Simulating phase noise for PFD-CP
cupric
I am trying to simulate the phase noise for a phase frequency
detector-charge pump. I have a couple of questions. Hope you can help
me out with this.
(a) In the pnoise tab of the analysis window, if I choose my sweep
type to be relative, what should I choose the relative harmonic to be
(what does this option mean?).
(b) Once I get the noise current plotted. How do I integrate the
waveform over the frequency of interest. When one does a strobed noise
analysis, one gets the option of plotting the integrated noise. But,
this option doesn't appear when I do the normal pnoise analysis (or am
I missing something here). Or, do I need to use the calcultor to do
the integration. Again, how do I use the calculator for this.
Thanking you in anticipation.
cupric
Vikas Vijay
The relative harmonic should be 1 (which is also the default) because
you need to measure the noise across the central periodic freq of the
output (i. harmonic 1 of the pss analysis). You just need to look at
noise skirts around the central frequency, the harmonics will anyhow
be filtered out by the following LPF stage.
Regarding ur second question i am not very clear.
-vikas
cupric...@yahoo.com (cupric) wrote in message news:<7d250b17.03080...@posting.google.com>...
Andrew Beckett
Answers below:
On 3 Aug 2003 20:37:18 -0700, cupric...@yahoo.com (cupric) wrote:
>Hi!
>
>I am trying to simulate the phase noise for a phase frequency
>detector-charge pump. I have a couple of questions. Hope you can help
>me out with this.
>
>(a) In the pnoise tab of the analysis window, if I choose my sweep
>type to be relative, what should I choose the relative harmonic to be
>(what does this option mean?).
What it means is that you are specifying the frequency sweep as an offset
relative to a multiple of the PSS fundamental. Therefore if you sweep
from f1 to f2, the actual frequencies it sweeps over is
f1+n*Fpss to f2+n*Fpss
where n is the relative harmonic number, and Fpss is the PSS fundamental.
When the phase noise is plotted relative to the magnitude of the output at the
harmonic you picked (so it is in dBc).
Another benefit of the relative sweep is when simulating oscillators, because
you want to look at the noise skirts around the oscillator frequency, but
you don't know the oscillator frequency (accurately enough) before
running - so letting PSS find it itself, and doing a relative sweep makes
sense.
>(b) Once I get the noise current plotted. How do I integrate the
>waveform over the frequency of interest. When one does a strobed noise
>analysis, one gets the option of plotting the integrated noise. But,
>this option doesn't appear when I do the normal pnoise analysis (or am
>I missing something here). Or, do I need to use the calcultor to do
>the integration. Again, how do I use the calculator for this.
>
You can integrate the noise output using the Results->Print->Noise Summary
menu from the ADE window.
Regards,
Andrew.
>Thanking you in anticipation.
>
>cupric
--
Andrew Beckett
Senior Technical Leader
Custom IC Solutions
Cadence Design Systems Ltd
game7world
For PFD/CP pnoise simulation, is it OK to set sweeptype=absolute？
What I found is that the noise performance of using sweeptype=absolute is better than using sweeptype=relative=1.
Thank you.
G7W