Re: [time-nuts] More 60 Hz graphs

[Poul-Henning Kamp]

In message <20110916001417.d49d3800...@ip-64-139-1-69.sjc.megapath.net>, Hal Mu
rray writes:

>Here is the frequency:
>  http://www.megapathdsl.net/~hmurray/time-nuts/ppstest/60Hz-freq.png
>I don't know where the wobbles are coming from.  They might be in my 
>collection setup.

They are very likely resonance-effects in the power grid.  Power-grids
are not inherently stable things, and most of them have resonance
frequencies like that, depending on which generators run and what
the impedance are between them.

A few grids have operational SPICE-like models to warn them about
configurations which are unstable, but in general the thinking that
it would be possible to build an analytical model of the grid died
with decentral productions uptake a decade ago.

Try running a suitably windowed FFT on your data...

-- 
Poul-Henning Kamp   | UNIX since Zilog Zeus 3.20
p...@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.

___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Re: [time-nuts] More 60 Hz graphs

[Tom Van Baak]

Hal,

Now that you're mastered sampling every cycle, it's a good time
to check how much of the noise or instability you see is due to
the h/w and s/w you're using vs. how much is due to the power
line itself.

Do you have a hp 3325 or similar synthesizer handy? Instead
of using 60 Hz from the power line, feed in a known stable or
accurate 60 Hz pulse source into the DCD pin and then see
how your plots look in comparison. It's an easy experiment to
do. The results will tell you how good/bad your PC-based time
interval counter really is.

Note it doesn't have to be 60 Hz; anything between say 10 and
100 Hz should work fine. By using a known stable and artificial
input signal all glitches and instability you see in the plots are
due to the sum total of h/w and s/w you're using. In effect the
PC is measuring its own noise without knowing it.

/tvb



___
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.