FST4W and WSPR - 2 Hz Tx difference

[PE3ES / F4VTQ]

When setting the same band and same dial frequency (in my case 30 meter and 1437) sending WSPR and FST4W intermittently produces a static 2 Hz difference between the decoded signals.

A stable receiver like G3ZIL reports :

FST4W 10.140.132

WSPR 10.140.134

All other stable stations show the same frequencies and 2 Hz difference.

Where is that unexpected 2 Hz difference coming from ?

The difference from 1437 is my own calibration difference, but now I do not know which frequency to tune to, 1432 or 1434.

Thanks

Erwin

[ahop...@googlemail.com]

Hi Erwin,

that is interesting, I had a very quick look at the code and the signal for both modes uses the tx frequency for the lowest tone and the others are higher,  maybe I should centre them around the tx frequency, I don't remember details about this in the spec. I shall have a play and compare self  decodes with the same signal routed to wsjtx. With a tone spacing of 1.46Hz the centre of the signal would be just over 2Hz higher so maybe wspr and fst4w decoders use a different measure of the tx frequency.

73 Alan M0NNB

[PE3ES / F4VTQ]

I have used my RFZero as test transmitter today. I see SparkSDR + HL2 spotting that 30 meter signal :

- FST4W 1430/1431

- WSPR   1433

So in spotting the same frequency also decodes show  2 Hz difference where WSPR is again the higher one.

Op zaterdag 4 maart 2023 om 13:35:02 UTC+1 schreef ahop...@googlemail.com:

[PE3ES / F4VTQ]

Yesterday afternoon I used another setup to do the same test:

Yaesu FT817nd - WSJTx 2.6.1

After a few hours of comparing FST4W with WSPR, surprise, NO difference between the 2 frequencies ....

Op zondag 5 maart 2023 om 22:53:35 UTC+1 schreef PE3ES / F4VTQ: