Content-Transfer-Encoding: 8bit
Message-ID: <3617AC3245C2D1118A840000F805359C01206E39@PDW-MERCURY-1>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: G4JNT Now back on LF
Date: Mon, 14 Feb 2000 09:32:59 -0000
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>

Hello all.  

After an absence of well over a year I'm now back on LF with a rebuilt
transmitter (similar to G0MRF type) and 250 Watts reliably.  Antenna
tested out - still OK ! - but as its only a 7m high Tee is probably a
bit small by modern standards on the bands.  ERP estimated at -22dBW
based on an antenna current of 1.5 Amps and effective height roughly
equal to it's actual figure.

Now using a transverter driver from an IC746 so linear modes from DSP
are no problem. 
I'm looking for LF operators who have access to a 56002EVM board in
order to try PSK08.   This is my adaptation of PSK31 for use on LF but
running 4 times slower.  The operating signal bandwidth is in the region
of 5 Hz and the data rate supports slow typing speeds - still
considerably faster than Aural CW

I also have an embyronic version of PSK01but this is more difficult to
implement on the EVM due to the need for longer length filters.  At the
moment it works but has responses every 16 Hz,  rolling off gradually.
Frequency has to be set to within 0.2 Hz or 0.1 Hz for the QPSK mode.
Frequency setting is only within 1 Hz at the moment due to the
oscillator accuracy in the 746, but this is stable and I can separately
calibrate to sub mHz accuracy for specific tests.

Re. Michael's comment below,   -29dB in what bandwidth ?   Remember the
393km QSO I had back in the dark ages of LF with G3PLX on 73kHz.  That
used a bandwidth of 70mHz and (Very) Slow CW with a 40 second dot
period.   The S/N for that signal would have been in the region of -26dB
in the 'normal' CW bandwidth of 300 Hz, and there were a few dB in hand.
Also bear in mind that Spectrogram type reception relies on incoherent
visual assessment of colour scales.  Proper coherent signal integration
can give probably 3 - 6 dB improvement on this - hence my interest in
being able to run tests with PSK08 and PSK01.

Noise levels on the band seem higher than they used to be a couple of
years ago.  Is this my imagination or have they really increased ?

Andy  G4JNT


> Who has experience with <-29dB to noise level signal identification?
> CoherentCW, PSK, SlowCW, dedicated DSP algotithms? There are lock-in
> amplifiers using similar techniques maybe we can benifit there.
> 
> Best 73's and gl on lf
> Michael Sanders, PA3BSH
> 
> 

-- 
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.