THE ART AND SCIENCE OF CONCERT HALL ACOUSTICS

EXECUTIVE SUMMARY

This study was conducted over a period of three months during a summer internship. In addition to researching the field for basic background information about concert hall acoustics (Beranek)(Shulman)(Talaske, Wetherill, and Cavanaugh), I learned how to use the ODEON package, a state of the art computational acoustics package from the Department of Acoustical Technology, Technical University of Denmark (Lynge). With ODEON, I studied three existing concert halls. Studies of these halls began with a baseline study of each hall in its basic current configuration with one or multiple performers at different locations to see how sensitive the results were to different types of performances. Several additional studies were done to see how modifications to each hall or to the material properties of reflecting surfaces would impact the sound throughout each hall. Additional studies of several hypothetical halls were also conducted. The final case was an attempt to design an “ideal” hall from the lessons learned from the earlier cases.

A total of 31 hall configurations were modeled. For each case detailed information about the quality of the sound was collected throughout the hall (results displayed on a 0.5meter by 0.5meter grid covering the audience). These displays included loudness, early decay time, reverberation time, lateral fraction (envelopment), definition (clarity), and speech intelligibility. Additional information was collected at several specified locations in the hall (center of main floor, near a side wall of the main floor, center balcony, side balcony, and under center balcony). This information included a sound rose, a circular representation of the direction from which a listener receives sound impulses, and an echogram. It is somewhat of an art-form to convert all this information into a single assessment of acoustical quality. For the extreme cases, both good and bad, the interpretation is rather clear. In most cases there are clearly preferable seats and less desirable seats in the hall.

Here is a brief summary are conclusions from these studies.

Ø Proscenia are not good for hall acoustics. They put the orchestra in a separate room from the audience, which results in uneven sound, poor blend, and a lack of definition.

Ø Reverberation Chambers can be used with much success to alter the Reverberation Time without changing the important measurements of Early Decay Time, Clarity and Definition.

Ø In most cases, the best seats for sound are on the main floor more than half-way to the back of the hall and in the center balcony(ies). The acoustics for seats under a balcony and far from the side walls are surprisingly good.

Ø Carpeting, considered one of the worst materials for acoustics, CAN be used effectively for aisle softening without hindering the sound of the hall considerably.

Ø Halls should be taller if the choice is between height and width/length. Flatter halls decrease the feeling of intimacy.

Ø Halls are not evolving towards a dome shaped “Bubble hall”. In fact, such a hall, while getting ideal lateral reflections, would have problems with echoes.

Ø The current simulation technology is reasonable to learn and use. An experienced user can analyze a hall using ODEON rather thoroughly (covering several cases) in one to two weeks. A quick analysis to provide useful information to a visiting performer/sound engineer can be done in 2 to 3 days.

Ø The current simulation technology captures many features of a hall, but does not handle two important aspects. Sound diffusion from small-scale features in the hall (i.e. decorations) is not handled. Resonance chambers that radiate sound into the hall the way a piano radiates sound from its “sound board” is also not treated. Including either of these processes will require a significant leap in the technology. I can not see how this leap can be made using the ray-tracing methods in use by all the packages we studied. In absence of information about sound diffusion and resonance chambers, the designer is left with educated guesses concerning some design issues (decorations & resonance chambers) that may separate the good halls from the best halls.