atomic clocks

[Jeff Liebermann]

On Mon, 23 Jun 2014 05:29:50 -0700, "William Sommerwerck"
<grizzledgeezer@comcast.net> wrote:

"Jeff Liebermann" wrote in message
news:1skfq9pr23olbr1e0793glaavhhod4v8hg@4ax.com...

Loopstick vertical:
http://802.11junk.com/jeffl/WWVB%20test/loopstick-vertical.jpg
Pure noise and no signal.

Bottom line is that the loopstick should be mounted horizontally and
perpendicular to Ft Collins CO. Vertical mounting does NOT work.

Interesting. I have the Time Machine and the Time Machine with a projection
clock. With the loopsticks vertical, neither has yet lost sync
That is... the sync annunciator (or what I interpret to be the sync
annunciator) in the display is still visible.

I suspect the loss of lock delay is rather long. It would not do to
have the sync indicator flashing on and off as the signal fades in and
out. Besides, the receiver is only powered on for 5 minutes. If you
don't mind tearing it apart, just ground the PON (power on) line, and
monitor the output with an oscilloscope. No need to use the sync
indicator.

Incidentally, with a 1 baud data rate, an LED attached to the data
line should give a tolerable visual indication of signal quality. On
the Temic[1] chip, the output is rather high impedance, so something
like a CMOS buffer will be needed.
<http://psn.quake.net/wwvbsdr.html>
I might modify mine into something like that when I have time.

>Both show the same time, which is unlikely if either had lost sync.

I beg to differ. Reception could be miserable and the sync indicator
probably won't show a problem for 24 hours. I think (not sure) that
it really means that it hasn't received an update for 24 hrs.

"Of course" an antenna's orientation should match the signal's. Why the La
Crosse only works when it doesn't, and it doesn't seem to matter for the
Oregon Scientific products -- I don't know.

I've seen worse RF design mistakes. I couldn't find a teardown of the
WS-811561 or any similar La Crosse radio clock. One possible problem
is that it's impossible to test a WWVB clock in China as the signal
doesn't go that far. It's possible that they designed and tested it
using a simulator, which will not have the cross polarization problem.
Just guessing.


[1] Temic was bought by Atmel in 1998:
<http://www.cpushack.com/2011/02/05/atmel-buys-mhs-again-the-twisted-history-of-atmel-temic-and-mhs/>

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[David Harmon]

On Sun, 22 Jun 2014 16:54:40 -0700 in sci.electronics.repair, Jeff
Liebermann <jeffl@cruzio.com> wrote,

On Sun, 22 Jun 2014 09:07:46 -0700, "William Sommerwerck"
grizzledgeezer@comcast.net> wrote:

Just "Most-Accurate Clock". But it's the same thing.

Heath model GC-1000 I assume.
https://www.youtube.com/watch?v=c8fD_3MgfDw
I have one that I'm "rebuilding". The biggest change is replacing the
overheating 7805 linear regulator with a switching regulator

We had one of those. We ran coax up two floors to the roof for the
antenna. The shield was grounded at the roof to a drain pipe or
something like one.

The clock ran internally on +5V and +12V supplies. The coax shield
was also connected to the supply ground. In order to get plus and
minus voltages for the RS-232 port, they connected the port ground
pin to the internal +5 supply and switched between +12 and gnd.

We connected the serial port to a PC. The PC serial port ground was
connected to the case, and the third prong on the power cord.
So, we had the clock's +5 supply driving current through a fifty
foot loop of whatever unknown conductors happened to be between that
drain pipe and the third prong.

The 7805 overheated.

>

 

[Michael Black]

On Mon, 23 Jun 2014, Ian Field wrote:

"Arfa Daily" <arfa.daily@ntlworld.com> wrote in message
news:NsKpv.56803$7m3.46486@fx11.am4...


"Leif Neland" <leif@neland.dk> wrote in message
news:mn.b3eb7de6c7e48bab.130671@neland.dk...
Michael Black har bragt dette til os:
On Sun, 22 Jun 2014, Arfa Daily wrote:



"

I have another atomic clock with an external antenna. I'll set it up so
it's vertical, and see what happens over the next day or two.


Surely 'atomic clock' is the wrong term for these devices ? As I
understand it, an atomic clock is a laboratory time-standard instrument
based on the decay rate of some atomic isotope, usually caesium ? The
devices to which you are referring are radio-synchronised clocks (also
referred to, again wrongly in my opinion, as "radio controlled clocks"
), deriving their synchronisation from data broadcast from a number of
low frequency transmitters around the world. Otherwise, in between the
synchronisation times, these clocks are just free-running, much like any
other clock.

It may be the wrong term, but it's common useage now. Most people don't
know what it's about anyway, other than that they keep time, so they
won't be mislead into thinking there's a cesium standard inside. They
are aware of "sync'ing up" so I don't think they have any problems once
they get it.

Michael

The clock's accuracy is based on a cesium standard (or some other atomic
timebase)

Whether the standard is inside the clock itself, or some hundred
kilometers away, controlling or sync'ing over radio is just a matter of
detail.

Gotta disagree with you on that one. The thing at the far end is an "atomic
clock". The thing that we are talking about, even though it is synced to
that atomic clock, is not one itself. It is an ordinary bog-standard
free-running clock, with a radio and a bit of data decoding circuitry in
it, to sync it to the real atomic clock once a day

All of my desk clocks sync on the hour every hour - my Casio Waveceptor watch
does it every 24h - if it fails it tries again on the hour for the next 3 or
4 hours.

Aren't you in the UK? If so, you have a different time signal, and likely
quite different "atomic clocks". One reason a lot of the clocks here in
North America (and the Waveceptor) try for the signal late at night is
because noise is lower then. Fewer electronic devices on at that point to
make noise. It's variable whether I can get a clock to sync in the
daytime. But if you are in the UK, you'd be closer to the station than
someone like me who is kind of at the outer ring of th reception area for
WWVB, which is in Colorado.

If we're nitpicking about the name, I found on the sidewalk a couple of
years ago a clock that didn't need setting. I got all exicted a desktop
"atomic clock" with LED readout and which ran off the AC line. But it wsa
really misleading, it was set at the factory and had a backup battery or
something, it didn't sync to anything.

Michael

 

[Ian Field]

"Michael Black" <et472@ncf.ca> wrote in message
news:alpine.LNX.2.02.1406231403480.26046@darkstar.example.org...

On Mon, 23 Jun 2014, Ian Field wrote:



"Arfa Daily" <arfa.daily@ntlworld.com> wrote in message
news:NsKpv.56803$7m3.46486@fx11.am4...


"Leif Neland" <leif@neland.dk> wrote in message
news:mn.b3eb7de6c7e48bab.130671@neland.dk...
Michael Black har bragt dette til os:
On Sun, 22 Jun 2014, Arfa Daily wrote:



"

I have another atomic clock with an external antenna. I'll set it up
so it's vertical, and see what happens over the next day or two.


Surely 'atomic clock' is the wrong term for these devices ? As I
understand it, an atomic clock is a laboratory time-standard
instrument based on the decay rate of some atomic isotope, usually
caesium ? The devices to which you are referring are
radio-synchronised clocks (also referred to, again wrongly in my
opinion, as "radio controlled clocks" ), deriving their
synchronisation from data broadcast from a number of low frequency
transmitters around the world. Otherwise, in between the
synchronisation times, these clocks are just free-running, much like
any other clock.

It may be the wrong term, but it's common useage now. Most people
don't know what it's about anyway, other than that they keep time, so
they won't be mislead into thinking there's a cesium standard inside.
They are aware of "sync'ing up" so I don't think they have any
problems once they get it.

Michael

The clock's accuracy is based on a cesium standard (or some other
atomic timebase)

Whether the standard is inside the clock itself, or some hundred
kilometers away, controlling or sync'ing over radio is just a matter of
detail.

Gotta disagree with you on that one. The thing at the far end is an
"atomic clock". The thing that we are talking about, even though it is
synced to that atomic clock, is not one itself. It is an ordinary
bog-standard free-running clock, with a radio and a bit of data decoding
circuitry in it, to sync it to the real atomic clock once a day

All of my desk clocks sync on the hour every hour - my Casio Waveceptor
watch does it every 24h - if it fails it tries again on the hour for the
next 3 or 4 hours.
Aren't you in the UK? If so, you have a different time signal, and likely
quite different "atomic clocks".

The waveceptor can be set up for use in a list of countries, but its a
fiddly process of pressing various buttons in the right order.

The clocks were bought in Lidl and AFAIK mostly German made - so they
probably sync to the 77kHz DCF in Frankfurt rather than the 60kHz MSF in
Anthorn, Cumbria (Formerly sited in Rugby).

 

[jfeng@my-deja.com]

On Tuesday, June 24, 2014 9:38:37 AM UTC-7, Michael Black wrote:

And the Garmin certainly was off a second or so from the "atomic
clocks", so it's back to the old problem of "which is right?".
Michael
After compensating for propaagation delays between my place and Fort Collins, the audiio tone on WWV is good enough for me, and deinitely more reliable than any of my consumer-grade "atomic" clocks decoding WWVB. I do not know how much latency there may be between the process of computer decoding and the LCD display.

 

[dave]

On 06/23/2014 12:04 PM, Ian Field wrote:

The waveceptor can be set up for use in a list of countries, but its a
fiddly process of pressing various buttons in the right order.

The clocks were bought in Lidl and AFAIK mostly German made - so they
probably sync to the 77kHz DCF in Frankfurt rather than the 60kHz MSF in
Anthorn, Cumbria (Formerly sited in Rugby).

Isn't GPS a lot less trouble? I'd worry about the elimination of the
leap second if I were you. FYI I have a dozen WWVB clocks of every
description (except 24hr analog).

 

[Michael Black]

On Tue, 24 Jun 2014, dave wrote:

On 06/23/2014 12:04 PM, Ian Field wrote:




The waveceptor can be set up for use in a list of countries, but its a
fiddly process of pressing various buttons in the right order.

The clocks were bought in Lidl and AFAIK mostly German made - so they
probably sync to the 77kHz DCF in Frankfurt rather than the 60kHz MSF in
Anthorn, Cumbria (Formerly sited in Rugby).

Isn't GPS a lot less trouble? I'd worry about the elimination of the leap
second if I were you. FYI I have a dozen WWVB clocks of every description
(except 24hr analog).

GPS tends to be more expensive. Yes, I got a Garmin used at a garage sale
for five dollars, and then last year a TomTom One for ten dollars at a
Rotary Club sale, but that doesn't happen much. I can get an "atomic
clock" for $20 (or 2.00 for that one at a garage sale), even my Casio
Waveceptor was only $20, some very special deal.

And the Garmin certainly was off a second or so from the "atomic clocks",
so it's back to the old problem of "which is right?". I know the "atomic
clocks" keep the same time, it's so much fun to watch two of them change
time in sync. And, GPS doesn't work so well in a house.

Michael

 

[Ian Field]

"Michael Black" <et472@ncf.ca> wrote in message
news:alpine.LNX.2.02.1406241235220.28455@darkstar.example.org...

On Tue, 24 Jun 2014, dave wrote:

On 06/23/2014 12:04 PM, Ian Field wrote:




The waveceptor can be set up for use in a list of countries, but its a
fiddly process of pressing various buttons in the right order.

The clocks were bought in Lidl and AFAIK mostly German made - so they
probably sync to the 77kHz DCF in Frankfurt rather than the 60kHz MSF in
Anthorn, Cumbria (Formerly sited in Rugby).

Isn't GPS a lot less trouble? I'd worry about the elimination of the leap
second if I were you. FYI I have a dozen WWVB clocks of every description
(except 24hr analog).

GPS tends to be more expensive. Yes, I got a Garmin used at a garage sale
for five dollars, and then last year a TomTom One for ten dollars at a
Rotary Club sale, but that doesn't happen much. I can get an "atomic
clock" for $20 (or 2.00 for that one at a garage sale), even my Casio
Waveceptor was only $20, some very special deal.

When I bought the Casio it was Ł60, I paid an extra Ł20 for a stainless
bracelet instead of the extremely nasty "resin" strap, which was probably
what they chose to call PVC.

Last time I saw the price advertised, was touted as a "special offer" in a
newspaper at Ł145.

 

[Jeff Liebermann]

On Tue, 24 Jun 2014 12:38:37 -0400, Michael Black <et472@ncf.ca>
wrote:

>GPS tends to be more expensive.

GPS receivers also burn more power making battery life problematic.

12ft GPS Wall Clock
<https://www.sparkfun.com/tutorials/47>

And the Garmin certainly was off a second or so from the "atomic clocks",
so it's back to the old problem of "which is right?". I know the "atomic
clocks" keep the same time, it's so much fun to watch two of them change
time in sync. And, GPS doesn't work so well in a house.

It depends on what you're trying to accomplish.
<http://leapsecond.com/java/gpsclock.htm>
Note that GPS time is 16 seconds ahead of UTC. There's also UT1 for
astronomers, which is where the leap second originates.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Geoffrey S. Mendelson]

dave wrote:

Isn't GPS a lot less trouble? I'd worry about the elimination of the
leap second if I were you. FYI I have a dozen WWVB clocks of every
description (except 24hr analog).

I've been researching and am looking for old android phones for this
purpose. There is a GPS time setting APP and an NTP app, but they don't
talk to each other.

My goal, which at this point is just speculation, is to put together
a combination APP which converts an old android phone into a GPS based
time server on your wifi network. My hope is if it works and enough people
use it, then not only will people have accurate time on their home networks
without the extra internet traffic, but each one is a bit of old electronics
rescued from a landfill.

To use, you just load the app, connect your phone to AC power and stick it
in a window. It syncs the clock with GPS and makes it available to all
your computers, etc.

If anyone knows of such an app, or wants to write one, feel free to, I
am not looking for anything more than a free timeserver and less trash.

Geoff.

--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379

 

[Jeff Liebermann]

On Wed, 25 Jun 2014 11:50:15 +0000 (UTC), "Geoffrey S. Mendelson"
<gsm@mendelson.com> wrote:

dave wrote:

Isn't GPS a lot less trouble? I'd worry about the elimination of the
leap second if I were you. FYI I have a dozen WWVB clocks of every
description (except 24hr analog).

I've been researching and am looking for old android phones for this
purpose. There is a GPS time setting APP and an NTP app, but they don't
talk to each other.

The wi-fi only Android tablets use NTP to set their clock. Look at:
/etc/gps.conf
for the configuration. Usually, it points to pool.ntp.org.

Incidentally, I had to install a clock sync program on a friends
no-name Chinese Android tablet because it would sync before updating
the date/time, causing the time of last sync to get mangled.
<https://play.google.com/store/apps/details?id=ru.org.amip.ClockSync>

My goal, which at this point is just speculation, is to put together
a combination APP which converts an old android phone into a GPS based
time server on your wifi network.

Using an Android for the platform might be a bad idea. You can easily
find stand alone Linux cards and boxes that will do much the same
thing. For example, Raspberry Pi doing GPS clock:
<http://www.satsignal.eu/ntp/Raspberry-Pi-NTP.html>
Plenty of others with and without GPS for NTP. Incidentally, also
think about stuffing a RADIUS server into the device for Wi-Fi
authentication and security.

Also, why would one use the built in GPS receiver for time, when a
network connected router can deliver the time with reasonable accuracy
via NTP? I don't see the benefit.

My hope is if it works and enough people
use it, then not only will people have accurate time on their home networks
without the extra internet traffic, but each one is a bit of old electronics
rescued from a landfill.

Well, a noble goal, but I don't think the number of smartphones that
will be saved from the recyclers is going to be noticeably affected by
such a project. A few hundred (or thousand) smartphone NTP/GPS
servers versus 968 million smartphones sold in 2013 (world wide) is
not going to do much.

To use, you just load the app, connect your phone to AC power and stick it
in a window. It syncs the clock with GPS and makes it available to all
your computers, etc.

Not quite so simple. You need to guarantee that your smartphone can
see the sky in order to get the GPS time. You'll need to decide if
the clients gets their time directly from the smartphone (acting as an
NTP server), or 2nd hand from the router as an NTP relay. For
example, Windoze clients by default get their date/time from
time.microsoft.com. Are you going to support SNTP or the full NTP?
Etc.

If anyone knows of such an app, or wants to write one, feel free to, I
am not looking for anything more than a free timeserver and less trash.

Sorry, but I don't have the time to do that.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[jfeng@my-deja.com]

On Thursday, June 26, 2014 7:08:09 AM UTC-7, dave wrote:

The Baud rate is like 1. I don't think decoding it takes very long.
My "synchronized" atomic clocks sometimes differ by a few seconds with the WWV audio tone.

 

[dave]

On 06/24/2014 09:38 AM, Michael Black wrote:

On Tue, 24 Jun 2014, dave wrote:

On 06/23/2014 12:04 PM, Ian Field wrote:




The waveceptor can be set up for use in a list of countries, but its a
fiddly process of pressing various buttons in the right order.

The clocks were bought in Lidl and AFAIK mostly German made - so they
probably sync to the 77kHz DCF in Frankfurt rather than the 60kHz MSF in
Anthorn, Cumbria (Formerly sited in Rugby).

Isn't GPS a lot less trouble? I'd worry about the elimination of the
leap second if I were you. FYI I have a dozen WWVB clocks of every
description (except 24hr analog).

GPS tends to be more expensive. Yes, I got a Garmin used at a garage
sale for five dollars, and then last year a TomTom One for ten dollars
at a Rotary Club sale, but that doesn't happen much. I can get an
"atomic clock" for $20 (or 2.00 for that one at a garage sale), even my
Casio Waveceptor was only $20, some very special deal.

And the Garmin certainly was off a second or so from the "atomic
clocks", so it's back to the old problem of "which is right?". I know
the "atomic clocks" keep the same time, it's so much fun to watch two of
them change time in sync. And, GPS doesn't work so well in a house.

Michael

I was speaking to the non-compatibility of the VLF clocks across borders
vs a global protocol, for mission critical users with access to the sky.

 

[dave]

On 06/24/2014 10:43 AM, jfeng@my-deja.com wrote:

On Tuesday, June 24, 2014 9:38:37 AM UTC-7, Michael Black wrote:
And the Garmin certainly was off a second or so from the "atomic
clocks", so it's back to the old problem of "which is right?".
Michael
After compensating for propaagation delays between my place and Fort Collins, the audiio tone on WWV is good enough for me, and deinitely more reliable than any of my consumer-grade "atomic" clocks decoding WWVB. I do not know how much latency there may be between the process of computer decoding and the LCD display.

The Baud rate is like 1. I don't think decoding it takes very long.

 

[dave]

On 06/25/2014 04:50 AM, Geoffrey S. Mendelson wrote:

dave wrote:

Isn't GPS a lot less trouble? I'd worry about the elimination of the
leap second if I were you. FYI I have a dozen WWVB clocks of every
description (except 24hr analog).

I've been researching and am looking for old android phones for this
purpose. There is a GPS time setting APP and an NTP app, but they don't
talk to each other.

My goal, which at this point is just speculation, is to put together
a combination APP which converts an old android phone into a GPS based
time server on your wifi network. My hope is if it works and enough people
use it, then not only will people have accurate time on their home networks
without the extra internet traffic, but each one is a bit of old electronics
rescued from a landfill.

To use, you just load the app, connect your phone to AC power and stick it
in a window. It syncs the clock with GPS and makes it available to all
your computers, etc.

If anyone knows of such an app, or wants to write one, feel free to, I
am not looking for anything more than a free timeserver and less trash.

Geoff.

Neat!

 

[Tim R]

On Tuesday, June 24, 2014 2:02:04 PM UTC-4, Jeff Liebermann wrote:

Note that GPS time is 16 seconds ahead of UTC. There's also UT1 for

astronomers, which is where the leap second originates.

Is this always true? So GPS is precise to the nanosecond, but inaccurate by a quarter minute?

 

[jfeng@my-deja.com]

On Friday, June 27, 2014 7:44:01 AM UTC-7, dave wrote:

On 06/26/2014 07:53 AM, jfeng@my-deja.com wrote:
On Thursday, June 26, 2014 7:08:09 AM UTC-7, dave wrote:
The Baud rate is like 1. I don't think decoding it takes very long.
My "synchronized" atomic clocks sometimes differ by a few seconds
with the WWV audio tone.

You should reorient, you haven't had a lock recently.
Maybe, but the little signal strength icon is on solid, and I interpret that to mean it thinks it sucessfully synchronized the last time it tried (less than 24 hours earlier). And I see the discrepancy on all of my cheap WWVB clocks. I do assume that the only significant error in the WWV audio tone is due to the variability of the height of the ionospheric skip layer (this tone was always seems to be in good agreement with my Heathkit GC-1000).

 

[Michael A. Terrell]

Jeff Liebermann wrote:

On Tue, 24 Jun 2014 12:38:37 -0400, Michael Black <et472@ncf.ca
wrote:

GPS tends to be more expensive.

GPS receivers also burn more power making battery life problematic.

12ft GPS Wall Clock
https://www.sparkfun.com/tutorials/47

And the Garmin certainly was off a second or so from the "atomic clocks",
so it's back to the old problem of "which is right?". I know the "atomic
clocks" keep the same time, it's so much fun to watch two of them change
time in sync. And, GPS doesn't work so well in a house.

It depends on what you're trying to accomplish.
http://leapsecond.com/java/gpsclock.htm
Note that GPS time is 16 seconds ahead of UTC. There's also UT1 for
astronomers, which is where the leap second originates.

I have one of these: http://www.ebay.com/itm/331205597628 I have to
get someone to install the antenna for me, since I can no longer climb
ladders.


--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.

---
This email is free from viruses and malware because avast! Antivirus protection is active.
http://www.avast.com

 

[Michael A. Terrell]

Arfa Daily wrote:

I'm sure they don't, but most of us on here are engineers of one kind or
another, and we *do* know better, so we ought not to be helping to
perpetuate this wrong description ...

Or the imaginary DB9 connectors?



--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.

---
This email is free from viruses and malware because avast! Antivirus protection is active.
http://www.avast.com

 

[Tim R]

According to Garmin, GPS time is off by the number of leap seconds that UTC has added, since UTC adds these corrections and GPS does not.

However, a correction signal is sent with the transmission, and if the GPS unit has the proper software it should match UTC to within a second.

So when I get home I'll fire up the eTrex and see if it matches UTC.