[Elecraft] Reported RX Frequncies in WSJT-X?

[Richard Lamont]

On 23/11/16 11:16, MaverickNH wrote:

> Looking at WSJT-X 1.7rc2 ALL_WSPR log and seeing RX frequencies such as
> 14.0971282. Seems like too many digits unless reporting 10th of a Hertz as
> in 14,097,128.2 Hz. Perhaps this is calculated as an average over the RX
> window?
> 
> Looking at the log for RX over 500+ RXs, I find the RX frequency range is
> 206Hz. Actual TX 14,097,037 and average RX at 14,097,110, or +73Hz.
> 
> I have a GPSDO managed PC clock as well as a GPSDO managed frequency
> oscillator in my K3S transceiver, so my regulation is pretty tight.
> 
> What are the major components of the deviation in reported RX frequency? I
> posted on WSPRnet also, but one seldom gets replies there on any posts.

There are errors at the TX end, errors at the RX end and errors in the
middle: ionospheric doppler shift.

With a K3S + GPSDO, the biggest source of RX error might be the
tolerance of the sampling frequency of the sound card. The built-in USB
sound card has a 100ppm clock crystal, corresponding to ±0.15 Hz @ 1500 Hz.

-- 
Richard G4DYA