It has been a couple of weeks since this topic was discussed, but
I think I should add a bit.  "QST" which is the magazine published by the
American Radio Relay League has had articles from time to time about time
and frequency standard stations.  I detect a slight bit of confusion, here.

        The HF time signal stations of WWV and it's sister, WWVH in Hawaii
do use the 100-HZ carrier method for sending the time and date to receivers
capable of receiving it.  It takes a whole minute to receive a frame of data
which is clocked out at one bit/second.  The signal is mixed in with the
voice announcements, standard frequency tones, and ticks indicating the
passage of seconds.  These transmitters are meant to be heard by the human
ear as well as mechanically received.  The same is true for Canada's time
and frequency station, CHU.  It transmits voice time announcements, ticks,
and a binary time update in the form of a 300-baud Bell103-style data burst
sent each second between about the 31ST and 40TH seconds in every minute.

        Both the U.S. and Canadian short wave time services take at least
a minute to automatically update clocks so I would put my money on the
WWVB signal if one can receive it.  It should provide complete updates almost
continuously since its binary code is constantly being sent and is not
time-shared with voice.  WWVB should also not be quite as susceptible to
the random Solar flares and daily/seasonal changes in signal propagation that
plague short wave signals.  The same holds true for the other VLF time
signal transmitters around the world although I have no idea as to whether
they all use different digital data formats.  Even if they do, the signal
propagation characteristics should be the same so they all represent a good
technical solution for automatically synchronizing clocks.

Martin McCormick