Envelope-to: dave@picks.force9.co.uk
Delivery-date: Thu, 22 Sep 2005 16:08:15 +0100
Message-ID: <001401c5bf87$516cc460$b2098351@Main>
From: "Alan Melia" <Alan.Melia@btinternet.com>
To: rsgb_lf_group@blacksheep.org
References: <B333AB0EF6D4D4118BAF00508BAED8EA088DA778@mms12014.stv.uk.astrium.corp>
Date: Thu, 22 Sep 2005 16:06:27 +0100
MIME-Version: 1.0
Subject: LF: Re: DFCW
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Hi Chris, yes this is possible, and it all depends on what crystal
oscillator you are using. You may find that the "logic inverter" style
circuit is not quite stable enough for very slow DFCW but is certainly OK at
3 and probably 10 second elements. One beauty of visual decode is thatit
doesnt matter too much if it drifts....it may insense the purists (like me
.... :-))   but if it stays on screen you can read it. You must try to avoid
chirp though....often due to RF getting back into the keying leads.

I sent FSK RTTY in the "Good Old Days" on 2m and 70cms with this technique
just FMing the crystal before multiplication to get 170Hz shift. STOP
tittering 'YKB !!

Lets assume you have a logic osc as Dave used in the original amp (I think)
Take the crystal pin that is connected to an "input" and connect a silicon
diode cathode  to a 50pF trimmer the other terminal of which is connected to
the crystal. The anode of the diode is connected to ground. Apply bias an
signal to the diode from the slider of a 10k pot and through a 10k resistor.
Decouple the top of the pot well. The trimmer should be set to the minimum
value that will give the required shift. The pot will give a fine control.
It is probably advisable not to apply a ground (0v) as one signal element.
I would suggest a signal swinging between about +2 and +5 volts. (the higher
the voltage the smaller the capacity.)  The reason for this is that the
capacity of the diode changes most around zero volts bias. You will probably
find that any small-signal silicon diode works 1N4148, 1N916 etc...... dont
use a 1N400x power diode they have too much capacity (bigger area junction
to carry the current). If you use a mixer osc using two high frequency
crstals you only need to FM one of them. But be careful of element
inversion. (I think the protocol is Highest Frequency is "Dash" )

I find that inter-element spacing is not essential, but some find it easier
to read. The gap can be sub-dot length as this give a "glitch" on the
received copy....allowing an element count to be done, and also
characterising the signal ....not everybodies element length or "weight" are
the same, and it can be useful in identifying a signal, just as one can
audibly. You will have to play with the values to get the right shift, then
you may be able to rplace the trimmer with a fixed small cap  50pF max may
be too big as the min is around 6 pF. Set it at min an just key the osc into
the rx and set up on ARGO.

If the above circuit description is noy clear I will knock up a graphic for
you.

Best of luck, I look forward to having my front end blocked by your signal
Chris.

Cheers de Alan G3NYK

 ----- Original Message -----
From: "OSBORN, Chris" <Chris.OSBORN@astrium.eads.net>
To: "LF Group (E-mail)" <rsgb_lf_group@blacksheep.org>
Sent: 22 September 2005 13:56
Subject: LF: DFCW


>
>
> My new LF TX is nearly finished and I would like to add the option
> of DFCW.
>
> I was thinking of just placing a varicap diode in parallel with the
> crystal and thus pulling the frequency
> for achieving the dots / dashes.
>
> Can someone direct me to a circuit diagram and software to achieve
> this ?
>
> 73 Chris G3XIZ
>
>
>