[time-nuts] wwvb d-psk-r predictive approach success

Bob Camp kb8tq at n1k.org
Fri Sep 18 07:15:24 EDT 2015


Hi


Ok !!

That sounds very useful. 

One silly way to deal with the unknown bits:

Flip them half way through the bit time. You will be wrong on every single 
bit. You will only be wrong for half the time. The noise suppression in the older
receivers should deal with short(er) blips better than long(er) ones. 

Bob

> On Sep 17, 2015, at 9:37 PM, paul swed <paulswedb at gmail.com> wrote:
> 
> Hello fellow time nuts. I started a long journey to deal with the wwvb BPSK
> signal over 3 years ago. Since then I have released reactive solutions that
> have been shared. Generally in the form of costas loops for phase tracking
> receivers like the Fluke 207 etc and then the remodulator for the
> timeclocks like the spectracoms and true times. But there has always been
> other possible approaches. I sent an email over a year ago called the
> cheat-n way. Several of you had also suggested GPS assisted or predictive
> methods. Never lost but a suggestion doesn't get you to the end game.
> 
> That said I finally went down the path as many things came together.
> Learning C, Arduino, getting the NIST document etc and in reality a week
> and a half ago I really wrote the program that seriously works well.
> 
> Essentially it works as follows.
> Predict WWVBs time code in advance using the GPS sentance. Then march the 1
> second bits out exactly 107ms after the wwv/wwvb Tick and use the GPS tick
> to keep things in step.
> Phase change on the BPSK signal occurs at 100ms and the east coast is about
> 7.5ms from wwvb.
> The time code simply flips a invert non invert switch in the form of a
> simple home brew double balanced mixer. Not to many 60 Khz DBMs around. By
> the way the transformers are the ethernet filters mentioned here a year ago
> by another time-nut.
> 
> I started testing this a week ago and its been reliable. Its great to
> listen to wwvb again on the fluke 207 thats stable and locked. To watch the
> charts over days with the diurnal shift and such. All like the gud ole
> days. It works with the HP 117s and as a added bonus all of the old clock
> systems I have work. Thats the spectracoms and truetimes netclocks etc.
> 
> So how complicated is the solution as compared to others I have released
> its simple.
> NEO 6 GPS receiver, 2 simple logic chips 14 pin dip class, an Arduino UNO
> or Nano or Mini.
> All total $25 maybe? Nice big stuff you can use a soldering iron on. The
> modulator slips between the old preamp and old receiver. It passes the DC
> around to the preamp.
> 
> I will release all details schematics code as always. Sorry no kits will be
> made.
> 
> Additional notes.
> Not a bad first project for C and arduinos. Use 30% of the space so its
> kind of big.
> But the code is very linear and should be very clear. I also built in
> tables for 10 years worth of operations Leap years and such. A table that
> washes out any error in a given year. Always starts coding correctly on the
> nearest minute.
> I will say there are 5 to 10 bits in the sequence that simply depend and I
> have them coded but there isn't a good way to know what they should be.
> There is a 5 bit sequence called ls_DST in the nist standard and though I
> read the details its not clear how to code them in relation to some date or
> something for 10 years of predictions. Doesn't seem to matter because what
> I see is a single or so bit error doesn't upset the old receivers. Still
> curious minds would like to know.
> 
> 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.



More information about the time-nuts mailing list