| Member Comments |
Avg rating: 80
|
|
|
|
dlao
- Special case
|
2008-11-03 |
|
It’s a great article that is useful and easy to understand. I just want to point out that Equ. (16) is not valid when a_i happends to be -10.
|
|
|
|
|
maxclk
- Re: Special case
|
(no rating)
2009-02-02 |
For Eq. 16, when a_i = -10, we have a situation of 0/0. To evaluate Eq. 16 in this case, take the derivatives of both numerator and denominator with respect to a_i, respectively, and then let a_i = -10.
|
|
|
|
|
Sushil
- weighting factor to phase noise curve
|
2009-05-27 |
Hi,
I was going through following MAXIM app-note:
http://www.maxim-ic.com/appnotes.cfm/an_pk/3631/
and it says that phase-noise curve should be multiplied to "2*sin(pi*f*Tc)^2" before integration. In this article I do not see this weighting factor, so calculated period-jitter will be large and should not be correct?
Regards/Thanks
Sushil
|
|
|
|
|
maxclk
- Re: weighting factor to phase noise curve
|
(no rating)
2009-08-31 |
|
This is because that we only consider the phase noise. The additive noise is not considered here and it is assumed relatively smaller comparing to the phase noise, see (4).
|
|
|
|
|
dexter
- Power spectrum density units
|
(no rating)
2009-09-21 |
|
Shoudln't L(f) be in dBc/Hz instead of just dBc?
|
|
|
|
|
maxclk
- Re: Power spectrum density units
|
(no rating)
2009-09-28 |
It is not necessary. Since the jitter is the integration of L(f) over the frequency, the frequency unit will not affect the result.
|
|
|
|
|
tech
- appoximation of square to sine
|
2009-10-14 |
sq(t) = sgn(sin(wt)) sq wave of f0=w.
When a RP n(t) is added to sin(wt) and then sgn is taken, it can lead to glitches at zero crossings, affecting the frequency of the sq wave itself.
|
|
|
|
|
stefannm
- Problem convincing myself of equation (5)
|
(no rating)
2009-10-16 |
Consider the imaginary case where Phi(t) is equal to some constant. That means over all time Phi(t) is not changing, so the clock period would equal the ideal clock period except that it is phase shifted by Phi(t). According to equation (1), this would result in a Jper = 0; however, equation (5) implies that Jper = some constant. I believe equation (5) represents the accumulated jitter and not period jitter. Please explain if I am incorrect in my understanding.
Thanks
|
|
|
|
|
maxclk
- Re: Problem convincing myself of equation (5)
|
(no rating)
2009-10-29 |
|
Phi(t) is the jitter part only. If Phi(t) has a constant offset, the offset can be taken out and put into the sin(wt) term. The offset does not show in equation (6). So it does not affect the derivation.
|
|
|
|
|
Krsih
- values in Table 2 in eq 16 - Not working
|
2009-11-05 |
I am trying to substitute values in Table 2 in eq 16 and I am getting negative sum ( -0.085798809) , for which sqrt can not be found.
I went through my excel file with calculations couple of times. I always end up with negative sum . Any ideas ?
|
|
|
|
|
| Your Comments | |
|
Login or register to post a comment.
|