Hello all:

I monitored the transmission of John, WD2XES in WSPR mode on 9/20 to 9/21 -2009 
ending at 05:00 UTC. I copied him for about 2-1/2 hours during which time I 
experimented with different settings on my receiver, as well as my DSP audio 
filters to see what interaction it would have on WSPR decodes. John was 
transmitting on 137.577 KHz, and this was my first attempt using WSPR on 2200 
meters. Receive conditions were deplorable. The lightening
crashes with a 500 cycle filter were peaking at 40dB over S-9. I didn't have 
enough time to make many observations; however here are some things that I did 
find out about WSPR.
WSPR references its decodes against the noise level in a given audio passband. 
The more noise, lower S/N ratios of course. On my setup; WSPR could not decode 
when static crashes were above peaks of 10dB over S-9. 
I found this out by using my noise blanker. Without the NB on, static peaks 
were 40 over, and WSPR would not decode. My NB reduced the peaks
to S-9 to 10 over, and WSPR would then decode. It made no difference what audio 
BW I was using, without the NB, the Static simply masked the signal. The noise 
level using the 2.4 KHz SSB filter was out of the question. So, the narrower 
you can get your IF - BW,the better WSPR will do at decoding. My comment about 
the dB reading. With no audio filtering, the XES signal decoded around -26 to 
-29 dB. With audio BW as low as 20 Hz;
it would show +8 to +10 dB. This is not an exaggerated reading. When you narrow 
your audio BW, you lower the passband noise level. And the DSP filter also " 
Boosts " the desired signal level in amplitude. So, lower passband noise, 
coupled together with a higher signal level amplitude will naturally increase 
WSPR's dB reading. And just to make this remark, adjustment of the volume 
control does not have any effect on WSPR at all. That is unless you turn it all 
the way down. About narrow audio filtering and WSPR, I'll say this; try it and 
see what happens. I did find this out; audio filtering less than 20 Hz produced 
too much " filter ringing " and WSPR hears this as an un-intelligible signal, 
and the ringing will mask out the desired signal, and No Decode. My little one 
night, 2-1/2 hour
experiment showed this:  In the extreme receive conditions on 2200 meters, WSPR 
decoded OK at an IF - BW of 500 Hz. Decodes were better, and easier for WSPR 
when I used an audio BW of 220 Hz. And decodes came thru fine with the 20 Hz 
audio BW, centered on the desired signal. IF filtering, and IF shift will 
provide the best results from interfering signals within the passband. Notch 
filter would be beneficial if the interfering signal is far
enough away that the notch dosen't affect the signal you are trying to receive. 
The audio filtering is most effective at reducing the noise that WSPR has to 
deal with, and WSPR dosen't like extreme static.
I am fairly new to WSPR, and this is probably some elementary information, But 
I do hope that to any newcomers to WSPR, it might be of some value.

Thanks for your time in reading this, and ;  73 to all :::


Andy - KU4XR - EM75xr - Friendsville, TN.
LOWfer Beacon " XR " @ 184.322 KHz ( QRSS-60 )
Coordinates:  N:  35* 43' 54" - W:  84* 3' 16"
http://www.myspace.com/beaconxr