Motor control and efficiency

Posted on January 20, 2008


I’ve posted a lot about the motor I’m trying to build, but I don’t think it is yet clear what kind of motor it is. This may look unimportant, but trust me that it becomes very important when you’re trying to find the right kind of electronic power controller to get the thing spinning. 😉

I’ve picked a technology of motor that has:

  1. Permanent magnets, because they allow building a brush less machine that can act as motor and generator with the same efficiency;
  2. Radially oriented magnetic flux, because (as we have seen) the axial types have high undesirable axial forces;
  3. Asymmetric number of stator and rotor poles, because it allows for a motor practically without cogging torque;
  4. A rotor outside the stator, because it allows for higher torque with the same motor volume (explanation here, at the bottom);
  5. Three-phase current feeding, because it is sufficiently smooth for a traction system, and there are plenty of 3-phase controller implementations out there;

Judging by the German application of this kind of motor to city-bus wheels for the last decade, I’d say I’m in the right track.

Finding the right kind of controller for this motor was a bit of a challenge, because it has several names depending on who you talk to. The radio-modelling people call it an “LRK – Lucas, Retzbach, Kühlfuss” (although they wouldn’t know if it is an “outrunner” or “inrunner”, because it has rotors outside and inside the stator). The electrical engineer could call it a “PMSM – Permanent Magnet Synchronous Motor”, because the rotor spins synchronous to the electric AC field. The generic industry reseller could call it a “BLDC – Brush Less DC” motor, because the electronic control circuit hides the fact that this is an AC motor, and you just have to feed it with DC current.

So, if we can call it an AC or DC motor, what exactly is the kind of controller we should use? Well, I tend to see the electronics as something that does not belong to the motor itself, so a BLDC is an AC motor for me, whatever your opinion is.

Luckily, the cool folks at “Microchip” have a very good notion of these things. They note that there is practically no physical difference between a BLDC and a PMSM, except the fact that they are controlled with different electric field waves. It is common to use simple square waves (on-off switching), but it is more advantageous to use sine waves (PWM switching). If done properly, the sine wave strategy can result in a nearly perfect control, kicking the motor’s efficiency level way past the 95% mark, possibly into the 99% mark!!! 😀 Other advantages are smoother operation and lower vibration noise (because of smaller torque ripple).

Fig.1: Square 3-phase motor control
(copyright Jorge Zambada, Microchip Technology Inc.)

Fig.2: Sinusoidal 3-phase motor control
(copyright Jorge Zambada, Microchip Technology Inc.)

Don’t take my word for it, take this guy’s (scroll down a few pages to see his 99% efficient motor – patented here).

So, one of the tricks in getting a high level of efficiency out of an electric motor is to feed it with perfect sine waves, in order to avoid hysteresis losses and induced current losses caused by higher-frequency harmonics. But it also means using faster, more complex and expensive electronics, which is probably the reason why the radio-modelling folks have been doing the simpler method (using square or trapezoidal waves).

Fig.3 – Harmonic content of a square wave (FFT scope).
(copyright Harry Lythall – SM0VPO)

Fig.4 – Harmonic content of a sine wave (FFT scope).
(copyright Harry Lythall – SM0VPO)

Well, efficiency is paramount to me. I also don’t want the motorized wheels to make “singing” noises or to vibrate when accelerating or braking. So I’m obviously going for a sinusoidal controller. But I may start out with a simple “square switching” device, during the motor prototype phase.

Of course, I also have to look at the motor’s own parameters, but that’s another story.