Ric mentioned that he uses a DDS with /4 divider. That setup would only maintain absolute phase as long as it is running continuously without interruption.

Best 73,
Markus

 Von: Riccardo Zoli <[email protected]>
 An: LF Group <[email protected]>
 Verschickt: Mo, 11. Dez 2017 23:11
 Betreff: Re: LF: Little TX test at 8270.00046 from JN64
...
 I'm running with a GPSDO at 16384.0000kHz clocking a DDS AD9850. The output is divided by 4 (SDR quadrature generator). Tuning dword is 8671724. 
...

-----Ursprüngliche Mitteilung-----
Von: Paul Nicholson <[email protected]>
An: rsgb_lf_group <[email protected]>
Verschickt: So, 17. Dez 2017 21:49
Betreff: Re: LF: EbNaut test at 8270.00332 from JN64bw

> f = 8270.00332 Hz
> Start time: 21:30:00 UTC
> Symbol period: 30 s
> Characters: 3 (repeated)
> CRC bits: 24
> Coding 16K21A

Decoded with Eb/N0 = +2.6 dB, rank 1, BER 41.6%
Reference phase -120,-120,-150,-150.
S/N -29.86 dB in 1Hz.

Phase was -50.9 deg, previous nights were +19.1 and +35.7 degrees,
has something changed the phase of the transmission?

At Bielefeld, -9.0 dB;

At Cumiana, -8.4 dB;
(3 nights stacked = -4.2 dB, less than 2 nights stacked,
phase did not repeat)

At Warsaw, -13.8 dB;

Good luck with the 2 char message, maybe that will decode
at Cumiana.

--
Paul Nicholson
--