Power Constraints on Motors: DC/AC

[dozentrio]

What are the limiting factors for how much power your motor can consume? I have some ideas, but I'd like to know for certain. What is stopping us from pumping 20 kW through a 500 W motor?

Based on my experience with a Currie motor, heat seems to be a big limitation. You can increase the voltage, and the motor will spin faster, but also draw more current. The increased current will cause laminations to melt and magnets to demagnetize and in general bad things to happen. Also, I've heard of guys saying they had windings lifting off the rotor, and your maximum torque output is probably also limited by the strength of the motor shaft and the motor casing (by which you presumably mount the motor to something. So:

1. Heat dissipation
2. Magnets, if subjected to too high of a flux density
3. Physical construction

The heat dissipation is not a big deal, as you can do things to cool the motor, or simply mount a thermistor inside and keep an eye on the temperature. When temp gets too high, pull back. The magnets are rather a problem, because I can't think of any way of increasing this limit. The only thing I can think of is to use windings, instead of permanent magnets. Not sure about the ramifications of this. Probably increases size of the motor, or power consumption/efficiency. The last is simply a matter of using thick/strong enough materials.

So that brings us to power delivery systems. You want batteries with high C ratings or, simply a lot of batteries. And of course, you want as high a voltage as possible. Within limits, I suppose. I believe once you get above 100 Volts or so, things can get tricky with arcing through the air and such. Actually that probably doesn't become a problem for much higher than 100 volts... Perhaps it requires increased complexity in the circuitry, and expensive high voltage parts. Anyway, onwards. Thick wiring capable of delivering high current. And lastly, a motor controller that can deliver high voltage/current. So:

4. Batteries (High voltage/Capacity/C rating)
5. A hefty controller, capable of high voltage/current delivery
6. Thick wires.

What inspired this was the Millenium Falcon. That ship seems like it's jerry-rigged and modified to run far beyond it's original design specifications. And bursts of high speed/performance are possible only by temporarily pushing the systems past their limits, and being aware of these limits. And what to replace when things blow up Now, this is obviously conjecture. I don't know anything about the Millenium Falcon, and if anyone thinks they do then they're one of those crazed fanatics, who has trouble distinguishing between fantasy and reality. But, the thought has inspired me, and now I am curious what it would take to build an e-bike, or e-vehicle, that had the capability for short bursts of power/performance way outside original tolerances. I think with electric vehicles this is far more possible than with ICE, for example. Sure, you can say spray NOS into the engine and run it at a screaming 10,000 rpm... but in general I think they are less tolerant to being run above specifications.

 

[dogman dan]

You have it right that mainly it's a matter of heat for the motor. Temp monitoring is a great idea even for use at the rated wattage for long periods. Most of the design work on motors was done when nobody had more than 10 mile range. Now we have 30 mile range with lithium and if your town is hilly heat can build up on long rides. Typically limiting the motor to under 180F measured inside the hub will prevent any motor damage. Some go higher, since the magnets can sometimes take it, as well as the windings. But some motors will have the magnet epoxy melt at 200F or so. I have heard of motors that were run at 250F without damage though. Methods for one, may always run at temps like that, but he does melt stuff at times.

I have personally melted a heinzmann motor to the point where the magnets no longer stuck to iron.

 

[amberwolf]

and your maximum torque output is probably also limited by the strength of the motor shaft and the motor casing (by which you presumably mount the motor to something.

And also by the shaft's connection to the rotor or stator, whichever it is part of. I've seen at least one thread where someone spun the stator in a BLDC around it's shaft.

The heat dissipation is not a big deal, as you can do things to cool the motor, or simply mount a thermistor inside and keep an eye on the temperature. When temp gets too high, pull back.

That depends on your sensor, and how good it's thermal connection to the part you're measuring is. If it's attached to say, the laminations in the stator, then the windings could have already melted before it reads a temperature high enough to worry about. Some things conduct heat poorly, like steel and iron, and some very well, like copper. Some are so-so, like aluminum. Ideally you'd want thermal sensors in a few places, since there could be inductive heating as well, at high currents, so that parts of the motor that might not normally get hot end up melting.

The magnets are rather a problem, because I can't think of any way of increasing this limit. The only thing I can think of is to use windings, instead of permanent magnets. Not sure about the ramifications of this. Probably increases size of the motor, or power consumption/efficiency. The last is simply a matter of using thick/strong enough materials.

The problem there is that you also increase the heat load on the motor. Instead of a passive magnetic source, you now have an active one that has current passing thru it, generating heat and also having magnetic fields passing thru it inducing currents in the field coils.

For motors that you really want to have fine torque control over, using field iwndings instead of magnets is a good idea; if you want really massive torque you probably *have* to use field windings, unless you go for huge diameter motors to get the torque that way. Even so, the greater the diameter the more torque it *could* have than the same motor in smaller diameter. Not sure of the physics words, but I think "longer moment-arm" is what I'm after.

 

[liveforphysics]

Sure, you can say spray NOS into the engine and run it at a screaming 10,000 rpm... but in general I think they are less tolerant to being run above specifications.

That's generally true, unless the engine has a big "H" on the valve cover, and a "K" or "B" on the engine block. Then they crave all the jug and RPM you can hit them with.


One of the greatest guys involved with the E-revolution, Justin from Canada, did some extensive testing to find the saturation point for some of the more common hub motors. At this point, the iron laminations saturate, and additional current results mostly in just additional heating, with only a slight additional torque/amp gained. They call this the "knee" in the slope of the torque/amp graph. Running motors past this "knee" remains linear for a bit at diminished returns, then other parts saturate, and only additional heat is gained with additional amps.

The cool thing is, this "knee" point often occurs way beyond the current levels most feed our motors, so there is a lot of overhead remaining for a boost sort of effect when needed for bursts.