In the last part of this article, we tried flipping the direction of current and the magnetic polarity in pickups to see what effect it had on the signal and hum being picked up by the coil. We found that flipping the direction of current inverted both the hum and the string signal, but flipping the magnet’s polarity only inverted the signal from the strings.
I’m sure you’ve jumped ahead of me here, but what about if we do both these changes at the same time? If we flip the direction of the current, then we invert the hum and the string signal. But then if we also flip the polarity of the pickup’s magnet, we will invert the signal again, meaning it goes back to how it originally was, but this time we don’t flip the hum.
Great – now we have a signal we can combine with our original pickup which will cancel the hum, without affecting the signal! You can see on the graph that the two hum lines are exactly opposite to each other. If we combine the two “total” signals from the two coils, we get all the string signal and none of the hum.
At this point it’s worth pointing out that we can never really achieve a level of cancellation this precise. The only way to do this would have the two coils be exactly physically identical to each other, and for them to be in the exact same physical location at the same time, which is clearly impossible. Therefore the noise and signal picked up by each coil will be slightly different and the there will still be a tiny bit of noise, and the string signal will not quite be exactly the same as if there were only one coil.
The only consideration left is whether to connect these two coils in series or parallel. In series, the waves are summed. In parallel, the waves are averaged. This makes no difference to hum cancellation, because (for example) the sum of 5 and -5 is zero, and the average of 5 and -5 is also zero.
In a normal humbucker, it’s normal to wire the two coils in series. This is because it further increases the signal to noise ratio – while the hum is inverted against itself, the string signal strength is effectively doubled.
On a guitar with single coil pickups, however, when two pickups are connected at the same time (for example in the 2 and 4 positions on a Strat), they are connected in parallel. This makes sense because otherwise the signal strength would be twice as much in the “in-between” positions, when two pickups are connected, as when only one pickup is active.
I hope this has demystified how coils can be combined to cancel hum without cancelling signal. This knowledge helps us when we are planning the wiring of a guitar, and we are trying to use the properties of the various coils available to minimize hum.