This is just an appendix to the previous post on alnico poles (the reason for adding it will be more obvious at the end). I ran some sweeps through a humbucker coil with different pole pieces. Here is the result:
Ths chart compares the self-resonance of a humbucker coil with various pole-piece/core material: Air, Alnico 5, standard 1010 steel slug, mild steel standard humbucker adjustable screws, button-head hard black-oxide-coated steel bolt (like a duncan Invader), core ferrite, and m19 transformer laminations (the ferrite and laminations were so close to each other that I only included one sweep). Obviously the laminations/ferrite were the most efficient in terms of highest permeability and lowest loss.
NOW… how does this relate to the alnico pole post? Well look at this graph:
The top sweep is an air core, the middle sweep is a solid brass core, and the bottom is the control. Eddy current loss in the brass flattens the peak amplitude by a few dB without changing the frequency too much. I experimented with simulating an alnico 5 single-coil style pole piece in a humbucker format (ferrous/non-magnetic pole with magnetizing bar underneath at a 90 degree angle) and was able to exploit the permeability and eddy current loss to so that:
The green trace is alnico 5 and the red trace is a thin ferrite rod wrapped in a single layer of copper foil (the adhesive-backed type typically used to shield a guitar control cavity), sitting perpendicular to a humbucker bar magnet. So that might be useful information to squirrel away for a future project. I got similar-ish results with thin, zinc-coated steel machine screws (not sure what size, they are the type that i use to mount small triode sockets/shields to a chassis), but I could not get the peak amplitude sufficiently high to match the alnico. It was easier to tune the frequency with the size and saturation of the ferrite, and the amplitude with the copper.
There will be more humbucker-specific discussion in the future after I have a chance to cover the basics.
As an addendum, here is a chemical analysis from some 1018 steel I bought for pole-piece use:
Here is a sweep of that 1018 bar vs a stack of laminations the same size (1/8″ x 1/4″) with a 1Meg resistor load instead of a simulated guitar+amp input load like the above sweeps, just to show the effect of the eddy currents a little better:
You can see that the steel is actually carrying slightly more signal, but the eddy current loss causes the high frequencies drop to -3dB (relative to the laminated core) around 6kHz.