Brian Lloyd wrote:
On Sat, Sep 25, 2010 at 1:54 PM, M. Jenssen <[email protected]
<mailto:[email protected]>> wrote:
*Hi Dave,*
*as mentioned last year when I sent you screen shots, the problem
with the occasional frequency shifts in Olivia and Contestia
continue. Means: at times the signal sent out is not symmetric to
the guide tones and there is a frequency difference.*
I have seen this too on occasion and just chalked it up to me
mistuning the signal.
--
Brian Lloyd, WB6RQN/J79BPL
sending my response to the entire list.
I believe you are seeing the result of fft processing in fldigi.
Because of the interaction between the fft block sampling rate and the
signal baud rate you may see a smear of the waterfall in a downward
direction. You can reduce this by setting the waterfall "fft latency"
to 1 and disable FFT averaging.
I have posted wav files for you to playback and also Audacity analyses
of these waveforms that clearly show that the precursor tones and the
data tones are concurrent.
http://www.w1hkj.com/spectrum_tests/4-500.wav
http://www.w1hkj.com/spectrum_tests/4-500-tones.csv
http://www.w1hkj.com/spectrum_tests/4-500-data.csv
http://www.w1hkj.com/spectrum_tests/8-250.wav
http://www.w1hkj.com/spectrum_tests/8-250-tones.csv
http://www.w1hkj.com/spectrum_tests/8-250-data.csv
Each wav file includes both a video and RSID preamble. You should be
able to set RxID ON and automatically lock to the playback waveforms.
Properly locked the s/n should be 29 dB and the f/o = 0.0 Hz.
The *tones.csv file are tab delimited files that you can open in Excel,
Gnumeric, or Open Office. Try plotting the spectrum and observing the
tone placement.
Ditto for the *data.csv which is an Audacity spectral analysis of the
data portion only.
I hope this explains what you are seeing. If it something other than
what I describe and the files indicate then I need substantive data
capture to allow me to analyze the problem. Ie: Rx waveform capture
when you observe the problem. I can provide you with an upload url for
large files.
73, Dave, W1HKJ
_______________________________________________
fldigi-alpha mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/fldigi-alpha
