Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C0175286C@PDW-MERCURY-1>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: RE: Spectran, Spectrogram and extreme QRS - a test ?This mail
 is UNCLASSIFIED
Date: Mon, 17 Apr 2000 12:16:33 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

I have not played with Spectran yet, but on the question of dot
length......
The optimum dot  will depend on the FFT and integration time of the
software.  Ideally the dot length will be a multiple of the FFT period;
three periods will give perfect resolution unless the software uses too
excessive an interleave factor, two will be adequate but dots could get
masked by impulse noise (an impulse can be several seconds in this
mode).   What does the 0.03 Hz refer to, is it the FFT bin size or the
minimum resolution which is usually two to three bins depending on the
window employed.   I use either EVMSPEC for the faster rates, or the
Dopplergram software for the EVM for slower monitoring,  both have rates
at binary submultiples of 8000Hz.

30s dots is still fast CW as far as I am concerned !

If your transmitter is REALLY stable, and I mean significantly better
than 0.1ppm / 0.01 Hz and with the frequency known to this accuracy, I
should be able to do an LF Dopplergram on your signal and see what day /
night ionospheric effects are visible.  These effects do show up on the
DCF 77.5 kHz signal and 137 should show a lot more.   Here the sampling
rate of the signal will be 8000 / (2^18)  =  0.03 Hz and a 256 point FFT
allows signal monitoring in a 240uHz bandwidth.  It won't take much
notice of the 30 second dot period !
Andy  G4JNT


>Since we are used to 3 second dots with Spectrogram, it seems
appropriate 
>to go to 30 sec with Spectran at maximum resolution. 

>Now, I want to arrange a test, if someone is interested, of course. I'm
>planning to beacon for some days at low power, sending a message with
QRS in DFCW 
>mode on 137.790 / 137.792, at 30 sec/dot, with 20 sec interval. 
>I will reduce RF power to 10 Watt, giving an approximative ERP of 10
mW. With
>this power setting my 900W TX will run unattended for days (I hope...)
>, if you have a stable RX with a stable BFO !


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.