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 >> Period Jitter is different from Jitter obtained from Eye Diagram
https://designers-guide.org/forum/YaBB.pl?num=1494239388

Message started by Trinath on May 8^th, 2017, 3:29am

Title: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Trinath on May 8^th, 2017, 3:29am

This is very basic question , which is confusing me.
The problem here is , I am supposed to calculate Period Jitter for a clock signal . I am hoping period jitter to be same as the Jitter obtained in Eye diagram .
But I see a huge difference among these two measurements.
Can someone please tell me why this difference is seen.
The way i calculated Period Jitter is by Calculating the period of each cycle and subtracting it with my ideal Period.
Whereas eye diagram is measured in adexl.

Title: Re: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Alberto-G on May 8^th, 2017, 6:48am

Hi, period jitter variance (true or Allan) should be defined on the difference of the time instants (crossing edge)
j(i)=(t(i+1) - t(i))
I think period jitter is a self referred measure so no ideal period should be used.
If you fix an ideal period for instance in an oscillators you get an unbounded rms jitter value.

I'm not an expert on this topic, maybe other users can answer better :)

Title: Re: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Trinath on May 9^th, 2017, 12:45am

Thank You , but can you also tell me why the Jitter seen in Eye Diagram is much more than Period Jitter

Title: Re: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Alberto-G on May 9^th, 2017, 2:38am

I think because jitter based on eye diagram takes into account also deterministic jitter components instead period jitter acts like a filter each n×f0 component.

Title: Re: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Trinath on May 10^th, 2017, 3:48am

let me be precise here:
I am intentionally giving a supply noise on my DC supply,and for this setup am calculation period Jitter .
The period Jitter I got was 20ps where as in Eye Diagram am seeing around 40ps.
Which among these two values should I consider. I wonder why these two are not same as my circuit is disturbed only by Supply Noise .
I am checking this Jitter for a Clock signal not data signal.

Thank You. .

Title: Re: Period Jitter is different from Jitter obtained from Eye Diagram
Post by Hercules Poirot on Aug 8^th, 2017, 5:49am

If you are giving to some synthesized logic, period jitter is the one that really matters as the physical design will be constrained by the smallest period of the clock.
If you are using the clock to run a Tx of a serial protocol which has embedded clocking, your accumulated jitter matters.