From: email@example.com (Bob Wilson)
Subject: Re: Class B amplifiers: what are the large low-value resistors for?
Date: Sun, 17 Nov 2002 04:09:39 -0000
Organization: Your Organization
X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit)
In article ,
>Kevin McMurtrie wrote in message
>> firstname.lastname@example.org (N. Thornton) wrote:
>> >Paul Burridge wrote in message
>> >> On Thu, 14 Nov 2002 17:34:21 +1100, "Phil Allison"
>> >> opined thusly:
>> >> This one hasn't: it's just a breadboard experiment. BTW, there's no
>> >> reactive load, just a loudspeaker. Okay, I know even a loudspeaker is
>> >> technically a reactive load, but we're only talking low audio
>> >> frequencies here, so I guess it'd be purely resistive, yeah?
>> >> What I'm outlining here is class B *audio* of a few watts. Sorry if I
>> >> should have made that clearer.
>> >I'm thinking you're not using class B here :) More like AB, which is
>> >somewhat different. Class B would be simpler, and can in fact be made
>> >to work without any problematic level of crossover distortion.
>> >If you're willing to use geranium output trs you can do a much simpler
>> >cct that's totally stable and sounds nice. I always do that for
>> >quickie amps.
>> >Use an LM324 opamp to drive 2 small 1 watt Ge output trs, no bias.
>> >Just connect the bases together and hook them straight to the LM324
>> >output. A few Rs for the opamp feedback, one output cap, and thats it!
>> >It only works with geraniums though: silicons give terrible crossover
>> >dist. And its runaway proof, despite the lack of any obvious runaway
>> >Theres no thermal or short protection, and no capacitive load
>> >proofing. Them is extra!
>> >Regards, NT
>> The LM324 itself has terrible crossover distortion! It's meant to
>> deliver current in only one direction so there's a couple volts of slack
>> in its push-pull transistors. That's what "Single Supply" means.
>If that's true, can you explain why it worked nicely? I've heard some
>attempted explanations, but I've never been really at peace with them
>myself. I have used this cct btw, it isn't just a paper idea.
Actually, part of what he says is true. The 324 has a rather odd output
stage that transistions from Class A to Class B as output load rises. If you
have no significant load on its output, the output distortion can be pretty
bad. The answer is simply to hang a load resistor on its output. After that.
it works just fine.