What is the difference between Class AB amplifier and Class D amplifier?

2025-10-27 14:56:24 25

This is a classic comparison in the world of audio electronics. The fundamental difference lies in how they amplify the signal: Class AB uses linear amplification, while Class D uses switching amplification.

Here’s a clear, engineer-friendly compare—plus real parts you can buy.

What is the difference between Class AB amplifier and Class D amplifier?

Big picture

  • Class AB (linear): Output devices conduct as analog valves with a small bias so both halves overlap → low crossover distortion. Power is burned as heat when output transistors drop voltage while passing current.

  • Class D (switching): Output devices are on/off switches driving a PWM (or similar) waveform → LC filter (and the speaker’s inductance) recovers audio. Very little time spent in the lossy linear region → high efficiency.

How they differ (at a glance)

Aspect Class AB Class D
Output stage Linear emitter/source followers with bias (push-pull) Half/Full-bridge MOSFETs switching (PWM/ΔΣ)
Efficiency (real-world) ~35–65% at medium power (peaks ~70% near full power) ~85–95% across a wide range
Heat & size Runs hot → big heatsinks, bulky PSU Runs cool → small heatsinks, lighter PSU
Distortion “gotchas” Crossover if bias is off Dead-time & switching residue if loop/filter not tuned
Noise/EMI Low RF noise Needs good layout, LC filter, and EMI control
Idle power Moderate (bias current) Very low
Load interaction Mostly resistive model; stable on tricky loads Output filter interacts with Z_load at HF; design matters
Complexity Simple, forgiving, easy to prototype More complex: gate drive, dead-time, LC, feedback compensation
Latency Essentially none Small (µs-level), usually irrelevant for audio
Typical uses Headphone amps, studio/hi-fi classics, guitar amps Portable/battery, automotive, PA/pro audio, smart speakers

Note: Class D isn’t “digital.” It’s an analog switching amplifier that often uses PWM or noise-shaped modulation.

When to choose which

  • Pick Class AB if you want super-simple, low-EMI designs, or you’re building a low-power, ultra-low-noise pre/headphone amp and can handle the heat.

  • Pick Class D for anything power/size constrained (battery, automotive, PA). You’ll get far less heat and cheaper magnetics/PSU for the same output power.

Concrete examples

Class AB ICs & modules

  • TI LM3886 — ~68 W into 4 Ω, classic hi-fi AB power op-amp (through-hole friendly).

  • ST TDA7293 / TDA7294 — 80–100 W AB audio power amps, widely used in DIY/pro amps.

  • Onsemi NJM4556A (JRC4556) / TI OPA1622 — headphone-class AB drivers with very low THD+N.

  • Discrete AB designs — e.g., many Yamaha/Marantz integrated amps (traditional hi-fi) use Class AB push-pull stages.

Class D ICs, controllers & modules

  • TI TPA3116D2 — 2×50 W @ 4 Ω, super popular board-level Class D (great for DIY).

  • TI TAS5630 / TAS5614 — 2×300 W-class, higher-power Class D stages.

  • Infineon MERUS MA12070 — efficient multilevel Class D, great for compact hi-fi.

  • Infineon IRS2092S — Class D controller that drives external MOSFETs for >500 W amps.

  • Hypex UcD / Ncore modules, ICEpower (B&O) — audiophile/pro modules used in many modern amplifiers.

  • Portable/IoT: MAX98357A (I²S in, Class D out) for small speakers.

Design tips (quick hits)

Class AB

  • Set and thermally track bias to avoid crossover distortion but keep idle heat reasonable.

  • Use adequate heatsinking and SOA margins; consider a dual-supply linear PSU for low noise.

  • Star-grounding and short feedback loops keep hum and oscillation down.

Class D

  • Keep the loop area of the switching half-bridge tiny; use tight gate-drive and proper dead-time.

  • Output LC filter: pick L/C for your load and switching freq; watch damping (speaker impedance varies with frequency).

  • Add common-mode chokes, RC snubbers, and solid ground planes for EMI; pass CISPR/FCC with enclosure filtering.

  • Close the feedback after the LC (post-filter feedback) if you need better load-invariant response (many modern parts do this internally).

Rule-of-thumb math

  • Heat (AB): Pdiss≈Vdrop⋅Iout — worst near mid-power; plan big heatsinks.

  • Efficiency (D): dominated by switching + conduction losses; efficiency stays high over a broad power range, so battery life and thermal headroom are much better.

Harendra Kumar
Harendra Kumar
Harendra Kumar holds a Ph.D. in Electrical Engineering with a specialization in power electronics. His academic expertise and years of experience allow him to break down complex concepts into clear, actionable information for his audience. Through his work, he aims to bridge the gap between advanced technology and its real-world applications. Harendra is an accomplished writer who specializes in creating high-quality, long-form technical articles on power electronics for B2B electronics platforms. His content combines deep technical knowledge with practical insights, making it a valuable resource for professionals in the electronics industry.