Overview
Class A/B amplifiers are the most common amplifier topology found in car audio. Understanding how they work — including their transistor pairs, dual voltage rails, signal path, and efficiency characteristics — helps explain both their strengths and their limitations. This article covers the fundamental operating principles of a classic Class A/B design, including how biasing is set to eliminate crossover distortion.
Voltage Rails: Single vs. Dual Supply
A basic amplifier may operate from a single positive supply rail, but a classic Class A/B design uses dual voltage rails: one positive and one negative (for example, +12 V and −12 V). Together, these two rails provide a full 24 V of voltage swing for the output stage.
This dual-rail arrangement is important because it allows the amplifier output to swing both above and below ground (0 V) without requiring a large output coupling capacitor to block DC — though some designs still include one at the output for protection.
Signal Path: From Source to Speaker
1. Common Ground Reference
All major components share a common ground bus: the negative supply rail, the positive supply rail return, the signal source (e.g., a phone or head unit), and the speaker return. This shared reference is fundamental to how the circuit operates.
2. Input Coupling Capacitors
The audio signal from the source enters through input coupling capacitors. These capacitors block any DC offset present on the source output and pass only the AC audio waveform into the amplifier's input stage. This is sometimes called AC coupling.
3. First Stage: Small-Signal Transistor Pair
The AC signal is fed into a pair of small-signal transistors:
- An NPN transistor (Negative-Positive-Negative) — conducts when the signal is above approximately +0.6 V
- A PNP transistor (Positive-Negative-Positive) — conducts when the signal is below approximately −0.6 V These two transistor types conduct in opposite directions, so each one handles a different half of the audio waveform. The NPN transistor amplifies the positive half-cycle; the PNP transistor amplifies the negative half-cycle.
4. Driver Stage: Intermediate Transistors
The small-signal transistors cannot directly drive a speaker — they lack the current capacity. Instead, they drive a set of intermediate (driver) transistors, which in turn control the large output transistors. This cascaded arrangement allows a very small input signal to control a very large output current.
5. Output Stage: Power Transistors
The final stage uses large power transistors — matched NPN and PNP pairs — to deliver the amplified signal to the speaker load. These transistors must handle the full output current and are typically mounted to a heatsink to manage thermal dissipation.
Class A/B Operation: Why Both Classes Are Involved
A pure Class B amplifier would switch each transistor on only for exactly its half of the waveform. This is efficient but creates a problem at the zero-crossing point — the moment the signal transitions from positive to negative. Because transistors require a small minimum voltage (approximately 0.6 V for a bipolar junction transistor) to begin conducting, there is a brief dead zone where neither transistor is fully on. This produces a characteristic distortion artifact called crossover distortion (also called a crossover notch).
Class A operation keeps both transistors conducting at all times, eliminating crossover distortion entirely — but at the cost of very high idle current and heat.
Class A/B is a compromise: both transistors are biased to conduct slightly even when no signal is present, so neither transistor is ever fully off during the crossover transition. This eliminates the crossover notch while keeping idle dissipation much lower than pure Class A.
Biasing: Eliminating Crossover Distortion
Biasing is the process of setting the idle operating point of the output transistors so that crossover distortion is eliminated. It is one of the most important adjustments in a Class A/B amplifier build or repair.
What Crossover Distortion Looks Like
On an oscilloscope, crossover distortion appears as a notch or flat spot in the output waveform at the zero-crossing point — the moment the signal transitions from positive to negative (or vice versa). The waveform looks clean on both the positive and negative halves but has a visible glitch or dead zone in the middle.
The Role of Bias Potentiometers
To eliminate this dead zone, potentiometers (bias pots) and resistors are used to apply a small standing voltage to the bases of the output transistors, keeping them slightly on at all times. In a dual-rail design, there is typically one bias pot tied to the positive rail and one tied to the negative rail.
Adjusting these pots changes how much the transistors overlap — that is, how long both the positive and negative transistors are conducting simultaneously during the crossover transition.
Setting the Bias: The Procedure
The goal is to find the point where:
- The output waveform is clean and continuous through the zero-crossing — no notch, flat spot, or discontinuity visible on an oscilloscope
- The positive and negative halves of the waveform are symmetrical — equal amplitude on both sides
- There is a controlled overlap region where both transistor pairs are conducting simultaneously, ensuring a smooth handoff Too little bias (under-biased): The crossover notch remains visible. The output waveform has a dead zone at the zero-crossing. This produces audible distortion, particularly on low-level signals and at high frequencies.
Too much bias (over-biased): The transistors conduct heavily even with no signal. Idle current is excessive, the amplifier runs very hot, and efficiency drops significantly toward Class A operation.
Correct bias: The waveform is smooth and continuous. A typical target is approximately 60% overlap — meaning both transistor pairs are conducting together for roughly 60% of the crossover transition — providing a clean handoff without excessive idle dissipation.
Practical Adjustment
With an oscilloscope connected to the amplifier output and a test signal applied:
- Start with both bias pots at minimum (transistors barely on)
- Slowly increase the positive-side bias pot while watching the waveform — the positive half of the signal will begin to clean up
- Repeat for the negative-side pot
- Adjust both pots until the zero-crossing is smooth and the waveform is symmetrical
- Verify that the amplifier is not running excessively hot at idle — some warmth is normal, but the heatsink should not become too hot to touch within a few minutes of idle operation
Efficiency Characteristics
Class A/B amplifiers are less efficient than Class D designs but more efficient than pure Class A:
| Class | Typical Efficiency | Crossover Distortion | Idle Current |
|---|---|---|---|
| Class A | ~25% | None | Very high |
| Class A/B | ~50–70% | None (when biased correctly) | Low–moderate |
| Class B | ~78% | Present | Very low |
| Class D | ~85–95% | N/A (switching topology) | Very low |
The efficiency penalty of Class A/B compared to Class D means more heat is generated per watt of output power, which is why heatsinking and thermal management are important in Class A/B designs.
Related Videos
4. Output Stage: Large Power Transistor Pair
The amplified signal from the first stage drives a second, much larger pair of transistors — the output transistors. These are the devices that actually deliver current to the speaker:
- The push transistor (PNP type) draws from the positive voltage rail and drives the output positive
- The pull transistor (NPN type) draws from the negative voltage rail and drives the output negative Together, the push-pull pair reproduce the full audio waveform across the speaker load by alternating which transistor is active on each half-cycle.
5. Output Coupling Capacitor
In some Class A/B designs, the speaker is connected through a final output coupling capacitor. This capacitor:
- Blocks any residual DC from reaching the speaker (protecting the driver)
- Passes the AC audio signal through to the speaker terminals
- Allows the speaker to swing positive and negative relative to ground as the signal alternates
Why "Class A/B"?
The "A/B" designation refers to the bias point of the output transistors — the amount of quiescent current flowing through them even with no audio signal present.
- Class A amplifiers bias the output devices so they are always conducting, regardless of signal. This eliminates crossover distortion but generates significant heat at all times.
- Class B amplifiers bias the output devices to conduct only when the signal is present on their respective half-cycle. This is more efficient but introduces crossover distortion at the zero-crossing point where one transistor hands off to the other.
- Class A/B is a compromise: the output transistors are biased to conduct a small amount of current even with no signal. This overlap eliminates crossover distortion while keeping idle power dissipation lower than a pure Class A design.
Efficiency and Heat
The theoretical maximum efficiency of a Class A/B amplifier is approximately 50%. In practice, real-world efficiency is often lower.
A key characteristic of this topology is that the output transistors and bias circuitry generate heat even with no audio signal present. You can measure current draw from the supply rails even when the amplifier is idle and no music is playing. This idle dissipation is the cost of the A/B bias point that eliminates crossover distortion.
This is in contrast to Class D amplifiers, which use switching output stages and can achieve efficiencies of 85–95%, generating far less heat for the same output power.
Key Takeaways
| Characteristic | Class A/B |
|---|---|
| Output topology | Complementary push-pull transistor pair |
| Voltage supply | Dual rail (positive + negative) |
| Signal handling | NPN for positive half-cycle, PNP for negative half-cycle |
| Crossover distortion | Minimized by small quiescent bias current |
| Theoretical max efficiency | ~50% |
| Idle heat dissipation | Yes — current flows even with no signal |
| Common application | Car audio amplifiers, home stereo amplifiers |