|
capacitor in front of their solid-state amplifiers, yet have no problem with the cheesy electrolytic capacitor that terminates the feedback loop of their amplifier. Somehow, this capacitor does not count; somehow, they imagine it is not in the signal path; it is.) The second possibility is to replace the output transistor with a transistor-MOSFET compound circuit. This would require adding a P-channel MOSFET and one resistor.
|
|
|
In the above circuit, a compound circuit is displayed. This topology is known for its linearity (it contains its own short feedback loop from the MOSFET's to the transistor's emitter) and some problematic switching distortion when used in a lean Class-AB or Class-B amplifier. But in this application, wherein the MOSFET is never completely turned off, this distortion does not occur. The P-channel MOSFET requires quite a bit of drive voltage, which explains the slightly increased rail voltage.
Class-AB Buffers In spite of a few hundred pages of advertising copy, Class-A operation is brutal and not easily achieved and single-ended, constant current source loaded, Class-A operation is obscenely brutal. Thus the 99.99% prevalence of Class-AB and Class-B output stages. Yes neither class is as linear as Class-A, but given enough feedback, the amplifier will test well and sometimes even sound good. The big problem is the discontinuity at the crossover point between output devices. With transistors, increasing the idle current does not work well because of the gm doubling at the overlap of conduction.
|
|