Gallery, Projects and General > Project Logs
Extending the soundscape of a hexaphonic guitar pickup
sorveltaja:
So this is going to be an attempt to make sense, of what kind of simple circuits could be used with analog hexaphonic guitar pickup.
I felt that it isn't fair to clog the optical pickup thread with this subject, so here we are.
But to the point. To get started, I'm aiming for outputs, that have more harmonics in them, than the plain square waves, that the pickups produce(there are clips on the optical pickup -thread).
Here is the first circuit, that I have already simulated, and tested on breadboard:
That waveform isn't in sawtooth form, that I was after earlier, as it was used on many analog synthesisers. It sounds like this(saved from the simulator, but the breadboarded circuit sounds practically same):
4040_testi.mp3.zip
It has one drawback in it, though. The fundamental frequency of the output is input divided by 8(In this case the input frequency was 1000Hz, and output was 125Hz).
There was only one thing, that I could think of, to multiply the input by 8, to get the input and output frequencies to match - 4046 pll(phase locked loop).
As the form of the input is square wave, it wasn't problem. That 4046(standard version) requires input to be close to the supply voltage, which was 9 volts.
The amplifying circuits of the optical pickups provide enough output, so that the 4046 is able to track the input signal.
I'm not too good on drawing schematics, but here is the circuit, as it was breadboarded(10k resistors on the right present resistor network):
In practice, it seems to work.
Then, after some searching on the net about "staircase generator", I found something like this, kind of "sampler" that is based on 4051(8-channel multiplexer):
It sounds like this:
4051_testi.mp3.zip
For some reason, it also has fundamental output frequency, that is divided by 8. The 10k resistor network might not be necessary, but the point is, that the resistors on the 4051's outputs could be replaced with trimmer pots, to adjust the combined output to make arbitrary waveforms.
I haven't breadboaded that yet, but what the heck, there is only one way to find out.
sorveltaja:
The 4051-based idea of arbitrary waveform 'generator' is now tested on the breadboard, and it works. The circuit:
And same on the breadboard:
The stepped output waveform is easily modified with trimpots. As there are 8 outputs, there are also 8 adjustable steps in the wave. Triangle-ish one:
And same after filtering:
One thing, that I noticed, when looking the combined waveforms on the scope, is that it couldn't track them too well. All kinds of wiggling happened. Maybe the produced waveforms are more complex, than they look.
But on the other hand, I think I'm going to build(at least) two or three of those 4051-based contraptions. Perhaps some with fixed waveforms. For the adjustable ones, some kind of futuristic, retro indicators would be nice to have, to visualise the form of the wave without the need of using scope. Something like 8x8 Led matrix comes to mind. But we'll see.
As mentioned before, that circuit also divides the input frequency by 8, and therefore needs 4046 and 4017 before it, to multiply the input frequency first by 8.
sorveltaja:
So now I'm hooked into testing/making a visual waveform indicator for that 4051-based generator. As usual, the simpler it is, the better it is.
One possible candidate is something like this:
I already have the two 3914's(actually from an ancient led-oscilloscope project) and 4017's on my shelf. Plenty of leds also(from the same scope project).
I tried to simulate the above circuit, but the results were very quirky and useless. That's why I'm not a big fan of simulations. Certain rules must be obeyed, or the result is garbage.
So not much of 'diy circuit bending' allowed.
Usually at that point I move straight on to breadboarding, but this time both boards are occupied by other circuits, that are still under testing. Jolly good.
But no matter what, I'm going to test that circuit above, to see if it's clumsy or useful in practice.
In fact, even if there are several of those 4051-based devices, only one "plugin"- 'display unit' is actually required. After all, they need visual indicator only, when adjusting the waveform. Multiple, permanent displays would be a lot nicer, though.
In the end, a spin-off of the recent brain hickup was to make a steampunk-like, victorian style, mechanical indicators for each of the trimpots, that are used to adjust the waveform.
All the gears and stuff.. some of them could be rather easily printed out... and the end result would be - bulky.
And other than that, I had an idea of using the pickup's outputs to control some kind of mechanical devices. Sort of like Mellotron(https://en.wikipedia.org/wiki/Mellotron), that has a tape loop for each keys of the keyboard. Very heavy yes, and the tape technology is getting rare these days.
Instead, for more precise(like industrial) response, stepper motors could be used to rattle whatever objects, or what the heck, even plain cheapo small electric motors for more "organic" sound.
sorveltaja:
Led matrix isn't easy to breadboard, so I decided to build one, for 10x8 leds. Last two rows will be added tomorrow, and rest of the wiring:
So far, all the rows seem to work, as expected. Needless to say, that it's rather tedious to build things like that. Local suppliers have a very limited selections of the led displays, so one should see what's available on Ebay. But for now, I'll skip that option.
On the other hand, what could be better, than recycling components, that have gathered dust for years on the shelf?
What comes to that 10x8 led matrix, if the idea, that I'm after, works, it could be used as a single display for something, like three 4051-based generators. Not really multitasking, but for one generator at a time, when checking/adjusting the waveform.
While scavenging leds from the old led-oscilloscope project(it had 16x10 leds), I got an idea, that the 8-step waveform generator could be extended to have 16 steps for finer adjustment. Probably not worth doubling the parts count, as it could still sound the same as 8-step one does.
After all, using simple filter(s) to alter the stepped waveforms could be easier to deal with.
sorveltaja:
So the led matrix is built, and the circuit is breadboarded:
The led "display" shows different waveforms, but not exactly in a way, that I was after.
As on the oscilloscope, there should be a trigger function, which keeps the waveform from wandering back and forth. It is possible to do, by adjusting the 4017's frequency, but the "triggering" depends also on 3914's input signals frequency.
One might wonder, why I'm so eager to add a visual indicator for that 4051(8-channel multiplexer)-based waveform generator. It's simply because I just love all the bells and whistles it could be a lot easier to remember, what kind of set of adjustments sounds good or bad.
Like a connection between aural and visual 'image'. For example, a guitarist, that uses one or more effect pedals, has an image in his mind, of what his favourite sound 'looks' on those pedals in a form of settings.
Currently, the 4051-based circuit is on the breadboard, and has 10k multiturn trimpot for each channel. Dirt simple option could be to use slide potentiometers instead, for a good visual reference.
Although the slide pots tend to be quite pricey(at least the decent ones), I consider that as a plan b, if that led display-based idea fails.
Navigation
[0] Message Index
[#] Next page
Go to full version