Re: [SI-LIST] : Jitter Measurement

About this list Date view Thread view Subject view Author view

From: David Instone ([email protected])
Date: Wed Apr 04 2001 - 02:31:10 PDT


Pat, see comments below

Zabinski, Patrick J. wrote:
>
> Please excuse my ignorance and enlighten me...
>
> I was under the impression that "noise" had both
> phase and magnitude components, so the 750 uV noise
> floor spec does add amplitude 'jitter' as you state.
> However, isn't there an additional phase component
> in there somewhere? If so, does Tek provide
> this spec?
On anything other than the (impossible) truly vertical edge amplitude
noise also appears as timing jitter. Draw yourself a trangle wave and
add some high frequency noise ie make the line very thick, you should
see that the line has become wide in both the y and the X (timing)
axis. Or have I missunderstood your comment.
>
> Also, regardless of the noise floor of the SD-24
> head, shouldn't we be taking into account the
> phase stability (a.k.a., jitter) of the reference
> oscillator internal to the TEK11801?
Yes

> Is this where
> the 2 ps spec comes in?

No I think the 2ps is for the trigger circuit. From the Tek 2001 catalog
entry (page 114)for the 11801-

Trigger system
.
.
Delay jitter- 1.1ps +4ppm of a position typical. 2ps +5ppm (max)

>
> Again, this is a weak spot of mine, but it would seem
> that you should "add" (simple add? RMS? other?) the
> various jitter components together to obtain the composite jitter.

For components that are gaussian you rms add, any bounded components
then have to be convolved together and then the gaussian convolved in.

May I recommend you download
ftp://ftp.t11.org/t11/member/fc/jitter_meth/99-151v0.pdf
and read Annex A

Dave

>
> Comments?
>
> Pat
>
> >
> > This is not quite true. It depends on the slewrate of the
> > signal being
> > measured. assuming you are using the SD24 head, this has a
> > noise floor
> > of 750uV (can't use the averaging function when measuring jitter). Now
> > if the slew rate of the signal at the jitter measuring point is say
> > 2.6mV/ps (approx 500Mhz 1V p-p sinewave) then that 750uV of
> > noise looks
> > like 750/2600 ps of jitter but with a 1V 50MHz signal (260uV/ps
> > slewrate) it would look like 750/260 ps of jitter. A similar effect
> > would occur for a 0.1V 500Hhz signal. Now the same principle also
> > applies to the trigger. A slow slew rate gives more trigger
> > uncertainty,
> > which appears as jitter.
> > So get the biggest amplitude fastest risetime trigger signal you can
> > and keep the Y amp attenuation to a minumum.
> >
> > Dave.
> >
> >
> > Mike Hughes wrote:
> > >
> > > A while ago, a TEK apps engineer told me that the "noise
> > floor" of the 11801B is
> > > around 2pS RMS. My impression was that this limit is
> > independent of the
> > > frequency being measured. (As long as you are within the BW of the
> > > scope/sampling head of course.)
> > >
> > > Mike
> >
>
> **** To unsubscribe from si-list or si-list-digest: send e-mail to
> [email protected]. In the BODY of message put: UNSUBSCRIBE
> si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu
> ****

-- 
Regards

Dave Instone. Compliance Engineer Storage Systems Development, MP24/22 Xyratex, Langstone Rd., Havant, Hampshire, P09 1SA, UK. Tel: +44 (0)23-92-496862 (direct line) Fax: +44 (0)23-92-496014 http://www.xyratex.com Tel: +44 (0)23-92-496000

**** To unsubscribe from si-list or si-list-digest: send e-mail to [email protected]. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu ****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Jun 21 2001 - 10:11:26 PDT