paws,
you see different values of cap/resistor to give a certain amount of feedback at a certain frequency without coupling collector dc to the transistor base, there are other ways to do it but the cap resistor is most commonly seen in cb amps,
the .1uf you see in dave clones and the .01's seen in many others give very little reactance @27megs, you are mainly dealing with just the resistance,
capacitive reactance ( xc ) = 1/ 2*pi* f* c , caps also cause current to lead voltage by 90 degrees of phase,
cap/resistor feedback circuits are ok for single band amplifiers but not for broadband amps which usually add an inductor in series with the cap/resistor,
inductive reactance ( Xl ) = 2*pi* f* L, inductors cause current to lag voltage by 90 degrees of phase,
the result of the series connected capacitor/inductor is reactance X = Xl-Xc , plus the resistor gives you a impedance Z which determines the level feedback at any particular frequency,
the reactance of the cap/inductor in series with the resistor gives you a feedback curve that if designed correctly will flatten the naturally rising gain of bipolar transistors as frequency is decreased,
amps that dont use an inductor in the negative feedback such as rmitaly have a rising gain as frequency is reduced, the cap does little and what it does do is the opposite of what is needed in a negative feedback circuit if you hope to use the amp over a wide frequency range,
they are also often unstable under certain drive/load conditions and frequencies, you can make them self oscillate, even the ones they sell as hf amplifiers with filters vswr and overdrive protection bling,
better designs use L/C/r networks and should not start break dancing when you hit a sweetspot,
negative feedback has other effects, it closes an open loop, reduces gain, extends frequency response, lowers effective input and output impedance, improves damping, lowers distortion and can improve stability,
the designer must calculate the best tradeoffs based on the transistor characteristics hes going to use and the intended frequency and/or bandwidth of operation amongst other things,
i have repaired many 1446 amps that have got so hot the feedback components are fried or dropped out of the board and they dont self oscillate @27mhz like a palomar which makes me think the people behind the palomar ripoffs dont have a clue what they are doing, they are far from alone,
68ohm in series with .01uf cap is typical for cb amps using 1446 transistors, its a carryover from the days when the same amps used mrf455's which have a less enthusiastic character than 1446's but those values seem to work @27mhz, its when you go well below 27mhz and gain gets hairy that the shit can hit the fan with a big swing on your current meter and a puff of smoke,
the maths behind it all when using reactive components applied to real amplifiers are pretty complex,
if you can get your head around it thats great, if you can't then its your choice of what to try next,
i agree with boo, 1/2watt is not nearly large enough wattage, even 2w is marginal,
