I have tried what ripdivot did (on page 1 of this thread), which is basically what bassetrox (on page 2) outlined.
I use a set of microcoils in my strat, which are the finest pickups I ever had. They are non -RWRP, just like original Strat pickups. They are not made to particularly cancel noise, but have very high, clear output, which makes it better. But there was room for improvement, so tried the method described. After some experimentation yet to come I will put up a series of photos and publish them here.
I used core of pcb material (epoxy coated glass fibre about 2mm) with two outer covers made from Pertinax 1mm each. I did not have slim size expoxy but the Pertinax seems quite stable in the sandwich. The bobbin slot therefore is 2mm x 15mm overhead. Using 0.15mm wire (which would be equivalent to AWG 7) allows for 500 turns of wire.
I did, however, decide to use less. 350 turn results in a 150 Ohms resistance. I had much thinner wire available, but we suspected that the winding machine (for transformers) would not be able to handle such thin wire.
The inductance would be negligible compared to the real pickup.
I for stock parallel operation, the guitar is now dead silent. You can sit on top of a tube amp combo and hear - silence. This is very remarkable. I can detect no microphonics issues, but I will pot the coil later.
The circuit described in the patent does something, but I am not sure what. The series cap does some noise-shaping. The remaining buzz sounds different. You´ll know immediately, if the coil is in the right way or not, if noise does not decrease.
Using the super-strat-switching (three toggles) allows me to dial in any serial and parallel combination I imagine. So single, parallel2 and parallel3 all sounds dead silent, but most of the noise is back as soon as I switch to serial
I have no explanation to that yet.
I can imagine that a series connection would decrease the hum-bucking somewhat
, because the coil area times turns product must be in a comparable range. Double the area (two pickups in series) and double the turns means four times the resulting product. On the contrary, parallel mode makes the product smaller.
Indeed I have the impression, the more pickups are in parallel, the more silent it gets.
I suspected that in series mode the cancelling coils somehow becomes in-phase which would increase the noise signal, but two solder connections later I was wiser
Maybe series and parallel operation would ask for a different number of windings on the cancelling coil.
Tone-wise, I can indeed not hear the remotest change. The additional resistance and, much more important, inductance does not seem to be a problem. I am happy with the improvements that already are, but since the series mode with the microcoils is so useful, noise cancellation on those would be great.
PS: if anybody could point out a method for tension-less feeding of delicate wire to a winding machine don´t hesitate to tell us.