10

I like to build a device that needs long term time stability. In germany, one of the best simple clocks use the mains power frequency to obtain a time measure. While the frequency is allowed to change a little with the power net load, the errors are nullified over longer periods, usually over night. So in the long term, a mains driven clock is actually atomic clock driven and very precise.

This applies to the whole european power network I guess.

So I wonder how about other countrys worldwide? Would it be feasable to use mains frequency for timing issues after adjusting to a country's official frequency?

Li-aung Yip
  • 8,931
  • 27
  • 50
dronus
  • 551
  • 3
  • 15
  • http://wwwhome.cs.utwente.nl/~ptdeboer/misc/mains.html – jippie Sep 27 '13 at 16:17
  • Your easiest and probably cheapest solution is to use a GPS receiver with a 1pps or static frequency output, like this one: https://buy.garmin.com/en-US/US/oem/sensors-and-boards/gps-18x-oem-/prod27594.html – jippie Sep 27 '13 at 16:19
  • 1
    A GPS receiver would need GPS radio signal, which is not available at all locations (eg. indoors). – dronus Oct 04 '13 at 13:29
  • I now decided for an temperature compensated crystal clock. Just to many questions on stability still unanswered when it comes to unknown destinations. My feeling is a mains driven clock that detects for the presence of 50 or 60 hz would do well worldwide, but no clear picture if this holds in practice.. – dronus Oct 17 '13 at 22:20
  • Maybe some day someone answers with a comprehensive study here and a result of yes or no, wether it would be reliable... – dronus Oct 17 '13 at 22:21
  • 1
    I was in Mongolia the other day, a huge country with small villages every here and there. Maybe they're better called 'camps'. They all have their own diesel generator and frequency error will not be even close to zero. I don't know about the grid in Ulan Bator ( the country's capital) though. – jippie Jan 17 '15 at 06:10

2 Answers2

4

In Australia, the interconnected grid frequency is held to a nominal frequency of 50 Hz, to be within ± 0.15 Hz 99% of the time. The accumulated time error may not be more than five seconds.

Excerpt from the AEMO Frequency Report Reference Guide:

enter image description here

The various Australian power utilities do a pretty good job of maintaining the target frequency and accumulated time error within the stated limits. AEMO publishes reports on this monthly - for the month of June 2013 the accumulated time error in Queensland varied between + 2.40 and -3.77 seconds.

enter image description here

From this, we can conclude that the 50 Hz power system frequency is an OK clock signal - but only so long as your time-keeping device has uninterrupted access to the power grid!

Here is a list of conditions that may cause you to lose your mains power, and therefore your time reference:

  • Someone turns off the mains power switch to your device
  • A circuit breaker trips in your house
  • The power goes out on your street
  • A transient fault occurs on the distribution network - enough to result in a brief 200ms auto-reclosing cycle
  • A major fault occurs on the transmission network - the grid goes black in your area

You would almost certainly be better off finding some other way to keep time. A GPS time receiver (ref. "IRIG B") is common in applications where precise time keeping is required, such as power systems protection relaying.

Li-aung Yip
  • 8,931
  • 27
  • 50
  • Nice one, someone at the local utility company told me things had been relaxed here in Tasmania but I didn't have a reference for it before. – PeterJ Sep 27 '13 at 22:49
  • @PeterJ: If you follow the first link, the document does include a separate set of tables for Tasmania. The requirements are more lax because you have a smaller power system - there's less rotating inertia to keep things steady. (I am being entirely serious.) – Li-aung Yip Sep 27 '13 at 22:51
2

In North America, whenever the error exceeds 10 seconds for the east, 3 seconds for Texas, or 2 seconds for the west, a correction of ±0.02 Hz (0.033%) is applied. Time error corrections start and end either on the hour or on the half hour.

This quote was taken from here - scroll down for the heading "Long-term stability and clock synchronization". You might also be interested in this question:

Community
  • 1
Andy aka
  • 434,556
  • 28
  • 351
  • 777