Applications processors with integrated speaker amplifiers often compromise loudspeaker performance. This means that full output power rating of the speaker cannot be reached when connected to the battery supply rail and so limits the maximum output sound-pressure level, especially when raw output power is preferred over audio quality.
The ability to maximise speaker distortion, when maximum volume is preferred, requires careful consideration, especially to prevent speakers being overdriven and damaged. Audible system noise artefacts, such as TDMA noise, can couple into the signal path, resulting in audible hiss when listening to headphones and speakers (6). This can be attributed to several factors, including poor headphone grounding and poor power supply rejection in the amplifiers.
Figure 6 – Sources of power supply noise and power supply rejection of outputs
On top of this, the quest for smaller, slimmer products means PCB space is at a premium. Keeping external components to a minimum not only minimises BOM cost but saves PCB area. Analogue audio interconnects often need passive decoupling networks and employ EMI-reduction techniques, such as ferrite beads. Additionally, maintaining system signal integrity, whilst routing high frequency speaker signals with fast switching currents across the PCB and via flexible connectors, is difficult and can reduce efficiency due to parasitic impedances.
Using a dedicated high-performance class D speaker amplifier will enable optimal acoustic output power improvements, minimise hiss and give users audibly louder speakers. This can be achieved without fear of damaging speakers, by employing a range of automatic gain control and threshold limiting techniques, as found on WM9090 audio subsystem.
Examples of this include having the ability to program a maximum signal headroom level in the amplifier to limit distortion, and a corresponding power limiter level, to set a maximum electrical output power (7). More recently, the advent of advanced low-power dynamic range compression has seen quiet signals artificially boosted to give a higher overall power level (noticeable as a louder output level).
A dedicated class D speaker amp is usually found in a small package, enabling it to be located very close to the loudspeaker, effectively minimising the distance over which the high-frequency speaker signals need to travel, minimising the EMI risk. The next generation of speaker amplifiers from Wolfson will further address this problem by moving to a noise-immune digitally-interconnected speaker amp, which can be located close to the speaker module without any external passive components.
Figure 7 – Power supply limiting and programmable headroom