[time-nuts] Mains frequency

[Tim Shoppa]

Filter families that aim for super flat passbands and super steep skirts
have the craziest phase response. They also "ring like a bell" when hit
with impulses.

Other filter families aim for well controlled phase response but give up
flat passbands and super steep skirts. They do not "ring like a bell" when
hit with impulses.

Gaussian filters work very well for well controlled phase response and are
used in e.g. oscilliscopes and other applications where the data is eyed
critically in the time domain.

In communications and time-domain work, it is common to use a filter that
has well controlled phase delay across the passband, but still has somewhat
steep skirts and can have less well controlled phase delay outside the
passband. In the literature these are called "Gaussian to 12dB" and
similar. They are a joy to listen with, in environments with impulse noise,
compared to the typical steep-skirt filters.

Tim N3QE

Tim.



On Tue, Nov 19, 2013 at 10:06 AM, Didier Juges <shalimr9 at gmail.com> wrote:

> "Alternatively you can put mains through a 40-60 or 50-70 Hz
> bandpass filter to suppress anything but the fundamental"
>
> The problem with any filtering is precisely that the phase shift through
> any filter is highly dependent on the actual signal frequency, so if your
> purpose is timing, the filter will probably do more damage than help. Tom's
> data shows that "no filter" works well at his location, which is probably
> typical, aside from the large number of precision time references in the
> basement :)
>
> I have some data about filtering the mains I am trying to put together in
> a web page. Will take a few days. That data was collected with regard to
> power factor correction, not timing, but it is within the scope of what has
> been discussed in this thread and it may be interesting to the curious mind.
>
> Didier KO4BB
>
>
> Poul-Henning Kamp <phk at phk.freebsd.dk> wrote:
> >In message <77A742979E2849FD9CA585192F2F5AFC at pc52>, "Tom Van Baak"
> >writes:
> >
> >>For tau 0.01 to 0.3 s it would take more work to quantify which pieces
> >are
> >>contributing to the instability pie.
> >
> >It doesn't really make sense to talk about taus less than approx
> >0.1s for Mains.
> >
> >It is not atypical to see 10% "harmonic distortion" on mains in a
> >household, most of it from switchmodes, but also single-phase
> >motors etc.  Music amplifiers, in particular played at "teenagerockgod"
> >volumes add a very interesting challenge too.
> >
> >In theory harmonic distortion does not change your zero-crossings,
> >but it is not really "harmonic" in the Fourier sense, it's just
> >mostly centered around the overtones.
> >
> >Either you include the "harmonic" distortion, and then your measurement
> >applies only to that specific outlet, when the dish-washer is running
> >exactly 20 minutes into programme 3 and no kids listen to loud
> >music.
> >
> >Alternatively you can put mains through a 40-60 or 50-70 Hz
> >bandpassfilter to supress anything but the the fundamental, but
> >what exactly do you measure then ?
> >
> >Once you get below about 1Hz, the most interesting thing you can
> >do with mains is to sample it at ~4kHz, make a water-fall plot and
> >try to identify your house-hold appliances by their distortion
> >patterns :-)
> >
> >--
> >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.
> >_______________________________________________
> >time-nuts mailing list -- time-nuts at febo.com
> >To unsubscribe, go to
> >https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> >and follow the instructions there.
>
> --
> Sent from my Motorola Droid Razr 4G LTE wireless tracker while I do other
> things.
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>