[time-nuts] Accuracy of 10mhz - a simple explanation?

[John Ackermann N8UR]

N3IZN at aol.com said the following on 01/28/2007 06:11 PM:
> Accuracy of 10mhz - a simple explanation?
>  
> So can any one provide a link or want to try and explain to this layman  what 
> the deal is with the accuracy spec provided by (or should I say claimed by)  
> the various manufactures and kits out there?
>  
> For example my Trimble claims to have accuracy of 1.16 X 10 -12.
>  
> I'm guessing that the exponent at the end of the spec, the higher the  better?
>  
> Just how high can you get? How high is needed to say lock a amateur  
> transverter to for 10 or 24 ghz?

That kind of notation is called "fractional frequency offset" and it
means that the error will be less than 1.16x10e-12 Hertz (1.16
trillionths if I did the math correctly) at a nominal frequency of 1
Hertz; that error scales with frequency, so if you multiply up, that
means that the signal will be 1.16 Hz off if the carrier frequency is 1
THz (1,000 GHz).

An easy set of "road markers" is that 1x10e-6 is 1 Hertz at 1 MHz, and
1x10e-10 is 1 Hertz at 10 GHz.

There are two main kinds of measurement that the time-nuts talk care
about: frequency offset or error, and frequency stability.  Measuring
stability is where most of the fun is...

Hope this helps.

73,
John