Audio hubs extend battery life in portable entertainment devices

Adding more features to a handset usually increases the drain on the battery, whether as a result of power-hungry video and imaging functions such as large color displays or multi-megapixel cameras, or using additional signal processing to perform more advanced tasks such as speech recognition or MPEG decoding.

In the audio domain, supporting movie playback, mobile TV, gaming and other multimedia features also draws more battery current. While older handsets only required a mono loudspeaker for playing a short ring tone, recent multimedia phone designs incorporate stereo speakers which are active for longer periods, e.g. during TV streaming or gaming.

These application features also require higher audio quality at the speaker outputs. However, stereo speakers require twice the power of a mono speaker, and a movie clip which lasts 10 minutes will drain 120 times more energy from the battery than a mono ring tone which lasts only 10 seconds. Higher levels of volume are also expected nowadays, with 1-W speaker output power a typical requirement, placing further demands on the battery.

Adding features to a handset usually goes hand-in-hand with the addition of circuitry, and in the ever-shrinking handset that means less space than before for the battery. Adding yet more power-hungry audio features to a handset while embedding the smallest possible battery forces designers to look closely at every potential inefficiency in the handset, so that precious battery power can be saved wherever possible. This need for longer battery life is driving the trend towards Class D amplifier technology, which can remove the largest source of inefficiency in the audio circuitry.

Reducing form factors are also driving integration of mixed-signal audio functions, but integration also presents new challenges of improving levels of audio quality without further increasing power consumption or requiring additional external components such as regulators or passive components. This complex design puzzle is increasingly being solved by a new generation of audio hub devices. Three major considerations are driving innovation of these devices:

1. Improving audio quality. This isn't simply limited to improving SNR and THD performance of key components. It also involves rejecting noise created by other components; eliminating pops, clicks, zipper noise, and other transients; maintaining high performance at higher volume levels, all of which contribute to a better audio experience from the perspective of the end user.
2. Minimizing power consumption (in active and standby modes).
3. Reducing pcb footprint and component count.

Audio hub concept
Portable multimedia devices such as mobile phones typically contain a number of analog and digital audio sources in diverse data formats and are required to convert and mix combinations of these audio streams before outputting to the real world via various transducers (ear speakers, loudspeakers, headphones, headsets). To save space, cut cost, and reduce design complexity, it's advantageous to group these audio processing functions into a single device, the audio hub.

The audio hub must be able to interface with analog signals of varying magnitudes, source impedances, dc offsets, and bandwidths, such as FM receivers, microphones, send/receive voice data, ring tones or hi-fi line input. Flexible input configurations can support these diverse signal characteristics in different system architectures while minimizing pin count, saving space, and reducing costs.

Digital data sources can also exist in different formats, word lengths, and sample rates. While telephony usage modes normally only require the audio hub to handle mono, 8-kHz data in PCM format, the integration of digital music playback features requires audio-hub devices to handle different sample rates, word lengths, and data formats (e.g. stereo, 16-bit, 44.1-kHz I2S data). A flexible digital audio interface and clocking scheme on the audio hub together with hi-fi-quality data converters enable digital music playback on the handset with no additional mixed signal components.

Mixing in the analog domain in the audio hub can eliminate sample-rate conversion difficulties, and flexible mixing paths can enable new application features that can combine microphone input, digital music, FM receiver, and received voice data (Fig. 1). These devices provide the ability to re-digitize this mix, enabling features such as karaoke recording.

Diverging wafer-process trends between digital multimedia processors and mixed-signal devices further strengthen the drive to integrate mixed-signal audio functions in one audio hub.

Power-supply requirements for the audio functions in this signal chain are most diverse in the audio hub device, where three or four separate supply domains are typical, each with its own voltage, current capability, and noise characteristics. Audio hubs must be carefully designed to operate within the various limitations of these supplies (Fig. 2). Minimizing power consumption without degrading the audio signals is the key to providing portable devices with hi-fi quality sound without unreasonably reducing battery life. Different power-saving techniques must be used for each power-supply domains.

Save power with digital supplies
Because audio quality is unaffected by reducing digital supply voltages, the digital core will be powered by the lowest possible voltage to save power. At these low voltages, the use of dc/dc converters can bring significant efficiency savings compared with linear regulators, while the supply ripple caused by high-frequency switching of the converter can be tolerated by the digital circuitry more easily than an analog block which would require a stable supply voltage to keep noise levels as low as possible.

In a similar way, the digital I/O buffer supply will consume less power at low voltages and audio quality won't be affected, although for practical reasons this supply is sometimes higher than the digital core supply (e.g., for maintaining compatible signaling levels between devices that communicate with each other).

Page 2: next page
Page 1 | 2