A THR amplifier (specifically referencing the popular Yamaha THR desktop series) is a compact audio system that relies on Class D pulse-width modulation (PWM) amplification and a switch-mode power supply (SMPS) to deliver high-fidelity sound from a low-voltage DC source. It changes the physical footprint and thermal management of audio gain, replacing massive linear transformers and heavy aluminum heat sinks with high-frequency switching MOSFETs and compact surface-mount inductors. People commonly confuse Class D amplification with digital audio processing, assuming the "D" stands for digital, when it simply denotes the topology's alphabetical sequence after Class C.

The Core Architecture: How a Class D THR Amplifier Works

When you open the chassis of a modern THR amplifier, you will not find the massive toroidal transformers or the giant TO-3 output transistors typical of vintage Class AB designs. Instead, the heavy lifting is done by a Class D amplifier IC (often a Texas Instruments TPA3116D2 or similar) and an external SMPS brick.

The fundamental principle of Class D amplification is switching, not linear dissipation. The analog audio input signal is fed into a comparator alongside a high-frequency triangle wave carrier (typically between 300 kHz and 500 kHz). The comparator outputs a PWM square wave where the duty cycle is directly proportional to the instantaneous voltage of the audio input. This high-frequency square wave drives an H-bridge of MOSFETs, switching the load between the positive and negative supply rails.

Bench Safety Warning: When probing the output stage of a Class D amplifier with an oscilloscope, the high dv/dt (rate of voltage change) of the switching nodes can induce massive ground loops if you use standard long ground-lead probes. Always use a short ground spring attachment or an isolated differential probe to avoid blowing the MOSFETs or your scope's input stage.

Because the MOSFETs are either fully on (low resistance, low heat) or fully off (high resistance, zero current), the theoretical efficiency approaches 100%. In practice, a well-designed Class D stage operates at 85% to 95% efficiency, which is why a 30W THR amplifier can run entirely off a small external laptop-style power brick without melting its enclosure.

Worked Numeric Example: Sizing the LC Output Filter

The raw output of the H-bridge is a violent, high-voltage PWM square wave containing both the desired audio frequencies (20 Hz to 20 kHz) and the high-frequency switching carrier (e.g., 400 kHz). If you fed this directly to a speaker, the voice coil's inductance would naturally filter some of it, but the tweeter would likely burn out from the high-frequency RMS energy. We must strip the carrier using a passive LC low-pass filter.

Let us calculate the cutoff frequency for a typical output filter found in these desktop amps. The standard formula for the -3dB cutoff frequency of an LC filter is:

fc = 1 / (2π√(LC))

Assume the design uses a 10 μH shielded power inductor and a 1 μF film capacitor.

  1. Convert to base units: L = 10 × 10-6 H, C = 1 × 10-6 F.
  2. Multiply L and C: (10 × 10-6) × (1 × 10-6) = 10 × 10-12.
  3. Take the square root: √(10 × 10-12) ≈ 3.162 × 10-6.
  4. Multiply by 2π: 2 × 3.14159 × 3.162 × 10-6 ≈ 19.87 × 10-6.
  5. Invert for fc: 1 / (19.87 × 10-6) ≈ 50,327 Hz.

The cutoff frequency is approximately 50.3 kHz. This is a perfect sweet spot: it passes the entire 20 kHz human hearing spectrum with minimal phase shift, while aggressively attenuating the 400 kHz PWM carrier frequency (which sits nearly three decades above the cutoff, resulting in massive dB reduction). For a deep dive into filter design topologies, the Texas Instruments Class D amplifier resources provide excellent application notes on LC component selection and parasitic resistance.

Where You Meet This Topology in Practice

While we are analyzing this through the lens of a desktop audio unit, the exact same Class D H-bridge and LC filter topology is ubiquitous across modern electrical engineering. You will encounter this exact architecture in:

  • Variable Frequency Drives (VFDs): Industrial motor controllers use the same PWM-to-H-bridge concept, just scaled up to 480V AC and heavy-duty IGBTs instead of small MOSFETs.
  • Switch-Mode Power Supplies (SMPS): The power brick feeding the amplifier uses a variation of this switching theory to step down mains AC to 19V DC efficiently.
  • Active Noise Cancelling (ANC) Headphones: The tiny drivers in modern earbuds are driven by micro-Class D amplifiers to preserve battery life.
  • Automotive Audio: High-power aftermarket subwoofer amplifiers use interleaved Class D phases to push thousands of watts without draining the alternator.

Real-World Scenario Walkthrough: The SMPS Brownout

Understanding the amplifier is only half the battle; understanding its power delivery is where most DIYers and bench technicians fail. Here is a real-world failure mode involving a THR-style amplifier and a generic replacement power supply.

  1. The Setup: A user loses the OEM 19V power brick for their desktop amplifier and replaces it with a generic, unbranded 19V laptop charger found in a drawer. The amp powers on, the DSP boots, and quiet playing sounds fine.
  2. The Numbers: The OEM brick is rated for 19V at 3.42A (65W). The generic replacement is rated for 19V at 2.5A (47.5W). The amplifier's Class D stage is capable of outputting 2 × 15W RMS (30W total). With 90% efficiency, 30W of audio requires ~33W of DC input. The internal DSP and preamp stages draw an additional 8W continuously. Total peak system draw is roughly 41W.
  3. The Outcome: During a track with heavy low-frequency transients (like a kick drum or bass drop), the audio suddenly clips violently, the digital display flickers, and the amplifier reboots.
  4. What Went Wrong: Audio RMS ratings do not capture transient peak current. A sharp bass transient demands instantaneous current that spikes well above the continuous 41W calculation, briefly demanding 55W from the rail. The generic 47.5W SMPS brick hits its over-current protection (OCP) threshold. Instead of shutting off completely, the brick's voltage folds back to 12V to protect itself. This 12V drop starves the internal 5V buck converter that powers the digital signal processor (DSP). The DSP experiences a brownout, crashes, and triggers a system reboot.

The Fix: Always match or exceed the OEM current rating. If you must use a third-party SMPS, verify its transient response curve on the datasheet, or add a low-ESR bulk capacitor bank (e.g., 4700μF) at the amplifier's DC input jack to supply transient current locally without sagging the SMPS voltage.

Frequently Asked Questions

Why does my Class D amplifier emit a faint high-pitched whine when idle?
This is usually magnetic noise from the LC output filter inductors or the SMPS transformer. If the switching frequency (e.g., 400 kHz) drifts or if sub-harmonic oscillation occurs due to poor PCB layout or cheap shielding, the physical vibration of the inductor windings (magnetostriction) can fall into the upper limits of human hearing. Replacing unshielded drum-core inductors with molded, shielded power inductors usually eliminates this.

Can I bridge two channels of a standard Class D amp IC for more power?
Unlike Class AB amplifiers, you cannot simply bridge the outputs of most modern Class D ICs (like the TPA3116) unless the datasheet explicitly supports PBTL (Parallel Bridge-Tied Load) mode. Standard Class D outputs are already configured as BTL (Bridge-Tied Load) internally. Shorting or bridging them externally will cause a direct short across the H-bridges, instantly destroying the output MOSFETs.

Does the switching frequency of the amplifier interfere with my radio or Wi-Fi?
It can. A 400 kHz switching frequency generates strong harmonic spikes that extend well into the AM radio band and can cause intermodulation distortion in nearby sensitive RF receivers. Proper layout requires keeping the high di/dt traces (from the MOSFETs to the LC filter) as short as possible and using a continuous ground plane to contain the electromagnetic interference (EMI). For more on EMI mitigation in switching circuits, refer to SparkFun's guide on switch-mode power supply layout.