WWVB Celebrates 50 Years of Broadcasting Time 97
First time accepted submitter doublebackslash writes "On July 5th, WWVB, NIST's timekeeping radio station transmitting near Fort Collins, will celebrate 50 years of continuous operation. Operating at 60kHz, the signal actually follows the curvature of the Earth via a trick of electromagnetics, allowing nearly the entire globe to receive an accurate time signal, which has in recent years reached an accuracy of 1 part in 70 trillion. Recent upgrades, which came in $15.9 million under budget will allow the station to be better received even in large buildings, giving it an edge on timekeeping that not even GPS can touch, with its need for open skies to receive a signal."
I implemented a teensy WWCB transmitter once (Score:5, Interesting)
Some 15 years ago, when they were at their original low power, my area was so fringe that my fancy new WWVB wristwatch just wouldn't pick it up.
The protocol is really quite straightforward and well documented at their site. The 60kHz signal sends binary by sending either full power or a bit less (I forget how many dB). I used a computer synced with NTP and a plain old soundcard generating 60kHz from a sound card into an audio amp, and I just did either full on or full off. The output ran into a big coil that I had wound to be roughly resonant around 60kHz.
Much to my amazement, it worked. So I just kept the watch near that coil overnight and it synced perfectly, until WWVB cranked up their power at which point I retired the mess.
Re:Nearly the entire globe- except not really (Score:2, Interesting)
The problem I have is there's no error correction/checksum -- the signal is binary-digit encoded, so it's possible to get a clock that think it's synced but is off by X hours or X minutes or X seconds while otherwise thinking that all is well. This isn't a big problem on smarter systems that continuously monitor the signal but most wall clocks/etc. only check from time to time so they can spend 12 hours thinking it's another time entirely. This isn't incredibly common because the signal is 1 bit per second, so it's pretty robust even at low strength, but it does happen.
The German version of radio timekeeping includes a checksum to avoid these issues. Then again, their version requires a lot more math to calculate the date, and only uniquely identifies dates within like a 200-year range (which sounds like a long time, but it short enough that you can get into trouble with hard-coded assumptions).
Re:Accuracy... (Score:2, Interesting)
The broadcast signal is that accurate. And even the received signal is that precise, given a stationary receiver. So baring multi-path issues (which are detectable) you can do a one-time calibration for your local propagation delay to make your local accuracy match your precision.
Moreover, they also provide a frequency reference, in addition to the time signal, so you can do your time timekeeping just by counting peaks. Such references are immune to propagation delay (though not to Doppler effects, if your receiver is mobile) as they do not encode specific point-in-time data.
Re:Accuracy... (Score:4, Interesting)
You do have a point.
Given that light travels about 0.3 m in one nanosecond, a variation in the signal path-length of about 300 m would induce a smudge on the arrival-times of about a microsecond. Realistic path-length variations could no doubt be larger, and could vary over a time-period of minutes or hours, depending on ionospheric conditions. This would of course be much larger than the inferred time-accuracy of 1.14e-14 s in the single second between each broadcast 'tick'.
However, GPS [wikipedia.org] is subject to the same vagaries of ionospheric conditions, as well as error in signal-interpretation. It has a theoretical accuracy of 14 ns, but more typically it is 100 ns.
No doubt the received accuracy of both WWVB and GPS could be improved by frequently collecting and applying the appropriate ephemera corrections for a given geographical location.
Software Defined Radio & WWVB? (Score:2, Interesting)
So anyone here tried using a software defined radio setup to receive and decode the WWVB signal?
Re:I implemented a teensy WWCB transmitter once (Score:4, Interesting)
This one could, and I don't claim to know why. But I saw it clearly on my oscilloscope: 60kHz.
Actually it wasn't exactly 60kHz, it was 59 point something because of quantization according to a frequency counter, but apparently it was close enough to keep the watch happy.
Re:Accuracy... (Score:4, Interesting)
For some time I plotted the jitter of reception of DCF-77 (a similar transmitter in Germany) and I found there was a clear cycle of increase and
decrease of the jitter of the pulses output by my receiver (measured over one minute) over the day.
At daytime the jitter is around 20us, at nighttime it is more like 200us.
This is most likely explained by path length variations that apparently are depending on propagation.
(although texts about such transmitters often boast that there is no propagation effect like the one seen at shortware at those frequencies)
The claimed accuracy is of course at the source, and maybe when you started receiving WWVB years ago and perform some kind of averaging
over a long interval, you could eventually get an accuracy like that, but there is no way it can be achieved over short intervals, let alone for
individual second pulses.