An amplifier's class (A, B, C, or D) defines the fraction of the input signal cycle during which its active output devices conduct current, directly dictating its efficiency, heat output, and signal fidelity. In a real circuit, the class you choose dictates your power supply sizing, heatsink mass, PCB layout constraints, and whether you need heavy copper pours to handle high-frequency switching noise. Before we break down the topology, let's clear up the most persistent myth on the bench: 'Class D' does not stand for 'Digital.' Class D is a high-frequency analog switching topology (PWM), not a digital-to-analog converter. It processes analog audio by modulating pulse width, and confusing the two leads to disastrous PCB layout mistakes regarding EMI and grounding.

The Conduction Angle: How Classes A, B, C, and D Work

To understand amplifier classes, look at the conduction angle—how many degrees of the 360-degree input waveform the output transistors are actually turned on. Think of it like a water valve controlling flow to a hose:

  • Class A (360° conduction): The valve is always wide open, bleeding excess water out of a bypass to maintain linear pressure. The transistor conducts through the entire signal cycle. It is perfectly linear but wastes massive amounts of power as heat just sitting at idle.
  • Class B (180° conduction): You use two valves: one opens only for positive pressure, the other only for negative. In a circuit, one transistor handles the positive half of the waveform, the other the negative. This eliminates idle current waste but creates 'crossover distortion' at the zero-crossing point where the handoff occurs.
  • Class C (Less than 180° conduction): The valve only snaps open for brief, high-pressure pulses. The transistor conducts for a fraction of the cycle. This is highly efficient but introduces massive distortion, making it useless for audio but perfect for RF applications where a tuned LC tank circuit reconstructs the sine wave.
  • Class D (Switching/PWM): The valve rapidly snaps fully on and fully off thousands of times per cycle. By varying the width of the 'on' pulses (Pulse Width Modulation), the average flow matches the desired analog signal. The transistors act as switches, not linear resistors, resulting in minimal power loss.
Bench Note: Pure Class B is almost never used in audio due to severe crossover distortion. You will almost always see Class AB, which biases the transistors slightly into conduction (around 200°) to smooth out the zero-crossing handoff, trading a small amount of efficiency for vastly improved audio fidelity.

The Thermal Reality: A 50W Numeric Breakdown

Theory is fine, but thermal management is where amplifier classes dictate your physical build. Let's run a worked numeric example for an amplifier delivering 50W RMS into an 8-ohm load.

Class A (Single-Ended)

A single-ended Class A amplifier has a maximum theoretical efficiency of 25%, but in real-world bench conditions, it often operates closer to 20%. To get 50W of audio out, your power supply must deliver roughly 250W. That means 200W of continuous heat must be dissipated by the heatsink, even when playing silence. You need a massive extruded aluminum heatsink with a thermal resistance of roughly 0.2°C/W to keep the junction temperature safe, which often weighs over 10 lbs.

Class D (Switching)

A modern Class D IC like the Texas Instruments TPA3116D2 operates at roughly 90% efficiency at 50W. To get 50W out, the power supply delivers about 55.5W. The amplifier only dissipates 5.5W of heat. This requires nothing more than a small surface-mount thermal pad on a 2-layer PCB with a ground plane. The physical footprint shrinks by 95%, and the power supply can be a lightweight switching brick instead of a heavy toroidal transformer.

Where You Meet These Amplifier Classes in Practice

Knowing where each class dominates helps you recognize them in commercial gear and plan your own projects.

  • Class A: High-end studio headphone amplifiers, boutique guitar pedals, and low-noise preamplifier stages. You meet this when absolute lowest Total Harmonic Distortion (THD) and zero crossover artifacts are required, and power consumption is irrelevant.
  • Class AB: The workhorse of the 1990s and 2000s. Found in AV receivers, powered studio monitors, and guitar amp heads (like the classic LM3886 chip amp builds). It bridges the gap between Class A fidelity and Class B efficiency.
  • Class C: HAM radio transmitters, RF oscillators, and induction heaters. You will never see this in an audio signal path. It requires a resonant LC circuit to filter the harsh pulses back into a clean sine wave at a single specific frequency.
  • Class D: Bluetooth speakers, active subwoofers, soundbars, and battery-powered portable gear. According to Texas Instruments' audio topology guides, Class D has entirely replaced Class AB in consumer and prosumer portable audio due to its thermal and battery-life advantages.

Decision Tree: Picking the Right Amp IC for Your Build

Stop guessing and use this decision matrix to select the exact topology and part number for your next workbench project.

Application Scenario Primary Priority Recommended Class Concrete IC / Part Pick
Battery-powered portable Bluetooth speaker Max efficiency, minimal heat Class D TI TPA3116D2 (Up to 50W, widely available on breakout boards)
HAM Radio 100W RF Transmitter final stage High power RF efficiency, tuned output Class C (or AB for linear SSB) NXP MRF101AN (LDMOS transistor for HF/VHF RF stages)
High-fidelity desktop headphone DAC/Amp Ultra-low noise, zero crossover distortion Class A (Output Stage) TI OPA1612 (Bipolar op-amp configured for high Class A bias current)
DIY Audiophile Chip Amp (Vinyl listening) Warmth, simple layout, good THD Class AB TI LM3886 (Classic 68W 'Gainclone' topology, requires large heatsink)

Common Layout Pitfalls for Switching Amplifiers

If your decision tree points you to Class D, you must respect the physics of high-frequency switching. The most common reason a DIY Class D build fails or sounds like a buzzing hornet is poor PCB layout.

  1. The Output Inductor Placement: Class D amps use an LC low-pass filter (typically a 10µH shielded inductor and a 1µF film capacitor) to strip away the 300kHz+ PWM carrier frequency before it hits the speaker. Place this inductor as close to the IC output pins as possible. If you route a long trace to the inductor, that trace becomes an antenna, radiating EMI that will couple into your I2C, SPI, or ADC lines.
  2. Ground Plane Splits: Never route high-current PWM return paths across the same ground plane used by your sensitive analog inputs (like a DAC's I2S lines). Use a solid, unbroken ground plane and place your decoupling capacitors (100nF ceramic + 100µF bulk) within 2mm of the IC's VCC pins to provide local charge storage for the rapid switching transients.
  3. Snubber Networks: If you see high-frequency ringing on your oscilloscope at the switch node, you need an RC snubber. A typical starting point is a 10Ω resistor in series with a 1nF ceramic capacitor placed directly across the switch node and ground. This damps the parasitic inductance of the PCB traces.
Safety & Code Caveat: When designing high-power Class D or Class AB amplifiers that connect directly to mains-derived power supplies (e.g., +/- 50V DC rails from a toroidal transformer), ensure proper creepage and clearance distances on your PCB. Mains voltage work requires fused primary sides and proper chassis grounding to prevent lethal shock hazards. Always defer to local electrical codes for chassis bonding.

Frequently Asked Questions

Why do guitarists still prefer Class A and AB amplifiers over Class D?

Electric guitar relies on harmonic distortion and natural compression for its tone. Class A and AB tube or solid-state circuits clip 'softly,' generating even-order harmonics that sound musical to the human ear. Class D amplifiers, when overdriven, clip 'hard' and generate harsh, high-order odd harmonics. While Class D is catching up in clean pedalboard amps, the physical interaction between a tube's Class AB output stage and a speaker's voice coil remains the gold standard for guitarists.

Can I use a Class D amplifier to drive a subwoofer?

Yes, Class D is actually the optimal choice for subwoofers. Subwoofers require massive amounts of power to move heavy cones at low frequencies, but human hearing is less sensitive to high-frequency THD (Total Harmonic Distortion) at those ranges. The slight high-frequency noise floor inherent in Class D switching is completely masked by the subwoofer's physical low-pass filtering and the acoustic limits of the driver. For a 300W subwoofer build, a Class D module like the ICEpower 1200AS is the industry standard.

What is 'Class G' or 'Class H'?

These are essentially Class AB amplifiers with a dynamic power supply. Instead of running the output transistors off a single high-voltage rail (which wastes power as heat during quiet passages), Class G/H uses multiple voltage rails or a continuously variable rail. When the audio signal is quiet, it runs off a low-voltage rail for high efficiency; when a bass transient hits, it instantly switches to a high-voltage rail to prevent clipping. You frequently meet these in heavy professional PA amplifiers where rack weight and thermal limits are critical constraints.

For deeper reading on switching topologies and filter design, the Electronics Tutorials amplifier guide and All About Circuits' Class D primer provide excellent schematic-level breakdowns.

The Default Recommendation: If you are building a battery-powered device, a compact desktop speaker, or an active subwoofer in 2026, default to a Class D topology using the TI TPA3116D2 or a similar modern switching IC. The thermal savings, PCB footprint reduction, and battery life extension far outweigh the layout complexity for 95% of maker projects. Reserve Class A and AB strictly for high-end headphone staging or specific guitar tone applications where heat and power draw are acceptable trade-offs for absolute signal purity.