Class A amps

Rick: The sonic difference between Class A and Class AB amplifiers happens because transistors are not "perfect." It turns out that if you have an input voltage of less than about 0.6 volts, a transistor won't work at all; the output is nil even though there is an input. In a Class A amplifier, one ggets around this by "adding" a large constant "bias" to the input signal (about half of the maximum input voltage), so that the transistor will always be in the "active" region, where it amplifies the way one expects it to. In a Class AB amplifier, one uses a pair of transistors (or pairs, or more) to drive the output, one for the plus part of the signal and one for the minus part, as Michiel said. This way, there isn't a need for a huge constant bias, which wastes lots of power as heat when there is no signal. Instead, one just needs to add a little bias to get the transistors above the little part near zero when the transistors don't amplify. Unfortunately, it's near impossible to eliminate the little distortion just by biasing, and it will manifest especially in quiet, delicate passages. With the little Class B amplifier I built, this makes percussive elements sound like crumpling wax paper and things like flutes sound like kazoos. Certainly not acceptable, but fortunately it isn't anywhere near that bad in commercial amplifiers (or so I hope).

I have an Accuphase power ampliefier. If I play in class A. the music will be more clean and soft, but I can not play so laud as in class b

it amuses me when a class "a" amplifier does not generally have a high output. Tell that to krell .

All this discussion about operating class is really moot if it involves ASSUPTIONS. Since I have been designing amps for over 40 years I am somewhat of an expert on the subject. The most important thing is the OVERALL RANGE OF LINEARITY. This property is obviously much easier to achieve with a class A design albeit at great expense. In the last 25 years or so, this whole situation has become rather moot as the better designs by most competent engineers have long overcome the biasing problem. The sad truth of the matter is that the average audiophile layman has absolutely no way of verifying that any given unit has been correctly designed. One must un-fortunately trust the reputation of the designer and today even that is no guaratee as there are a lot of what I call dilletantes touting themselves as great engineers. The real truth is that everyone can't be "King of the mountain". James Bongiorno formerly of Dynaco, SAE, GAS, & Sumo.