[Milsurplus] A Chicken Band "Leaner Applefryer" for Our Rigs

[David Stinson]

The little RCA AVR/AVT rig is running well 
for hours at a time and I've been using it 
for QSOs and nets, so I've set the vibrator problem 
aside for now and am concentrating on something
 that will help with this and several other rigs.
I will get back to the vibrator... honest (crossed fingers).

Getting these SCR-183, SCR-288 and Light Aircraft HF sets 
working is a lot of fun and I've made many QSOs 
with them "as-is," but there are a lot of times when 3-8 Watts
ain't gonna "cut it."  So I got hold of a Chicken-Band so-called 
"150 Watt, two-pill Leaner" and decided to make 
a "foot-warmer."  The trucker won't miss it.... I hope....

Very common bipolor P-P amp with a pair of MRF455s.
This thing is a real "piece of work"   (eye-roll).
No bias- it ran hard Class-C.
No base stabilization.  No output low pass filtering.
Any signal so unfortunate as to find itself forced 
to drive this train-wreck came out 
sounding like a water buffalo gargling Drano.

The first step was to stabilize the bases and bias the thing linear.
I used .004 at the bases.  Yes, I know that's alot.
I'm trading output and drive ease for stability and that settled
the thing down just fine.

Next, biasing the thing linear.  
First thing that had to go was the wienie RFC in the 
base circuit.  It was like 220 uH and, at 3.9 MC, 
might as well have been a short.  I replaced that with
a manly, non-potted, non-ferrited 1 mH CHOKE
as the Good Lord intended!

(Everything following bypassed for RF, of course)
Used a voltage divider down from the +12 V supply, 
33 Ohms 10 W  and 5 Ohms in series, the 5 Ohm 
connection point clamped with a power diode to ground
at one diode drop, about .7 volts and the bias picked-off
at this point.  This worked, after a fashion.  The amp was
fairly linear and the transistors, being mostly always on,
got just a touch warm.   There sure was a lot of  power
getting wasted as heat in that divider.

Next I tried an adjustable voltage regulator through
series diodes.  Amp looked best at about .9 volts bias,
but the transistors were "on " harder and got a little
warmer.  No "scary" warm but still.

So I tried another divider, but this time with 2-ohms
before the clamping diode to provide just a touch more
forward bias than one diode drop.  That divider still
got warm.  Until I touched it during a test and it was cold.
Turned-out my lead that provided the +12 to the divider
had fallen-off and the base circuit was self-biasing
at about 1 volt full drive.  Transistors run a little warm
but not alarming and the linearity is pretty good- acceptable.
And no heat/wasted power in a divider.

Here's the circuit:

http://home.netcom.com/~arc5/AVRAVT/newbias.jpg

OK, Smart People:  Why is this working?

My antenna tuner is being my LPF for now.
A "real" filter set is next.

73 DE Dave AB5S