Measure of jitter and phase noise on a square wave does not match
时间：04-04
整理：3721RD
点击：
Hello everyone,
I have done a measure of jitter and phase noise on a square wave, but two measures do not match.
In detail, I've measured the jitter with an oscilloscope and the phase noise with a spectrum analyzer, exploiting a special feature of the instrument. By a post-processing computation, I've converted the phase noise data into a jitter value. This value does not match with the jitter measured by oscilloscope: it is 10 times grater than one obtained from oscilloscope.
Why does it happen this? Maybe I measure the phase noise only on the central frequency, which corresponds to the frequency of the square wave. Could it be or there are other causes?
I am a little confuse
May you help me to understand?
Thanks and best regards,
Antonio
I have done a measure of jitter and phase noise on a square wave, but two measures do not match.
In detail, I've measured the jitter with an oscilloscope and the phase noise with a spectrum analyzer, exploiting a special feature of the instrument. By a post-processing computation, I've converted the phase noise data into a jitter value. This value does not match with the jitter measured by oscilloscope: it is 10 times grater than one obtained from oscilloscope.
Why does it happen this? Maybe I measure the phase noise only on the central frequency, which corresponds to the frequency of the square wave. Could it be or there are other causes?
I am a little confuse
May you help me to understand?
Thanks and best regards,
Antonio
First question: what is a Fourier frequency range for a phase noise measurement?
Second question: how many samples do you use to calculate a rms jitter?
Read this: https://www.sitime.com/api/gated/AN1...easurement.pdf
well they are two different things. Turning phase noise into time jitter requires an integration...and the limits of that integration, and any frequency windowing you use in your test equipment, will change the end result.
Also, "Phase Noise" assumes a mostly random noise density. If there are harmonics, or transient spike effects...a phase noise measurement will probably MISS them, although they might contribute to time jitter.