Hi Alberto ...
> value = 11024,80450540374 .
>
> A long number but reflects what I see on most softwares that don't
> have offset adjustments. An 800Hz tone is measured as 796,4Hz ...
How can you have a 0.2 Hz error at 11025 Hz and then 3.6 Hz at 800 Hz ?
Or did I understand wrongly your numbers ?
This 11.024XXX divided by 2 is the "corrected clock value" that JVComm32
needs to print a straight vertical line when in HF-FAX mode , 60 lines per
minute. It's not some measured reference value.
The procedure was:
- JVcomm using 11025 as default sampling rate and 60LPM mode.
- WWV or GPS 1pps pulse to receiver.
- Some minutes later I had a slanted line, which reveals the sound card
clock inaccuracy.
- Applying the "slant correction", software pop-up an info "I'll set the
clock value as 11.024XXX/2". That's it.
W95SSTV needed offset at the same "direction". But this software doesn't
show the adopted value.
Other numbers for the same "800Hz" tone:
= Analyzer2000 measured 796,4Hz.
= Argo measured 802,25 Hz - don't remember the decimal portion
= Freq Counter measured 798.8 <> 801.6 - alternating
= Spectran 127 , 8KHz sampling rate, needed -6.7Hz offset to show 800.0 Hz
= PSK31 for SB16 gave -5 degree (AFC off) - (6 degree = 1 hz?)
I have K6STI DSP-Blaster with its I/Q vector display, SbFFT1.2 from KW5Q,
Spectrum for EVM (G3PLX). I'll do some measurements tonight to see what I
got.
Question: the modulating tones used by WWV HF signal are accurate? How much?
73
Marcus
PY3CRX/PY2PLL
S. B. do Campo - GG66rf
http://py.qsl.br
Active from "DC" to 24GHz
|
