Switchable additional winds

[okabass]

No PM, please. High level info here.

[Joe Gwinn]

"Terman's method" is fine, at least as long as inductance does not change with frequency, one takes care of parasitic capacitances of the measuring setup and the system (PU) has only one resonance. Maybe a little time-consuming. The theory behind is simple. Always nice to have different options. But each method has its drawbacks. I am not going to revive an old, fruitless discussion, though. Prefer being pragmatical instead, as PU capacitance is not a primary influencer of sound.

Not really. The inductance does in fact change, and one can quantify the change using Terman's Method. It's exactly those parasitic capacitances (of the coil) that one is measuring. Parasitic capacitances of the leads are far smaller, and more or less represent wiring inside the guitar.

I published the method some time ago in a separate thread.

[Sooner Boomer]

I've been lurking for a while, and decided to post. I know quite a bit about electronics, but have only recently gotten into guitar-related stuff. It's pretty hard to separate lore from fact when discussing guitars, and especially pickups. Putting number to descriptions helps immensely.

Please keep the discussions public. Yeah, you might be going over old ground, but each time a small kernel of knowledge is unearthed.

Has anyone ever wound a bi-filar pickup?

[Antigua]

I've made a four conductor tapped single coil, and I stand corrected: there outer coil brings the resonance of the inner coil way down, regardless of whether it's connected at one end or not. I'm not sure exactly what is happening in terms of circuit analysis, but obviously some combination of inductive and capacitive coupling. I'll post more details a little later today.

[Helmholtz]

I've made a four conductor tapped single coil, and I stand corrected: there outer coil brings the resonance of the inner coil way down, regardless of whether it's connected at one end or not. I'm not sure exactly what is happening in terms of circuit analysis, but obviously some combination of inductive and capacitive coupling. I'll post more details a little later today.

The phenomenon is called reflected capacitance by magnetically coupled windings. Now you may also measure interwinding capacitance between the windings.

[Antigua]

I finally got around to creating a 4 lead tapped single coil of my own:



Here are measurements of the four lead tapped single coil:

Inner coil without outer coil in place, 4000 turns
DCR: 2.935k
L: 578mH
Peak f: 16.3kHz

Inner coil with 4000 turn outer coil in place, not connected
DCR: 2.900k
L: 575mH
Peak f: 10.4kHz

Outer coil, 4000 turns, inner coil not connected
DCR: 3.117k
L: 711.8mH
Peak f: 10.0kHz

Both coils in series, 8000 turns total with connections at 4000 turns:
DCR: 6.028k
L: 2.282H
Peak f: 9.48kHz

So, the presence of the outer coil does bring the resonant peak of the inner down by quite a lot, from 16.3kHz down to 10.4kHz, but interestingly the 100Hz inductance measurement with the DE-5000 didn't change.

Here are several plots:



The most important takeaway is that the outer coil limits the resonant peak of inner coil, regardless of whether it's connected at one end or not, so the current industry practice of using three leads doesn't represent a drawback. There is very slight difference, but that would surely disappear in a load context. When the inner coil was by itself on the bobbin, the resonant peak was 16.4kHz, but with the presence of the outer coil, it dropped to 10.4kHz, and having a single connection at the ground side only reduced the resonant peak by another ~200Hz. And as was seen with production tapped single coils, the resonant frequency of the tapped coil(s) is nearly the same as that of both coils connected in series.

The two overlapping lines that dip way down show that when the outer coil is shorted, it becomes a huge load upon the inner coil, similar to eddy currents caused by covers and base plates, but much more extreme. The fact that they overlap means that the result is the same even if the shorted outer coil is connected to ground at the short point.

With this four conductor tapped coil there are other things that can be tested for, but overall it looks like
Helmholtz is correct, that the practical value of a tapped single coil is very limited by an unavoidable interaction between the two coils.

[bbsailor]

Antigua,

You made a nice 4 wire strat test pickup. I have one more test for you to try on this pickup that will put other issues to bed. Using a dual trace scope connect one scope channel to the inner winding and the other channel to the outer winding.

Hand hold the pickup above guitar strings and note the peak output voltage from each coil. Swap scope channels and coils to see if you get the same result. This will eliminate any bias due to the scope channels not being amplitude balanced.

If you follow this to its logical conclusion you could make a pickup coil with an upper winding and a lower winding and see the interaction of the coils in this configuration.

What at you want to do if form a good mental model about what happens to the pickup when you make a flat and wide pickup versus a tall and narrow pickup.

Another good new thing to try is to wind the pickup with AWG 32 to AWG 34 to about 500 to 600 turns and feed the pickup into 2 conductor shielded cable with an XLR connector on the amp end of the cable. Feed the XLR into a mic mixer input and listen to a different sound without the resonant hump in the typical electric guitar sound location. As an alternative you can feed the XLR ditectly into a Shure A95U mic matching transformer directly plugged into the guitar amp input. At the low impedance mic input impedance of 150 to 250 Ohms, the cable capacitance and winding capacitance will have a minimal audio effect.

These are just some tips from tinkering with guitar pickups for over 50 years.

Thanks

Joseph J. Rogowski

[Antigua]

Antigua,

You made a nice 4 wire strat test pickup. I have one more test for you to try on this pickup that will put other issues to bed. Using a dual trace scope connect one scope channel to the inner winding and the other channel to the outer winding.

I know some people aren't satisfied that an inducer coil accurately represents a moving guitar string, but the inducer coil I use come pretty close to resembling the magnetic orientation of a magnetized guitar string

...

What at you want to do if form a good mental model about what happens to the pickup when you make a flat and wide pickup versus a tall and narrow pickup.


This plot shows shows the inner versus the outer winding, not plotted simultaneously, but with the inducer coil not having been moved at all between plottings:



The outer coil shows a 2 dBV increase over the inner coil. It would be interesting to experiment with a Jazzmaster style coil, add winds, then test it, add more winds and test again, etc. For a given load, the voltage generated is proportional to the total flux change through the loops of wire, so it seems apparent that the loops of the outer coil experience more overall flux change than the loops of the inner coil. I'm guessing there's a coil area of peak efficiency, beyond which the magnetic field becomes increasing parallel with the loop, and eventually even containing portion of the return path of the magnetic field, which would mean less overall flux change, less overall voltage.

Also note the extremely similar resonances, apparently because the coils share capacitance, which helps ensure the peaks will be close, if not identical.

If you follow this to its logical conclusion you could make a pickup coil with an upper winding and a lower winding and see the interaction of the coils in this configuration.

We kind of have a top/bottom coil in the form of stacked pickups. I've compared the two halves of those before, and as to be expected, the top coil produces a greater voltage. I'll have to look at it again though and see if there was a strong "reflected capacitance" in that case.

One thing I think is worth trying though, is designing a pickup that puts the greatest number of turns possible, as close to the strings as possible, so you'd use 44AWG, or 45AWG, the smaller the better, and created a short coil of 8,000 turns, and then compare the voltage production of that small, short coil, compared to the a fatter, taller 42 AWG coil. In theory, since the smaller coil places more loops of wire closer to the string, it should capture a greater flux change than a typical 42 AWG coil, where many of the winds end up being farther away from the guitar strings.

If you think about what a PAF humbucker is doing, it's similar in theory: you have a tall Strat pickup with 8000 turns, you take 4000 off the bottom, give them their own coil, put it beside the first coil close to the strings. Those 4000 turns that were unproductive on the bottom are now just as productive as the 4000 turns that were above them, so it's like you have two "tops", and a greater voltage in turn.

AlNiCo poles pieces are less permeable than steel pole pieces, so the issue of distance between string and wire loop would disproportionately effect AlNiCo pickups.

[David Schwab]

I would have expected the inner coil to be the larger of the two. What happens when it's done that way?

And what if the two coils are stacked instead of coaxial?

Note: I haven't read all the posts yet

I finally got around to creating a 4 lead tapped single coil of my own:

[ATTACH=CONFIG]48601[/ATTACH]

Here are measurements of the four lead tapped single coil:

Inner coil without outer coil in place, 4000 turns
DCR: 2.935k
L: 578mH
Peak f: 16.3kHz

Inner coil with 4000 turn outer coil in place, not connected
DCR: 2.900k
L: 575mH
Peak f: 10.4kHz

Outer coil, 4000 turns, inner coil not connected
DCR: 3.117k
L: 711.8mH
Peak f: 10.0kHz

Both coils in series, 8000 turns total with connections at 4000 turns:
DCR: 6.028k
L: 2.282H
Peak f: 9.48kHz

So, the presence of the outer coil does bring the resonant peak of the inner down by quite a lot, from 16.3kHz down to 10.4kHz, but interestingly the 100Hz inductance measurement with the DE-5000 didn't change.

Here are several plots:

[ATTACH=CONFIG]48600[/ATTACH]

The most important takeaway is that the outer coil limits the resonant peak of inner coil, regardless of whether it's connected at one end or not, so the current industry practice of using three leads doesn't represent a drawback. There is very slight difference, but that would surely disappear in a load context. When the inner coil was by itself on the bobbin, the resonant peak was 16.4kHz, but with the presence of the outer coil, it dropped to 10.4kHz, and having a single connection at the ground side only reduced the resonant peak by another ~200Hz. And as was seen with production tapped single coils, the resonant frequency of the tapped coil(s) is nearly the same as that of both coils connected in series.

The two overlapping lines that dip way down show that when the outer coil is shorted, it becomes a huge load upon the inner coil, similar to eddy currents caused by covers and base plates, but much more extreme. The fact that they overlap means that the result is the same even if the shorted outer coil is connected to ground at the short point.

With this four conductor tapped coil there are other things that can be tested for, but overall it looks like
Helmholtz is correct, that the practical value of a tapped single coil is very limited by an unavoidable interaction between the two coils.