[sdiy] VCO - sine output - why bother?

[rsdio at audiobanshee.com]

The problem with the usual chain is that you have to disable all of the VCO waveform outputs if you want to use the VCF as a sine generator. Any input to the VCF will affect the resonant frequency, making tracking problematic. Sine from a VCF is great, but I'd prefer to be able to mix other waveforms as well.

In other words, not only is a self-resonating VCF a more expensive circuit than a pure sine oscillator, but you don't want to lose the VCF on other waveforms just to generate sine.

Once you have a decent VCF and sine oscillator that can work independently, it only makes sense to mix the sine in after the VCF. Pure sine has no harmonics, so VCF on sine is equivalent to simple volume control.

There have been many times where I painstakingly adjusted resonance and keyboard tracking on a VCF to get a pure sine wave that was playable, only to have the tracking ruined when I mixed in other waveforms or even noise. Try it out!

Brian


On Aug 26, 2016, at 2:38 AM, Steve <sleepy_dog at gmx.de> wrote:
> well not a dedicated VCF, but one that you already have in the usual chain. That's why it looked to me like a lot of effort for dubious gain - but since it is so common, there had to be a good reason, which is why this thread exists. :)
> Probably I'm thinking more in terms of voicecard, while the other reply I already got on this thread and yours are more in the realm of "modular and free" ;-)
> (or / and I just haven't seen enough architecture differences in typical fixed-topology synths, I've only seen a few, all rather similar, and assumed "this is how it's done")
>  
> Then again, when I think in the box of "subtractive synthesis", in a strict sense anyway, the suggestions of using a single sine wave to add a fundamental after a VCF feels kinda odd :-D
> I guess users of more complex (routing wise) synths have more tricks like that up their sleeves, which kinda forms this idea in my mind that next to my imagined designs to build there needs to be more room for MUX ICs on the PCB...
>  
> Steve
>  
> Von: "Roman Sowa" <modular at go2.pl>
> if you're going that route then why do you think VCF is simpler than
> sine waveshaper, even if it uses OTA? VCF will usually take 4 OTAS.
> 
> Roman
> 
> W dniu 2016-08-26 o 10:30, Steve pisze:
> > Now I'm not a super synth expert, also not in sound design with such.
> > What I've always wondered, if I look at VCO circuits, why people go to
> > the lenghts that they do wave-shaping a more or less pure sine as a
> > waveform output, my heart ached when I saw a circuit that used an OTA +
> > yet other stuff just for that.
> > What more does a dedicated sine output provide that one can't get with
> > using triangle -> lowpass VCF with keyboard tracking on? So maybe it
> > doesn't shave off all harmonics totally, but is the sound difference
> > really that big?
> >
> > Steve
>