Problem in getting jitter from phase noise of pll

kooller

Hi all,
7 a. p1 O1 Z+ S4 X! c5 y; t9 ?
$ u- ?$ |9 X3 ~, v) W% p0 d5 _2 CI'm recently doing the jitter simulation of PLL, and I first got phase noise of PLL by adding every part of noise of PLL.

Then calculate jitter by the following formula:
, B1 Q# ]5 ^+ y. t7 h
* _* ^& a0 r9 h! G; ]' Wphase jitter = sqrt(integral(fre,2*Lpll))/(2*pi*f);
9 x+ ?, A$ e. i. ^8 Fperiod jitter = sqrt(8*integral(fre,Lpll.*(sin(pi*fre*T)).^2)/((2*pi*f)^2));

Frequency of PLL is 5GHz, and the integral band of phase noise is 100Hz~100MHz, phase noise of PLL is as follows: -85dBc/Hz @100Hz; -64dBc/Hz @100kHz; -80dBc/Hz @1MHz; -102dBc/Hz@10MHz; -125dBc/Hz@100MHz.

$ y! C- L2 ?, V( [+ DThe calculated jitter results is :
phase jitter: 18.9ps.
. b; W+ d7 c0 b1 g2 @( |1 A) bperiod jitter: 0.04ps.
& P2 S" R  n$ f% @
This shows that the period jitter is so small, and only 1/450 times of phase jitter, while our test results shows that the period jitter is about 1.8ps.
' [. l" ^1 q, W' Q' x5 o6 |* A; s
So can somebody tell me where is the problem of my calculation, and how much will phase jitter larger than period jitter?

anita66

看您的PN那麼差,jitter應該不可能不到20ps,其實PN是ㄧ個積分所得到的數值,同樣,jitter換算回來時,也是要累加才對,所以,你的算法有問題,可以再找一下正確的計算公式,3Q~