FDA803D and FDA903D STMicroelectronics Digital-Input Audio Amplifiers with Auto Diagnostics

STMicroelectronics has launched two new 1x45W class-D fully digital automotive audio amplifiers with the aim of simplifying design, reducing BOM costs and helping to certify systems to the required ISO 26262 ASIL (Automotive Safety Integrity Level). Designated FDA803D and FDA903D, applications include telematics, emergency-eCall equipment and Acoustic Vehicle Alerting Systems (AVAS, for hybrid/electric vehicles).

The parts offer I2S (Inter-IC Sound, a serial bus interface standard used for connecting digital audio devices together) and TDM (Time-division multiplexing) digital-audio inputs and a feedback-after-filter topology for low noise and high sound quality. Combining these features with full I2C automotive-grade diagnostics including offset and open-load detection in play mode, the new products extend ST’s FDA family of 2-channel and 4-channel devices up to 4x135W.


By delivering high audio quality with low distortion, the FDA803D and FDA903D also can be used for adding extra audio channels to high-end infotainment systems. They can drive 2Ω-load speakers and feature built-in output-power limiting to protect small or low-cost speakers. Efficiency is claimed to be up to 90% (4Ω, 1kHz, 20W), and thermal performance from the PowerSSO36 package with an exposed underside heat spreader allows heatsink-free designs that permit smaller modules and simplify mounting in the vehicle.

The FDA803D and FDA903D have a wide supply-voltage range of 3.3V to 18V, allowing use in a variety of equipment powered from the car battery or other sources. STMicroelectronics reports the parts are tested according to CISPR class V to help ensure compliance with EMC specifications for automotive equipment.

Thanks to the solutions implemented to solve EMI problems, either device is intended to be used in the standard single DIN car-radio box together with the tuner.

ST’s feedback loop architecture includes the output L-C low-pass filter, enhancing frequency response linearity and reducing distortion. Following is some background on ST’s feedback-after-filter concept:

The audio performance of a Class D amplifier is heavily influenced by the characteristics of the output L-C filter. The choice of its components is critical because a lot of constraints have to be fulfilled at the same time: size, cost, filter for EMI suppression and efficiency. In particular, both the inductor and the capacitor exhibit non-linear behavior; the value of the inductance is a function of the instantaneous current in it and similarly the value of the capacitor is a function of the voltage across it.

In the classical approach, where the feedback loop is closed right at the output of the power stage, the L-C filter is placed outside the loop and these nonlinearities cause Total Harmonic Distortion (THD) to increase. The only way to avoid this phenomenon would be to use components which are highly linear, but this means they are also bigger and/or more expensive. When the L-C filter is outside the loop, its frequency response heavily depends on the impedance of the loudspeaker; this is one of the most critical aspects of Class-D amplifiers. In standard class D this can be mitigated, but not solved, by means of additional damping networks, increasing cost, space and power dissipation.

Since the demodulator group is now in the feedback path, some constraints regarding the inductor and capacitor choice are still present but are less stringent than in the case of a typical switching application

FDA803D or FDA903D also can be used with the “classical” configuration of feedback on output (before L-C filter), through I2C configuration, allowing maximum flexibility. The choice depends mainly on EMI target requirements and could affect (slightly, according to ST) other performance factors like damping or THD.

FDA903D is a single bridge class D amplifier intended for any automotive audio application and is distinguishable from FDA803D by means of a load-current monitoring feature that allows continuous monitoring of the speaker, thus enabling advanced diagnostics as well as speaker performance enhancement.