Moving Coil Pickups for the Technically Curious

[cabezon]

I was looking at utilizing this system a year ago as I have been looking for a setup that allows for individual string outputs. I abandoned this idea when it struck me that the conductive frets would create an issue. Since then I have been thinking about how to utilize such a simple setup and have come up with several possibilities.

1. Refret the neck to where the frets do not extend the width of the neck. The smaller frets would be only as wide as necessary for the string. It would require a separation space or a nonconductive material between itself and its neighboring fret so as not to pass the current over.

2. Find or create a nonconductive fret material

3. Go fretless!

I am very interested in this chat as I have been looking for some time for the information posted here.
Cabezon

[bbsailor]

I was looking at utilizing this system a year ago as I have been looking for a setup that allows for individual string outputs. I abandoned this idea when it struck me that the conductive frets would create an issue. Since then I have been thinking about how to utilize such a simple setup and have come up with several possibilities.

1. Refret the neck to where the frets do not extend the width of the neck. The smaller frets would be only as wide as necessary for the string. It would require a separation space or a nonconductive material between itself and its neighboring fret so as not to pass the current over.

2. Find or create a nonconductive fret material

3. Go fretless!

I am very interested in this chat as I have been looking for some time for the information posted here.
Cabezon

Cabezon,

There is no need to go to all the trouble to splitting the frets. Here is why. Consider the nut end of the string as the common ground end. Use the truss rod as the common ground return path connected to the metal nut. This way if you fret multiple strings across a fret, the hot ends of the strings are still isolated behind the bridge end of the strings and are still free to produce an independent output. The output of each string is between the common ground and each individual string isolated from touching a common metal bridge. Use nylon inserts or Graphtech bridge inserts to prevent shorting out of the strings. The best transformers to try are miniature audio output (4 or 8 ohms to 20K to 50K) used with the low impedance side connected between each string and the common ground.

The string impedance is so low that making the strings shorter will not affect the output level.

I hope this helps.

Joseph Rogowski

[earthtonesaudio]

The string impedance is so low that making the strings shorter will not affect the output level.

I hope this helps.

Joseph Rogowski

That's just what I was looking for, thanks!

[earthtonesaudio]

Alright, a new related question:

How much hum should one expect to hear from the string output, after impedance matching and/or amplification to "normal" (compared to a standard pickup) levels?


By the same token, what would be the most practical method of humbucking, if you wanted to keep the individual string signals separate for hexaphonic purposes?

My first thought is to run a dummy string through the neck, and subtract it's output from each individual string's output.

[bbsailor]

Alright, a new related question:

How much hum should one expect to hear from the string output, after impedance matching and/or amplification to "normal" (compared to a standard pickup) levels?


By the same token, what would be the most practical method of humbucking, if you wanted to keep the individual string signals separate for hexaphonic purposes?

My first thought is to run a dummy string through the neck, and subtract it's output from each individual string's output.

earthtonesaudio,

I'm glad someone has tried this.

Hum should be rather low because the string is effectivly grounded at one end, the nut end through the ground return. Make sure that all the following are grounded.

1. Truss rod or ground return from the nut side of the string
2. Transformer secondary ground connection from each string
3. Transformer metal frame

To humbuck, simply reverse the phase of alternate strings. Use an oscilloscope to see the phase of the input to each transformer and the outputs. Any common noise should be neutralized like in a traditional humbucker.

Dummy String? The easiest method is to use the truss rod as the ground return or remove the fingerboard and lay a thin copper strip under the fingerboard.

If you use two strings for one string path you do not get individual string output. You need the common ground return from the nut to get an individual hex output for midi.

Please describe what you discovered when you tried this moving string pickup? How did you implement it? What is your output from each string?

Here is tip. The longer the magnetic field under the string the higher the output. If you put too strong of a magnet under the string, it will damp the string vibrations. Try using the Radio Shack 1" rectangular ceramic magnets, part number 64-1879. There are 5 magnets per pack so three packs should give you a 5" long magnetic field under the strings with the 1" side of the magnets in line with the strings, or three across. Put some foam under the magnets and you can use screws in the holes in the magnets to fine tune the distance from the magnets to the strings. Position the magnets closer to the string as they get closer to the bridge to balance the upper harmonics where the string has less vibration amplitude.

This is fun stuff and an interesting alternative to traditional pickups.

Joseph Rogowski

[earthtonesaudio]

Thanks Joe,

I haven't actually built anything yet, but I plan on prototyping something simple (maybe just a 2x4 board with strings) to test out these ideas.

My main concept is to use an active preamp to boost the weak signal from the strings. I figure it's going to be a lot of gain, so any common mode noise induced in the strings has to be eliminated.

Your comments about using pairs of strings to cancel common mode noise leads me to believe that a "dummy string" buried in the neck would probably work. I guess the key would be to keep the dummy as physically close to the signal string as possible.

[bbsailor]

Thanks Joe,

I haven't actually built anything yet, but I plan on prototyping something simple (maybe just a 2x4 board with strings) to test out these ideas.

My main concept is to use an active preamp to boost the weak signal from the strings. I figure it's going to be a lot of gain, so any common mode noise induced in the strings has to be eliminated.

Your comments about using pairs of strings to cancel common mode noise leads me to believe that a "dummy string" buried in the neck would probably work. I guess the key would be to keep the dummy as physically close to the signal string as possible.

earthtonesaudio,

The so called "dummy string" is actually the ground return. What you want is a very low resistance return path so that the signal feeding the passive input transformer (3.2 ohms to 20K ohms impedance) without the noise of active circuits. Try measuring the impedance of each guitar string with the Exech LCR meter. You want the return path to be about 10 times lower in resistance than the lowest resistance string which is at about 0.5 ohms.

The fun comes in when placing magnets under the strings to emphasize different harmonics.

Joseph Rogowski

[earthtonesaudio]

Yes, I was imagining using a low-resistance path (truss rod, aluminum beam, etc.) as the ground return for all strings, and separate, 1-per-string "dummy strings" solely for noise rejection.

[David Schwab]

and separate, 1-per-string "dummy strings" solely for noise rejection.

I don't think it works that way. On a pickup with a coil, the coils are wired in reverse and each has an opposite magnetic polarity. Whereas the string has a resistance and will get current flowing when in a magnetic field, it's not going to buck hum with another string unless you are presenting an opposite polarity for the common mode noise.

Now if you had the strings in some kind of pairs with opposite magnetic polarity, or at least wired into the transformer, that would work.

[earthtonesaudio]

I was thinking something like this (see attached drawing).

The signal and dummy strings connect together at the nut end of the guitar, along with the ground return. The signals are taken at the bridge end of the guitar. The dummy string runs parallel to the signal string, buried in the fretboard or somewhere close like that.

The dummy string doesn't move, so the magnetic field does not affect it at all. It picks up pretty much the same environmental noise as the signal string, and the differential amplifier cancels the noise while providing voltage gain and line driving ability.


The ground return has to traverse the length of the neck, so in order to have comparatively low resistance and inductance, it would have to be a thicker "wire," possibly like a truss rod or metal reinforcement beam.

[David Schwab]

The dummy string doesn't move, so the magnetic field does not affect it at all.

It will effect it if there is a moving string carrying current over it. Pickup coils don't move either.

Look up these patents by Clevinger:

4408513

4499809

They are something similar

[earthtonesaudio]

It will effect it if there is a moving string carrying current over it. Pickup coils don't move either.

Hm. I think you're right.

But I imagine it must be a small effect, if the strings are a fretboard's thickness apart. This geometry can't possibly make for a very good transformer.

[Steve Conner]

I don't see the point of any of these dummy strings. All that matters is the loop area formed by the strings and the ground return, the system will pick up magnetic hum in proportion to that area. The impedance is so low that it shouldn't pick up electrostatic interference at all, provided of course that you ground one point of the loop, which you should.

I think Mr. Rogowski recommended reversing the polarity of some of the strings and also reversing the magnets on those strings. If you can get equal loop areas that contribute to the output in opposite polarities, you've got a humbucker. You may not even need to reverse the magnets: does the phase of one string relative to another matter?

If you connected the odd-numbered strings in parallel to the non-inverting input of a diff amp, and the even-numbered ones likewise to the inverting input, that's probably about as good as the humbucking would get, though at the cost of losing individual string outputs.

I'm not sure if a ground return down the neck would even be needed in this case, as long as the strings are all connected by a brass nut or the like. Electrostatic interference would hit the diff amp's inputs in phase and be cancelled.

Then again, a copper fingerboard with ceramic frets, wouldn't that look neat?

[Joe Gwinn]

The dummy string doesn't move, so the magnetic field does not affect it at all. It picks up pretty much the same environmental noise as the signal string, and the differential amplifier cancels the noise while providing voltage gain and line driving ability.

As drawn, IC1 is not a differential amplifier. If you change the IC1 circuit to be a true differential amplifier, humbucking could be achieved.

However, you also need to design to tolerate a static electricity discharge to either input of IC1, or the circuit will not survive the first dry day.

[Steve Conner]

Oh, yeah good point... I forgot about ESD, we haven't had a dry day in Scotland since before ICs were invented.

I prefer the passive idea with the step-up transformers myself. What if you connected the strings all together at the nut, then at the bridge: connect all the odd strings to point A, all the even strings to point B, and then hook the step-up transformer between points A and B?

You could ground the system at A or B, or the nut, it makes no difference that I can see.