Strings as pickups?

[Cernael]

If I understand it all correctly, normal magnetic pickups work thusly:

1. The pickup magnet generates a constant magnetic field.
2. When a string moves relative to this field, a current (proportional to the string's movement) is induced in it.
3. This current generates a temporary magnetic field around the string (also proportional to the string's movement, of course; in addition, the field itself moves along with the string).
4. This second field, and its movement, in turn induces a current in the coil around the magnet; this current is what we amplify and use, under the honorific "The Signal".

Now, I wonder, would it be possible to take the current in the string itself, and use THAT as "The Signal"? As an additional, blendable pickup or something?

Issues I foresee:

1. The pickup magnets along each string should have the same polarity, otherwise different magnets would cancel out each other's effect on the string current. Thus, regular humbuckers are out, as is a matched set of J p'ups; P's or split-coil J's would work fine, and there's a chance quad coils might, too - I'm not sure.

2. You'd have to make a special bridge for this, or have an extra "hot" wire running down the neck; the most feasible solution I can think of, is to have the bridge split in two parts, and have "The Signal" go in one direction in the top two strings (or three, if it's a six-stringer), and the other direction in the rest of the strings. Then, have the strings electrically joined together up at the headstock, and have one half of the bridge as "hot" and the other half as "ground".

3. You probably don't want to ground the strings with your fingers as you play, as that might interfere with "The Signal" - so it would probably be a good idea to prescribe strings coated with some electric insulator or other. Nylon is somewhat available, isn't it?

4. The may be some impedance mismatch involved - I have no idea.

5. String-to-string volume balance might be affected by the different string gauges. Thicker strings will tend to be louder, I think.


That's it. Is it possible, according to the laws of physics? Is it feasible, according to electrical engineering praxis? Will the sonic difference be worth the effort? Other opinions?

[mcgraham]

It's an interesting idea, although the induced eddy currents in the string are incredibly small and highly localised around the position of the magnets; therefore I'm really not sure how much signal (if any!) you'd get at the bridge. Also bear in mind the massive difference in frequency range from lowest frequency to highest, so even if the signals did propagate along the string there'd be some real discrepancy/offset in frequencies once it arrived at the bridge... unless perhaps the magnet was located right against the bridge, but that brings with it its own issues.

These are just my initial thoughts, I'd be interested to hear other's views.

Mark

[Telebass]

You'd never have enough signal. You need the magnet AND all those turns to have enough induction taking place. In addition, there'd be terible problems of frequency response, as stated by mcgraham...

[Cernael]

I'm not sure I understand.

Is it not as I outlined, the magnet inducing a current in the string, which in turn varies the magnetic field, inducing a current in the coil? Even if all the energy from the string, that went into the magnetic field variations, were picked up by the coil, how could it have stronger current than its source? Lower voltage? Something else?


Could the frequency response issue be taken as a non-issue, with the simple attitude of "well, that's just how strings-as-pickups sound"?

Low signal - strong preamp?

Somewhere I thought that this might be a good way - if it is good at all, that is - to electrify a harpsichord, or somesuch. To do it like a guitar/bass, you'd need a very long coil, or just many coils. With this way, one single (homogenous? Ish?) magnetic field. You might even have two rows of magnets on either side of the strings, and tilt the rows in relation to each other to vary the strength of the field across the strings, thus varying the signal level somewhat...

[GreeneKing]

Firstly I'm no expert.

Surely the movement of the string in the magnetic field induces a current in the coil of the pickup. The string/pickup is a generator. The value of this voltage signal being in some sort of proportional relationship to the mumber of turns in the coil.

So the pickup itself is a voltage anplifier? (current being reduced in a similar proportion given that the energy loss of the string is finite).

Ergo without the coil there's no useable signal.

So what you are suggesting is therefore a problem that the pickup was created for solving!



Peter

Edited November 15, 2008 by GreeneKing

[Cernael]

Transformer rather than amplifier...and probably not a very efficient one, given that the magnetic medium is air (well, the magnet counts too, I guess, but there's still a lot of air) rather than iron/steel/whatever...

...isn't it? Take a normal transformer, two coils around a common, circuitous core; change half the core to air; have the coil around the air part be just a single loop, and call it "string". If we for a second ignore the inefficiency issue, we have a transformer that scales up voltage by a factor equal to the number of turns of the coil; at least 10-20 thousand, IIRC. Current is scaled down by the same factor.

Unless, of course, it doesn't act as a transformer.

Edited November 15, 2008 by Cernael

[GreeneKing]

I think you are right, and spelling the very doom of your proposal?

Peter

[Count Bassy]

Moving a wire in a magnetic field will induce a voltage in the wire. Similarly having a static wire in a changing magnetic will induce a volatge. The later will is how a classic pickup works, as moving a magnetic material in a magnetic field will disturb the magnetic field, and this disturbance induces the a voltage in the pick up winding.

What you are suggesting is using the first effect, and a voltage would be induced along the length of the wire when the string vibrates. However the voltage would be miniscule as the pickup wire (the string) only passes through the field once, and is not closely coupled to it.

To pick this up you would need to tap the signal at both end of the string. I would have thought that the frequency response would be good. After all the inductance and resistance of the pickup would be very low which should give it a very even response and very low losses before it got to the (extremely high gain) pre-amp.

BUT! the string will also act as a wonderful ariel for all sorts of interference, and the chances are this would totally overwelm any of the signal you wanted, which is probaly insurmountable in any practical sense.

PS this may be a load of bolleaux

Edited December 8, 2008 by Clive Thorne

[Cernael]

[quote name='Clive Thorne' post='330404' date='Nov 17 2008, 03:20 AM']To pick this up you would need to tap the signal at both end of the string. I would have thought that the frequency response would be good. After all the inductance and resistance of the pickup would be very low which should give it a very even response and very low losses before it got to the (extremely high gain) pre-amp.

BUT! the string will also act as a wonderful ariel for all sorts of interference, and the chances are this would totally overwelm any of the signal you wanted, which is probaly insurmountable in any practical sense.[/quote]
Well, of course you need to tap it at both ends, in a sense. Else you'd just have a wierd disembodied capacitor thingy (with excellent antennae properties, as you say) going on.

My thought was that if you connect the strings to each other up at the headstock, and use individual string saddles (that are isolated from each other), you could tap the strings at the bridge, have the signal from both ends of each string (through other strings), and you don't need a separate wire from the headstock. That would probably work on the antennae issue a bit, too - as the signal goes in both directions, and picks up interference both ways, don't you get something reminiscent of the humbucker effect? I guess you'd have to be careful when selecting how to connect the strings, in order to balance out the hum, though. E and G strings on one (electrical) end, and A and D on the other, would be a good way to go, perhaps.

Edited January 2, 2009 by Cernael

[The Funk]

I know I'm not really a Nobel prize-winning physicist or anything but doesn't electricity in the strings = fried bass player?

[Count Bassy]

Well, I did say it meight be a laod of bollocks, but I actually meant that above the comment, not that below it. The latter being a result of not deleting the left over bits of crap at the end!!

I like the humbucking idea, but what would you do with the left over string on a fiver?

[Cernael]

[quote name='Clive Thorne' post='348284' date='Dec 8 2008, 02:18 AM']Well, I did say it meight be a laod of bollocks, but I actually meant that above the comment, not that below it. The latter being a result of not deleting the left over bits of crap at the end!!

I like the humbucking idea, but what would you do with the left over string on a fiver?[/quote]
Yeah, that bit kinda bothers me, too. Actually, the different gauges of the strings do, too; I don't know in what direction that'll affect it all. With some luck, B+E=A+D+G, or close enough.

As for the finger fry issue, I think you'd need coated (=insulated) strings anyway, as grounding the strings to your body (as you do when you play - that's what happens when you've got that kind of hum that stops when you touch the strings) would disrupt the electrical path greatly. Then you'd just strip the insulation away at the bridge and nut, so you get a clear path there.

And further, no, you only get fried when current from the mains outlet creeps past your amp - which should, I think, contain a few safety valves to prevent that from happening. What we're talking about here is a very weak current, that is created in the strings through magnetic induction no matter if you tap it or not.

[octavedoctor]

It wouldn't work and there would be no point in doing it.

The strings impedance would be so low (effectively shorting the input of the amplifier) and the voltage generated so small that it would be impossible to amplify it effectively.

The existing arrangement of search coil + magnetic core interacting with the mass of the string is far more efficient and that's why it works and why we use it in preference to the idea you propose.

[Cernael]

[quote name='octavedoctor' post='364859' date='Dec 28 2008, 03:50 PM']It wouldn't work and there would be no point in doing it.

The strings impedance would be so low (effectively shorting the input of the amplifier) and the voltage generated so small that it would be impossible to amplify it effectively.

The existing arrangement of search coil + magnetic core interacting with the mass of the string is far more efficient and that's why it works and why we use it in preference to the idea you propose.[/quote]
Just saw this thread again.

Actually, you don't give me enough info here to convince me that you're right - which means there actually IS a reason to try it, namely to see if it works or not - to see if you're right, or just a naysayer.

I can come up with a number of different reasons why it would be impractical - and I have, and have shared those above - but none strong enough to say, "It's pointless".

As for your second sentence, as I've come to understand the impedance issue (which, admittedly, I don't do very well), the biggest problems comes along when you have a high output impedance (like piezos), feeding into a too low input impedance (in the amp). This, if I understand it, is the opposite situation, which shouldn't be that much of a problem. And anyway, can't you change the impedance and/or voltage to acceptable levels with the right buffer/preamp anyway?

[tauzero]

[quote name='Cernael' post='368940' date='Jan 2 2009, 11:02 PM']As for your second sentence, as I've come to understand the impedance issue (which, admittedly, I don't do very well), the biggest problems comes along when you have a high output impedance (like piezos), feeding into a too low input impedance (in the amp). This, if I understand it, is the opposite situation, which shouldn't be that much of a problem. And anyway, can't you change the impedance and/or voltage to acceptable levels with the right buffer/preamp anyway?[/quote]
The resistance of a bass string is tiny. I'm not sure what the impedance implications are - I think you'd need a very high impedance coupled with huge gain and the sort of signal-to-noise levels that you would be very unlikely to find at an economic price. The high impedance means that the input would be susceptible to induced noise, which would be induced whether the strings were coated or not (if they were coated, it would be a capacitive coupling with the coating acting as the dielectric).

[Cernael]

Okay, thanks; it's beginning to make sense.

Bad signal-to-noise ratio is an objection I can understand, can't really get around, etc.

[Alien]

I've seen it done somewhere, but I'm buggered if I can recall where. It was done on a single-string instrument, and the string ran back down the instrument (inside perhaps?) so it could be connected to both poles of the jack. I'll try to find a link this evening.

Andy

[Beer of the Bass]

I remember reading a discussion of this years ago on the musical instrument makers forum (MIMF). [url="http://www.mimf.com/archives/make_pickups.htm"]Here[/url]'s a link to the archived discussion. Apparently it's been tried, and can work...

[Cernael]

O_o

Niftyness!

Thanks for the link!