There you go. Federal statute and chain down to the standard.
The point is that there is a clear rationale for a standard that keeps the long term average of the frequency very very accurate, accurate enough that you can set you watch by it, and run your clocks from it. And not only is there a good reason for doing this, there is a federal statute that provides for regulation of power, and a standard that the regulating authority enabled by that statute uses to perform exactly this regulation of long term average frequency. And this standard is tied back to NIST.
So not only is long term accuracy a good idea, it is actually performed, and is performed via federal statute.
There are three main power grids in the US - eastern, western, and Texas (yes, Texas is on its own grid). The eastern and western grids extend up into Canada, and Quebec has its own grid.
I don’t know about Quebec, but the three US grids used to keep the long term frequency at exactly 60 Hz. It’s important to note that this was only long term though. During the day, heavy loads tended to drag the frequency down. At night, they would speed the generators up to “catch up”.
This was done mainly to keep clocks running on synchronous AC motors accurate. Individual states (except for Texas) had no control over the grid frequency since each grid (except for Texas) spanned multiple states. This was regulated by the Federal Energy Regulatory Commission.
Over the years, synchronous clocks have mostly gone away. Nobody really knew how much stuff out there really depended on the long term accuracy of the AC frequency though, so in 2011 they announced that they had decided to run an experiment. The power companies would stop working so hard to maintain the long term AC frequency, just to see who would notice.
This article has more details:
The experiment was intended to last for one year. There were numerous news articles in 2011 when the experiment was announced, and since then, I haven’t seen anything else about it one way or the other. There weren’t massive complaints. As far as I am aware, they never went back to tightly regulating the long-term frequency.
So the long-term accuracy of the AC frequency now seems to be a thing of the past.
Thanks ALL, for sharing what you know in this somewhat hijacked thread. Its been a reminder of things forgotten, the 2011 NERC test for example. Thank you engineer_comp_geek.
Excuse my earlier ramblings, I wandered so far off base, I confused even myself. My rub was originally with the comment by Dr Strangelove:
“They calibrate against atomic clocks” to which I composed, but didn’t post, the following screed.
The USNO clock is utilized as a baseline for regulatory statistics. As said upthread, deviations from the ideal (appx. 60 Hz) over various time frames
are corrected by process adjustments, either by faster, or slower turbine speed. However, this is not calibration of speed control but rather,
an adjustment of instrumentation set point. This process change hopefully brings numbers on a spread sheet into an acceptable range
of deviation from the ideal, the goal being to satisfy regulatory requirements.
Again, this is not instrument calibration, any more than one is calibrating a speedometer by altering the pressure applied to a gas
pedal when travelling above or below the speed limit. Changing the operating parameters does not equate to a calibration of the instruments
that control or indicate those parameters.
The USNO clock’s resolution is so incredibly fine, it has no practical use in our industry for instrument calibration, even for the
larger units up the line. I calibrate with (mostly) Fluke equipment, NIST traceable, and have never been approached as to why I don’t
use an atomic clock. The bean-counters and regulators use the USNO as a baseline for compliance and record keeping…
The people that actually maintain the instrument calibrations don’t give a fig about the clock.
After re-reading the entire thread several times, (I think) we’re all singing out of the same book. Sorry for getting my ass up in the air.:smack: