[time-nuts] WWVB teensy BPSK early experiments

[paul swed]

John thats exactly what Chris did. Its the teensywwvb.ino. It decodes the
AM timecode just as the DCF code does. In fact Chris's code is derived from
the DCF code with the changes you are asking about and it works. Download
his solution and read the comments throughout. The original DCF comments
are in it also.

Its the starting point for the tinkering I am doing towards a BPSK SDR.
Thats a big leap for me.
What the code gives you is the frontend gain, followed by a bandpass
filter. Multiplier that mixes the LO and incoming signal to a lowpass
filter. Then on to AM detection and AGC.

If you don't care to install the arduino IDE with teensy extension just use
notepad ++ to look at the code. Pretty exciting.

Regards
Paul
WB8TSL

On Mon, Nov 2, 2020 at 8:24 PM John C. Westmoreland, P.E. <
john at westmorelandengineering.com> wrote:

> Hello Time Nuts,
>
> I have a question -
>
> Has anyone been successful getting WWVB to decode using the example for
> DCF77 and making the changes for WWVB?
>
> Chris has his own approach - it's interesting - but on my setup at least I
> haven't been successful with it - maybe it's my antenna - not sure just
> yet.
> Also - my screen eventually 'whites-out' after running that for a while -
> meaning - the screen goes completely white and I don't know what's doing
> that yet - I suppose it's a bug of some sort.
>
> But, with the DCF77 approach - it appears to identify bits OK - with the
> code having only minimal changes - so I'd like to go down that path.
>
> Note - I have a fork here - but it's debug - and not sure it's worth
> anything other than looking at some of the data that's being dumped.
> https://github.com/jwestmoreland/Teensy-DCF77
>
> For DC7FF - the AM is just 0.2ms (1) and 0.1ms (0) with no AM at the minute
> marker, but WWVB is a little more complicated -
> 0.2ms (0), 0.5ms (1), and 0.8 ms (Marker) - 2 consecutive markers mark the
> end of current minute/beginning of next minute.
>
> The DCF77 code seems to take some advantage of the signal level/timing and
> the dimensions of the TFT display for
> displaying the signaling - at least that's what it appears to me right now
> unless I'm looking at this wrong - I guess that isn't
> a big deal until you try to deal with the marker timing for WWVB - of
> course I could be looking at this wrong right now too.
>
> Maybe trying something like the Goertzel algorithm like KE9NS has done in
> his work with WWV is worth taking a look at too with
> this setup.
>
> 73's,
> John
> AJ6BC
>
>
>
>
>
> On Sun, Nov 1, 2020 at 10:50 AM Mike Feher <mfeher at eozinc.com> wrote:
>
> > Too bad I am an old retired hardware engineer and know nothing about
> these
> > new SW controlled devices. Have Fun - Regards - Mike
> >
> > Mike B. Feher, N4FS
> > 89 Arnold Blvd.
> > Howell NJ 07731
> > 848-245-9115
> >
> > -----Original Message-----
> > From: time-nuts <time-nuts-bounces at lists.febo.com> On Behalf Of paul
> swed
> > Sent: Sunday, November 1, 2020 12:48 PM
> > To: Discussion of precise time and frequency measurement <
> > time-nuts at lists.febo.com>
> > Subject: Re: [time-nuts] WWVB teensy BPSK early experiments
> >
> > John sent the link on the teensy wwvb AM receiver. It can easily be
> > shifted to other frequencies. Using the work Frank dcf77 and then Chris
> > wwvb did with the PJRC audio design tool gives quite a bit of insight to
> > the work that was done.
> > The system assumes a sample rate of 192Khz. That limits the receive range
> > to about 98KHz.
> > I have not been able to confirm the rate I am using is 192Khz and when
> the
> > set command is used its not recognized. So somethings up.
> > The design Chris has essentially is a microphone input with some 38 db of
> > gain feeding a bandpass filter around 60 KHz to a multiplier (RF mixer)
> > then a low pass filter. Chris and Franks design is much more than these
> few
> > words with spectrum display, signal level, and AGC.
> > So without any effort the frontend solution is sitting in the teensy that
> > cuts out a bunch of typical wiring. Granted its not as tight as discrete
> > components can be. But its a chunk of software code. No digi key orders
> for
> > L & Cs... :-) The output of the chain is 600 Hz audio that does contain
> the
> > phase changes.
> >
> > But the trick is creating the true I & Q channels. Not an issue to add a
> > second multiplier and chain did that last night. But at someplace a 90
> > degree phase change or delay needs to occur.
> > Or as suggested a hilbert transform.
> > Regards
> > Paul
> > WB8TSL
> >
> >
> >
> >
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at lists.febo.com
> > To unsubscribe, go to
> > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> > and follow the instructions there.
> >
> _______________________________________________
> time-nuts mailing list -- time-nuts at lists.febo.com
> To unsubscribe, go to
> http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
> and follow the instructions there.
>