So you might go back to tying the emitter to gnd, and use a 10K in series with the 220 uF cap on the base.
Nah, that just killed the sound altogether (the 10K resistor).
I don't know what works best from 5000 miles away, so you have to play with values, :-).
But I would definitely keep the cap in line to the base and the R to gnd on the base [but
larger than 1K], so the amp is turned off with no signal input, and so no DC flows through
the speaker. With these in place, you can remove the 10 ohm R from the speaker line.
And, given the choice of Class A, Class AB, Class B, Class C or Class D, what would YOU call it?
The emitter resistor isn't really needed because the bias-stabilizing negative feedback comes from the fact that the top base resistor is connected to the COLLECTOR and not to VCC.
My feeling was, the original ckt was an "attempt" at a Class A linear design, but not
especially well done, ;-).
And you're right, I overlooked that the upper base R is really connected for "self-biasing"
and not normal Class A linear operation, which brought the comment about an emitter R.
So, I get a demerit for that [bad dog]. OTOH, I've tried self-biasing ckts like that, and
never found them to be very stable with beta and temperature variations. The emitter R
is the way to get real gain stability.
Also, of course, if you are going to do either bias scheme, then you need to isolate
the speaker using a really big cap, as previously mentioned.
So, would Nick's ckt as is be a Class C?