[TowerTalk] 2000 Antenna Forum Speakers

[Tom Rauch]

> >By the way, I measured some rigs using Maxwell's measurement 
> >method as well as one I devised, and indeed the rigs I measured 
> >were very close to conjugately matched when properly tuned. IMO, 
> >Maxwell is much more on target than the people desperately trying 
> >to disprove a conjugate match can exist.
> 
> I thought conjugate matches were a well-proven concept. It's news to me
> that Maxwell's work has come under such scrutiny.

I guess it depends on what you are trying to prove.

If you look at the QST curves for the ETO amplifier, they cross 50 
ohms source Zo at just over 1200 watts output. I'd wager $1000 
that was the point where the amplifier was actually tuned for 
maximum energy transfer.

If you reduce drive, the source impedance always INCREASES in a 
non-class A PA system. That's because the anode source voltage 
is fixed but the operating current decreases.

I measured this two ways:

One was by injected a small amount of signal near the operating 
frequency through a high power attenuator and observing the 
reverse injected voltage levels on a frequency selective voltmeter. 

The other was using Maxwell's method of delta E and delta I to 
measure source impedance when the load impedance is perturbed.

A PA certainly does not have to be conjugately matched, but when 
it is conjugately matched energy transfer is maximized and 
efficiency is at or near its maximum value.

I measured a class C PA at over 70% efficiency when it presented 
a 50 ohm source Zo to a a 50 ohm load. Both a standard load-pull 
and the reverse generator confirmed this.

Of course I can mistune the PA, and it won't be conjugately 
matched....but then it won't be as efficient either.


73, Tom W8JI
w8ji@contesting.com

--
FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm