From: "M. Sanders \(PA3BSH\)" <misan@xs4all.nl>
To: G0MRF@aol.com
Cc: "LF-reflector RSGB" <rsgb_lf_group@blacksheep.org>
Subject: RE: LF: RE: Sound cards.
Date: Sun, 9 Jan 2000 21:34:36 +0100
Message-ID: <000501bf5ae0$f208cb60$6c976dc2@v3q4v1>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
In-reply-to: <df.df070c24.25a90d45@aol.com>
Importance: Normal
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dear David,

G0MRF wrote:
> Yes I also run mine at 11k and 16k FFT.  The results are very
> good, but they
> would be 3dB better if I could access it at 5k5 and 16k FFT.  0.3Hz
> resolution  and not 0.7Hz.
>


I am not shure about the 3dB. Conventional filters improve the wanted signal
when decreasing bandwith. The amount of energy/bandwith or signal to noise
ratio that goes into the soundcard input is not changed when applying a more
refined FFT algorthm. We should bear in mind the orriginal transmitted
signal is about 5 to 15 Hz wide. Spreading the spectral reading accross more
fine lines in the display does not improve things as far as i can see. There
is an other feature however in Spectogram that is helpfull. The relative
signal strengths of signals can be identiefied in colours. When choosing the
30dB scale for the full spectrum one can identify wanted signals with a
better signal to noise perception. Signals with a more or less constant
strength appear in one colour. Our wanted signal will be in one colour when
the AGC is turned off. When filtering the other colours out the signal to
noise ratio (by eye) increases dramatically.

The topic interest me so i have posted this message on the reflector as
well, hoping someone has some more info about the effects on signal to noise
ratio.

Greetings,

Michael Sanders, PA3BSH