> The basic idea is the following:
> 1. Multistage amplifiers are used to achieve the desired gain. Certain
> stages have been inserted to 'compensate' for input and out loading
factors.
> 2. The problem is probably with the output (final) stage. I have inserted
> the quasi-complementary class AB, if this is removed and a more "standard"
> (the more common in text) circuit (see attached). I have not changed the
> resistors, but they should probably be much lower (for the a higher power
> output such as (say) 4 W). Say about 50 ohms (this would be R8_ and R9_ in
> the attachment).
> 3. The circuit seems operational if much lower power outputs are desired.
> (Biasing of certain stages have been done from the previous stage, but it
> should not really matter, the output till just before the penultimate
stage
> is fine).
>
>
> > There are so many things wrong from that point of view that I don't know
> where to begin.
>
> Perhaps this is so. I am not really sure. It may have been that this is a
> very text oriented circuit. To achieve the given specifications, the
initial
> stages were designed, additional stages placed in to take care of loading
> effects. The final stage inserted to achieve better power efficiency, the
> diodes to take care of partial cross-over distortion (not completely, for
> obvious matching defects in discrete implementation).

Why are you building this circuit?  Are you just trying to get to a working
amplifier, or are you trying to learn about electronics.  If the former, buy
a module of have someone else do it because you are in way over your head.

If the purpose is to learn something about electronics, then doing a project
like this is a good idea.  However, you have to take it one step at a time.
Stop trying to get an overall result and instead focus on undertanding every
piece of this circuit, one stage at a time.  You will also need to learn
more about electronics and amplifiers as you go along.  Nobody here has time
to walk you thru a couple semesters of electronics theory, so you need to
find a book, take a course, or whatever on your own.  In particular you seem
to be missing the whole concept of feedback.  We can help with specific
questions, but you have to understand enough of the theory first to make
sense of the answer.

The circuit you attached of the new final stage makes a lot more sense than
the previous one.  I suggest you start with it by driving it from an opamp
and watching the various waveforms on a scope.  Then take out the diodes and
see what happens.  Can you partially correct for this effect using feedback?
Is it oscillating now?  If so, why?  How do you stop it?


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.