💾 Archived View for bbs.geminispace.org › u › decant_ › 18608 captured on 2024-08-31 at 15:43:57. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2024-08-19)
-=-=-=-=-=-=-
Thermal compensation (TC) has long been the problem for accurate time keeping. During the mechanical era of horology, the problem was consider solved with the invention of alloys with high dimensional stability vs temperature. Invar is one such alloy, Charles Édouard Guillaume won the Nobel prize for physics for his work on Invar.
The current industry standard is to strap a TC ASIC to a single crystal oscillator. But other methods have been used in the past.
For example, a carefully designed and calibrated network of thermistor that achieve TC by control the input voltage of the crystal.
I think the twin quartz approach is the most elegant[1]. It uses two tuning fork crystals calibrated to different temperature to achieve thermal compensation. [1] is a very good read
The least elegant way for a high accurate quartz by FAR is what is called OCXO. With OCXO you entirely give up on governing the frequency of the quartz crystal, Instead, you seal the crystal into a tiny oven that is normally held at 80 Celsius!
What about atomic clock? Well, all atomic oscillator have very bad jitter, to make the output of an atomic oscillator useful, it is locked to a very high quality quartz. So jitter from the atomic is corrected by the quartz and the long term frequency drift of the quartz is corrected by the atomic.
[1] Malcolm Pipes, “The Magic of the Twin Quartz Watch,” British Journal of Horology, December 2023 pp. 512-512.
Jul 17 · 7 weeks ago · 👍 ResetReboot, klimperfix