> One question, though. How exactly does negative feedback reduce IMD? Is it
> by preventing or reducing overdrive or is it something else?
> Bill, W6WRT
Hi Bill,
I'll encourage the other guys answer the finer details of how NFB works...
But, for every DB of negative feedback, you get a db of IMD improvement.
And also, for every db of negative feedback, you get a db of gain reduction.
You simply need to make up for that gain reduction with more drive at the
input to the NFB stages.
Using a grid driven tetrode driving another grid driven tetrode:
A small sample of the final's output [through a 20pf capacitor] goes from
the final's plate to the unbypassed choke/resistor of the cathode for the
driver. This provides a 180 degree out-of phase signal to buck and smooth
out the distortion variations. Just like an unbypassed cathode resistor
you've seen in class A audio amplifiers.
The amplifier should be tuned up on freq BEFORE the NFB is connected. Once
the resonant tank circuits in the loop are at pure resistance, then the NFB
will be pure 180 degrees. Otherwise out-of-resonance tanks can introduce
inductive or capacitive reactance back thru the loop - the amp could take
off. I plan to use a short feedthrough insulator thru the chassis from the
final's plate to the driver cathode. Best chance for pure 180 degree path
w/o reactance, etc.
Once the NFB is connected, the amplifier will be even more stable, with a
20db gain reduction, assuming the NFB is pure 180 degrees. In addition, the
tetrode driver and final should both be neutralized, cancelling G-P
capacitance, giving an additional slight gain reduction and stability.
[Another form of NFB, really]
73,
Tom, K1JJ
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|