------------------------------ From: francois@mot.com (Andre Francois) Subject: Re: Amplifier Classes.... Date: Wed, 6 Apr 1994 14:36:13 GMT Let's start with amplifiers in general. The class of an amplifier usually defines how efficient the amplifier is and also what the distortion of it's output is. I work in the RF realm a great deal and we user FM power amplifiers that are class C and D and other undefined classes (or undefined to me at least) for the sake of efficiency. Now specifically in the audio realm we refer to the Class A, Class AB and Class B. I admit that I cannot go into the n'th detail of amplifier design (I have never designed an amp although I can follow a design pretty well). Class A amplifiers are designed with a one tube or transistor biased so that it is always on. If you think of the DC operating point of the collector of plate as swing from 0 to 10v and at one extreme (say 10v) the output current (collector current or plate current) is zero then the bias point with no signal applied would be 5v. This means that with no signal, a Class A amplifier is drawing half of the current it can (at 0v it would draw maximum current). This is very ineffient, but allows an open loop gain with relatively little distortion and guarantees that small signal distortion is the same as large signal (assuming our input signal is not large enough to cause clipping). Class B amplifiers are usually two tube or transitors in a push-pull configuration. That is, for the positive half of the input signal the top transitor amplifies the signal, and for the negative half of the input signal the bottom transistor amplifies the signal. What we have hear are two output devices biased to conduct NO current in standby when there is no signal. This make for an efficient amplifier. The disadvantage is in the variation in parts between the top and bottom transistor sometime requiring local as weel as global feedback to prevent one type of class B distortion. Second, the small signal response or distortion might be very distorted because there is usually a small current that the transistor or tube draws before it turn on and begins to conduct (for a transistor this is the base current I think there is a grid bias voltage for tubes?). Therefore the resolution of a Class B amplifier (the smallest signal it will amplify) will not be as great as a Class B amplifier and the two transitor variances lead to part tolerance distortion. Class AB amplifiers solve the low level signal problems of Class B amplifiers by biasing the tubes or transistors to always be on. They do not bias them in the middle of the operation point like Class A, but instead guarantee that the output devices conduct a small current during standby with no signal. This new biassing element adds distortion (the more complicated a circuit the more distortion and noise as a general rule). So in a nutshell: Class A: low open loop distortion, but inefficient (runs hot all the time). Class AB: medium open loop distortion, but medium efficiency (runs warm all the time). Class B: terrible open loop distortion and resolution, but efficient (runs cool all the time). I have an old article (I think from Audio magazine) that I will send if you request (self addressed stamped envelope please). -- Regards, Andre A. Francois Software Engineer Test Software Engineering Radio Products Group, Land Mobile Products Sector 8000 West Sunrise Blvd., RM 1C15, Fort Lauderdale, FL 33322 Phone: (305) 475-6115 Pager: 4694 Fax: (305)475-6006 Internet: EAF003@email.mot.com