This article has permanently moved here from the Natural Frequency Journal, though you can still view a copy of the original at the Internet Archive.
Acoustics has long played a part in the design of buildings. In ancient times this was mainly due to limitations on the audibility of the human voice. Whilst a load shout can be heard up to 100 meters away, most people cannot maintain that vocal level for more than a few minutes. The strain required to produce such a volume usually means that complex sentences are almost completely unintelligible anyway. The Greek amphitheatre was one of the earliest solutions to this problem, a means by which actors could clearly vocalise to a large number of people seated some distance away. This same problem has been faced by many designers since, and their response has drawn heavily on the basic principles first developed in these amphitheatres.
A mixture of historical reference, empirical analysis and trial-and-error is pretty much how acoustic design progressed up until the end of the nineteenth century. In 1895 the lecture theatre of the recently completed Fogg Art Museum at Harvard was found to have intolerable acoustic problems. Wallace C. Sabine, a young physicist at the University, was given the task of remedying the situation. Upon examining the auditorium he determined that the biggest problem was the incomprehensibility of speech due to excessive reflection of sound back and forth between its internal surfaces. He tested this by bringing in large numbers of cushions from the seats of a nearby theatre, and noted an immediate improvement. This is considered by many to be the first significant scientific attempt to determine the fundamental parameters of good room acoustics.
In 1898 Sabine derived a mathematical formula for the time taken for sound to decay in a room based solely on its volume and the amount of absorption its internal materials provided. Taken over 60dB, this is known as the Reverberation Time of an enclosure and is still in popular use today. Whilst there have been many proposed modifications to better accommodate specific conditions, this formula has proved very useful and still forms the basis of most acoustic design work carried out today. However, given the wide range of building types and the activities taking place inside them, the problems faced by modern acoustic designers are often very different from those found in a traditional lecture theatre. In the case of open plan offices, the acoustic landscape is all too often a neglected aspect of a building’s interior.
Open Plan Offices
When it comes to the acoustic design of open-plan offices, what is ideally required is perfect communication out to a specific distance, for example twice that from one side of a desk to the other, and then zero intelligibility beyond that. Obviously people need to be able to communicate clearly, however other people’s conversations can be an annoying source of distraction - especially when they are on the telephone.
Research has shown that it can take around 10-15 seconds for someone to refocus their attention on a task once they have been momentarily distracted. The human voice has the greatest distraction potential of any other source, apart from sirens and high pitch discordant sounds. Moreover, it has information content which the brain finds difficult to ignore and will process, leading to periodic and long term attention diversion. Obviously this is not an issue unless you really have to concentrate on something which is why, in general, executive and managerial groups have been shown to be more disturbed by noise than clerks and secretarial staff.
The amount of space per person is also an important influence on sound levels in offices as there is a tendency for people to raise their voices when the room they are in is more densely occupied, particularly when talking on the telephone. Overcrowding produces tension and therefore a heightened awareness of the surrounding environment. Areas of between 8.5 m² and 14 m² per person are a satisfactory range, with around 10 m² being preferable (Boyce, 1974 and Croome 1977). However, below 12.6 m² there will often be a small number of people who still feel overcrowded.
Dealing with the Ceiling
Open plan offices usually have relatively low ceilings and large floor areas, which means that the surrounding walls play a negligible role in the sound field. As well, internal partitions tend to be low level and relatively lightweight, so they too have limited effect. With hard-wearing carpeted floors being all but mandatory, the only real variable within the office that the designer has left to play with is the ceiling.
Given cost constraints and the need to accommodate services, ceilings in most office buildings provide nothing more than a large expanse of relatively lightweight acoustic tile. Whilst there are many high quality and well designed tiles available, they can only do so much to solve these problems. Moreover, with the increasing use of chilled beams and natural ventilation systems that require exposed thermal mass for night-cooling, there simply may not be enough surface area left in the ceiling to rely solely on good absorption. However, with some careful design, even a highly reflective exposed concrete ceiling can provide excellent acoustics, and a more successful and interesting solution than just highly absorptive tiles.
With a flat ceiling, there really is nothing to stop sound reflecting off into the distance. Whilst partitions, furniture and planting can offer some obstruction at low level, sound traveling outwards and upwards can continue relatively unimpeded, as shown in Figure 1. As a result, the sound field in an office is not the same as in a large open volume in which propagating sound obeys the inverse square law - falling off at around 6dB per doubling of distance. Instead, the open plan space tends to create a tunneling effect, which means there can be areas where the sound reduction is as little as 3dB per doubling of distance.
Some well designed acoustic ceiling tiles make use of incidence angles to reflect useful sounds whilst at the same time absorbing unwanted reflections. Those sounds that go straight upwards are useful as they can be quickly reflected back down quite near to the speaker, which is desirable as this helps increase sound levels for those near by. Those sounds that travel closer to the horizontal strike the ceiling much further away and at a much lower incidence angle, making them unwanted reflections.
If, as shown in Figure 2, the ceiling consists of areas of flat reflective horizontal surfaces, set at different heights, then highly absorptive material can be placed on the vertical surfaces that separate them. This ensures those sounds that would ordinarily travel quite far are instead absorbed by the ceiling, whilst those that travel near to vertical are reflected straight back down. As shown in Figure 3, this can be a good first step in reducing more distant levels.
Taking this same process and scaling it up makes it possible to design a ceiling configuration that promotes acoustic privacy. Figure 4 shows just one example of how this might be done. Whilst there are many other configurations that would work similarly, their main feature is the division of the ceiling area into boxed sections.
In this example the vertical elements that provide the separation are made highly absorptive to reduce unwanted reflections traveling into adjacent sections. Also, the top of each section is made highly reflective and angled slightly to ensure that sound generated within each space remains as much as possible within that section exclusively.
Obviously the ceiling geometry by itself cannot prevent all sound from escaping into adjacent areas. However, when faced with a situation where the ceiling has a number of different roles to play, it is still possible to provide an acceptable acoustic landscape in an open plan office that maintains the privacy, and therefore the productivity, of its occupants.
Boyce, P.R. 1974. Users’ assessments of a landscaped office, Journal of Architectural Research, 3(3), 44-62.
Croome D.J. 1977. Noise, buildings and people, Oxford; New York: Pergamon Press.
Click here to comment on this page.