[time-nuts] Re: Has Line-Frequency Timekeeping Become Less Reliable in the U.S.?

[Jay Hirschfeld]

Commercial versions of this exist. KCC Scientific makes
crystal-stabilized precision AC frequency standards (their
Hertz/Cronos/Mercury line) aimed at vintage synchronous clocks and
turntables, with selectable 50/60 Hz output.

One wrinkle on "GPSDO divided down to 60 Hz": 10 MHz to 60 Hz isn't a
clean integer chain. 10^7/60 leaves a factor of 3, so a pure
decade/binary divider string won't get there. Two easy routes: a
4046-style PLL with the odd factor in the feedback divider, or a small
DDS (an AD9833 clocked from the 10 MHz reference gets you 60.000000 Hz
with sub-uHz resolution for about $5). After that it's an audio amp
and a step-up transformer. A synchronous clock motor only wants a watt
or two.

This thread is a nice snapshot of a broader, somewhat-concerning
pattern: the free public time-distribution channels are being retired
one by one (TEC now, LORAN-C years ago, WWVB perennially on the budget
chopping block) while everything downstream quietly consolidates onto
GNSS. My hedge against that has turned into a bench project that's
gotten entirely out of hand: a surplus FE-5680A rubidium disciplined
by GNSS, with a 74HC14/74LS90 divider chain generating 1 PPS into
chrony, headed toward a public NTS server. The same 10 MHz reference
driving a DDS would make a grid-independent 60 Hz source that keeps a
synchronous clock honest straight through an outage. Longer term the
plan is for several of these nodes to compare time against each other
over the network, so any one receiver being wrong (or lied to) shows
up as an inconsistency rather than propagating. I'm happy to share the
divider schematic if anyone wants to build another one; a perfboard
version is in progress and a proper PCB is planned.

-- 
Jay
https://clockmesh.com

On Thu, Jul 16, 2026 at 12:38 PM Bob kb8tq via time-nuts
<time-nuts at lists.febo.com> wrote:
>
> Hi
>
> There are still a multitude of mechanical “plug in the wall” timers out there that do 24 hour
> or longer device control stuff. Most of it is pretty mundane gear. Most of the timing isn’t all that
> precise. Still, they are out there.
>
> Bob
>
> > On Jul 15, 2026, at 6:06 PM, Jim Lux via time-nuts <time-nuts at lists.febo.com> wrote:
> >
> >
> >
> >
> >
> >
> > I would think that ovens, appliances, and the like don't count cycles any more.  They all have microcontrollers in them, and they'd count oscillator cycles.
> > Even my 20 year old bedside clock doesn't use line frequency (because it has battery backup, it can't).
> >
> > Yes, my grandmother's old stove with the clock built in, manufactured some time pre-1960, certainly is line frequency dependent.
> > A washing machine with a mechanical timer perhaps.  But the old microwave oven I have in the garage that has a "turn knob to set cooking time" uses a *mechanical* spring wound timer.
> >
> > Is there anything that uses line frequency as a time standard any more (other than weird stuff like PDP-11's using line frequency for RTC input)
> >
> > Older timeclocks controlling lighting (the ones with the pegs to turn and off) probably use synchronous motors. Same with the household timers that one might use for holiday lights.
> >
> > I note that in a new installation for lighting controls, almost invariably, for exterior and sign lighting they're required to be "astronomical" timers that know when sunrise and sunset are. (CA TItle 24, ASHRAE 90.1) (or photodetectors). And for interiors, they're required to be able to automatically deal with holidays and daylight saving time. That pretty much means the old mechanical timers have limited uses.
> >
> >
> >
> >
> > On Wed, 15 Jul 2026 15:15:10 +0000, "Thomas D. Erb via time-nuts" <time-nuts at lists.febo.com> wrote:
> >
> > I recently came across information suggesting that the way the North American power grid handles accumulated time error may be changing, and I had not realized this was even being discussed.
> >
> > Historically, synchronous clocks-and some electronic appliance clocks that count the 60 Hz AC waveform-benefited from long-term grid-frequency correction. Although the instantaneous frequency varied slightly, accumulated error was corrected so that these clocks remained reasonably accurate over time.
> >
> > Does anyone here know the current status of Time Error Correction on the different North American interconnections?
> >
> > More specifically, are traditional synchronous clocks now likely to accumulate noticeable long-term error? And what about microwave ovens, ranges and other digital clocks that may use AC zero crossings rather than a crystal or real-time clock for their time base?
> >
> > I initially saw a reference to an April 14, 2026 NERC/FERC action, but on looking more closely, that appears to concern retirement of BAL-002-WECC-3, a contingency-reserve standard-not BAL-004 time-error correction. NERC still lists BAL-004-WECC-4, Automatic Time Error Correction, as active in the Western Interconnection.
> >
> > I would be interested to hear from anyone familiar with current utility frequency-control practices, particularly in the Eastern Interconnection.
> >
> >
> > Thomas D. Erb
> > o: 508-359-9684
> > p: 508-359-4396 x 1700
> > f: 508-359-4482
> > a: 97 West Street, Medfield, MA 02052 USA
> > e: tde at electrictime.com
> > w: www.electrictime.com
> > Tower & Street Clocks Since 1928
> >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe send an email to time-nuts-leave at lists.febo.com
> >
> >
> >
> >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe send an email to time-nuts-leave at lists.febo.com
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe send an email to time-nuts-leave at lists.febo.com