What is a Class D amplifier — and why does it appear in everything from a £30 Bluetooth speaker to a £3,000 audiophile power amplifier? The answer is that Class D amplifies audio in a fundamentally different way. Instead of linearly following the input waveform, Class D switches its transistors on and off at very high speed — typically 300kHz to 1MHz. That switching signal is then used to reconstruct the audio. Consequently, a Class D amplifier wastes very little power as heat. It delivers high output from a small chassis and — in modern implementations — sounds genuinely excellent.
That last point is where the story gets interesting. For most of audio history, Class D carried a reputation for harsh, digital-sounding reproduction that audiophiles avoided categorically. Early implementations in the 1990s and early 2000s earned that reputation fairly. Modern Class D has closed the gap to Class AB dramatically. Advanced modulation, precision feedback loops, and high-quality output filters have enabled this — and in some measured dimensions, Class D has surpassed Class AB. Understanding why requires understanding how the technology actually works.
Quick Answer: A Class D amplifier is a switching amplifier that converts audio signals into high-frequency pulses rather than linearly amplifying them. It is highly efficient — typically 85–95% versus 50–65% for Class AB — which means less heat, smaller chassis, and lower running costs. Modern Class D designs measure and sound excellent. The trade-offs compared to Class A or Class AB are subtle rather than fundamental — and depend more on implementation quality than topology.
How Class D Amplifiers Work
The switching principle
At the heart of Class D amplification is pulse-width modulation (PWM). Here is what happens: the audio signal is compared against a high-frequency triangle or sawtooth carrier wave. Specifically, wherever the audio signal exceeds the carrier, the circuit switches the output to the positive supply rail. Where it falls below, the output switches to the negative rail or ground. As a result, the comparison produces a stream of rectangular pulses. Each pulse encodes the audio amplitude at that instant — hence “pulse-width modulation.”
Furthermore, these pulses switch at a frequency far above the range of human hearing — typically 300,000 to 1,000,000 cycles per second. Because the transistors are fully on or fully off, waste is minimal. Audio engineers call Class D a “switching amplifier.” Additionally, a low-pass filter at the output — an inductor and capacitor — smooths those pulses back into the audio waveform. Additionally, the speaker itself acts as a partial filter because of its inductance. What arrives at the speaker terminals is a high-power replica of the original audio signal.
Why switching is more efficient than linear amplification
In contrast, in a Class A or Class AB amplifier, the output transistors are always conducting to some degree — they operate in their linear region, which means they are always dissipating power as heat. In essence, the transistor acts like a variable resistor. As it passes more current, it drops voltage across itself — and that voltage drop multiplied by current equals wasted heat. Indeed, this is inherent to linear amplification — you cannot eliminate it within the topology.
However, in a Class D amplifier, the output transistors are either fully on or fully off. When fully on, the voltage drop is very small — a fraction of a volt. When fully off, no current flows at all. In both states, the transistors waste very little power as heat. Consequently, the transistor only dissipates significant power during the brief switching transitions, and those transitions happen very quickly. The result is efficiency ratings of 85–95% for Class D versus 50–65% for Class AB and 15–35% for Class A.
Notably, the output filter is the critical component
Specifically, the low-pass filter at the output of a Class D amplifier is one of the most critical components in the design. It must pass audio frequencies cleanly up to 20kHz while attenuating the switching frequency above 300kHz. Doing so without introducing phase distortion or audible frequency response errors is the engineering challenge. The quality of this filter is one of the primary factors that determines the audio quality of a Class D design. In practice, a poorly designed output filter produces the harshness and high-frequency smearing that characterised early Class D implementations. A well-designed filter is essentially transparent in the audible band. This is the standard achieved by modern audiophile-grade Class D designs.
Class D vs Class AB vs Class A
Understanding the amplifier class system
Amplifier classes describe how the output transistors are biased — that is, how much current flows through them when no signal is present, and how they handle the full range of the audio waveform. The classes are not a quality ranking; they describe topology. Indeed, a poorly implemented Class A amplifier sounds worse than an excellently implemented Class D. The classification tells you about efficiency and operating principle, not about sonic quality directly.
Amplifier class comparison at a glance:
- Class A: Output transistors conduct for the full 360° of the waveform cycle. Extremely low distortion, runs hot, very inefficient (15–35%). Used in high-end audiophile equipment where cost and heat are secondary concerns.
- Class AB: Output transistors conduct for slightly more than 180° each — one transistor handles the positive half-cycle, one the negative, with a small overlap to eliminate crossover distortion. Efficient enough for practical use (50–65%), lower distortion than Class B. The dominant topology in consumer and professional audio for decades.
- Class D: Output transistors switch fully on and off at high frequency. Highly efficient (85–95%), runs cool, compact. Modern implementations achieve distortion figures comparable to high-quality Class AB. The growing topology for portable, integrated, and high-power applications.
- Class H and Class G: Variants of Class AB that modulate the supply voltage dynamically to reduce heat at lower output levels. Common in professional audio and some high-power home amplifiers.
The crossover distortion question
However, Class AB has a historical weakness: crossover distortion. This non-linearity occurs where the positive transistor hands off to the negative transistor in the output stage. Notably, modern Class AB designs manage crossover distortion very effectively through careful biasing and feedback, but it remains the characteristic measurement signature of the topology. Meanwhile, Class D has its own distortion signature — switching frequency harmonics and output filter artefacts — different in character and frequency range. Neither topology is inherently distortion-free; both require careful engineering to minimise their characteristic artefacts.
The tube amplifier comparison
Tube (valve) amplifiers operate on a completely different principle — thermionic emission rather than semiconductor switching — and are outside the Class D technical framework entirely, though they can be configured in Class A or Class AB biasing schemes. Notably, the sonic differences are profound and well-documented. Tube amplifiers produce characteristic even-order harmonic distortion that many listeners find musically pleasing. Class D, by contrast, aims for low distortion across all harmonic orders. The full comparison is in the tube vs transistor amps guide.
Efficiency — Why It Matters More Than You Think
The practical consequences of efficiency
An amplifier’s efficiency rating determines what fraction of mains power reaches the speakers versus how much becomes heat. At 90% efficiency, a 100W Class D amplifier draws roughly 111W from the mains. The remaining 11W becomes heat in the amplifier. At 60% efficiency, a comparable Class AB unit draws roughly 167W — 67W of that becomes heat. Furthermore, the practical consequences are significant in several directions.
Indeed, chassis size is the most visible consequence. Consequently, heat must be dissipated through heatsinks — the larger the heatsink, the more heat it can absorb and radiate. In particular, Class A amplifiers require enormous heatsinks. Some 50W designs have heatsinks the size of a small car radiator. Class D amplifiers produce so little heat that they require only small heatsinks or sometimes none at all, enabling genuinely compact chassis at high power outputs. A 500W Class D amplifier can fit in a chassis smaller than a 50W Class A design. Understanding how wattage ratings relate to real-world listening levels the amplifier wattage guide explains in detail.
Running costs and environmental impact
Moreover, for always-on environments — a dedicated listening room, a recording studio, a commercial installation — power consumption differences become financially significant over time. A 50W Class A amplifier drawing 200W continuously costs considerably more annually in electricity than a Class D equivalent drawing 20W at the same listening level. For most listeners this is minor. But for audiophiles who leave equipment powered continuously for thermal stability, the efficiency difference is real and cumulative.
Battery-powered and portable applications
Furthermore, Class D’s efficiency advantage becomes decisive in battery-powered applications. A portable Bluetooth speaker, a battery-powered headphone amplifier, or a car audio system all depend on drawing the minimum possible power from their energy source. Class AB in a battery-powered device would drain batteries far faster than Class D for the same output level. This is why virtually every modern portable audio device uses Class D internally. It is not a compromise — it is the only topology making battery-powered high-fidelity audio practically viable.
How Class D Sounds
The early reputation — earned but outdated
Class D amplifiers gained their reputation for harsh, brittle, and “digital-sounding” reproduction from genuinely poor early implementations in the 1990s and early 2000s. The output filters in those designs were inadequate, the switching frequencies were too low, and the feedback mechanisms were insufficiently sophisticated. Listening tests confirmed that experienced listeners could identify Class D amplification and rated it sonically inferior to equivalent Class AB designs. That reputation was accurate for those products at that time.
However, the landscape changed substantially through the 2010s. Switching frequencies increased, GaN transistors replaced silicon MOSFETs, feedback topologies improved, and output filter design advanced significantly. Blind listening tests with Hypex, Purifi, and Ncore amplifiers consistently show experienced listeners cannot reliably tell them from high-quality Class AB designs at matched levels. This is not marketing language; it reflects genuine engineering advancement in the technology.
What modern Class D measurements show
Consequently, modern audiophile-grade Class D amplifiers measure extraordinarily well. THD+N (total harmonic distortion plus noise) figures below 0.001% are achievable across the full audio bandwidth. SINAD (signal-to-noise and distortion ratio) scores above 120dB place the best Class D designs at the very top of amplifier measurement rankings. Output impedance is very low, enabling tight control of speaker drivers. Frequency response is flat to within a fraction of a dB from 20Hz to 20kHz. The measurements of the Purifi 1ET400A module — used in amplifiers from several audiophile manufacturers — consistently outperform most Class AB designs at any price.
Where subtle differences remain
Nevertheless, the honest position is that subtle differences between topologies remain audible to some experienced listeners under controlled conditions. Class A amplifiers, in particular, have a characteristic smoothness in the upper midrange that some listeners find more natural than Class D’s presentation. Whether those differences reflect genuine audible distortion components or the listener’s familiarity with a particular sound is a genuinely contested question in audio science. What is not contested: the gap has narrowed dramatically. It is subtle rather than obvious, and at matched price points Class D often measures better than Class AB competitors.
Where Class D Is Used
Consumer audio
Indeed, Class D now dominates portable and compact audio applications. Every Bluetooth speaker uses Class D internally. Most soundbars use Class D. The majority of AV receiver power stages in the last five years have incorporated Class D elements for the surround channels. Compact integrated amplifiers use Class D almost universally. The topology enables meaningful output power from physically small chassis — a key reason desktop mini amplifiers have grown in popularity. Understanding whether an integrated amplifier or a separate power amplifier makes more sense for your system is covered in the integrated vs power amplifier guide.
Professional audio
Similarly, Class D dominates professional live sound amplification. Line arrays in a concert venue — driven by tens of thousands of watts — use Class D almost exclusively. The weight and heat reduction over equivalent Class AB designs at that power level are operationally transformative. A Class D rack weighing 20kg replaces an equivalent Class AB rack weighing 80kg. In touring sound production where equipment travels daily, that difference is significant. Crown, Lab.gruppen, and QSC all manufacture touring-grade Class D amplifiers that deliver consistent, reliable performance across tens of thousands of hours of use.
Audiophile applications
Meanwhile, the audiophile market’s adoption of Class D has been more gradual and more contested. Brands like Merrill Audio, Mola Mola, and Acoustic Imagery produce reference-grade Class D power amplifiers at several thousand pounds. These compete directly with the finest Class AB and Class A designs. The Hypex Ncore and Purifi modules have been particularly influential. Several boutique manufacturers buy these OEM modules, build their own chassis and power supplies around them, and produce genuinely excellent amplifiers at competitive prices. Meanwhile, budget chips like the TPA3116 and TPA3255 have brought legitimate amplification to sub-£50 — impossible with Class AB at that cost.
Should You Buy a Class D Amplifier?
When Class D is the obvious choice
Specifically, Class D is the correct choice for portable audio, desktop systems, and any application where compact size and low heat matter — not just acceptable, but optimal. A quality Class D mini amplifier delivers genuinely excellent sound from a paperback-sized chassis — cool to the touch, drawing minimal power. The best mini amplifiers available today use Class D almost exclusively. For home theater, Class D power stages in AV receivers and powered subwoofer amplifiers are standard and entirely appropriate. For budget integrated amplifiers, Class D chip-based designs from SMSL, Topping, and similar manufacturers deliver measured performance that competes with Class AB designs costing significantly more. The best mini amplifier guide covers the strongest Class D options at compact price points.
When to think carefully
However, the cases where Class D requires more careful evaluation are in high-end stereo listening applications where the last 5% of sonic refinement matters. High-sensitivity horn speakers can reveal audible noise floors in some Class D implementations that Class A designs avoid. In a critical listening environment, some experienced listeners continue to prefer Class A or high-quality Class AB for specific musical content at moderate volumes. These are genuine considerations rather than marketing-driven objections.
The implementation matters more than the topology
Ultimately, the single most important point about Class D amplifiers is that topology is not destiny. A £50 chip amplifier and a £5,000 Purifi-based design both use Class D topology. They sound dramatically different from each other. Similarly, a poorly engineered Class AB amplifier sounds worse than a well-engineered Class D design at the same price. Implementation quality determines sonic results far more than the choice of topology. The output filter, power supply, feedback design, and PCB layout matter more than choosing Class A, AB, or D.
Compatibility check for Class D with passive speakers: Some Class D amplifiers — particularly early implementations and budget chip designs — interact poorly with specific speaker impedance curves. If a speaker’s impedance drops sharply at certain frequencies and the output filter targets a flat load, the frequency response can shift. This is less common in modern Class D designs, but worth verifying if pairing with speakers known for complex or low impedance curves.
Frequently Asked Questions
Class D basics
Is Class D the same as digital amplification?
No — despite the terminology, Class D amplification is not inherently digital. The “D” does not stand for “digital.” A Class D amplifier can take an analogue input and process it entirely in the analogue domain. No digital conversion is required. However, some Class D amplifiers do incorporate digital signal processing (DSP) and accept digital inputs, but the switching output stage itself operates on analogue principles. The confusion arises because Class D uses high-frequency switching that superficially resembles digital on/off states. But the process is continuous-time modulation — not discrete digital sampling.
Does Class D sound worse than Class AB?
Not inherently — and not in modern implementations. Indeed, early Class D amplifiers from the 1990s and early 2000s had genuine sonic shortcomings related to inadequate output filters and low switching frequencies. Modern Class D designs using advanced modulation techniques and high-quality filters measure and sound excellent. Blind tests with Hypex Ncore and Purifi designs consistently show experienced listeners cannot reliably distinguish them from high-quality Class AB. Whether Class D sounds worse than Class A at the highest levels of refinement is more nuanced, but the gap has narrowed dramatically.
Heat, efficiency, and compatibility
Why do Class D amplifiers run cool?
Class D amplifiers run cool because their output transistors waste very little power as heat. In a linear amplifier, the output transistors are always partially conducting. They drop voltage and dissipate that voltage drop times the current as heat. In a Class D amplifier, the transistors are either fully on or fully off. When fully on, the voltage drop is minimal — a fraction of a volt. When fully off, no current flows. Heat is only generated during the brief switching transitions, and those are so fast that the total heat generated is small. Specifically, a Class D amplifier rated at 85–95% efficiency dissipates only 5–15% of its drawn power as heat. A Class AB amplifier at 60% efficiency dissipates 40% as heat — several times more.
Technical questions about Class D
What is PWM and how does it relate to Class D?
PWM stands for pulse-width modulation. It is the method Class D amplifiers use to encode the audio signal. The input audio waveform is compared against a high-frequency triangle wave (the carrier). When the audio signal is higher than the carrier, the output transistor switches to the positive rail; when lower, it switches to the negative rail. The result is a series of pulses whose width at any moment encodes the amplitude of the audio signal at that moment. A low-pass filter at the output then converts these pulses back into a smooth, high-power audio waveform. PWM is used because fully-switched transistors (fully on or fully off) waste far less power than transistors operating in their linear region.
Can Class D amplifiers drive any speaker?
Generally, most modern Class D amplifiers handle the majority of consumer speaker loads without issue. The compatibility concern arises with speakers that have very low or variable impedance curves. Some Class D output filters target a specific nominal load, and significant deviations can cause mild frequency response shifts. High-sensitivity horn speakers can sometimes reveal noise floor issues in budget Class D designs that are inaudible with less sensitive speakers. For most speakers in the 4Ω–8Ω range with 85–92dB sensitivity, reputable modern Class D amplifiers present no meaningful compatibility issues.
Class D vs other amplifier classes
What is the difference between Class D and Class H or Class G?
Class H and Class G are Class AB variants. They use linear output stages but add dynamic power supply rails that step up as output increases. This reduces the voltage drop across the output transistors at lower levels, improving efficiency over standard Class AB without switching the output. By contrast, switching output transistors rather than linear ones is the Class D approach — achieving 85–95% efficiency. Class H achieves efficiency of around 70–80%, bridging the gap between standard Class AB (50–65%) and Class D (85–95%). Class H is common in professional live sound where a linear output stage is preferred. It sidesteps Class D’s historical output filter interaction with complex speaker loads.