The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Spectrum of an oscillator with phase noise
https://designers-guide.org/forum/YaBB.pl?num=1549470605

Message started by iVenky on Feb 6^th, 2019, 8:30am

Title: Spectrum of an oscillator with phase noise
Post by iVenky on Feb 6^th, 2019, 8:30am

We can represent oscillator as

y(t)= A cos (wt + Φ(t)), where Φ(t) is phase noise.

If Φ(t)=cos(w_nt) then, we can show using Narrowband approximation that

y(t)≈ A cos (wt) + A sin (wt) sin ( w_nt)

Which one of the figures attached (a or b) is the actual spectrum of the oscillator?

I used to think it's (b) and the entire carrier power is spread throughout, but I can't explain it mathematically. Only (a) makes sense mathematically since we have a carrier term (highlighted above) and then a phase noise term.

Question 2:

Also when we measure SSB phase noise it's overlaying both left and right of the carrier, resulting in 3 dB more phase noise compared to DSB phase noise, right? In the plot (a), the dBc that I am plotting is DSB phase noise, which is 3 dB lower compared to SSB phase noise, right?

Title: Re: Spectrum of an oscillator with phase noise
Post by Ken Kundert on Feb 7^th, 2019, 2:47pm

It's b.

The narrow band approximation does not work with oscillator phase noise. Fundamentally the narrow band approximation is a small-signal approximation, it assumes Φ(t) is small. But in oscillator phase noise, Φ(t) is unbounded over long time intervals. It is this unbounded nature that leads to the Lorenzian distribution.

Also, when you plot S(f), you should distinguish between S_Φ or S_V.

-Ken

Title: Re: Spectrum of an oscillator with phase noise
Post by iVenky on Feb 9^th, 2019, 7:04pm

Hi Ken,

Thanks for the reply, I got it now. I was plotting Sv, I understand S_phi goes to infinity due to flicker but not Sv.