[ICOM] Truth in advertising?

[Adam Farson]

Hi Alex,

Couple of points:

 << Once upon a time there was something called shape factor of a filter >>
There still is. Icom specify their IF filter shape factors in all their
published specifications, in term of -6 dB and -60 dB bandwidths.

<< radios at that time didn't use LNA's >>

Ah, but they did. They called them RF amplifiers. The purpose of the RF
amplifier was threefold: to overcome the mixer NF (the RF amplifier had a
lower NF than the mixer), to overcome the insertion loss of the tuned
preselector, and to prevent local-oscillator re-radiation (vitally important
in military applications.) The Telefunken E.52 "Koeln" HF receiver,
developed in 1940, had two cascaded RF amplifiers embedded in a preselector
with 5 tuned circuits. As well as protecting the mixer from overload,
cross-mod etc., the relatively narrow preselector improved the image
rejections of superhet designs with a comparatively low IF.

As you mentioned, mixers used in modern receivers tend to have a lower NF,
so the RF amplifier is unnecessary under normal conditions where antenna and
band noise is 10 to 12 dB above the receiver's noise floor in any event. But
the RF amplifier is still there, and can be switched in when needed for
weak-signal reception.

<< they come back and place the selectivity back where it belongs: as close
up front as practical, right after the mixer. >> and as permitted by the
current state of technology. In the current, up-converting designs with DSP
in the lower IF, the first IF "roofing" filter is not a
selectivity-determining element in the sense of having a BW close to the
occupied bandwidth of the received signal. It serves two purposes: it
suppresses the image response created by the down-conversion to the second
IF, and it reduces the statistical likelihood that strong out-of-band
signals will ride down the IF chain and overload the ADC.

The DSP IF filters determine the receiver selectivity; they have far lower
shape factors than can ever be achieved with a crystal filter at an IF in
the VHF range. They also offer facilities such as continuously-variable
bandwidth and -70dB tuneable notching. As long as the analogue signal power
entering the ADC is held well below the "all ones" point, the DSP filters
will do the job they were designed to do - very well.

Ideally, we would like to see DSP at signal frequency, or at least at the
first IF - but for cost and technology reasons, we are not there yet, at
least in equipment affordable to most radio amateurs.

Cheers for now, 73,
Adam VA7OJ/AB4OJ


Scanned by WinProxy
http://www.Ositis.com/