Talk:15 kV AC railway electrification

Proposed Rename

The existing name gives no hint that the topic is other than a general purpose electric transmission system. The name should make it clear that this is localized and special purpose. To most of the world AC by default means 50 or 60 Hz.LeadSongDog 18:30, 29 August 2007 (UTC)

Error

It is actually 16 kV that is used in Sweden and Norway. Bjorn Knutson 18:21, 7 December 2007 (CET).

The nominal voltage is 15 kV (see for instance http://banportalen.banverket.se/Banportalen/upload/1080/Jarnvagsnatsbeskrivning_T07.pdf ). On Banverket's net the maximum voltage at the point of consumption (the locomotive) is specified as 17.25 kV (15 kV +15%) and the minimum voltage is specified as 13,5 kV (15 kV -10%). To compensate for voltage drop along the catenary, the voltage at the feeder station's terminals are usually 16 kV or slightly higher. Unfortunately, Banverket adds to the confusion by stating 16 kV in many non-technical texts, such as press releases regarding accidents. All of this is for Sweden, I can't say anything about Norway but I would guess they have adapted the same EN-standard for interoperability reasons. JTragardh (talk) 07:54, 30 December 2007 (UTC)

First tested...

The technique for this electrification system was first tested in Germany on the Höllentalbahn during the 1930s and 1940s.

Sounds "a bit" questionable to me since the Gotthardbahn was fully electrified using this system in 1922... I think the first tests in Switzerland were run between Seebach and Wettingen around 1918, see also List of stock used by Swiss Federal Railways. --Kabelleger (talk) 22:20, 15 January 2008 (UTC)

It seems that this sentence was torn out of the right context with one of the last edits. It seems to belong to the 50 Hz stuff. --Kabelleger (talk) 00:18, 16 January 2008 (UTC)

It was a stray paragraph at the bottom which I moved up the page. The whole article still needs redoing - it looks like it started a a babel fish translation of the German article. I'd give it a go but I can't read the German reference materials. Railwayfan2005 (talk) 18:33, 16 January 2008 (UTC)

I've done some more copyediting, and removed that very inaccurate sentence over; I have references at my side stating that this system was used in Norway in 1922, so it could not have been developed during the 1940s. Arsenikk ^(talk) 15:57, 8 June 2008 (UTC)

Synchronous v. asynchronous

Quote:

"The frequency of 16.7 Hz depends on the necessity to avoid synchronism in parts of the rotary machine, which consists principally of a three phase asynchronous motor and a one phase synchronous generator. Since synchronism sets in at a frequency of 16+2⁄3 Hz (according to the technical details) in the one phase system, the frequency of the centralized system was set to 16.7 Hz".

I don't understand this. Where a synchronous motor is used, the output frequency would be exactly one-third of the input frequency. Where an asynchronous (induction) motor is used, the output frequency would be slightly less than one-third of the input frequency because of rotor slip. Assuming 10% slip, the output frequency would be about 15 Hz, not 16.7 Hz. Can somebody explain this please? Biscuittin (talk) 23:51, 6 December 2009 (UTC)

I think I have over-estimated the slip. A synchronous speed of 1,500 rpm normally gives an output speed of about 1,440 rpm, which implies only 4% slip. However, the principle is the same. Biscuittin (talk) 00:05, 7 December 2009 (UTC)

It would be quite difficult to do frequency conversion with an asynchronous motor. With a synchronous motor there is a simple factor of 3:1, e.g. a six pole 50 Hz motor driving a two pole 16+2/3 Hz generator. With an asynchronous motor running at 4% slip the factor would be about 20:7 so it would need a 40 pole motor driving a 14 pole generator. The alternative would be to use gearing. Biscuittin (talk) 17:36, 7 December 2009 (UTC)

Unfortunately I'm not proficient in the english terminology of power electrics, and the explanation is a bit half-baken (because in the german wikipedia no-one seems to understand this stuff well enough either), but anyway, I'll try to explain the parts that I understood :)

Traditionally, the rotary frequency converters were indeed totally synchronous, e.g. 12 poles in the 50 Hz machine and 4 poles in the 16 2/3 Hz machine. While for a synchronous motor/generator one could use a permanent magnet as the rotor, larger machines typically use electromagnets powered by direct current, which is supplied to the rotor by collector rings.

Now imagine, you have the construction of a synchronous motor and generator, but instead of supplying direct current to the rotor, you supply a very low frequency (e.g. one Hz) alternating current. What would happen? Of course, the speed of the rotor (or the frequency of the generated current) would be "off" from the "normal" frequency (when the rotor is supplied with direct current), by exactly the frequency of the current in the rotor.

This all works quite well if the frequency in the rotor is not too low. If that happens, you basically get direct current in the rotor, which somehow causes more thermal losses than an alternating current (unfortunately the details are not really clear to me), leading to overheating rotors.

So, what they did is change the nominal frequency of the net to something "a bit off" from 16 2/3 Hz, so they would never get a constant direct current in the rotor of asynchronous machines. The choice of 16.7 Hz is pretty much arbitrary I think (of course it should not be too far away from 16 2/3 Hz, or you would get problems with the locomotives).

    For the reason look at the Nyquist Sampling Theorem.  If the sampling frequency is an exact multiple of the highest allowed frequency, you can have 
    a DC term in the product.  Something very similar happens here, and obviously you don't want DC in your motor - it does nothing constructive, and 
    it generates heat.  So, by moving the frequency slightly off (from 16-2/3 Hz to 16.7 Hz), you do not get the DC term.  It is a subtle fix, but it 
    works.

What is *not* clear to me: Everybody talks about "alternating *current*" in the rotor. I imagine that should be alternating *voltage* on the rotor, because if the current in the rotor reached zero, there wouldn't be a magnetic field and you couldn't transfer any power during that phase (but on the other hand, I'm totally unsure if what I just wrote makes any sense).

    "Alternating Current" is the accepted term.  Yes, the voltage alternates, and so does the current.  Hopefully both are sine waves.  The real power 
    (watts) is the product of the voltage times the current times the sine of the phase angle between them.  Reactive power (vars) is the product of 
    the voltage times the current times the cosine of the phase angle between them.  Reactive power does no work.  As for the reason why, well, 
    everyone knows a potched watt never toils (sorry, couldn't resist).

You probably wonder why they bothered with all that in the first place. Well, if you only have synchronous motors and generators, you wouldn't have any control over which machine transfers how much power in which direction. This can be a problem if you have a large connected network, and if all the power is needed close to a weak motor/generator, and the ones with more capacity are far away - in such a situation, you could overload the weak machine, while the others don't do much at all. If you have control over the current in the rotor, you can control the machine in a way that it transfers a given amount of power in a certain direction, no matter what. This way railways can even sell power or use their infrastructure to transport electricity through the country. --Kabelleger (talk) 19:38, 7 December 2009 (UTC)

Thank you, Kabelleger. Your explanation does make sense and I am beginning to understand the concept. I think it is about keeping all the converters "in phase" with each other. By using the system you describe, it would be possible to make small adjustments to the speed of an out-of-phase converter to synchronize it with other converters. Biscuittin (talk) 09:22, 8 December 2009 (UTC)

That's true, but being able to get converters in sync is more of a nice side-effect, I think. The real value comes from being able to control the amount of power that is converted. --Kabelleger (talk) 22:56, 8 December 2009 (UTC)

I think the two effects are connected. If the 50 Hz system and the 16+2/3 Hz system were out-of-phase (as they probably would be) the motor would try to pull into synchronism with the 50 Hz system and the generator would try to pull into synchronism with the 16+2/3 Hz system. The result would be a tug-of-war which would waste energy and generate heat. The ability to make fine adjustments to the converter speed would overcome this problem. Biscuittin (talk) 10:28, 11 December 2009 (UTC)

The situation that you describe can only ever happen if you have asynchronous converters. If all converters are synchronous, they keep both networks in sync using all of their power (if that's not enough, then they are simply overloaded and either shut down or, well, blow up, possibly leading to a chain reaction which eventually shuts down the entire power supply of the 16 2/3 Hz network). So yes, if you have two networks that are not in sync then you'll need asynchronous converters, but if you only have synchronous converters then you'll never ever get two networks that are out of sync, and therefore don't need asynchronous converters. --Kabelleger (talk) 13:04, 11 December 2009 (UTC)

Sorry, I disagree with you. Networks can easily be out-of-phase with each other which is why back to back converters were developed. Biscuittin (talk) 20:32, 11 December 2009 (UTC)

In general that's true, but for railway applications things are a bit different, as far as I understand. Until the 1990s (and in some places still today), synchronous converters were used, which means that the two sides *had* to be in sync, otherwise it just would not have worked. I think the key why this was possible is that the 16 2/3 Hz side was not a large connected network, but rather many small, independent networks, which all had their own converters. In Germany they call that "decentralized power supply" (http://de.wikipedia.org/wiki/Bahnstrom#Dezentrale_Versorgung). The issue with large networks is that while two networks may be in sync at one geographic location, they will probably be out of sync at another location, because the cable lengths between the two locations are likely to be different, and if you have a few hundred kilometers of difference in length and a frequency of 50 Hz, you easily get a notable phase shift, because 2/3 c is not that fast any more in this context. But with small networks (especially when they are only connected at one point), you shouldn't get these problems.

And well, later, railway companies wanted to use their infrastructure to do more than just supply power to the trains near a converter station, so they had to connect the independent networks, which gave them all the trouble of large networks (reason #1 to build asynchronous converters) and had to be able to control the amount and direction of the power conversion (reason #2 to build asynchrounous converters). --Kabelleger (talk) 21:46, 11 December 2009 (UTC)

Yes, I agree. I think the earlier disagreement was due to a misunderstanding. Biscuittin (talk) 12:38, 12 December 2009 (UTC)

Inversion

The frequency change is described as "inversion" which could be misleading because inversion normally means conversion from DC to AC. In the case of solid-state converters, it is technically correct because there would be conversion from AC at 50 Hz to DC, followed by conversion from DC to AC at 16.7 Hz. However, I think a different term should be used. Biscuittin (talk) 00:00, 7 December 2009 (UTC)

I have changed "inverter" to "converter". Biscuittin (talk) 09:43, 7 December 2009 (UTC)

Disadvantages

16.7Hz system is not suitable for Shinkansen. —Preceding unsigned comment added by 121.102.47.215 (talk) 07:31, 22 February 2010 (UTC)

Argentina

If Buenos Aires-Rosario-Córdoba high-speed railway (planned) use standard gauge, it will use 15kV AC 162⁄3 Hz. 121.102.47.215 (talk) 00:57, 13 March 2010 (UTC)

Two problems with information like this -- it's speculation about the future, and it's unsourced.
Prari (talk) 13:30, 14 March 2010 (UTC)

Even a third problem. It is highly uncommon to use 15kV/16.x Hz for new railroad systems. All high speed and heavy freight systems that do not have a surrounding network imposing something other than that are going for 25kV/50Hz (or 60Hz), or even in some rare cases 50kV. Even the countries that do have huge DC-networks have a tendency to use 25kV for there new freight or highspeed lines or even to convert to that system. A likely information might go away without heavy proof. That would be, I would be more than willing to trust the information, if it said that Argentina were going for 25kV/50Hz, because that is what is done everywhere, what is sold on the world market as standard equipment etc. The only reason to go for 15kV/16.xHz would be to buy used rolling stock from Sweden, Norway, Germany, Austria and Switzerland. Anyway, a surprising information or even one that is hard to believe does need some credible source to create evidence. For the time being, let's keep it with the information that this project is "on hold", as could be read in some wikipedia article. I would like to add that using an IP-address (which is perfectly ok, off course!!!) as user name does not make it easier to persuade others about this.--Bk1 168 (talk) 14:06, 21 May 2010 (UTC)

50kV system is never using for high-speed rail. 121.102.47.39 (talk) 02:17, 29 May 2010 (UTC)

50kV is used for some isolated freight lines. All new high speed lines either use what is the electrification system in the country (15 kV/16.7 Hz in Germany, Sweden, Norway, Switzerland & Austria, 3 kV on older highspeed lines in Italy) or 25 kV AC, like some new high speed lines in Italy. I would be very surprised if Argentina did not use 25kV AC, but it is off course not clear if the Argentine high speed railroad is coming at all.--Bk1 168 (talk) 07:34, 29 May 2010 (UTC)

History

I am thinking of adding a History section. The 15 KV AC system is a development of the 6.6 KV AC system used in Germany and (with German equipment) in England, see London,_Brighton_and_South_Coast_Railway#Railway_electrification. Biscuittin (talk) 10:26, 26 March 2010 (UTC)

Reasons for low frequency

Quote: "In particular, the lower frequency reduces flashover problems in the motors". I'm not sure this is correct - I think it is mainly about efficiency. Commutator motors are inefficient when run at 50/60 Hz but the efficiency can be improved by reducing the frequency. Biscuittin (talk) 10:33, 26 March 2010 (UTC)

15 kV 162⁄3 Hz system has more disadvantage and no advantage than the other voltage system. 121.102.47.39 (talk) 14:00, 10 May 2010 (UTC)

Some complete nonsense being talked above. 25 Hz was chosen as it made possible the use of the AC Commutator motor (an a.c. version of a series d.c. motor). Any higher frequency would have made commutation harder to achieve, and would have caused issues with flashover across the windings of the armature. The a.c. commutator motor was able (with the addition of some reistances in the armature windings) operate on a.c. and (with the resistances removed) also on d.c., which was no mean feat in the early part of the 20th century. 15 kV 162⁄3 Hz systems have been immensely successful, and there is nothing wrong with them per se, except that today they need there own power plants or have to convert the 50/60 hz supply. The system has one third of the impedance of 50 Hz systems, as impedance is proportional to frequency, so it is quite efficient, and can transmit power long distances without causing voltages to drop below acceptable limits. So, whatever you may think, 15 kV 162⁄3 Hz system is a very good one. Bhtpbank (talk) 19:36, 10 May 2010 (UTC)

Nordic people want to convert to 3 kV DC, while Alpine people and Mediterranean people want to convert to 25 kV 50 Hz. 121.102.47.39 (talk) 12:09, 11 May 2010 (UTC)

The reasons to use 15 kV/16.x Hz are historical. 100 years ago it was difficult to handle 50Hz in the engine. Now this creates two major disadvantages:

• with 50 Hz the railroads can, but don't have to use electricity from the public networks without frequency conversion. (Sometimes having a separate railroad electricity network is considered an advantage, but if that is the case, it is possible to build it even with 50 Hz where useful, but to use the exisiting infrastructure otherwise)
• transformation efficiency increases with frequency. That means that either the transformation is less efficient, loosing energy to heat or that it needs to be much heavier, which is a hazard to carry around in the engine.
• Anyway this system is not so bad. 25kV/50Hz is better, but the common DC-systems are not as good because they require to heavy wires.

This could be mentioned in the article. I am quite sure about the transformation loss issue, but it would be useful to find sources and to quantify this. Btw. I have no evidence of the Sweden and Norway wanting to convert to 3000V DC, which would not make any sense, nor any serious evidence of any country moving from 15kV to 25kV, although that might make sense. --Bk1 168 (talk) 14:27, 18 May 2010 (UTC)

People in Sweden, Norway and northern Germany want to convert to 3000V DC because they are more blond. While people in Switzerland, Austria and southern Germany want to convert to 25 kV AC because they are more Alpine race and Mediterranean race. 121.102.47.39 (talk) 17:47, 18 May 2010 (UTC)

So in short: This talk about 3kV-conversion is total junk! I do not see any relationship between races and electrification currencies and I think this is the wrong place for talking about mediterrenean race and nordic race. Please keep that at least out of the article and write it into the places where it belongs, if useful, or leave it away otherwise.--Bk1 168 (talk) 05:27, 19 May 2010 (UTC)

15 kV 162⁄3 Hz is worse than either 3 kV DC or 25 kV 50/60 Hz because 15 kV 162⁄3 Hz associate poorer power to weight ratio, and useless for French TGV trainsets or Japanese bullet trains. Nordic people want to convert to 3 kV DC, while Alpine people and Mediterranean people want to convert to 25 kV AC. Also French people want wider carbody than UIC profile and they hate 15 kV 162⁄3 Hz. 121.102.47.39 (talk) 07:25, 19 May 2010 (UTC)

In general it is clear that 25kV/50Hz or 25kV/60Hz is superior to 15kV/16.7Hz. But the effort to convert is quite significant and I have never heard of any real projects in that direction. As a matter of fact, in Germany and Sweden quite successful high speed operations are done with 15kV/16.7Hz, with up to 300km/h between Frankfurt and Cologne. 3kV DC is not at all a serious option. Railroads that are using 1.5kV or 3kV are really seriously converting to 25kV or considering such a conversion. So technically speaking, the following order in technical usefulness of the four most common electrification schemes in Europe exist:

1. 25 kV / 50 Hz
2. 15 kV / 16.7 Hz or 16 2/3 Hz
3. 3 kV / DC
4. 1.5 kV / DC

Problem with low voltage is that it require very heavy wires and is difficult to support the power of heavy and fast trains. 15kV is not bad, but it is a minor disadvantage that the frequency is so low.

Anyway, this has nothing at all to do with nordic race or alpine race or whatsoever. So let us keep to the topic, which is technical and let us keep to the reality, which is that no 15 kV/16.x Hz railroad is even slightly considering conversion to one of the DC systems, which would be simply absurd and they currently are not even seriously considering the conversion to 25kV, even though that might make some sense in the very long term.--Bk1 168 (talk) 09:58, 19 May 2010 (UTC)

Just adding some evidence to this. No major railroad is seriously considering a switch to 15 kV / 16.7 Hz or 16 2/3 Hz, but many are actually switching to 25 kV / 50 Hz. This gives some evidence that 25 kV / 50 Hz is preferred to 15 kV / 16.7 Hz or 16 2/3 Hz. Again from those who actually have 15 kV / 16.7 Hz or 16 2/3 Hz, nobody is seriously considering to switch to 25 kV / 50 Hz, which the ones with the DC-systems are doing in quite a few cases, at least for new railroad lines. That suggests that 15 kV / 16.7 Hz or 16 2/3 Hz is preferred over the DC systems. That 3 kV DC is better than 1.5 kV should be almost obvious, it is the weight of the wires, that is one of the problems with this currency and that is worse with 1.5 kV. Always talking about heavy railroad infrastructure, not light rail, where issues might be slightly different.--Bk1 168 (talk) 07:46, 20 May 2010 (UTC)

four most common electrification schemes in Europe exist:

1. 25 kV / 50 Hz
2. 3 kV / DC
3. 15 kV / 16.7 Hz or 16 2/3 Hz
4. 1.5 kV / DC

power to weight ratio:

1. 3 kV / DC
2. 25 kV / 60 Hz
3. 25 kV / 50 Hz
4. 15 kV / 16.7 Hz or 16 2/3 Hz
5. 1.5 kV / DC

16.7Hz is the worst because 16.7Hz needs heavy transformers. 121.102.47.39 (talk) 08:25, 20 May 2010 (UTC)

You have to consider the whole system. 3 kV DC might or might not be slightly favorable for building the rolling stocks, but it is simply very expensive to build and maintain the overhead wires for this voltage. So the reality is that railroad companies are slowly moving away from the 1.5kV/3kV DC systems to the 25kV system for heavy rail. You can observe that 1.5kV is not replaced by 3kV, but by 25kV. So you can claim that they know less than you do or whatever, but here we need to describe what real railroad systems really do, not what we would do on our model railroad or so.--Bk1 168 (talk) 08:44, 20 May 2010 (UTC)

Standard gauge not broad gauge 1.5 kV DC lines will replaced by 3kV DC. And standard gauge not broad gauge 3 kV DC lines won't convert to 25kV 50/60Hz. Any 3 kV DC lines won't convert to 15 kV 162⁄3 Hz. 121.102.47.39 (talk) 08:59, 20 May 2010 (UTC)

Quote: 'You have to consider the whole system.' Very true. The 25 kV voltage for example was chosen because it balanced out the larger investment in clearances and insulators and the gain in energy efficiency and cheaper wiring. Same goes for the choice of power system. For trams and metros low voltage DC current (600 and 750 V) is still used, even for new systems. 82.139.114.136 (talk) 19:35, 31 July 2012 (UTC)

High-speed rail lines needs heavier wire and either 25 kV AC or 3 kV DC. Nordic people want to convert to 3 kV DC, while Alpine people and Mediterranean people want to convert to 25 kV AC. And any people want wider car body. 121.102.47.39 (talk) 08:31, 20 May 2010 (UTC)

I'm sorry, but that is not true. Germany is operating real high speed lines with up to 300 km/h (180 mph) using 15 kV/16.7 Hz. And no matter how nordic or how mediterranean people in Germany are, there is no evidence for any serious plan to change the electrification system to 25 kV/50 Hz and definitely there are no plans to switch to 3kV DC. no way!. The other question is, what does the technical electrification system have to do with races? I think that your articles about different human races might be interesting or not, but they are very off topic in this technical article. And in areas where this might be on topic, you should at least consider the required neutrality in Wikipedia articles. So, please keep your nordic/meditarranean/alpine-race stuff out of the railroad articles and their discussions in the future. Nobody wants to read it here. Thank you so much.--Bk1 168 (talk) 08:44, 20 May 2010 (UTC)

In addition, standard gauge not broad gauge lines with 3 kV DC won't convert to either 15 kV 162⁄3 Hz or 25 kV 50/60 Hz. Sweden don't want high-speed trains. 121.102.47.39 (talk) 08:43, 20 May 2010 (UTC)

Sweden does a huge operation with high speed trains with 200 km/h (125 mph). Some lines will allow higher speeds in the future.--Bk1 168 (talk) 08:47, 20 May 2010 (UTC)

Sweden don't want high-speed trains 300km/h or faster. Sweden don't want skyscrapers 300m or taller, too. 121.102.47.39 (talk) 13:25, 22 May 2010 (UTC)

Oh yes, Sweden does have very real plans for a high speed line from Malmö and Gothenburg via Jönköping to Stockholm with speed >= 300 km/h. But I would use the term high speed for speeds >= 200 km/h (125 mph for our American friends).--Bk1 168 (talk) 09:50, 24 May 2010 (UTC)

Sweden don't want world's fastest trains. Sweden don't want world's tallest building, too. 121.102.47.39 (talk) 06:36, 28 May 2010 (UTC)

I do not know if this will be the fastest train in the world, but I do not think so, but it will be without any doubt a high speed train running around 300 km/h. It is like a rotated letter Y, with lines coming from Malmö and Gothenburg joining in or near Jönköping and continuing to Stockholm from there. It is quite easy to find information about this. And I am sure they will not use 3kV DC for this line.--Bk1 168 (talk) 06:57, 28 May 2010 (UTC)

Any people want wider car body. Especially French people. 121.102.47.39 (talk) 09:06, 20 May 2010 (UTC)

The reason for the low frequency was the motor poles of the old AC motors were made of a solid material. This caused eddy currents in these materials because magnetic materials of the days were electrically conductive (usually some kind of iron), heating up the motor unnecessarily. This effect gets less with lower frequency and can be made irrelevant at about 15 Hz. At the time the electrification schemes were introduced most countries did not have a centralized power distribution system. Also, a lot of industrial users also used large AC motors and these lower frequencies so sourcing them was actually rather easy. In the US the railways settled at 25 Hz due to its availability, even though more efficient motors could be constructed with 15 Hz.

Newer designed AC motors have laminated pole pieces, virtually eliminating the eddy current problem. This development came too late however to stop the massive investments that were already in place by then. In parallel the grid frequency got standardised at 50 Hz in Europe and 60 Hz in the US, leaving the railways with a system that needed a separate infrastructure. Investments in changing the system would be excessive because all rolling stock would have to be adapted simultaneously. At the time though the investment seemed logical because it made power levels possible that weren't available in any other way.

The low frequency system is suited for high speed railways. The 15 kV system has roughly the same losses as a 25 kV system. The voltage difference is insufficient to cause significant differences in energy losses. The lower frequency does require a heavier transformer to convert the same amount of energy, so the rolling stock will most likely wind up significantly heavier. The ICE is actually heavier than the TGV, even though I don't know if this is the primary cause. A few TGV POS (in French) SNCF TGV POS trains have been built to work under the German and Swiss wires, but have a significantly lower power output then, reflecting the unsuitability of the transformers for this frequency. The other way around is feasible without an increase in weight.

High speed rail travel is possible under lower voltages too, but have usually been limited to 200 km/h (in France some 1500 V routes are suited for 200 km/h with appropriate rolling stock like the Sybic and overhead wires that have been tensioned sufficiently. Currents of in excess of 4000 A will flow with power levels like these which are at the limit of practical usability. 82.139.114.136 (talk) 11:46, 31 July 2012 (UTC)

redirect from "15 kV AC railroad electrification"

Shouldn't we have a redirect from "15 kV AC railroad electrification"?--Bk1 168 (talk) 05:39, 19 May 2010 (UTC)

25 kV/50 Hz engines running on 15 kV/16.x Hz

It seems to be quite easy to build engines that operate in 25 kV / 50 Hz in such a way, that they can also run somehow on 15 kV / 16.x Hz. The reason is that it is now common to do the following steps in the engine:

1. transform voltage from the overhead wire to an intermediate voltage
2. convert that from AC to DC. This is where our 121.x.x.x-friend comes in: Using 3kV DC would save step 1 and the installation on the engine, but as I have written before, the whole system has to be considered and what is saved on the engine is not justifying the extra effort for the overhead wire, so actually railroads are moving away from DC and 25kV/50Hz is becoming the standard for highspeed and heavy trains.
3. convert that DC to whatever frequency is neded, probably higher than 16.x Hz.
4. transform that to whatever currency is needed.

So with 15kV/16.xHz the transformer in step 1 can either be used as is, just yielding a lower voltage and thus lower power, or its primary part can easily be accessed at 60% of its length, yielding the same currency as with 25kV. In either case the transformer is not optimized for this usage, so it is too small. Big disadvantage of 16.x Hz, that a huge, heavy transformer needs to be in the engine to get a decent efficiency. Anyway, by reducing the power, this works. I am aware that the french TGV trains did something like this to operate in Switzerland, where they could not go beyond 140 or 160 km/h anyway, but actually they can and may go up to 180 km/h on 15kV/16.x Hz. The same is true for Danish 25kV engines, which can run at reduced speed and power in Germany. Probably it applies to many more cases. But off course it is possible to build engines or trainsets that are dedicated to running full power in both systems by simply implementing transformers that are capable of working with full power and efficiency on 15kV/16.xHz. Making that work with 25kV/50Hz should be relatively easy, but needs to be done, off course. So the Czech engine is just one of many examples.--Bk1 168 (talk) 13:59, 21 May 2010 (UTC)

I can't give you a source at the moment, but I'm certain that the voltage of the DC circuit is lower than 3000 volts, more like 500 volts (of course, this gives very high currents, but it seems that it is easier to handle high currents than high voltages within the locomotive).

Second, step 3 and 4 don't work this way. You either power a DC motor (using pulse-width modulation), which is what was done in the 70s and 80s, but starting in the 90s, all modern designs use three-phase motors. These have a unique property that the motor turns synchronously to the "rotation" of the three-phase supply, so the direct current is converted to a three-phase current with the required frequency and a phase-shift of 1/3 for each phase. In any case, no transformer is needed at this stage. --Kabelleger (talk) 04:51, 22 May 2010 (UTC)

ITYWF that they are induction motors not synchronous motors. 109.156.49.202 (talk) 18:09, 14 November 2011 (UTC)

Very true, but they both require frequency control because especially in large high-efficiency induction motors the slip (the difference between the input frequency and rotational speed) is only a few percent. 82.139.114.136 (talk) 12:06, 31 July 2012 (UTC)

It is hard to find sources indeed, but in high power machines the voltages are actually quite high. I believe there to be a standardized bus voltage of 3600 V. When I find the source I will give it, but it does seem to reflect the improvement in insulation since the early 1900s when a maximum voltage of 1500 V was common.

-edit: I found the bus voltages in the Bombardier MITRAC system to differ according to the power supply used. 82.139.114.136 (talk) 12:22, 31 July 2012 (UTC)

I reread your comment Bk1 168, making a transformer is actually more difficult than just tapping it at the appropriate part of the transformer. Due to the lower frequency the magnetic core needs to be larger too to contain the stronger magnetic field without saturating, and will need more windings to have sufficient inductance for this same reason. I expect that transformers that have been designed for both systems have multiple primary windings that can be connected in series or in parallel in such a fashion that all these criteria are met, and have multiple secondaries that are switched accordingly to keep providing the correct voltages.

In theory a transformer will always be able to convert more energy when the frequency increases. Transformers in computer power supplies are so small due to the high frequency used (10s to 100s of kHz). Higher frequencies do have the disadvantage more eddy current and skin effect losses come in to play, so designing a universal transformer does require some thought. Older transformers built for 16⅔ Hz might therefore not always work satisfactory when used for higher frequencies.

It is therefore not only an efficiency issue, but a design issue too and in theory a low frequency transformer will always have a lower power to weight ratio. 82.139.114.136 (talk) 12:32, 1 August 2012 (UTC)

border crossings

Germany - Switzerland:

1. Konstanz (DE) - Kreuzlingen (CH): electrified
2. Singen (DE) - Etzwilen (CH): just a museum railroad with steam
3. Singen (DE) - Bietingen (DE) - Thayngen (CH) - Schaffhausen (CH): electrified, owned by German railroad DB
4. Erzingen (DE) - Trasadingen (CH) - Schaffhausen (CH): not electrified, but long term project to electrify, owned by German railroad DB
5. Waldshut (DE) - Koblenz (CH): electrified
6. Laufenburg (DE) - Basel Bad (CH): not electrified
7. Weil (DE) - Basel Bad (CH) - Basel (CH) - Muttenz (CH): electrified, Basel Bad - Weil owned by German railroad DB
8. Jestetten (DE) - Neuhausen (CH) - Schaffhausen (CH): electrified: owned by Swiss railroad, even the section in Germany as part of connection Zurich - Schaffhausen
9. Jestetten (DE) - Lottstetten (DE) - Rafz (CH): electrified: owend by Swiss railroad as part of connection Zurich - Schaffhausen

--Bk1 168 (talk) 00:34, 22 May 2010 (UTC)

Slight correction:

Austria - Switzerland:

• Buchs (CH) - via Liechtenstein - Feldberg (AT).
• St. Margrethen (CH) - Bregenz (AT).

Germany - Switzerland:

• Singen (DE) - Etzwilen (CH) - not sure this line reaches Singen. According to [1] it terminates before the border.
• Correction: Eglisau (CH) - Ratz (CH) - Lottstetten (D) - Jestetten (D) - Neuhausen (CH) - Schaffhausen (CH). Owned and operated by SBB, electrified. No connection to DB network, but goes via German territory.
• Correction: Waldshut (DE) - Koblenz (CH) is only electrified as far as Waldshut, the first station in Germany. No electrified connection with DB network.
• Add: Basel Badischer Bahnhof (CH) - Riehen (CH) - Lörrach (D) - Zell in Wiessental (D). Electrified.

References: [2] [3]. Except, do you count border crossings as the actual border, or the property border between DB and SBB?. In Basel, Basel Badischer Bahnhof and the three lines leading to it from Germany are all owned by DB. The property border between SBB-owned and DB-owned tracks is actually on the link-line across the Rhine between Basel Badischer Bahnhof and Basel SBB.

Now we need to feed this into the article. TiffaF (talk) 19:53, 29 September 2010 (UTC)

Singen - Etzwilen is not electrified. No regular passenger service on that line. I think it is connected from Etzwilen all the way to the southern part of Singen. Within Singen it was interrupted during a highway expansion project, so only two stubs ending in southern Singen, one from Singen, one from Etzwilen.--Bk1 168 (talk) 20:29, 29 September 2010 (UTC)

Dead link

During several automated bot runs the following external link was found to be unavailable. Please check if the link is in fact down and fix or remove it in that case!

• http://de.wikipedia.org/wiki/Liste_von_Bahnstromanlagen
  • In 15 kV AC railway electrification on 2011-06-01 18:18:13, 404 Not Found
  • In 15 kV AC railway electrification on 2011-06-20 05:19:30, 404 Not Found

--JeffGBot (talk) 05:20, 20 June 2011 (UTC)

Duplication

Do we need the lists of substations in this article? I think they just duplicate the lists in List of installations for 15 kV AC railway electrification in Germany, Austria and Switzerland. Roberttherambler (talk) 22:09, 3 September 2017 (UTC)

Done it, there's literally no reason for that to be duplicated that way, it only fluffs the page and makes it harder to navigate TuxyQ ^{(talk|contrib)} 22:41, 19 January 2022 (UTC)

Move?

How about calling the article 16 2/3 Hz railway electrification? The oddball frequency is more of a defining feature of this system than the voltage.Nankai (talk) 05:37, 1 June 2020 (UTC)

Oddball? It's just 1/3 of the European 50Hz standard. Martin of Sheffield (talk) 09:03, 1 June 2020 (UTC)

Every other railway electrification article is named by voltage; I think it would be confusing if this one were different. Mackensen (talk) 12:00, 1 June 2020 (UTC)