[sdiy] NEWBEES are alive and kickin' !

[harry]

Hi bees (inline)

Peter van Hamersveld wrote:

> Yes,
>
> No, we did not give up.
>

Thats GOOD !

>
> After some thinking, and considering, we decided to start REALLY simple.
> So we (Peter) soldered a very simple square generating VCO, consisting of:
> 2 capacitors
> 2 NPN transistors
> 4 resistors
> 1 trimpot parallel to the source to fake a Control Voltage from a keyboard.
>
> it worked :)
> though not being a standard 1 V / Octave VCO, but having a linear proportion between CV and frequency.
>
> We noticed VCO's can be arse-kickin simple. A battery, a resistor, a capacitor and a 'four-layer-diode' (sorry i do not know the proper english word, but it's such a diode that starts conducting at a certain (positive) voltage (NOT being almost 0)) can already make a SAW Oscilator.

Yep... its still four layer diode.   If it has three legs... possibly a PUT or UJT but
the method is the same.

>
>
> So, questions:
>
> 1) Why are most VCO's (e.g. in tom's cookbook) having a quite complicated structure. Just to make it 1 V/Oct? (than that must be hard.)

Yes... that is hard. The trick is even tiny variations of temperature of the transistor
junction will throw it out of tune !!!

>
>
> 2) A bit the same question: Are there any circuits converting a CV linear with frequency to a CV linear to note (1V/oct)? Or do 1V/Oct need to generate the waveshapes in a completely different way in order to be so?

Yes there are such circuits. You should check the archives of the list, as well as
member sites. Also check National Semiconductor AN-311
Theory and Applications of Logarithmic Amplifiers... for a discussion and some
circuits. Most are a variation on this theme.

H^) harry

>
>
> thank you!
>
> Your blockrocking' Bee,
>
> -- Peter van Hamersveld, plons at gmx.co.uk on 06-03-2001