I have three digital-read out clocks in my house. All three clocks are about 2 minutes slow after one month. These are not battery-powered clocks but plugged into the 110VAC outlets in my home. All are on the same circuit in the house. My computer however gains about 30 seconds a day on its system time. The computer is plugged into a surge protector/UPS. What is going on here?
a) Where are you? In most places the mains frequency is controlled so that over the space of a week or so, a simple electric clock that derives its timing from the mains will agree with the reference clock (usually an atomic clock) that the electricity utility uses.
b) The clock in your computer derives its time from an internal crystal oscillator. It doesn’t make any difference whether the computer is plugged into the mains or a UPS.
I’m in a small town in Minnesota USA. Utility company is run by the city. They buy power from the grid and distribute through the town. Thanks for the post.
Ok, you’re connected to the main US grid.
Since making my earlier post, I’ve looked at the latest NERC standards. The requirement for matching mains frequency to the atomic clock seems to have been dropped. I don’t know why.
I’d guess it’s because these days, most digital clocks rely on a quartz crystal oscillator.
And I’d guess that the reason has absolutely nothing to do with your suggestion, but rather federal/state regulatory issues. 
Does the quartz crystal oscillator expect an exact 60 cycles/second source or does a variation in the cycles have no impact on its ability to keep time accurately?
No. A crystal oscillator doesn’t care what the mains frequency is.
Why else would they have bothered to control AC mains frequency that tightly, assuming they ever did?
Does the quartz crystal oscillator expect an exact 60 cycles/second source or does a variation in the cycles have no impact on its ability to keep time accurately?
That’s the sole reason, AFAIK.
If you want perfect time, get rid of your clocks and get ones that are reset each day by radio. There are clocks (and even wristwatches) that are reset each day by radio from the atomic clock in Colorado. This works anywhere in the U.S. except for Alaska and Hawaii (and it also works in parts of Canada and Mexico).
A Frequency (and phase) match is very useful to minimize losses wehn sharing or bridging sections of the grid. Of course, in practice, it is impossible to maintain a strict lock under varying and somewhat unpredictable loads, but it’s a good idea to try.
Frequency match between “sections of the grid” is a bit more than useful. It’s an inherent property. It has nothing to do with minimising losses.
I’m not sure what you mean by phase matching. At a given instant, the phase angles at various parts of the grid will all be different.
Phase matching is only important at the instant of connecting something to the grid. E.g. a generator that’s just been started can be connected to the grid once its frequency (roughly) matches that of the grid. You do that connection when the phase angle of the generator (roughly) matches the local phase angle of the grid. Thereafter, the frequency of the generator is locked with the frequency of the grid.
I assume here that you’re talking about a strict lock with an independent reference, like an atomic clock, rather than a strict lock with other “sections of the grid”. It’s impossible not to maintain a strict lock with other sections of the grid.
::Ducks::
Even a stopped clock is right two times a day.
Thanks Meeko but these guys are not stopped - they are just SLOW. It’s either cheaply made clocks bought for a cheap price by me at Walmart or something with the current in my house or my small Minnesota town. A real pisser for me, an anal Virgo (is that redundent ?), who likes everything perfect. And time should always be perfect.
An anal Virgo or an anally probed Virgo? Perhaps you and your home were swept up and away at near light speed, probed, and then returned again at near light speed. That would explain the time difference.
a1997xf11, if you have a voltmeter handy, check the voltage at a convenient outlet. Chances are that you are running slightly less than 120v. This could be caused by being at the end of a circuit, a voltage regulator that is either ‘locked down’ or improperly set, or it just may be that your municipality is running slightly undervoltage. My folks (also in rural Minn.) have the same problem occasionally.
Desmostylus is correct in his statement that it’s impossible not to maintain a ‘strict lock’ on the grid as was recently demonstrated in the massive outage in Eastern US states. If enough generators (supply) are lost, the load (which has not lessened) will be such that the additional strain on the remaining generators will cause them to slow down. Relays in substations will detect this ‘underfrequency’ and trip out load until the frequency recovers. If the relays are set incorrectly, or if a relay otherwise fails to operate, the protective relays on the generators will open causing further loss of supply resulting in a ‘cascading blackout’. The underfrequency relays are set such that x amount of load will be dropped at, say, 59.6hZ, y amount will open at 59.4hZ, z amount at 59.2, etc. This allows individual utilities to somewhat prioritize their loads leaving the most critical on for the longest time possible.
As for quartz clocks, the quartz crystal will vibrate at a ‘regular’ frequency if provided with a ‘regular’ voltage. If the quartz crystal is ‘tuned’ to 120 volts, any deviation in voltage will result in a higher or lower rate of vibration. As i believe it would not be cost effective to test each crystal to determine its exact frequency for a given voltage, i would expect that there is an allowable ‘range’ that is acceptable depending on the quality of the timepiece.
I am unfamiliar as to whether there is or was a ‘standard’ that mandated all generators run strictly to the atomic clock, as (here in the US) the frequency was agreed upon for economic reasons well before the advent of the Cesium Atomic Clock.
Curiously, railroads, whose signal lines operated at 25hZ, used to send a ‘spike’ down their lines once per day to ‘reset’ all their station clocks to the hour. In days past, the railroads were the ‘official’ timekeepers.
Only an analog clock. A dark plastic square is never the right time.
In fact, it was the need for a standardization in time in regards to delivering mail that prompted the railroads to develop and implement the time zone system that we use today.