Why do tramways use 750V DC instead of an equivalent AC Voltage?

[AidanCroft]

Mod note - split from Liverpool-Manchester electrification thread

A slightly off topic question - why do tramways use 750V DC instead of an equivalent AC Voltage?

I know that DC transmission has lower transmission losses but because of the cost of rectifiers 25kV AC is instead favoured.

Also, I understand why low Voltage must be used in a street environment but why not at AC?

Kind regards,

Aidan.

 

[Domh245]

A slightly off topic question - why do tramways use 750V DC instead of an equivalent AC Voltage?

I know that DC transmission has lower transmission losses but because of the cost of rectifiers 25kV AC is instead favoured.

Also, I understand why low Voltage must be used in a street environment but why not at AC?

Kind regards,

Aidan.

I suspect that it'll be to do with the fact that with DC, there is no need for the transformers, which means that it'll be a lot lighter, as well as being low-floored - it's difficult to sling a transformer under a tram, and you can't put it on the roof without substantial support. Instead, the feed from the pantograph can just go straight into the main bus-line to power the motors, ancillaries etc.

 

[AidanCroft]

I suspect that it'll be to do with the fact that with DC, there is no need for the transformers, which means that it'll be a lot lighter, as well as being low-floored - it's difficult to sling a transformer under a tram, and you can't put it on the roof without substantial support. Instead, the feed from the pantograph can just go straight into the main bus-line to power the motors, ancillaries etc.

Trams have to be low Voltage also because you can't have 25kV in streets but my confusion is why not have low Voltage AC instead of low Voltage DC?

Aidan.

 

[po8crg]

You can't have high voltage DC ever. 25kV DC would be insanely dangerous.

At lower voltages, DC means you don't need a rectifier in the vehicle, which keeps the vehicle lighter. This is much more important for light rail than for heavy rail.

 

[AidanCroft]

You can't have high voltage DC ever. 25kV DC would be insanely dangerous.

At lower voltages, DC means you don't need a rectifier in the vehicle, which keeps the vehicle lighter. This is much more important for light rail than for heavy rail.

Basically, I am visualising a light rail network where there are either AC to DC rectifier substations or AC to AC transformer substations. Why can't we have low Voltage AC-supplied trams that have AC motors - is it simply because the required rectifier/DC link/inverter set would take up all the space you mention?

As for third rail, obviously it has to be DC because AC would be unacceptable due to the skin effect.

But lastly, why can't we have high Voltage DC (HVDC) OHLE? It would have lower transmission losses - is it just the cost of all the rectifier substations offsetting the benefits of lower losses? Why do you suggest it is more dangerous?

Kind regards,

Aidan.

 

[ironstone11]

Basically, I am visualising a light rail network where there are either AC to DC rectifier substations or AC to AC transformer substations. Why can't we have low Voltage AC-supplied trams that have AC motors - is it simply because the required rectifier/DC link/inverter set would take up all the space you mention?

As for third rail, obviously it has to be DC because AC would be unacceptable due to the skin effect.

But lastly, why can't we have high Voltage DC (HVDC) OHLE? It would have lower transmission losses - is it just the cost of all the rectifier substations offsetting the benefits of lower losses? Why do you suggest it is more dangerous?

Kind regards,

Aidan.

I think you are making a false assumption here. The skin effect at 50Hz is negligible. What is the difference between 750V DC OHLE and third rail at 750V DC? No obvious difference in losses other than resistive as far as I can see.

If AC transmission is so lossy why do we straddle the country with AC power lines?

As an earlier reply pointed out, if you use AC for trams you have to carry a heavy transformer, which is just what do don't want.

 

[AidanCroft]

I think you are making a false assumption here. The skin effect at 50Hz is negligible. What is the difference between 750V DC OHLE and third rail at 750V DC? No obvious difference in losses other than resistive as far as I can see.

If AC transmission is so lossy why do we straddle the country with AC power lines?

As an earlier reply pointed out, if you use AC for trams you have to carry a heavy transformer, which is just what do don't want.

If we used low Voltage AC we wouldn't need the transformer but each tram would need a rectifier, which explains now why we use DC for tramways.

AC does have lower transmission losses. When it comes to pylon distribution we use AC as it's easier to step up and down. For very long distances HVDC is used because the cost of converting to DC is offset by reduced losses. Due to dielectric capacitance underwater cables of more than 50km MUST be HVDC.

I was wondering why we can't have HVDC instead of HVAC on our railway OHLE?

Aidan.

 

[edwin_m]

Basically, I am visualising a light rail network where there are either AC to DC rectifier substations or AC to AC transformer substations. Why can't we have low Voltage AC-supplied trams that have AC motors - is it simply because the required rectifier/DC link/inverter set would take up all the space you mention?

As for third rail, obviously it has to be DC because AC would be unacceptable due to the skin effect.

But lastly, why can't we have high Voltage DC (HVDC) OHLE? It would have lower transmission losses - is it just the cost of all the rectifier substations offsetting the benefits of lower losses? Why do you suggest it is more dangerous?

I think the reason trams are DC is because there was no real reason to do otherwise. You also have to consider that all the electrification systems were invented in an era when modern power electronics weren't available, and most railways and tramways have to maintain compatibility with what they have already.

Most electric railways have adopted AC because it allows a higher line voltage, which in turn can transmit much more power without huge losses in the supply. However trams can't use higher voltages in the street. Being fairly small and short-distance, power losses weren't a huge problem. Going to AC would have involved a lot more complication, as traction motors don't work so well on AC and (until semicondutor rectifiers in the 1960s) converting AC to DC needed equipment that was too cumbersome to carry around on a tram. Regenerative braking is also useful for trams, but would have been difficult on an AC supply with the technology of the early 20th century. Trams thus established themselves universally with a supply of a few hundred volts DC, and anyone wanting different would have had to buy non-standard technology which would have been more expensive and potentially unreliable.

As to why trains don't use high voltage DC: before power electronics there would have been no easy way to step it down on board the train to a voltage that the traction motors could take without frying them. Stepping it up in the feeder station would have been pretty difficult too. It's probably possible to have a HVDC traction system today, but it would be incompatible with all existing systems and again require non-standard equipment on each train. Nor am I convinced the traction losses would have been any lower on DC, as the skin effect isn't significant in an OLE wire at 50Hz and the quoted AC voltage is RMS which has the same heating effects as the same DC voltage.

 

[AM9]

If we used low Voltage AC we wouldn't need the transformer but each tram would need a rectifier, which explains now why we use DC for tramways.

AC does have lower transmission losses. When it comes to pylon distribution we use AC as it's easier to step up and down. For very long distances HVDC is used because the cost of converting to DC is offset by reduced losses. Due to dielectric capacitance underwater cables of more than 50km MUST be HVDC.

Aidan.

HVDC is also used to connect different AC networks. If the HVDC Cross-Channel interconnector was an AC link, the GB and French power supply systems would need to be synchronised. Not only would this be politically undesireable, it would also complicate re-connection as a minor phase slip would cause massive power surges through the interconnector when the circuit was remade.
For interconnectors over large land masses, e.g. the between large regional networks in the US, DC is essential as parallel paths via different routes could have vastly differing lengths.For example, with a 60Hz supply, a difference between two arms of a parallel link of less than 14 Km would result in a 1 degree phase difference. This would create massive error currents. Using DC for the links means that synchronisation is applied only to the inverters at each end of the links.

 

[HSTEd]

With new SiC power electronics you could easily have a ~30kV DC system.

You could size the voltage based on the peak voltage of the AC system, which would reduce losses and increase substation distances. Instead of 25kVac it would be 35kVdc - which could cut losses by half.

 

[Nym]

Mod note - split from Liverpool-Manchester electrification thread

A slightly off topic question - why do tramways use 750V DC instead of an equivalent AC Voltage?

I know that DC transmission has lower transmission losses but because of the cost of rectifiers 25kV AC is instead favoured.

Also, I understand why low Voltage must be used in a street environment but why not at AC?

Kind regards,

Aidan.

The standard railway transmission equipment results in DC having higher transmission losses than AC, due to lower voltages and the need for semiconductor (these days) devices as well as simple wound devices.

I suspect that it'll be to do with the fact that with DC, there is no need for the transformers, which means that it'll be a lot lighter, as well as being low-floored - it's difficult to sling a transformer under a tram, and you can't put it on the roof without substantial support. Instead, the feed from the pantograph can just go straight into the main bus-line to power the motors, ancillaries etc.

With the filtering required by modern drive systems of recent years this benefit it quickly disappearing (See class 455 conversion to AC drive for filtering requirements). However, for a small isolated system the required filtering is not as significant.

Basically, I am visualising a light rail network where there are either AC to DC rectifier substations or AC to AC transformer substations. Why can't we have low Voltage AC-supplied trams that have AC motors - is it simply because the required rectifier/DC link/inverter set would take up all the space you mention?

As for third rail, obviously it has to be DC because AC would be unacceptable due to the skin effect.

But lastly, why can't we have high Voltage DC (HVDC) OHLE? It would have lower transmission losses - is it just the cost of all the rectifier substations offsetting the benefits of lower losses? Why do you suggest it is more dangerous?

Kind regards,

Aidan.

You don't *need* to use a DC Link Variable Speed Drive, other types of drives, such as cycloconvertors are available provided you have more than one phase available. However, polyphase OHLE transmission is rare.

You can't have HVDC OHLE primarily (IMO) because of the lack of self commutation, so any arc will not self extinguish as it does with an AC supply.

Just have a look at the comparative size of DC contactors compared with AC contactors that operate at LV (let alone HV), for example, the 3TC4417-0AB4 series compared with the equivalent AC contactor.

With new SiC power electronics you could easily have a ~30kV DC system.

You could size the voltage based on the peak voltage of the AC system, which would reduce losses and increase substation distances. Instead of 25kVac it would be 35kVdc - which could cut losses by half.

Except life isn't that simple, the efficiency of power electronics compared with wound devices it just isn't there yet...

 

[Holly]

If you use low voltage then, for a certain power, you must use high current.
If you use high current and AC then because of the so-called skin effect the outer of the conducting cable or power rail must not be ferromagnetic (iron or steel) otherwise losses are great.
So if you use AC you must use Copper or Aluminium (which may have a steel core for strength but not steel on the outside).
Conversely, if you use DC then the third rail or overhead wire may be made of steel, which is cheaper.

 

[JamesRowden]

I suspect that the answer to this question is simply that 750V DC is safer if the overhead-line were to fall and come into contact with a human underneath. AC electricity at a frequency of 60Hz is more dangerous to humans than DC at the same voltage because 60Hz is similar to the frequency of the electrical signals that human bodies use to transfer data. This is dangerous because it could subsequently interfere with the bodies functions. I know this because I was told this while doing the safety assessment for my PhD project to produce a DC power system.

 

[Taunton]

Tramways, and early electric railways, used dc supply and particularly dc motors because they were straightforward technology, which ac is not. What was that dynamo-powered (or even lemon powered) light bulb circuit we built from first principles for GCSE Physics in school? DC of course. Some early town public electricity supply was even DC at the start.

The highest power dc rail systems are 3,000 v, used by a number of countries which developed their systems in the 1930s. Belgium, Italy and Russia are among these, their trains have motors permanently in series. Britain at that time went for 1,500 v for several projects, as did France, Netherlands, and elsewhere. Most rail dc systems with overhead as opposed to conductor rail used these higher voltages.

750 v on tramways and 3rd rail suburban systems (it started at about 550 v and slowly crept up over the years) does have the advantage that it can be directly applied to straightforward motors. High voltage AC is not everything it's cracked up to be. You need less lineside substations, but instead you get substations in each train; a 12-car unit has three transformer/rectifier units (ie substations) in it. Tramways use overhead wires purely for safety reasons on public roads, if they are fully fenced (eg the DLR in London) they go for 3rd rail, easier to maintain.

Notably BART in San Francisco, a wide-ranging outer suburban system, which researched every engineering aspect when it was being designed from scratch in the 1960s, going for a non-standard wide 5'6" gauge for example (stupid; it means they can't hire in standard permanent way equipment etc) went for 1,000 v DC (another non-standard) third rail.

One system I've never understood is the 3-phase, as used in Germany, Austria, Switzerland, Scandinavia etc (at 15 Kv, 16 2/3 Hz). I don't understand how you can have three phase with only one supply wire, so if one of you electrical engineers here can explain, I would be grateful, also what are its benefits and disbenefits. As I understand it the current is supplied to the motors as AC and is not rectified. Some of the 3-phase systems like the Jungfraujoch mountain railway have two wires - I might understand three, but how do two give 3-phase, and why do they need a second one anyway. Incidetally, the Jungfrau is 1,125 v 3-phase AC, which is getting down to the AC voltage levels the original question started with. I suspect the fact that it's only 6 miles long and generates its own power so doesn't need lineside substations has something to do with it.

 

[HSTEd]

Except life isn't that simple, the efficiency of power electronics compared with wound devices it just isn't there yet...

Solid State Transformer research seems to be advancing at a rapid clip these days.
Got a lot of work suggesting higher efficiencies than wound devices are now possible with semiconductor ones these days. At least in the 11kV-400V regime.

There is also the matter that a series ganged MOSFET stack, even with all the fancy driving circuits that are currently necessary, is drastically lighter than a traction transformer.
Which will save energy which might lead to interesting effects in overall end to end efficiency.

Definitely think we will see 380Vdc auxiliaries start to appear on trains though, no matter what happens with traction circuits.

You can't have HVDC OHLE primarily (IMO) because of the lack of self commutation, so any arc will not self extinguish as it does with an AC supply.

Just have a look at the comparative size of DC contactors compared with AC contactors that operate at LV (let alone HV), for example, the 3TC4417-0AB4 series compared with the equivalent AC contactor.

Also derivatives of that ABB HVDC hybrid circuit breaker appear to be slowly reducing the contact erosion problems with DC contactor. Although they have not yet managed to overcome the size issues.
Improved semiconductors means this is likely to happen sooner rather than later.
Just look at an SiC device compared to a similar voltage Si one.

Some of the 3-phase systems like the Jungfraujoch mountain railway have two wires - I might understand three, but how do two give 3-phase, and why do they need a second one anyway. I

With a clever array of transformers (the Scott-T system) you can convert a balanced three phase load into a balanced 'two-phase load'.
A two phase system normally has four conductors to provide two isolated circuits but I see no conceptual reason you could not use another isolating transformer (or just rearrange the previous ones) to allow two live conductors with the third conductor being provided by the running rails as in normal systems.

 

[edwin_m]

One system I've never understood is the 3-phase, as used in Germany, Austria, Switzerland, Scandinavia etc (at 15 Kv, 16 2/3 Hz). I don't understand how you can have three phase with only one supply wire, so if one of you electrical engineers here can explain, I would be grateful, also what are its benefits and disbenefits. As I understand it the current is supplied to the motors as AC and is not rectified. Some of the 3-phase systems like the Jungfraujoch mountain railway have two wires - I might understand three, but how do two give 3-phase, and why do they need a second one anyway. Incidetally, the Jungfrau is 1,125 v 3-phase AC, which is getting down to the AC voltage levels the original question started with. I suspect the fact that it's only 6 miles long and generates its own power so doesn't need lineside substations has something to do with it.

The 15kV system used on the main lines in these countries is single phase. I assume on the two-wire routes the rails are effectively the third phase and they don't bother with a neutral conductor because the loads will always be balanced.

 

[AM9]

The 15kV system used on the main lines in these countries is single phase. I assume on the two-wire routes the rails are effectively the third phase and they don't bother with a neutral conductor because the loads will always be balanced.

That's true as the circuit is run in a delta configuration. In that case the voltage between any two of the phase connections is the square root of 3, i.e. approx 1.732 of the voltage to neutral. The voltage stated for the Jungfraujoch line is 1125v. Assuming that to be the phase to phase RMS voltage, then each phase will be near to 650V, - a voltage that is practical for a traction motor. If the loads are reasonable balanced, there is virtual neutral voltage in which no net current flows therefore no conductor is required.
A benefit of the system is that currents are lower for a given power needing lighter OHLE, (although it is a twin wire system). Regeneration is easily performed as the motors run sychronously with the supply at a fixed speed on decent.

 

[edwin_m]

That's true as the circuit is run in a delta configuration. In that case the voltage between any two of the phase connections is the square root of 3, i.e. approx 1.732 of the voltage to neutral. The voltage stated for the Jungfraujoch line is 1125v. Assuming that to be the phase to phase RMS voltage, then each phase will be near to 650V, - a voltage that is practical for a traction motor. If the loads are reasonable balanced, there is virtual neutral voltage in which no net current flows therefore no conductor is required.
A benefit of the system is that currents are lower for a given power needing lighter OHLE, (although it is a twin wire system). Regeneration is easily performed as the motors run sychronously with the supply at a fixed speed on decent.

Yes I assumed it would be something along those lines. I assume if every train has three-phase motors the load will always be pretty much perfectly balanced unless a phase is lost somewhere.

 

[MarkyT]

The 15kV system used on the main lines in these countries is single phase.

The 16 2/3 Hz frequency was a compromise dating from the early 20th century. It was still transformable with acceptable losses, whilst also being low enough to use in early series wound 'universal' motors without too much destructive commutator flashover or excessive eddy current losses, both of which would have been a major problem at 50Hz or 60Hz. Hence the traction units did not need to carry a rectifier, which especially in the early days were fragile, heavy and expensive mercury arc units, yet the railway gained the advantages of a higher transmission voltage. In many places electric railways built their own generating stations so could specify a different frequency. Elsewhere rotary converters could be employed easily, with the frequency being exactly 1/3 that of standard 50Hz supplies.

 

[edwin_m]

The 16 2/3 Hz frequency was a compromise dating from the early 20th century. It was still transformable with acceptable losses, whilst also being low enough to use in early series wound 'universal' motors without too much destructive commutator flashover or excessive eddy current losses, both of which would have been a major problem at 50Hz or 60Hz. Hence the traction units did not need to carry a rectifier, which especially in the early days were fragile, heavy and expensive mercury arc units, yet the railway gained the advantages of a higher transmission voltage. In many places electric railways built their own generating stations so could specify a different frequency. Elsewhere rotary converters could be employed easily, with the frequency being exactly 1/3 that of standard 50Hz supplies.

This also illustrates the problems tramways would have had if they had adopted AC power in the early years, without getting the benefits of a higher voltage. And having chosen DC then, there was no good reason to change later.

 

[341o2]

I suspect that the answer to this question is simply that 750V DC is safer if the overhead-line were to fall and come into contact with a human underneath. AC electricity at a frequency of 60Hz is more dangerous to humans than DC at the same voltage because 60Hz is similar to the frequency of the electrical signals that human bodies use to transfer data. This is dangerous because it could subsequently interfere with the bodies functions. I know this because I was told this while doing the safety assessment for my PhD project to produce a DC power system.

to expand this, from my training by BR regarding the GN suburban electrification, we were advised not to get within 3 feet ( I suppose 1m today) of the wires because the current can arc across and once 25kv hits you, you're a crisp. This happened to a steam lcomotive fireman who went onto the loco's tender with a pricker at Preston

 

[stuartl]

AC electricity at a frequency of 60Hz is more dangerous to humans than DC at the same voltage because 60Hz is similar to the frequency of the electrical signals that human bodies use to transfer data

I've never heard of this and anyway the mains frequency in UK is 50Hz. DC is often said to be more dangerous than ac due to the hold on effect. If you get hold of an electric cable your muscles contract and you can't let go. With ac the voltage goes to zero every 20ms so giving you a slight chance to let go.

 

[talltim]

Of course the Sheffield Tram-Train will be capable of running on 750V DC and 25kV AC.
Presumably there must be a bit of 'give' with regards to voltage, especially on DC systems. A line I have been researching recently, the Clarens-Chailly-Blonay in Switzerland was a typical Swiss 'slightly more than a tramway' and used 750V DC. However at Clarens it ran on the lines of the Vevey–Montreux–Chillon–Villeneuve tramway which used 600V DC and at Fontainivent it shared the station with the Montreux–Oberland Bernois railway which uses 800V DC. I suspect the voltage was chosen as a halfway house.
The MOB shares a station at Chamby with the museum line of the Blonay-Chamby museum which is actually owned by the Chemins de fer Electriques Veveysans which uses 900V. The stock owned by the museum comes from lines across Switzerland, obviously anything that originally used AC or a very different DC voltage must be modified, but I would imagine most just copes with the higher or lower voltage.

 

[455driver]

If we used low Voltage AC we wouldn't need the transformer but each tram would need a rectifier, which explains now why we use DC for tramways.

Aidan.

All the AC Desiros have a transformer because the AC voltage isnot smooth enough so it is converted to DC and then back to (smooth) AC.

 

[Taunton]

I would imagine most just copes with the higher or lower voltage.

True generally.

An example is at Ventimiglia station, border between France and Italy on the Mediterranean. French line is 25Kv AC, while Italian is 3,000v DC. However much French rolling stock is dual voltage for 1,500v DC as well, as this is extensively used in other parts of France, so Ventimiglia station and the approaches is a little island of 1,500v DC, French trains change voltage in their normal manner on the approach, while Italian trains run around the station area quite satisfactorily at half their normal voltage.

 

[Domh245]

All the AC Desiros have a transformer because the AC voltage isnot smooth enough so it is converted to DC and then back to (smooth) AC.

Is that why they make that weird high pitched sound, not the traction inverter warble but the screeching kind of sound from the PTSOL?

 

[snowball]

AC electricity at a frequency of 60Hz is more dangerous to humans than DC at the same voltage because 60Hz is similar to the frequency of the electrical signals that human bodies use to transfer data. This is dangerous because it could subsequently interfere with the bodies functions. I know this because I was told this while doing the safety assessment for my PhD project to produce a DC power system.

I've never heard of this and anyway the mains frequency in UK is 50Hz. DC is often said to be more dangerous than ac due to the hold on effect. If you get hold of an electric cable your muscles contract and you can't let go. With ac the voltage goes to zero every 20ms so giving you a slight chance to let go.

The distinction between 50Hz and 60Hz is unimportant in this context. Both are more dangerous than significantly higher or lower frequencies. I once saw a graph of danger against frequency. It had a maximum somewhere in the 50-60 Hz region but it was not a sharp maximum.

I can't remember whether the graph extended far enough to include DC.

One wonders how the graph was produced!!

 

[Bletchleyite]

You can't have high voltage DC ever. 25kV DC would be insanely dangerous.

It wouldn't be useful because you couldn't, without turning it back to AC, use a traditional wound transformer. But why more dangerous? 25kV of anything will fry you to a crisp if you get anywhere near it.

Neil
--- old post above --- --- new post below ---

I would imagine most just copes with the higher or lower voltage.

Yeah, older stock just goes slower on a lower voltage than intended, and can probably get away with a slightly higher one if the motors are only run in series, provided the insulation is up to it. (Isn't this how the LUL stock works on the shared Bakerloo section?)

Neil

 

[swt_passenger]

All the AC Desiros have a transformer because the AC voltage isnot smooth enough so it is converted to DC and then back to (smooth) AC.

The other main reason it is converted to DC is to form the input to an inverter/(converter) that has a 3 phase output to the motors at varying voltage and frequency (VVVF).

If that wasn't done you'd have huge issues with speed control, because a simple AC motor is a single speed device.

It is also why the traction equipment can be exactly the same on an AC and DC Desiro - in a block diagram with the power supply on the left, everything to the right of the 'DC link' would be the same on a 350 or 450.

Unless I've not kept up with the technology the production of the 3 phase VVVF supply from a single phase AC overhead is just too much to do in one big 'black box'.

 

[Bletchleyite]

Unless I've not kept up with the technology the production of the 3 phase VVVF supply from a single phase AC overhead is just too much to do in one big 'black box'.

And in any case some of the 350s are dual-voltage - the 350/1s can have (and have had, but removed again) shoegear fitted.

Neil