Single string pickups - 6 of 'em

[David Schwab]

I mean, someone like Pat Metheny has a nice ferrington ha^rp guitar with something like 40 strings...

You mean the 42 string Pikasso I, created by Canadian luthier Linda Manzer.

I have a photo of her with that guitar that I snapped back in '96... I'll have to scan it one of these days.

[Satamax]

Ok, that's Linda Manzer who did this one, what did Ferrington do for Metheny then????

[mjmiller]

Are the extras sympathetic or intended to be played?

I wrote a reply to Satamax's comment and then accidentally his the back button on the mouse while putting down my coffee. It was like... 45 minutes of thinking and writing... sigh

let me just say this:
I AM trying to keep it simple (I prefer KIS - no need for name calling
This is part of a big ball of wax that I just explained and deleted by accident

There are 4 possiblilties I have come up with to do what I want to do
This is one possibility that I wanted to kick around with experienced folk

And Yes, it's worth the effort to figure out if its worth the effort.

[Satamax]

Are the extras sympathetic or intended to be played?

I wrote a reply to Satamax's comment and then accidentally his the back button on the mouse while putting down my coffee. It was like... 45 minutes of thinking and writing... sigh

let me just say this:
I AM trying to keep it simple (I prefer KIS - no need for name calling
This is part of a big ball of wax that I just explained and deleted by accident

There are 4 possiblilties I have come up with to do what I want to do
This is one possibility that I wanted to kick around with experienced folk

And Yes, it's worth the effort to figure out if its worth the effort.

Hey MJ, don't worry, i was taking the mick! (huh pulling your leg?) I mean, what you tend to head to i've tried to think about a few years back, and gave up Good luck on your project.

Bye.

Max.

[mjmiller]

So here's the deal:
I don't like how harmonic distortion and intermodulation come as a package deal. The tone I choose to play with shapes my playing possibilities, and this bothers me.
I use a lot of highly nonstandard tunings, two of which place 2 strings a tone apart, and one of which has 2 strings as a unison. When I use these tunings I am forced to play with an extremely clean tone. I don't have anything against clean, I just want to be able to dial in something grittier when I like. Grit and flat seconds don't go well together - sounds like the 'skronk' that Korn's guitarist is fond of. I'm more of a melodic kind of guy. I've been messing around with lines... like... I can't remember what is called, but where each note rings out through the beginning of the next note. Like legato, but overlapping. Anyway, it's got a sweet sound(kinda haunting), but don't try it with any bite - it becomes unintelligible.
I'd also love to be able to play something other than 5th type power chords with blistering distortion, and the wide major 3rd of even temperment with high gain is the bane of my existence.
Combine these things and it becomes obvious to me that my world would be a better place if I could separate (to as high a degree as possible) harmonic and intermodulation distortions. I've come up with 4 ways to do it.
"Everything should be made as simple as possible, but no simpler." Didn't Einstein say something like that?
It all started with (Anderton's?) quadrafuzz. Hmm... distort bands separately to reduce intermodulation... but for it to help with my freaky melody lines, I'd need one band for each note, at least in that range. Totally impractical and far from simple. Scratch that idea
12 commuting filters, one for each note in the 12 tone scale could be cool. If you played a note and executed a bend, it'd drift into another band, which could make for some sweet ramifications, plus intermodulation would be smacked down big time. But I don't want to deal with a 12 channel preamp. That's still not simple enough for me.
The third method is being kicked around in my ADC/DAC post, but it'll be years before it's to the point where I can plug in a guitar, start playing, and have it do what I want. Also, I think it's performance could be drastically increased if I feed it one signal per string. Everything could be optimized (and I could distribute the necessary processing between multiple computers) Which brings us to the best solution I have come up with:
One pickup per string.
It requires the fewest preamp channels and gives me the best separation of signals. Its the simplest, most intuitive solution to my problem. The only problem is that I've never built an inductive, capacative, piezoelectric, or optical pickup, and I'm not about to dive headlong into any of those four endeavors without kicking the idea around who knows something about the subject, which is why I started this thread. I want to find out if its worth the effort.
It sounds like inductive pickups have resulted in nothing but heartache... but I wonder if all the possibilities have been exhausted. I have a dual rail humbucker where all the lines of magnetic flux curve from the blades inwards, towards each other. Shrink that, rotate it so the blades are parallel to the string (and short for decent HF response - don't want the strings in the field for an inch), and voila.
Piezos are probably the hardest to work with as far as eliminating crosstalk is concerned - way too much vibration is transferred from any part of the guitar to any other.
I don't know if anyone has ever made a capacitive guitar pickup, but my Pops drew out an idea for one that seems like it would work very well. If anyone's interested, I'll try to dig up his sketch and remember how he said the thing would work (he's a pretty brilliant, if rusty, old school electronics engineer by degree, but he hasn't had an engineering job for decades) I remember that it was a bit like an electret microphone
I also don't know if anyone's ever tried building an optical pickup and I don't know what kind of frequency response curves would be generated (different ones based on different types of optical pickups), but it would be freaking easy to build a few simple ones and also very easy to achieve perfect isolation between strings and eliminate optical noise from the environment.

[Satamax]

MJ MJ, i get your point!

Well, now you're talking. Remember that thoses coils are quite small, so they don't have a lot of output, and you will have to use a lot of gain to generate a decent volume, lots of gain results often to a bad noise to music ratio. Myself, i think i'd go for piezzo, and six guitar synth (if you can manage theses to get old roland to sound like anything good for you, it could be a cheap option) You know, if you go to the mimf, optical and inductive etc pickups have been discussed to some extent, even Hexa coils etc. To have acces to the library, you'll have to register thought. But im my understanding, you'll have at some point to mix the signal even if it's in the air after six amps, and at a certain point, i'd bet intermodulation distorsion will come back in the picture!

[mjmiller]

I'm going to have to google mimf... I've seen it referenced a lot on here. Sounds like it could be a good resource. I'm okay with mixing the signal at any point after the distortion is done. If I want that tube power amp distortion tone, I WILL have to send each string out through a separate amp and speaker, but I'm okay with that. I've seen a lot of tube amps with upwards of 6 output tubes. Just look at some of Ampeg's stuff. I think the BAGA has 12 output tubes.... and there's something... natural, I guess, about treating each string separately through the whole process. I can have cute little 1 watt ax84 firefly type amps for the unwound strings, moving up to a 5 watt amp for my low A (I'm strung like a baritone on the bottom end)

The other option is to mix it all before an ultralinear power amp. I'm okay with that, too.

As for coils - my guitar is very well shielded, and if I shield the coils as well and buffer the coils onboard the guitar, don't you think I could get a good signal out to the amp?

Well, I'm going to go see what I can dig up in this mimf... thanks for the reference.

[Satamax]

http://www.mimf.com/cgi-bin/WebX

[mjmiller]

So I just got back from the mimf... thanks for the reference
Went through what I could find on opticals there... lots of smackdown posts, but the real gem was the link to the lightwave systems site. Wow! Amazing results, but they achieve them through what I have to avoid - encapsulating part of the string in a dark chamber. Presumably, they shine an IR LED on one side of the string and pick up the vibration of its shadow on the other side. I use the whole darn string, though. Sometimes I play with a retired bow from my cello and I get right down at the bridge to pull out some freaky harmonics, so I can't encapsulate the string
My idea was to mount a laser diode and phototransistor under the string and catch the refracted light. Driving the laser with a carrier frequency then passing the output of the phototransistor through a narrow filter would allow removal of any environmental interference and each string could have its own carrier to eliminate any possibility of crosstalk. There might be too many variables, though, such as the reflectiveness of the underside of the string - could change during the course of playing if I sweat on them or get them dirty or something, and this could drastically reduce the signal strength.

Might be worth a trip to the Rad Shack and an afternoon of tinkering to see what I can get out of it..

'nother wacky pickup idea - the capacitive pickup. This one might work better. I couldn't dig up the idea my Pops sketched out, but I was thinking about it and I think I have a pretty good way to do it. More amplitude modulation of carrier signals... Way simpler than an electromagnetic pickup to construct - works like a theremin antenna.

you've got a vibrating metal string connected to ground
take a small metal plate or piece of wire or whatever and pass a carrier signal through it. As the string gets closer to the plate, the capacitance increases, and more of the carrier signal is shunted to ground. Each string has it's own carrier frequency and you pass each through a bandpass filter (again, eliminating crosstalk and environmental noise - just use freq's you're not going to pick up with your little antennae) then a lowpass to get rid of the carrier. Audio signal comes out, clean as a whistle if the circuit is well designed, dead flat frequency response, no magnetic drag or "stratitis" and one signal per string.

Any thoughts? Did I miss something big here? This is all new territory for me.
Lets's see.... say we're talking about a 1mm diameter string and our pickup it 1mm x 5 mm and is 5mm under the string. base capacitance should be... .001 x permittivity of air and I have no idea what that is
Anyway, the distance from the string and placement along its length can be freely messed with. If you think it would be really noisy, or quiet, go play a theremin for a minute They are deriving control signals, not audio ones, but they're sensitive over a couple of feet, and have a crappy connector to ground (the player) and noise present in the control signal would modulate the frequency or amplitude of the output signal, and I don't recall hearing anything like that...

Theoretically, this should have all the advantages of the Lightwave system (BTW they're like 5 miles from my old house and I never knew! Same with ... is it Seymour Duncan in Santa Barbara?) but with the high frequency response I'd like (no piezos to mix in to compensate for lack of HF response) and no string encapsulation.

And it seems like it would be way easier to build than an electomagnetic pickup - all that mucking about with magnet wire and bobbins and magnets and polepieces
Granted, there's more electronics involved - you have to make a tuned oscillator / filter pair, but I'm not sure if you'd even need the lowpass to get rid of the carrier, and the oscillator could easily drive a nice long cable

Any thoughts?

[Satamax]

Hey MJ, you're getting way out of my league! May be a HB of some king but with only three polepieces on each bobin with a one string gap between each polepiece and alternated from each bobin.

[Mark Hammer]

1) Pole-per-string pickups need to be skinny, so as to situate them close enough to the bridge where the lateral string travel will be minimized, and bleedthrough will be similarly minimized. I have an old Guild Tri-Oct polyphonic octave-down unit from what I believe is the late 60's that came with a proprietary divided hex pickup. Unfortunately the pickup has a footprint about the size of a P90 so tracking is abysmal unless you place it in lieu of a bridge pickup (won't fit between bridge pickup and bridge) and use heavy gauge string. Unfortunately, to achieve the sort of profile that lets you snuggle the pickup next to he bridge, both the magnet and coil need to be pretty small and a preamp is absolutely necessary to bring individual string levels up to anything reasonable. On top of this, the string doesn't move all that much so the signal is tiny to start with. None of this justifies abondoning it, but it does mean that one has to be realistic in expectations.

2) There are several different purposes for divided pickups. One is certainly to provide individual string sensing for synthesis, such that he circuitry can recognize a pluck on string X as distinct from a pluck on string Y and Z. But another purpose is also to provide better stereo spread, and simply make a normal guitar sound bigger. Both are legitimate purposes, but the second can tolerate greater variation in pickup design.

3) Harry Bissell turned me onto the G-Vox pickup (http://home.epix.net/~joelc/cheapgit.html), which can often be found inexpensively on E-bay. I tried his 95% analog guitar synth (based on an Ibanez Strat-type guitar) that uses such a pickup to drive his modules, and I was very impressed with its performance. Although it is pitched as a MIDI-pickup, it is simply a pickup with 6 little humbuckers. Having looked at the schems for the control modules on a number of MIDI guitars using similar pickups, I can assure you that they are easily adapted to purely analog processing.

[Merlin]

Here is a picture of Ovation bass pickup: http://www.merlinpickups.com/Ovation%20009.jpg

[David Schwab]

Here is a picture of Ovation bass pickup: http://www.merlinpickups.com/Ovation%20009.jpg

I used to like those basses.

[Odd I/O]

Has anyone ever tried making a set of pickups, 1 per string, and preamplifying them seperately to eliminate (drastically reduce) intermodulation distortion? You could get the 2 guitar lead sound or strike complex chords with loads of gain without producing mush, it would seem...

I would've done this a long time ago, but I'm way too lazy

And you'd either look like bootsy but with 6 cables instead of 4 , or have to find / make a fancy freakin cable...

Michael Miller

The new digital Gibson Les Paul has a Hex PU at the Bridge, I think. and Yep, per the previous posts, It might be easier in the long run to shell out the bux for something professionally made than DIY...might even be less expensive too in the long run.
----On cables for such a widget: It seems to me that if the PuP's six coils are in parallel, then couldn't they just have a common that also served as a braided shield for the cable? Perhaps beneficial to twist the inner "hot" wires. Hmmm ---a squirming coil of wires strings and coils-----Quite a can of worms.
----"Big Electric Cat, Big Electric Cat, Big Electric Cat..."
---------Adrian Belew

[David Schwab]

It all started with (Anderton's?) quadrafuzz.

I love the Quadrafuzz! I have a plugin version of it as well.

Anderton made some cool shit. Now with all the digital effects and stuff, no one bothers to make funky analog stuff anymore.

Getting back to the pickup... I'd just get a Roland hex pickup, or something similar. The coils are so damn small, you need really thin wire, and then they are still only big enough to wind low impedance coils (which is fine though).

I've been wanting to get one and make a hex fuzz for a while now.