Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Electric Trains (in general)

  1. Jun 19, 2016 #1

    I thought I remember reading that transmitting power over 800km's is passed the break even point for DC vs AC. And I read that modern rail is starting to use 25KV AC, but aren't many networks bigger than that, specifically in NSW Australia, the according to wikipedia it's 815km, I'm not sure if that is from top to toe, I assume it is. So wouldn't it actually be more efficient to have it as DC? From a transmission perspective.

    Side point: [The benefit of transmitting HV is less current needed for necessary power. I know that the power loss of transmitting is in the I^2 term, which is of importance, there is also parasitic inductance for AC transmission. But side point, I know the electrons in AC just sitting there oscillating, and the slow but present electron drift of DC doesn't effect this does it? Like adding to the the current in power losses? I'd have thought the mode of AC or DC wouldn't effect the loss in I^2.]

    I am aware that series DC motors are great because of the torque characteristic.
    In the old days, couldn't they still use a wound rotor induction motor, just just vary the resistance of the rotor to get maximum torque on start up? And so use AC?

    Now that we can control the frequency easily with solid-state technology, is the switch to AC trains just based on the fact that induction motors are really robust and you don't need to replace commutators all the time?

  2. jcsd
  3. Jun 19, 2016 #2


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    There is no single full length of 815 km. That distance covers ALL the small side lines etc
    the electrified part of the network doesn't stretch too far beyond Sydney.
    1) the north line from Sydney to Newcastle is around 120km
    2) the west line from Sydney to Lithgow is around 115 km
    3) the south line from Sydney to Kiama is around 100km

    back in the 1990's there were plans to electrify the Newcastle to ~ Muswellbrook ( ie up through the Hunter Valley) using 25kVAC but the plans were abandoned

    ... but don't forget there can be multiple power feeds into the system along its length that negate the problem you assumed existed

    Last edited: Jun 19, 2016
  4. Jun 19, 2016 #3
    the electrons in AC are indeed drifting back and forth but they themselves don't add to the "current in power loss" as you say , current is current there isn't anything that adds to it or takes away from it , if you have a given length of wire with resistance X , and you apply a potential to it's both ends, you have a current through that wire , which is directly proportional to the applied voltage.This is by Ohms law.

    from reading your post I got the feeling you are asking whether DC has no ohmic loss over a length of wire is that correct?
    If that is what you thought , it is wrong.
    The only reason at long distance DC is preferred over AC is because AC has additional inductive loss also higher capacitive losses , also DC has fewer conductors necessary.But it's distribution is much more complicated.
  5. Jun 19, 2016 #4


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    going by that response ... I'm not sure if you actually realise the topic of the discussion ?

  6. Jun 19, 2016 #5
    Ah, interesting figures, now that I see them it seems obvious. But what "problem I assumed existed" is negated by multiple power feeds? About my question as to why no AC over DC? Because I thought the issue may have been commutators:
    But playing devils advocate, if there was a long electrified line over 800kms (which probably would be very rare) would it not be more economical to have it DC because it's past the 'break even point' with AC?

    Sort of, not really what I meant.
    I haven't done any electrical theory in over 7 months, but then last night on researching this topic I was thinking about AC current, and how weird a concept that is. Even though we say there is a current traveling from ac generation source to load, or for that matter from fault back to neutral. (I've done Electric Systems, and there was a significant +,- & zero sequence component of fault studies in Power And Machiens) So I took it for granted that there actually is an AC current, but I can't imagine there actually being, if the drift velocity of an electron is 8cm an hour and they're oscillating at 50Hz, then they're not going anywhere. As I envision it in a copper wire the electrons in a wire stop the electric field from leaving out the sides, and allow the electric field to be transmitted.
    So I know that there are inductive losses from long distance over head power transmission, and if I'm running a large motor from the supply that is generated miles away then that current that the AC motor draws is going to add to the I^2.R losses of the transmission line, I'm just trying to understand how if the electrons aren't really moving, also I'm trying to compare that to an equivalent DC I^2.R loss where the electrons in the transmission line do actually move.
    The only thing I can conjure to explain this is that the electrons are actually moving very fast, because If I'm running an inverter then surely electrons are actually jumping over the junctions very fast.

    thanks very much!
  7. Jun 19, 2016 #6


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    read my comment again you seemed to miss the point as you even queried it .......

    I said earlier .......
    there wont be single end fed very long electrified train systems .... I can almost guarantee they will have multiple DC feedpoints along the system
    as is done here in Australia

    As far as I'm aware, the drift is still occurring over and above the oscillation, so there will still be a net movement in a given direction

    no, they still have the same slow drift velocity

    what junctions ???

  8. Jun 19, 2016 #7
    I'm afraid you're going to have to humour me and spell it out, I have to many things running through my head (red herrings). I don't see what multiple feeds has to do with mitigating parasitic inductance? Does it have anything to do with increasing efficiency of transmission line (by breaking it up)? At first when I re-read what you wrote I thought 'oh because it has to be less than a quarter of the standing voltage wave at 50Hz, but that's 1500km, so the length of the line isn't an issue to AC.
    Yeah I'm sure you're right.
    Huh, fascinating...and odd.
    The PN junctions of a rectifier. Sorry for the confusion.

    Yeah I'm really having difficulty understanding how AC current is actually flowing, maybe because I've always stuck with the theory and I've never married it to a physical model before, but hopefully I'll snap to my senses soon and realise I already understood it.

    Cheers Dave!
  9. Jun 19, 2016 #8


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    :smile: I don't claim to be an expert ... there's a good few guys on here that are way better than I, specially when it comes to the deeper theory

    Just remember, in an AC circuit in particular, the energy isn't carried in the electrons but in the electromagnetic field that oscillates back and forward along the outside of the conductor. Electrons are accelerated back and forward. An accelerated charge, eg an electron, will radiate EM radiation

    In a DC circuit, there is an initial pulse of an EM field through the circuit followed by the continuous flow of charge

    It's the fast moving EM field that accounts for the near instantaneous light etc coming on when the power switch is initially turned on
    The EM field is moving at the speed of light - the velocity factor of the cable. House power cabling will have a Vf of around 0.90 - 0.95 of the speed of light
    a coaxial cable used for RF communications have various Vf's of between ~ 0.5 - 0.7 of the speed of light, depending of the construction of the cable
    ( types of insulation etc)

  10. Jun 20, 2016 #9
    I do know what we are talking about Dave.I just couldn't understand Tim's wording in his first post and was under the impression he thinks certain way.

    As for the multiple points of feeding the railway line it's true for all lines whether DC or AC , here I have a 70km line between capital city and where I live , the line uses 3kV DC yet it's fed atleast in 3 places that I know of , maybe more.

    as for why the 25kV AC system over the 3Kv DC i'm not sure probably the 25kV ac system has more power and potential, otherwise the AC line actually introduces more complexity , older trains were almost all running with synchronous aka universal motors , they can be run on both DC and AC.
    Wikipedia indeed says that newer trains use three phase asynchronous motors.
    Although from a technical point of view the DC overhead line is simpler , we have some older trains running here with the overhead DC at 3kV, they have synchronous motors that are controlled using thyristors.

    As much as I read Dave , AC in general has no net movement in any direction.As for what Tim is confused about , well I think that it doesn't matter whether the physical charge carriers get somewhere after time or not all it matters is that they make a movement, because current is measured as the amount of charge that passes through some given conducting area in a given amount of time , say second.
    Much like Dave said , the fields carry the energy the electrons are just horses they ride.
    As for energy , you can run from point A to B and that will mean your doing work yet you can run back and forth like they made you do in school gym class and your doing just as much of a work.

    This thread inspired me to make my own as I too have some questions regarding electric trains, some of them came up while reading your posts.
  11. Jun 22, 2016 #10
    What's the link to the thread you started?
    Also @Salvador you can send me your Faraday motor design if you still want a hand, I could use the practice.

    Thinking about it, AC motors probably are much easier to control efficiently, but I wonder is it because now that we can control the frequency easily with solid-state technology, is the switch to AC trains just based on the fact that squirrel cage induction motors are really robust and you don't need to replace commutators all the time? (that is the main advantage?)
    Also, I wonder, was there some problem with using wound rotor induction motors in the old days, (vary the resistance of the rotor to get maximum torque on start up)? Maybe it was too inefficient or they couldn't get resistors big enough, so they had to use DC motors instead? I may have to ask this in a new thread.

  12. Jun 22, 2016 #11
    haven't created the thread yet , need to focus a bit on it.
    As for the Faraday thing , I though it through and sadly it won't work the exact way I envisioned.But that's for Pm.

    as for the railway motors thing I hope someone else could come and help with that answer as I too would love to know.
  13. Jun 22, 2016 #12


    User Avatar
    Science Advisor
    Gold Member

    Last edited by a moderator: May 8, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted