A Class H audio amplifier is a hybrid design that pairs a traditional analog Class AB output stage with a dynamically tracking or multi-step power supply to minimize heat dissipation and maximize efficiency during high-power audio peaks. What this topology changes in a real installation is the thermal management profile: by drastically reducing the voltage drop across the output transistors, it allows a 1000W pro-audio amplifier to fit in a lightweight 2U rack chassis without requiring massive extruded aluminum heatsinks or noisy cooling fans. Because the audio signal path remains strictly analog, it avoids the high-frequency switching noise and complex output filtering required by Class D designs. However, builders and audiophiles commonly confuse Class H with Class D (which switches the output stage itself via PWM) and Class G (which uses discrete, fixed voltage rail steps rather than a continuously modulated tracking supply).
The Core Mechanism: Tracking Rails vs. Fixed Rails
In a standard Class AB amplifier, the power supply rails are fixed. If you are driving a speaker with a 10V peak signal, but your power rails are set at ±40V to handle occasional transient peaks, the output transistors must drop the remaining 30V. That voltage drop, multiplied by the current flowing to the speaker, is burned off purely as waste heat.
To visualize this, think of a standard Class AB amplifier like a municipal water system that maintains 100 PSI at all times, even when you only need 10 PSI to fill a glass, wasting massive energy through the pressure-reducing valve at your tap. A Class H amplifier uses a smart pump that dynamically adjusts its output pressure to exactly 12 PSI whenever you open the tap, drastically reducing the energy lost to restriction.
In a true Class H circuit, a high-speed modulator monitors the audio input signal and dynamically adjusts the power supply voltage to remain just a few volts above the instantaneous peak of the audio waveform. In modern 2026 implementations, this is often achieved using high-frequency switching power supplies (SMPS) with GaN (Gallium Nitride) FETs that can modulate the rail voltage hundreds of thousands of times per second, ensuring the tracking supply can keep up with rapid 20kHz audio transients without introducing rail-switching artifacts into the audible band.
Worked Numeric Example: Heat Dissipation at Peak Output
Let us look at the exact thermal math for a 100W RMS output into an 8Ω resistive load.
Peak Voltage (V_peak): √(100 × 8) = 28.28V
Peak Current (I_peak): 28.28V / 8Ω = 3.54A
Scenario A: Standard Class AB (Fixed ±35V Rails)
At the exact moment the signal peaks at 28.28V, the voltage drop across the output transistor (V_ce) is the rail voltage minus the output voltage: 35V - 28.28V = 6.72V.
Power dissipated as heat per transistor at peak = 6.72V × 3.54A = 23.8W.
Scenario B: Class H (Tracking Rail with 4V Overhead)
The tracking supply modulator detects the 28.28V peak and instantly raises the rail to 32.28V (output + 4V overhead). The voltage drop across the transistor is now only 4V.
Power dissipated as heat per transistor at peak = 4V × 3.54A = 14.1W.
By simply tracking the rail, the Class H topology reduces the peak thermal load on the output devices by roughly 40%. Over a continuous complex music signal (which has a high crest factor), the average power dissipation drops even more dramatically, often pushing total system efficiency from the ~50% typical of Class AB up to 80% or higher.
Where You Meet Class H in Practice
You will rarely find Class H topology in budget consumer receivers. Instead, it dominates specific high-performance niches where analog fidelity and thermal efficiency must coexist:
- Touring Pro Audio: Legacy and current high-end touring amplifiers (like specific models in the QSC PLX and Crown XTi lineages) utilize Class H or Class G/H hybrids to keep rack weight down while maintaining the ultra-low noise floor required for massive line arrays.
- Active Studio Monitors: High-end nearfield monitors often use Class H for their woofer amplifiers. The analog output stage avoids the high-frequency EMI that Class D amps can inject into sensitive studio environments, while the tracking supply keeps the cabinet cool during extended mixing sessions.
- Smartphone and Portable DACs: If you are using a high-end smartphone or portable audio player, the headphone amplifier driving your IEMs is likely a Class G/H hybrid chip (such as those from Cirrus Logic or Texas Instruments). In battery-powered devices, the 'tracking' is done via internal charge pumps stepping between 1.8V and 3.3V rails to extend battery life during dynamic music playback without clipping on bass transients.
Topology Comparison Matrix
When selecting an amplifier IC or designing a power stage, it is critical to understand how Class H stacks up against its closest relatives. Below is a benchmark comparison based on typical 2026 silicon implementations.
| Criteria | Class AB | Class G | Class H | Class D |
|---|---|---|---|---|
| Typical Efficiency | 40% - 55% | 60% - 75% | 75% - 85% | 85% - 95% |
| Output Stage Type | Analog (Linear) | Analog (Linear) | Analog (Linear) | Switching (PWM) |
| Power Supply | Fixed Rails | Discrete Stepped Rails | Continuously Tracking Rails | Fixed or Variable DC |
| THD+N (Total Harmonic Distortion) | Ultra-Low (<0.001%) | Very Low (switching glitches possible) | Very Low (requires fast modulator) | Low (requires heavy LC filtering) |
| Heatsink Requirements | Massive | Moderate | Small to Moderate | Minimal / None |
Source context: Efficiency and THD+N figures reflect typical modern implementations detailed in Texas Instruments Audio Amplifier design resources and Analog Devices audio topology guides.
Frequently Asked Questions
Is a Class H amplifier better than Class D for studio monitoring?
For critical nearfield studio monitoring, Class H often holds an edge over Class D due to its inherently lower electromagnetic interference (EMI) and lack of required output LC filters, which can interact unpredictably with the complex impedance curve of a studio monitor's voice coil. While modern Class D amps (using post-filter feedback) have vastly improved their THD+N specs, the purely analog signal path of a Class H output stage guarantees zero high-frequency switching noise, making it the preferred choice for ultra-low-noise environments where the slight efficiency penalty over Class D is acceptable.
What is the exact difference between Class G and Class H topologies?
The distinction lies in how the power supply rails are managed. A Class G amplifier uses multiple fixed voltage rails (e.g., ±15V and ±40V). When the signal exceeds 15V, the output stage physically switches to the 40V rail. This creates a highly efficient system but can introduce 'switching glitches' or notch distortion at the exact crossover point between rails. A Class H amplifier uses a modulator to continuously vary the rail voltage (or uses so many high-speed steps that it mimics a continuous curve), keeping the overhead voltage constant and eliminating the hard rail-switching transients characteristic of Class G.
Why do some Class H amplifiers still use heavy toroidal transformers instead of SMPS?
While the tracking rail modulator in a Class H amp is a switching circuit, the primary power supply feeding it does not have to be. Many high-end audiophile and studio Class H amplifiers still use massive toroidal linear power supplies with large capacitor banks to provide instantaneous, noise-free current reserves. The Class H tracking modulator then acts as a highly efficient 'pre-regulator' between those massive fixed DC rails and the output transistors. This hybrid approach gives you the ultra-low noise floor of a linear power supply combined with the thermal efficiency of a tracking output stage.
Can I retrofit a standard Class AB amplifier to Class H?
Retrofitting an existing Class AB board to true Class H is generally not practical for a hobbyist. It requires inserting a high-speed, high-voltage tracking modulator between the main power supply and the output stage, along with specialized high-speed diodes and compensation networks to ensure the tracking supply reacts faster than the audio signal's maximum slew rate. If the tracking supply lags even slightly behind a fast 20kHz transient, the output transistors will clip against the collapsing rail, resulting in catastrophic 'rail-sticking' distortion. It is almost always better to design the modulator into the PCB layout from the start to minimize parasitic inductance.
