Why 16 1/3 Hz for some electric locomotives?
John Byer
the world. I think I pretty well understand the diesel/generator/motor
setup for locomotives. What I don't understand, is that many of the
electric locomotives use 30-50K volts ac, (this I understand,) at some
wierd frequency, like 16 1/3 Hz (this, I don't.) Inasmuch as most ac
in the world is distributed at 50 or 60 Hz, why do they go to the
trouble to generate 16 1/3 Hz? Surely there must be some rational
reason, but I sure haven't been able to come up with it. Forgive
me if this has been covered before, I've only recently started reading
this newsgroup.
Thanks,
{uunet|sun|apple}!unisoft!byer
Douglas W. Jones,201H MLH,3193350740,3193382879
> ... What I don't understand, is that many of the
> electric locomotives use 30-50K volts ac, (this I understand,) at some
> wierd frequency, like 16 1/3 Hz (this, I don't.) Inasmuch as most ac
> in the world is distributed at 50 or 60 Hz, why do they go to the
> trouble to generate 16 1/3 Hz?
There's a rational reason, don't worry. At low frequencies, induction
motors run slower, and that allows them to use lower gear ratios and lower
speed bearings in the drive train. In the early years of electrification,
there wasn't much problem about compatibility with commercial 50 or 60 hz
power, since everyone was used to using motor generator sets for voltage
conversion in DC electrification, and AC power was not yet universal on
the commercial front.
Note that 16.666 is exactly 0.3 times 50hz. This means that a synchronous
motor-generator with a 3 to 1 ratio can convert commercial 50hz power to
16.666hz railroad power.
The down-side of using such low frequency power is that it requires huge
transformers to step the voltage up or down. In the engine, these added
useful weight, but simple weights are more convenient. (In aerospace
applications, 400hz is common because, at that frequency, transformers are
even lighter than the 50 and 60hz transformers we use on the ground.)
If you're stringing new wire, you don't worry about such odd low frequency
power. You run at the commercial frequency, using DC traction motors and
thyristor controls. Even 50 years ago, motor-generators in the engine
to convert from line voltage (whatever it was) to DC for traction motors
were coming into use.
Doug Jones
jo...@herky.cs.uiowa.edu
pie...@cimnet.dec.com
>What I don't understand, is that many of the electric locomotives use 30-50K
>volts ac, (this I understand,) at some wierd frequency, like 16 1/3 Hz
>(this, I don't.)
Without a reference, I think most of the 25Kv nominal and all of the
50KV nominal is 50 or 60 Hz, as appropriate. The lower frequency
usually shows up in the 11-15Kv range.
>Inasmuch as most ac in the world is distributed at 50 or 60 Hz, why do they
>go to the trouble to generate 16 1/3 Hz?
As someone else (Doug(?), sorry, I'm stuck in the editor...) points
out, the LF has advantages for induction motor design. In addition,
it is possible to build more efficient motors of ANY sort, when designed
for the lower frequencies. In the old days, this was significant
enough to make the decision valid to use the lower frequency. In many
case the commercial companies had substantial LF generation capacity
to supply motor type loads, trolley company rotary converters
ferinstance. As most of the traction motors were series commutator
motors, the motor efficiency was seen as critical.
Plus, as pointed out, the RR's commonly had their own plants, so it
made sense to design to suit. With time, it became possible, then
economical, to build 50/60 Hz motors efficient enough to utilize the
industrial frequency. This allowed tapping the normal grid for power
(and added some complications around wave shape, single phase loading,
etc, etc...)
The cutting edge, currently, turns the AC supply to DC, then chops it
back to controllable frequency AC, allowing the use of induction motors
as traction motors. The same technique is used on some diesel-electric
locos, also.
thanks
dave pierson |the facts, as accurately as i can manage
Digital Equipment Corporation |the opinions, my own
600 Nickerson Rd
Marlboro, Mass
01752 pie...@cimnet.enet.dec.com
Lucius Chiaraviglio
Jones,201H MLH,3193350740,3193382879) writes:
>From article <27...@unisoft.UUCP>, by by...@unisoft.UUCP (John Byer):
>> ... What I don't understand, is that many of the
>> electric locomotives use 30-50K volts ac, (this I understand,)
Actually, not many electric locomotives use over 25000 V (exceptions
that come to mind are the Black Mesa & Lake Powell dedicated coal-hauling
short line in the U. S. and some long mineral-hauling line in the Karoo desert
in South Africa, both of which use 50000 V -- at 60 Hz in the U. S. and 50 Hz
in South Africa).
>> at some
>> wierd frequency, like 16 1/3 Hz (this, I don't.) Inasmuch as most ac
>> in the world is distributed at 50 or 60 Hz, why do they go to the
>> trouble to generate 16 1/3 Hz?
In France, Britain, the Soviet Union, and parts of Japan and South
Africa (and other parts of the world I'm sure) the standard has for some time
been 25000 V @ 50 Hz (note, however, that these countries also have
significantly large DC electrification systems of lower voltage -- these are
holdovers from earlier days, except for some extensions to existing DC lines).
In other parts of Japan (seems to have 2 different domestic/industrial
electrification systems -- figure that one out) and a microscopic part of the
U. S. the current standard is 25000 V @ 60 Hz (also 12500 V @ 60 Hz in the
U. S.).
Now the fun stuff comes in when we get into old AC electrification
systems. In Germany, Switzerland, Austria, Sweden, and Norway (and maybe some
other countries I can't remember right off hand) the standard was and still is
15000 V @ 16+2/3 Hz; in the U. S. the old AC electrification standard (present
in only microscopic areas, mind you) was 11000 V (or 11500 V) @ 25 Hz. Some
low-frequency 3-phase AC systems existed, but these are extinct except for 5
short tourist-hauling mountain railways (I think that it's 5 and that's what
they all are -- somebody correct me if that's wrong).
>There's a rational reason, don't worry. At low frequencies, induction
>motors run slower, and that allows them to use lower gear ratios and lower
>speed bearings in the drive train. In the early years of electrification,
>there wasn't much problem about compatibility with commercial 50 or 60 hz
>power, since everyone was used to using motor generator sets for voltage
>conversion in DC electrification, and AC power was not yet universal on
>the commercial front.
This is true only for the 3-phase electrification systems -- the
single-phase electrification systems used (and still use to a fair extent, in
the case of the low-frequency systems) DC motors running on AC current. The
reason for using low-frequency AC rather than DC was because transformers
don't work on DC, and DC motors using over 3000 V (and that only with divided
armatures and double commutators -- otherwise the limit is 1500 V) are
difficult to design. The reason for using low-frequency AC rather than
domestic/industrial-frequency AC is that DC motors function progressively more
poorly as they are fed AC current at higher frequency (I still don't
understand exactly why this is, but I think it has something to do with
worsening inductive kickback generating overvoltages at the commutator and
causing flashovers -- does this sound right, anyone?).
>Note that 16.666 is exactly 0.3 times 50hz.
Nit to pick: actually, you want 1/3, which is closer to 0.3333 than
to 0.3.
| Lucius Chiaraviglio | Internet: chia...@silver.ucs.indiana.edu
BITNET: chia...@IUBACS.BITNET (IUBACS hoses From: fields; INCLUDE RET ADDR)
Internet-gatewayed BITNET: chiaravi%IUBACS...@vm.cc.purdue.edu
Alt Internet-gatewayed BITNET: chiaravi%IUBACS...@cunyvm.cuny.edu
Douglas W. Jones,201H MLH,3193350740,3193382879
by chia...@silver.ucs.indiana.edu (Lucius Chiaraviglio):
> ... Some
> low-frequency 3-phase AC systems existed, but these are extinct except for 5
> short tourist-hauling mountain railways (I think that it's 5 and that's what
> they all are -- somebody correct me if that's wrong).
One would be the Gornergrat Bahn up from Zermatt. It's a narrow guage rack
and pinion railroad with dual overhead wires. The cars all have two
side-by-side pantographs on the roof.
Doug Jones
jo...@herky.cs.uiowa.edu
Bill Brown
>In article <5...@ns-mx.uiowa.edu> jo...@pyrite.cs.uiowa.edu (Douglas W.
>Jones,201H MLH,3193350740,3193382879) writes:
> This is true only for the 3-phase electrification systems -- the
>single-phase electrification systems used (and still use to a fair extent, in
>the case of the low-frequency systems) DC motors running on AC current. The
>reason for using low-frequency AC rather than DC was because transformers
>don't work on DC, and DC motors using over 3000 V (and that only with divided
>armatures and double commutators -- otherwise the limit is 1500 V) are
>difficult to design. The reason for using low-frequency AC rather than
>domestic/industrial-frequency AC is that DC motors function progressively more
>poorly as they are fed AC current at higher frequency (I still don't
>understand exactly why this is, but I think it has something to do with
>worsening inductive kickback generating overvoltages at the commutator and
>causing flashovers -- does this sound right, anyone?).
Well - sort of. Traction motors are typically series-wound motors which
will run on either DC or AC. The catch is that in the case of AC large
amounts of power go into creating eddy currents in the stator. These eddy
currents just generate heat; they produce no useful energy. In order
to avoid this problem the stator has to be made out of laminations which are
insulated from each other. The higher the frequency, the thinner the
laminations; therefore the more volume occupied by the insulating material.
Since the ammount of iron required for a given field strength is constant,
the motor has to get bigger.
Years ago there was quite a lot of 25 Hz power used in heavy industry.
The characteristics of series-wound, shunt-wound, and compound-wound
motors (which share the same field problems) are ideal for driving
certain types of loads. There are trade-offs however, since the lower
frequencies require more iron, and therefore more copper to go around
the greater volume of iron, in transformers. Also generating and
distributing 25 Hz power is an extra expense as it requires a seperate
physical plant from the normal 60 Hz or 50 Hz system. Now 60 Hz is
converted into DC where needed, or frequency-changers (usually solid-state)
are used.
Seems like we went thru this a couple of months ago. Should be in the
archives somewhere.
-bill
wlb...@lbl.gov
Disclaimer: These opinions are my own and have nothing to do with the
official policy or management of L.B.L, who probably couldn't care
less about employees who play with trains.
MEDBERRY MICHAEL G
and other western roads?
jobst_brand...@hplabs.hp.com
generators were large. In Europe, where 50 Hz is the norm, early
AC railroading used 1/3 of 50 Hz or 16 2/3 Hz for generating DC on
board for DC motors that drove the wheels through connecting rods.
DC, in contrast to AC, was not convertible from one voltage to another
and low voltage required high current. So in order to go long
distances, high voltage AC was used and because low speed equipment
was used in the notor-generator sets, 16 2/3 Hz came about. Today,
most European countries still run 15000v 16 2/3 Hz. Among these are
France, Germany, Austria, Switzerland, and others.
France is also a leader in 25000v 50 Hz lines whose current is compatible
with the national power grid at 50 Hz. While this is a reasonable
alternative, the 10000 VDC lines are also being developed with the
advantage that less power is radiated and inductive losses are avoided.
New solid state converters make DC convertible to traction motor levels.
At the same time single phase AC is being used to drive three phase motors
by solid state circuits. Some diesel electric three phase units are being
built by GE and Siemens in the US or Canada. France also runs a number of
mainlines with 1500VDC while FS Italia runs 3000 VDC. It's a big mess of
voltages, AC DC and catenary profiles. We should have it so good.
Lucius Chiaraviglio
>Does anybody know what the Great Northern or Milwaukee Road used in Montana
>and other western roads?
I might as well list all that I know. The Great Northern and some
railroad in Virginia (forgot exactly which) used 11500 V @ 25 Hz for their
electrified sections. It should be noted that the electrified section of the
Great Northern was a segment between 70 and 80 miles long spanning a tunnel in
the Cascade mountains, and was proceeded by a 3-phase electrification (don't
have technical data, sorry) which it replaced in the 1920's. The
electrification on both railroads is now defunct (as of the 1950's for the
Great Northern), but the formerly-electrified tracks are still in use (by
diesel-hauled trains). The Pennsylvania and the New Haven railroads used
11000 V @ 25 Hz for their Northeast Corridor lines from Washington D. C. to
New Haven, Conn. via Philadelphia and New York Penn Station, and also for the
Philadelphia to Harrisburg line of the Pennsylvania. This electrification is
still largely in use, but much of the New Haven electrification (somebody
posted here exactly how much of it a few months ago, but I can't remember) is
at 12500 V @ 60 Hz and/or 25000 V @ 60 Hz.
The Chicago, South Shore and South Bend railroad (now bankrupt) and
the Illinois Central use 1500 V DC for their electrified trackage (in the
Chicago vicinity only); earlier somebody posted on here about how the CSS&SB
used to have a 6000(?) V @ 25 Hz electrification in the 1920's and before.
The Milwaukee Road had fairly extensive electrification at 3000 V DC -- this
was removed only shortly before the whole railroad went under in the 1970's
(this should (but almost certainly won't) serve as a lesson to anyone else who
is considering being so short-sighted -- if they had just hung on they might
have been able to make a comeback using their electrification during the oil
crisis).
Lucius Chiaraviglio
MICHAEL G) writes:
>>Does anybody know what the Great Northern or Milwaukee Road used in Montana
>>and other western roads?
>
Bonehead! How could I forget the Black Mesa & Lake Powell
coal-hauling dedicated short line electrified at 50000 V @ 60 Hz (which I just
mentioned in another recent message), or the electrifications of the New York
Central and the Long Island RailRoad electrified by third rail at 660 V DC
(undercontact) and 600 V DC (overcontact) respectively. The latter two are
only in the vicinity of New York City and for sure aren't western roads, but
if I'm going to mention all I know I should put them in. Also, for
completness I should mention the Baltimore & Ohio No. 1 - 3 electric
locomotives (followed by more electric successors until dieselization in 1952)
which hauled trains to the Mount Royal station -- these ran on 675 V DC
(originally from overhead third rail; later from conventional third rail (I'm
not sure whether overcontact or undercontact)).
I should add that my information, when not obtained from ancient
rec.railroad postings or otherwise, comes from _The Great Book of Trains_ by
Brian Hollingsworth and Arthur Cook (some of it may also be found in _The
Illustrated Encyclopedia of Railway Locomotives_ by Robert Tufnell).
Mike Williams
Why 16 1/3 Hz? Seems a very peculiar frequency. Why not 15 or 20 Hz? Or
is the reason for this, like the reason for the exact widths of different
guages lost in the mists of time?
Bill Brown
>and other western roads?
The original GN electrification was 6600 V, 25 Hz 3 Phase. When the new
tunnel was dug the new electrification was 11,000 V, 25 Hz, single phase.
The dual-catenary required for the 3 phase system was a real problem.
The Milwaukee Road electrification was 3000 V DC.
From _When the Steam Railroads Electrified_ by Middelton (sp?). An
excellent book if you can find it. I got mine at the Roseville meet.
I understand that a revised edition is going to be issued Real
Soon Now.