A Class A amplifier is an analog circuit design where the output transistor conducts current continuously over the full 360 degrees of the input signal cycle, meaning it never fully turns off. By keeping the output device biased firmly in its active linear region at all times, the circuit entirely eliminates crossover distortion—the dead-zone glitch that occurs when one transistor hands off the signal to another in push-pull topologies. The trade-off for this pristine linearity is brutal thermal inefficiency.
The Core Concept: Biasing and the 360-Degree Conduction Angle
In a standard Class B or Class AB push-pull amplifier, you use two transistors: one handles the positive half of the AC waveform, and the other handles the negative half. They take turns turning on and off. In a true Class A topology, the DC bias current is set deliberately higher than the peak AC load current.
If your speaker demands a peak current swing of 2 Amps to hit maximum volume, your Class A output stage must be biased to idle at a minimum of 2 Amps (often slightly more). When the audio signal swings positive, current increases above 2A; when it swings negative, current decreases toward zero but never actually reaches it. Because the transistor never cuts off, there is no switching delay, no base-emitter turn-on voltage threshold to cross, and no high-frequency switching noise. You get a pure, continuous analog reproduction of the input wave.
The Math: Efficiency, Heat, and a 10W Numeric Example
The defining characteristic of Class A on the workbench is heat. Because the output devices are drawing maximum current even when you are playing no music (idle state), the power supply must deliver massive continuous wattage.
Let’s look at a worked numeric example for a standard resistor-loaded, single-ended Class A amplifier designed to deliver 10W RMS into an 8-ohm speaker.
- Target Output Power (P_out): 10W
- Maximum Theoretical Efficiency: 25% (for a resistively loaded single-ended stage)
- Total DC Power Draw (P_DC): 10W / 0.25 = 40W
At idle, with zero audio playing, the amplifier draws 40W from the wall and dissipates all 40W as heat through the heatsink. When you play music at absolute maximum clipping volume, the amp delivers 10W to the speaker and dissipates the remaining 30W as heat. Paradoxically, a Class A amp runs cooler at full volume than it does at idle.
| Amplifier Class | Conduction Angle | Max Theoretical Efficiency | Crossover Distortion | Typical Use Case |
|---|---|---|---|---|
| Class A | 360° (Continuous) | 25% (Resistive) / 50% (Inductive) | None | Audiophile power amps, headphone amps, RF pre-stages |
| Class AB | 180° to 360° | ~50% to 78.5% | Low (requires careful bias trimming) | Most commercial AV receivers, pro audio PA amps |
| Class D | Switching (PWM) | ~90% to 95% | N/A (Switching noise requires filtering) | Subwoofers, portable Bluetooth speakers, car audio |
Where You Meet This in Practice
You will rarely find true, high-power Class A topology in mass-market consumer electronics because the heat output and power consumption violate modern energy standby regulations. Instead, you meet it in specific niches where linearity justifies the thermal penalty:
- High-End Audiophile Power Amps: Brands like Pass Labs (founded by Nelson Pass) build massive, heavy Class A amplifiers like the XA series. The physical enclosure is essentially a giant radiator.
- Dedicated Headphone Amplifiers: Because headphones require only milliwatts to a few watts of power, the heat penalty of Class A is manageable on a desktop. Designs like the xDuoo TA-03S (tube) or various solid-state DIY builds use Class A to drive high-impedance planar magnetics without crossover artifacts.
- Guitar Pedal Buffers and Preamps: The input buffer stage of legendary pedals like the Klon Centaur operates in Class A to preserve the high-frequency harmonics and transient attack of the guitar pickups before the signal hits the clipping diodes.
- Low-Power RF Transmitters: In radio frequency applications where signal purity and strict spectral emission masks are required, Class A prevents the intermodulation distortion that switching or push-pull stages generate.
What it changes in your installation: Choosing Class A dictates your physical enclosure design (requiring massive extruded aluminum or forced-air cooling), your power supply sizing (transformers must handle continuous high-current draw without sagging), and your room's ambient thermal load.
Bench Scenario: Building a 5W Headphone Amp and Melting a Heatsink
To understand why Class A requires rigorous thermal math, let’s walk through a real-world bench failure.
The Setup: I was building a DIY single-ended Class A headphone amp using an IRFP240 N-channel MOSFET, targeting 5W of output into 32-ohm planar magnetic headphones.
The Numbers: At 25% efficiency, delivering 5W to the load required 20W of DC draw. The MOSFET needed to dissipate 20W of heat continuously at idle. I mounted the TO-247 package to a small extruded aluminum heatsink rated at 2.0°C/W. My bench ambient temperature was 25°C.
The Outcome: Using basic thermal math: Junction Temp = Ambient + (Power × Thermal Resistance).
25°C + (20W × 2.0°C/W) = 65°C. This seemed well within the IRFP240’s 175°C maximum junction rating. I powered it up and set the bias.
What Went Wrong: I forgot to account for the junction-to-case thermal resistance (0.8°C/W) and the case-to-sink resistance (0.5°C/W due to a poorly applied, cheap thermal pad). The actual total thermal resistance from silicon junction to ambient air was closer to 3.5°C/W.
The real junction temperature hit 25°C + (20W × 3.5°C/W) = 95°C.
As the silicon reached 95°C, the MOSFET’s gate threshold voltage shifted. In a poorly compensated circuit, this causes the bias current to increase. The amp drew more current, generating more heat, which lowered the threshold further—a classic thermal runaway loop. The dissipation spiked to 35W, the junction hit 147°C, and my bench supply tripped its overcurrent protection, frying the 10-ohm gate stopper resistor in the process.
The Fix: I added a 0.22Ω, 5W source degeneration resistor. As current increases, the voltage drop across this resistor increases, effectively reducing the gate-to-source voltage (Vgs) and providing local negative feedback. This locked the bias point, stabilized the thermal loop, and allowed the amp to idle safely at a toasty, but stable, 72°C case temperature.
What People Commonly Confuse With True Class A
The term "Class A" is heavily abused in audio marketing. Here is what it actually means versus what manufacturers claim:
- "Sliding Bias" or Marketing Class A: Many commercial amplifiers advertised as "Class A" are actually Class AB designs biased very high. They might operate in pure Class A for the first 1 or 2 watts—enough for quiet listening—before the bias current runs out and the amp seamlessly slides into Class B operation to deliver higher power. True Class A does not slide; it clips abruptly when it hits its current limit.
- Single-Ended vs. Class A: People often use these terms interchangeably. A single-ended output stage (using only one output device per channel) is almost always Class A. However, you can build a push-pull Class A amplifier using two transistors per channel, both biased to conduct fully through the entire cycle. Push-pull Class A cancels even-order harmonic distortion and doubles the efficiency to 50%, but it is rarely seen outside of high-end tube amplification.
- Tube Amps (SET vs. Push-Pull): A Single-Ended Triode (SET) tube amp is inherently Class A. But a massive 100W push-pull tube amp (like a Marshall JCM800 or a high-end McIntosh) is almost certainly running in Class AB to achieve that power level without melting the output transformers.
FAQ: Biasing, Distortion, and Topology
Why do Class A amps sound "better" if they are so inefficient?
They don't necessarily measure better in terms of total harmonic distortion (THD) compared to a well-designed Class AB or modern Class D amp. However, because they lack crossover distortion, they eliminate high-order odd harmonics that occur at the zero-crossing point of the waveform. The human ear is highly sensitive to these high-order artifacts, making Class A sound subjectively smoother and more "liquid," especially at low listening volumes where Class AB amps spend most of their time crossing that zero-point.
Can I use a switching power supply (SMPS) for a Class A amplifier?
Yes, but with strict caveats. Because Class A draws a massive, continuous, steady DC current, the SMPS must be rated for 100% continuous duty cycle at that specific amperage. Many cheap SMPS units are rated for peak dynamic loads and will overheat if asked to deliver their maximum rated current 24/7. Furthermore, you must implement rigorous LC (inductor-capacitor) filtering on the DC rails to prevent high-frequency switching noise from modulating the audio signal, as Class A's high power supply rejection ratio (PSRR) is often poorer than push-pull designs.
How do I measure if an amp is truly Class A on the bench?
Connect an oscilloscope to the output and a current probe to the power supply rail. Play a sine wave and increase the volume. In a true Class A amp, the DC current draw from the power supply will remain virtually flat and unchanged regardless of the audio signal level, right up until the exact moment the waveform clips. If you see the DC current draw rise and fall in time with the audio signal's volume, you are looking at a Class AB or Class B amplifier.
