[time-nuts] 60 Hz measurement party

[Alberto di Bene]

On 6/26/2011 6:03 PM, Tom Van Baak wrote:
> Alberto,
>
> That's a cute plot. I see 53 x 50 = 2650. What is the x axis scale?
Tom,

    that capture was taken a few months ago, from memory it should cover about 15 or 20 minutes.
> But I'm curious -- the pending issue with 60 Hz in the US is more about
> uncorrected long-term accumulated time drift and less about short-term
> frequency stability.
>
> So it seems like you would have to integrate all the wiggles in the 53rd
> harmonic of your waterfall plot in order to compute time drift, yes?
Yes, if what you are interested in is the potential drift of a line-controlled clock, just integrating
the signed deltas from the nominal value will give that info. In any case, I have near to my bed
such a clock. Thanks God the power line failures are very rare where I live (let's say once every
three months), and that clock keeps admirably the time. I would guess just a few seconds per month.
It doesn't have a crystal oscillator inside, it relies just on the 50 Hz.
> Is there any chance you could run both the waterfall frequency monitor
> and also a 50 Hz phase comparator for a couple of days to see how
> well (or how poorly) one can compute the net long-term mains time drift
> from a set of independent short-term frequency deviation measurements?
I don't have a 50 Hz phase comparator, but I can arrange for a long waterfall capture, a day long,
if that can be of help.

73  Alberto  I2PHD