[time-nuts] Question on Tom's ADEV plots
javier.serrano.pareja at gmail.com
Fri Oct 17 12:55:29 UTC 2008
Yes, I am wondering what sort of reference would be ten times better than a
good OCXO at tau=0.1 second and stay good to tau=10^4 seconds. I am looking
at the very first plot of http://www.leapsecond.com/pages/fury/. Also, what
do you mean by " to avoid having to subtract background noise from the ADEV
plot"? Is there a standard recipe for this type of operation?
On Fri, Oct 17, 2008 at 11:17 AM, Tom Van Baak <tvb at leapsecond.com> wrote:
> Hi Tom,
>> I've been looking at some ADEV plots in your website, like
>> the ones for the Fury, and I was wondering if there is a page
>> in the site that explains your measurement setup. I am
>> specifically interested in the reference oscillator you use. It
>> should in principle be much better than the Fury's OCXO for
>> short tau and better than GPS for medium tau, right?
>> PS. Feel free to answer to the list. I was not sure I should post
>> this since it's a direct question to you.
> That's a good question and it sort of depends on which year I
> made the measurement or what performance level I'm expecting
> out of the unit being tested.
> I also found out the hard way that it also depends on the tau,
> since different references have different "sweet spots". Now
> that I have quite a range of oscillators here, from BVA to ULN,
> from cesium to maser, I can usually pick one that will give the
> most accurate ADEV or PN plot for the range of tau that I'm
> interested in. Not surprisingly I have no one standard that has
> both the best PN and the best ADEV. For a lot of work I use
> an unlocked Z3801A as the PN ref and a cesium or a locked
> 58503B as the ADEV ref. But it varies according to need or
> what's not in use at the moment.
> There's a good chance for any given plot that I have a record
> here of which instrument generated the plot and what reference
> was used. If it's important, either I mention it, or just ask me.
> On your GPSDO question, yes, to avoid having to subtract
> background noise from the ADEV plot, one needs to use a
> reference (and measurement system) that is several times,
> maybe even 10x, better than what's being tested.
> http://www.LeapSecond.com <http://www.leapsecond.com/>
More information about the time-nuts