The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> SpectreRF simulate divider phase noise
https://designers-guide.org/forum/YaBB.pl?num=1200582367

Message started by neoflash on Jan 17^th, 2008, 7:06am

Title: SpectreRF simulate divider phase noise
Post by neoflash on Jan 17^th, 2008, 7:06am

Hi:

I've read the doc on this web as "Simulating the Phase Noise Contribution of the Divider in a Phase Lock Loop".

In this paper, it is suggested that using spectreRF get voltage noise at output and post process it through calculation. However, any way to get phase noise directly?

Regards,
Neoflash

Title: Re: SpectreRF simulate divider phase noise
Post by neoflash on Jan 18^th, 2008, 1:56am

In the simulation results, there are also phase noise available. Is this the result we want?

Title: Re: SpectreRF simulate divider phase noise
Post by lianggu on Feb 15^th, 2008, 5:02pm

If you set noise type to jitter or timedomain in Pnoise, only voltage noise results will appear, no phase noise available at all.

Title: Re: SpectreRF simulate divider phase noise
Post by borisk on Feb 27^th, 2008, 9:05am

Hi,

I'm encountering the same problem. It seems to me that for oscillator phase noise for instance, you set the noise type to sources, and when you run pnoise there will be a phnoise option. This is I think the total noise, i.e., AM and PM, but luckily in oscillators the PM dominates such that when you look at phnoise it'll correspond to the PM component. The other thing to keep in mind is that it is in the form of voltage noise, but the PSD's of the voltage noise process and the phase noise process are related.

Anyway, back to the question. If looking at the divider or even simply inverters and such, we want the jitter and in this case, if you set the noise type to sources and check the phase noise it's much higher than expected. Here the AM dominates the PM, and again, the phnoise gives them combined in terms of voltage.

You can set the noise type to modulated and you'll be able to see the AM and PM contributions separately. I think that here we want to look at the PM only which turns out to be generally smaller (at least for some buffers I was looking at). What I don't know is whether the AM --> PM conversion, say from a noisy input, is captured correctly by this.

Any other comments? What about setting noise type to timedomain or jitter?

Cheers,

Boris

Title: Re: SpectreRF simulate divider phase noise
Post by Frank Wiedmann on Feb 27^th, 2008, 11:54pm

See http://www.designers-guide.org/Forum/YaBB.pl?num=1201654647/3#3 and also http://www.designers-guide.org/Forum/YaBB.pl?num=1192438637/1#1.