Scalable Output Power Approach Saves Time-To-Market In Audio Amplifier Design

Employing a common base circuit, a single class D audio amplifier design platform with scalable output power can generate multiple power levels up to 500 W, enabling designers to unify Class D audio designs for numerous products.

While home theatre systems, AV receivers, musical instruments, car entertainment, and other high-performance portable consumer devices are rapidly taking advantage of class D amplifiers for superior audio performance, their power needs are a lot different. For instance, requirements could vary from 50 W for AV receivers to 500 W for high-quality professional amplifiers used in powered speakers.

Designing individual high-performance class D audio amplifiers for each of these products is time consuming and costly. Furthermore, any delay in marketing the final product could hamper its success. Alternatively, a single design platform that features scalable output audio power can simplify a designer’s job, resulting in faster turnaround of the amplifier with fewer components.

A single platform then can improve the end product’s timeto-market and trim its cost. Consequently, simply changing the output MOSFETs with appropriate voltage ratings enables designers to use the same base circuit design to create multiple power levels up to a total of 500 W for a stereo output, unifying a class D audio amplifier design for a number of products.

To achieve that goal, International Rectifier has developed a reference design platform for a two-channel class D audio power amplifier with scalable output power. Besides allowing a designer to scale the output power stage from 25 W per channel to 250 W per channel, the IRAUDAMP7D audio design platform offers selectable half-bridge (stereo) and full-bridge (bridged) topologies.

For that, it incorporates the integrated high-voltage class D audio driver IC IRS2092, along with its digital audio dual MOSFETs, such as the IRFI4024H-117P, IRFI4019H-117P, IRFI4212H-117P, and IRFI4020H-117P, on a single-layer printed-circuit board (PCB). Additionally, this design platform presents an optimum PCB layout for peripheral circuits using a single-sided board. As a result, it requires a small heatsink for normal operation (one-eighth of continuous rated power). Furthermore, all the required housekeeping power supplies and protection circuits are included.

Although the class D audio driver IRS2092 is crafted for applications between 25 and 250 W per channel in half-bridge mode, it can also be used in a full-bridge topology to boost the total output power to 1000 W for some specific applications. A designer can extend the power to higher than 1000 W by including external gate buffers.

AUDIO DRIVER PLUS MOSFETs

Using G5 HVIC technology, the 200-V class D audio driver IRS2092 integrates features and functions that facilitate the design and construction of a high-power class D amplifier using through-hole components on a single-sided board. This high-voltage audio driver is combined with optimized digital audio power MOSFETs to attain up to 250 W per channel or 500 W total audio power from a single-layer PCB.

Implementing the advanced process technology, the new audio driver chip integrates four essential functions required for high-performance class D audio amplifier design. These include the error amplifier, pulse-width modulation (PWM) comparator, gate driver, and robust protection circuitry (Fig. 1).

The built-in protection circuitry simplifies the complex task of overload protection with self-reset control and under-voltage lockout protection (UVLO). Multiple protection features enabled include over-current protection (OCP), high-side and low-side MOSFET over-voltage protection (OVP), high-side and low-side MOSFET dc protection (DCP), and over-temperature protection (OTP).

In addition, the chip delivers programmable preset dead time for improved total harmonic distortion (THD). The dead time can be selected for optimized performance according to the size of the MOSFET, minimizing dead time while preventing shoot-through. As a result, no external gate-timing adjustment is required. Since the dead time is set based on the voltage applied to the DT pin of the chip, the audio driver circuit makes this selection easy and reliable by employing only two external resistors connected to the DT pin.

To combine output scalability with optimal audio performance, this design taps the benefits of digital audio MOSFETs that offer parameters and die sizes optimized for class D audio amplifier applications. In fact, each digital audio MOSFET comprises two power MOSFET switches connected in half-bridge configuration to achieve low on resistance per silicon area.

Moreover, the gate charge Qg, body diode reverse recovery Qrr, and internal gate resistance RG(int), as well as packaging, are all optimized to enhance class D audio amplifier performance in areas like efficiency, THD, and electromagnetic interference (EMI). Since one size is not the answer for all power levels, the design selects the most appropriate MOSFET for the desired output power.

SCALING THE OUTPUT

Output MOSFETs can be configured in half-bridge and full-bridge topologies. The half-bridge also can be used to drive speakers with both 8-Ω and 4-Ω load impedances. Unlike the half-bridge, the full-bridge is architected to drive only an 8-Ω load. So using IRFI4024H-117P MOSFETs with 55-V BVDSS capability in a half-bridge topology, the amplifier offers 25 W/channel with 8-Ω speakers and 50 W/channel with 4-Ω speakers. Using the IRFI4212H-117P in a half-bridge method delivers 60 W/channel with an 8-Ω load and 120 W/channel with 4-Ω speakers.

The same devices in a full-bridge version with 8-Ω speakers increase the audio amplifier’s output power to 100 W and 240 W, respectively. Simply replacing the medium voltage output switches with MOSFETs handling higher breakdown voltages further amplifies the output power to a higher level. For example, with the ability to withstand 100-V BVDSS, the IRFI4019H-117P furnishes 125 W/channel for 8-Ω speakers and 250 W/channel for 4-Ω speakers in half-bridge mode. In a full-bridge structure, the IRFI4019H-117P boosts the total class D output power to 500 W (Fig. 2). Likewise, the 150-V IRFI4020H-117P provides 250 W/channel for an 8-Ω load, but it does not support 4-Ω speakers as well as the full-bridge topology.

The output of the power stage comprising the high-side and low-side digital audio MOSFETs is amplified PWM waveform. The LC low-pass filter at the output recovers the audio signal from this PWM waveform. It also filters out the class D switching carrier frequency and leaves the audio signal at the speaker load. To damp any LC resonances and prevent peaking frequency response with light loading impedance, though, an RC filter called the Zobel network follows the low-pass filter.

The switching frequency used is 400 kHz (Fig. 3). It drops to around 0.05% as the output power level moves upward to 120 W/channel with a 4-Ω resistive load in a stereo configuration. The signal and switching frequencies remain the same. As you move up the power ladder using ±70-V supplies and replacing medium-voltage parts with high-voltage audio MOSFETs such as the IRFI4020H-117P, the distortion performance remains below 0.05% at 250 W/channel for an 8-Ω load with similar signal and switching frequencies (Fig. 4).

Although these measurements were conducted at 1 kHz, the distortion performance of the class D audio amplifier design is equally good over the entire audio band. THD plus noise (THD+N) versus frequency test indicates that the distortion is low (<0.03%) and remains consistent over the entire audio range of 20 Hz to 20 kHz, even as the output power is increased from 10 W/channel to 50 W/channel with a 4-Ω load impedance (Fig. 5). Similar tests conducted for noise indicate that the noise floor remains below –80 dBv over the entire audio range.

The low on resistance of the digital audio MOSFETs combined with low input capacitance results in lower power (conduction + switching) losses and higher efficiency for the class D amplifier. In addition, the secure dead time provided by the audio driver chip prevents cross conduction to eliminate any further losses and keep the efficiency bar high.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish