[time-nuts] Square to sine wave symmetrical conversion

[Bob Camp]

Hi
> On Jul 27, 2015, at 1:28 PM, Charles Steinmetz <csteinmetz at yandex.com> wrote:
> 
> Bob wrote:
> 
>> In tis case the question is "do you *need* low harmonics in the oscillator
>> stage to get low phase noise?"
> 
> Note that there are actually two questions.  One is WRT the phase noise of the oscillator itself, and the other WRT the phase noise of a system that integrates the oscillator.  In particular, even harmonics in the oscillator proper generate additional phase noise in the system when the signal is AC-coupled and/or DC-restored, and when it is fed to a zero-cross detector or other circuit that is sensitive to the symmetry of the waveform.
> 
> NIST published a paper on this.[1]  There is other research describing and quantifying the phenomenon, as well.
> 
> Best regards,
> 
> Charles
> 
> 
> [1]  "The Effect of Harmonic Distortion on Phase errors in Frequency Distribution and Synthesis," Walls and Ascarrunz  <http://tf.nist.gov/timefreq/general/pdf/1437.pdf>
> 

An interesting (similar) experiment: 

Rather than measuring “timing error” as they did in the paper, do the same experiment (variations in mixer level) and measure the phase noise instead. The basic 
configuration of a DBM as a phase detector is pretty well documented. Set up the signals in quadrature and fire up the audio spectrum analyzer. Calibrate the beat
note in the usual fashion for each power setting and take a look at the indicated phase noise at some convenient offset like 1K or 10 KHz. 

If mixer saturation / harmonics are an issue, you should see the measured phase noise go up as the levels get closer to each other (mixer in saturation). Indeed what you 
see is that (if anything) the phase noise improves as the sensitivity of the system gets better with the higher level of drive. There are a whole raft of papers on this going
back several decades. It’s the normal approach to measuring very low levels of phase noise with the older DBM based instruments. 

Bob

> 
> 
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.