Class B wandering toward Class A?

[treez]

A class B amplifier is a distorted (crossover distortion) , but efficient version of a Class A amplifier.

So, considering a guitar amplifier......is it always best done in Class A (from a sound point of view, not efficiency)?

....Also, if done in class B, can you use DSPs at the pre-amp stage to entirely get rid of the crossover distortion?

....Also, Class AB is to make class B like class A.......but is it possible to do a "Class BBBBA" amplifier, which is very similar to a class B, but slightly dabbling into class A.......ie, with the transistors slightly ,(very slightly) biased on when on no load)

.....has the designer complete choice over the degree of "Class A" that he/she puts into the class AB amplifier?

[Elvee]

Originally Posted by treez

A class B amplifier is a distorted (crossover distortion)

It needs not be. An accurate class B amplifier can be perfect: see example below

....Also, if done in class B, can you use DSPs at the pre-amp stage to entirely get rid of the crossover distortion?

You could, but open-loop correction relies on the perfect matching of the amplifier's non-linearity and the corrector, which will be difficult for a wide range of factors like temperature.

....Also, Class AB is to make class B like class A.......but is it possible to do a "Class BBBBA" amplifier, which is very similar to a class B, but slightly dabbling into class A.......ie, with the transistors slightly ,(very slightly) biased on when on no load)

.....has the designer complete choice over the degree of "Class A" that he/she puts into the class AB amplifier?

Yes, the designer has the choice, but class AB is always a (bad) tradeoff.
It is a mongrel, and always creates unwanted artifacts.

Class AC of J. Broskie is a better solution (but a hot one).
Another option is a sliding bias class A, like the Circlophone.

[HJWeedon]

Hi all.

As I understand it, The various classes of amplifiers were adopted with tubes and various tubes were designed to work optimally in a specific class. For Audio power the choice was single tube class A, push-pull class B. Specific grid characteristics were designed for the tubes to work optimally in a specific class. A typical class B tube was the 6N7, a dual power triode. Then came the power petodes and beam-power pentodes, those tubes worked best in something between class A and class B and class AB was adopted. The 6L6 and the 6V6 were tubes in those last categories.

The class B category tubes had characteristics that were quite linear over most of the grid characteristic, but deviated with a curved bow just before cutoff. By biasing a second tube to pick up where the other cut off, the output was still pretty linear.

Class AB became the class where both tubes in a push-pull pair contributed equally to the output at low power, By tweaking the the grid-wire spacing ever so slightly it was possible to get very high performance in the crossover region where both tubes contributed about equally.

Along come transistors, first "bipolar" transistors and later FETs. Using solid state devices it was much more difficult to tweak the transfer function to fit the application better. Class A was too hot and the transistors burned up. Class B was better, but because the cut-off characteristic was much sharper and hard to match the term crossover distortion was invented. Crossover distortion is due to the difficulties of crossing over from producing output power controlled by one transistor to producing output power with a second transistor seamlessly.

Crossover distortion is quite audible even at quite low percentage levels because it is such an un-natural type of distortion. People simply do not like that type of distortion.

There is a second equally disturbing effect from using solid state circuits in class B push pull. The solid state devices are high current, low voltage devices, whereas tubes are high voltage low current devices. At high currents power supplies ring and that causes havoc in amplifiers. Power-supplies are supposed to have little, or no, effects on an amplifiers performance. It took several 10ns of years before design engineers learned how to bypass and wire supplies.

This last problem may very well be why solid state amplifiers got a bad reputation among Audio-Files in the beginning. Amplifiers produced sounds that nobody had heard before and that ruined the reputation of many amplifier constructions.

Lots of rambling about amplifier classes and stuff.

Have a good day,

Hans J Weedon