Matching source and input impedances in power amplifiers

[Joerg]

Hello Ken,

It depends on how you define "output impedance". If you define it as the
variation in output voltage vs load current, the impedance is as I stated.

That's how I define it.

The protection reduces the output voltage when the load current increases
therfor, it is part of the output impedance of the circuit.

Not really. In an RF amp such protection schemes are typically much
slower than the slew rate of the amp, usually by orders of magnitude.
That is one reason why, for example, an over-voltage condition after an
antenna cable break blows the final transistors. The protection circuit
isn't fast enough to curb power.

It is possible to design true feedback that sets the impedance as you
describe but I have rarely seen that in RF gear. Usually there is just a
fwd-ref bridge plus a current shunt and those signal to a uC or (slow)
discrete parts.

You changed the output impedance by those modifications. If you hadn't
made them, the output voltage would have been less at that load current.

Not the short term "true" output impedance. In this quite typical case
the bridge and shunt signals went into multipliers (fast), then into a
micro controller (really slow compared to the RF). Only after the uC had
executed its respective code would the protection scheme do something.
Probably in the order of milliseconds.

Actually, the shunt signal also had an override path, curbing power
while bypassing the uC (still slower than the RF stage though). But this
only means that the output impedance was curbed to a lower limit for the
high power range. When operated at, say, 20% power it wouldn't do
anything to prevent you from connecting that 30ohms load versus 50ohms
nominal.

 

[Ken Smith]

Hello Ken,




Not really. In an RF amp such protection schemes are typically much
slower than the slew rate of the amp, usually by orders of magnitude.

Yes, I am aware of this. But you will note that the way I defined output
impedance is that the load is varied slowly while making the measurement.
This is how I defined what I meant by the "output impedance".

Not the short term "true" output impedance.

If you called it just "the short term" I'd agree. The long term output
impedance of a very non-linear situation is just as much the "true"
impedance as the short term impedance.