[NLRS] Link -- New phase-modulation WWVB signal

[Dr. Gerald N. Johnson]

On 3/12/2013 7:46 PM, Donn, WA2VOI wrote:
> According to the NIST web pages, they're a little better than that,
> though they do acknowledge the difficuties of 60KHz work. :-)
>
> "The antennas are spaced 857 m apart. Each antenna is a top loaded
> monopole consisting of four 122-m towers arranged in a diamond shape. A
> system of cables, often called a capacitance hat or top hat, is
> suspended between the four towers. This top hat is electrically isolated
> from the towers, and is electrically connected to a downlead suspended
> from the center of the top hat. The downlead serves as the radiating
> element.
>
> As the length of a vertical radiator becomes shorter compared to
> wavelength, the efficiency of the antenna goes down. In other words, it
> requires more and more transmitter power to increase the effective
> radiated power. The north antenna system at WWVB has an efficiency of
> about 50.6%, and the south antenna has an efficiency of about 57.5%.
> However, the combined efficiency of the two antennas is about 65%. As a
> result, each transmitter only has to produce a forward power of about 54
> kW for WWVB to produce its effective radiated power of 70 kW."

I'm sure I have that data here somewhere. But it will take another year 
to find it once I start looking for it.

I have used WWVB and used to use WWVL to check my in house standard. I 
did it short time and I could see either random walk in my standard, 
variations in propagation time, or I figured movement of the vertical 
wire radiator in the wind. It might have been random variations in my 
test set up also.

What I did was with a TRF receiver tuned to 60 kHz, I put that on the 
scope vertical (Tek 475) and I triggered the scope sweep with 100 kHz 
from the standard (Manson Labs, double oven, 5 MHz 5th overtone crystal 
in a sealed glass tube about the size of a 6146). The sine wave 60 kHz 
signal with 100 kHz trigger gives the appearance of three power line 
phases. I selected a crossing of two sine waves and followed it across 
the scope, only when they were at high power part of each second. That 
way I maximized my S/N. When I got the standard tweaked so the crossing 
stayed within 100 ns for 15 minutes I figured I had it as good as I 
could do because the crossing was then wandering randomly in that 100 
ns. That's 1.1x10-10, as good as an oscillator rated for 10-9 per day 
aging gets for such a short term measurement.

It might have been better to have used the local standard's output for 
the horizontal sweep to take scope triggering out of the measurement, 
but then it would have been much harder to figure the error having to 
look for drift of a major fraction of that 100 kHz cycle lasting 10 
microseconds (10,000 nanoseconds). That standard has a 1 MHz output too 
which might have given some help but the scope sweep made it much easier 
to measure the amount of drift.

I now have an HP 10811 oscillator without battery back up and I warm it 
up before using it with my microwave counter, but I have not checked it 
with WWVB or GPS. I am content with pretty good, say a part per 10 
million which it should do unless savaged and I've not had to argue 
frequency that close in my life so I'm in a don't really care all that 
much. I'm not going to set up to run the station from that high quality 
frequency standard when I have a tuning dial on the IF.

At 60, kHz QRM from NTSC TV sweep is a big problem and sometimes 
harmonics from arcs on the AC power line appeared to contribute 
considerable QRM.

More conventional VLF frequency standard techniques use a phase locked 
loop with about a 10 hour time constant, maybe longer so the twice an 
hour phase shifts used on WWVB for identification don't kick the local 
standard out of lock. So I presume phase shift keying as proposed won't 
mess up those either. Though that way it can take a week to get a 
standard on frequency so they can plot the received phase and detect the 
ID shifts and detect when aurora is on the way. The VLF propagation 
shifts considerably several hours before aurora arrives. At the Iowa 
frequency standard maintained by W0PFP at ISU (dismantled when he 
retired early) it too three 6' tall racks of equipment to perform that 
function.

I wonder how well the clock handles the known QRM sources when its based 
on the binary coding and the amplitude shift keying.

73, Jerry, K0CQ
>
> Basic clock is <1x10-12. Compensating for path delay allows UTC to
> <100usec. Not too bad, actually, for consumer use (i.e., $20 clocks).
>
> 73 Donn
> WA2VOI/0
>
>