Auto Alternators As Motors

[jimraysr]

50 / 60 Hz

Many Electrical items have been rated for both 50 or 60 HZ and yes they run slower on 50 HZ. As I recall they were de-rated when running on 50 HZ. 2 pole motors should be 3600 RPM on 60 HZ, but are generally listed as 3450 RPM. 4 pole motors should be 1800 RPM but are listed at 1725 RPM and finally 6 pole motors 1150 RPM rather than 1200 RPM.

By the same logic, 50 HZ motors would run at 3000, 1500 and 1000 RPM synchronous speed. I have no experience with 50 HZ motors.

Much of the South American power grid was (and maybe still is) 50 HZ (or 50 cycle as was the common term)

It sounds like in fact that there are 3 phase RC controllers which could operate a three phase alternator if it has the current capacity.

Do the sensorless three phase RC motors have 6 leads?

The whole point of poly phase motors is to come closest to having a continuous push like DC motors have. Three phase motors see three lines feeding them and the sign wave of each phase is 120 degree apart. It is getting three pushes on each revolution. A washer or cooler motor (swamp box for desert rats for cheap cooling) has only one phase in it, but most have a start capacitor which gives the start winding a 2nd push with a phase shift from the capacitor and it is taken out of the circuit by a mechanical switch or a relay operated by back EMF operating a relay. Permanent split capacitor motors have a 2nd capacitor which is connected all the time to the aux. winding which is the same as the start winding. They may or may not have a starting circuit.

I am throwing all this AC motor stuff in here as it does apply if one was to motorize a three phase alternator. Electrical laws are universal and not subject to interpretation by a judge. )

Jim

 

[arceeguy]

Sensorless RC motors have three leads. The field coils are conected in a delta or wye configuration. Sensored RC motors hae the same three leads to the field, along with an additional connector for the hall effect sensors. The sensored motors have superior low rpm performance since the controller knows precisely what position the rotor is in. Sensorless systems use back EMF to determine rotor position and complicated firmware algorithms for acceleration. Early sensorless systems were prone to cogging and stuttering under acceleration, but todays controllers run a lot smoother.

 

[jimraysr]

Rc Motors

OK, vedy interesting! No mention of a power lead (s)? So are they rotating the fields and the motive power is constant?

Jim

 

[arceeguy]

Yes.
PWM is employed for speed control.

 

[safe]

From my understanding, and induction motor IS a synchronous motor as its RPM is synchronized with the "rotating" field.

No, not really. Sorta kinda maybe...

The synchronous motor has a rotor that is energized by some external power source. The induction motor "short circuits" the rotor, so the electricity comes from itself and is not externally applied.

You might think of the analogy of a supercharger verses a turbocharger. The supercharger does better at the low end, but then sort of has nowhere to go after that. The turbocharger gets more and more powerful as the rpms and load increases.

The induction motor (if driven with a single frequency) is not very impressive. The main concept here is that the induction motor needs to be fed a variable frequency depending on the rpm and also it's a good idea to control the current too. Some of the controllers I've read about are using PWM in a sort of AC kind of way. It's sort of weird, but the results are impressive apparently.

A RC motor ESC should probably work, but like with the brushless RC motors they are used on they need things like current limits to become more usable. On a model airplane the torque is easier to deal with than on an ebike. There are two guys selling custom circuits to deal with the primitive nature of RC ESC's, but then you are adding another $100 to your project. (and you are still within the permanent magnet world)

More Poles, More Wires?

I'm a little unclear on this myself, but I'm pretty sure that you don't need individual wires for each pole. Let's say you have 6 poles, then you can use 3 wires and let things overlap using Three Phase power. 12 poles is the same. The main thing is that as the pole count goes up the rpm goes down.

For ebikes many poles are better than fewer.

For Three Phase... 3, 6, 9, 12, 15, 18, 21...

However I would be skeptical after a certain point that maybe too many poles causes unwanted interactions, so 12 or so is probably enough.

 

[arceeguy]

OK, I think I'm getting it a little better now.

So a PMDC brushless motor is a synchronous motor, that's why they "freak out" when the rotor loses sync with the field.

The induction motor can run at varied RPM's because of "magnetic slip". Like how a ceiling fan (induction motor) varies speed by reducing voltage and not changing field frequency.

Your concept is to vary the frequency of the field, but since it is an induction motor, you won't have to worry about rotor position like a PMDC brushless would.

My head hurts!

 

[jimraysr]

POLES and IRON

An other consideration is, as the poles go up so does the iron. 2 pole generators are maybe 1/2 the size of a 4 plole of the same rated output.

Jim

No, not really. Sorta kinda maybe...

The synchronous motor has a rotor that is energized by some external power source. The induction motor "short circuits" the rotor, so the electricity comes from itself and is not externally applied.

You might think of the analogy of a supercharger verses a turbocharger. The supercharger does better at the low end, but then sort of has nowhere to go after that. The turbocharger gets more and more powerful as the rpms and load increases.

The induction motor (if driven with a single frequency) is not very impressive. The main concept here is that the induction motor needs to be fed a variable frequency depending on the rpm and also it's a good idea to control the current too. Some of the controllers I've read about are using PWM in a sort of AC kind of way. It's sort of weird, but the results are impressive apparently.

A RC motor ESC should probably work, but like with the brushless RC motors they are used on they need things like current limits to become more usable. On a model airplane the torque is easier to deal with than on an ebike. There are two guys selling custom circuits to deal with the primitive nature of RC ESC's, but then you are adding another $100 to your project. (and you are still within the permanent magnet world)

More Poles, More Wires?

I'm a little unclear on this myself, but I'm pretty sure that you don't need individual wires for each pole. Let's say you have 6 poles, then you can use 3 wires and let things overlap using Three Phase power. 12 poles is the same. The main thing is that as the pole count goes up the rpm goes down.

For ebikes many poles are better than fewer.

For Three Phase... 3, 6, 9, 12, 15, 18, 21...

However I would be skeptical after a certain point that maybe too many poles causes unwanted interactions, so 12 or so is probably enough.

 

[safe]

An other consideration is, as the poles go up so does the iron. 2 pole generators are maybe 1/2 the size of a 4 pole of the same rated output.

Rotor iron or stator iron?

As I see it the iron is sort of "extra credit" anyway because it's there in the stator to amplify the field. In the rotor the iron should be adequate for the task so that core saturation doesn't take place, but other than that all you would need is to increase the number of little slots in the core to equal (or nearly equal) the pole count.

I'm not sure if there is a direct relationship of iron to poles or whether it's a design convention. They do a lot of stuff with big industrial motors because they don't have to think about weight considerations. For something for an ebike you might have to tweek the designs so that maybe you sacrifice in one area in order to keep the weight down.

More iron is almost always better from an efficiency standpoint... but iron weighs a lot... so there might need to be a compromise.

I have this feeling that "we" the "ebike world" are more or less wandering in an unknown wilderness because weight trumps most everything on our ebikes. I have a feeling that the "ideal" ebike induction motor is either extremely rare or just doesn't exist in production.

 

[Tinker1980]

I know a bit about large three phase induction motors... I work in a factory, I used to work in a part that did powder coating. The booth draft blowers and conveyor line drives used induction motors, (actually, nearly anything that moved used induction motors.) I mention the booth draft blowers and conveyor drives, because they had the ability to vary the speed. This was done very simply, with a Variable Frequency Drive, but we are lazy and called it a VFD. The motor RPM would change depending on the frequency, no load speed was typically 5% or so less than synchronous speed. There is a formula:


f*120p
S= -----------
P

Where S is synchronous speed of the rotating field, in RPM, f is the line frequency, P is the number of poles, and p is the number of phases. Running a motor slower than the 60 Hz power line frequency doesn't seem to hurt it or make it get hot - the booth draft blower on my booth ran around 47 hz all the time, which came out to 2800 RPM or so. I could tell the VFD to run to 60 Hz so I could clean the inside of the booth without powder going everywhere.

Problem is... those were 480v motors, and the controllers were made to run them, far too large and expensive for a bike! Perhaps the thing to do, instead of making a permanent magnet rotor for an alternator, is make a squirrel cage rotor, with steel plates and small copper rods? A stack of steel plates, stars with 12 blunt points, copper rods in the low parts, short-circuiting rings in the ends? That would be a true induction motor. Not sure what it would take to construct.

-Mark

 

[safe]

Everything you say is true.

I think the problem is to first of all "know" how many turns you need in the coils if you are to construct a motor and then (of course) being able to actually fabricate a motor from scratch.

Most industrial induction motors have some problems for ebikes:

They are hopelessly overbuilt for efficiency and reliability and weigh too much.

They run on voltages that are way too high for ebikes.

----------------------

What is needed is to find a specialty induction motor that runs at the voltage we need and is designed for low weight while giving away as a sacrifice some of the reliability features that the industrial motors have.

VFD makes sense... also... for my application I would want "Torque Control" too. The idea of this EBRR (Electric Bicycle Road Racing) racing series is to allow exactly 1000 watts of power to leave the battery (sort of like restrictor plates in NASCAR) and no more. The motor designer is then in a situation where they try to get the most overall efficiency from the power allotted. Induction motors have the flatter and wider powerband so as long as you are running the motor with full load then the efficiency should be good.

It's frustrating... I can imagine what I want, but for the time being I can't do anything about it. My actual development has been cut short this summer because I've got some major home repair issues to deal with. (my dishwasher leaked and ruined a bunch of stuff)

One of these days...