The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Measurements >> Phase Noise and Jitter Measurements >> does phase noise of ripple counter accumulate?
https://designers-guide.org/forum/YaBB.pl?num=1038425039

Message started by rfic on Nov 27^th, 2002, 11:23am

Title: does phase noise of ripple counter accumulate?
Post by rfic on Nov 27^th, 2002, 11:23am

Dear Ken,

I have read your papers about how to predict phase noise of PLL, it's very helpful.

I have one question about the phase noise of ripple counter (programmable counter).

In programmable counter, we know there's a critical path(including the 1st DFF and End of Counting circuit, etc) which will determine the speed of the counter; for the same consideration, I think only the phase noise contribution of the critical path will affect the phase noise. That means phase noise from other stages of ripple counter is NOT accumulated.

So do you think I should only consider the contribution from critical path or all the stages? :-[

thanks and happy thanksgiving day :)

rfic

Title: Re: does phase noise of ripple counter accumulate?
Post by Jitter Man on Nov 27^th, 2002, 2:47pm

It is easier to work with noise in dividers as jitter rather than phase noise. The total jitter of a divider is simply the sum of the jitter of each series stage (one has to perform the summation in terms of the jitter variance). For this reason, the jitter contributed by the later stages, which rarely limit the speed of the divider, can be on par or even larger than the jitter contributed by the first stage.

So don't assume that you can neglect the phase noise produced by the "non-critical stages" simply because they are not limiting the speed of the divider.

[glb]Jitter Man[/glb]

Title: Re: does phase noise of ripple counter accumulate?
Post by rfic on Nov 27^th, 2002, 5:38pm

hi, Jitter man,

thanks for your answer.

Let me explain my idea:

We know in time domain, the jitter means the rising-edge of signal (suppose everything is rising-edge triggerred) happens randomly(sometimes early, sometimes late) .

In ripple counter case, when critical stage change state, the other stages should be already "stablized" (normally already be zero), the rising-edge of the output signal totally depends on critical stage, unless the jitter from other stages are huge that it even affect the rising-edge of critical stage.

Does that make sense?

thanks

rfic

Title: Re: does phase noise of ripple counter accumulate?
Post by Jitter Man on Nov 27^th, 2002, 10:39pm

RFIC,
Of course I don't know your design, but that is not my understanding of how most ripple counters work. In fact, the name ripple counter stems from the fact that an output transition only results after the terminal input transition ripples through all of the stages. Thus, the jitter accumulates. If the jitter is due to random device noise, than the jitter contributed by each stage is uncorrelated, so it accumulates in an RMS fashion. If the jitter is due to the supplies, it is a vector sum.

[glb]Jitter Man[/glb]