Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Message-ID: <xPGsOBAVGWj4Ewmg@dennison.demon.co.uk>
Date: Tue, 25 Jan 2000 08:40:53 +0000
To: rsgb_lf_group@blacksheep.org
From: "Mike Dennison" <mike@dennison.demon.co.uk>
Subject: Re: LF: Narrow Bandwidth reception
In-reply-to: <LPBBJIAEFBFDMPGAFFKBAEFBCAAA.g3xtz@rgcomms.demon.co.uk>
MIME-Version: 1.0
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

>I would like to know other users experiences using the Spectrogram programs.
>I have frequently found that I am only able to detect the weakest stations
>using a sample rate of 5.5k with a FFT size of 16384. This gives an
>on-screen bandwidth of 86 Hz. If I attempt to see more of the band, either
>by increasing the sample rate, or decreasing the FFT size, I am unable to
>detect the signal that I know is there.  This seems fairly obvious, in that
>decreasing bandwidth should improve signal to noise, but I wonder if all
>users have the same results?
>73 de Graham B. Phillips.  G3XTZ.

This is exactly my experience which, as you say, is what theory
suggests. I am sometimes surprised to see spectrograms with wider
bandwidths (though I understand some people have computer problems with
the higher sample rate). Another very important control is the averaging
which must be set to the highest that will allow you to see dots
properly - note that this will vary with dot length. Spectrogram 5.09
calls this Spectrum Average and is best set at just above the dot length
(I set it at 4). Earlier versions needed a much bigger figure - say 10
times the dot length. This makes a huge difference to the S/N and
especially the rejection of burst interference such as QRN. Note that if
this control is set correctly, the full advantages of the ON7YD VCW
system are realised because the dots can be longer for the same duration
of message.
-- 
Mike, G3XDV
IO91VT
http://www.dennison.demon.co.uk/activity.htm