern Railway/Region legacy prevailed, hence that we're stuck with the two systems at the moment, and possibly for the foreseeable future, unless more infrastructure money is pumped into the Southern Region lines, in order to allow for a universally-electrified system like what most countries in Europe have. Why we're so behind the others is absolutely beyond my savoir-faire in railways, even Germany are lightyears ahead in rail technology (probably mentions why Siemens tends to be the winner in terms of rolling stock IMO).

 

[HSTEd]

And quite a huge saving in costs! In 3rd rail land, I've read there is a substation on average every 2 miles (confirmation please), where as on the ECML and WCML it's an average of 25 miles. (oleman, can you confirm please?)

That alone will save a huge amount as there will be a lot less duplication, plus the staff that are no longer required to maintain the excess substations can be re-deployed elsewhere.

The problem with the comparison between the ECML, WCML and the southern region is that they arent really that busy (certainly the ECML isnt) and thus the substations are primarily limited by voltage drop rather than actual power demand.

Trying to supply the entire southern regional terminal region of London with one 25kV substation would be fun, the power draw requirements are absolutely insane.
Additionally noone has explained what would happen to routes like the Watford DC lines or the Northern City Line? Would these be closed or dual voltage stock and expertise be retained to service them?

The employment of 25kV power supplies would almost certainly require significant hardening work to be done to teh supply systems to counter the unbalanced phase effect that the 25kV system induces in the utility power supply.

As I said the changeover would likely be a nightmare.

I am not by any stretch of the imagination anti 25kV, but I do not believe that the benefits outweigh the risks and problems associated with changing over.

 

[ole man]

Most feeder stations are 25/30 miles apart, then you have TSC (Track Section Cabins) then you have MPTSC (Mid Point). These are used to keep the current at a acceptable level, the more traffic the closer the TSC AND MPTSC will be to each other.
Now we are starting to Install Auto-Transformer which gives the 25kv a another 25kv feed, this will be national wide soon, and is currently used on the congessed routes (MML south, West Coast South)

 

[HSTEd]

Most feeder stations are 25/30 miles apart, then you have TSC (Track Section Cabins) then you have MPTSC (Mid Point). These are used to keep the current at a acceptable level, the more traffic the closer the TSC AND MPTSC will be to each other.
Now we are starting to Install Auto-Transformer which gives the 25kv a another 25kv feed, this will be national wide soon, and is currently used on the congessed routes (MML south, West Coast South)

Has Network Rail actually commited to nationwide installment of Autotransformers? Last I heard they had baulked at the cost.

 

[Old Timer]

And quite a huge saving in costs! In 3rd rail land, I've read there is a substation on average every 2 miles (confirmation please), where as on the ECML and WCML it's an average of 25 miles. (oleman, can you confirm please?)

That alone will save a huge amount as there will be a lot less duplication, plus the staff that are no longer required to maintain the excess substations can be re-deployed elsewhere.

Historically that used to be the distance on OHL Electrified lines, however in recent times the power demands from trains have become much greater, as has the average frequency and length of trains (Number of units drawing power). This means that in future I would expect to see the Feeder Stations brought closer together as there is an upper limit in the amount of voltage that can be passed down the OHL system.

 

[tbtc]

If you can do the 25kV electrification work with no additional disruption or expense relative to the current 3rd rail being renewed as individual segments reach life expiry, and provide a fleet of dual voltage vehicles throughout the transistion

You can start putting up the masts now, in advance of a move to 25kV, so that there's no "gap" between using one system and the other.

As previously mentioned, most modern EMUs (in Third Rail land) are dual voltage - since the 319s - so they are intended to be converted to overhead wiring if/when needed.

3rd rail hasn't came to the end of its life

The infrastructure has a limited life though, and will need replacing at some stage. That would be the best time to replace it.

 

[jopsuk]

Even before the masts, I'd guess there's quite a lot of bridges and other structures that may require modification before wires can be installed. With the recent inporvements in the headroom in the tunnels at Southampton, would there be space for a conductor-rail in the roof as well?
What about the Brunel tunnel on the ELL- again, a rail rather than wires would seem the way it could happen?

As for train conversions- I wonder if even something such as the 465 and 466 would be that difficult to convert, borrowing somewhat from the 365 plans. By the time it is done the 455 and 456 will certainly be life-expired, and the 460s may or may not be part of 458s. Everything else has pantograph wells or even (in the case of some 375s) never-used pantographs.

 

[Bald Rick]

Has Network Rail actually commited to nationwide installment of Autotransformers? Last I heard they had baulked at the cost.

Not nationwide, just where appropriate. They're in and working on parts of the WCML and the MML south of Elstree. GEML will be equipped to Shenfield for Crossrail, fed by one mother of a supply point at Pudding Mill Lane.

 

[thelem]

Would it help to introduce 750V DC OHLE capable of running at 25kV AC? This would bring some of the benefits of OHLE now and would mean that when you were ready to change a route to 25kV you wouldn't need to erect miles of wires.

 

[ole man]

The WCML is getting done now, starting with Bletchley and MK area, then onto Stafford North to Carnforth area, the company doing this is Keltbray/Aspire Rail

 

[OxtedL]

Would it help to introduce 750V OHLE capable of running at 25kV? This would bring some of the benefits of OHLE now and would mean that when you were ready to change a route to 25kV you wouldn't need to erect miles of wires.

Not really, as precisely zero stock is equipped to use 750V OHLE (excluding trams ), whereas the newer ~50% can be adapted to use 25kV reasonably simply.

 

[Old Timer]

Would it help to introduce 750V OHLE capable of running at 25kV? This would bring some of the benefits of OHLE now and would mean that when you were ready to change a route to 25kV you wouldn't need to erect miles of wires.

It is a good thought in principal but I presume you mean 750V DC, which simply would not work.

The reason for the HV system is that it make stransmission over longer distances feasible.

DC equipment requires very heavy conductors because of the amount of current that would be required, and this equipment is not compatible with AC equipment.

By way of example I attach a copy of a 1500v DC pantograph from an EMU, just look at the size of the pantograph head, which is completely made of copper.

 

[WatcherZero]

Some early AC wires were overengineered in cable thickness and can be switched to DC but it depends on the age of the wires and not many wires that old still around.

 

[Old Timer]

Some early AC wires were overengineered in cable thickness and can be switched to DC but it depends on the age of the wires and not many wires that old still around.

The original GE OHL was fixed tension which was thus not a problem then - it is now - whereas new equipment has to be tensioned.

The cross-sectional diameter would still have to be far bigger than anything we have now and that would bring about problems with tensioning and support. Everything including the Structures and foundations would have to be redesigned to take this into account, which is no easy or cheap job.

 

[jopsuk]

anywaym, the internal wiring of the train would presumably have to be very different- conversion of DC OHL stock to AC OHL would be expensive. Conversion of what's currently set up for AC OHL to DC OHL and then back again? Madness.

 

[MK Tom]

So do people think that Waterloo-Exeter (for example) should be electrified as 25Kv AC overhead throughout, in addition to the existing third rail east of Basingstoke, or as 750V DC third rail throughout, or as half and half changing at Basingstoke?

 

[HSTEd]

So do people think that Waterloo-Exeter (for example) should be electrified as 25Kv AC overhead throughout, in addition to the existing third rail east of Basingstoke, or as 750V DC third rail throughout, or as half and half changing at Basingstoke?

Im of the opinion that while converting to 25kV as far east as possible would be preferable, converting at Basingstoke would cause problems with having to dual signal the section containing the station and Worting junction.

Probably most cost effective to extend third rail operation to Overton and then switchover there as its a nice twin track station with at most a couple of crossovers, reducing the neccesary signalling overhead in the dual electrified area. This is if the electrification is to be rapidly extended all the way to Exeter Central/St Davids.
If the electrification was only being extended to Salisbury I would support third rail since the length of the track we are talking about there would be rather shorter and the result would be a 25kV "island" tacked onto the third rail network.

EDIT:
There is also the Reading-Basingstoke line to consider, which would require Basingstoke to be dual signalled anyway, unless ofcourse we were willing to accept the XC express service being forced to stop somewhere between Reading and Basingstoke to allow a future EMU to change from 25kV to 750V.

Unless we were willing to remodel Basingstoke station to extend the Reading Bays into through platforms by demolishing the station building and part of the car park and making the train switchover there, while removing the connections that allow trains from Reading to reach the other platforms, preventing the signalling circuits concerning the main station area having to be dual electrified.

EDIT #2:
It occurs to me that dual electrification hardening for Basingstoke's signalling infrastructure would probably be less expensive than such an extensive rebuild so yes, if Reading-Basingstoke was commited at the same time as Basingstoke-Salisbury/Exeter then I would support 25kV with transistion at Basingstoke, otherwise see the post above.

 

[WatcherZero]

anywaym, the internal wiring of the train would presumably have to be very different- conversion of DC OHL stock to AC OHL would be expensive. Conversion of what's currently set up for AC OHL to DC OHL and then back again? Madness.

Its not that hard usually because the traction motors and onboard electrics usually operate around the lower voltage (roughly 600v) and the equipment to convert 25kv is already carried onboard as a normal design function of a 25kv AC vehicle or can be easily added to a 3rd rail vehicle built with roof space for pantographs.

 

[Nym]

^What he said...

When it comes on board a modern unit, it goes through a transformer to a DC link, and then through a 3phase control invertor for the motors, the DC voltage isn't nesseserally 750V though, and there isn't nesseserally one transformer either, Pendos have several dotted throughout (*hence the 25kV bus on top).

Converting from AC to DC or DC to AC would simply mean the addition of a transformer, or a DC-DC convertor, (The latter is expensive).

 

[ushawk]

So do people think that Waterloo-Exeter (for example) should be electrified as 25Kv AC overhead throughout, in addition to the existing third rail east of Basingstoke, or as 750V DC third rail throughout, or as half and half changing at Basingstoke?

Id say Basingstoke to Salisbury 3rd rail (along with the "Salisbury 6") then Salisbury to Exeter OHL. For Basingstoke to Reading, i think OHL into the bay platforms, but for services not terminating, they should be able to change over from AC to DC and the other way around just north of Basingstoke, so there isnt any need to have wires all the way into the station making a mess and so the train doesnt have to wait as long in the station.

 

[HSTEd]

^What he said...

When it comes on board a modern unit, it goes through a transformer to a DC link, and then through a 3phase control invertor for the motors, the DC voltage isn't nesseserally 750V though, and there isn't nesseserally one transformer either, Pendos have several dotted throughout (*hence the 25kV bus on top).

Converting from AC to DC or DC to AC would simply mean the addition of a transformer, or a DC-DC convertor, (The latter is expensive).

DC-DC converters are nowhere near as expensive as they used to be, especially if it is a step down converter.
Modern trains have very high voltage traction busses however, with the AGV aiming for (apparently, according to Alstom) 3.3kV DC, which requires something a little more sophisticated than a chopper to run from 1.5kV.

EDIT: Does this mean that all the transformer vehicle for the 22x bimode project would need to run on third rail as well would be a chopper to turn the 750V DC into ~750V AC which could be fed into the main transformer using a very thick auxiliary winding?

 

[WatcherZero]

Ive heard the modifications for the 22x bi-mode project would include creating the major modification of a high voltage bus along the length of the vehicle.

 

[Nym]

DC-DC converters are nowhere near as expensive as they used to be, especially if it is a step down converter.
Modern trains have very high voltage traction busses however, with the AGV aiming for (apparently, according to Alstom) 3.3kV DC, which requires something a little more sophisticated than a chopper to run from 1.5kV.

EDIT: Does this mean that all the transformer vehicle for the 22x bimode project would need to run on third rail as well would be a chopper to turn the 750V DC into ~750V AC which could be fed into the main transformer using a very thick auxiliary winding?

Don't know the spesifics of the 22x series but IIRC it's a HVDC bus, so the chances are if they are adding a transformer it will just be a reasonably simple rectifier with an insanely large choke..

 

[HSTEd]

Don't know the spesifics of the 22x series but IIRC it's a HVDC bus, so the chances are if they are adding a transformer it will just be a reasonably simple rectifier with an insanely large choke..

Yeah, and if the bus is a higher voltage than the third rail you would have to invert it (square wave is fine for this) and feed it into the transformer in order to get the correct voltage out unless you want to fit a boost converter which could get nuts with the size of the capacitors involved.

All this needs is a massively thick winding on the transformer and an extra chopper, which isnt too much extra weight methinks.

 

[Nym]

If I was stepping up in a high power application like this I'd proberbly use a high frequency ZVS DC-DC convertor.

DC (750V) - High Frequency IGBT (MOSFET would be thick) ZVS Chocked Invertor (Square Wave) - HF Transformer - HV Rectifier (Thyristor Controlled to prevent feedback, when not desired), Output Choke and Capacitor - HVDC (Whatever voltage you want)

You're drawing about... well... 500kW per carriage, thats a rather large peice of power electronics there, but IGBTs and using an HF transformer cuts down on the non-switching component sizes, or if you are being clever, you could intergrate it into the AC transformer at HF to have one multiwiding transformer, but the former is more an off the shelf design. And DC Buses are easyer to feed in and out of than AC ones, don't need phase syncronising.

 

[apk55]

DC-DC converters are nowhere near as expensive as they used to be, especially if it is a step down converter.
Modern trains have very high voltage traction busses however, with the AGV aiming for (apparently, according to Alstom) 3.3kV DC, which requires something a little more sophisticated than a chopper to run from 1.5kV.

EDIT: Does this mean that all the transformer vehicle for the 22x bimode project would need to run on third rail as well would be a chopper to turn the 750V DC into ~750V AC which could be fed into the main transformer using a very thick auxiliary winding?

All designs are a compromise, with parameters being optimized for minimum cost, weight and efficiency. A higher voltage means thiner conductors, but more insulation, and creapage and clearance distances.

For motors making them higher voltage may mean a larger motor frame. The higher voltage means more turns of thiner conductor per slot, and the insulation must be thicker to withstand the higher voltage. Therefore more of the slot is filled with insulation so there is less conductor. And as electrical insulation is also a good thermal insulator the conductors can not be cooled as easily so the current density in the conductors may need to be reduced. The optimum voltage for a motor depends on its rating, for multiple units 750 V is probably a good compromise but for higher power locomotives 1.5KV is often used.

IGBT's are now available in voltage ratings of 6KV so I can understand why a 3KV DC link is being considered as only a single device is required. It is not easy to connect devices in series but they may be paralleled easily as they have a positive temperature coefficient.

3 phase motor inverters can be run off a variable voltage (although the control will become more complex as modulation would need to be varied in response to voltage.) Running on reduced voltage would not reduce the maximum starting and low speed torque, only the maximum power (or torque at speed) which would fall in line with the voltage. This is probably the situation for many third rail units where the inverter can be supplied direct from the line via nothing more than a filter and circuit breaker. It is probably cheaper, easier, lighter and more efficient to do this than use a voltage converter on the input.

On an AC unit (and DEMU,s) the DC link voltage can chosen at any value so possibly they can be a higher voltage. However many multiple units probably use common parts with DC units so probably 750V is still used. However the transformer rectifier is normally replaced with a 4 quadrant converter to improve the power and form factor of the current waveform.