class A,class A/B

[Cpt.Beaky]

'fraid you're just plain wrong about that. While what your illustration shows is indeed crossover distortion, what is depicted is more along the lines of what a Class B design would yield. Class AB is designed to have a little bit of bias current running so that the transistors do not have to turn on, as they are already idling. An optimally biased Class AB amplifier can have vanishingly low distortion.

The real problem is the transistor, which has a gain that changes with the amount of current flowing through it. In class A, this current is fixed, thus less variance in gain (and lower distortion). Modern audio transitors go a long way towards solving this problem, as they are more linear.

Mmm, that's interesting. I'd heard some of that, but did not know that switching was not a problem in A/B amps.

 

[gouldsborogr]

nice thread. i never really paid much attention to this in the past; after reading this, i will in the future.

 

[Shonver]

Mmm, that's interesting. I'd heard some of that, but did not know that switching was not a problem in A/B amps.

It's the same problem as ever, switching is still a problem in Class B amps, but in class AB the problem is not so much switching, but rather the gain variance I wrote of earlier. The link given above to the ESP site makes for very good reading. You can get a bit more detail from here http://www.dself.dsl.pipex.com/ampins/dipa/dipa.htm. It's a bit technical, but still worth the effort, even for the non-technically minded.