I think this is what you want, but unfortunately, I don't know the price:

https://timetoolsltd.com/products/synchronized-network-clocks/analog-poe-ntp-ip-clock/

Rick

Sent from ProtonMail, Swiss-based encrypted email.

------- Original Message -------
On Tuesday, December 13th, 2022 at 2:09 PM, Bob Camp via time-nuts time-nuts@lists.febo.com wrote:

Hi Bob,

Seiko Corporation has produced a range of GPS receiving clocks under its “Space Link” name for several years. They have also a produced a range of GPS synchronised quartz wristwatches under their “Astron” name for more than ten years.

See www.seiko-clock.co.jp/product-personal/wall_clock/satellite/

And Silicon Chip Magazine, a fantastic Australian electronics hobby magazine, has published several GPS time projects over the past 7 years or so. Their most recent project in September details modifying a regular quartz clock to use a GPS receiver. You can find a preview of that project and purchase a kit of the parts at:

www.siliconchip.com.au/Issue/2022/September/New+GPS-Synchronised+Analog+Clock

Hope these help you find what you’re looking for.

Roger.

I have a GPS 'synced' analog clock.  I'm using an old synchronous
motor clock and driving it with 90vac generated by an old synthesizer
clocked off my house 10MHz feeding a bipolar amplifier.  I think this
is essentially the same setup as TVB mentions at
http://leapsecond.com/pages/atomic-nixie/

Here's the clock in my home office:

https://nt4tn.net/clock/clock.jpg

Here is the equipment on the other side of the wall supporting it.

https://nt4tn.net/clock/equipment.jpg

Of course, it only sets the frequency frequency not time... but you
can just set the time of the clock via mk1 eyeball to whatever
precision you like and once its set it keeps time.

I've also had fun running the clock on mars time when the rovers were
the current thing and (more often) sidereal time for astronomy.

I'm not sure I'd recommend this approach in that it draws a lot of
power and ties up otherwise more flexible equipment (and noisy).  I
put it into place expecting to replace it with some dedicated 'fake
mains'  (like... running a wall wart transformer in reverse from a
dedicated little synthesizer and an audio amplifier) but have never
gotten around to it.  The power usage is an actual issue too since it
gets in the way of providing long lasting backup power.

One great thing about using synchronous motor clocks and a fake mains
is that you could easily support a bunch of clock this way and there
are many interesting synchronous motor clocks that exist.

On Thu, Dec 15, 2022 at 5:10 AM Bob Camp via time-nuts
time-nuts@lists.febo.com wrote:

Relatively easy DIY project.  Most battery operated wall clock movements are driven
by a two phase motor - positive pulse followed by negative pulse.  Many GPS modules
provide a single phase pps output.  The problem is to generate the bi phase from the
single phase pps.  I have a Brooks Shere system running that has a 'heart beat' signal
available (one second on, one second off).  If that signal is applied thru a series capacitor
(about 47-100 microF) it will be differentiated to provide the bi phase pulses.
Remove the battery from the clock.  Locate the motor winding and apply the bi phase signal.
I have had a couple of wall clocks driven by GPS for years.  It only missis a beat if the GPS goes down.

Alternate solution is to use a micro processor on a chip (about $2.50) to generate bi phase
from pps.  That takes about 4-6 lines of program, depending on the chip.

Regards.
Bill, NL7F

On Tue, 13 Dec 2022 15:09:26 -0500, Bob Camp via time-nuts wrote:

Bill Beam
NL7F

Bob-

They don't hang on the wall or have mechanical movements but...
Here's one of my favorite semi analog clocks I have.
http://badnixie.com/Black_Emerald_Nixie.html

I have two modsix clocks running off the same GPS receiver wirelessly.  It
outputs 4/6 letter WORD generator, selectable Time Sync out,  integrated
GPS, Barometric Pressure, Humidity, Temp , and real  GPS coordinates.

You can see them in one of my YouTube videos below.  It's the two with
alphanumeric Nixie tubes the B-7971 and B-8971 tubes from Boroughs.  I
consider them pretty analog as pure analog GPS receiver isn't something I'm
sure anyone has tried to build ie. no integrated circuits.  If someone has
I'd love to see it:

https://youtu.be/_p3gDBU524A

I consider these pretty analog and nuts.  170V and you're good to go plus I
love neon plasma too!

Best Regards,

Bill

On Wed, Dec 14, 2022, 10:11 PM Bob Camp via time-nuts <
time-nuts@lists.febo.com> wrote:

On Thu, Dec 15, 2022 at 5:09 AM Bob Camp via time-nuts <
time-nuts@lists.febo.com> wrote:

The only thing I can see a need to feed-back is the local time offset,
possibly including daylight savings time. So I think I'd use a conventional
crystal-based clock mechanism but replace the crystal with a 32768Hz source
locked to a GPS module in the same way as a GPSDO. It would also need a
reasonably good holdover oscillator as, assuming you want a self-contained
clock, the GPS signal indoors (once considered impossible) may be
inadequate in some locations.

So the time display would be completely under your control, but the tick
rate would be as accurate as you can get.

Hi

Indeed that gets me part of the way there. It at least moves the task from building a
clock over to feeding an existing gizmo the type of time I want to see.

What I want to display is actually the time that GPS runs on. The formal name appears
to be GPS System TIme. UTC without leap seconds is another common way of talking
about it.

Since you are off from the “rest of the world” by some number of seconds, you can’t just
sync up to a typical source and move on. Somehow you need to deal with the (variable
over long time spans) offset between GPSST and UTC.

The “why?” behind this is that I’m in the middle of a bunch of GPS based file this and
that. They all sync up to GPSST. Checking on them to see that all is well involves
working out when they should have done this or that. Having a clock to look at should
make that easier.

Bob

1PPS into a toggle flip-flop such as a 74AC74 w/ feedback (AC has strong
output drive, can run off 2V to 5.5V) followed by a differentiator
(could be just an appropriate capacitor) could generate the bipolar
pulses for a standard battery clock movement.

David N1HAC

On 12/15/22 2:59 AM, Bill Beam via time-nuts wrote:

Hello to the group.
Interesting subject. There is another approach to this and I and Rodger are
working it.
Thats to reuse broadcast television clocks like Evertz. They make wall
clocks, rackmount versions all driven by LTC timecode. These can be had on
ebay at times for reasonable cost.
How to drive these. The normal way is a gps locked broadcast generator.
However in progress and working is an Arduino gps locked LTC generator.
Uses a cheap ublox GPS receiver and arduino and literally 2-3 cmos chips
74HC series.
OK that nasty soldering iron.
But it is working well. Still reworking some of the code.
Though I have the magical broadcast generator it consumes 30 watts and
given the power cost in New England now the arduino solution is .5 watts
all in.
I will share with timenuts all of the details when ready.
Not expecting this to take long but ordered a LEA-6T from China should be
here any week/day now. Its better than the NEO-6. The LEA has an advantage
it can generate a very good 2 KHz. Eliminating a stable 2 KHz osc in the
solution today.
This uses some code from the open LTC generator. That arduino code has some
very serious issues that are eliminated by the few chips added to build a
proper LTC encoder.
Regards
Paul
WB8TSL

On Thu, Dec 15, 2022 at 9:53 AM Bill Notfaded via time-nuts <
time-nuts@lists.febo.com> wrote:

Does it need to be a mechanical clock or would a computer display do ?
I think Lady Heather will display GPS time rather than UTC. So a raspberry
pi board (even the very oldest spec, not one of the hard-to-find ones)
would be adequate.
I realise that's different to what you asked for, but it seems as though it
gives you what you want.

On Thu, Dec 15, 2022 at 3:25 PM David G. McGaw via time-nuts <
time-nuts@lists.febo.com> wrote:

As pointed out by other people already, there are several analog wall
clocks out there that can be set by external means like NTP, GPS, etc.
In my mind they all suffer from the same flaw, which is their
functionality is restricted to advancing time in precise increments but
don't "know the time". I.e. if power goes away and comes back, you'll
have to set such clocks by hand, even though their motor function can be
influenced. What is lacking is a set of 3 absolute position rotary
encoders with 6° resolution like described at [1] which referenced [2].
Somebody^{TM} should put such a clock together. 3 such encoders, a
stepper motor with gears to arrive at 6°, the rest is just software.

-Alex OE2AHL

[1] https://www.instructables.com/Absolute-Position-Encoder-With-Single-Track-Gray-C/
[2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6111776/

On Tue, Dec 13, 2022 at 03:09:26PM -0500, Bob Camp via time-nuts wrote:

Hi

One advantage of going straight from a GPS module to the clock is that
the setup should run through the next leap second with no operator intervention.
Having a setup that simply offsets UTC by X seconds and then feeds that
to the clock gets you into updates.

Since this is an “eyeball” measurement, anything better than milliseconds is plenty
good enough. My tired old eyes can’t detect a clock tick that is off by even 10’s
of milliseconds. Not a lot of need for sawtooth correction or fancy pps stuff.

Yes, this is a very different sort of clock and time requirement than most of what
we chat about. Running a clock at the “wrong time” will be interesting. Noticing
that it’s 18 or so seconds off should be pretty easy to do.

Bob

On 12/15/22 9:57 AM, Alexander Huemer via time-nuts wrote:

There are clocks that pulse forward with one polarity, and reset to "top
of the hour or day" with a negative pulse.  My middle school had them in
the 1970s, and it was interesting to watch the reset cycle (at noon). 
Or, to be honest, it might be some sort of "power switching" protocol.
That is, use 60 Hz to run, but some pattern of interruption sets it to a
known time.  But I remember watching the minute hand snap to 12, then
advance in discrete steps.

Some time ago I bought a NOS slave clock on eBay for about $25. It runs on
110V 60 Hz. It has a solenoid to reset time to 12:00 and all the clocks in
a building lijj my e a school would get the reset pulse at the time.

Well being ham fisted, I broke or burned out the reset solenoid, so no way
to set the time any more.

But I love this clock! So, I built an inverter and used a microcontroller
to generate a precise 60 Hz sine wave. I’m running it at 90 VAC and it’s
dead quiet. The second hand has continues movement. I wrote the code to
allow running the clock at different frequencies to simply setting the
time. E.g., running it at 59 Hz makes it lose 1 second per minute.

I get the frequency from a GPSDO. Once I set the time, it never drifts.

I spent a bunch of time fretting about power loss, and how to automatically
adjust the time afterwards, but realized I actually didn’t care because the
process of setting the time by running the clock at the wrong speed was
kinda fun.

I can’t begin to tell you how much time I’ve spent staring at this thing
while listening to WWV on the phone, and tuning it so the new minute begins
when the second hand is in the center of the black dot at the 12 position
on the dial.

I do have lights that blink if it loses power, or loses the 10 MHz
reference so I know I need to set it.

I was careful to make a very clean inverter as I didn’t want any harmonics
causing vibration in the synchronous motor.

I can post pics, but it’s just an institutional wall clock like you used to
see in schools or railway stations.

Neil

On Wed, Dec 14, 2022 at 9:11 PM Bob Camp via time-nuts <
time-nuts@lists.febo.com> wrote:

On Fri, Dec 16, 2022 at 12:49 AM Alexander Huemer via time-nuts <
time-nuts@lists.febo.com> wrote:

Yes, but mechanically is an expensive way to make a small memory. This is
why I suggested a nonvolatile ram/eeprom. It can be in the clock or in the
clock controller : the difference between those is that in the latter,
cabling faults cause memory loss.

The key point is that you align it once (at manufacture) and thereafter
writes (ticks) change both mechanical and electronic representations and
reads (on recovery) read only the electronic one. If a synchronome clock
with a capacitative impulse is used, it can probably be designed to
function even over a power failure synchronised with the tick.

The downside is difficulty of doing DST adjustments, especially autumn, But
since that's not wanted and the only correction is recovery forwards after
a power failure, it fits nicely. I don't know if you need to know that it's
stopped/recovering but a status led could show that.

Another option is to do as Alexander says and fit mechanical
feedback/readout. But it could be simplified with a top-dead-centre marker

• the extra information from encoders is nice to have, but at a cost. TDC
  wil give you an alignment check every 12 hours and can, again, indicate
  that it's uncalibrated if there have been power failures since the last
  check.

I think one of the Synchronome competitors did this using a write-only
protocol. As the hour approached, a special signal was sent that put the
clock mechanically on the exact hour and then the next pulse restarted it
from there. However you need to tolerate a short synchronising error every
hour which sounds as though it wouldn't suit your needs.

Hi

We had a system like that in one school I was at. For whatever reason, the clock in
one room always went insane at least once a week. It loved to do this in the middle of
class. Needless to say, all focus on learning went away when the clock started wildly
spinning through however many hours or minutes it though it needed to ….

Turns out they very much do still make those clocks, mechanical solenoids and all.
Slam them with some number of seconds of 24V or 110V AC on that solenoid and
they reset this or that way. Multiple systems came from all sorts of outfits and spares
still get made to match up with them.

I “discovered” that 110V signaling is something to be avoided pretty early on. I would
prefer to avoid that if I can for this project. That said, I’ll always remember the IBM clock
on the wall in 8th grade English class ….

Bob

Quite a long thread and the last post was about analog clocks that had
typically been used in schools and office buildings. Bobs request I believe
was for a clock that was analog and lived in the family room?
I used the school/office clocks for a number of years and honestly the
click thumps will drive you crazy in a quiet area. Plus they take 24 volts.
I went to the very small 14" motor clocks and have been very happy ever
since. The clocks run off a battery operated divider and osc. Nothing
magical but I hate clocks that are out of sync. There are 4 clocks across
time zones.
To wrap up also consider noise.
Good luck Bob.
Regards
Paul

On Fri, Dec 16, 2022 at 5:02 AM Tom Van Baak via time-nuts <
time-nuts@lists.febo.com> wrote:

On 2022-12-15 20:57, Lux, Jim via time-nuts wrote:

That sounds like an impulse clock that would use a DC pulse from a
master clock to pick a solenoid that operated a ratchet mechanism that
advanced the minute hand and the hour hand was gear driven off of it.  I
have an old IBM impulse clock that used a 24 VDC  pulse to advance the
minute hand.  Over the ratchet wheel there is a two way contact that is
operated by a bump on the ratchet wheel that would disconnect the normal
pulse at 1 minute before the hour.  There where both two wire and three
wire system, mine is wired for the three wire so the top normally open
contact is connected to the third wire, so with it stopped at 1 minute
to the hour, a pulse on the third wire would advance the clock on step
and the normally closed contact would close again and it would
continue.  On the two wire system there is a diode in series with   I
believe the normally open contact and the signal line so that when the
normally open is closed, it will only advance when a pulse of reverse
polarity is applied.

There was a later system that used clocks with a synchronous motor and
used signals superimposed on the AC line to synchronize the clocks. 
These clock are easy to recognize as they have a second hand where as
the impulse clocks do not.

It should also be noted that IBM did make wall clocks that that where
just normal AC clocks that where not part of a system, like the 15" one
in front of me built at the IBM Canada plant in Don Mills in the 50s and
still runs great although there is a little slop in the gears now.

I believe the impulse system was initially developed by International
Time Recorder (ITR) one of the predecessor companies that was merged  to
form CTR later renamed to IBM.  IBM would later sell it timeclock
division to Simplex that continued to produce many of the same designs.

Paul.

Another option might be one of these:

http://www.hrclockmovement.com/cpdetail?product_id=741

I've been meaning to get one of these but haven't got round to it, so I
can't vouch for it's longevity. But it does have position sensing hands
and you can point it to an ntp server of your choice. Only updates once
a day to it may not be time nutty enough...

On 14/12/22 07:09, Bob Camp via time-nuts wrote:

Hi

Turns out that the folks at American Time:

https://www.american-time.com/

Make a mind boggling range of wall clocks with various methods of synch ( or not …).. Unlike
a lot of outfits these days, they are quite happy to deal with / sell to individuals. There’s now
a couple of PoE clocks headed this way.

Very nice folks to deal with and highly recommended !!!

Their product also shows up on eBay, but mostly at prices that make one wonder. Given the range
of products, you might have to shop for quite a while to find one that made sense at an attractive
price.

( yes most of their products have a definite “school clock” look …)

Bob