Fri, 21 Jul 2000 10:03:46 +0200
To perform serious tests, not the signal from the antenna to receiver,
but the transmitter power has to be reduced.
Most HF receiver ground floor is much lower than external noise and
there's not a great meaning in reducing noise and wanted signal at same
rate and amount.
Like that the S/N remains substantially unmodified S/N and changes will
be only noted when the signal will approach the internal receiver
ground floor, or when the overall receiver gain is low and the AUDIO
gain has to be increased and delivers the "hisses" that will mask the
demodulated signal, in spite of the good front end NF and in spite of
the 250Hz IF filter used.
The test MUST be performed under this perspective.
Stated a noise floor that's mostly external to HF receivers, how much
the S/N vary and operator brain detect informations when a signal is
close to the existing noise and its power is changed.
Someone appear not to believe it, but close to the noise level, even a
single dB is noticeable amount in readibility.
It's real that a good antenna pattern would help more to receive, but
we can't force someone else changes his antenna or to move from a noisy
area to be able in hearing better our signal.
If he doesn't change his antenna, how can we enhance the way he listen
to us if not increasing our ERP ?
> ---------- Initial message -----------
> From : Ray LaRue <firstname.lastname@example.org>
> To : "email@example.com" <firstname.lastname@example.org>
> Cc : email@example.com, firstname.lastname@example.org
> Date : Thu, 20 Jul 2000 12:17:55 -0400
> Subject : Re: [TowerTalk] perspective
> Been at this business over 40 years. Been lurking the list a long
> to, but I gotta get into the fray.
> It seems like all the hyperbole about 1 db making much difference is
> easily settled. (It's amazing to me, how many people want to strain
> gnats, while swallowing elephants (or something like that)).
> Those of you that are concerned enough with the truth of the matter,
proof the question for yourself, rather convincingly. You can build a
> -1 db power attenuator which is shunted by a toggle switch.
> Operate it over a period of time during your QSO's with and without
> attenuator inserted between your exciter and your amplifier, or
> direct (if no amplifier). You will be amazed how little effect it
> have on your "contacts", at the weak signal or strong signal ends, of
> the spectrum.
> Pi-attenuator formulas calculate 2- 828 ohm and one 5.7 ohm, will
> you -1 db of loss. With CW and SSB duty cycles, 5 watt carbons should
> tolerate up to 100 watts drive for short term tests..
> Select values you can find as close as possible to this. I took some
> old carbon 5 watters that were near and filed them to the correct
> values. One six ohm and 2- 828 ohm resistor will give you -1.05 db,
> close enough. An SPST switch, 2 UHF connectors and a small box,
> complete the package.
> If you build it with short leads inside, in the smallest box that will
> hold it, it should be good to at least 30 Mhz without a lot of stray
> reactance. Make sure you use a decent DVM for the resistor checks.
> instrument you use will determine your tolerances and thus accuracy,
> your tests.
> Try each QSO with and without the attenuator inline. It's real hard
> under any circumstances to justify worrying about the difference.
> For me any DX station that is that close to the noise, I don't want to
> bother with. Most of us don't enjoy S/N ratios much less that 6 to 10
> db. I guess I don't want to struggle that much for my QSO's. I'm not
> sure that's what I would call a legitimate contact anyway.
> OK ya'll, the floor is yours. Here is a way for possible resolution
> the question. Get out you soldering irons... Let's hear the results
> your tests.
> Ray LaRue, W4BYG
> A1-Op, BPL, PSA, ARRL life, past SCM Georgia
> List Admin: aea-tempo, areu, azden, elmers, fieldday
> kachina, morrow, test-equipment, tvi-rfi-emi @qth.net
> VP Engineering, Dominion Video Satellite, Inc. (SkyAngel DBS)
FAQ on WWW: http://www.contesting.com/FAQ/towertalk
Administrative requests: towertalk-REQUEST@contesting.com