[time-nuts] Rb as source for ADEV?

Tom Van Baak tvb at LeapSecond.com
Thu Feb 6 18:13:56 EST 2014

Hi Bob,

Short-term measurements like that do a good job using a OCXO and TIC to reveal TIC and GPS resolution limits. It's a useful exercise, but it usually doesn't tell you anything about the OCXO itself.

Long-term measurements like that do a superb job using GPS to measure the long-term stability and frequency drift rate of the OCXO (or Rb). This, because for intervals of days to weeks to months you can't beat GPS as an inexpensive ultimate time/frequency standard. And for those long intervals, the TIC resolution is usually more than enough.

Just collect the raw phase data and raw DAC data. In a few weeks you'll be able to make some really nice plots.

Also consider freezing the DAC for a week (unlocked) and compare that data with your locked data. You can then make plots that look like: http://leapsecond.com/pages/gpsdo/

On Rb -- since you're now a time nut, do not let the Rb warm up at all -- the data get you the first hour or first day will be just as interesting as the data you collect the next month.

On ADEV -- I know the source code calls them adev and oadev, but you should just call it ADEV. No one uses the old back-to-back formula anymore; the overlapping formula is the default. And then plot using the "all" or "many" tau method I mentioned.


----- Original Message ----- 
From: "Bob Stewart" <bob at evoria.net>
To: "Discussion of precise time and frequency measurement" <time-nuts at febo.com>
Sent: Thursday, February 06, 2014 2:23 PM
Subject: [time-nuts] Rb as source for ADEV?

I'm getting a little more familiar with ADEV and OADEV now. With a little help from Tom, I've put together gnuplot scripts and his software to get a decent plot. But, as some time-nuts have pointed out, my measurements are of one part of my system being compared to another. Would I learn anything useful by hooking my Rb standard up to my 5335A and comparing that to the OCXO? How long should I let the Rb heat up before starting the test? I'm a bit uncertain about the result of comparing a stable clock of questionable accuracy to an accurate clock of questionable stability. Hopefully I said that right.


Bob - AE6RV

More information about the time-nuts mailing list