News 12-07 (No.295)
Issued : July 25, 2012
[ Japanese Version ]
Isabella Stewart Gardner Museum - Calderwood Performance Hall's Acoustics
By Motoo Komoda
Calderwood Performance Hall plan
The ceiling's skylight with pyramid-shaped glass
Perforated wall panels and waffle-patterned panels
attached to balcony underhang
In our March, 2012 newsletter, I wrote about the opening of the Isabella Stewart Gardner Museum's new wing and mentioned that we were waiting until the museum hosted Calderwood Performance Hall's opening series of concerts to take acoustical measurements in the new hall. In this article, I will discuss key features of the hall's acoustical room design and some particularly memorable highlights of the hall's first concerts.
As can be seen on the accompanying architectural drawings, the basic shape of the new hall is a cube that measures 14 m. on each side, a configuration that posed several difficult challenges from the acoustical design perspective. Because of the acoustical obstacles inherent in a cube-shaped hall configuration, most standard acoustical engineering textbooks categorically advise against adoption of the cube shape when designing a concert hall.
<< Mitigating the Acoustical Challenges of a Cube-shaped Hall >>
Parallel floors and ceilings can cause a hall to have flutter echoes. To prevent this undesirable phenomenon from occurring in Calderwood Performance Hall, we negotiated the use of pyramid shapes in the glass of the hall's skylight, thereby creating surfaces in the ceiling that are not parallel to the hall's floor.
The hall's balcony tiers also have parallel side walls. On most of these wall surfaces we installed 2.5 mm. wood panels with 40 percent open perforations. These panels are acoustically transparent and promote diffusion of high frequency sounds. The three layer gypsum board wall behind the panels has sufficient rigidity to produce excellent sound reflections while the perforated wood panels ensure the prevention of flutter echoes. In addition, each balcony underhang connects to the hall's walls at 90-degree angles. We pursued an iterative design process for this part of the hall and achieved a design that also effectively produces sound reflections from each of these corners.
<< Testing in Scale Models >>
Our room acoustical design process for this project included building a 1/24 scale model and conducting acoustical testing of the hall's configuration in the scale model. In addition, we decided to do additional testing of one element that we wanted to install at the concrete balcony underhangs. We had developed a plan to use waffle-patterned ceiling panels at the concrete balcony underhangs to promote sound diffusion, but this kind of repetitive pattern (a diffraction grating) can sometimes be the source of undesirable acoustic phenomena (as explained in our September, 2005 newsletter). To best understand how the waffle-patterned panels would perform as acoustical elements in Calderwood Performance Hall, we built a 1/10 scale model of a balcony with a waffle-patterned panel attached to the balcony underhang and conducted tests. By testing in the scale model, we checked that these panels would achieve the desired sound diffusion result.
Calderwood Performance Hall's
reverberation time characteristics
As a result of our final acoustical testing, we further confirmed that echoes cannot be detected in the new hall. The reverberation time measures 1.0 seconds (at 500 Hz, in an unoccupied hall), from which we calculate that the reverberation time measures 0.9 seconds when the hall has a full audience. The reverberation characteristics are mostly consistent across the entire frequency spectrum. Based on the results of our measurements, our design achieved the quantitative acoustical physical properties that we set as the objectives for this hall.
<< Concerts in Calderwood Performance Hall >>
As I mentioned in the March article, one of Calderwood Performance Hall's most distinguishing characteristics is the placement of the audience seating. The 300 seats entirely surround the stage, starting on the ground level and continuing in single-row balconies that rise vertically on all sides of the hall. When I sit in this hall, the sense of intimacy is so strong that I feel as if I could touch the performers merely by extending my arm. The single-row arrangement of 80 percent of the seats means that almost all of the audience enjoys unobstructed "VIP" seating. (Only the ground level audience section has two rows of seats.)
However, the equal proximity of the audience from all sides of the hall also means that the hall has a square stage, which creates a totally different stage environment compared with the stages of most other performance venues. There is no directionality from the stage to the seating and, not only does the audience surround the performers for 360 degrees in the horizontal dimension, but much of the audience looks down on all sides of the stage from the balconies. For both solo and ensemble performances, the performers need to think carefully about how they configure their seating on this hall's stage.
Paavali Jumppanen's piano with the lid removed
Circular ensemble seating configuration of
Yo-Yo Ma's cello concerto performance
When Paavali Jumppanen gave his solo performance, he tried removing the lid from his piano. He found that the piano's notes rose toward the ceiling and that the reverberations blended well throughout the hall. Soprano Kiri Te Kanawa deftly and slowly turned to face each side of the audience as she sang, greatly impressing me by her attention to connecting with each and every member of the audience. It was truly a wonderful experience to listen to her beautiful voice at such close proximity.
When Yo-Yo Ma began to rehearse the Haydn Cello Concerto, he placed his cello riser in front of the ensemble and sat with his back to the other musicians. This was not a perfect configuration and he decided to both abandon the riser and to sit together with the ensemble in a circular configuration. This enabled the musicians to have eye contact with each other as well as with Mr. Ma and improved their ease of playing.
I watched and listened to these and other rehearsals and concerts that celebrated the hall's opening. Each musician and group of performers brought a creative approach to performing in the hall's unique environment. I look forward to learning about how other musicians creatively adapt to this intimate and delightful new venue at the Isabella Stewart Gardner Museum.
You can read more about the museum's music program or listen to a podcast at the music page of the museum's website.
Acoustical Design Legacies and Lessons of Older and Bygone Halls - Part 4
-- Reverberation Time Measuring Devices and Meters --
By Dr. Minoru Nagata, Founder of Nagata Acoustics
In this series' previous article I wrote about a special slide rule I used many years ago as a handy tool for calculating room reverberation times. This month, I turn my attention to devices I used in the 1950s and thereafter to measure reverberation time.
<< Using an Electromagnetic Oscillograph to Monitor the Decay of Sound Pressure Waveforms >>
As I wrote in the previous article, when a sound being generated in a room is suddenly interrupted, its reverberation time is defined as the number of seconds required for the sound's intensity to decay to one one-millionth of its original strength. When I joined NHK Research Laboratories in the autumn of 1949, my first assignment involved watching an electromagnetic oscillograph's recording of sound reverberation waveforms on the device's light-sensitive paper. My job was to monitor the successive peak values recorded on the time axis.
The oscillograph of this period predated the general availability of cathode-ray oscilloscopes. Instead, the electromagnetic oscillograph relied on minute vibrations to record data and the following design: A small mirror placed in a magnetic field would vibrate when there was an audio signal. A beam of light from the mirror then reflected onto the roll of light-sensitive paper and when the paper was developed, the recorded waveform could be observed. When I think of the technological progress from the days of that oscillograph to our present world of easy-to-use digital cameras and other gadgets, it's mind boggling to remember the amount of effort routinely expended in the late 1940s to record sound and vibration waveforms using an electromagnetic oscillograph. (I referenced Amino and Arai's "The Electromagnetic Oscillograph" in the Journal of the Acoustical Society of Japan, Vol. 66, No. 8 (2010), page 404 to write this section.)
<< Sabine's Method of Measuring the Duration of Reverberations after Interrupting a Sound Source >>
The above-discussed device focused on the decay of waveforms as the measured phenomenon. At the turn of the 20th century, one acoustics professor was interested in an entirely different perspective, a perspective that focused on how long or short a sound's reverberation lasts in a room. This professor was none other than W.C. Sabine, the father of sound reverberation theory. Sabine used the sound generated from an organ pipe and a stopwatch to measure the duration between the moment the sound was generated and when it became inaudible. He conducted his experiment in Harvard's Fogg Lecture Hall. This may have been the first ever instance of a measuring device being used to determine the amount of reverberating sound in a room. Fortunately, the lecture hall's acoustics created sound reverberations that were audible for several seconds.
The amount of reverberations in a room is a relative value. Sabine also experimented with bringing chair cushions into the lecture hall to measure the decrease of reverberations and found that the reverberation time was inversely proportional to the amount of chair cushions. This discovery became the core of the reverberation theory that he developed.
<< A Measuring Device Based On the Decay Processes of Sound Reverberations >>
With the proliferation of microphones and cathode ray tube (CRT) screens, it became possible to design a device that visually represents oscillating waveforms and sound reverberation times. At NHK Labs, we developed a device based on the decay process of sound reverberations and of electrical circuits. The following paragraphs describe the science we used to develop this CRT oscilloscope (then still called an "oscillograph") around the year 1950.
According to Sabine's reverberation theory, after a sound source is interrupted, the intensity of sounds in the room's sound fields decays exponentially over time, as shown in the below Equation (1):
In Equation (1), " Io" represents the original intensity of the sound, "It" represents the intensity of the sound after "t" seconds, "e" represents the mathematical constant "2.7182…". The value of exponent "α" is the amount the sound's intensity decays in 1 second and is referred to as the decay constant. When the sound's intensity decays by one-tenth per second, then the decay constant "α" equals 0.1.
This decay can be replicated by a simple electrical circuit that includes a resistor "R" and capacitor "C", as shown in Fig. 1. When the circuit's switch "SW" is opened, the voltage "Et" decays in the same way as expressed in Equation (1) and also as expressed in Equation (2):
In Equation (2), "Eo" represents the original voltage and "Et" represents the voltage after "t" seconds. The "RC" value of this equation is a constant known as the "the time constant" because it is expressed in seconds (that is, the dimension of time). 1/RC represents the voltage rate of decay. Because the reverberation time of Equation (1)'s "It/Io" is defined as a decay of 60dB, when expressed by Equation (2) as voltage decay, 60dB becomes 1/1,000(10-3). Therefore, reverberation time "T" (in seconds) can be expressed using the time constant "RC" as shown in Equation (3) below:
In the device configuration shown in Fig.2, reverberation signals picked up from a microphone are connected to the CRT's Y-axis and decaying signals from the RC circuit are connected to the CRT's X-axis. When both signals are interrupted at the same time, if they have the same decay ratio, the CRT's waveform appears as a 45 degree isosceles triangle, as shown in Fig. 3(a). In order to improve the ease of observation using this device, the device can be made to rectify the reverberation signal and its angle can be made to display as a 45 degree angle, as shown in Fig.3(b). By adjusting the "RC" value, and specifically, the "R" value in Equation (3), the reverberation time "RT" can be calculated.
The above measuring device and calculation method comprise the early CRT reverberations meter we developed in Japan and how we used it. Together with two of my colleagues, Mr. Nagatomo and Mr. Sakamoto, I delivered a draft paper on "A Simple Reverberation Measuring Device Using a CRT Oscillograph" in April, 1950, at the Fourth Research Presentation Meeting of the Japan Broadcasting Corporation.
<< Our Reverberation Time Measuring Device Using Storage Tube CRTs >>
My recollection is that a simple reverberation time measuring device that did not have a logarithmic compression circuit could only be used to observe reverberation signals in the range of 20 dB. However, by watching the fluctuations on the device's CRT screen, we could visually experience the decay process of the sound reverberations while they were decaying. We achieved this enhancement of the device through the joint effort of the (then) Television Research Department and Acoustical Research Department in developing a new reverberation meter that incorporated storage tube CRTs.
Unfortunately, neither technical documentation nor photos exist of this advancement in reverberation time measuring devices. Nevertheless, it's clear that this measuring device, which made use of the most advanced television technology of the day, represented a pioneering innovation in the design of acoustical measuring devices. In the half-decade before 1955, this device played a major role in our acoustical measuring activities for broadcasting studios and halls.
The storage tube CRT device for acoustical measuring was a physically heavy piece of equipment and when we added the power supply to the device for off-site work, the very heavy combined weight meant that transporting the device to project sites outside of Tokyo was a major chore only to be attempted by individuals with strong muscles and the willingness to work up a sweat doing manual labor. In addition, because we built the device by hand, transporting it revealed other weaknesses. I remember that we spent half of our first day at one project location retightening the screws that held the device together and fixing other parts damaged during transport.
<< Bruel & Kjær's High Speed Sound Level Recorder and Later Devices >>
We only relied on the above-described, painstakingly developed and transported device for some two or three years until around 1955 when Bruel & Kjær's high speed sound level recorder became available in Japan. Thereafter, for the next 35 years or more, Bruel & Kjær's logarithmic sound level recorders, with their incorporated analog technologies, ruled the market for reverberation time measuring devices.
By the dawn of the 21st century, the work of Manfred R. Schroeder led to use of inverting or integrating squared impulse responses to develop yet more precise technologies for measuring reverberation time. This advance brings us to the current state of reverberation time measuring devices. Today, these devices incorporate "Schroeder Integration" in user-friendly digital devices available in many sizes and price points for the needs of both professionals and non-professionals interested in measuring a room's reverberation time.
Nagata Acoustics Inc.
Hongo Segawa Bldg. 3F, 2-35-10 Hongo
Bunkyo-ku, Tokyo 113-0033, Japan
Tel: +81-3-5800-2671, Fax: +81-3-5800-2672
2130 Sawtelle Blvd., Suite 308
Los Angeles, CA 90025, U.S.A.
Tel: +1-310-231-7878, Fax: +1-310-231-7816
75, avenue Parmentier
75011 Paris, France
Tel: +33 (0)1 40 21 44 25, Fax: +33 (0)1 40 21 24 00
[ Japanese Version ]