New atomic clocks will make GPS even more accurate

[BajaTaco]

A news story on the BBC reports of the launch of a new satellite, "Grove-B" which has a hydrogen maser atomic clock on board. Apparently, the most accurate clock ever built.

"Following its launch from the Baikonaur Cosmodrome in Kazakhstan, it will become the most precise time piece to orbit the Earth. It will be accurate to one billionth of a second per day, or one second in three million years. By comparison, a typical wristwatch is accurate to about one second per day."

In the Fall 2007 issue of Overland Journal, Tom Sheppard has an article summarizing the use of super-clocks in space to provide our GPS units with an accurate fix. While accuracy has improved substantially in recent years, it may become much better with the Grove-B.

"If the technology is shown to be successful, it will be built into all 30 of Galileo's operational satellites, eventually allowing users to pinpoint their location with an error of just one metre, compared to the several metres experienced with current GPS technology."

Very cool! Hats off to the individuals with the brain power and talent to invent these things :REExeSquatsHL1:

 

[Schattenjager]

That is too cool! I just got a G-shock solar / atomic watch and have become completely geeked out over the whole "atomic clock" thing.

Hopefully GPS's will be able to take advantage of this soon, but I have to think it will be a while - upgrading satellites can't be an inexpensive proposition.

Thanks for the post!

 

[DaveInDenver]

Some additional information.

For GALILEO they are still flying two traditional atomic clocks and the maser clock is an experiment. The TKS (time keeping system) on the NAVSTAR Block IIR vehicles is already based on a rubidium atomic clock. These are pretty close to the same level of accuracy, around an Allan frequency variation of about 10^14 per 24 hours period and on the vehicles are doing about one part per 10^13th over a 6 hour period in testing. We've been flying them since 1997.

The clock itself is generally not the root of errors, but rather drift of the VCXO (voltage-controlled, crystal oscillator) in the PLL (phase locked loop). The supervisory system will generate non-standard messages (telling your receiver to ignore it) if it detects more than 16 ns (a nanosecond is 1 billionth of second) of error between epochs, which are 1.5 seconds apart. IOW, the bulk of our GPS system is considered to be in error if it measures more than ~10 billionth of a second of message time error. It's considered phase corrected when the time returns to less than 2 ns skew between epochs. All of this is already at least two orders of magnitude less accurate than the atomic clock itself can do.

But all that really doesn't matter. In real world observation, multi-path receiver errors as vehicles move through your view have a much great effect on received time accuracy than the clock or PLL itself.

BTW, we (as in the U.S.) send up a new NAVSTAR satellite at the rate of about 2 or 3 per year on average. Some years it's only 1 and some as many as 5. The lifespan of our first Block II and Block IIA vehicles is about 7.3 years and the Block IIR (current ones) is about 7.8 years. The Block IIF (the new design) is supposed to be 10+ years, but we don't have any in orbit yet and so that's still only projected. We've launched (IIRC) 48 Block II vehicles so far.