As a comment on this dialogue: > > Such as temperature and humidity. One of the Cesiums seemed to have a rathe r > > large frequency offset. Did you drift-compensate your ADEV measures or not?
> > No, I ran the ADEV in Stable32 without removing drift. My understanding > is that linear drift shouldn't affect the calculation. I was also a > little concerned whether drift removal would work very well given that > both Cesiums had periods where the offset changed for quite a long > period (CS1 near the end of the data, CS2 at the beginning) and I wasn't > sure what impact that might have. The straight ADEV will certainly be affected if there is a drift. The ADEV for delta time=T is simply the RMS difference between the clock phase at time t and time t+T. Any rate offset will appear in this calculation. If you are using STABLE32, you might want to fit out a smooth polynomial with the tools that are supplied. Or as an alternative, use the Modified ADEV; this computes the difference to both earlier and later times and implicitly includes the removal of a simple offset. John -- you can get info on STABLE32 at [1]http://www.wriley.com/. Even if you don't get the software, the manual is a good bible and can be fetched at [2]http://www.wriley.com/Manual146.pdf. Pay special attention in the manual to pages 31-49 for a discussion of the types of variances that are of use in the community. FYI -- The folks who maintain the Rb and Cs clocks on the GPS satellites are sold on the Hadamard variance as a long-term performance measure. To learn how to read an interpret the slopes on an ADEV plot, refer to pages 77-83. On page 79, you will see that the Modified ADEV gives you more information in the short-term (i.e. the area where the VCXO is the principal contributor) part of the curve for White vs Flicker phase modulation. I highly recommend STABLE32 -- many of us have found it well worth the $400 price that Bill Riley charges for his labor of love! 73, Tom References 1. http://www.wriley.com/ 2. http://www.wriley.com/Manual146.pdf _______________________________________________ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts