Stacked pickups - I'm missing something

[ScottA]

It's actually the area-turns product, multiplied by the effective permeability, that governs the voltage generated by a given hum field. It's a bit of math to see if this is proportional to the self-inductance of the coil.

Good point. Hence the effectiveness of the hum cancelling with large coils like the Suhr and Illitch system.

My comment is appropriate to the case being discussed here where the "active" coil and the "dummy" coil are similar in dimension.

[Mike Sulzer]

Two coils in series have too high an inductance if you are trying to get a single coil sound and a reasonably high output. The large inductance results in too low a resonant frequency.

I think stacked humbuckers are the best option for a medium impedance (or lower; remember the Less Paul low) design. You can resonate as you wish with a cap, and use a low noise preamp in the guitar or out. Stacked humbuckers give the best hum cancellation.

I wonder what the nature of the loss is? The top coil produces a signal. The bottom coil is just like the top one, but is shifted by about 1/4 of an inch, and is hooked up in the reverse direction. If that 1/4 of an inch shift simply attenuates the signal by some factor that is independent of frequency, then the loss wouldn't really affect the frequency response of the pickup. But it seems at least plausible that some important filtering happens here.

[Joe Gwinn]

Good point. Hence the effectiveness of the hum cancelling with large coils like the Suhr and Illitch system.

That's a good bounding-case example. The large coil has the same area-turns product as the small pickup coil, but the inductance of the large hum-cancelling coil is far less than the inductance of the far smaller pickup coil, so the frequency response of the assembly is mainly that of the pickup coil alone.

My comment is appropriate to the case being discussed here where the "active" coil and the "dummy" coil are similar in dimension.

Hmm. In a given hum field, as I add turns to either coil, its induced voltage changes linearly with the total number of turns (the area of a single turn not changing), while the inductance varies with the square of the number of turns. In this case, matching inductance will also match induced voltage, and inductance is easily measured. But voltage will vary with the square root of inductance.


Background: I too have many times heard the rule of thumb about voltage output being proportional to inductance, but have never been able to find out where it came from. I would hazard that there is at least one kind of device where this is true, or true enough, but counterexamples exist. Perhaps it came from the radio art. The actual rule may have been that the signal-to-noise ratio of loop antennas is improved if the inductance is greater. This would be tangled up in the optimum Q for a resonant radio antenna, and how best to match such an antenna to a vacuum tube amplifier. Somewhere I have a copy of the classical article on the design of such antennas.

[ScottA]

How about this? "Close enough for Rock n' Roll".

Practically, it works in pickups.

That's a good bounding-case example. The large coil has the same area-turns product as the small pickup coil, but the inductance of the large hum-cancelling coil is far less than the inductance of the far smaller pickup coil, so the frequency response of the assembly is mainly that of the pickup coil alone.
Hmm. In a given hum field, as I add turns to either coil, its induced voltage changes linearly with the total number of turns (the area of a single turn not changing), while the inductance varies with the square of the number of turns. In this case, matching inductance will also match induced voltage, and inductance is easily measured. But voltage will vary with the square root of inductance.

Background: I too have many times heard the rule of thumb about voltage output being proportional to inductance, but have never been able to find out where it came from. I would hazard that there is at least one kind of device where this is true, or true enough, but counterexamples exist. Perhaps it came from the radio art. The actual rule may have been that the signal-to-noise ratio of loop antennas is improved if the inductance is greater. This would be tangled up in the optimum Q for a resonant radio antenna, and how best to match such an antenna to a vacuum tube amplifier. Somewhere I have a copy of the classical article on the design of such antennas.

[ScottA]

I will add, that I suspect there might be material effects that impart more subtle responses on this as well.

Consider two different pole piece materials that might impart the same inductance with very different mass/volume in the core, I think there may be differences in the efficiency of hum generation, probably due to eddy current effects and the different "lossiness" of the materials.

I did a fair amount of work on this a few years back, and I saw some glimpses of this stuff. I think I could study it in a much more directed manner now given what I've learned since.

For my current purposes, the 3x3 matching I do in my design works well enough (as well as anything else).

I may have cause to revisit this when I go to prototype my next platform.

[ScottA]

Also, at the moment I'm deconstructing a commercial stacked pickup, to characterize and quantify the sources of loss from the coils coupling together and also the coil coupling to the upper coil shield in this design. I'll be pulling the data together and I'll post it here when I'm done.

[David Schwab]

This is what I didn't appreciate. Thanks. Although, it surprises me. That extra 1/4 of an inch distance makes that much difference? I guess it's all a matter of degree. If one of these coils is wound hot (Seymour Duncan winds one of these to 10.7k, and the bobbin is around the same size and shape of a humbucker bobbin), and the second coil cancels "a bit" of the signal, output could still be reasonable if "a bit" was quite a large fraction.

If you listen to only the bottom coil, you will see that it doesn't sound anything like the top coil. It's a muffled tone without much top end, and less output.

The reason why these pickups are wound fairly hot is some of the low frequencies is canceled out. So they tend to sound very bright with less output. So they are often wound fairly hot to compensate.

The Hot stack is 21.4k. 10.7k is one coil.

[David Schwab]

Basically, it's a very lossy design. It's physically impossible to magnetically isolate those coils from each other. That's the big reason why stacked hum cancelling pickups don't really sound as good as conventional pickups.

An interesting design is both the Kinman stacked pickup, with the laminated steel bottom bobbin;




And the DiMarzio Virtual Vintage pickups, which are more-or-less copies of the Kinman design.

In the DiMarzio, there is a steel plate under, and around the top coil (which has the magnets), and the bottom smaller coil is wound with heavier gauge wire and has steel slugs in place of the magnets, and between the magnets, to increase the inductance.

On the Virtual Vintage Solo Pro I examined, the top coil read 8.53k, and the bottom was 2.45k.

These work very well and sound very much like single coil pickups.

[Mike Sulzer]

If you listen to only the bottom coil, you will see that it doesn't sound anything like the top coil. It's a muffled tone without much top end, and less output.

Even if the top and bottom oil are wound the same, they "sound different"? Why? The sound level from the bottom coil is less, and so you hear less bass and highs; you have heard of this effect before; it is related to human hearing, not the physics of the pickup. The actual frequency response of the coils can be the same, but you still would hear less bass and treble.


They are wound hot to get the level up, not to fiddle with the frequency response.

[Mike Sulzer]

In the DiMarzio, there is a steel plate under, and around the top coil (which has the magnets), and the bottom smaller coil is wound with heavier gauge wire and has steel slugs in place of the magnets, and between the magnets, to increase the inductance.

You mean to increase the output. I think the idea here is the opposite of what you said. By using fewer turns, the iductance goes down by the square. Using a higher premeability material (and more of it?) increases both the inductance and output by similar amounts. The net effect is hum cancellation, with somewhat less total inductance than using two identical coils. Thus a higher resonant frequency and more highs.

[David Schwab]

Even if the top and bottom oil are wound the same, they "sound different"? Why? The sound level from the bottom coil is less, and so you hear less bass and highs; you have heard of this effect before; it is related to human hearing, not the physics of the pickup. The actual frequency response of the coils can be the same, but you still would hear less bass and treble.
They are wound hot to get the level up, not to fiddle with the frequency response.

No, the bottom coil has lots of bass and very little highs. The output was nearly the same. The bottom coil would be perfect for jazz guitar, as it had that dark mellow tone. Keep in mind that the poles expend into that coil, and then extend up to the strings, so it's not at all like taking a regular single coil and moving it farther from the strings.

Have you built any of these Mike? The first one I made was a stacked Tele bridge pickup that was wound to around 12k with 43 AWG wire. It was super bright and kind of thin sounding. The output was about the same as a standard single coil. Perfect for a vintage Tele, but not what you would expect from a pickup wound to 12k, and I was looking for something with more mids. So I wound it much hotter and then I got more low end and less high end, just as with any pickup. In the end I went with a dual blade design though.

I've made quite a few stacked pickups trying out different designs for both guitar and bass.

Remember the two coils are out-of-phase, so you are losing some low frequency content. Duncan has a wiring trick when you switch the two coils in phase for more output and low end.

Wiring Diagram

[David Schwab]

You mean to increase the output. I think the idea here is the opposite of what you said. By using fewer turns, the iductance goes down by the square. Using a higher premeability material (and more of it?) increases both the inductance and output by similar amounts. The net effect is hum cancellation, with somewhat less total inductance than using two identical coils. Thus a higher resonant frequency and more highs.

I'm going by what the patent says. Patent #5908998:

The present invention overcomes the prior art limitations by providing additional ferromagnetic material in the gaps between the pole pieces of a pickup bobbin and/or beneath, around and/or in close proximity to the bobbin. Thisarrangement has the advantage of increasing the inductance of the pickup without causing a corresponding increase in the pickup's DC resistance and impedance. Thus, a sound with better tonality and less harshness results.

And with the different gauge wire in the bottom coil:

It is also preferable in these embodiments for the electromagnetic pickup to have a first and a second humcancelling bobbin. Preferably, the first hum-cancelling bobbin has a gauge of wire different from the second hum-cancelling bobbin such that the second hum-cancelling bobbin has a resonant frequency lower than that of the first hum-cancelling bobbin. In this arrangement, the ferromagnetic material is, preferably, positioned within or beneath the body of the second humcancelling bobbin, thereby further emphasizing and accentuating the differing frequencies

Note that this idea was "borrowed" from the Kinman patent, where he uses less wire in the bottom coil, but has it wound on a laminated steel bobbin. And then "borrowed" by Duncan/Beller for patent #7166793. The idea is by using less wire of a heavier gauge in a bobbin with increased inductance, and then shielding the top coil, you can get the same level of hum cancelation without adversely affecting the tone of the top coil as much. My experience with these pickups is that it works very well.

Using a higher premeability material (and more of it?) increases both the inductance and output by similar amounts.

That's what I just said, they added extra steel slugs to increase the inductance. Obviously increasing the inductance would increase output.

[Mike Sulzer]

I do not care what the patent says. The purpose of a patent is to protect a market; do not expect the inventor to understand what he invented or care much about communicating the idea beyond what it takes to get it approved.

I am not sure you understand the point I am making. Consider the Suhr idea. You use a coil with a very large area, and so it only takes a relatively few turns to get the same hum signal level, but the coil has a lower inductance because of the usual squared relationship: inductance depends on the flux of each turn linking each other turn. So you cancel hum without adding a lot of inductance in series; some, but not as much as with a standard stack, and therefore, you do not lower the resonant frequency as much. The Kinman idea uses different materials to do something similar: same hum sensitivity in the two coils, but the bottom one has less inductance, and therefore affacts the resonant frequency less.

The magnetic shielding helps keep the hum down in the top coil so that you do not need to pickup as much in the bottom coil, allowing you to keep the inductance down more than otherwise possible. It also helps prevent signal from getting down into the bottom coil, I think.

They did not add the extra steel in the bottom bobbin to increase the inductance; they added it to get snough hum signal without while keeping the inductance down with fewer turns.

I'm going by what the patent says. Patent #5908998:



And with the different gauge wire in the bottom coil:



Note that this idea was "borrowed" from the Kinman patent, where he uses less wire in the bottom coil, but has it wound on a laminated steel bobbin. And then "borrowed" by Duncan/Beller for patent #7166793. The idea is by using less wire of a heavier gauge in a bobbin with increased inductance, and then shielding the top coil, you can get the same level of hum cancelation without adversely affecting the tone of the top coil as much. My experience with these pickups is that it works very well.



That's what I just said, they added extra steel slugs to increase the inductance. Obviously increasing the inductance would increase output.

[Mike Sulzer]

It is an obvious fact, verifiable by different means, that the cores guide and contain some of the flux but that not as much reaches the bottom as enters the top. You get less signal from the bottom coil. Beyond that it gets complicated; the fact that the two coils are connected out of phase means that the coupling between the coils lowers the total inductance, but the self inductance of the two tends to matter more because the coupling is not all that large. (If it were, you would get little signal from the strings.)

Yes, I have made them. it is my favorite for a medium impedance pickup.

No, the bottom coil has lots of bass and very little highs. The output was nearly the same. The bottom coil would be perfect for jazz guitar, as it had that dark mellow tone. Keep in mind that the poles expend into that coil, and then extend up to the strings, so it's not at all like taking a regular single coil and moving it farther from the strings.

Have you built any of these Mike? The first one I made was a stacked Tele bridge pickup that was wound to around 12k with 43 AWG wire. It was super bright and kind of thin sounding. The output was about the same as a standard single coil. Perfect for a vintage Tele, but not what you would expect from a pickup wound to 12k, and I was looking for something with more mids. So I wound it much hotter and then I got more low end and less high end, just as with any pickup. In the end I went with a dual blade design though.

I've made quite a few stacked pickups trying out different designs for both guitar and bass.

Remember the two coils are out-of-phase, so you are losing some low frequency content. Duncan has a wiring trick when you switch the two coils in phase for more output and low end.

Wiring Diagram

[Mike Sulzer]

Duncan has a wiring trick when you switch the two coils in phase for more output and low end.

Wiring Diagram

The link says nothing about more low end. You made that up. It says that get more output, but lose the hum canceling. All you are doing is put the two coils in phase so that the signals from the strings and from hum add. You are just turning it into a standard single coil pickup, probably with more turns than you really wanted.