Recently I've been thinking of building an RF power amplifier (for HF,
as it happens), and I ran into the age-old problem where I'd love to
have high efficiency but I need to be able to reproduce AM and SSB
signals without distortion.

There are many ways to do this, but I find myself strangely attracted
to this idea that just came to me. It seems so simple and obvious
that it must be at least 50 years old, but I've never seen it before.
Has anyone seen this technique before, and/or does anyone foresee any
drawbacks to using it?

Note: the ASCII art drawing must be viewed with a fixed-width font,
such as Courier, in order for it to look right.


FEEDBACK ENVELOPE RESTORATION FOR CLASS C (or D or E) FET RF AMPLIFIER

This feedback technique allows a class C (or D or E) FET RF amplifier
to be used in some situations where a linear amplifier would normally
be required. Even though the amplifier still distorts at radio
frequencies, producing harmonics which must be filtered out, the AM
envelope is correctly restored so that the entire functional block
behaves as an efficient linear amplifier.

+--RFC--V+
|
| | ------------------
|---+-| OUTPUT NETWORK |-+-RF_OUT
| | | &LOWPASS FILT. | |
RF_IN-+-------------------||-----+---| | ------------------ |
| | |---+ |
| R | | |
| ------------ F GND |
+-| AM |---+ C |
| DETECTOR | | |\ | |
------------ +--|+\ | gate bias adjusts v
| ---+ conduction angle |
------------ +--|-/ to correct envelope |
+-| AM |---+ |/ |
| | DETECTOR | |
| ------------ |
| feedback matches output AM to input AM |
+-------------------------------------------------------+

David
--
http://www.qrz.com/n5izu

This method is often employed in NDB transmitters

--
Gregg
*It's probably useful, even if it can't be SPICE'd*
http://geek.scorpiorising.ca

This method is often employed in NDB transmitters

--
Gregg
*It's probably useful, even if it can't be SPICE'd*
http://geek.scorpiorising.ca