All motors are generators, right? 3

[Renovator1]

Sorry about wandering over from the circuit engineering section, but something occurred to me that I could not find the answer to using google...

If all motors can be generators (and I can clearly understand how even an induction motor can be a generator - simply apply a lower frequency to it than it is currently spinning at), how, exactly, can a shaded pole motor be a generator? Is it simply a matter of over-speeding the rotor, as in a conventional induction machine, relying on residual magnetism to initiate a then self-reinforcing magnetic field build-up?

A bit obtuse of a question, I know, but it really has stumped me this evening!

 

[davidbeach]

I wouldn't buy the supposition that all motors can be generators. Synchronous motors, DC motors, and 3-phase (or more if such exist) squirrel cage induction (asynchronous) motors can function as generators. There are all manner of single phase motors that would never work as generators.

 

[electricpete]

Single phase motors have a variety of means to generate rotating field for purposes of starting (shaded pole is one means). Once they are started, the reverse rotating field effects diminish due to the high rotor impedance at those high frequencies... acts similar to a 3-phase motor. If a 3-phase motor can act like a generator, why a 1-phase motor?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

I guess we should mention the induction generator typically draws its magnetizing current from the line.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Renovator1]

Thanks for the replies so far, folks. Note that I am not contemplating actually using a shaded-pole motor as a generator, just trying to understand how it *theoretically* could be used as one!

davidbeach - I'm sure there are motors which are not practical to use as generators, but according to Faraday, they'll still work as such. More specifically, my understanding is that *any* induction machine will send current back to the AC line whenever the rotor is spun faster than the synchronous speed and that maximum current occurs in either motoring or generating mode at the "slip" speed (either below synchronous for motoring or above synchronous for generating). Have I got that right or have I bollixed something somewhere?

electricpete - the magnetizing current lags the applied voltage by 90 deg. when an induction machine is unloaded but comes into phase with the voltage as the load approaches maximum, correct? This would apply whether the machine was a generator or a motor, I would imagine. Thus, most of the "magnetizing" current in generator mode would go back to the line and only a small amount would be consumed fighting hysteresis/eddy/whatever losses (e.g. - those things that determine the efficiency).

I've just about wrapped my head around how even a series-wound DC motor can function as a generator (briefly short the motor terminals together while spinning then "catch" the inductive kickback, a la a boost converter), but the shaded pole motor really eludes me...

 

[waross]

A series generator will run a series motor at a fairly constant speed.
The magnetizing current stays 90 deg. out of phase with the applied voltage. The vector sum of the load current and the magnetizing current approaches unity as the load increases but the magnetizing component of the current stays at 90 deg. to the voltage.
I suspect that any motor that limits the load current by generating a back EMF will regenerate. In some special motors the power regenerated may be very little. An induction disk kWHr revenue meter comes to mind.
Stepper type motors or other electronically driven motors may not function as generators at first consideration. However, inductive kick is an indication of back EMF and any motor that generates an "inductive kick" is trying to return energy to the supply.
A shaded pole motor will regenerate energy but given the high resistance of the winding and the load of the shading coil, they may not be able to generate enough to suply their own losses. Even an efficient induction generator may have an operating region where the generated power is less than the losses. Some small motors may never get out of this operating region where losses exceed generated energy.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Renovator1]

Talk about an exercise in futility: using an eddy current "motor" as a generator (induction disk kWHr revenue meter)! Still, it sounds like something the over-unity nuts will want to try, doesn't it? (all you have to do is spin a disk of metal in between the windings and it will magically produce electricity!)

Anyway, thanks for your reply, Bill. It is now starting to make sense to me how a shaded pole motor can act as a generator, albeit a crappy one. The shorted turn that produces the salient pole, and leads to this type of motor being so inefficient as a motor, would also "rob" most of the energy when operated as a generator. Correct?

 

[Skogsgurra]

For the record:

The inductive kickback is not proof that the motor is able to generate. Every coil or other inductive circuit produces a counter-EMF when energized and an "I don't like to be interrupted" voltage aka kickback when de-energized.

The generating mechanism is something quite different and depends on voltage generated by the rotating rotor field.

I think that it is important to draw a clear line between these two phenomena.

The shaded pole motor is not able to produce any energy that you can use. With efficiencies down in the 10 % region, one cannot hope to get any useful energy out of them. All that can be observed is that power consumption is slightly reduced when such a motor is driven over synchronous speed. A good illustration of what Bill said: "Some small motors may never get out of this operating region where losses exceed generated energy"

A stepper motor has a PM rotor and can be an efficient generator.

The SR motors probably also can generate since they run synchronously and every attempt to make them move faster should produce a braking torque, which means that power is absorbed by the shaft and therefore must be delivered to the mains. If losses do not consume that power.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Renovator1]

Hey Gunnar... I realize that *practically speaking* a shaded pole motor cannot be used as a generator; I was instead wondering how it *theoretically* could act as one.

Also, just to be clear, when I referred to "inductive kickback" in a previous post it was in regards another not-so-suitable motor as generator, the series dc motor (but for very different reasons from the shaded pole motor).

As I understand it, the way you get a series dc motor to act as a generator is to short it out periodically as if it were the inductor in a boost converter. The transfer function for the series dc motor is pretty non-linear so things can get out of hand rather quickly but, non-intuitive though it seems, it theoretically can be a generator...

 

[Skogsgurra]

If you leave the losses out, you have the same rotating field in a shaded pole motor as you have in a three-phase motor. And the same voltage induced in the windings. So, it is not a question of different principles - just more or less good implementations of the induction motor principle.

What you say about the series motor and shorting it like the inductor in a boost converter is not anything I would subscribe to. It works just like any DC machine, albeit not so easily controlled as a shunt wound or separately exited machine.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[waross]

Hi Gunnar;

The inductive kickback is not proof that the motor is able to generate. Every coil or other inductive circuit produces a counter-EMF when energized and an "I don't like to be interrupted" voltage aka kickback when de-energized.

The generating mechanism is something quite different and depends on voltage generated by the rotating rotor field.

I think that it is important to draw a clear line between these two phenomena.

I happily accept your correction of my statement.
Re the stepper motor, I was considering the motor and the drive electronics as a unit that may not be able to accept regeneration. Again I stand corrected. The motor itself may be a fairly efficient generator.




Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[cswilson]

Both the Lorentz torque generation mechanism and the reluctance torque generation mechanism are fundamentally reversible, and so devices using them can act as either motors or generators.

The common stepper motors with permanent magnets are really AC synchronous motors, and so can generate as easily as they motor. (Superior Electric markets the same motors as stepper motors with 4 leads or "slow synchronous" motors with 3 leads.) Many stepper users do not realize this and are unprepared for the generated voltage on a fast deceleration. Typical stepper drives these days are able to handle this well, and it is seldom that even a shunt resistor is needed.

Switched reluctance (SR) motors, which are really the same thing as variable reluctance (VR) steppers, also can generate, and with efficiencies that exceed 90% (I have seen papers proposing their use in dedicated generating applications), but there are some interesting issues.

The direction of torque generated from current in an SR motor phase is not dependent on the direction of current in the phase. Before magnetic saturation, the torque is proportional to the square of the current (which eliminates the sign dependency). So to generate torque in the direction you want, you should only have current on during half of the cycle. Any current in the other half of the cycle-- even current in the opposite direction -- will work against you. Of course, this means that SR motors cannot work from the AC mains at all.

So, to decide whether you will use the SR motor as a motor or generator, you are fundamentally just deciding which half of the cycle you will be driving current through the phase.

To generate effectively, the power electronics must provide a proper path using flyback diodes to permit the current to "pump up" the DC bus capacitors. What is then done with this regenerated energy is dependent on the application, and is now the same as with any electronically controlled motor.

Curt Wilson
Delta Tau Data Systems

 

[waross]

Thanks Curt. lps

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Renovator1]

Excellent explanation of the SR motor/generator behavior, Curt. Now that finally makes sense to me, too. And since the shaded pole motor has been dispensed with, how about your take on my understanding of how a series dc motor can be a generator? Gunnar is not the only one who is skeptical of me and I have found precious little written on the subject.

To easily flip back and forth between motoring and regenerating modes with a series wound motor I envisioned the setup shown in the attached file, where instead of the traditional "chopper" configuration (switch and FWD in series, motor in parallel with FWD), a half-bridge is used with FWDs across both switches. When the switch in series with the motor is width modulated the motor is a motor (buck converter); when the switch in parallel with the motor is width modulated the motor is a generator (boost converter). This makes perfect sense to me conceptually but I have yet to see a book on power electronics and/or motors even explain it as much as I have in this measly post. In fact, all I have seen on the subject is the occasional "it's really difficult to use a series dc motor as a generator"... Okay, the same curves that imply infinite speed at zero load or infinite current at locked rotor also imply some really wild behavior when such a motor is a generator, but is the problem simply a matter of loop compensation being difficult or is there some other more fundamental/practical issue with using the series motor as a generator. E.g.: if a 200VDC bus is bucked down to 100V at the motor and this results in an average motor current of 100A then why wouldn't it be possible to flip over to the boost mode and turn that into 200V at 50A flowing back into the DC bus? Neglecting I^2R losses, etc, of course.

 

[waross]

A Series wound DC generator is pretty much unusable.
Given enough residual magnetism to start generating, the output voltage is roughly proportional to the current.
When a load is connected:
As the current increases the field strength increases.
As the field strength increases the voltage increases.
As the voltage increases the current increases.
As the current increases the voltage increases.

This runaway may be limited by saturation of the magnetic circuit, stalling of the prime mover, or electrical or mechanical failure of the load.

A generator may be compounded by adding a series winding. This gives a rising voltage with loading. This may be used to compensate for line voltage drops, particularly if the feeders from the generator to the load are very long.

Because of its inherent characteristics, a series generator has almost no practical application. I can not recall ever seeing a series generator. I have encountered heavily over-compounded welding generators for special applications.
I suppose that in smaller sizes a series generator may be made more usable by design techniques such as higher than normal winding resistance and/or designing the machine to use magnetic saturation or a variable speed drive as a limit but I can't recall seeing an application.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Skogsgurra]

Renovator. Can't see what you are trying to achieve in that circuit. Field winding connected to a dangerously primitive transistor arrangement and armature not connected at all. What is the purpose of that?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Renovator1]

Skogsgurra - that was a quick drawing I did in my EDA software that does not, alas, have a nice symbol in it for a motor, so I cobbled one together out of circle and an inductor. Just try to pretend that's a series motor with the field and armature connected, well, in series.

waross - what you described is exactly what I imagined would occur. So, it would seem to me that as long as the control loop response time is much faster than the L/R (or mechanical?) time constant of the motor then it should, theoretically, be possible to tame this beast.

As to where you might find, and actually want to use so ill-suited a motor as a generator? To allow regeneration in an electric vehicle...

(and yes, I know that the induction motor gives you regeneration for "free" (just command a lower electrical speed than the rotor speed); series dc motor systems still cost about 1/4 what an AC system does.)

 

[Skogsgurra]

OK, Reno. A short wire between the two would have been useful.

Traction motors in locomotives are series wound. And they regenerate in many cases. The resistors used in older machines are there for starting (and for controlled braking).

Controlling a series wound DC generator is probably as easy (not) as controlling an inherently unstable air-plane. It can be (and is being) done. But it can get very nasty when out of control.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Renovator1]

Yah, yah, I hear ya.. I need to make better symbols for my EDA software.

This actually sounds like a really interesting project to try out - seeing if a series dc motor can be made into a reliable generator, that is. Most (all, I think) SMPS topologies present a constant power load to a source so you ought to be able to tamp down some of the squirrelly behavior of the series DC motor by melding the two together. Keeping the regen current to a fraction of the motoring current should make things easier, too - much as you don't want to either stall a series motor nor let it run unloaded (confining operation to a much narrower part of the transfer function curve, that is).

Am I on the right track here or what? Surely somebody has done this already... any references would be most interesting.

Now, about the eddy current motor used in kW/hr meters...

 

[waross]

Now, about the eddy current motor used in kW/hr meters...
An induction motor with an eddy current brake.

Bill
--------------------
"Why not the best?"
Jimmy Carter