Trains.com

Traction motors

5936 views
41 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    June 2002
  • 19,376 posts
Posted by daveklepper on Wednesday, May 11, 2022 4:44 AM

An AC commutator motor, as used in nearly all PRR and NYNH&H electric locomotives, exceptions being the New Haven's EF-2s (rotary converter and DC motors), and New Haven (ex-Virginian-N&W) and PRR post-WWII freight electrics and New Haven's "Jets" (EP-5s) which were all rectifier-DC motor types, is basically a DC motor and will work OK on DC as done on the New Hacvewn into GCT, but separate control equipment is usual (transformer taps for AC, with or without series-parallel transition, resistor insetion for DC, always inclkuding series-parallel transition). Cleverly desigmed "copmpensating coils" mitigate the loss of efficiencu from the alternating polarity.  Model train "universal motors," as typical for classic Lionel, are small variations of the same idea.  But no locomotives using such motors have been built since the early 1950s.  (Europe and North America)  25Hz is about the highest frequency such a motor built for heavy duty will tolerate.  If fed 50 Hz or 60 Hz, it will sit and hum and heat up. possdibly to self denstruction.  Small motors at 60Hz are possible.

Today's locomoitives, rapid-transit and  commuter and high-speed MU cars usually employ a variation of the "squirrel-cage" induction motor that is the non-synchronous hysterisis induction motor, where the rotating bars  are slanted with respect to parallel to the center shaft.  As with the sychronous motor, speed is determined by the frequency of the power to the field coils which number three or multiples of three.  But the slant-bar armeture does not rotate exactly at the speed of the rotating field but slips slightly, the  degree of slip depending on the load.  At overload, it simply won't turn before it destructs, unlike a brush-commtator motor or an AC slip-ring motor.  (If already turning before overload, it will simply not provide power and slow-down.)

Variations include rotating permanent magnets instead of rotating aluninum or copper bars.  Efficiency is greater, but so is weight.  I don't believe any North American builder uses this approch.

Other types are "inside-out" versions, the "wheel-motor." where the rotaring slanted bars or magnets are on the perimeter, and the multiple-of-three field coils form a non-rotating drum in the center.  Manfacturers include Alstom, Stored Energy Systems (Derby, England), and Magnet Motor (Pressberg, Germany).  They are used by most of the low-floor terminal-to-aircraft airport buses in Europe and by most Eropean trolleybuses.

All these locomotives and other vehickles take DC directly or rectiftify AC from catenary or alternator (AC generator) to DC and then use inverters to prioduce the 3-phase AC  at the right freqency and voltage to drive the motors.

  • Member since
    November 2021
  • 124 posts
Posted by JayBee on Tuesday, May 10, 2022 1:23 PM

The Siemens GTO Inverters used in EMD SD70MAC locomotives, use four different modes depending on locomotive speed. The first mode is "Asynchronous or Free Running" mode and is used from 0 to 14 mph, the switching frequency increases from 1 to 15 Hz. The second mode is called "Sinusoidal Switching" and operates from 14 to 28 mph, with output frequencies of 15 to 31 Hz. The third mode is ""Rectangular" for speeds from 28 to 35 mph using output frequencies from 31 to 38.5 Hz. Above 35 mph the Inverters are in "Block" mode and use Pulse Width Modulation, and frequencies from 38.5 to 120 Hz.

Power output from the Inverters is 2025 VAC, the full power rating of the diesel is available above 9.3 mph. Below that it is limited to control torque and wheelslip.

All of the above is from the EMD "SD70MAC Locomotive Service Manual 2nd Ed"

  • Member since
    January 2002
  • 4,509 posts
Posted by M636C on Monday, May 9, 2022 4:18 AM

Erik_Mag

 

 
Enzoamps

Never occurred to me they would have power oscillators driving the motors.

 

 

Do note that the switching frequency for locomotive inverters is a few kHz at most, where class D audio amplifiers are running 500 kHz or higher. The inverters provide a variable voltage variable frequency drive, as the motors want a relatively low voltage and high current at low frequencies, but higher voltages at high frequencies - where low frequency may be under 10 Hz and high frequency may be 60 Hz.

The Siemens inverters relied on varying the DC Link (rails) voltage when operating where the output frequency was the switching frequency, thus preventing use of pulse width modulation to vary the output voltage.

I first ran across Class-D amplifiers in a 1965 issue of Electronics World that I read in fifth grade.

 

The "Siemens Inverters" referred to were those in the earliest EMD locomotives. These used Gate Turn Off (GTO) Thyristors, and in electric locomotives these had a distinctive sound that might not have been audible in EMD diesel locomotives.

EMD later used Mitsubishi Insulated Gate Bipolar Transistors (IGBT), and these are being used in some rebuilds of earlier EMD AC locomotives.

Of course Siemens now make IGBT devices and EMD's choice was presumably based on lower purchasing costs rather than availability.

The actual motors can be driven by either type of inverter, so traction motors can be retained if GTO inverters are replaced by IGBT or later devices.

Peter

  • Member since
    January 2019
  • 1,313 posts
Posted by Erik_Mag on Saturday, May 7, 2022 10:32 PM

Enzoamps

Never occurred to me they would have power oscillators driving the motors.

Do note that the switching frequency for locomotive inverters is a few kHz at most, where class D audio amplifiers are running 500 kHz or higher. The inverters provide a variable voltage variable frequency drive, as the motors want a relatively low voltage and high current at low frequencies, but higher voltages at high frequencies - where low frequency may be under 10 Hz and high frequency may be 60 Hz.

The Siemens inverters relied on varying the DC Link (rails) voltage when operating where the output frequency was the switching frequency, thus preventing use of pulse width modulation to vary the output voltage.

I first ran across Class-D amplifiers in a 1965 issue of Electronics World that I read in fifth grade.

  • Member since
    January 2002
  • 4,509 posts
Posted by M636C on Saturday, May 7, 2022 6:52 PM

timz

Speaking of AC motors --

Does any RR in the world still use straight-AC locomotives, that run over a normal speed range sending constant-frequency AC to the traction motors, like a GG-1? I assume no one builds such locomotives now; when was the last one built? Did such engines always need 25 Hz or less in the catenary?

 

Although it is some time since I've been to Europe, I think the last locomotives with AC commutator motors built in Germany were the Br 111 and in Switzerland were the Re6/6. Both these classes may have different classifications now, in Germany post the absorption of East Germany and the DR and in Switzerland due to the adoption pf six digit numeric classes. I think these would have last been built in the 1980s. I think both these types and some older locomotives are still in use in Germany and Switzerland. There may be similar locomotives in service in Norway but possibly not in Sweden now.

All the locomotives described here worked on 16.66 Hz, 15kV.

The Germans and French built locomotives that could operate using commutator motors on the grid frequency of 50Hz, at 20kV or 25kV respectively. These were experimental and not adopted for general use (although the French tried).

The main problem involved the commutators on the motors as at higher frequencies the current tended to "flash over" causing a short circuit and damaging the commutator. There was more deterioration at the commutator with the higher frequency, even if flashovers were avoided.

Peter

  • Member since
    April 2015
  • 469 posts
Posted by Enzoamps on Saturday, May 7, 2022 4:42 PM

Thanks.  I totally do understand class-D, and so on as my career was in electronics, and specifically professional Audio.  Just I rarely encountered power amplifiers in excess of 4000 watts.   Never occurred to me they would have power oscillators driving the motors.

  • Member since
    February 2005
  • 2,147 posts
Posted by timz on Saturday, May 7, 2022 3:45 PM

Speaking of AC motors --

Does any RR in the world still use straight-AC locomotives, that run over a normal speed range sending constant-frequency AC to the traction motors, like a GG-1? I assume no one builds such locomotives now; when was the last one built? Did such engines always need 25 Hz or less in the catenary?

  • Member since
    January 2019
  • 1,313 posts
Posted by Erik_Mag on Saturday, May 7, 2022 12:25 PM

[N.B. My reply was being typed about the same time that OM was replying).

A more detailed description is that the latest locomotive traction inverters act pretty much like a Class-D audio amplifier. That is the outputs are rapidly switched between the positive and negative "rails" (or buses) of the DC supply with varyiing intervals of being connected to the positive and negative rails.

The "switches" are a key point of inverter technology as a locomotive inverter needs to supply at least a good fraction of a megawatt. The original Siemens inverters used on the EMD AC locomotives used Gate Turn Off thyristors (GTO's) that were incapable of switching at much more than line frequency, at higher motor speeds the inverters were switching at the drive frequency. The gate drive circuitry for GTO's is very substantial, so EMD shared the inverter with the three motors on the truck. In the mid to late 1990's, high power IGBT's (Insulated Gate Bipolar Transistors, a modification to FET's) with MUCH simpler gate drive requirements. This allowed GE to use an inverter per motor, which gives finer control over the traction motors.

For line frequency induction motors, there is a tradeoff between starting torque and efficiency (one work around was wound rotor induction motors as used on GN's three phase Cascade tunnel electrification, the N&W and VGN phase splitter locomotives. With a variable frequency drive, it is possible to design the induction motor for high efficiency and still have high starting torque. In fact, the high efficiency design helps with adhesion as motor torque drops rapidly when the wheels start slipping.

  • Member since
    September 2003
  • 19,652 posts
Posted by Overmod on Saturday, May 7, 2022 11:49 AM

The situation with AC motor control is complicated if operation with other locomotives using a form of 8-notch MU or DPU is needed.  These control the speed of the Diesel engine.  The engine governor controls fuel and hence developed horsepower as load 'in notch' increases and decreases; it is not like a typical car where engine speed is used to vary road speed.

If you were to vary engine speed in some proportional synchronous speed of an induction motor, you'd have ghastly low-end torque -- these motors self-induce a current in a conductive but non-initially-magnetic armature and then use 'rotating' fields to pull the 'induced' armature around.  A slow enough speed on the motor/alternator to do this at slow road speed would lug or stall the engine, unless you used a great multiplicity of poles as in an Alexanderson machine... sized and insulated for the peak current and protected from flashover issues between near-adjacent poles in the structure...

It might be possible to use a mechanical variable-speed drive, like the Bowes drive or a Beier gear with a great multiplicity of discs, but you would still not have the ability to utilize full engine power-in-notch at any physical road speed -- which a synthesized-AC drive can do very well.

The 'traditional' method of generating ripple-less DC from any of the alternator frequencies requires what might be considerable current and voltage storage -- inductors and capacitors, in regular DC motor control.  Erik knows more than I do about current best practices (no pun intended) in producing flexible near-synchronous torque from synthesis.

Note the difference between varying rotational speed of an alternator and accelerating a locomotive whose alternator is constantly turned at proper frequency generation for HEP (on traditional GE and EMD engines, that is frequently (again no pun intended) 720rpm.  At that speed, the engine can produce power with field excitation as quickly as the governor can get the engine to hold rpm with applied load (the engine and alternator rotor having substantial rotational inertia) so loading and subsequent amp generation for rapid acceleration is 'on tap'... just not up to the full horsepower the engine could develop at its nominal eighth-notch speed (which might be 1050rpm on a GE marketed to compete with a comparable-horsepower EMD 2-stroke)

You could approximate multiple-speed control of induction motors by changing the mechanical number of poles 'active' in the motors (as was done for some early AC electric locomotives) or in an alternator.  There are reasons this was not and is not common practice for modern power...

  • Member since
    March 2016
  • From: Burbank IL (near Clearing)
  • 13,014 posts
Posted by CSSHEGEWISCH on Saturday, May 7, 2022 9:56 AM

Enzoamps

AC motor variable frequency?  Determined by the rotational speed of the engine/alternator?  Or is it a more complex system?

 
It's a bit more complicated.  The main alternator feeds AC into rectifiers which convert the current to DC.  The DC current is then fed into inverters of various designs which convert it to variable frequency AC.  Somebody else can provide the particulars of how this is done.
The daily commute is part of everyday life but I get two rides a day out of it. Paul
  • Member since
    April 2015
  • 469 posts
Posted by Enzoamps on Saturday, May 7, 2022 5:29 AM

AC motor variable frequency?  Determined by the rotational speed of the engine/alternator?  Or is it a more complex system?

  • Member since
    January 2019
  • 1,313 posts
Posted by Erik_Mag on Friday, May 6, 2022 11:06 PM

DC traction motors are almost always series motors, exceptions including some European electrics had separately excited field windings (e.g. AEM-7). "Separately excited means that the current for generating the field is from a separate source of power than what supplies the armature. A large fraction of DC series traction motors had field taps or shunts to provide for reduced field (higher speed) operation in lieu of the separately excited field.

The DC motors do indeed use brushes, which has been a maintennance issue since the 1880's.

Most AC traction motors are three phase induction motors driven by a variable frequency drive. Some are synchronous motors, but those require a bt more complicated control scheme, but have the advantage of being more efficient at very low speeds.

  • Member since
    April 2015
  • 469 posts
Traction motors
Posted by Enzoamps on Friday, May 6, 2022 10:00 PM

I know something about small motors, like model trains, or recording equipment (tape decks, disc drives), juke boxes, etc.   But reading about locomotives I find I know squat.  I can assume a locomotive traction motor is not a permanent magnet type, but do they use brushes?  SLip rings?  The stationary windings are easy to figure.  But how does energy get into the rotor?   And assuming the AC motors are not induction types (like an alarm clock) same questions.

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy