I've been reading around Gnuts, and although I can find mention of wiring a dummy coils in parallel as opposed to series, I didn't find any definitive comment about the supposed usefulness (or non-usefulness) of each configuration.
The assumption seems to be to wire a dummy coil in series. Obviously, a series dummy coil will tend to give a 'bassier' tone compared to parallel, but is it necessarily 'better'? Would an 'option' to switch between series and parallel be useful?
Switching the dummy coil between a series and parallel arrangement with the pickup would certainly be useful in comparing the tonal difference. But in general, maybe not-so-much.
If you're using a dummy coil that has identical characteristics to the SC pickup you're trying to de-hum, it seems like you can't win all ways round. There always seems to be some unattractive side-effect.
- In series, the additional inductance will cause a significant loss of high frequencies. There will also be an overall loss of signal but this is likely to be too small to notice. This seems to be the preferred configuration.
- In parallel, I should expect the tonal change to be relatively benign. Probably slightly brighter than the SC alone. But the attenuation of all frequencies would be significant. Approximately 1/2 the voltage the pickup would produce without the dummy in parallel. 1/2 voltage means -6dB. To put that in perspective, -3dB is just noticeable. -10dB sounds half as loud. The volume loss makes parallel high unattractive.
You could use a dummy coil that is dissimilar to the pickup you're trying to de-hum. If used in series with the SC pickup, you would want a coil that produces the same (hum) voltage but has much lower inductance and resistance. If used in parallel the dummy coil should have much higher inductance and resistance than the pickup. However, designing such coils might be a daunting task.
There might be a relatively easy fix when using a dummy with the same characteristics as the pickup. Not perfect, but easy. Put the dummy in series with the pickup and shunt the dummy with a small capacitor. I reckon something on the order of 20~50 nF. I expect this would do a good job of retaining the anti-hum function but mitigate the high frequency loss. The "hum" should be dramatically reduced but if there is a "buzzy" component (from a motor or fluorescent light) it won't be as effective as it would without the capacitor. JMO. To my knowledge, this path has not yet been explored. It just makes sense to me, at least on paper.
Yeah, I was wondering about switching in practical use. My feeling was similar to yours: probably not that useful. But, I did not find any expressed opinions on this - just a general assumption to have the dummy coil in series.
I read about using a dissimilar coil (larger area, fewer turns, much lower resistance and impedance; but with similar inductance). An interesting idea, but not very practical for most people. Though, there is a(n expensive) commercial product out there based upon those principles.
I too wondered about the use of a capacitor to change the characteristics/effects of a dummy coil, but did not (could not) think it through as far as you. But, if motor and fluorescent hum will be an issue, it does not seem that useful (to me) in practice since those frequencies are so common (and have become even more common with newer light bulbs).
So, rather than switching a dummy coil from series to parallel, a more useful approach might be the ability to switch between single coil, and single coil with dummy coil in series (perhaps with a capacitor shunting the dummy). This too I had considered, but without the capacitor.
Thanks. And thanks for the links - I had already read those pages, but they are useful for other readers.