SoylentNews is people
SoylentNews
from the you're-not-just-bored:-the-wall-clock-IS-slow dept.
Electric clocks on continental Europe that are steered by the frequency of the power system are running slow by up to 5 minutes since mid-January according to a news release from the European Network of Transmission System Operators for Electricity ('entsoe'). The transmission system operators (TSOs) will set up a compensation program to correct the time in the future.
Many electric clocks rely on the transmission system frequency to provide a source that minimises long-term drift. Quartz crystals have good short term stability, but dreadful long term stability, so plant and machinery that requires power to be turned on or off at a specific time each day without maintenance over a long period historically used clocks slaved to the power-system frequency, which is kept long-term stable by the system operators to prevent problems in power generation and transmission across national and supra-national grids - for example, attempting to switch supplies to generators that are not synchronised to the grid frequency can severely damage the generator.
It is normal for transmission system operators to allow the frequency to drop slightly at periods of high demand, thus slowing clocks, but usually, the frequency is increased during periods of low demand to ensure the long-term average frequency remains stable.
(Score: 2) by VLM on Wednesday March 07 2018, @02:42AM
IQD has some "low performance" products for $7 in bulk that are like 0.3 ppm. In stock at Mouser not imaginary datasheet stuff.
Right now clock accuracy is changing like crazy, like how a couple decades ago when I was a kid a switching FET that could survive an amp at a couple volts at audio switching freqs was amazing and then specs started changing by a factor of 10 like annually or so.. for decades? So your specs really aren't more than a couple years out of date, like 2016 or so. An exciting time to be alive, etc!
A couple years ago I built a 10 mhz GPS-do using the usual ebay stuff for my workbench for my counter, oscillators and spectrum analyzer and its almost like "why bother" with modern oscillators. If lightning took out the GPSDO antenna I'd probably toss it and buy a new "late 2010s" era TCXO.
Note that over a long term like hours (weeks?) the theoretical error of my gpsdo might be zero but the equivalent phase noise or thermal drift or whatever short term might be worse for my gps-do made of weird (old) surplus parts than for modern nifty gadgets... or maybe the other way around? In my infinite spare time I'll have to evaluate this.