V.C.O. - Whadda ya know?

[flurng]

Howdy! - flurng here. I'm just looking for some thoughts, opinions or insights anyone may have regarding the subject of V.C.O.'s. (For those who may be unfamiliar with the term, the V.C.O. is a Voltage - Controlled Oscillator, or an oscillator whose operating frequency is determined by a given input voltage.) To put my question into context, I am currently working on the design of an analog music synthesizer, a device which almost invariably employs one or more V.C.O.'s as it's signal generators. I have successfully assembled a handful of audio oscillators, mostly derivatives of the Armstrong or Hartley. My problem is determining which technique to use for frequency selection; varactor diodes, variable source voltage, a transistor-capacitor filter arrangement, or a full-blown V.C.F. ( Voltage Controlled Filter ). I know that varactors are used quite often for RF oscillators, but I'm not convinced they would provide a sufficient range of audio frequencies for my needs. Ideally, I would like to create a circuit which can produce any note over at least 5 octaves, preferably 8. I've heard others have tried substituting a transistor for the resistor(s) in an r/c filter circuit, with some success, although this scheme seems to suffer from frequency instability. Any thoughts, or comments? I'd love to hear 'em! Thanks in advance!
flurng

 

[Nigel Goodwin]

Varicap diodes are completely unsuitable.

I would suggest you try looking at how VCO's etc. were done back in the analogue days, don't forget a log response is essential.

I built my younger brother one from a kit back in the 70's, it was this one:

Matrixsynth: Powertran TRANSCENDENT 2000

I was recovering from glandular fever at the time, and built it once I was able to move out of bed (I lost three and a half stone in ten days!).

 

[RadioRon]

I'm coming to this question from the RF direction, and I agree that varactor diodes used in LC oscillators would not be appropriate at all. My gut reaction to the question is first of all, why limit yourself to analog when it would be so easy using digital/DSP and DDS techniques? Anyway, assuming analog techniques are necessary, perhaps have a look at using a transconductance amplifier as a variable resistor, or perhaps a MOSFET but include a feedback system that phase locks or frequency locks the oscillator as you would any synthesizer.
This note looks useful:
https://www.calvin.edu/~pribeiro/courses/engr332/Handouts/oscillators.pdf

If expediency is an issue, have a look at a ready-to-go oscillator like the MAX038.

I mention transconductance amps only because I recall seeing a voltage controlled sine oscillator many years ago that used such an amp. I don't have a reference for this. I note that National's LM13700 data sheet shows a number of interesting applications. Have a look at Figure 17 in this link:
https://www.electro-tech-online.com/custompdfs/2009/02/LM13700.pdf

 

[audioguru]

The MAX038 is not made anymore for a few years.

 

[RadioRon]

I guess that means the old 8038 from Intersil or Exar are long gone too.

 

[audioguru]

The Intersil ICL8038 is not made anymore but some people posted info that Exar still makes the XR2206 function generator IC.

 

[gerty]

Here's the xr2206: ICs: Linear
you have to scroll halfway down the page, ordered/recieved a couple of them a month ago.

 

[flurng]

To all who have responded: Thank you kindly for the wonderful insights and/or food for thought regarding my "sonic quest".
To Mr. Nigel Goodwin: I, too, was skeptical regarding the use of a Varicap diode. As I mentioned in my post, I need a much wider range than I believe is possible with a varicap. Thanks for the heads-up about the Transcendent 2000 - I will certainly research it further. Also, sorry to hear about your illness, but I'm glad you are back in peak form again!
To RadioRon: I'm sure you're right about the ease of using digital techniques, and I suppose they would provide excellent results, however, there's just a certain something, that analog synths possess, a certain fat, funky quality that digital units can't quite seem to reproduce faithfully. Thanks for the tip regarding transconductance amplifiers - I'm not familiar with those particular devices, but I'll be certain to Google them & find out more about them next chance I get. Finally, you suggest using pre-existing components such as the MAX038, and I suppose that would save quite a bit of time and effort, but I'd really like to create something from the "ground up", partly to improve my knowledge of such circuits, and partly because I'm hoping my blind ignorance might actually help me to "stumble across" some interesting or novel new design / sound. I'd really like to come up with something completely unique and different from what's commercially available to date.
Thanks, once again to everyone for their input! You've certainly given me much to think about, and I'll be sure to keep everyone posted as to my progress! A good day to all!

 

[Nigel Goodwin]

To Mr. Nigel Goodwin: I, too, was skeptical regarding the use of a Varicap diode. As I mentioned in my post, I need a much wider range than I believe is possible with a varicap. Thanks for the heads-up about the Transcendent 2000 - I will certainly research it further. Also, sorry to hear about your illness, but I'm glad you are back in peak form again!

It was 30 years ago

I believe my brother still has his 2000, and still uses it - I certainly don't have the assembly instructions any more, the details were all in ETI magazine - but I might have disposed of the magazines by now

I'll have a look and see if I can find them.

I also know the guy partly responsible for the original Moog synthesisers, Gene Zumchak - if you google his name you will find a fair amount of useful material.

Heres a site that looks pretty useful: