Owen Duffy wrote:
Hello Ian,

Ian White GM3SEK wrote in
:

...
To cut the story short, noise temperature is the only concept that
will always give correct results. As Owen points out, some of the
numbers are large and ugly - but the important thing is that they are
correct. The results can easily be converted back into a more
comfortable format... and those results will likewise be correct.

I make the observation that hams *like* Noise Figure, the the roll up of
a system component's Noise Figure into whole of system impact is often
(very often) not done well.

I was explaining to a local EME enthusiast that a certain two stage 1296
LNA that represents NF=0.51dB when the FET specs give NF=0.78dB for the
first FET alone, is very creative. When the effects of input circuit loss
and roll up of the second stage noise is included, it is unlikely that
such a preamp would have a guaranteed NF better an 0.9dB.

For a narrow band application, it is indeed possible to construct a
circuit which has lower noise temperature than the active devices. Look
up "cold fet noise source". (a quick google turns up, for instance,
patent 6439763..)



In high performance systems, I perceive a preference to not use G/T as a
metric for receive system performance.

This is hams, the preferences of which you speak? In the rest of the
microwave station world, I think G/T is a popular "one metric for all",
at least for things pointed at the sky.


Rather, hams will quote (brag) Sun