Message-ID: <004c01bee76d$2decd5a0$5802063e@default>
From: "Alan Melia" <Alan.Melia@btinternet.com>
To: "rsgb_lf_group" <rsgb_lf_group@blacksheep.org>
Subject: LF: No Cluster postings so Alan,s Ramblings & a new waterfall program
Date: Sun, 15 Aug 1999 23:23:53 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Hi all, disappointment this week I am afraid, there is a big NIL on postings
to the local cluster. The system was rebooted on Friday by the look of the
banner. This was probably due to local thunderstorms and power glitches, but
no postings to either 136 or 1800..

Not a great deal of personal logging done either. On Friday evening at abt
2000Z, I heard Lech G3KAU (who could miss him!!) working Dave G3YMC
....sorry
still can't hear you Dave. At least it shows my rx set-up is consistent
(even
if its consistently poor)

On Saturday no QRS logged, and at 0900Z I went to listen for Mike on
71.8kHz.
Sad to say nothing heard, but I did have a highish noise level on that
frequency. This was a sweeping 50Hz modulated signal which gave a
herringbone
effect on the waterfall display. Probably local but not under my control, as
I
switched everthing in sight off. Listened again on Sunday for Mike but again
not the faintest sign from him. I obviously need to do more work on the 73k
loop.

Sunday
Heard ON4ZK on QRS calling .....PV at 0925Z but no sign of a 2nd over, I
guess this was a sked with Marco. G3XTZ worked PA0CC at 0845 on normal cw,
and heard Dick PA0SE calling cq later. It seemed fairly quiet here and I
was surpprised at what seemed a low level of activity, probably the holiday
season effect.

Remembering the comments about detuning 'T' aerials and longwires when using
a loop for rx, I thought I would recount a silly episode from this weekend.
I
often get a low level of 70Hz harmonics from the VGA monitor displaying on
the waterfall (FFTDSP4). I have used this a a confidence booster that the
sensitivity is normal. The comb is not audible but would be 'M' as a QRS
signal. Because of the lightening I disconnected the local TV and
2m co-linear feeders and connected them to the central heating rad (probably
not a good idea if I ever do get a main strike!) I became aware that the
noise level on the AOR 7030 reduced if I connected the RF ground to an
earthed
unused computer chassis (good!). Reading the manual a little more carefully
it has separate RF and DC/Audio grounds. A little later I reconnected the 2m
co-linear to the 2m FM rig and found that the 70Hz comb reappeared.
The rig is fed from a 12v psu which is earthed (and metal cased, passes PAT
test), but neither the +12v or return side are actually bonded to ground. A
little further checking showed that the comb was getting into the rx via
some
common coupling in the mains earth, because removing the psu plug also
removed the
comb. I also noticed changes in noise level as I moved wires near a 4 way
distribution block. It just goes to show that a good RF earth is important
for RX as well and some care in routing leads and power feeds will pay off.
Signal levels do not seem to have declined, but it seems as though the
evening noise level which could be affected by TV/Video PSUs seems to be
less
troublesome, at least on 136. I had expected to have trouble with earth
loops
around the RX to computer connections, but these have not surfaced. I also
came across a commercially produced (but not moulded plug) 'IEC lead' that
did not have a through earth connection (that one slipped though my PAT
test!) This gave me a tingle when I went to connect up the audio from the rx
to the computer sound card on the machine it was powering. SO BE CAREFUL, I
GUESS IT COULD HAVE COST ME A SOUND CARD, if I had not realised the problem.
I have seen 2 core moulded cable advertised, probably for use with
'double-insulated' kit, but I am not sure whether you can tell this just by
looking at it.

I have also downloaded a copy of Spectrogram 5.09 and was hoping to run it
on a 2nd computer, in parallel with FFTDSP4. (pity there were not more qrs
signals this weekend)
I did a test with a locally generated weak signal, and found that the
indicated S/N on Spectrogram was about 10dB better than FFTDSP4. That was
with a 1.3 Hz resolution on 'gram and 2Hz on FFTDSP4. I must admit the
screen
did not look that much better. I look forward to a longer test.

My parameters were 5K6/16bit/mono,60dB,200mS,4096, 300-644, avg=1. I found
that unattended logging uses an enormous amount of disc, and I prefer FFTDSP
for this purpose (12 hours is abt 25Meg) It will be interesting to get
pictures for a weak QRS station to compare, particularly as you seem to be
able to reprocess the 'gram file with different parameters.

I have also found a shareware package (30days) called Analyser 2000 from
Brownbear
Software (German). Their download version runs for 15mins per session. There
are some nice facilities but I did not find out how to 'tune' the zoom to
the part of the
audio spectrun where the signal was! The registration fee is $98 and a
'machine ID' must be provided, which could be a little expensive if you want
to use it /P on a laptop as well.
The URL is http://members.aol.com/btf1


73 de Alan G3NYK
Alan.Melia@btinternet.com