From: "james moritz" <j.r.moritz@herts.ac.uk>
To: rsgb_lf_group@blacksheep.org
Date: Mon, 10 Jan 2005 16:55:13 -0000
Message-ID: <000001c4f735$272eea90$e6a4c593@RD40002>
MIME-Version: 1.0
Importance: Normal
In-Reply-To: <000001c4f727$00356120$6507a8c0@Main>
Subject: RE: LF: Re: Pre-amps on LF
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org

Dear LF Group,

I think Jan-Martin is right about the PIN diode losses - at LF they will
just work as "ordinary" diodes which won't greatly affect their loss -
although they will lose their advantages in terms of linearity and ability
to handle an RF signal larger than the bias current, so might cause more
intermodulation. However, there is the effect of the diode biasing chokes
and DC blocking capacitors - these are usually quite small (100uH, 10s of
nF), so not really big enough for LF, especially when the signal has to pass
through a number of these before reaching the mixer. The mixer itself is
also a problem, since it usually has an impedance
step-up/unbalanced-balanced transformer at the input, which is only a few
turns of wire on a small ferrite core - this will have too little inductance
for LF, making the input impedance of the mixer very low and generally
compounding the other problems. I'm sure some people have described mods
where this transformer is changed to improve LF performance, but I can't
think who at the moment. A lot of RXs seem to use a combination of low-pass
filter and attenuator for LF/MF reception, presumably to prevent the thing
being hopelessly overloaded by MF broadcast stations when a long wire is
attached, but not helpful if you are trying to receive weak LF signals. I
have also noticed that the internal noise floor tends to increase at LF - I
think this might be due to noise sidebands of the synthesiser leaking
through the mixer and getting into the 1st IF, since the LO frequency is
very close to the IF frequency when receiving a low frequency (the RA1772
actually has a crystal notch filter on the synthesiser output for this
reason).

It isn't hard to make an RX front end with good performance at LF as well as
HF, but the designers of amateur rigs haven't actually tried to do that -
they have produced a circuit designed primarily for HF, and LF reception is
just tacked on as an afterthought. However, since this isn't their main
objective, I suppose one should not expect too much!

Cheers, Jim Moritz
73 de M0BMU