[time-nuts] wwvb d-psk-r update
paulswedb at gmail.com
Mon Oct 13 18:21:54 EDT 2014
Yes indeed the project that keeps on giving. Its been a bit.
Some real progress also.
Currently using the NAA downconverter scheme. That is downconvert wwvb at
60 KHz to 100 Hz, The LO was a free standing xtal at 5.99000X divided by
100 to 59.90x KHz as the LO. It does fairly well.
I then tried various attempts to leverage the data to figure out which way
to flip a inverter to remove the PSK.
They were failures.
I want to understand that a carrier flip has occurred within 200 ms worst
case. I believe there is flexibility in that figure.
Backed up the whole ball game to the cheat'n method of using the GPS tick
and then sample the stream. Really just wanted to see if I could see the
wwvb psk data accurately. That works well and I can actually easily see the
Now since we are discussing "Cheat'n ways" I did one other thing, changed
the LO from the xtal to a HP3335 locked to a RB reference. You simply could
not have more stuff to do a job then whats running right now.
The GPS tick feeds a uproc along with the downconverted wwvb psk. About 10
lines of code and the output is just the bit stream. Absolutely not a drop
of magic in the code. A monostable and flip flop could do the same job. But
hey that's not 2014 is it. The code get GPS tick wait 110 ms sample the
wwvb psk code. Flip a bit.
But at least for the first time I am watching the psk code. Now to get rid
of 80 literal pounds of junk.
Back to the hard work.
If you are reading I may have to go to your 5 Mhz 62.5 Khz IF at 2.5 Khz
More information about the time-nuts