Sound principles

In the search for perfect pitch, technology is influencing traditional instrument design. Philip Sen explains.

Musical performance isn’t easy to control. Aesthetic quality is influenced by subjective traits such as a musician’s training, talent and emotions. Likewise, physical characteristics such a brass players’ lips or a violinists’ hands can make a big difference to how they play.

Meanwhile, there is often little a musician can do to improve the acoustic qualities of a venue, which can be affected even by the size and movements of the audience. The only stable interface between the unique and unpredictable player and the unique and unpredictable audience is the instrument itself.

Though many believe that instrument design is set in stone along rigid historic guidelines, there are ways that detailed adjustments and a little exact science can greatly improve even the subjective quality of tone.

Craftsmen use their own judgment to make constant adjustments to each instrument and achieve some kind of tonal consistency. But with few people willing to take on decade-long apprenticeships, manufacturers can exploit engineering techniques to introduce consistency and enhance quality in top-of-the-range instruments.


Dr Richard Smith, formerly chief designer at instrument producer Boosey & Hawkes, is one of a growing breed of engineering musicians and craftsmen. His company, Smith Watkins, specialises in trumpets, cornets and flugel horns customised to the player not by trial and error but by technology.

Smith’s work involves the calibration of lead (pronounced as in ‘leader’) pipes, the part of a trumpet between the mouthpiece and the body that can most deeply influence whether a player can get a good or a great sound. In experiments that involved closing off a trumpet mouthpiece with a rubber diaphragm, he proved that the common perception that a trumpet’s sound depends on airflow through the mouthpiece is incorrect.

It is instead the resonance and vibration of the air within the instrument (Smith draws an analogy with AC as opposed to DC current in an electric circuit) that affects the tonal quality. Only 0.2 per cent of the energy put into the trumpet is transformed into sound, so design needs to be enhanced for efficiency. And a narrower, not a wider, lead pipe can sometimes be better for a particular player.

This knowledge is exploited to help tailor instruments for the individual. Most trumpets, even handmade ones, come with standard (but not precision-made) lead pipes that may or may not suit the musician. ‘The assumption is that everyone will like it,’ says Smith. ‘There’s a mystique and mystery about it. I am trying to get away from that and look at variations.’

So he offers his customers a lead pipe selection, allowing them to choose the best pipe for their own style and physique. By measuring the internal diameter of lead pipes to within fractions of an inch he can produce trumpets to varied bore-width standards for different customers.

He is now building upon this with pulse reflectometry techniques to accurately ‘profile’ the inside of the tube even after it is bent into shape. Sound reflections inside the lead pipes are measured to detect the tiny dents and imperfections that can affect the instrument’s sound quality. Unsuitable tubes can be rejected and Smith can engineer the lead pipes to unheard of levels of perfection.


Unlike brass, wooden instruments are less compliant to engineering because of the inherent variations between individual pieces of timber. Woodwind instruments are also unsealed so some of the airflow escapes from the key holes as well as from the bell, which also affects tone. Though no two oboes or clarinets can ever be the same,University of Edinburgh physicists are examining methods of mitigating their naturally unpredictable properties.

Prof Murray Campbell’s team is looking in particular at the design of the side holes. Since small changes in the sharpness of the holes’ edges can affect the sound, Campbell examines airflow around them using a technique called particle-imaging velocimetry. A sheet of laser light is fired in pulses across the plane of the holes as air is blown through the instrument. The results are correlated to ‘map’ air velocity vectors and thus the airflow. Comparison of these maps aids assessment of what produces different sound qualities.


While some UK engineers are seeking ways of exploiting technology to measure and influence the musical qualities of traditional instruments, one US academic and performer takes another view. Tod Machover, Professor of Music and Media at the Massachusetts Institute of Technology, says that rather than trying to optimise the existing qualities of an instrument he seeks to bridge the gap between live and studio performances.

He cites the 1990s-scandal when award-winning band Milli Vanilli was found to have been lip-synching to other people’s voices recorded and mixed in a studio as an example of how technology can remove music from live performance. ‘Technology allows perfection,’ says Machover, ‘but it can be inhibiting. There’s a danger of CDs sounding clinical.’

Machover was inspired to invent ‘hyperinstruments’, based on traditional designs but technologically-enhanced to introduce new elements to live performance. These aim to embrace and exploit technology while preserving spontaneity.

Working with virtuoso cellist Yo-Yo Ma, rather than adapting the musician’s precious Stradivarius, Machover built a cello integrated with technology.

Reworking the basic principles of an electric guitar, piezo-electric sensors inside the cello body transduced the air vibrations, sending them to a computer to be analysed, amplified and processed. Metal strips were also positioned on the fingerboard to detect where and how hard the artist pressed down on the strings, and microchips were inserted on either end of the bow. These created a signal detected by an antenna on the cello to measure the bow’s movement, and another piezo-electric strip on the bow was used to assess the amount of pressure the artist applied.

A violin Machover has worked on more recently has similar devices, plus strain gauges and accelerometers for more comprehensive and accurate measurement. The purpose was to exploit Ma’s actual movements as he played to make the sound multi-layered. As the computer detected certain movements or changes of pressure and force, not only could it amplify the cello but add to it with a virtual orchestra, interpreting the nuances of his movements to add extra sounds or amplify certain moments for dramatic effect. ‘I know that when I get a strong F sharp, for example,’ says Machover, ‘I can trigger an explosion of notes. It turns the cello into multiple instruments.’

While working with Ma, Machover realised that minute natural electric currents within the body were affecting the hypercello’s readings. He exploited this phenomenon in the shape of a ‘media medium’, or sensor-covered chair with flexible glass rods on either side that could detect levels of electrical charge in the air. A member of the audience could sit on the chair and electric distortions caused by movements of the hands would ‘characterise gestures into sound’ via the sensor apparatus.

‘Technology makes it very easy to surround oneself with stimulus,’ says Machover, ‘but it doesn’t encourage concentration and focus. We’re losing the ability to listen.’ Describing the sensor chair as ‘a virtuoso instrument for non-professionals’, Machover was inspired to design more new technology-filled instruments that departed from the traditional to allow the artist or untrained listener take part in the ‘experience’.

His Brain Opera project – in which the public could experiment with hyperinstruments and go on to listen to computer-generated music reconfigured from their activities – is now a permanent display in Vienna. The next idea was Toy Symphony, aimed at giving children access to music via toys such as Beatbugs, which are drum-machine-like pressure sensors that respond to how and where they are hit.

These are networked via infrared signals so participants can send each other rhythms and build multi-layered musical sequences upon them. A central computer modifies and complexifies the sounds for more melodic effects. Also part of Toy Symphony are Shapers, grapefruit-sized objects with built-in sensors designed to detect electrical charge in the fingers as they are manipulated. This is used to guide the generation of computer-synthesised notes.

The music’s origins are in the modern classical vein, but with added electronic trills and melodies layered above and below the main theme.

Machover’s ideas are undoubtedly not to every musician’s taste. He is, however, rethinking how technology can contribute to musical composition and performance aside from the more familiar mixing studio and synthesiser. His next idea is a ‘robot orchestra’. Working with MIT robotics and artificial intelligence expert Dr Cynthia Breazeal, he wants to make ‘a jungle of physical objects’ that can interact with people and each other to produce new forms of music. Perhaps one day we can all be musical prodigies.

Excerpts from the Toy Symphony can be heard on