[time-nuts] WWVB PM Receiver

[J. Forster]

Been tried. Doewn't work. Among other things, if you multiply by 2 and
then divide by two, you can have extra flips or missed flips. A Miller
Divider has the same issue.

..........

Interesting that you should bring up TV.

A REQUIREMENT of the conversion from B&W to color was compatability with
the then existing installed base.

In the conversion from NTSC to HDTV, compatability was not assured, but
Congress MANDATED low cost converters for the enite installed base.

A WWVB (or LORAN-C) timing receiver were far more expensive than all but
the largest TVs of the day.

Yet, AFAIK, there is no effort whatsoever to mitigate the effects of this
change, not from the government, not from the sole-source promoter of this
scheme.

YMMV,

-John

========================




> Hello All -
>
> I am new to this forum but have read it for a couple of years. The
> present fulminations on the WWVB format change should be reconsidered
> in the light of prior art. As an old RfFengineer I do not see any issue
> with the format and the business about patents is not really applicable
> as these modulations have been in use since the 1920's.
>
> For one thing the tube analog color TVs managed to have a PLL recover
> the color sub-carrier using less than 10  cycles of the 3.5 mHz color
> burst stuck on top of the horizontal sync pulse. Here we have a
> continuous signal with occasional phase reversals and some amplitude
> steps. This is a much simpler problem that is not begging for
> sophisticated solutions, however much tun they may be to implement.
>
> The following or variations should work.
>
> 1. Receive with a decent antenna but keep the bandwidth such that it
> does not store the phase information.
> 2. Use preselection  with care because the filter is going to have to
> flywheel down and then back up as the phase reverses. This will reduce
> the carrier level during that interval. More bandwidth will shorten the
> time of the ring down-ring up period. The filter must be phase stable
> with temperature.
> 3. AGC with gain may be desired for weak signals. If so implement with
> one of the Analog Devices handy
> dandy VGAs. They provide 50 or more dB of voltage controlled gain with
> absolutely NO phase variation as the gain is changed.
> 4. Now get rid of the short term amplitude variation - for this you
> need a limiter with no AM to PM
> conversion. Again Analog Devices has suitable parts with phase
> variation of less than 5 degrees for more than 70 db dynamic range. The
> AM is now gone.
>
> These parts also output a log of the analog input - so the AM keying
> could be recovered using that if
> desired for some complex sampling scheme.
>
> 5. OK - now you have an amplitude stable BPSK modulated carrier.
> 6. Get rid of the BPSK.
>
> Double the frequency - then divide by 2 and the carrier is recovered
> less the PSK.
>
> For the doubler you may wish to consider a Quadrature Driven Mixer
> Frequency Doubler - has no Hi Q circuits and the output is easily
> filtered. See my article in QEX last year March April.
>
> Or the classic Class C or diode doubler can be used as long as they
> preserve the input phase.
>
> 6a. A Costas Lopp is another classic method that should work fine.
>
> 7. Narrow the bandwidth (say 10 Hz) with a PLL  to clean up the output
> of the doubler. This will get rid of wide band noise and improve the
> SNR. Again the PLL must preserve phase - so some temperature test may
> be a good idea. The PLL also provides a constant level clock to the
> Divide by 2 to avoid phase jitter.
>
> 8. Divide by 2 - and you have a clean 60 kHz signal with no modulation.
> This can the be used for frequency comparison by the phase method or to
> lock up a good local standard.
>
> So it should be possible to implement a receiver without infringing any
> patents and without reams of signal processing code.
>
> Hopefully this is helpful.    -73 John C.Roos  K6IQL  Spring Hill Kansas
>
>
>
>
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>