Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C0175285E@PDW-MERCURY-1>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: "'LF Group'" <rsgb_lf_group@blacksheep.org>
Cc: "'KZY'" <Paul@HF-Inst.co.uk>
Subject: LF: DSP Experimentation,  Antenna book etc.
Date: Fri, 14 Apr 2000 12:16:39 +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>

If anyone wants to experiment with simple DSP techniques beyong just
plugging in Gram etc, or even wants to try some programming without
having to learn Windows API calls and how to use the soundcard,  ICs for
the serial port interface described in RadCom a good few years ago are
still available.   

For those who don't remember the article, this was an 8 bit A/D
converter digitising audio at 10000 / 7142 / 2500 Hz  and sending the
samples up the serial port to a PC.  The idea being that this was a
simple way for users to develop their own DSP type software  in standard
high level programming language without having to bother with the
digitising side of matters.   PC software available with the module at
the time generated waterfalls and spectrum plots, although I have to
admit that for these functions alone, these days you are better off
using Gram and the Soundblaster.   With slightly different PIC code for
the interface, and a change of crystal, it will also work at 7200 Hz
sampling instead of the overcomplicated sigma-delta interface specified
for the VE2IQ software

One function also available with this module was a simple downconversion
from a 1000 Hz tone to zero frequency, the output consisting of Inphase
and Quadrature components allowing phase measurements and vectorscope
type programmes to be written easily.  As far as I know, this function
is not yet freely available for the S/B (sits back and waits for howls
of refutation).

A quick plug now - a commercial version of this interface with
additional slow speed data logging functions built in, is available from
HF Instruments - see their advert in the latest RadCom and web page
HF-INST.CO.UK.  The data logging function could be useful, for instance,
for recording the AGC line over many hours while listening to a
propagation beacon in the same way that a chart recorder would be used. 

To Larry and Co. if you have a 56002EVM available I now have some
software for the PC that will let you sample a signal at 0.035 Hz (much
faster rates are also of course possible), and produce waterfalls this
wide  looking in a bandwidth of  0.00012 Hz (yes, 120uHz).  For QRSS
that would require a dot length of 4 - 7 hours and around 12 days for a
callsign.  Rather silly but quite possible! 
I produced an overnight Dopplergram plot of the DCF77 time transmission
and managed to clearly see some ionospheric spreading at night not
present during the daylight hours.  There really are visible ionospheric
effects on 77 kHz signals but the picture is marred by the few mHz drift
due to the Codec crystal changing the sampling rate as the room cooled
in the night.   If anyone wants a copy of this contact me directly on
G4JNT@ARRL.NET


After reading all comments / discussions / arguments / heated debate
about antennas recently ........
Just found a very good book on antenna design on a work colleague's
bookshelf.  

"The Handbook of Antenna Design" Vols 1 & 2  by A Rudge, K Milne et al.
ISBN 0-86341-052-9.  

Thought by many to be one of the best books on antennas there is.
After a serious bit of arm twisting I managed to borrow it from him for
a few days.   There is a decent sized chapter (over 100 pages) on VLF /
LF antennas and another large chapter on HF ones.    
More importantly it does not have pages upon pages of squiggly maths -
just simple equations that the average non-mathematician can actually
use and plenty of diagrams and  graphs to just read off the results.

And finally ...........

To those who complained they have never heard me on LF.  I don't, and
won't, operate normal speed CW.  I hate the mode but will just about use
it under protest on 10 GHz where it works well for rainscatter (and we
keep to decent speeds with much repetition anyway).   On LF I intend to
stay with the far more effective (in S/N terms even if not in
simplicity) PSK and other systems even if it means very few people to
work !    Morse might be cheap and simple, but that is it's only
advantage these days. CW may well be a skill, but not one I'm prepared
to practice at keeping up  (Sits back and waits for even more flak !)

Andy G4JNT

-- 
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.