Volumetric diffusers: a new paradigm for acoustic treatment

Lead Research Organisation: University of Salford
Department Name: Unlisted

Abstract

The experience you have when you listen to music in a concert hall, or speech in a classroom, depends on the design of the room. Different rooms have different acoustics: in some concert halls music sounds good, in others very poor; in some classrooms the teacher can be heard easily, in others it is a struggle to hear and understand what the teacher is saying. So, one of the major issues in designing a room is to get the acoustics right.One important property that needs to be controlled is the reflections from the walls. Overly strong reflections can distort the sound you hear, changing the tonal quality of instruments, or creating echoes which makes speech garbled. One solution would be to make the walls strongly absorbing, which would remove the reflections, but this often unacceptable because it creates a completely 'dead' sound. If you have heard music outdoors, you may have noticed how poor the sound is; so we want to hear some reverberation, but without the distortion.The way this is solved is to make the reflections from the walls diffuse - sound waves coming in from one direction are reflected back over a broad range of angles, rather than a strong reflection in one direction. This is usually achieved by fixing structures on the walls, such as wiggly surfaces, which break up the reflection. These structures are known as diffusers, and the acoustic engineers on this project have been at the forefront of acoustic diffuser design for sometime, with their designs being used in studios and auditoria worldwide.Now we are looking to diffuse sound in a different way, by hanging scattering objects inside rooms to diffuse the sound as it propagates around the space. We are trying to see if we can make more effective diffusers this way. When diffusers are hung on walls, then sound only strikes the diffuser from one direction; put the diffuser in the middle of the room, and then it interacts with sound from all directions.The physicists on the project work in photonics, the science of generating and controlling light. When light waves are shone onto crystals, the light interacts with the crystals in a similar way to how sound waves interacts with the diffusers; for this reason the sound structures are also called sonic crystals. We hope that by bringing together people working on light and sound, new ideas and concepts can be developed through the exchange of ideas and concepts across the disciplines. One of the differences between light and sound is that acoustic diffusers must operate over a much wider range of frequencies than is normally considered in photonics. The acoustic wavelength of musical notes ranges from a few metres to a few centimetres, and to diffuse sound over this range requires objects that include small and large parts. For this reason, we have propose to use fractals, objects that have elements on different scales. An everyday example of fractals are ferns; take a frond from a fern and you will see it is a miniature replica of the whole. In similar ways we can have acoustic scatterers which work at different scales, and so scatter sounds of different wavelengths. We have also proposed some other approaches to the design. For example, it is possible to exploit special binary number sequences, like the ones used to encode digital audio signals, to maximise the diffusing properties of the diffusers.The work involves several workpackages. To enable us to design the diffusers and understand what is going on, we need to develop computer-based prediction models which tell us how sound scatters from the diffusers. We propose to use simple and complex models; simple models often give more incite into what is going on, whereas more complex models should offer more accurate predictions. We will carry out measurements using scale models in the anechoic chamber. We will also examine how the diffusers might be applied to real rooms; to demonstrate to practitioners where and when they would be useful.

Publications

10 25 50
publication icon
Pogson M (2010) Multiple Scattering Between Cylinders and a Schroeder Diffuser in Acta Acustica united with Acustica

publication icon
Hughes RJ (2010) Volumetric diffusers: pseudorandom cylinder arrays on a periodic lattice. in The Journal of the Acoustical Society of America

publication icon
Pogson MA (2010) Diffusive benefits of cylinders in front of a Schroeder diffuser. in The Journal of the Acoustical Society of America

publication icon
R Hughes (2008) Volumetric Diffusers

 
Description The experience you have when you listen to music in a concert hall, or speech in a classroom, depends on the design of the room. Different rooms have different acoustics: in some concert halls music sounds good, in others very poor; in some classrooms the teacher can be heard easily, in others it is a struggle to hear and understand what the teacher is saying. So, one of the major issues in designing a room is to get the acoustics right.

One important property that needs to be controlled is the reflections from the walls. Overly strong reflections can distort the sound you hear, changing the tonal quality of instruments, or creating echoes that makes speech garbled. One solution would be to make the walls strongly absorbing, which would remove the reflections, but this often unacceptable because it creates a completely 'dead' sound. If you have heard music outdoors, you may have noticed how poor the sound is; so we want to hear some reverberation, but without the distortion.

The way this is solved is to make the reflections from the walls diffuse - sound waves coming in from one direction are reflected back over a broad range of angles, rather than a strong reflection in one direction. This is usually achieved by fixing structures on the walls, such as wiggly surfaces, which break up the reflection. These structures are known as diffusers, and the acoustic engineers on this project have been at the forefront of acoustic diffuser design for sometime, with their designs being used in studios and auditoria worldwide.

We have been looking at diffusing sound in a different way, by hanging scattering objects inside rooms to diffuse the sound as it propagates around the space. We were trying to see if we could make more effective diffusers this way. When diffusers are hung on walls, then sound only strikes the diffuser from one direction; put the diffuser in the middle of the room, and then it can interact with sound from all directions.

The collaborating physicists in Sheffield, who were funded by the related grant EP/D031621/1 work in photonics, the science of generating and controlling light. When light waves are shone onto crystals, the light interacts with the crystals in a similar way to how sound waves interacts with the diffusers; for this reason the sound structures are also called sonic crystals. We found that by bringing together people working on light and sound, new ideas and concepts were developed through the exchange of ideas and concepts across the disciplines.

Salford focussed its research on developing practical structures and investigating their properties via accurate numerical simulation and measurement. We initially extended the theory developed for diffuser structures that use both absorption and reflection by simulating single layer structures using slats of material. This allowed us to both examine their performance and also develop figures of merit for their performance. We found that, although simple, these structures did not perform well until their density was quite high, and that their ability to redirect sound was not optimum.

From the results of this pilot study we developed other types diffuser. One was inspired by percolation fractals and the other was based on cylinders, which could be either uniform or of varying diameter. Both approaches showed promise when simulated. However, when scale models were measured in the anechoic chamber the cylinder based structures performed better.

We also found it was possible to exploit special binary number sequences, like the ones used to encode digital audio signals, to maximise the diffusing properties of the diffusers. In particular, we could design cylinder-based diffusers with good performance based on sequences that are used for frequency hopping spread spectrum systems. This is a useful outcome, as it allows us to make cylinder based volume diffusers with good performance based on explicit design rules rather than by chance.
Exploitation Route We have published design methodologies and both simulated and measured results that would allow a manufacturer, learned in the art, to construct volumetric diffusers that could be used to achieve a given specification.
Sectors Aerospace, Defence and Marine,Construction,Environment,Leisure Activities, including Sports, Recreation and Tourism,Culture, Heritage, Museums and Collections

 
Description To date our findings have not been used
 
Description University of Sheffield 
Organisation University of Sheffield
Country United Kingdom 
Sector Academic/University 
Start Year 2006