Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Atomic Signal Transmitter/Repeater???

2,924 views
Skip to first unread message

mag3

unread,
Jun 13, 2011, 8:09:35 AM6/13/11
to
Hello all, new to this group. If an FAQ exists, please point me there and
I'll read up.

Anyway, things where I live have changed such that I no longer receive
the radio broadcast time signals from Ft. Collins, Co. etc. etc. My "Atomic"
wall clocks, even my watches no longer pick it up. My watch does pick up
the signal when I'm staying elsewhere in the country. I have no issues with
NTP protocol for my SOHO at home. That works well. MY PDC picks up the
NTP signal, and then re-broadcasts it to the other machines via Windows
Time Service.

What I'm looking for (and asking the NG), if they exist, are one or both of
the following:

1) A small, low power "transmitter" device that attaches to your SOHO,
receives the NTP signal, and then re-broadcasts it in "radio broadcast"
format for all your other atomic clock devices in the house that receive
via that signal?

2) Failing the above (and I'd imagine how the US Authorities might have issues
with it), then perhaps a set of wall clocks (digital or analog) with either USB
or RJ-45 NIC ports that can synchronize directly via NTP signals... I mean
"Cheap" ones, not "Industry Grade" devices in the $1000+ etc. range.


Pie in the sky, perhaps, but there's no harm in asking... :-)

Thanks much.

____________________________________________
Regards,

Arnold

Terje Mathisen

unread,
Jun 13, 2011, 10:20:51 AM6/13/11
to
mag3 wrote:
> What I'm looking for (and asking the NG), if they exist, are one or both of
> the following:
>
> 1) A small, low power "transmitter" device that attaches to your SOHO,
> receives the NTP signal, and then re-broadcasts it in "radio broadcast"
> format for all your other atomic clock devices in the house that receive
> via that signal?

I have a feeling that the only thing missing would be an actual mW
transmitter, tuned to the proper frequency, and gated by a signal line
from an rs232 or printer port:

The actual sw to wiggle said line according to the US/UK/GER transmitter
rules would be trivial to write. :-)

In fact, if anyone have a suitable transmitter for DCF, I'd be happy to
write the sw...

Terje

--
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"

Chris Albertson

unread,
Jun 13, 2011, 11:17:54 AM6/13/11
to
I'm actually working(slowly) on something like this. A good sixe
antennal will go up inside a 20 foot tall pole in the back fence.
This will receive the 60KHz signal from WWWV Ft. Collins, Co. There
are several antenna types but I've found a 4" loopstick works here.
Them the signal is converted to a binary time code and put on a RS442
serial cable. rs442 is good for at least 1,000 feet.

But if all you want is an accurate clock, any cell phone is pretty
good, iPhones and iPads stay on-time most people now have one of there
with them full time. You see very few peole with wrist watches
anymore and no one under 30.

My purpose for getting WWWV is to compare it with GPS time and I hope
maybe learn something about the atmosphere. Any different in the two
times must be because of the path delay.

> _______________________________________________
> questions mailing list
> ques...@lists.ntp.org
> http://lists.ntp.org/listinfo/questions
>

--

Chris Albertson
Redondo Beach, California

Steve Kostecke

unread,
Jun 13, 2011, 3:03:35 PM6/13/11
to
On 2011-06-13, mag3 <zmpma...@yahoo.com> wrote:

>1) A small, low power "transmitter" device that attaches to your SOHO,
>receives the NTP signal, and then re-broadcasts it in "radio broadcast"
>format for all your other atomic clock devices in the house that
>receive via that signal?

You'd have to check the legality of this sort of repeater with your
country's equivalent of the FCC.

If you can receive DCF, JJY60, MSF and WWVB you can use this to
rebroadcast it:

http://www.c-max-time.com/home/index.php

Time Signal Generator Shop Floor Booster

TSG100 TSG100 is a Time Signal Generator especially designed for shops
and exhibition halls. Once started it will automatically receive the
time and transmit the selected signal to syncronize all RC products in a
6m radius.

The external loop antenna connection allow to enlarge the coverage area
up to 50m radius.

$550 from digikey

NIST suggest this antenna -->
http://tf.nist.gov/service/pdf/1936_antenna_wwv.pdf

Also see this thread -->
http://www.radiobanter.com/showthread.php?t=62837

There is also the CHU shortwave time signal from
Canada on 3330 kHz, 7850 kHz and 14670 kHz -->
http://www.nrc-cnrc.gc.ca/eng/services/inms/time-services/short-wave.html

>2) Failing the above (and I'd imagine how the US Authorities might have
>issues with it), then perhaps a set of wall clocks (digital or analog)
>with either USB or RJ-45 NIC ports that can synchronize directly via
>NTP signals... I mean "Cheap" ones, not "Industry Grade" devices in the
>$1000+ etc. range.

Either of these will acquire the time over a network:

http://www.chumby.com
http://dash.chumby.com

If you're up for a project you could build your own Arduino based SNTP clock
(which polls your local NTP server.)

Starting points:

http://arduino.cc/en/Tutorial/UdpNtpClient
http://www.faludi.com/code/open-source-arduino-clock/
http://www.youtube.com/watch?v=_4zJcqZKzms
http://7in7.tumblr.com/post/38066438/open-source-arduino-clock-project
http://hacknmod.com/hack/arduino-based-led-alarm-clock/

--
Steve Kostecke <kost...@ntp.org>
NTP Public Services Project - http://support.ntp.org/

Chris Albertson

unread,
Jun 13, 2011, 3:19:47 PM6/13/11
to
On Mon, Jun 13, 2011 at 7:20 AM, Terje Mathisen <"terje.mathisen at
tmsw.no"@ntp.org> wrote:
> mag3 wrote:
>>
>> What I'm looking for (and asking the NG), if they exist, are one or both
>> of
>> the following:
>>
>>        1)      A small, low power "transmitter" device that attaches to
>> your SOHO,
>>                receives the NTP signal, and then re-broadcasts it in
>> "radio broadcast"
>>                format for all your other atomic clock devices in the house
>> that receive
>>                via that signal?
>
> I have a feeling that the only thing missing would be an actual mW
> transmitter, tuned to the proper frequency, and gated by a signal line from
> an rs232 or printer port:
>
> The actual sw to wiggle said line according to the US/UK/GER transmitter
> rules would be trivial to write. :-)

You would not have to write it. An time code generator is included
with the NTP source (the source file is .../utils/tg.c) By default it
is not compiled.

Quote for the source code follows:

* This program can generate audio signals that simulate the WWV/H
* broadcast timecode. Alternatively, it can generate the IRIG-B
* timecode commonly used to synchronize laboratory equipment. It is
* intended to test the WWV/H driver (refclock_wwv.c) and the IRIG
* driver (refclock_irig.c) in the NTP driver collection.

mi...@niftyegg.com

unread,
Sep 29, 2014, 9:48:30 PM9/29/14
to
On Monday, June 13, 2011 8:17:54 AM UTC-7, Chris Albertson wrote:
......
>
> My purpose for getting WWWV is to compare it with GPS time and I hope
> maybe learn something about the atmosphere. Any different in the two
> times must be because of the path delay.

Partly solved problem:
There are interesting ways to discover time delay data.
All you need do is keep the GPS receiver fixed in space.
Any apparent motion over time belongs to atmospheric changes.
The apparent wandering of multiple GPS receivers might give
additional information.

GPS time is accurate and precise---
Research differential GPS.
"http://en.wikipedia.org/wiki/Differential_GPS"
GPS assisted surveying has proven to have astounding
quality and many Automobile GPS devices have a parallel
receiver to collect current data and apply it.

Back of the envelope GPS reacts to picosecond time signals
to generate a default accuracy of 15-meter nominal accuracy
and when augmented to about 10 cm in case of the best implementations.

The back of my envelope demands no wire longer than 15 meters
and if I am serious 10 cm (about 1.x nano seconds). All logic and
amplifier delays insert additional errors and even when understood
must be stable over time and temperature.

Inside a single GPS receiver clock deltas can be measured in picoseconds
even if the absolute time is +/- 100ns...

Look hard at individual GPS receivers because many can lock on numerous signals:
"receiver that tracks both GPS and GLONASS satellites simultaneously. When using them together, the receiver has the ability to lock onto 24 more satellites than using GPS alone". If you can extract the raw data
you can look for differential transit time data from all 24 paths. i.e.
does the local location appear to move as each satellite is added to or
removed from the computation.

Small systems like the Raspberry Pi now have GPS add on devices inspect the device driver code to see how much you can dig out of the hardware. An old
Android phone likely has GPS hardware and Android is mostly open source.
0 new messages