Hi JP:

Have you tried this at midnight local time?

Have Fun,

Brooke Clarke
http://www.PRC68.com
http://www.End2PartyGovernment.com/

Justin Pinnix wrote:

Justin,

A while back, I did some crude experiments with the C-Max CME6005 IC which is, I believe, used in that evaluation module. If memory serves, there were some rather severe restrictions documented in either the datasheet or an application note regarding the loading on the TCO/TCON pins. As I recall, it recommended a very minimal load (much less than what an LED might draw) and to not use the TCO and TCON outputs simultaneously. They claimed violating these recommendations would interfere with the receiver's operation, although I find it hard to believe. I'm certainly not suggesting this explains your results, only that you might want to look into it.

Regards,
Brent

On Oct 30, 2011, at 16:09, Justin Pinnix justin@fuzzythinking.com wrote:

You can use WWVB for timing but it is not as easy as buying the kit.
First you need a better antenna and then a good location for it that
is away from noisy electronics.    then you have to let it disincline
a local clock for a while.  It will not be as good as GPS and you may
never get 24 hour reception, even with a larger loop antenna.

On Sun, Oct 30, 2011 at 2:07 PM, Justin Pinnix justin@fuzzythinking.com wrote:

--

Chris Albertson
Redondo Beach, California

Hi Brent,

I actually saw that warning partway through the experiment.  I disconnected
it and didn't get any measurable difference.  So, I put it back - it makes
it easier to see what's going on :-)

Based on several suggestions here, I think the next thing I'll try is
capturing actual event arrival times.  That will allow me to plot a
histogram and let the experiment run all night (and day) to take advantage
of different local times.  Plus it gives me an excuse to buy (or build) a
universal counter.

Thanks,
-Justin (AJ4MJ)

On Sun, Oct 30, 2011 at 7:00 PM, KD0GLS kd0gls@mninter.net wrote:

Thanks for posting that, Justin, it was most interesting, and the
reference to Brooke Clarke's page, which I did not know.

It would be interesting to know what the analogue signal looks like, as
many pieces of electrical and electronic equipment can radiate
significantly on those frequencies.  I expect you optimised the signal,
though, as you've mentioned it.

Are you sure the images are all correct?  To me, your Fig 2 and Fig 3
images look identical, as do Fig 6 and Fig 7.

"Precision" - well, that depends, doesn't it?  Perhaps more precise than a
very poor Internet connection?  Better than no Internet connection at all!
Certainly not GPS level.

Cheers,
David

SatSignal software - quality software written to your requirements
Web:  http://www.satsignal.eu
Email:  david-taylor@blueyonder.co.uk

On Mon, Oct 31, 2011 at 1:54 AM, David J Taylor <
david-taylor@blueyonder.co.uk> wrote:

Hi David,
Thanks for pointing out the problem with the figures in my webpage.  I have
fixed it.  I'm going to attempt to capture the analog signal by probing the
QOut pin.

Good point about "precision" - it's all relative.  This is an extension of
a previous project of mine - a clock I built around an LPro.  Setting it to
the Thunderbolt is a bit of a pain and I was hoping to build in the C-Max
to make it more "fire and forget".  I was hoping for a time signal within
several microseconds of UTC without the need of an outdoor antenna.  Like
most of my projects, the journey is more interesting than the destination,
so even if I don't accomplish that goal I'm still having fun.  Embedding a
consumer GPS is probably the more practical solution.

Thanks,
-Justin

The updated images make the effects clearer.  It will be interesting to
see the image capture.

Microseconds, no.  Milliseconds perhaps.  But sometimes nothing at all
depending on the time of day and season of the year etc. etc.  Let's hear
more about the project as it progresses.

Cheers,
David

SatSignal software - quality software written to your requirements
Web:  http://www.satsignal.eu
Email:  david-taylor@blueyonder.co.uk

Update -

I attempted to look at the analog signal.  I looked at the output of the
crystal.  With the scope's low-pass filter turned on I was able to see a
very weak sine wave with a period of roughly 15us.  It was too weak for any
analysis.

http://www.fuzzythinking.com/wordpress/wp-content/uploads/2011/11/60khz.png

I also looked at the demodulator out.  This is a much larger amplitude
signal, though it has already been rectified.  However, I was able to see
good correlation between it (blue) and the TCON output (yellow)

http://www.fuzzythinking.com/wordpress/wp-content/uploads/2011/11/dem_out.png

Most importantly, I've been reading up on the subject and have adjusted my
expectations.  I didn't realize that frequency measurements (like those
done by the HP117 and Fluke 207) don't actually use the timecode part of
the signal.

Page 11 of http://tf.nist.gov/general/pdf/1383.pdf tells me that WWVB is
3-4 orders of magnitude better than WWV for frequency measurements, but
that the two services' time uncertainties overlap and WWVB is at best 1
order better.  As a matter of fact, my measurements of the C-MAX line up
with the top end of that estimate (and Tom's :-)).

I'm still planning on capturing a long set of timings, but I need to do
some hardware mods first.

Thanks,
-Justin

On Mon, Oct 31, 2011 at 1:04 PM, David J Taylor <
david-taylor@blueyonder.co.uk> wrote: