Acoustical Design of the Tokyo Opera City (TOC) Concert IBall, Japan

Takayuki Hidakat, Leo L. Beranek*, Sadahiro Masudat, Nor&o Nishiharat and Toshiyuki Okanot

ffahwnaka R&D Lab., l-5-1 Otsuka, Inzai, Chiba 270-13, Japan, *975 Memorial Dr., # 804, Cambridge, MA 02138

Abstract: The TOC concert hail seats 1636, volume 15,300 m3, and reverberationtime, with audience and orchestra, 1.95s.
Measwzmentson CAD computer and 1: 10 wooden models and fi&ized materials sampleswere conducted over a 5-yr. pexid The
hailin planis rectangular.The ceilingis a distortedpyramid,with its peak nearer the stage than the rear of the hall and 28 rn above the
main floor. T5.s uniqueshapewas anal@ on the models so that all interiorsurfaces combine to distributesources on stage uniformly
over the seatingareas and to yield optimum values for RT, EDT, [I - IACC,], BR, ITDG, diDfusionand loudness(for deiinitionssee
Beranek1996, Concd and Opera Halls: How TheySound, Acous. Sot. Amer.) 0x1 the long ceiling facing the stage, Schroeder QRD
dihsers providediffusion,eliminatea possibleecho, and stnxqthen lateralreflqtions. Performersjudge the acoustics excellent7
INTRODUCTION
The TOC Concert HaJl (TOC) is contiguous to the New National Theatre complex. A joint-venture design team was
assembled by an Executive Committee, with Takahiko Yanagisawa of TAR Associated Architects as lead architect. The
requirements presented to the acoustical consultauts were: Q approximately 1630 seats; 8 rectangular shape in plan; 0
primarily for coneer@ and recital pertbrm~ ; @Ireverberation tune in the range of 1.8 to 2.0 seconds with full occupancy;
&I pyramidal ceiling; and @ natural wood interior. Acoustics before April 199 1 was the responsibility of S. Masuda.
On receiving the commimion of Acoustical Design Consultant for the TQC in April 199 1, L. Beranek met to lay out the
program with the acoustical stafT of the Takenaka R. & D. Institute of Chiba, Japan, headed by T. Hidaka, which was
retained to make models and perform acoustical measurements. Beranek cominued a systematic effort to assemble
drawings, photographs, details on materials and acoustical data measured by world-wide acoustical engineers for sixty-six
concert halls in regular use in twenty-two countries. The technical literature was reviewed to assist in establishing the
important acoustical characteristics that should be determined at various stages of its design and after its completion. The
results of these studies, in&ding drawings, photographs, descriptions, and acoustical data are combined in Beranek (19%).
Hidaka and staff made measurements of the important acoustical parameters, including those recently developed in
nineteen concert halls. The next step was their design of a CAD computer model. This was followed by construction of
a 10: 1 wooden model in which the audience was simulated, and where impulse signals were radiated from a tiny loudspeaker
at various positions on the stage to spherical heads in the audieuce with 1/8-m. microphones as “ears”.
DESIGN AND MEASUREMENTS
The acoustical characteristics that appeared from the studies to be most meaningful, which were measured in existing
halls, in models and in the completed hall, and the designs employed in TOC to accomplish them were:
A. Reflected sound waves from the surfaces of&e hall ikludiig, direction &d magnitude: The goal was to
achieve uniform sound projection over the audience from all points on the stage. Because early sound yields musical
clarity, the CAD model was used to determine the direction of the early waves arriving at each position and the 10: 1
model was employed to furnish information on the sequence of wave arrivals and their strengths by use of sound-
pressur&me plots called “refleetograms”. The unique pyramidal ceiling and other features of the hall are shown in
Figs. 1 and 2. The stage area is IO?/0greater than that in Boston Symphony Hall. Its sides are flared to better project
the sound to the audience The first balcony overhangs the back of the stage and is shaped to enhance communication
among orchestral sections and the conductor. A flat, irregular surfaced canopy, 9.6 rn square is hung over the front
part of the stage. The architect’s handling of the nnique pyramidal concept and hung canopy is shown in Fig. 2. To
enhance the energy in lateral reflections, and to avoid echoes as heard by musicians on stage, the long pyramidal
surface facing the stage is covered by Schroeder QRD di&sers.
B. Reverberation tlme (RT) occupied, and early decay tiPne (EDT) unoccupied: These quantities both a&et the
clarity and add Wness of tone (liveness) to the music. They were measured both in the 10: 1 model and in the concert
hall at various stages of its completion. The RT goal at mid-frequencies was 1.9 to 2.0 seconds.
C. Bits5 streragtlt (BR), tite support tbat the BnaIl(occupied) glvea to the low note5 of maask h relatio8n to the higher
notes: We chose as a measure of BR the ratio of the sum of the RT’s at 125 and 250 Hz to the sum at 500 and 1000
Hz. Achieving a ratio greater than unity was particularly diflicult in TOC because of the architect’s requirement that
the interior surfaces of the hall be wooden. A satisfactory ratio was achieved by employing chairs that were lightly
upholstered modeled after those in Vienna’s Grosser Musikvereinssaal.
D. htlmacy, the lmltial-&ne-delay gap t,: The hall should return the first sound reflection to the center part of the
main floor within 20 ms after arrival of the direct sound.. The t, was obtained from impulse responses and drawings.
E. Spaciousness, measared by the quantity one minus the interaural-cmss-comelatlon coefflcieslt, (Hidalca,
Beranek and Okano, 1995). A large value of spaciousness means that a considerable portion of the sound energy
at listeners’ ears arrives from near-lateral directions. Reflections from the parallel sidewalls, shaping of the balcony

1795
 faces, scattering elements on the balcony faces and the QRD diffusers resulted in high values of [ 1 - IACC,].
F. Dision: Irregularities on the walls, balcony faces and ceiling give the sound a rich acoustical patina. The fine-scale
diffusion on the side walls was caretilly developed Tom laboratory measurements to be effective above 1 kHz.
G. Clarity, C,, the mtio of the early energy (0 to 80 m) to the reverberant energy (80 to 3OOQms) im dB.
H. Quieti The HVAC noise, hall unoccupied, is about NCB-20. There are no troublesome ethos.
I. Loudness: The measure of loudness used is (V/EDT)xlO‘* (Beranek 1996a).
TOC concert hall was completed nearly a year in advance of its opening, September 10, 1997. This afforded the
opportunity for measurem ents, separately, before and after installation of seats, pipe organ and full occupancy. ln the form
of cloth covering, a simulated audience was developed. An unique opportunity led to data that suggest a new variable in
the awustics of concert halls, namely, the need for balance of energy in lateral reflections in different frequency regions.
This potential factor obviously needs testing under controlled conditions.
CONCLUSIONS
The values of the measured quantities in the TOC Concert Hall (given first) and the ranges of those measured in the
VkTmB ~~~?~rri~~~~~~~~~~~aal~A.en&rdaRl cQ~wtigcbw.%~ and B0&3n Syqhq 1LU, 3,-$&d StSndm-dS (in ~rq&:~~~). sz:
R&== 1.9&s (1.8-2.0 s)lfDTsE;= 2.59 s (2.4-3.0 s); t, = 15 ms (15-21 ms); C,= -2.6 dB (-3.7~2.7 @; [l - &KC,:]
=0.71 (0.62-0.71);@$= 1.05 (1.03-l.ll);(V,~&Tjx~~~=6.4~&(5.4to?.8~j;STl = -12.1 &(-I7.&+13.7~&).
T’bs n<,mk $&j&g t&g?t&c T()C has >qhqw2 ~~y&,~&&~, &y&z, &&q, q&,l:mq QT~&J & &q$ &
sound. The responses of mwsic critics, musicians, and wndactors have been excellent, indicating by their statcmeE% W
the hall takes its place among the best in the world. This success with an entirely new architectural solution, the pyramidal
ceiling, indicates that the need to duplicate precisely previously successful halls to achieve excellent results is no Ilongez
necessary, provided the principal acoustical parameters listed &ove have va.&s w;&zt with ?bose for the b& !&I~.

FIG. 1. The T&o Opera City Concert Ha.ll (TOC). The

REFERENCES Fig. 2. TOC Cocri H&l

1796