Content-Transfer-Encoding: 8bit
Message-ID: <3.0.1.16.20010219180729.333f6882@mail.cc.kuleuven.ac.be>
Date: Mon, 19 Feb 2001 18:07:29
To: rsgb_lf_group@blacksheep.org
From: "Rik Strobbe" <rik.strobbe@fys.kuleuven.ac.be>
Subject: Re: LF: Slower and slower
In-reply-to: <3A9144F9.27870.17F5A6B@localhost>
References: <6454168.982595788787.JavaMail.imail@bronty>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

At 16:08 19/02/01 -0000, G3XDV wrote:
>There's nothing to stop you integrating a signal over days, except 
>that the frequency stability and accuracy required (not the same 
>thing) gets tighter and tighter.

With tools as ARGO you can watch a 'frequency window' that is many hundred
times the resolution, so I believe that accuracy is not so critical as the
frequency stability.
To take optimal advantage of ARGO the TX (and RX) stability needs to be
better than 1/dotlength over the duration of a dash.
So for 300 sec. dots / 600 sec. dashes the combined drift (of TX and RX)
must be better than 0.003Hz over 10 minutes. Assuming the worst case (TX
and RX drifting in opposite directions, as favoured by our old friend
Murphy) the stability of both TX and RX should be better than 0.0015Hz over
10 minutes, what equals 0.01ppm (or a drift of 0.1Hz at a 10MHz reference
oscillator).
I'm not a specialist in oscillator stability, but it seems to me that a
making a reference that fullfills this requirements is in the category
'difficult but not impossible'.

Using DFCW instead of QRSS will reduce the QSO time to abt. 1/3 and the
required stability to 1/2 (0.02ppm). More sophisticated methods (FDK etc..)
might even give better results.

Another question : how stable will the signal be after passing trough the
ionosphere ? Will the doppler shift be less than 0.01ppm ?

But please, don't let the above let anybody stop experimenting. One year
ago not even the wildest mind would have thought about a QSO between G and
VE3 that would take several days ... and now it is a fact !

73, Rik  ON7YD