[time-nuts] WWVB Dephaser Question

[John C. Westmoreland, P.E.]

Paul,

Doesn't the post by Chris Howard above appear to be a good/great starting
point?
Seems like that approach could be extended.

https://github.com/chris-elfpen/Teensy4WWVBsdr

73's,
John
AJ6BC


On Sun, Oct 11, 2020 at 9:23 AM paul swed <paulswedb at gmail.com> wrote:

> Good morning to the group. Both Rodger and I can answer the last question
> on the questionable bits very well.
> I have 7 antique devices spectracoms truetime fluke tracor Dymec etc.
> They handle it very well each has at least a 2 second integration time. The
> dpskr has always taken advantage of this fact. Literally we have run days
> without flips in the charts. Even in summer.
> Some surprises in doing this. Numbers of bits that could change don't seem
> to. Then some of the bits have a really odd pattern but they follow the
> pattern. Thats been coded into the dpskr. In fact anything over several
> years we have figured out are coded in.
> One thing we did not do was automatic change from summer to winter offsets.
> Its really a PIA so just not that excited when this happy switch flipper
> can do it in 1 second.
>
> The other bit of fun for the magical new DPS wwvb receiver is after you get
> the bits you must decode them and its nothing like the AM code. Its a M/N
> FEC code. Its all in the NIST papers but was fun coding and I imagine every
> bit as fun decoding. The dpskr software at least gives you hints to the
> process.
>
> Here is an offer. Anyone that gets the BPSK bits working through a SDR/DPS
> arduino/STM cheap chip. Please no $100 FPGA development kits.
> I will be happy to dig into the decoder. The solution needs a 1pps and the
> bit.
> Whats really funny about the wwvb bpsk is that there has never been sort of
> a open development platform. Anyone who has played with the ES100 will run
> into some issues that are annoying for a time nut to tinker with.
> Regards
> Paul
>
> On Sun, Oct 11, 2020 at 10:39 AM Bob kb8tq <kb8tq at n1k.org> wrote:
>
> > Hi
> >
> > The “simple” approach is to generate the full modulation pattern
> > for the signal based on a “known good” time source. There are a
> > couple of ambiguous  bits so it will only be close. Feed that into
> > your inverter and the result will be (near) clean WWVB. Since you
> > never demodulate the WWVB, there isn’t a lot to the RF side ….
> >
> > There will be some drops and pops. The question will be just how
> > well this or that antique device deals with them.
> >
> > Bob
> >
> > > On Oct 10, 2020, at 10:01 PM, paul swed <paulswedb at gmail.com> wrote:
> > >
> > > The additional article John sent us is a pretty good read. Having
> > soldered
> > > all the little wires together with heat shrink, I can see the advantage
> > of
> > > the epoxy approach.
> > >
> > > A consideration for the discussion. So far its been about the various
> > tried
> > > and true methods for a RF frontend. But the real challenge that is in
> > front
> > > of us is the software that uses that frontend. How about a very easy to
> > > build a phase flipper so that those that are software inclined do not
> > need
> > > to deal with the frontend to get going. The dpskr has a phase flipper
> in
> > it.
> > > But it can be even simpler than that.
> > >
> > > A 60 KHz logic signal ( divide 6 Mhz down or anything else thats easy)
> > > Feeds an inverter to generate the 180 degree phase. A gate to select 0
> or
> > > 180 degrees. All of the gates/inverters can actually be a single quad
> > > nand gate. A D flip flop with the clock from the 0 degree 60 KHz logic
> > > level. Your data into the D input. The D flip flop synchronizes the
> data
> > to
> > > the clock. On the output you can filter the signal or not and cut the
> > level
> > > down or not.
> > > Its a BPSK source.
> > > Granted in a pure gate approach the actual bpsk flip will not be the 2
> X
> > 60
> > > KHz for 1/2 cycle. But in real receivers the 120 KHz never comes
> through
> > > the various stages and filters. So no real harm. This also doesn't
> supply
> > > the AM signals 14-17 db modulation. But its good enough to allow
> software
> > > to be developed and its simple.
> > > Good luck
> > > Paul
> > > WB8TSL
> > >
> > > On Sat, Oct 10, 2020 at 2:05 PM John Magliacane via time-nuts <
> > > time-nuts at lists.febo.com> wrote:
> > >
> > >> For WWVB reception, I use a single turn of 40-conductor ribbon cable,
> > >> configured as a 40-turn loop, brought to resonance with a parallel
> > >> capacitance, that differentially drives an instrumentation amplifier.
> No
> > >> electrostatic shielding is needed to eliminate e-field pickup with
> this
> > >> approach.
> > >>
> > >> The antenna hangs in my attic with thumbtacks and does a commendable
> > job,
> > >> day or night, 1622 miles east of WWVB.  See attached JPEG image.
> > >>
> > >> The March 2017 issue of "Circuit Cellar" magazine described an
> > "improved"
> > >> version of my antenna/preamp combination (which I haven't looked
> into).
> > >> See attached PDF document.
> > >>
> > >> During my early experimentation, I realized that the preamp would need
> > to
> > >> have a high dynamic range in order to perform well in the high-noise
> > >> environment that is LF.   And if the preamp is going to feed a
> receiver
> > >> through any reasonable length of coax, it will need to be able to
> drive
> > a
> > >> high capacitance load as well.
> > >>
> > >>
> > >> 73.000 de John, KD2BD
> > >> _______________________________________________
> > >> 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.
> >
> >
> > _______________________________________________
> > 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.
>