[time-nuts] Overheard from NASA

Jim Lux jimlux at earthlink.net
Tue May 10 00:21:53 UTC 2011


On 5/9/11 8:25 AM, William H. Fite wrote:
> Overheard from a senior NASA research metrologist:
>
> "The only reason we're doing it is because we *can* (improving clock
> accuracy, said in the context of the aluminum clock).  We can already time
> so accurately, just as an example, that if we launched a spacecraft today
> toward Sirius we could predict its location when the craft arrived many
> thousands of years from now, to within a thousand miles or so."
>
> That's not a precise quote but it is a close paraphrase.
>
> Heck, I thought that was why time nuts did it, anyway.
>


When it comes to good clocks on spacecraft, we're a long way away from 
"better than we need", particularly for small power/mass/volume.

Having a atomic clock on board would let you do things like one-way 
ranging, particularly techniques such as delta DOR, which can give you 
"cross range" measurements (i.e. azimuth).

Knowing the position to 1000s of km may not be particularly useful, even 
at long distances, but as a practical matter, we want to know distances 
to cm or mm at Jupiter or Saturn distances.

Given that Jupiter is about 600-800E9 meters away (call it a round 1E12 
meters), that's a precision of 1 part in, say, 1E14.

We use precise measurements of range rate (on the order of mm/s) to 
determine the gravity field, and from that the internal structure of a 
planet.  The Juno spacecraft has a coherent transponder that contributes 
Allan deviation of around 1E-15 or 1E-16 over 1000 seconds, with the 
rest of the measurement system (transmitter on earth, receiver on earth, 
propagation uncertainty at 32/34 GHz) contributing roughly comparable 
amounts.

The transponder (KaTS) receives a signal at 34 GHz from earth at a 
fairly low SNR and generates a carrier at 32 GHz with a fixed ratio of 
phase/frequency to transmit back.  The SNR is limited by the power we 
can transmit on Earth (tens of kW, with BIG antenna gain) and the size 
of the antenna on Juno.

IF we had a "good" clock on board, we wouldn't need to worry about the 
"transmitter on earth" and "one way propagation uncertainty" for the 
outbound path.

A USO (quartz oscillator in a temperature controlled dewar) isn't in 
this class of performance (and is big and power hungry to boot).


If you had a good onboard oscillator, you can do VLBI type measurements 
to measure not only range, but angle to a higher precision than is 
currently possible.




More information about the time-nuts mailing list