> There are very good amplifier designs that exhibit no positive (thank
> god) or negative feedback and which are more-or-less the equivalent of
> "open loop" (in op-amp terms), like the 1990 ARRL design with a
> 4CX1000A on 144 (the onr which runs the screen at RF/DC ground and
> cathode below).
Is it grid driven?
> Correct me but... a 1/2-wave length of coax at the test/operating
> frequency in question between the amplifier and radio will present the
> 'real' Zi of the amplifier to the radio and the 'real' Zo of the radio
> to the amplifier, ie. the coax and any transform that it was
> contributing to the test will vanish.
Sure that is true, but it is obviously not the concern Jim
has...unless he is using 75 ohm cable between a 50 ohm amp and
his exciter.
> Its easy to say "just make sure the amp is matched..." but afterall
> that was the point of the experiments he was doing... my point was
> that just as in the real world the amp doesn't have an input Z of
> 50R+0j nor does the radio's output have a source impedance of 50R+0j
The radio's output impedance has nothing to do with SWR readings
on the line, although I agree many solid state radios "like" load
impedances other than exactly 50 ohms.
> Have you ever 'looked into' the output impedance of a radio
> transceiver with something like a Wiltron Network Analyser -
> sure isn't 50R+0j, hence the whole 'problem' (if indeeded it is
> one) is a mixture of tuning amp input versus radio output - and
> the two interact.
You can not look into the PA and get a meaningful reading, unless
it is on. You have to do a load pull, or use a reverse source that is
unaffected by transmitter power.
> Placing a 1/2-wave coax between the two simply eliminates one
> factor, making it easier to workout what the interaction is.
Eliminating the most predicable and easiest measured item in the
system isn't necessarily a major advantage.
> In the smaller station at home I use an FT847 with a Commander-II
> 3CX800, it has 'strange' modes of tune-up - the input circuit
> interacting with the FT847 and length of coax between the two. For
> some coax lengths I 'appear' to need > 30W drive for full output while
> other lengths I need only about 18W for the same output. What's
> intersting is the difference in grid current and aparrent efficiency
> of the amplifier.
Must have a poor input circuit. What type is it? Is it a low pass or
bandpass with enough Q, and one that provides a low impedance
at the cathode for harmonics of the drive frequency??
73, Tom W8JI
W8JI@contesting.com
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|