NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Peter Monta
Date: 2024 Dec 20, 11:24 -0800
Hi Sean,
The claim of "less than a second per year" should be taken with a large grain of salt. Quartz oscillators have a noise process called "aging", the gradual drift of the resonator with time. You can compensate for temperature all you like (as their web page describes), but aging sets a fundamental limit on quartz accuracy.
The right way to improve things within the wristwatch form factor, in my opinion, is to use an auxiliary MEMS oscillator. These can have very low aging because of their construction (silicon resonator, high temperature sealing). Power requirements are too high for a wristwatch when used continuously, but this could be mitigated by powering it up only sporadically, say for a few seconds a week, as a check on the main quartz oscillator.
Such a device might be okay to tide me over while I'm waiting for my thorium-229 wristwatch. This interesting talk describes recent progress in nuclear clocks (talk starts at 18:55), and he does indeed mention future prospects for small, low-power systems:
Cheers,
Peter
