Frequency controlled oscillator?

[yano]

Anyone know of schematics for a frequency controlled oscillator?

It seems to me that a guitar controlled synth would be relativly easy to build if you could simply generate a signal at the same frequency as the incoming "control" signal. I'd be interested in anyone's ideas.

[puretube]

AGC...

(followed by any waveshaping circuirt...)


monophonic, though

[Paul Marossy]

I tried designing such a circuit, which I guess I should revisit sometime. It's not as easy as it sounds. Seems to me that the most direct approach is to use a frequency to voltage converter chip and build upon that. However, those chips do not seem to track guitar frequencies very well, especially since they are sporadic in nature and in strength. These chips are designed more for continuous input, like monitoring the speed of a motor or something.

Maybe there is an easier way to accomplish this with a guitar, but I don't know what it is. Mark Hammer has a page at his website somewhere that has a cool little project on it, but it's not exactly simple, IIRC.

[lovekraft0]

A CD4046 phase-locked loop will track the frequency of an input signal - seems like somebody used one in a guitar synth a long time ago, but I can't remember where. Again, strictly monphonic.

[Ge_Whiz]

John Hollis's 'Crash Sync' does a reasonable job with simple parts.

[punkaled]

Doesnt the "shoctave" circuit (posted on this board a while back) use something similar to this to create the synth "organ" which plays along with the guitar at one octave down?
iirc it uses three transistors to do it, one to sense the frequency and two in the oscillator.

[ExpAnonColin]

You need to use a phase-locked loop, IE the 4049.  There are others out there.  There's a datasheet on my page as well as a schem for a DF-2, which operates on the same principal.

There are many things you're going to want to consider-for example, you need to filter out other frequencies than the fundamental.  Otherwise the note will be jumping all over the place or you might not be getting the right note... and you have to worry about the output waveform as well.

-Colin

[Maneco]

Also,don´t forget the guitar synth project of the japanese book posted some time ago...it's based on the f to v converter concept,and has an interesting filter...

[toneman]

The EML PolyBox used a PLL to lock to the fundamental.
then generated a *chord* in the key U played in.
It was radical for it's time.
But, the PolyBox was made for 9MONOPHONIC)synth input, not guitar.
U would need a good (tracking)filter to remove 2nd harmonic
otherwise,  fundamental extraction would B glitchy.
afn
tone

[StephenGiles]

Try this on Mark Hammer's site:
mhammer.ampage.org

This has a very efficient set of tracking oscillators which is a million miles from Uncle Penfold's 4046 offering.

[StephenGiles]

Sorry, I buggered up the previous post, look here:
http://ampage.org/hammer/files/EH-Guitar-Synth.ZIP
You will not be disappointed - there is all sorts of other good stuff in the circuits which people seem to choose to ignore in favour of ....fuzz boxes - my voice goes up at the end of that sentence!!!!!!!!

[yano]

How about a MIDI controller? If the frequency detection could be done maybe with a microprocessor? And then output to MIDI, that could be cool.

[Transmogrifox]

If you have the knowledge and the means to use a microprocessor or DSP chip, don't fool around with analog on this thing!

You could implement the FFT in real time (you may have to fool around with windowing periods) and then write a program to generate waveforms at some selected peaks or even pick out the harmonics...do a sample/hold effect on the harmonics.  Now I'm going crazy, but digital is a good way to go on this effect if you are able to program a DSP.

I worked a little with the ADSP2191 digital processor from Analog Devices.  You can get the EZ-KIT evaluation board and learn how to program it in a few months.  It's quite a steep learning curve, even for someone with some programming experience, however it will interface with C (and C++ I think)...and maybe Java.  The nice thing about the EZ-KIT is that it has audio input jacks and a very high-end A/D converter, plenty of processing speed and a decent amount of RAM and EEPROM to handle a program that does an FFT and generates tones.

[puretube]

still monophonic, and depending on what the axe ( and its artist) has to offer

[StephenGiles]

Thought I'd re-introduce this to the proceedings because it was largely ignored, and it took me a long time to trace out the circuits! Please inspect the Adaptive Filter for starters which turns any input into a sinewave. I've seen a good demo of this on a scope back in EH UK. Next look at the tracking oscillators - they track very well - thanks to the 2 4047s.

[puretube]

absolutely right!
but to respond to Yano: you gotta somehow "keep the incoming signal going" by a good sustainor-circuit (AGC/compressor);
(like in the MS, e.g. ...)

[StephenGiles]

I think the compressor in the EH guitar synth did just that.

[puretube]

err, that`s exactly what I meant to express  

[puretube]

supposed to be a fast tracking filter:
http://www.pat2pdf.org/patents/pat4032857.pdf

[StephenGiles]

Nice to see that word "circuit" rather than the USAian "schematic"!!!!!
Stephen