[time-nuts] WWVB New Modulation scheme...
Burt I. Weiner
biwa at att.net
Fri Oct 26 00:46:41 EDT 2012
If I sent these thoughts before, pardon me as I'm suffering from CRS Syndrom.
Part of the normal work I provide is off-air measurements of
broadcast stations. Would you believe, they pay me to do
this! Anyway, I measure AM carriers, FM carriers, and TV, both
analog and digital pilots, as well as other stuff. For FM stations I
also measure any subcarrier including the RDBS signal at 57 kHz.
My method for measuring is described
Here: http://www.k5cm.com/k6OQK%20FMT%20NEW.htm The methodology
described here is used for primarily for AM Broadcast signals, but
can be used for FM subcarriers also.
Some time ago I tried measuring the RDS signal in the broadcast FM
baseband With my trusty 3586B looking at the baseband of my FM
receiver I tried tuning the 3586B SLM to 57 kHz and looked at the
resultant 15625 Hz I.F. out of the 3586B as described in the above
article. It was a useless mess. Due to the nature of the 57 kHz RDS
modulation, which is not FM, I just couldn't get a meaningful
One day while poking around and looking at the 57 kHz RDS signal,
rather than using 20 Hz bandwidth, which is really about -3 dB at 6
Hz, I used the 3.1 kHz bandwidth position. Lo and behold, I now had
a very well defined display that looked very similar to the eye
pattern used for examining digital transmissions. I was allowing
both sidebands into the 3586B and was now able to measure down to
milliHertz resolution with no problem. All I had to do was adjust my
3336B (read the above linked article) to get the pattern to stand
still as either a circle or and X on the scope.
I still need to dig into this a bit farther because the RDS signal
really does not have a carrier as such. It has the two sidebands
depending on the phase shift determined by the data. However, I do
know that RDS generators that are GPS referenced give me a stationary
pattern when I'm tuned to 57000.000 Hz. The RDBS signal is supposed
to be locked to the third harmonic of the 19 kHz pilot, which has a
+/- 2 Hz tolerance. The truth is that they're not always locked, and
I know of several stations that do not feed a Pilot reference to the
RDBS generator. FM broadcast RDS injection levels are typically
about 2-5 percent of +/- 75 kHz deviation, but it is not a FM signal
when you look in the FM baseband. Anyway, just a thought that maybe
if you look at the WWVB signal with a wider bandwidth it might be
interesting. I really need to take the WWVB loop off of my
Symetricom 8170 and hook it to my 3586B SLM and see for myself.
For what it's worth,
Burt I. Weiner Associates
Broadcast Technical Services
Glendale, California U.S.A.
biwa at att.net
More information about the time-nuts