Moving Coil Pickups for the Technically Curious

[jrfrond]

It's always refreshing to see new approaches to an old idea. It proves that most technically-oriented people are always thinking.

The downside to all of this is that many musicians, and not just guitarists, are hardcore traditionalists and things like this don't fly. I saw a prototype system similar to this a LONG time ago, where the strings, which were magnetized ala the early Rickenbacker horseshoe magnet system, went through small circular coils near the bridge, which were fed to opamps. Then there was a guitar that I THINK was called the Lightwave, using infrared LED's and optocoupler's to read string vibration. Both were touted as revolutionary, both died before production commenced. Even the Gibson Robot has flatlined.

The problem is that musicians always SAY that they are open-minded and want something different, when in fact, they just want better versions of the old stuff. Go figure!

[Steve Conner]

Totally, I mean we're still using vacuum tubes, for goodness' sake!

I need an EMI-proof guitar for my Tesla coil guitar act, so the Lightwave thing would have been great.

[Mike Sulzer]

I think you need the step up transformers in order to get good enough signal to noise ratio (over preamp noise). Also this has the advantage of providing the possibility of a differential input if you need it.

[David Schwab]

Then there was a guitar that I THINK was called the Lightwave, using infrared LED's and optocoupler's to read string vibration. Both were touted as revolutionary, both died before production commenced.

Lightwave is still in business!

LightWave System Pickup

Then of course there was the Hoag optical pickup. He was the first.

HOAG OPTICALGUITARS SITE

Roland did this string as coil idea in reverse with their first guitar synth the GR-500. It had a large magnet under the strings by the neck and it ran the output of each string back to the actual string. That created infinite sustain.

I always thought this was really exciting stuff. This got me into messing with optical pickups and hall effect sensors back in the 70's. I should try that stuff again.

[bbsailor]

It's always refreshing to see new approaches to an old idea. It proves that most technically-oriented people are always thinking.

The downside to all of this is that many musicians, and not just guitarists, are hardcore traditionalists and things like this don't fly. I saw a prototype system similar to this a LONG time ago, where the strings, which were magnetized ala the early Rickenbacker horseshoe magnet system, went through small circular coils near the bridge, which were fed to opamps. Then there was a guitar that I THINK was called the Lightwave, using infrared LED's and optocoupler's to read string vibration. Both were touted as revolutionary, both died before production commenced. Even the Gibson Robot has flatlined.

The problem is that musicians always SAY that they are open-minded and want something different, when in fact, they just want better versions of the old stuff. Go figure!

John,

The purpose of starting this thread was to offer another perspective on building a string transducer. This may not take over the world like traditional replacement pickups have done when they easily fit a standard pickup humbucker, single coil, or P90 footprint. However, this new technique offers technical guitar people a way to easily experiment with using technologies that were common in the early days of radio and ribbon microphones but with each string functioning as an individual and isloated ribbon or as part of a collective set of ribbons. This thread simply explains how approaches different from thousands of turns of very fine wires around a magnet can also produce sound and provide individual, isolated string outputs for those who might want to use that characteristic.

Add to this the advent of digital signal processing, midi, hex pickups where there is an interest in saying; what if i can use the signal induced on a moving string in a magnetic field? Then, what happens is a discussion like has evolved in this thread.

The ear is the final arbitrator in what sounds good, Science attempts to explain what people seem to like and also tries to replicate it. Knowledge gives guitar builders and techies additional alternatives.

Joseph Rogowski

[earthtonesaudio]

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.

You're quite right. Here's a low-power quad of fully differential amplifiers, enough for four strings. That ought to simplify things somewhat, as long as you're comfortable soldering the 32-pin TSSOP!
I agree about the static protection; a couple caps or diodes should do the job.


@ David Schwab:
That Roland guitar sounds interesting.
Perhaps a novel system could use both. Optical pickups to sense the string, and electromagnetics to control its vibration.

[bbsailor]

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!

earthtonesaudio,

Look for inexpensive, miniature audio output transformers. 3.2 ohms to 8 ohms are common low z measurements and 2K to 200 K are available high impedance sides. Even with a lower turns ratio (3.2 ohms to 2K will even work as it has 1:25 turns ratio), it will allow you to make a lower noise design than using active at the front end. With this setup you could probably get away with a single IC stage with 6 input mixer set to a gain of between 10 and 20 to provide an output in the 300mv (peak) range (plus or minus 100 mv).

Based on the photo of your guitar, try this.

1. Remove the metal tailpiece. Replace it by making a replacement out of .5" thick Plexilass or Lexan.

2. Have the strings go through .125" copper riviets which has the hot connection for each transformer connected to it and the common ground return.

3. Run a tamporaty thich, fine stranded awg 10 wire which conncts the brass nut to the rear of the bridge as the ground return noted above and connected to the other end of each string tranmsformer.

4. Mount all six transformers on the plastic tailpiece on a citcuit board along with the next item.

5. Use miniature trimmer pots to adjust the level of each string output.

6. Decide whether you want to mix the output to a single sum of all six string output or provide the output of each transformer to drive a midi systen with a separate input from each string.

7. Send the outputs of the low E and A strings to a footswitch that selects which string you want to be sent to an octave divider to provide synthesized bass in real time in sync with the isolated output from either bass string even while playing chords. The foot switch tells the octave divider which string contains the desired bass note. Octave dividers only like one note at a time as input.

Obtain a few packs of flat 1 inch Radio Shack magnets 64-1879. I like them because they can be put on top of a foam pad and screwed into my test jig (guitar neck screwed to a pine board), and the heights of each magnet can be adjusted to fine tune the harmonics. I use them with the 1 inch side length running under each string... three across is 2.25 inches.


Joseph Rogowski

[Joe Gwinn]

You're quite right. Here's a low-power quad of fully differential amplifiers, enough for four strings. That ought to simplify things somewhat, as long as you're comfortable soldering the 32-pin TSSOP!

Looks good, but be sure it's low enough noise.

I agree about the static protection; a couple caps or diodes should do the job.

The classic is reverse-biased 1N4148 diodes between each input and ground, and between the same input and Vcc, a big filter capacitor between Vcc and ground, and a zener diode across the big filter capacitor. The net effect is to constrain the inputs to remain between zero and Vcc volts.

[Steve Conner]

Looks good, but be sure it's low enough noise.

It won't be... you'll need a super low-noise discrete front end like audiophiles use on their moving-coil record pickups. Or a transformer.

[Joe Gwinn]

It won't be... you'll need a super low-noise discrete front end like audiophiles use on their moving-coil record pickups. Or a transformer.

Audiophiles? Thanks for reminding me. There are some specialist makers of analog ICs intended for use in high-end audio systems, and they may have what you seek. For instance:

That Corp: THAT Corporation 1510/1512 Audio Preamplifier ICs.

AD829 is often used as well.


The audio folk also sell matched transistor arrays:

THAT Corporation 300 Series Matched Transistor Array ICs

Linear Integrated Systems - Press Release LSK389

A matched transistor array can be used as the input differential amplifier for a lesser opamp.

[earthtonesaudio]

Ugh. The 1510/1512 looks great, but that's a minimum $100+ order. Too rich for my blood. If a standard opamp I-V converter is too noisy, that would pretty much be a dealbreaker for me. I want minimum complexity, minimum size. So transformers and discrete circuitry are both a bit off-putting to me.

I think I ought to do some real-world experiments before I overthink everything.

[Steve Conner]

Try the INA217 or whatever it's called. I still think a transformer is worthwhile, it should take up less space than the battery you'll need for an active system.

I think I ought to do some real-world experiments before I overthink everything.

Too late, it's already overthought.

[David Schwab]

I'd go with a transformer too, even if it was before an active circuit.

[Joe Gwinn]

Ugh. The 1510/1512 looks great, but that's a minimum $100+ order. Too rich for my blood. If a standard opamp I-V converter is too noisy, that would pretty much be a dealbreaker for me. I want minimum complexity, minimum size. So transformers and discrete circuitry are both a bit off-putting to me.

Umm. We all want cheap and simple. Low enough noise too. The question isn't desire, it's practicality.

But there are lots of analog ICs available. Perhaps someone knows of a more common and yet adequate one.

[Mike Sulzer]

You would not expect to find an IC for this application; it is pretty special. The requirement is for extremely low noise voltage; the noise current does not matter, of course, since the source impedance is so low. Junction transistors generally have lower nose voltage (and higher noise current) than FETS or tubes. Transistors with large areas, or multiple devices in parallel are two ways to go. This is not what you tend to find in ICs. But the transformer wins for simplicity. You can match normal devices, and the differential input is free.

Umm. We all want cheap and simple. Low enough noise too. The question isn't desire, it's practicality. If you can you avoid differential inputs since you get noise from both teh positive and negative inputs

But there are lots of analog ICs available. Perhaps someone knows of a more common and yet adequate one.