Power Amplifier Classes: A, B, AB, and C

Choosing the right power amplifier is a fundamental engineering decision that balances competing goals: high-fidelity signal reproduction versus energy efficiency. Whether you're designing an audio system, a radio transmitter, or a motor driver, the amplifier class dictates the core trade-off between linearity and power loss. The four classic amplifier configurations—A, B, AB, and C—are examined through their transistor biasing strategies, which lead to distinct performance characteristics in efficiency, distortion, and application suitability.

The Fundamental Trade-Off: Linearity vs. Efficiency

At the heart of every power amplifier is an active device, typically a transistor, that acts as a controlled valve for current. How you bias this valve—the steady-state current when no input signal is present—defines its class of operation. Linearity refers to an amplifier's ability to produce an output signal that is a perfectly scaled replica of the input, with no added distortion. Efficiency, specifically collector efficiency (or drain efficiency for FETs), is the ratio of useful AC power delivered to the load to the DC power drawn from the supply, expressed as a percentage: $\eta =(P_{out,AC}/P_{in,DC})\times 100\%$. A perfectly linear amplifier wastes significant power as heat to maintain its bias condition, while a highly efficient amplifier severely distorts the signal. This inverse relationship is the central theme governing the design choice between Class A, B, AB, and C topologies.

Class A Amplifiers: Maximum Linearity, Minimum Efficiency

A Class A amplifier is biased so that the transistor conducts current continuously throughout the entire 360-degree cycle of the input waveform. The quiescent point (Q-point) is set near the middle of the transistor's linear operating region. This design ensures that the transistor never cuts off or saturates for the intended input swing, providing excellent linearity and low distortion.

The major drawback is catastrophic inefficiency. Even with zero input signal, the transistor draws full DC bias current, which is dissipated as heat. For a standard resistive load, the maximum theoretical efficiency is just 25%. This occurs only at the absolute maximum output swing; at lower volumes, efficiency plummets further. Consequently, Class A amplifiers are large, require significant heat sinking, and are impractical for high-power applications. They are reserved for situations where signal fidelity is paramount, such as in premium audio preamplifiers or low-power RF stages.

Class B Amplifiers: Push-Pull Efficiency and Crossover Distortion

To overcome the efficiency limits of Class A, the Class B amplifier uses a push-pull configuration with two complementary transistors (one NPN, one PNP). Each transistor is biased at its cutoff point, conducting for exactly 180 degrees of the input cycle—one handles the positive half-cycle, the other the negative. When one transistor is active, the other is completely off, drastically reducing DC power consumption.

The theoretical maximum efficiency of an ideal Class B stage is $\pi /4$, or approximately 78.5%. This is a revolutionary improvement over Class A. However, this comes with a significant flaw: crossover distortion. As the input signal crosses zero volts, there is a dead zone where neither transistor is conducting, causing a flat spot or notch in the output waveform. This introduces severe distortion for low-amplitude signals, making pure Class B unsuitable for audio reproduction without correction.

Class AB Amplifiers: The Practical Compromise

Class AB amplification is the industry-standard compromise, blending the best traits of A and B. Each transistor in the push-pull pair is biased slightly above cutoff, so they conduct for slightly more than 180 degrees of the cycle. This small, continuous quiescent current eliminates the dead zone, dramatically reducing crossover distortion to inaudible or acceptable levels.

The efficiency of a Class AB amplifier lies between that of Class A and Class B. At low output power, it operates similarly to Class A with lower efficiency. As power increases, its behavior shifts toward Class B, approaching the 78.5% theoretical maximum at peak output. This makes Class AB the dominant choice for almost all audio power amplifiers, car audio systems, and medium-power RF applications where both good fidelity and reasonable efficiency are required.

Class C Amplifiers: Specialized High-Efficiency Operation

A Class C amplifier is biased far below cutoff, such that the transistor conducts for less than 180 degrees of the input cycle—often between 120 and 150 degrees. It operates in a strongly nonlinear mode, producing pulsed, highly distorted output current. This yields the highest possible theoretical efficiency, which can exceed 80% and approach 90% in practice.

The extreme distortion means a Class C amplifier cannot be used with resistive loads for amplifying complex waveforms like audio. Its application is exclusively in tuned RF circuits. The output is fed through a parallel LC tank circuit (a tuned load) resonant at the input frequency. The tank circuit rings and filters the current pulses, reconstructing a clean sinusoidal output waveform from the harmonic energy. This makes Class C ideal for fixed-frequency, high-power radio transmitters, signal generators, and RF heating equipment, where efficiency is critical and the signal is a constant carrier wave.

Common Pitfalls

1. Ignoring Thermal Management in Class A Designs: A common error is underestimating the heat dissipation. A 10-watt Class A audio amplifier may draw 40 watts or more from the supply continuously, with the excess 30+ watts converted directly into heat. Failure to implement a sufficiently large heat sink will lead to rapid thermal runaway and device failure.
2. Misapplying Class C Amplifiers: Attempting to use a Class C stage to amplify AM or audio signals without a tuned output network will result in unusable, severely distorted output. Remember, Class C is only suitable for constant-envelope or pulsed RF signals where a resonant tank can filter the output.
3. Inadequate Bias Stability in Class AB: The small bias voltage in a Class AB circuit is critical and highly temperature-sensitive. A poor design without proper thermal tracking (e.g., using a diode or transistor mounted on the heat sink to compensate) will cause the quiescent current to drift. As transistors heat up, current increases, leading to further heating and potential thermal runaway, or conversely, colder temperatures can reintroduce crossover distortion.
4. Overlooking Crossover Distortion in Class B: While textbooks often calculate ideal Class B efficiency, real-world designs must account for the non-zero turn-on voltage of transistors (about 0.7V for BJTs). This makes the dead zone and resulting distortion even worse than in the ideal model, necessitating the move to Class AB biasing for any fidelity-sensitive application.

Summary

• The class of operation (A, B, AB, C) is defined by the transistor's conduction angle, which creates a direct trade-off between linearity and efficiency.
• Class A conducts over 360°, offering perfect linearity but a maximum of 25% efficiency, making it suitable only for low-power, high-fidelity stages.
• Class B uses a push-pull pair conducting 180° each, achieving up to 78.5% efficiency but suffering from crossover distortion at the signal zero-crossing.
• Class AB applies slight forward bias to a push-pull pair, creating a conduction angle slightly over 180°. This practical compromise minimizes crossover distortion while maintaining good efficiency, making it the standard for audio and general-purpose power amplification.
• Class C, biased for conduction under 180°, achieves the highest efficiency (>80%)