Moving Coil Pickups for the Technically Curious

[bbsailor]

A guitar or bass with metal strings has the potential to utilize a new way to sense the vibrations from the string. Just think of a metal string as a "moving conductor" or a "wire in a magnetic field" (web search on the words within quotes for more information).

I'll share with you what I have learned from tinkering with pickups for almost 50 years.

A voltage can be generated when a wire is moved through a magnetic field. To see this with a quick experiment do this. Use an acoustic guitar and attach a scope probe across one string then move a cermic magnet near the string. Set the scope the a low mv range. Plunk the string and you will see an output in the 2 to 5 mv range. Just like traditional high impedance pickups with thousands of coil turns around magnets, the output of the string motion is a function of the actual motion vector at right angles to the field direction. This is defined a Lorentz Voltage.

Lorentz Voltage is defined as E=LB dx/dt sin (/) (my poor excuse for a Greek letter) where:
dx/dt is the velocity of the moving conductor;
B is the strength of the magnetic field;
L is the length of the conductor in the magnetic field; and
(/) is the angle of the direction of conductor motion relative to the direction of the magnetic field.

Based on the above formula and my own experiments, the voltage induced within the string increases as the length of the magnetic field is increased. I used a .5" wide by 6.5 " long by 0.125" thick rubberized magnet running under the string from the end of the fingerboard to near the bridge. By adjusting the magnet height to be closer to the string near the bridge I can balance the harmonic output. This arrangement puts a full quarter string length within the magnetic field.

Now for the fun part. Obtain a low frequency current transformer such as a "CSE187L" for about $2.50 (single unit price) from Allied or Jameco. If you have a scope current probe you can use this also by just putting the probe around the string behind the nut and use a thick jumper wire to connect the string behind the probe and the bridge end of the same string. This now forms a single turn loop with the string acting as the ribbon in a "ribbon microphone". Use the CSE187L by attaching the primary (single turn transformer wire) to the string with jumper wires attached to the string behind the nut and behind the bridge. Attach your scope or amp to the secondary and observe or listen to the sound. The output of the CSE187L is about 21 ohms DCR and matches a low impedance XLR to high Z matching transformer 500 to 50,000 ohms at a 1:10 turns ratio very well or a microphone mixer XLR input.

The CSE187L has a 1:500 turns ratio and is rated for a 50 to 400Hz frequency range but it goes well beyond that. You can even connect two strings in series by electrically connecting the individual machine heads of the bass E and A strings together and attach the CSE187L primary behind the bridge to these two strings and make a quick two string pickup for test purposes.

For a little more output you can obtain a "low frequency torroid current transformer" with turns ratios up to 1: 3000. Just run a jumper from the string behind the nut, loop it through the torroid as the primary loop and connect the other end behind the bridge.

You might ask, why has this pickup not been popular?

1. You need to have a ground return from the strings at the nut end into the body of the guitar near the bridge.

2. It does not just drop in with a standard single coil or humbucker footprint.

3. The strings can act as an antenna and pickup hum.

As another experiment take a soldering gun and use it as a hum source. When a single string or two series strings are connected to an amplifier, move the soldering gun near the string and listen to the hum.

Here is the challenge for the technically curious.
(1) How could you apply traditional humbucking techniques from traditional high impedance pickups to this moving coil pickup?

(2) Using the current transformer techniques described above, how could you adjust the output level of each string to obtain a more acoustically balanced string output.

(3) What other musical applications can be derived form having individual and isolated string outputs.

I hope this stimulates some thinking and introduces another approach to making guitar pickups. I hope some of you try these quick experiments and share your results.

Joseph Rogowski

[StarryNight]

Neat Ideas! Prelim. questions...

so if this were applicable to all six strings would there be a need for six transformers, or would it be possible to connect all strings in series?

If all strings were connected in series to one transformer, then if you played a barre chord with six strings vibrating would the total voltage increase making chords much louder than when playing single strings?

Would there be a problem with eddy currents with that much magnet under the strings, or would it be relative to the proximity of the tranformer(s)?

[bbsailor]

so if this were applicable to all six strings would there be a need for six transformers, or would it be possible to connect all strings in series?

You could do it with six transformers to have individual string outputs. However, you could also use a single toroid and loop each string through a separate primary toroid winding, thus mixing the six strings in one transformer. On my first guitar modification with this pickup, I removed the guitar fingerboard; put a strip of copper sheeting down the neck, soldered the top of the copper sheet to a brass nut; routed the other end of the copper sheet into the body of my Applause acoustic guitar; mounted six 6 ohm to 20K transformers on the heel of the neck wrapped in copper foil for additional shielding. I used a copper rivet in each bridge hole to thread the ball of the string through; soldered a wire lead to each copper rivet that attached to that string's transformer and the common ground return from under the neck. The output from each string 20K side went to a separate pin of an 8-pin microphone connector with a common ground. Then I made a breakout cable with the 8-pin connector on one end and the broken-out leads with a 1/4" plug so I could plug the guitar into a 6-channel microphone mixer. I had individual volume, EQ, and panning of each string on the mixer. I mounted ceramic magnets on a wood bar that mounted in the sound hole like a clamp-in acoustic guitar pickup.

If all strings were connected in series to one transformer, then if you played a barre chord with six strings vibrating would the total voltage increase making chords much louder than when playing single strings?

Not necessarily.

Would there be a problem with eddy currents with that much magnet under the strings, or would it be relative to the proximity of the tranformer(s)?

The rubberized magnets have less intensity than ceramic magnets and can be installed with three .5" wide strips, one strip for two adjacent strings. Eddy currents are not a problem. The transformers are typically located behind the bridge away from the magnets.

I hope this encourages you to give this a quick try to experiment on one string with a current-based pickup. You can try this with just two alligator clips and one 8 ohm transformer (70 volt, low wattage speaker matching transformers also work) to better match the low-Z of the string to something more useful for an amp input. Use any hand-held ceramic magnet to listen to the sound and the better upper harmonics masked by high-Z pickups.

Joseph Rogowski

[Joe Gwinn]

A guitar or bass with metal strings has the potential to utilize a new way to sense the vibrations from the string. Just think of a metal string as a "moving conductor" or a "wire in a magnetic field" (web search on the words within quotes for more information).

I'll share with you what I have learned from tinkering with pickups for almost 50 years.

A voltage can be generated when a wire is moved through a magnetic field. ....

There are a number of old patents on just this. With modern rare earth magnets, it may have become practical at last.

[bbsailor]

There are a number of old patents on just this. With modern rare earth magnets, it may have become practical at last.

Joe,

You are correct about early patents.

When you try to maximize the voltage from each individual string through a separate transformer you want a transformer with 5 to 10 times the impedance of the string typically between about 0.5 ohms to about 2 ohms. The size of the core wire will primarely govern the impedance of each string with the low E string having the lowest impedance and the high E string having the highest impedance. This is sort of like matching the ribbon in a ribbon microphone. Now you have six string ribbons to balance with each string having a different center frequency and impedance range.

However, when you try to combine the strings using a single transformer, using a current transformer (CT) approach produces a better and more economical approach. You can think of putting the CT in the common ground return leg of the circuit using one CSE187L or by using a toroid with several thousand turns and wrap a lead from each string through the toroid back to the ground return to attempt to balance the output from each string.

With a little creative alternate phasing of adjacent strings, you can even achieve a degree of humbucking to keep the strings from acting as hum antennas. This approach opens up the economical possibilities of using a single CT costing less than $10.00.

New methods of combining the truss rod neck strengthing/adjustment function
with a ground return electrical function can offer some new design opportunities for building a wide bandwidth pickup.

Putting too strong a set of magnets will dampen the strings. That is why I used the rubberized magnets over a longer length of the string for more higher harmonics. There is room for experimentation to optimize for a good, balanced sound by using various magnet configurations.

Right now I am thinking the guitar output should be more aligned with the range of microphone output at low impedance to maintain fidelity. 500 ohm to 50,000 ohm matching transformers are cheap and you only need one at the amp end or use an outboard microphone preamp to feed the guitar amp.

I welcome your thoughtful comments.

Joseph Rogowski

[Possum]

...

I ran across that patent as well, but have no idea what number it was or if I saved it, its not a new idea. Problems that could happen is with the bass strings being fatter would generate a higher output than the treble strings, but I guess with a regular pickup they do anyway. Its a cool idea but would probably give a very hifi sound and not appeal to mass market, but then again I guess Lace sells those Alumatone things.....

[jonson]

probably tinking of these Possom.

[David Schwab]

Another one, but more complicated.

[Possum]

...

So much for new ideas, eh? Yeah those are it, though haven't seen the last one before. I think it would be a hard thing to pull off and probably wouldn't sell. Just getting good electrical contact with the strings would require special hardware and different strings would probably conduct differently maybe. Still would be interesting to hear what it sounds like...

[bbsailor]

So much for new ideas, eh? Yeah those are it, though haven't seen the last one before. I think it would be a hard thing to pull off and probably wouldn't sell. Just getting good electrical contact with the strings would require special hardware and different strings would probably conduct differently maybe. Still would be interesting to hear what it sounds like...

Possum,

Here is a quick 5 minute way to try this assuming you have a 4 or 8 ohm primary transformer with a 5K to 20K output. Radio Shack small 70 volt speaker transformers work well using the lowest wattage tap which gives the highest turns ratio.

Alligator clip the low impedance side of the transformer to two adjacent strings on the guitar neck behind the nut. These two strings will go over the metal bridge (shorted together) and will be then seen by the transformer as being in series. Attach the high impedance side of the transformer to a 1/4" phone jack. Connect the jack to an amplifier with a coax guitar cable. The magnets on the traditional guitar pickups will give you some sound but it will be rather weak because the magnetic field is narrow. Take a hand-held ceramic magnet or single coil pickup and hold it or place it parallel with the strings for a longer magnetic field and you will hear the output increase. Listen to the higher harmonics.

By moving the magnet closer to the neck it will have a more mellow sound and when moved closer to the bridge it will have a sharper sound.

Use double stick tape to mount the magnet under a string set so you can play these two adjacent strings to listen to it while playing simple music. Put it on the two bass strings and you can fingerpick and send the bass notes to a separate amplifier. No permanent mods to your guitar are required to do this quick test.

No special hardware is needed to make contact with the strings. Just ensure that the bridge inserts are metal and clean, and that the nut is non-conductive (for this quick test).

I am not trying to change the pickup-maker-market but just trying to share ideas to offer more alternatives. Ultimately, the sound is in the ear of the beholder. Give it a try and listen to it. Read the patents and compare what you hear to the claims made by the inventors who mention better higher harmonics.

Using this technique shifts the pickup-maker requrement to wind thousands of turns of very thin wire to using off-the shelf components and mechanical assemblies to package and embed these pickups inside or on the guitar. Here is where creative adaption by the pickup-make community where the minds many are much better than the worn-out mind of one.

Joseph Rogowski

[Joe Gwinn]

When you try to maximize the voltage from each individual string through a separate transformer you want a transformer with 5 to 10 times the impedance of the string typically between about 0.5 ohms to about 2 ohms. The size of the core wire will primarily govern the impedance of each string with the low E string having the lowest impedance and the high E string having the highest impedance. This is sort of like matching the ribbon in a ribbon microphone. Now you have six string ribbons to balance with each string having a different center frequency and impedance range.

Yes.

However, when you try to combine the strings using a single transformer, using a current transformer (CT) approach produces a better and more economical approach. You can think of putting the CT in the common ground return leg of the circuit using one CSE187L or by using a toroid with several thousand turns and wrap a lead from each string through the toroid back to the ground return to attempt to balance the output from each string.

With a little creative alternate phasing of adjacent strings, you can even achieve a degree of humbucking to keep the strings from acting as hum antennas. This approach opens up the economical possibilities of using a single CT costing less than $10.00.

New methods of combining the truss rod neck strengthing/adjustment function with a ground return electrical function can offer some new design opportunities for building a wide bandwidth pickup.

All true, but I suspect that it didn't work enough better than a traditional pickup to justify the complications. That said, rare earth magnets may have tipped the balance. Or maybe not.

Putting too strong a set of magnets will dampen the strings. That is why I used the rubberized magnets over a longer length of the string for more higher harmonics. There is room for experimentation to optimize for a good, balanced sound by using various magnet configurations.

Actually, it isn't really (or wholly) the strength of the magnet that matters. If you have a steel string vibrating perpendicular to a strong but uniform field, there will be no wolf tones. If you do not complete the electrical circuit and short the string, there will be no damping either. If you provide a path for the voltage induced in the vibrating string is drive a current, there will be damping in proportion to the magnitude of the driven current, but no wolf tones.

It's the non-uniformity of the field that causes wolf tones, and also causing damping. It's the side pull that varies with distance of string from the magnet that causes the problem.

The test is a aspirin-pill sized magnet with two parallel strip-shaped mild steel polepieces. The string passes between the polepieces, where the magnetic field from the pill (also between the polepieces) is fairly uniform. The field is parallel to the pickguard, so vibration perpendicular to the pickguard will induce max voltage. This is easy to arrange on a test guitar without modifying the guitar. The magic ingredient is duct tape.

Right now I am thinking the guitar output should be more aligned with the range of microphone output at low impedance to maintain fidelity. 500 ohm to 50,000 ohm matching transformers are cheap and you only need one at the amp end or use an outboard microphone preamp to feed the guitar amp.

This can work. But high fidelity isn't necessarily desired.

[earthtonesaudio]

About to try this!

I have Graphtech-style (non-conductive) saddles installed in a Tune-O-Matic style bridge, and I've got hold of some polycarbonate sheeting for the tailpiece.

My guitar looks just like this.


My plan is to not do any permanent mods, so the fretboard is staying on and I'll be running some metal tape along the back (or front if it fits!) of the neck for string ground.

I'll make the poly tailpiece section large enough to hold the additional electronics and controls. Right now I'm strongly leaning toward BJT common base amplifiers for each string, and actively canceling common mode hum/noise. Looking for some opinions/guidance in this area.

It will probably take me a while to get everything together, but I'm really looking forward to this!

[David Schwab]

I have Graphtech-style (non-conductive) saddles installed in a Tune-O-Matic style bridge, and I've got hold of some polycarbonate sheeting for the tailpiece.

Is that going to be strong enough? You might want to pick up a cheap stop tail and then drill out the holes bigger and line them with some kind of non conductive tubing.

My plan is to not do any permanent mods, so the fretboard is staying on and I'll be running some metal tape along the back (or front if it fits!) of the neck for string ground.

If you are doing that to ground your hand, you will then be shorting out the strings when you touch them. I wouldn't worry about the string ground.

I'll make the poly tailpiece section large enough to hold the additional electronics and controls. Right now I'm strongly leaning toward BJT common base amplifiers for each string, and actively canceling common mode hum/noise. Looking for some opinions/guidance in this area.

It will probably take me a while to get everything together, but I'm really looking forward to this!

Sounds like a cool project!

[earthtonesaudio]

The signal will be taken from the bridge end and the nut end needs to be grounded for there to be a complete circuit. I hadn't really thought about it before, but I think compared to the milli-ohms resistance in the strings, my body is basically an open circuit, so I'm not really worried about shorting the string signal before it makes its way to the bridge.

I plan on using a large piece of poly to make the tailpiece (maybe sort of like a trapeze style, extending all the way to the strap button), so the forces from the anchor posts vs. the string tension will be spread over a large area.

Another reason for using polycarbonate is that it's clear and I want to put silly lights inside.

[earthtonesaudio]

I've been thinking about this and I have a technical question.

First is, the frets are conductive. So if the ground return is the nut, won't the effective resistance be altered whenever you fret two or more strings on the same fret?

For instance, if you played the high E string, fretted at the 12th, you have the vibrating portion of the string providing an AC signal and the stationary part of the string providing some resistance to ground. Then if you were to fret the low E at the same fret, there are now two parallel paths to ground. Would this cause the gain of the system to change?