Appliance Clock Anomaly
October 29, 2010 11:46 AM   Subscribe

Why would three digital clocks lose time at exactly the same pace?

We have 7 different digital clocks in our house. The ones on the stove, the microwave and the thermostat each lose one minute every week. It just seems odd.

Would this have something to do with the electricity going to the kitchen and the dining room?
posted by Mr. Yuck to Technology (10 answers total) 2 users marked this as a favorite
What are you comparing them to?
posted by chesty_a_arthur at 11:50 AM on October 29, 2010

They're probably using the 60Hz power supply to measure time. It's reasonably common.
posted by Mwongozi at 11:51 AM on October 29, 2010 [2 favorites]

Response by poster: I'm comparing them to our computers, the coffeemaker, an alarm clock, and a couple of Bose receivers. All of which keep proper time.

Do none of those devices use the 60Hz cycle?
posted by Mr. Yuck at 11:55 AM on October 29, 2010

Best answer: Some digital clocks rely on the frequency of the alternating current supply (60Hz, in the US) to keep time. Others have an internal crystal oscillator and do not rely on the AC frequency to keep time. It's possible your AC frequency is slightly off, and only the three clocks which lose time base their timekeeping on the AC frequency, while the ones that do not have an internal oscillator.

On preview: if your computer is regularly internet-connected, it probably checks the time against time servers and adjusts as necessary.
posted by DevilsAdvocate at 11:57 AM on October 29, 2010

Yes, it's probably caused by your power. As mwongozi says, they most likely use the (supposed to be) 60Hz cycle of your electricity as a timing device. Could be noisy power lines, could be something causing noise on the line within your home. Short of an oscilloscope and maybe an electrician, hard to track this one down.
posted by empyrean at 11:58 AM on October 29, 2010

Best answer: Most digital electronic devices do not use AC line frequency as their time source. They use their own internal crystal oscillators, which are typically accurate at best to no more than about half a second per day (for example a quartz wristwatch). Cheaper electronic devices will use cheaper oscillators that have much less accuracy.

On the other hand, AC line frequency is regulated by law to be extremely accurate for clock purposes and is synchronized to the National Bureau of Standards clocks. While line frequency can deviate by several seconds for short periods throughout the day due to load generation factors, over longer periods adjustments are made to bring line frequency back into synchronization with the Bureau clocks so that over days, weeks, or years, the exact required number of cycles occur. Errors are not permitted to accumulate. Synchronous AC clocks are more accurate than any common digital clock over long periods of time (months to years). So the $10 kitchen wall clock plugged into AC might be more accurate than your $1000 chronometer wristwatch.

If three out of seven of your digital clocks drift by about the same amount, it is most likely coincidence.
posted by JackFlash at 12:30 PM on October 29, 2010 [8 favorites]

Response by poster: Thanks!
posted by Mr. Yuck at 12:45 PM on October 29, 2010

To remove any ambiguity about the precise time, call the NIST (formerly the National Bureau of Standards) time line at 303 499 -7111.
posted by Kevin S at 2:18 PM on October 29, 2010

You are not alone. Scientist Christiaan Huygens had the same problem after he invented the pendulum clock (1657).
posted by exphysicist345 at 10:59 PM on October 29, 2010

Wow, JackFlash, I did not know that. I'm used to AC power generation shipboard, where the electrical operator sets 60 Hz by eyeball on a meter. As such, it may be a bit more or a bit less, or even vary. He'll sometimes want to bring frequency up a bit on one bus to pick up load from another machine running in parallel. An AC synchronous clock, while "good enough for government work" with a manual adjustment once in a while, wouldn't be guaranteed squat.

So you're saying that if they run at 60.2 Hz for whatever reason for a few minutes, they're required to keep track of that and run below 60 Hz for long enough to compensate? Sounds like a accounting nightmare. So, if I may piggyback on the original question - why would they bother? Just for AC synchronous clocks? (reference links ok, no derail intended.)
posted by ctmf at 10:20 AM on October 30, 2010

« Older How to be supportive of my SO with job offer...   |   He got the funk Newer »
This thread is closed to new comments.