[time-nuts] Truetime dc 468 goes sat rcvr simulator basics working

paul swed paulswedb at gmail.com
Mon Dec 6 21:12:42 UTC 2010


Justin,
I might think quite a few time Nuts have these.
So making progress have build a gps sat message decoder for the GPRMC
sentence that gives time and data. I believe most GPS units put that
sentence out. Have it decoding time and next is date. Maybe tonight.
Then I have to glue this code into the simulator to set the clock.
Next will be a update subroutine which is quite tricky as to how the updates
done.

Lastly the dates a pain. Always has been. Its the number of days from new
years.
But it seems that the system doesn't actually care what those digits are.
Soooo I am thinking of using the LSD 2 as the day of the month and the
single 3rd digit as a hex month. I know cheeky. I can calculate days as the
system originally used. Maybe you can't use the 3rd digit for Hex months.
But I do know the other digits support hex numbers.
So thats were I am at.
Lastly I may use a seperate SX28ac as a 20 mc to 100hz divider would save
about 8 external divider chips. Just not sure I can make it accurate enough
to appease the dc468 gods.
Regards
Paul
WB8TSL

On Mon, Dec 6, 2010 at 3:44 PM, Justin Pinnix <justin at fuzzythinking.com>wrote:

> Paul,
>
> That is very much appreciated.  I have one of these boxes.  I didn't
> realize
> that the signal was no longer being broadcast, so at this point I'm all for
> whatever hacks will make it useful.
>
> Thanks,
> -JP
>
> On Fri, Dec 3, 2010 at 5:55 PM, paul swed <paulswedb at gmail.com> wrote:
>
> > Well I think this puts me in the dangerous zone of time nuttery.
> >
> > I had posted to the group about a month ago about a Austron 2201 gps rcvr
> > and any documents anyone might have. Russell responded and sent me some
> > great documents on the Austron.
> >
> > In a phone conversation with Russell we started talking about the old
> > Truetime dc 468s. Kind of got me thinking. I have been able to design a
> > small single chip goes timecode simulator to drive these older clocks.
> >
> > The dc 468 was always quite attractive with their panelplex 7 segment
> > displays. Have re-purposed the boxes for several other projects on
> occasion
> > using the supply nd the displays.
> >
> > I find the dc 468s at flea markets still for $5 in the northeast US.
> > Generally no converter assembly though it doesn't matter since the GOES
> > timecode signals been gone since 2005.
> >
> > At the moment the software fits in a small parrallax SXB micro. Its
> written
> > in basic language and compiled. Runs very fast with a 20 Mhz clock.
> >
> > At this stage the unit puts out a manchester encoded timecode signal and
> is
> > injected at the very first data stage on the decoder. You have to do this
> > to
> > get the PLLs to lock so things decode correctly. I have not attached a
> GPS
> > time signal to it yet and associated time decoding to automatically sync
> > the
> > clock. That will be the next item. But I suspect this will be a bit ugly.
> > Like sync once per day type of behavior.
> >
> > Another interesting thing I discovered in this project. That the GOES 100
> > Hz
> > timecode actually transports a very stable frequency reference. So you
> need
> > a pretty good 100 hz source. What this means is the clock will only drift
> > as
> > much as that source drifts. Thats why I think a single sync 1 time per
> day
> > is most likely just fine. Additionally the internal 10 MC clock locks to
> > the
> > 100 HZ and is pretty stable. (Ok now I am in trouble. What on earth does
> > that statement mean in this group?)
> >
> > As with the Loran simulator I will share the information with the group
> as
> > it shapes up.
> > For now I need to make sure the basic software actually is correctly
> > keeping
> > time.
> >
> > Regards
> > Paul
> > WB8TSL
> > _______________________________________________
> > 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.
> >
> _______________________________________________
> 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.
>


More information about the time-nuts mailing list