I always wondered how much time means 1 second?

Technically, the SI second is defined as

the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the caesium 133 atom.

(Source: Wikipedia.)

But how do watch makers set the second time in their watches? I don't think they use the above process. Do they actually implement the SI second? If not, what is the relation between the second as doled out by a quartz watch and the SI second?

  • 1
    $\begingroup$ en.wikipedia.org/wiki/Quartz_clock $\endgroup$
    – Edvard
    May 14, 2014 at 9:52
  • $\begingroup$ Watch makers use reference clocks to check the accuracy of the watches they make. The makers of reference clocks do ultimately use atomic clocks. $\endgroup$ May 14, 2014 at 10:01
  • $\begingroup$ Thanks, but how do they set the speed of rotation of second hand of clock or how they measure time , if they don't use quartz ? How they manage before quartz was invented ? $\endgroup$
    – bnil
    May 14, 2014 at 10:04
  • $\begingroup$ @RedGrittyBrick I think there are three or four levels of references for basic units. (So also reference clocks for reference clocks, etc), most countries have official institutions to regulate this, as far as I know $\endgroup$
    – Bernhard
    May 14, 2014 at 10:08
  • 1
    $\begingroup$ @user1650894: they used hour-glasses, pendulums, escapements and other mechanical devices. So far this isn't a physics question. I suggest reading Wikipedia starting at Clock and following all links to a depth of (say) 3. :-). Then come back here for any Qs about physics. $\endgroup$ May 14, 2014 at 10:14

1 Answer 1


Clocks, in general, can be fashioned out of any oscillatory phenomenon as long as it is regular enough. Initially these were mechanical: pendulums and springs, and so on. For quartz watches, this is still the case, and the oscillator is a tuning-fork-shaped quartz crystal that vibrates mechanically, though this vibration is read by an electronic circuit.

Regardless of the specific implementation, any clock (be it a pendulum clock, a quartz watch, or a caesium atomic clock) is simply an oscillatory system with a bunch of 'readout' apparatus around it. This oscillatory system will have a base frequency but it will also deviate from it to some extent. (And yes, this includes atomic clocks, because of things like magnetic noise.) One simply accepts these deviations, in the understanding that they set the limit to which the clock can be trusted. For quartz watches this isn't great, as they can slip by a few seconds every month or so. For atomic clocks this is more like one second in 300 million years, but it is still finite and it still means there are things one might want to do that are not yet possible because we do not yet have accurate enough clocks.

Finally, there is the issue of calibration. Once one has an oscillatory system, one needs to adjust the frequency to match, as well as possible, all the other clocks around. To do this, watch factories will have stable reference clocks (which you can assume are atomic clocks) which they use to calibrate their resonators. If they're too fast or too slow they are trimmed down to bring them to speed, and this is done until they are as accurate as their finite precision allows them to be.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.