[time-nuts] Upper limit on phase noise from two oscillators.

Mike S mikes at flatsurface.com
Wed Apr 27 11:27:41 EDT 2005

At 10:05 AM 4/27/2005, Tom Van Baak wrote...
>Now is low phase noise really what you're after, or
>are you more interested in low short- and long-term
>Allan Deviation?

As a rookie to this, please correct me if my understanding is off.

Phase noise is only really relevant to sine wave signals. If one had a perfect square wave source, it would have zero Allan deviation, but significant phase noise, due to it's difference from a pure sine wave.

If one is using the repetitive nature of a signal by triggering at a fixed level and slope, phase noise may not have a significant impact on the application, since a figure for phase noise doesn't necessarily imply that the wave shape varies across cycles, nor does it indicate poor Allan deviation.

Worse Allan deviation would always be reflected in worse phase noise (because it implies a less than perfect sine wave), but not the other way around. 

More information about the time-nuts mailing list