A new chip technology could usher in major change for the audio industry, says Roland Hemming.
For as long as anyone can remember loudspeakers haven’t changed. They have been packaged into many shapes and sizes, coupled together and had endless degrees of engineering refinement.
Despite some use of titanium and plastics, the loudspeaker is still essentially a piece of paper that flaps.
Ongoing effort goes into improving audio systems, making audio quality as good as possible. Yet when it gets to the last link in the chain we still rely on a moving coil of copper wire slung around a magnet. The high quality signal processing and distribution we have these days has minimal distortion and perfect frequency response but work on these niceties is then lost by the basic mechanics inside a loudspeaker.
It’s like making pasta sauce; carefully selecting the freshest plumpest tomatoes and fine herbs, cooking them gently and then just before serving, pouring in 10 cloves of garlic and half a bottle of chilli sauce.
A loudspeaker performs a simple task – it moves air. People have looked at many other ways of achieving this. Over the years I’ve listened to motorised drivers, crystals that expand when an electric current is passed over them, electrostatic loudspeakers and even technology that uses ultrasonic waves to direct the audio.
All of these solutions are either more expensive or less effective, or only work well for very specific applications. However, the next decade is likely to introduce a new era in loudspeaker technology.
Origin of MEMS
It all started nearly 30 years ago. In the late 1980s digital mirror projection was invented. Just 10 years after that it was a shipping project. This form of projection relies on a micro-electro-mechanical (MEMS) device. Essentially it’s a chip with a bunch of moving mirrors on it. You shine a light at it and each mirror reflects or deflects the light towards the screen. With each mirror representing one pixel and moving many times a second you can project moving video.
Now another iteration of this technology is becoming available, using the same MEMS technology to create a loudspeaker.
This technology isn’t even new for audio. Most laptops, smartphones and tablets use MEMS microphones. They use a chip which is covered in mechanical sensors to pick up sound. However reversing this process to reproduce sound rather than pick it up is a different deal but the technology is just being realised.
The result is a chip populated with hundreds or thousands of tiny transducers. Each of these move back and forth, compressing air to create sound waves. These chips can be reproduced in massive quantities so the price can plummet. Looking at how the cost of digital mirror projectors came down in price is only part of the story. Whilst the projector business sells in large volumes, the number of loudspeakers sold each year is hundreds of times as many. The volume manufacturing of these loudspeaker chips will be off the charts.
If you consider two for each pair of headphones, a few for each laptop, more for televisions and cars, it’s possible that these could be the amongst the most mass produced chips ever.
Consumer audio will be revolutionised with truly personalised hearing for headphones and the ability to create individual beams of audio for each listener watching TV or listening in the car.
Large panels for precision
On the professional audio side, these transducer chips could be assembled together to create large loudspeaker panels. The result will give audio professionals an unprecedented level of precision. Each individual transducer can be individually controlled. Current line array technology lets you play with the acoustic relationship between tens of transducers. The MEMS solution multiplies that by tens of thousands. You will be able to steer beams of audio with absolute precision and change the properties of the panel in realtime as content is being played.
And it will be just that – a panel that can be thin and light, able to be put in locations where you can’t currently put a loudspeaker. I spend my life being asked to hide loudspeakers. With this technology, near invisible audio becomes far more achievable.
The limitation of other loudspeaker panel technologies has been their ability to reproduce lower frequencies, but the sheer number of MEMS devices in one of these arrays will enable them to work together to combine waveforms to provide high sound pressure levels across a wide frequency range.
Small versions of these loudspeakers will be connected using PoE so a single cable can provide power, digital audio and control data. The amount of power required is a fraction of a conventional loudspeaker. The quality will be far superior when each movement is under fine digital control and does not have the same mechanical restrictions of a moving coil and paper cone.
This technology can be scaled from headphones through to large loudspeaker arrays.
Since it is a digital system you can accurately monitor the loudspeaker. Combining the array with MEMS microphones lets you monitor the performance remotely and much more accurately. This will transform some audio systems. Concert engineers can ensure that their system is performing optimally. Emergency audio systems can confirm that each individual element is working, assuring you that it will work in an emergency.
As with so many other technologies before them, the future of loudspeakers is probably solid state. This is the future of reproduced sound.