A slightly left-of-center idea for DIY optoisolators

[Mark Hammer]

So I'm sifting through a parts drawer of optical devices, wondering if I had put any homemade LED/photocell units in that drawer and forgotten about them (I do that sort of thing a lot  ).  One of the things I find in there a some 10-segment bargraph LEDs intended to accompany LM3914 and 3915 chips to indicate levels of things.

And it dawns on me: one could arrange a group of photocells to be maximally illuminated by the LED sections at different levels, such that LDR1 would be shone on by any level, LDR2, only when the level rises more, and LDR3 only when the level is much higher.

Now all I have to do is figure out an application and a way to keep them light-tight, and I'm in business!

[pinkjimiphoton]

that is cool... i was tinkering with trying to do a "color organ" that uses photoresistors to effect different frequencies.
can't wait to see what ya come up with!! a whole plethora of effect control comes to mind. well done mark!

[Seljer]

iirc those LM3914 chips have a 'dot' mode and a 'bar' mode. In dot mode if you fed an envelope to the LED array you could'd get a short pulse of light on both the rise and the fall of the signal. May be interesting for some more dynamic vocal filter sounds?

[Mark Hammer]

Yeah, a sort of quasi-addressable LDR.

[noisette]

Do you know this one?
http://www.musicsynthesizer.com/Corpse/VCEOM.html

Kind of related to your idea...

[Mark Hammer]

Very much so.  I guess the only difference between that and what I'm musing about is that the circuit you link to uses each 3914 output in a more discrete fashion.  I was thinking more in terms of a kind of "moving LED" that the 3914 would shift around among LDRs.

[noisette]

Well, I think I understand YOUR idea, but I don´t get what the dither oscillator is doing in that linked circuit... 
But the whole concept is intriguing, you might also look at this(that is if you don´t have already):

http://www.till.com/articles/scanner/
http://www.jhaible.com/tonline_stuff/jh_ipscan.html

Technically different, but conceptually similar...