Electronics Weekly Magazine
Loading

Sign-up for newsletters:

Electronics Weekly newsletters - Sign up for Made By Monkeys, Mannerisms, Gadget Master and Daily and Monthly newsletters

Need for efficiency stabilises audio amplifier chip design

Steve Bush
Wednesday 10 March 2010 13:26

Pushed by power efficiency requirements, integrated audio amplifier architectures are settling down, with Class-D at the top end and Class-AB at the bottom, leaving a little gap in the middle where Class-G is flourishing.

At the top is Texas Instruments 2x300W stereo Class-D TAS5630.

"It features the highest stereo output power in the industry, more than twice that of the closest comparable devices," claims TI.

Aimed at home entertainment - AV receivers, and Blu-ray players, it is said to be at least 88% efficient.

It is this high efficiency that makes Class-D popular, reducing heat dissipation compared with traditional Class-AB audio amps, allowing smaller heatsinks and power supplies to be employed, fitting more audio channels into a given size.

Like several other chip makers, TI is getting hi-fi performance from Class-D.

In this case, bandwidth is 73kHz, and total harmonic distortion (THD) is 0.03% at 1W into 4[ohms] load; staying pretty much below 0.05% from 100mW to almost 100W into 4 or 8[ohms].

A-weighted signal to noise ratio (SNR) is >100dB, and power supply rejection ratio (PSSR) is -80dB with the outputs bridged for 600W mono.

Phone battery life worries have motivated designers to push Class-D operation down to the smallest speaker amplifiers.

Power chip maker AnalogicTech has dipped its toe into audio with the AAT5101, a 2.5W Class-D amplifier for Li-ion cells.

Efficiency is up to 92%, quiescent current is 1.35mA - dropping to 60nA in stand-by.

It delivers 0.12% THD+N at 1.0W from 5V to an 8Ω, and can be cranked up to 1.35W into 8Ω with THD+N <1%, and 2.5W with a 4Ω load and <10% THD+N.

Typically PSRR is -60dB and SNR is 91dB.

Class-AB operation survives at the very bottom of the power spectrum - in audio op-amps.

TI is active here as well, releasing the super-low distortion (0.00005% at 1kHz) JFET input OPA1641, with 5.1nV/√Hz noise at 1kHz.

Quiescent current is 1.8mA per channel, the supply can be anywhere from +/-2.25 to +/-18V, and there are also dual and quad versions.

The gap between Class-D and Class-AB now lies somewhere between a few mW and a few hundred mW, which includes headphone amplifiers.

Although there are plenty of Class-AB designs here, these linear circuits suffer excessive I[super2]R losses, while fully switching designs like Class-D waste too much energy in transitions.

Here too phone chip firms have innovated, adopting Class-G designs where a Class-AB linear amplifier is supplied from two or more voltage levels, generally made by a switched capacitor power converter.

Wolfson has gone Class-G for a hi-fi quality stereo headphone amplifier for single battery applications.

Its WM9010 delivers 28mW into 32Ω, offering 100dB signal to noise ratio (A-weighted) and includes a charge pump to make a negative voltage rail - so no coupling capacitors.

The pump automatically switches its +ve and -ve outputs between two levels depending on the output amplitude required.

Rated supply voltage is 1.7-2.0V and the devices comes in a 1.84x1.34x0.7mm chip-scale package.

Showing just how low Class-D amplifiers can operate, the 9010 has a big brother called WM9090 which includes a 3.7V single rail 750mW mono Class D speaker amplifier.

 

Comments powered by Disqus

Share the content

Most Viewed

Products

Related Jobs

Resources