[Lowfer] Audio limiting using diodes, and capacitor

[Andy - KU4XR]

Hi all:

As many of you probably have done; I have been reading the thread
on the RSGB LF reflector where Jim - M0BMU has been using diodes,
back to back, across the audio line to the sound card. The purpose
being; to reduce the noise impulse action from lightening.
OK; curiousity got to me, so I had to try it out myself and see 
if there is any benefit. The finding here is; YES. And surprisingly
so. An entire evening in a nutshell; I found that Germanium diodes
worked better than Silicon diodes. I found that there is a " sweet -
spot " on the volume adjustment where the distortion to the audio
is minimal, but the reduction in impulse noise in quite notable.
Too much audio will reintroduce the impulse noise but in a wideband
clipped form, and will create all kinds of crud on the audio, and
the waterfall display. I found the best place to insert the diodes
was before the active audio filter. I obtained additional reduction
in noise by placing a .33 µF capacitor, and a variable resistor,
going to ground from the output side of the active audio filter.
The resistor is used like a tone control to reduce high frequency
response on the audio line. This works to offset what the diodes do
to the audio. The noise becomes wideband, and shifts a great deal
of audio impulse noise energy to higher frequencys. The " Tone "
control will roll off the high frequency noise. With the right 
settings of volume, tone resistor, and tone capacitor value, 
I was able to almost eliminate 20 dB over S-9 lightening crashes.
I could see them on the S-meter, but just barely hear them on the
audio.
I forgot to mention that I was using the carrier of DCF-39 to see
any changes on the waterfall of Spectran, plus it is audible here,
and I could " hear " the reduction in lightening noise, and any
changes to the desired audio. There was one setting that would give
maximum noise reduction, and best Signal to Noise ratio on the
signal I wanted to hear thru the noise.
I have a 60 second audio clip of the results. The first 20 seconds
is un-filtered audio. There is a 1500 cycle tone in the background,
that is DCF-39 set for a 1500 cycle offset. The next 20 seconds,
( 20 to 40 ); is with the diode clipper, and capacitor tone control
optimized for 1500 cycles. Then the final 20 seconds, ( 40 to 60 ),
is the filtered audio, and the radios Noise Blanker turned ON.

See if you can hear any difference in the static crashes, and the
received signal. And imagine this to be a WSPR signal:::

Hear the audio clip at :    http://webpages.charter.net/ku4xr/

Click on the Diode & Capacitor noise reduction MP3 file.
The recording was made with a 2.4 KHz Bandwidth for plenty of noise.


73; and have a great weekend everyone.


Andy - KU4XR - EM75xr - Friendsville, TN. USA
LOWfer Beacon " XR " @ 185.29875 KHz ( QRSS-60 )
Coordinates:  N:  35º 43' 54" - W:  84º 3' 16"
http://www.myspace.com/beaconxr
http://webpages.charter.net/ku4xr/