View Single Post
  #19   Report Post  
Posted to sci.electronics.repair
Jeff Liebermann Jeff Liebermann is offline
external usenet poster
 
Posts: 4,045
Default radio time code clock error

On Sat, 27 Jun 2015 23:37:42 -0400, Michael Black
wrote:

I thought low frequencies were used by WWVB (and WWVL before it) because
it was all ground-wave, and well penetrating.


The skywave doesn't magically go away at lower frequencies. For
example, here's a study on hearing both WWVB and JJY (Japan) in
Germany, which certainly would require skywave propagation.
http://df6nm.bplaced.net/LF/MSFgaps/MSFgaps.htm
See 3rd graph from the top. Unfortunately, I've never bothered to do
any DXing on VLF, but have been told that it's possible to hear Europe
and Japan in California if the conditions are right. VLF DX list:
http://www.dxinfocentre.com/time-vlf.htm

Unlike HF (high freq) propagation, which is "bent" by the ionosphere
back towards the ground, VLF frequencies are reflected off the D
layer.
http://vlf.stanford.edu/research/introduction-vlf

If it was on shortwave, you would be stuck with a given frequency for a
givne time, which of course is why WWV transmits on a bunch of
frequencies.


Actually, VLF has the same problem. Beyond some distance (which
varies with antenna size and power level), the various time sync
stations become very sensitive to ionospheric layer conditions, which
is mostly controlled by the sun. During the day, skywave propagation
sucks, while ground wave is unaffected. However, when there's dark
over most of the distance between the transmitter and the receiver,
the skywave returns. More on how it works:
http://solar-center.stanford.edu/SID/activities/ionosphere.html

The reason I've seen for late night sync'ing (and which makes sense) is
that in the wee hours of the morning, fewer people are up, thus much less
manmade interference.


That would be so nice if it were true, but it's not. It's not man
made interference that causes most reception problems, it's lightning
and atmospheric RF noise that are extremely high at VLF frequencies.
See the graph at:
https://en.wikipedia.org/wiki/Atmospheric_noise
https://en.wikipedia.org/wiki/Atmospheric_noise#/media/File:Atmosphericnoise.PNG
At 60 KHz, the noise level is about 80dB above the normal thermal
noise. When designing a receiver for 60 KHz, the problem is not the
usual NF (noise figure) and gain, but rather being able to remain
linear and handle the overload produced by the noise. That's why you
often see VLF receiver front ends with heat sinks on the RF
transistors.

Incidentally, the high noise levels is also why you can get away with
using tiny antennas with loss instead of gain. A bigger antenna will
amplify the noise and the signal equally which doesn't improve things.
A bigger antenna also produces more voltage at the receiver input,
which has to handled by an increasingly higher voltage and power input
stage. However, what a bigger antenna might do is increase the Q of
the antenna (i.e. decrease the antenna bandwidth) which would pickup
less off frequency noise, which will help increase the SNR (signal to
noise ratio).

So the CFLs are turned off, the tv sets are turned
off (probably a bigger issue when they were actually CRTs, all that
horizontal sync frequencies into the deflection coil), lots of other
things that might be on in the daytime are off so they aren't making as
much noise.


Look again at the noise graph, and notice the red line added showing
man made noises. At 60 KHz, the man made noise is about equal to the
atmospheric noise.

If you want to pursue such a noise source survey, I suggest you build
a loop resonant at 60 KHz and plug into an oscilloscope looking for
noise. I suggest a very high Q magnetic loop so that it only picks up
noise very close to 60 KHz. Just about anything with a switching
power supply should show up as a noise source. I have some opinions
on CFL lamps, LED lamps, and plasma TVs which I'm not quite sure are
correct, so I'll not comment on those. Let's just say I have some
very quiet LED lamps and some VERY noisy LED lamps.

I've noticed this. My Casio Waveceptor watch starts looking at midnight
local time, and if I have the CFL on at the time, it likely doesn't sync
up, but will at a later hour (the watch is better than most of the clocks,
since it tries multiple times). COnversely, if I have the CFL off, it
generally will sync at midnight.


That sounds about right. The chip, usually made by C-Max uses an
algorithm to determine when to listen. The problem is battery life.
It makes no sense to listen during daylight hours, so it doesn't. If
it sync at midnight, it doesn't try again later. If it fails, it
retries according to some pattern. I used to have the link to the
details, but can't find it right now. This is close but not the one I
was looking for:
http://www.c-max-time.com/tech/software6005.php

And I'm probably at the far edge of the reliable reception area.


Probably true with a wris****ch receiver or consumer device. Not true
with a proper loop antenna, which gets much better range. In the
early 1980's, I setup a WWVB receiver in an area infested with high
electrical noise. None of the commodity WWVB clocks would work. A
shielded loop antenna and overkill coax shielding did the trick.
Something like this, except my loop was circular and about 1 meter
wide:
http://www.ka7oei.com/wwvb_antenna.html

Of course, with at least some of the clocks, one does have the option of
pressing button (or putting the batteries back in) so the clock tries to
sync up, no matter what time it is.


I hacked one of clocks to do that so I could make measurements:
http://802.11junk.com/jeffl/WWVB%20test/
The receiver inside the antenna has an "enable" line to turn it on and
off. In this device, it's labeled PON. The clock/display/control
chip enables it according to the previously mentioned algorithm. I
tied the line high (after disconnecting it from the clock) so that it
would run continuously. It takes about 3 weeks to kill the two AAA
batteries when running continuously.

Clean signal that will decode correctly:
http://802.11junk.com/jeffl/WWVB%20test/loopstick-perpendicular-to-WWVB.jpg
Noisy signal that won't work:
http://802.11junk.com/jeffl/WWVB%20test/end-pointed-at-WWVB.jpg
http://802.11junk.com/jeffl/WWVB%20test/loopstick-vertical.jpg


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