Log In


Audio Amplifier

TFA9890: 9.5V Highly Power-Efficient DSP Embedded Class-D Audio Amplifier


The TFA9890A is a high efficiency Class-D audio amplifier with a sophisticated speaker boost and protection algorithm. It can deliver 7.2 W peak output power to an 8 Ω speaker at a supply voltage of 3.6 V. The internal boost converter raises the supply voltage to 9.5 V, providing ample headroom for major improvements in sound quality.

A safe working environment is provided for the speaker under all operating conditions. The TFA9890A maximizes acoustic output while ensuring diaphragm displacement and voice coil temperature do not exceed their rated limits. This function is based on a speaker box model that operates in all loudspeaker environments (e.g. free air, closed box or vented box). Moreover, advanced signal processing ensures that the quality of the audio signal is never degraded by unwanted clipping or distortion in the amplifier or speaker.

Unlike competing solutions, the adaptive sound maximizer algorithm uses feedback to accurately calculate both the temperature and the excursion, allowing the TFA9890A to adapt to changes in the acoustic environment.

Internal intelligent DC-to-DC conversion boosts the supply rail to provide additional headroom and power output. The supply voltage is only raised when necessary. This maximizes the output power of the Class-D audio amplifier while limiting quiescent power consumption.

The TFA9890A also incorporates advanced battery protection. By limiting the supply current when the battery voltage is low, it prevents the audio system from drawing excessive current from the battery, which could create a problem in the system. The advanced processor minimizes the impact of a falling battery voltage on the audio quality by preventing distortion as the battery discharges.

The device features low RF susceptibility because it has a digital input interface that is insensitive to clock jitter. The second order closed loop architecture used in a Class-D audio amplifier provides excellent audio performance and high supply voltage ripple rejection. The audio input interface is I²S and the control settings are communicated through an I²C-bus interface.

The device also provides the speaker with robust protection against ESD damage. In a typical application, no additional components are needed to withstand a 15 kV discharge on the speaker.

The TFA9890A is available in a 49-bump WLCSP (Wafer Level Chip-Size Package) with a 400 μm pitch.


  • Sophisticated speaker-boost and protection algorithm that maximizes speaker performance while protecting the speaker:

    1. Fully embedded software, no additional license fee or porting required
    2. Total integrated solution that includes DSP, amplifier, DC-to-DC, sensing and more (including LED)
  • Adaptive excursion control - guarantees that the speaker membrane excursion never exceeds its rated limit

  • Real-time temperature protection - direct measurement ensures that voice coil temperature never exceeds its rated limit

  • Environmentally aware - automatically adapts speaker parameters to acoustic and thermal changes including compensation for speaker-box leakage

  • Speaker error detection (detects faulty speakers and enclosures)

  • Output power: 3.6 W (RMS) into 8 Ω at 3.6 V supply voltage (THD = 1%)

  • Clip avoidance - DSP algorithm prevents clipping even with sagging supply voltage

  • Bandwidth extension option to increase low frequency response

  • Compatible with standard Acoustic Echo Cancellers (AECs)

  • High efficiency and low power dissipation

  • Wide supply voltage range (fully operational from 2.7 V to 5.5 V)

  • Two I²S inputs to support two audio sources

  • I²C-bus control interface (400 kHz)

  • Dedicated speech mode with speech activity detector

  • Speaker current and voltage monitoring (through the I²S-bus) for Acoustic Echo Cancellation (AEC) at the host

  • Fully short-circuit proof across the load and to the supply lines

  • Sample frequencies from 8 kHz to 48 kHz supported

  • 3 bit clock/word select ratios supported (32x, 48x, 64x)

  • Option to route I²S input direct to I²S output to allow a second I²S output slave device to be used in combination with TFA9890A

  • TDM interface supported (with limited functionality)

  • Volume control

  • Low RF susceptibility

  • Input clock jitter insensitive interface

  • Thermally protected

  • 15 kV system-level ESD protection without external components

  • 'Pop noise' free at all mode transitions

System Block Diagram

Application Block Diagram


  • TFA9890A SDS

Scan to follow

Open WeChat, click "Discover" at the bottom, and use "Scan" to follow.