How real electric motors work

John Storey

5. Printed circuit motors

Sometimes called "pancake motors", these are a particular cunning motor configuration whose operation is in some ways is easier to visualise than that of a conventional motor. They fit into confined spaces (say inside a car door, to make the windows go up and down) and, because the rotor is light and has little rotational inertia, can accelerate to full speed and stop again very rapidly. This feature isn't so important for car windows (unless you’re into drive-by shootings), but is essential for industrial robots and other servo mechanisms.

Advantages:
• Efficient – no hysteresis or "iron" loss
• Very low rotational inertia
• Light weight
• Flat, so fits into confined space.

Disadvantages:
• Expensive to make
• Armature has little mass, and therefore can overheat quickly

This is what a typical printed circuit motor looks like. Don't worry about the two black wires for the moment.

This is what's inside. Glued to each "face” of the motor are (in this case) eight magnets. Their poles alternate N-S-N-S etc as you go around.

Here's the other face, with its eight magnets. These line up exactly (north to south) with the magnets on the other face, creating a strong magnetic field across the gap where the rotor sits. (Try not to think about that black wire for the moment.) So, now we've set up a strong magnetic field running axially (ie, parallel to the motor shaft). That field threads back and forth eight times through the small gap that will be occupied by the rotor when we put it all back together. (Note that the end faces are made of iron and complete the magnetic circuit.)

Now, if the magnetic field is parallel to the shaft, and we want a tangential force on the rotor, which way does the current have to be flowing? Well, it has to be at right angles to both, and therefore radial.

Now it should all make sense. The brushes contact the rotor on that blackened area near the shaft. The current goes out along the copper wire, and is travelling almost radially as it goes through the region of highest magnetic field. (Almost radial but not quite, to reduce cogging torque.) So, a force is exerted on the wire that is at right angles to the wire and at right angles to the magnetic field, causing the rotor to turn. Now, if the wire just turned around and came back in towards the shaft again, the force on the bit coming back would be equal and opposite to what it was going out, cancelling out any useful torque and the whole thing would just sit there with smoke pouring out of it. So, once the wire has gone out past the magnet, let's take it over diagonally to the right and bring it back in to the shaft past the next magnet which, you'll recall, has its magnetic field in the opposite orientation. Now the force on the returning piece of wire will add to the torque, and away we go.

Once back on the blackened piece of rotor the current can pass out through the second brush and back to the battery or whatever it is that's powering the robot.

Try not to think about the black wire for a moment. In this picture you can see the two brushes. They simply rub on the rotor which, as you saw, consists simply of a flat piece of insulator with copper lines etched or stamped on it, like a printed-circuit board. The other side of the brushes ends up as brass terminals on the outside of the motor, as in the first photo.

So there we have it. The only problem now is that the magnets themselves cannot retain their strongest permanent field unless they are always in a completed magnetic circuit. So, you can't magnetise them and then assemble the motor. But, once you've assembled the motor, you can't get at them to magnetise them. So, we can't actually build this kind of motor. Pity, really, it was looking rather promising.

But wait a moment! Suppose we thread a black* wire back and forth between the magnets, as in the picture above. Let's bring the wire outside the motor, and once everything is assembled we zap a gazillion amps through the wire. Think about what direction the magnetic field created by the current through that wire will be in. Perfect! Admittedly it's not a very thick wire to be coping with such a large current (typically several thousand amps), but it's only for a few milliseconds and the wire doesn't have time to complain. Also, it only has to happen once...

*Actually, any colour would do.