[sdiy] oscillator jitter / phase noise

[Magnus Danielson]

From: jbv <jbv.silences at club-internet.fr>
Subject: Re: [sdiy] oscillator jitter / phase noise
Date: Wed, 02 Jun 2004 18:31:44 +0200
Message-ID: <40BE00EE.9A03E05E at Club-Internet.fr>

> 
> 
> Scott Gravenhorst a *crit :
> 
> > jbv <jbv.silences at club-internet.fr> wrote:
> > >>
> > >>
> > >> This would then lead to very interesting phasing pattern,
> > >> if two or more oscillators beat against each other.
> > >> Obviously a quartz stabilised oscilator, DCO or software
> > >> oscillator will not have this feature.
> > >
> > >this feature (phase noise/jitter) is very easy to implement
> > >in a DCO or software oscillator, by simply adding the
> > >output value of a random LFO to the phase accumulator...
> > >of course, each oscillator needs its own random LFO...
> >
> > I would wonder, however, whether this would accurately (or
> > accurately enough) simulate the conditions of an analog VCO.
> >  As was pointed out, there are many factors which contribute
> > to a VCO's non perfect nature.
> >
> 
> Yep...
> So why not add another random LFO to control frequency jitter,

"frequency jitter" is nothing but a derivate form of phase modulation jitter,
so if you take a noise generator and integrate it's output you can then add
that to the noise for final phase modulation. Phase-wrapping isn't a problem
in the digital domain as much as it is in the analogue domain (a fact that I
got an unexpected evidence of in the lab a few weeks ago!).

> and (why not) another random LFO to slightly modify the shape
> of the waveform ?

Hint: Frequency shape noise for different power-spectra shapes, adjust the
amounts of each variant and add into a final phase-modulation. Naturally, for
another oscillator you want an independent noise-source.

> What other parameters could be modified ?

What other parameters *need* to be modified? ;O)

Cheers,
Magnus - sharing some noise