I think so too.
However - my electronics knowledge isn't that good but I'm struggling to visualise a circuit that somehow charges a battery at the same time as power is being drawn from it.
-
Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
Transport for Wales Class 231 / 756 FLIRTs
- Thread starter DaveHarries
- Start date
I think so too.
However - my electronics knowledge isn't that good but I'm struggling to visualise a circuit that somehow charges a battery at the same time as power is being drawn from it.
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R
RailUK Forums
The train obtains power from the overhead this is then converted and sent to the traction motors which power the train.
The motors are not run off the batteries all the time.
When the train is powered by battery only is the only time the motors will be powered by battery.
I'm an electronic engineer, so let me explain a bit: there are two possible architectures for a system like this where there is an external supply (the overhead line in this case), batteries, and a load (in this case the traction motors and the rest of the train power):
- "on-line" architecture: the 25kV goes through a transformer and rectifier to provide DC at battery voltage and is connected to the batteries. The traction converters (which convert battery DC to variable-frequency AC for the motors) then draw from the batteries. This approach guarantees that there's always power fed to the motors regardless of what happens. It's perfectly feasible to have a battery being both charged and discharged at the same time (think of the battery in a petrol car when the engine is running - there's charge coming from the alternator, but there's also power being drawn by ignition system and accessories like the headlights and car radio). This model is slightly less efficient but has high reliability and with modern electronic traction converters you're generally converting via DC anyway so there's no big loss of efficiency
- "off-line" architecture: in this model the battery is switched in and out as required. When 25kV is available the power directly feeds the traction converters; if we lower the pantograph a relay will switch over to the batteries. This will produce a momentary loss of power during the changeover, and the relay needs to be capable of coping with the sudden changeover (it's a mechanical component so will wear out).
I presume the 756s use the online architecture, but I haven't seen any technical documents confirming this.
- "on-line" architecture: the 25kV goes through a transformer and rectifier to provide DC at battery voltage and is connected to the batteries. The traction converters (which convert battery DC to variable-frequency AC for the motors) then draw from the batteries. This approach guarantees that there's always power fed to the motors regardless of what happens. It's perfectly feasible to have a battery being both charged and discharged at the same time (think of the battery in a petrol car when the engine is running - there's charge coming from the alternator, but there's also power being drawn by ignition system and accessories like the headlights and car radio). This model is slightly less efficient but has high reliability and with modern electronic traction converters you're generally converting via DC anyway so there's no big loss of efficiency
- "off-line" architecture: in this model the battery is switched in and out as required. When 25kV is available the power directly feeds the traction converters; if we lower the pantograph a relay will switch over to the batteries. This will produce a momentary loss of power during the changeover, and the relay needs to be capable of coping with the sudden changeover (it's a mechanical component so will wear out).
I presume the 756s use the online architecture, but I haven't seen any technical documents confirming this.
I understand how this configuration works, but is the battery really being charged and discharged simultaneously?
I would have thought it would be doing one or the other at a given time, depending on whether the power demand is greater or smaller than the external power (which in the case of OLE I would imagine would always be more than required).
I.e. if there's external power to spare, the excess power charges the battery. If there's insufficient power, the battery contributes additional power. Or does it not work like that?
When I see batteries in circuit diagrams they generally only have two terminals, not a pair of charging and a separate pair of discharging ones.
I'm probably missing something so an explanation would be appreciated.
Peter Sarf
Established Member
Logically the batteries would be CAPABLE of being charged or discharged at any given time. Thats "online architecture"
The batteries will be being discharged when the OHLE is not available.
The batteries will also be being discharged if the OHLE is supplying less than the traction motors etc are demanding.
The batteries will be being charged when the traction (etc) demand is less than the what the OHLE can supply.
I would expect the above to be achieved by the voltage across the battery vs the voltage supplied (via a DC conversion).
The only alternative is if there is some switching/relay isolating leading to "off-line architecture"
Dare I mention regenerative braking.
- Joined
- 13 Dec 2018
- Messages
- 4,119
Yes, I'm aware, the previous poster was asking something different, which is what I was trying to explain.
That, I think, is what I said. It's certainly what I meant.
It seems a bit misleading to me to describe this configuration as having the motors run off the batteries, though clearly it's different from the off-line architecture described above. I would have thought it somewhat rare for the OLE alone to be unable to produce enough power to accelerate a train, in which case in both architectures the motors are powered from the OLE not the battery when OLE power is available, and from the battery when it isn't.
Fwiw I believe it does always go via batteries. At the very least this is true in the case of the diesel engines. They're not directly tied to traction rather the traction runs off the batteries and the battery gets topped up by the diesel engine.
Having said that flirts are very modular and I imagine can be setup in many configurations. The 231s for example do have the diesels converting to electric and then directly powering the traction.
Having said that flirts are very modular and I imagine can be setup in many configurations. The 231s for example do have the diesels converting to electric and then directly powering the traction.
Peter Sarf
Established Member
Actually I think this is all deserving of a separate thread - How do OHLE, Diesel and Battery interact in a Bi-Mode / Tri-Mode ?.
Last edited:
Yes, that's how it works - the battery is acting as a reservoir and is topped up if more power is available than is consumed by the motors (until it reaches full charge, of course). It's not being charged and discharged simultaneously as such, but it is simultaneously connected to both the power source and the load.
If you're in diesel mode then the battery helps "average out" the load on the diesel engine, which saves a lot of fuel.
Of course (stating the obvious) on the South Wales Metro system the overhead line sections necessarily have to supply enough power to the train to both run the motors *and* charge the battery so that the train has enough battery capacity to get through the unpowered sections.
tfw756rider
Member
Thanks all for the replies, I'll look at them properly in a bit.
On a different subject though - I'm not sure if this has occurred to anyone in here or been discussed in here before, but following the discussion in the 197 thread of a claim in the 197 Wikipedia article that 197125 was named "The Cheltenham Spa Express" (which turned out to be false, with @sd0733 saying that it wouldn't make sense to name a 197 after a destination that will ultimately be served by 231s):
231001 has a Caerphilly name (Sultan) and 231005 has a Barry name (Gavin & Stacey), but those destinations will ultimately be served by 756s. Could, when appropriate, Sultan be transferred from 231001 to 756001 or 756101, and Gavin & Stacey be transferred from 231005 to 756005 or 756105?
It would eventually make more sense to have the Caerphilly and Barry names on 756s, rather than have those names going to Cheltenham (or even Bridgend, Ebbw Vale and Maesteg).
On a different subject though - I'm not sure if this has occurred to anyone in here or been discussed in here before, but following the discussion in the 197 thread of a claim in the 197 Wikipedia article that 197125 was named "The Cheltenham Spa Express" (which turned out to be false, with @sd0733 saying that it wouldn't make sense to name a 197 after a destination that will ultimately be served by 231s):
231001 has a Caerphilly name (Sultan) and 231005 has a Barry name (Gavin & Stacey), but those destinations will ultimately be served by 756s. Could, when appropriate, Sultan be transferred from 231001 to 756001 or 756101, and Gavin & Stacey be transferred from 231005 to 756005 or 756105?
It would eventually make more sense to have the Caerphilly and Barry names on 756s, rather than have those names going to Cheltenham (or even Bridgend, Ebbw Vale and Maesteg).
Last edited:
I did a short trip on a 231 yesterday and at each station it told passengers to move further back down the train to alight at the next station, though it was silent on the matter when it actually started announcing the next station.
I was in the 3rd coach out of 4.
I'd love to know why it's so difficult to get a passenger information system to provide information not disinformation.
For a long time the 231s were announcing that the next stop was Grangetown on approaching Central heading South. Maybe they still are.
I was in the 3rd coach out of 4.
I'd love to know why it's so difficult to get a passenger information system to provide information not disinformation.
For a long time the 231s were announcing that the next stop was Grangetown on approaching Central heading South. Maybe they still are.
Wrong, the OLE charges the batteries.
The batteries are in use at all times to power the motors.
There is an option to run the unit on OLE but we are not authorised to use this function.
Even when not under OLE and battery drops below 20% engine starts and charges the battery.
Exactly this.
The batteries just replace the engines (231) and power the unit.
756 does have the ability to run off the OLE direct, but this option is only to be used with authority.
There's a VCB that opens/closes in between the PES sections and train automatically alternates between charging/discharging. As a driver it's seamless and unnoticeable.
Last edited:
So it’s not a tri-mode train then like they describe, if it can’t be powered by pure OHL which is definition of an electric powered train
Like I said above, twice. They can run purely off the OLE. But we don't use them in that mode.
Trainbike46
Established Member
I have a feeling various people using the same terms mean very different things in this thread.
I would like to see a circuit diagram of whatever the arrangement is, if anyone is in a position to provide this? It would certainly help clear up what is happening.
That makes sense. one diesel engine would not be enough to power the train without speed restrictions, so you don't actually want to have completely empty batteries.
As it should be.
If the arrangement is in fact:
OHLE (AC Electric energy)->transformer etc.->DC battery charger (DC electric energy)->battery (chemical energy)->DC electric energy->motor control systems->electric motors->kinetic energy moving the train
This is how I would interpret "Never direct of OHLE". Such an arrangement would be insane (but possible?), because of the DC -> chemical energy -> DC step, wasting both energy and battery lifespan. Note how this is a very different arrangement from the "on-line" arrangement described by @MikePJ, where when both OHLE and demand from the motors is present, most energy would not be going to or from the battery.
Last edited:
RichardAsh1981
Member
- Joined
- 28 Oct 2017
- Messages
- 123
Given that a battery works by moving ions between the electrodes and back again, then clearly it can only have two terminals, used for both charging and discharging, if the ions are to make it back to where they started (you could truly separate the functions with an all-liquid flow battery, but it would remain pointless).
As trainbike46 points out, even in the online architecture, a good chunk of the energy never gets converted from electrical to chemical and back again (and there is no need).
I suspect what is behind the briefing Logger79 has received is that the units are capable of operating in a mode with the battery isolated, so that the OHL must meet the instantaneous traction demand, but this isn't authorised for use. There are various possible reasons for this, but the obvious one is that the OHL draw will fluctuate more when run like this and the network as a whole may not like that.
I would be surprised to find a fully online architecture with separate battery charger and traction converter in this application, because of the use of Lithium chemistry batteries. Unlike most older chemistries with fairly constant terminal voltages, the terminal voltage of a lithium cell changes a lot with state of charge. This means a lithium charger has to be a continuously variable voltage charger, which varies the voltage depending on the state of charge of the battery, which is inferred from how much charge current the battery takes. Add a continuously variable load in parallel with the battery and the normal charging algorithms won't work.
Obviously with the correct measurement points the relevant currents can be measured, but it is getting very complex, and requires the "battery charger" and "traction controller" to be integrated into a single system - which is why there won't be a block schematic which shows this, and the full board diagram will be very complex.
As trainbike46 points out, even in the online architecture, a good chunk of the energy never gets converted from electrical to chemical and back again (and there is no need).
I suspect what is behind the briefing Logger79 has received is that the units are capable of operating in a mode with the battery isolated, so that the OHL must meet the instantaneous traction demand, but this isn't authorised for use. There are various possible reasons for this, but the obvious one is that the OHL draw will fluctuate more when run like this and the network as a whole may not like that.
I would be surprised to find a fully online architecture with separate battery charger and traction converter in this application, because of the use of Lithium chemistry batteries. Unlike most older chemistries with fairly constant terminal voltages, the terminal voltage of a lithium cell changes a lot with state of charge. This means a lithium charger has to be a continuously variable voltage charger, which varies the voltage depending on the state of charge of the battery, which is inferred from how much charge current the battery takes. Add a continuously variable load in parallel with the battery and the normal charging algorithms won't work.
Obviously with the correct measurement points the relevant currents can be measured, but it is getting very complex, and requires the "battery charger" and "traction controller" to be integrated into a single system - which is why there won't be a block schematic which shows this, and the full board diagram will be very complex.
chargesmith
Member
756005 and 756007 running on TAM lines today.
Last edited:
I think when running under a discontinuous electrification scheme there's a more fundamental reason than that not to isolate the battery!
Trainbike46
Established Member
Yes, there would be an issue if the battery is isolated and therefore not charging when you hit the end of the OHLE!
tfw756rider
Member
756005 entering service is great news!
This now means that 19 (of 24) 756s have been in service.
2 (of 7) 3-coach 756s have been in service (756005/007)
All 17 4-coach 756s have been in service (756101/102/103/104/105/106/107/108/109/110/111/112/113/114/115/116/117)
==============================
The number of 756s still to enter service is down to 5, all of which are 3-coach units (756001/002/003/004/006)
==============================
The overall number of new Transport for Wales (TfW) trains that have been in service is now up to 95 (of 148):
39 (of 51) 197/0s
All 26 197/1s
All 11 231s
2 (of 7) 756/0s
All 17 756/1s
This leaves 53 new TfW trains still to enter service:
12 197/0s
36 398s
5 756/0s
Peter Sarf
Established Member
Getting tantalisingly close to all new trains being in action.
I reckon 398s last.
197s might be dragging on though. Cambrian might still be a stumbling block. Especially if there are too many 197s sidelined with problems so not a full fleet switch over possible on Cambriam coast.
756/0s should be a formality (only held up by progress of knitting I can guess).
56xx
Member
Diesel engine was running throughout the stop at Merthyr Tydfil today on 756 007. 1M44 - 1A52, about 12 minutes, not sure of before arrival and after departure, so battery must have been low?
756004 ran from Barry Tourist Railway to Canton early this morning, so perhaps it's next for a turn? https://www.realtimetrains.co.uk/service/gb-nr:V16346/2025-05-12/detailed#allox_id=0
chargesmith
Member
Remember the 756s are designed to run on the Vale of Glamorgan line which means no OLE from Cardiff to Bridgend which is a fair distance - I suspect the system is designed so that the diesel gen set runs when the train is idle as I doubt that the diesel system is capable of providing direct traction and purely tops up the battery packs. Could see it was running at Taffs Well on 109 today as I could see the reflection of an exhaust expelling something while we were stopped.
Markdvdman
Member
I thought the 756s are to be designated for Rhymney to Barry Island? Aren't the 231s allocated for VOG line?
TT-ONR-NRN
Veteran Member
Barry Island is the principal terminus of the Vale of Glamorgan line…
The 231s are currently allocated on it, yes, but they’re destined for Maesteg - Ebbw Vale and Cardiff - Cheltenham.
Markdvdman
Member
Surely you mean Barry? The island is a spur off on its own
tfw756rider
Member
In any case, the 231s were always eventually destined for the non-"Valleys lines" valleys lines (Ebbw Vale and Maesteg) and Gloucestershire circuit... hence why they're branded as Wales & Borders like the 197s, rather than (South Wales) Metro...
...with the Metro-branded 756s always eventually destined for the Rhymney and Vale of Glamorgan (VoG) lines.
Once the 398s are in service on the Treherbert/Aberdare/Merthyr lines, that will allow the 756s to move to the Rhymney and VoG lines, which will in turn allow the 231s to move to the Ebbw Vale / Maesteg / Gloucestershire circuit...
...which will in turn (!) allow the quantity of 197s used on that to make more 197s available for the main 197 routes (as in, the ones that include Pembrokeshire-Manchester, to give just one example).