prototype testing of hum canceling P-90 design

[Joe Gwinn]

What does 'coaxial' mean wrt winding a pickup?

A coaxial humbucker looks like an obese singlecoil, but its "coil" is in fact two independent coils, one wound upon the other, so there is an inner coil plus an outer coil with a thick layer of dielectric between. The coils are connected series-opposing.

This works, to the degree it does work, because the music fields are quite nonuniform, being concentrated near the magnets, so the inner coil develops more music voltage than the outer coil, so the music doesn't quite cancel.

Hum fields are pretty uniform, so both coils see the same density of hum flux, so if the area-turns products of the inner and outer coils are the same, the hum will cancel. At least in theory - alnico magnets have a permeability of about three, so the inner coil will get more hum than expected from the area-turns product, so some tweaking of area-turns products will be needed to achieve complete hum cancellation.

[peskywinnets]

A coaxial humbucker looks like an obese singlecoil, but its "coil" is in fact two independent coils, one wound upon the other, so there is an inner coil plus an outer coil with a thick layer of dielectric between. The coils are connected series-opposing.

This works, to the degree it does work, because the music fields are quite nonuniform, being concentrated near the magnets, so the inner coil develops more music voltage than the outer coil, so the music doesn't quite cancel.

Hum fields are pretty uniform, so both coils see the same density of hum flux, so if the area-turns products of the inner and outer coils are the same, the hum will cancel. At least in theory - alnico magnets have a permeability of about three, so the inner coil will get more hum than expected from the area-turns product, so some tweaking of area-turns products will be needed to achieve complete hum cancellation.

Thanks for the clarification.

In that case, I've already made one about 6 months ago ....ie I wound a bobbin halfway, stopped, made a loop in the wire ....continued the wind until the bobbin was full, snipped the loop, ended up with four wires ...wired them as you've outlined - it sucked.

[Joe Gwinn]

Thanks for the clarification.

In that case, I've already made one about 6 months ago ....ie I wound a bobbin halfway, stopped, made a loop in the wire ....continued the wind until the bobbin was full, snipped the loop, ended up with four wires ...wired them as you've outlined - it sucked.

What was the exact problem?

I think you need a dielectric layer between the coils. And a bit of experimentation.

[Mike Sulzer]

This works, to the degree it does work, because the music fields are quite nonuniform, being concentrated near the magnets, so the inner coil develops more music voltage than the outer coil, so the music doesn't quite cancel.

It is indeed local nature of the field from the vibrating string that would make this work (perhaps not very well, I suspect), but it is not exactly the non-uniformity of the field. Remember, it is the flux enclosed by a coil that counts, and both the inner and outer coils include the field near the magnets. The difference is that the outer coil also includes some field lines that, having passed down through both coils, now return to the string and pass outside the inner coil, but inside the outer one. Such a field line has no net effect in the outer coil since the trip up cancels the trip down (opposite directions).

[Dave Kerr]

So, a dielectric layer and a layer of UltraPerm (Joe G started a thread on that stuff a while back) might isolate that outer coil a little better?

[Mike Sulzer]

So, a dielectric layer and a layer of UltraPerm (Joe G started a thread on that stuff a while back) might isolate that outer coil a little better?

The dielectric provides no more isolation than empty space. It is just there for mechanical reasons. A magnetic material could do a better job of guiding the flux back to the string, but this is not an easy thing to do in practice.

[Joe Gwinn]

It is indeed local nature of the field from the vibrating string that would make this work (perhaps not very well, I suspect), but it is not exactly the non-uniformity of the field. Remember, it is the flux enclosed by a coil that counts, and both the inner and outer coils include the field near the magnets. The difference is that the outer coil also includes some field lines that, having passed down through both coils, now return to the string and pass outside the inner coil, but inside the outer one. Such a field line has no net effect in the outer coil since the trip up cancels the trip down (opposite directions).

While true, why go to that level of detail here? It's a matter of pedagogy - start simple, albeit approximate, before exposing people to the full and confusing complexity.

And more generally, many beloved details have little practical impact in guitars, so it's best to neglect them until someone manages to detect them in a pickup. Guitar pickups are not precision scientific instruments.

[Joe Gwinn]

The dielectric provides no more isolation than empty space. It is just there for mechanical reasons. A magnetic material could do a better job of guiding the flux back to the string, but this is not an easy thing to do in practice.

Agree.

[Joe Gwinn]

So, a dielectric layer and a layer of UltraPerm (Joe G started a thread on that stuff a while back) might isolate that outer coil a little better?

It could work, but don't use too strong a magnet, or you will saturate the ultraperm.

[David King]

The Kinmans are just stacked with a magnetic iron shield to redirect the field away from the lower coil.

[Mike Sulzer]

While true, why go to that level of detail here? It's a matter of pedagogy - start simple, albeit approximate, before exposing people to the full and confusing complexity.

An admirable goal, but difficult to achieve in practice. What you wrote could be interpreted by the unwary as implying that the inner coil is more sensitive to the "music" field because its windings are closer to it. It makes perfect sense because the magnetic field does get weaker when you get farther away because the field lines spread out, lowering the flux density. But in this case the outer coil sees all the "music" flux the inner one does, and more.

And more generally, many beloved details have little practical impact in guitars, so it's best to neglect them until someone manages to detect them in a pickup. Guitar pickups are not precision scientific instruments.

Pickups are not scientific instruments, true. But designers and builders here need to be very precise in the selection of materials and techniques. Did you notice the differences in the sound resulting from different bobbin materials in Frank's post? Quite a bit of research goes on involving multiple variables, mostly without the help of a good theoretical understanding of how magnetism works. The research deserves praise, but much of it would be easier with better understanding. In particular, it is really helpful to understand the law of magnetic induction.

[David Schwab]

The Fralin hum canceling P-90, Lace hum canceling P-90, Gibson EBO neck pickup, Fernandes Sustainer driver/pickup, and Gibson Ripper bass pickups are all sidewinder designs. There are more, but those are ones I can think of off the top of my head. I think some Lawrence designs might be sidewinders too. Some have the magnets inside the coil bobbins- parallel to the strings, some have them on the outside of the bobbins coupled to a ferrous bobbin core, some have the magnet somewhere in the center core.

The Lawrence L-250s are sidewinders. His newer pickups aren't. Kent Armstrong makes some bass pickups like that too. Q-Tuners are sidewinders, and I make some as well.

I got a Jazz bass set that sounds really nice. It took several prototypes, but now I'm really happy with them.

[David Schwab]

Well, ... that narrows it down! You can stack 'em or lay 'em down side by side. The devil is in the details.

It would have to be a stack or a split coil if it's going to sound like a single coil.

Stacks are inherently bright, so you have to wind them hot to compensate. I made a 14K Tele bridge pickup and it was super bright. I was expecting a fatter tone. You look at some of the DiMarzio stacks, and they are 24K. Stacks phase cancel a lot of low end.

[freefrog]

The Kinmans are just stacked with a magnetic iron shield to redirect the field away from the lower coil.

There's more than that: the "lower coil" has a special design... it's made of 150 parts of laminated steel, allowing a full hum cancelling operation while the noise sensor is only a 600ohms device.
If my memory serves me, Kinman explains in the related patent that regular stacked coils with shields (like his first gen Strat designs) woould have created way too much inner capacitance to be effective - while the stray cap of a P90Hx is low enough to put the resonant frequency at the same place exactly than a regular P90. :-)

[David Schwab]

There's more than that: the "lower coil" has a special design... it's made of 150 parts of laminated steel, allowing a full hum cancelling operation while the noise sensor is only a 600ohms device.

The reason for that is the lower steel bobbin has higher inductance. He uses laminated steel to reduce eddy currents.

DiMarzio did a copy and makes them with the steel shield under the top coil, which has the magnets, and then the bottom coil has extra steel slugs to increase the inductance and is would with less turns, and I think heavier gage wire.

Here's a DiMarzio Virtual Vintage Solo Pro. The top coil reads 8.53k, while the bottom is 2.45k.