Gallery, Projects and General > Project Logs
Extending the soundscape of a hexaphonic guitar pickup
sorveltaja:
Today was a 'wire hassle' -day. I'm glad that it's almost done:
I tested the circuitry shortly with variable psu, as usual, slowly increasing voltage, and the current limit, while observing, to make sure, that if something goes wrong, there is less chance to damage the components.
So far, no smoke or sparks. Almost there, but still some stuff to do. To be honest, I'm looking forward to get a break from building, and do aural testings instead.
sorveltaja:
Not much progress today, except making first optoisolator for testing. Components used:
The leds are white, very bright(actually painfully bright, once they get some 'juice') ones. Ldr's are 'hermetically' sealed ones. There is no specific reason, why I chose to use those exact components.
They are just a part of the optical component stock, that I gathered back in the early phase of the 'optical hex pickup' -project.
Printed enclosure for the led and ldr:
They are secured in place with few drops of super glue. Then some heat shrink tube to keep the ambient light out:
Both ends need to be sealed also, using just about anything, that is thick enough, and opaque. As I have some black acrylic paint, that's what it'll be.
What comes to testing that optoisolator with the rest of the circuitry, I think the led part of it should be quite straightforward. There is an adjustable LFO(low frequency oscillator (running at ~0,2 - ~8Hz)) that drives it, and the involved series resistor needs to be adjusted to make the led to pulse so, that it fades completely, when the lfo wave hits the 'bottom'.
At this point, when thinking about it, when the lfo wave hits its highest spot, the leds, that I'm using, should glow very brightly(remember, this circuit was originally published ~40 years ago!), which shouldn't be a problem, unless that led starts to draw excess current. But we'll see.
The ldr -part(having a parallel resistor) is somewhat more difficult to imagine(at least for me), but in the end, it should be about adjusting its response, to fit the range of led's brightness variation.
sorveltaja:
It seems, that the resistors for led and ldr might not need to be modified that much, if at all. For my surprise, the led already pulses in a way it should do.
So I listened to the output, and there was an audible pulsing effect. It was rather lacking, as I tested the circuitry with only one(of four) optoisolator, and sine/rectangular waves. The lfo and optoisolator -part of the schematic:
As can be seen, there is already a limiting resistor(R6) at the output of the lfo(IC3A), that drives the leds. There is also an option to use an external source for modulating the leds(feeding it to pad P, while pad N is disconnected).
Now rest of the optoisolators are finished, and soldered to the board. Only few resistors(for leds and ldr's) needs to be added, and the circuitry should be completed.
If all goes well, then it's finally time for (lengthy)audio testings.
sorveltaja:
After playing with different settings for a good few hours, the phase shifter seems to be working now, as it should. A very primitive audio clip, where 50Hz rectangular wave was used as an input source:
phase_shifter_testi_1.mp3.zip
There are two things, that needs to be modified(not required, but for my personal taste), though. One is to increase the value of the capacitor(C9), to lower the speed range of the Lfo(IC3A, which drives/pulses the leds).
Another one is to alter the range of the 'offset' -control. I'm not sure, how to explain it clearly, but here it goes anyways: the Lfo wave, that modulates leds, has a very narrow 'effective area', where the phasing effect is audible, while rest of it is 'dead zone', where nothing happens. So that very 'effective area' is passed way too quickly.
What I'm after, is to make whole Lfo range 'effective'.
This waveform(internal, part of the circuitry), that drives leds, is triangular. I tested using the external source(function generator) with triangle, and also sine wave, but still similar results.
Then came the idea of testing the leds with plain dc, instead of pulsing waveforms. That way it was a lot easier to 'browse' through the Lfo range. As a result, the effective range is indeed very narrow, within 0,5 volts(of several volts range).
And of course, what a bummer, as I forgot to write down resulting numbers, while testing.
sorveltaja:
Today I simulated the Lfo part of the schematic, with different values for R27, R7, and R6:
I managed to get quite good results, but when I did the same modification in practice, they didn't work at all, as expected.
My aim was to bring the triangle wave, that IC3A(which modulates the optoisolators leds) makes, between 2,40 - 2,80V, because that's where the effective range seems to be.
I tested with function generator's triangle wave, having 0,40V with 2,40V offset, and that's where it sounded the best, without gaping dead zones.
When doing the mod to the circuitry, it messed the Lfo speed settings also. It's not worth to restore the original resistors, as with them, it was already too quirky. I think it's time to ditch the included Lfo section, and use something, more usable instead.
After searching alternatives from the net, there is something, but not quite what am looking for: A simple, variable speed triangle Lfo, with adjustable dc bias.
On the other hand, I already have built that step waveform generator(the one with slide pots), which shouldn't need bias adjustment, just some filtering to make a decent triangle waves, at needed 'biased' voltage level. Might well be worth testing.
If it works for modulating the leds, then build another, simplified version(without 'divide-by-8' -section), that uses multiturn trimpots for finer step level adjustments.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version