The optical guitar pickup.

[amptramp]

It sounds like the ideal receptor would be a photodiode or phototransistor, not an LDR, just for the sake of response time.  There are high-speed photoconductive components, but the standard LDR is not one of them.

[Taylor]

Are there any photodiodes or -transistors where the optical element is exposed? The ones I've seen are always sealed PDIP packages.

[Taylor]

> How does a varying resistance create an audio signal by itself?

moeburn later said "my amp has a bias in it". If an input grid, base, capacitor is leaking DC, a photoresistor will modulate the DC into a voltage.

I too have doubts about photoresistor speed; but we can get many volts out of an LDR in a fraction of a second, so we can get a teensy part of a volt at audio. And moeburn did say he needed heaps of gain.

So for this to work with any amp, I guess you'd need to connect it to something like a transistor preamp with no input cap. Varying the resistance between transistor base and ground does seem to do the trick in simulation.

[artifus]

http://www.creative-science.org.uk/lightbeam.html

[Taylor]

http://www.creative-science.org.uk/lightbeam.html

Crazy! Cool link.

It occurred to me that if you mounted such a "pickup" towards the neck, the light on the sensor would be blocked twice per string oscillation - once going one way, once going the other. So you get an octave up effect. 

[snakey]

http://www.youtube.com/watch?v=RheshR-9Q0w

[Jule553648]

I saw this thing:

http://articles.ircam.fr/textes/Leroy06a/index.pdf

They used reflective sensors - phototransistor and LED in same case and they measured reflections. This could be used more instead of a pickup. But they made things complicated by modulating the whole thing...

[petemoore]

  Apparently piezo assistance was chosen to accomodate and output the HF pickuping jobs.
  Either that is the demonstrationists preferred tone or, and, there didn't seem to be any great variance when the knobs were reset.
  2-D imaginative description of what my preconception of what photo-PU behaviour might be/when:
  Seems that any slowness in creation of the electron-analog following the string would be most obvious as the string is moving through the center portion of it's excursion when speed its the highest, or when otherwise nearly still except the uber-harmonic shudderings which continue the  as the string 'stops' to changing directions.
  ...to attempt to describe Pho-PU's ability to follow the 3-D motion of an oscillating string.
  It is expected that it will sound like itself, considering the sensors placement at almost the end of the string.
  To spread these photo sensing points down the string [like >1 mag pickup basses do] requires giving up some favorite places to access the string for picking, or mirrors on the string, reflecting the LED photons back into the light sensor, or LED's in the strings. Perhaps a tiny portion of string over the photo elements, 12-sided chrome polished zone with very strong lights on it, lots of output preamp...might put some cool dynamic-related-misread-anomolies in the output.
  Sure looks like a tidy approach to DIYing 1 pickup per string tech.
 
 
 

[PRR]

> I guess you'd need to connect it to something like a transistor preamp with no input cap.

No; just a resistor to a battery.

Hold the LDR in your ambient light, measure resistance, use a similar or somewhat larger fixed resistor. Battery may be 50V-100V before the LDR arcs/sparks, perhaps lower if you work in bright light and run into power dissipation limits. Lower battery makes less signal but is of course safer. 9V should be ample. 1.5V may be enough.

Since you will surely need an amplifier anyway, you can try to combine them; but it's hard enough to bias a transistor without also having a key bias element change with the light. We normally keep the light-sensor circuit simple, then cap-couple to a standard amplifier.

Photo transistors are much faster, often cheaper, readily available.

Photo-diodes are much-much faster (far faster than audio needs) but make less output.

An LED makes a workable photo-diode. Some glass-case signal diodes are moderately light sensitive.

[artifus]

http://en.wikipedia.org/wiki/Demodulation