[time-nuts] Is my Cs C-Field set OK or do I need to take more data?

[Poul-Henning Kamp]

In message <41E1D768.1090305 at pacific.net>, Brooke Clarke writes:
>Hi:
>
>For the last few days I've been tweaking the C Field setting on a FTS 4060/S24.  Here's some data taken using the 1 PPS output of a GPS disciplined Rb oscillator (PRS10) as the "A" trigger input to an SR 620 TI counter.  The "B" stop input is the 1 MHz output from the Cs source.
>http://www.pacificsites.com/~brooke/pdf/sn1013_C-Field.pdf
>
>The plot Y-zxis is ns of Time Interval and the X-axis is seconds of run time.
>
>A linear fit line has the equation
>ns of TI = 0.0059X + 30.283.
>Since this is ns/sec the drift rate of the Cs is 5.9E-12. 
>
>So when measuring against a GPS based reference is is fair to say
>that you need to measure until the TI gets above something like 200
>ns or the elapsed time exceeds some number? Or do I need to take
>more data before I change the C-Field setting?

"As you run your experiment you will get to a point where you are confident
in your result and conclusion and you will be tempted to stop it.  You
should run at least twice and preferably three times as long in order to
convince the rest of us"

				-- my physics prof


-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.