7 ms·

230/240v does have it's advantages, primarily being able to use smaller diameter wire for the same current carrying capacity. But 50 Hz as a system frequency i

by chiph 4y ago

230/240v does have it's advantages, primarily being able to use smaller diameter wire for the same current carrying capacity. But 50 Hz as a system frequency is just wrong. It results in oddities like running your railroad at 16-2/3 Hz.

eru 4y ago
I'm not sure what's so odd about running your railroad at 50/3 Hz? Do you not like rational numbers? (I guess you are talking about three-phase power in general, or is there anything special about railroads?) In practice, just like everything in engineering, you don't run your grid at 50 Hz nor your railroad at 50/3 Hz. You run the grid at some frequency that varies slightly around 50Hz. There's always engineering tolerances. So even if you don't like rational numbers, it doesn't really matter whether you run the railroads at 50/3 +-0.01 Hz, or at 16.66 +-0.01 Hz. (I don't know how tight the tolerances are in practice here. It doesn't matter for the argument.)
- lazide 4y ago
  Rational numbers are a lot more work than integers. Also, 60 is a common base for several things (degrees, time), which interact with the grid in numerous ways. It’s convenient to have them match.
  - eru 4y ago
    Just express everything in micro-Hertz and you can use integers just fine. (If you need more precision, go for nano-Hertz..)
- iggldiggl 4y ago
  > In practice, just like everything in engineering, you don't run your grid at 50 Hz nor your railroad at 50/3 Hz. You run the grid at some frequency that varies slightly around 50Hz. There's always engineering tolerances. > So even if you don't like rational numbers, it doesn't really matter whether you run the railroads at 50/3 +-0.01 Hz, or at 16.66 +-0.01 Hz. In practice the nominal frequency was changed to 16.7 Hz anyway due to some weird edge case created when running motor-generators [1] for prolonged periods of time at exactly the nominal 3:1 frequency conversion ratio. (See https://de.wikipedia.org/wiki/Bahnstrom#16_2%E2%81%843_Hz_gegen%C3%BCber_16,7_Hz https://de.wikipedia.org/wiki/Bahnstrom#16_2%E2%81%843_Hz_ge... and use an online translator at your own peril if necessary.) [1] These days, newly built (or probably occasionally re-built) substations prefer to use solid-state frequency converters, which are immune to this particular problem, but there still are enough of the older motor-generator sets around, too. Wikipedia also claims that motor-generator sets are more tolerant against earth faults, which might or might not preclude against getting rid of all of them even long-term.
jabl 4y ago
> It results in oddities like running your railroad at 16-2/3 Hz. 50 Hz railway systems are probably more common than 50/3, at least in Europe: https://en.wikipedia.org/wiki/Railway_electrification_system#/media/File:Europe_rail_electrification_en.svg https://en.wikipedia.org/wiki/Railway_electrification_system... 25 kV AC 50 Hz (60 Hz in countries using such grid frequency) is the modern standard: https://en.wikipedia.org/wiki/25_kV_AC_railway_electrification https://en.wikipedia.org/wiki/25_kV_AC_railway_electrificati...
thulle 4y ago
How does a 50 Hz system frequency result in railroads running at 1000/60 Hz?
- iggldiggl 4y ago
  For hysterical raisins: Before the advent of power electronics, which allowed three phase asynchronous motors to proliferate, commutated series-wound motors were the railway motor because of their beneficial characteristics (quoting Wikipedia: "high starting torque, can run at high speed, and are lightweight and compact"). A classic commutated series-wound motor is a DC machine. The problem with DC power is that without power electronics, you can't really change the voltage (except downwards by wasteful resistors), so your transmission voltage is limited by your maximum motor voltage and you incur relatively high transmission losses and need frequent substations every few kilometres. (Mainline railways can cheat a bit by equipping all of their rolling stock with two (or occasionally more) motors permanently linked together in series, so each motor only gets a fraction of the voltage, but you can take that approach only so far…) So you want to switch to AC power, which is more efficient because it can be easily transformed up and then down again, because transformers can actually be made small enough to fit into a locomotive. Fortunately with some adaptations a series-wound motor can be made to work on AC, too, but there are some trade-offs depending on the required amount of power and the frequency. I.e. the motor still works better if your AC current is more "DC-like", i.e. doesn't have too high a frequency. It turns out that for something more modest like a hoover or a domestic power drill, an AC series-wound motor ("universal motor") will work fine enough even at 50 or 60 Hz, but for railway purposes with their somewhat higher power demands things didn't work so well. Because at that time power electronics didn't exist or were still in their infancy (e.g. mercury arc rectifiers), there was no possibility of doing anything useful on board of the locomotive, so instead it was decided to reduce the frequency of the railway's power supply system, because at 16 2/3 or perhaps even 25 Hz motors could still be made to work reasonably enough. Also at that time changing the frequency meant using a motor-generator set, and presumably choosing some simple integer ratio for the two frequencies also simplified things. By the time other countries started considering switching to AC electrification, usable rectifiers existed that could be used on locomotives, so those countries could electrify at the full 50 or 60 Hz and then rectify the current to DC on board of the locomotive, thereby sidestepping the problems of running a commutated series-wound motor with AC power. Even later on we then got fully variable voltage and frequency inverters, which finally allowed serious usage of (usually asynchronous) three-phase induction motors. (Some railways used three-phase drives even before that, but without modern electronics this was a somewhat more cumbersome prospect, see e.g. https://en.wikipedia.org/wiki/K%C3%A1lm%C3%A1n_Kand%C3%B3 https://en.wikipedia.org/wiki/K%C3%A1lm%C3%A1n_Kand%C3%B3)

by chiph 4y ago

eru 4y ago

I'm not sure what's so odd about running your railroad at 50/3 Hz? Do you not like rational numbers? (I guess you are talking about three-phase power in general, or is there anything special about railroads?) In practice, just like everything in engineering, you don't run your grid at 50 Hz nor your railroad at 50/3 Hz. You run the grid at some frequency that varies slightly around 50Hz. There's always engineering tolerances. So even if you don't like rational numbers, it doesn't really matter whether you run the railroads at 50/3 +-0.01 Hz, or at 16.66 +-0.01 Hz. (I don't know how tight the tolerances are in practice here. It doesn't matter for the argument.)

lazide 4y ago
Rational numbers are a lot more work than integers. Also, 60 is a common base for several things (degrees, time), which interact with the grid in numerous ways. It’s convenient to have them match.
- eru 4y ago
  Just express everything in micro-Hertz and you can use integers just fine. (If you need more precision, go for nano-Hertz..)
iggldiggl 4y ago
> In practice, just like everything in engineering, you don't run your grid at 50 Hz nor your railroad at 50/3 Hz. You run the grid at some frequency that varies slightly around 50Hz. There's always engineering tolerances. > So even if you don't like rational numbers, it doesn't really matter whether you run the railroads at 50/3 +-0.01 Hz, or at 16.66 +-0.01 Hz. In practice the nominal frequency was changed to 16.7 Hz anyway due to some weird edge case created when running motor-generators [1] for prolonged periods of time at exactly the nominal 3:1 frequency conversion ratio. (See https://de.wikipedia.org/wiki/Bahnstrom#16_2%E2%81%843_Hz_gegen%C3%BCber_16,7_Hz https://de.wikipedia.org/wiki/Bahnstrom#16_2%E2%81%843_Hz_ge... and use an online translator at your own peril if necessary.) [1] These days, newly built (or probably occasionally re-built) substations prefer to use solid-state frequency converters, which are immune to this particular problem, but there still are enough of the older motor-generator sets around, too. Wikipedia also claims that motor-generator sets are more tolerant against earth faults, which might or might not preclude against getting rid of all of them even long-term.

jabl 4y ago

> It results in oddities like running your railroad at 16-2/3 Hz. 50 Hz railway systems are probably more common than 50/3, at least in Europe: https://en.wikipedia.org/wiki/Railway_electrification_system#/media/File:Europe_rail_electrification_en.svg https://en.wikipedia.org/wiki/Railway_electrification_system... 25 kV AC 50 Hz (60 Hz in countries using such grid frequency) is the modern standard: https://en.wikipedia.org/wiki/25_kV_AC_railway_electrification https://en.wikipedia.org/wiki/25_kV_AC_railway_electrificati...

thulle 4y ago

How does a 50 Hz system frequency result in railroads running at 1000/60 Hz?

iggldiggl 4y ago
For hysterical raisins: Before the advent of power electronics, which allowed three phase asynchronous motors to proliferate, commutated series-wound motors were the railway motor because of their beneficial characteristics (quoting Wikipedia: "high starting torque, can run at high speed, and are lightweight and compact"). A classic commutated series-wound motor is a DC machine. The problem with DC power is that without power electronics, you can't really change the voltage (except downwards by wasteful resistors), so your transmission voltage is limited by your maximum motor voltage and you incur relatively high transmission losses and need frequent substations every few kilometres. (Mainline railways can cheat a bit by equipping all of their rolling stock with two (or occasionally more) motors permanently linked together in series, so each motor only gets a fraction of the voltage, but you can take that approach only so far…) So you want to switch to AC power, which is more efficient because it can be easily transformed up and then down again, because transformers can actually be made small enough to fit into a locomotive. Fortunately with some adaptations a series-wound motor can be made to work on AC, too, but there are some trade-offs depending on the required amount of power and the frequency. I.e. the motor still works better if your AC current is more "DC-like", i.e. doesn't have too high a frequency. It turns out that for something more modest like a hoover or a domestic power drill, an AC series-wound motor ("universal motor") will work fine enough even at 50 or 60 Hz, but for railway purposes with their somewhat higher power demands things didn't work so well. Because at that time power electronics didn't exist or were still in their infancy (e.g. mercury arc rectifiers), there was no possibility of doing anything useful on board of the locomotive, so instead it was decided to reduce the frequency of the railway's power supply system, because at 16 2/3 or perhaps even 25 Hz motors could still be made to work reasonably enough. Also at that time changing the frequency meant using a motor-generator set, and presumably choosing some simple integer ratio for the two frequencies also simplified things. By the time other countries started considering switching to AC electrification, usable rectifiers existed that could be used on locomotives, so those countries could electrify at the full 50 or 60 Hz and then rectify the current to DC on board of the locomotive, thereby sidestepping the problems of running a commutated series-wound motor with AC power. Even later on we then got fully variable voltage and frequency inverters, which finally allowed serious usage of (usually asynchronous) three-phase induction motors. (Some railways used three-phase drives even before that, but without modern electronics this was a somewhat more cumbersome prospect, see e.g. https://en.wikipedia.org/wiki/K%C3%A1lm%C3%A1n_Kand%C3%B3 https://en.wikipedia.org/wiki/K%C3%A1lm%C3%A1n_Kand%C3%B3)