> I want to implement the amplifier discretely. It can be done - its not a
big
> deal. Let's say, its a circuit of a school assignment.

Ah, now we're getting somewhere!  What exactly is the assingment task?

> I have done a modular testing of each stage, if that's what you mean.

No, I mean to *understand* each stage, not just show that it works in some
cases.

> What about feedback? I am not sure, what you mean.

Exactly my point.  Feedback is a very important concept.  You can't design
good audio amplifiers without a decent understanding of it.  I'm not going
to write an introduction to feedback here.  There is much published
literature on the subject, go read some of it.

> Are you talking in terms
> of the final stage? RE feedback is implemented whereever necessary.
>
> Please explain a bit more.

No.  Quick answers won't serve your purpose and I'd have to include too much
background information anyway.  You need to do your part.  Go learn about
feedback.

> > 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?
>
> This doesn't seem necessary to me.

I think it will be more effective to understand this circuit and a class AB
audio amp in general by starting with the final stage and working backwards.
The opamp is so that you have something to drive the final stage while
trying to understand it.

> The diodes are more to the effect of
> partially cancelling the Vbe of the transistors.

Good.  Now think about this a bit deeper.  Suppose the diodes exactly
matched the transistor Vbe.  What is the quiescent power supply current?
What would happen if the transistors got a little warmer than the diodes.
What effect, if any, does that have on the power supply current?  The power
dissipation of the transistors?  The temperature of the transistors?

> I don't think this is addressing the problem - why do you think negative
clipping happened in the
> intial circuit?

Patience.  I may be willing to help you learn occasionally, but I'm not
going to do your homework for you.  Every unexpected reaction of the circuit
is an invitation to discovery.  You are missing much bigger issues than the
negative clipping.  But if you really want to follow that, what experiments
can you contrive to figure out why it's clipping?  What have you already
tried?

Here are some specific questions to think about:  What is the output
impedence of the Q10 stage?  Does it differ for sourcing and sinking?  If
so, what are both these impedences?  What is the input impedence of the
final stage?  Does it differ for sourcing and sinking?  If so, what are
they?

> The problem is only for high wattage output. The basic
> circuit configurations are fine.

Your circuit doesn't work as expected, yet you say the basic circuit
configuration is fine.  How can you possibly know this since it obviously
doesn't work fine!!?  In fact, the circuit configuration is major screwed
up.  If you just wanted a working amp, the right answer would be to toss
this mess and start over.  However, analysing all the problems with this
circuit will lead to better understanding than just copying someone else's.

> Okaie... you are allowed to get more technical, if it helps to answer. (I
am
> an EE, just grad.)(this is not exactly my area, so a little out of focus,
> but I think I can understand you if you go more detailed)

If a simple audio circuit is "outside your area" as an EE grad, what isn't?
What school did you receive your EE degree from?


*****************************************************************
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.