I don't recall; this oscillator was installed in a piece of gear I sold several years ago. I checked it every few months while I had it. All I remember is that the note I used to keep track of its calibration had at least four entries on it at the time I sold the generator, and I'm very sure it never left +/- 1E-9, always measured after the 8662A had been left running for at least a couple of hours. Whether it looked like a random walk or not, I couldn't say.
Lucky environmental factors must have played a big part because -- as Rick says -- they aren't normally that stable. There is a definite bathtub-shaped curve at work, as well. If the units that Rick saw at Agilent were all brand-new, then of course they weren't very stable compared to the OCXO in this 8662A. I could tell that it had been running for a long time when I received it, because I observed almost no accelerated aging during the first few weeks of service. Likewise, from what I have seen, the early-80s OCXOs don't tend to stay put as well as the newer ones do. Whether that's due to materials/manufacturing quality or simple aging is impossible to say, given the limited number that I have experimented with. -- john, KE5FX > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Hal Murray > Sent: Friday, December 29, 2006 1:29 PM > To: [EMAIL PROTECTED]; Discussion of precise time and frequency > measurement > Subject: Re: [time-nuts] HP 5345A > > > > > It probably wandered around much more than that during the year and > > you just happened to measure it when it was close to the starting > > point. Kind of like the broken clock that keeps perfect time every 12 > > hours. In any event, I can safely say, having looked at thousands of > > 10811's, that I have never seen one that had anything like that level > > of aging. > > I thought aging was generally uni-directional and reasonably > predictable if > you had enough data. > > Does it wander in both directions? > > _______________________________________________ time-nuts mailing list time-nuts@febo.com https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts