Just replaced an ocxo in my ST620 with a Moron ocxo, which had been unpowered for a few years in all probability. It took several days to come to a reasonable equilibrium. In a system with as many components as an ocxo, I suspect that seeking a new equilibrium will take an appreciable time well beyond design criteria, as Charles expresses. Hunting time constants may indeed be quite long. Don
> On Nov 27, 2014, at 1:48 PM, Charles Steinmetz <csteinm...@yandex.com> wrote: > > Mark wrote: > >> I've long had a nagging suspicion that OCXO's that are not adjusted will in >> practice have lower ADEV than ones that are tweaked regularly. Several >> days ago I noticed that one of my 10811's was performing quite well (I >> believe this is the first time any of 10811's have delivered adev / madev >> numbers in the 13's) and this was sustained for approx 2 days. After >> trimming the frequency and letting it sit for a number of hours I noticed >> the adev was notably worse. > > Every time a quartz oscillator is disturbed in any way, its stability is very > likely to go down for a time until it settles back in. The disturbance could > be adjusting its frequency (either mechanically or via EFC), interrupting the > power, changing the crystal temperature, physically bumping the unit, or > anything else that changes its operation. How long it takes to settle back > into "normally stable" operation depends on the crystal itself and on how > violent the disturbance was. I would never expect a quartz oscillator to be > back to normal stability for at least some days after a macro frequency > adjustment. [Tiny, tiny adjustments such as done by ongoing GPS discipline > do not seem to have a large effect on the stability of quartz oscillators.] > > But I note that you say an oscillator that had never delivered stability in > the e-13's had a couple of very good days, and then didn't go back to the > e-13's in a number of hours. First, you probably shouldn't expect it to go > back to the "couple of good days" level -- it's likely to go back to its > "normal" level. And second, you shouldn't expect it to get there for at > least a few days. > > Best regards, > > Charles > > > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
