Three phase drive motors

[cheesysmith]

 

This is a picture of a AC sine wave. As you can see it flows up and down. The square one overlaid is what a chopper/Gate Turn Off (GTO) thyrister would make. A modern electric unit, like a electrostar would take the AC coming in, and rectify it to DC (1500 volts IIRC). This then goes through the chopper controls to make the squared off wave for use by the motors. The motors speed would be controlled by altering the frequency (the gaps between the peaks) and the voltage (the height of the peaks). The actual motors are 3 phase, so you would have 3 waves overlapping. A unit on 750v DC would just have a transformer to up the volts to 1500v DC and feed the DC link directly. If you put another set of choppers at the transformer/25KV end, you can then run it in reverse and put power back into the OHLE, which is how regenerative braking works.

 

 

 

The chopper controls are used because if you only had a fixed frequency, the motors would only run at a set speed. With modern computer controls and electronics, the above is simple. Also, the 3 phase/AC motors used have 1 moving part and no brushes to wear out/short out/replace.

 

This system is used in the 68s and the refurbished 73s. The diesel engine/alternator replaces the AC incoming, and just supplies the DC bit directly. On the 73s, the DC motors at supplied by choppers that only do half the wave, just like the DCC chips do for your models. This is how the system works on the dual voltage DC motored 319 units. The 313s were actually just a DC unit with a transformer that took the 25kv AC and turned it into 750v DC for the traction systems.

 

If you go further back, you get thyristor control drives, as pioneered on BR by the 87/1. These only allow power to flow one way through them (lets just say the positive half of the AC sine wave). This is the same as what a rectifier does. But a thyristor can be turned on and off so only allowing part of the power through. Imagine you turned it off at the start of the positive wave, and only allowed 1 volt through. Then you allowed it to open for twice as long, so you would have 2 volt etc etc. They can do this very quickly, and as the incoming AC is at 50hz, you can have them turning off/on upto 50 times a second. The motors need to be slightly different because the DC power coming into them isn`t smooth/pure, but a series of DC pulses upto 50 times a second. This is been used on BR since the 317 units, and in the 90s and 91 locos.

 

I apologise if some of the terms used in this are wrong, but I ain`t a electrical engineer, and have just been trying to make a very simple explanation of how modern units, like the electrostars or the desiros are dual voltage. I`m sure someone will be along to correct any terminology I have got wrong. 

[Nearholmer]

If you have a look at the Wikipedia entry for variable frequency drives you will find that quite good.

 

A good place to start in this topic is with a few basics:

 

- two motor types yield near ideal torque vs speed relationships for traction, the dc series motor, and the multiphase induction motor;

 

- to control the speed of the dc motor, control the input voltage;

 

- to control the speed of the ac motor, control the input frequency;

 

- the Polyphase motor can be made lighter and more compact for a given duty, so is preferred, but until the maturity of power electronics, it was seriously difficult to obtain control of input frequency, so polyphase ac motors were only very rarely used for traction.

 

once one has got hold of those points, you can build up understanding of the many variants.

 

Kevin

[Zomboid]

I doubt electrostars and the like use 1500V DC, as there's no way to transform DC to a higher voltage, it would have to be inverted first. It be much more likely that the inverter runs on a nominal 750V DC so that on DC the line voltage can directly feed the inverter.

[cheesysmith]

I haven`t been able to find what the voltage of the DC link is, but the dual mode electro diesels built for america that are based on the TRAXX platform up the DC voltage to 2260v for the traction package inverters.

 

PS-No way to up the voltage from the DC line voltage, what does a transformer do?

[cheesysmith]

Something else I forgot to say was that with modern AC drives, if you know the actual speed of the train, and the size of the wheels, you can control the wheel speed directly with the electronics. This means wheel spin/slide would be impossible due to the fact the motor would only be able to turn as fast as the electronics would allow. Of course, in real life the speed of the control electronics and the precision required are different from reality.

[Zomboid]

Transformers only work on AC voltages. That's because they use the rate of change of flux in the core, and a DC voltage would put a constant flux on the core, so nothing would be induced in the secondary winding.

I don't really remember all of it, the important thing is that a transformer only works on AC.

[Nearholmer]

Cheesy

 

The control of wheel speed can be so precise as to allow controlled micro-slip, where the wheel is moving ever-so-slightly faster than the rail, which yields maximum tractive effort.

 

See about 2m in, although worth watching it all https://m.youtube.com/watch?v=bdLcipx37b8

 

K

[cheesysmith]

Sorry, i meant to say for DC you can have a step up converter, not a transformer.

 

Must say to myself 1000 times, "I`m a dumb ass"*

 

 

 

*too much big bang theory

[Edwin_m]

As mentioned you can hear the inverters increasing the motor frequency as the train accelerates and creating a rising tone, which then drops suddenly as if changing gear.  This happens when the inverters are running as fast as the electronics can cope with, or are about to hit a frequency that might interfere with signalling.  The inverters then start over again from the bottom of their frequency range, now generating a sine wave approximation at a higher frequency but with fewer on-off steps per cycle. 

 

The opposite happens when slowing down with dynamic braking.  This is most evident on the earliest three-phase drives (Networkers, 323s, Jubilee Line) where the limitations on Gate Turn Off thyristors mean they have to operate at audio frequencies.  Post-privatisation units were able to take advantage of the development of Insulated Gate Bipolar Transistors, which can work much faster so the inverter operates at frequencies that are too high to interfere with the signalling. 

[jjb1970]

Solid state inverters and variable frequency drives have transformed so many things. When I entered the work force in the early 90's it wasn't an especially exotic technology and we did the theory as it stood at the time at college but the cost of the drives restricted use to pretty high value applications. Now you see VSDs used for valve actuators and they're everywhere. I still remember the Ward-Leonard system of getting a variable speed motor drive from a fixed frequency AC input, anybody else remember those?

[Northroader]

The big plus in changing from dc motors to three phase is that you lose the commutator. The windings are set round a rotor and fed through copper bars around a drum type arrangement. Carbon brushes then feed the current through the bars. This needs regular maintenance, and has an inherent centrifugal weakness at higher rotational speeds. With a three phase, the windings are set in the outer part of the motor with permanent connections, and the rotor is a much simpler and stronger construction without the need of current feeds.

[Pannier Tank]

Solid state inverters and variable frequency drives have transformed so many things. When I entered the work force in the early 90's it wasn't an especially exotic technology and we did the theory as it stood at the time at college but the cost of the drives restricted use to pretty high value applications. Now you see VSDs used for valve actuators and they're everywhere. I still remember the Ward-Leonard system of getting a variable speed motor drive from a fixed frequency AC input, anybody else remember those?

I worked with Solid State Inverters for Induction Heating in the Steel Industry where we used SCR's and more recently IGBT Technology so found this article most interesting.

[Nearholmer]

JJB

 

Having started work in the mid 70s, I encountered Ward Leonard systems, and the even more fiendish Schraeger(sp?) motors, then thyristors, GTO, and latterly IGBT. Rotary converters, Mercury arc rectifiers, four-quad systems ........

 

In a way, it has been a good period to work in, because 1900s tech that was installed in the 1940s crossed over with the intermediate technologies of the sixties/seventies, before settling into digitally-controlled power electronics. In a way a bad time, because I freely confess that I couldn't get my head around every one, and keep it around them all!

 

Kevin

[jjb1970]

One of the odd ironies of the availability of relatively cheap and reliable solid state inverters and frequence control is the trend back towards DC power systems in marine applications. A variable speed AC generator delivers an output which is converted to DC for a DC power bus with consumers supplied through inverters from the DC bus. That uncouples engine speed from being fixed to whatever speed is required to maintain frequency, allowing better engine efficiency and lower emissions, with the DC bus offering good distribution efficiency. At one time using so many inverters and solid state conversion would have negated (and more) and efficiency gains but now it is very attractive.

[jjb1970]

One of the funny things about modern DC power systems is asking the HSE people what amendments they intend to make to safety procedures when there are no breakers to lock open or even rack out as all the switchgear is solid state too. Good fun that.

[royaloak]

Cheesy

 

The control of wheel speed can be so precise as to allow controlled micro-slip, where the wheel is moving ever-so-slightly faster than the rail, which yields maximum tractive effort.

 

See about 2m in, although worth watching it all https://m.youtube.com/watch?v=bdLcipx37b8

 

K

Isnt that what the class 60s have which allows them to shift such heavy loads from a standing start?

[Nearholmer]

I don't know the details for individual classes, but I'd be surprised if any modern heavy-haul loco didn't have it, plus high-tech sanding gear to go with it.

 

K

[Zomboid]

60s have (or had) creep control as I believe it was called when the magazine feature I read it in was written ("World of Trains part set, early 90s).

[jjb1970]

Didn't the Class 60 use seperately excited traction motors with individual controls to optimise traction and minimise slip, under the Brush SEPEX name? Whilst modern AC power systems have facilitated all sorts of clever performance optimisation techniques it is good to remember that older DC traction packages were also capable of doing some pretty clever stuff.

[Davexoc]

60s have (or had) creep control as I believe it was called when the magazine feature I read it in was written ("World of Trains part set, early 90s).

The 59s had similar. Each axle provided a feedback as to has fast it was rotating, which was compared with the loco speedo (radar type) and depending on the power demand, tried to turn each axle at road speed plus a bit. One early test had the 59 effortlessly pull a train out of Merehead including the two 56s DIT, and the rest is history...

 

Didn't the Class 60 use seperately excited traction motors with individual controls to optimise traction and minimise slip, under the Brush SEPEX name? Whilst modern AC power systems have facilitated all sorts of clever performance optimisation techniques it is good to remember that older DC traction packages were also capable of doing some pretty clever stuff.

IIRC 58050 was the SEPEX testbed

 

Dave

[Davexoc]

If you go further back, you get thyristor control drives, as pioneered on BR by the 87/1. These only allow power to flow one way through them (lets just say the positive half of the AC sine wave). This is the same as what a rectifier does. But a thyristor can be turned on and off so only allowing part of the power through. Imagine you turned it off at the start of the positive wave, and only allowed 1 volt through. Then you allowed it to open for twice as long, so you would have 2 volt etc etc. They can do this very quickly, and as the incoming AC is at 50hz, you can have them turning off/on upto 50 times a second. The motors need to be slightly different because the DC power coming into them isn`t smooth/pure, but a series of DC pulses upto 50 times a second. This is been used on BR since the 317 units, and in the 90s and 91 locos.

 

I apologise if some of the terms used in this are wrong, but I ain`t a electrical engineer, and have just been trying to make a very simple explanation of how modern units, like the electrostars or the desiros are dual voltage. I`m sure someone will be along to correct any terminology I have got wrong. 

Just to clarify, early thyristors were just triggered diodes. A diode only passes during the positve part of the cycle, it switching off as the voltage dropped below a certain threshold. Thyristors did the same, so to limit the 'on' time you switch them on later in the positive cycle, because it will not switch off again until the voltage drops below that minimum. The real progress was the GTO (gate turn off) thyristors which could then be switched on and off anywhere in the positive cycle.

 

Dave

[royaloak]

I don't know the details for individual classes, but I'd be surprised if any modern heavy-haul loco didn't have it, plus high-tech sanding gear to go with it.

 

K

As I understand it, the class 66s WSP doesnt work until they are doing about 5mph so isnt much use when trying to get a heavy load away from a standing start, its another shortfall/cost cutting exercise the class suffers from.

[Edwin_m]

All of classes 59, 60 and 66 have DC motors with creep control.  I believe EMD's view at the time was that the (then) more sophisticated and expensive AC drive being fitted to many of their American products wasn't necessary for UK duties. 

 

If I recall correctly the logic went like this.  With a huge freight train on many lines in the States the number of locomotives needed was determined not by the power necessary to keep up to speed across the prairies, but by the tractive effort requirement to climb the occasional severe gradient.  The better low speed adhesion of AC motors might mean the train could still climb the gradients with one locomotive fewer, but with less power available it would also climb more slowly .  The cost saving of fewer locos would often outweigh the slight increase in journey time on the hilly sections.  However on a mixed traffic railway such as the UK very slow hill climbing would normally cause unacceptable delays to passenger trains, and with few trains needing more than one loco there was no saving to be had there. 

 

The class 68 and 88 locomotives do however have AC drives - the cost balance may have changed in the meantime and DC drives are probably no longer available anyway. 

[Zomboid]

68s (and 88s) are European technology, which will have an impact on the thought patterns of the designers add opposed to the American approach. And they're not really heavy freight haulers in the way a 66 is supposed to be. 68s at least strike me as what the 67s should have been.

[cheesysmith]

The idea behind SERPEX motors was to provide better control of DC motors. Unlike model motors which use magnets, the motors on trains use electromagnet windings on both the commutator (the bit in the middle that rotates, so requiring brushes to transferes the power to then, just like the ones in and model train*). SERPEX applied computer control to the power applied to it and the windings around it at the same time, so the motor could not run away, and create uncontroled wheel slip. This was worked out using ixoin, the class 46 taken into RTC use.

 

The wheel creep system used by GM allowed the wheels to rotate slightly faster than the actual speed of the loco, and increased the percentage grip/tractive effort of the loco. When brush introduced the class 60s, GM asked how they got the engine speed to respond quickly enough with the power demands for the creep system, as the 4stroke system could not alter power as fast as the 2 stroke engine used by them. Brush just pointed out that the power demands of the SERPEX motors was independent of the power of the engine.

 

 

*there is some German model trains, the name escapes me, that use some complex electronics, that actually have AC motors.