Monograph ............................................................... 研究論文發表

台南市立化中心演藝廳之室內音響性能更新

ACOUSTICAL RENOVATION OF THE AUDITORIUM OF TAINAN MUNICIPAL CULTURAL CENTER

Weihwa Chiang, Yenkun Hsu, Chingtsong Huang, Ichao Chen, Liangkuang Yang, Jinchau Tsai

National Taiwan University of Science and Technology

43, Section 4, Keelung Road, Taipei City 106, Taiwan, R.O.C.

Fax: +866-2-27376721

E-mail: WHCH@mail.ntust.edu.tw

中文摘要

本研究探討台南文化中心演藝廳更新的室內音響性能設計以因應兩個市屬音樂表演團體的使用，設計目標除了提高室內整體餘響時間外，並包括加強樓座座席的強度、靠中軸線附近座席的側向反射能量。在維持原有的造型風格下，對應的設計手法包括將天花板及側牆後半部以及局部後牆吸音材改為反射材並於後牆裝設可調整吸音簾幕，天花板前段造型由梯度平面調整為弧度平緩的曲面，舞臺前移並於上方使用22塊懸吊反射版以及周邊使用MLS擴散材等。電腦模擬顯示中頻帶平均餘響時間(RT)(滿席)可至少提昇至1.7秒，1/20縮尺模型試驗顯示改變後的舞臺與原有舞臺的早期音響支持度(STearly)大致維持不變，但後期支持度(STlate)略微增加。完工後的音樂使用模式中頻帶餘響時間與初期衰減時間的平均值為1.95秒，音強指數(G)的平均值為4.6 dB，對35人國樂團與獨奏試演做評估的整體滿意度平均值為7分量表的5.3分，本研究將繼續探討完工後的舞臺以及劇場模式性能。

關鍵詞: 室內音響, 更新

ABSTRACT

Acoustical analysis was conducted regarding renovating the auditorium in Tainan Civic Center for the new symphony orchestra and the tradition orchestra.  The renovation was aimed to increase reverberation, lateral energy, and overall strength.  The absorbent real ceiling was replaced accordingly.  Major design strategies included replacing the absorbent ceiling walls by gypsum concrete board and installing retractable curtain, pushing the performing forward to and installing suspended reflectors above the stage and the audience, and splaying the rear parts of the side walls on the mail floor.  Computer simulation showed that occupied reverberation time could be at least increased to 1.7 sec.  Based on scale modeling, the renovated stage provided a similar early support as did the original stage but the late support was increased.  After completion the average value of early decay time and reverberation time was 1.95 in mid frequencies.  Hall averaged strength factor (G) was 4.6 dB in mid frequencies.  Evaluation of a rehearsal of the Tainan Traditional Orchestra with 35 players yielded an average overall impression of 5.3 out of a 7-point scale.  On going research is being made to measure the qualities on stage and the qualities of theater mode.

KEYWORDS:  Room acoustics, Renovation 

1. INTRODUCTION

In multi-purpose halls variable acoustics have to be provided to fulfill the needs for a variety of programs.  Variable reverberation time has proved to be the most valuable feature one can achieve by varying absorption and volume.  However, acoustics have often trailed behind staging and lighting.  Room width wider than 25 m, a fan-shape plan at the performing end and fragmented ceilings become common features in a large-size multipurpose hall with a proscenium stage. [1]

Besides reverberation, Barron also suggested the importance of controlling early reflections that would affect perceived strength, clarity, and apparent source width. [1]  For example, movable canopies were widely used by Artec Consultants, Inc. in multi-purpose halls.  Furthermore, early reflections arriving laterally could avoid the conflicting with stage lighting and be especially effective in large size halls.  They can be provided with tilted upper side-walls such as in Segrestroom Hall or terraced seating such as in the Berlin Philharmonie. [2] [4]

The stage volume in the range of approximately 1000-m³ to 2500-m³ has been derived by Gade to provide the optimum ST1 (or STearly) values 12±1 dB.  However, after orchestra shells being used for years, Jaffe suggested that a small stage-to-audience volume ratio would result in harsh sound on the stage and over-powered brass sound in the audience.  On the other hand, integrating the stage into the audience would result in insufficient support for the stage.  Placing surfaces suspended 6 to 8m above the floor would resolve this problem. [5] [6]

1.1 The auditorium of Tainan Municipal Cultural Center

 Except in Taipei metropolitan area, the majority of large size auditoriums built in Taiwan were multi-purpose and normally with proscenium stages. [7]  Opened in 1984, the auditorium of Tainan Municipal Cultural Center was a 2162-seat hall with a audience volume around 16350 m³.  The renovation was initiated mainly because the ceiling was damaged and two new orchestra supported by the city governments were recently founded.  The hall is very good reputed especially for its clarity and low noise.  However, the measured reverberation time (with the orchestra shell and unoccupied audience) in the range of 1.52 to 1.73 sec was too low for orchestral programs (FIG 1). [7][8]  This can be attributed to the absorbent rear ceiling, rear side walls, and rear walls.  Accommodation of all the seats in three audience layers resulted in a ceiling height of approximately 20m and a room width of 30 m. 

Fig. 1  Reverberation time published in 2 literatures. [7][8]

Fig. 2  Longitudinal section comparing the original ceiling to the remodeled ceiling.

Fig. 3  Computer generated perspective view of the suspended panels.     

2. DESIGN

The design goal was to provide the new house orchestra with more sufficient reverberation and lateral energy globally, and enhanced strength specifically on balcony seats.  The occupied mid-frequency reverberation time was set to 1.7 sec (with an orchestra on stage) and the unoccupied strength factor was set to be greater than 4 dB.  Guidelines reviewed previously were used as the possible solutions to improve the acoustical qualities.  The acoustical analyses were performed using computer models and 1/20-scale models.  Due to a low budget and the attempt to keep each side wall as an integral piece, attentions were placed upon the ceiling system.  All the catwalks were preserved.

2.2 Design Strategies 

The original stepped ceiling was modified to three major curves to provide multiple reflections to the first balcony (FIG 2).   The absorbent real ceiling and rear side walls were replaced accordingly.  Surfaces on real walls were replaced by gypsum concrete board but with retractable curtain installed as the variable absorption element to provide a low reverberation time for theatrical programs.   

Twelve 2.75-m² triangular reflectors in two row were suspended approximately 7.5 m above the stage to provide support on stage. (FIG 3)  Another 10 reflectors in two rows were placed 9 to 11 ms above the audience to provide early energy to the main floor.  The triangular arrays were adopted from the ones used in the Tanglewood Shed but the outer reflectors were tilted 20° to reflect sound laterally to the seats along the central axis.  The total area of the reflector was about 1/18 of the total area (1,190 m²).  The 60-% opening percentage among the reflectors allowed sound energy to propagate through the reflector layers to retain all spaces as an integral part.   

Two minor changes were also made:

1.      The rear parts of the side walls on the mail floor were tilted to reflect sound to the seats under the balcony. (FIG. 4)

2.      By including the orchestra pit as the platform for music performance, the depth of the shell was reduced by approximately 3 m.  The total performing excluding the triangular-shape edges on both sides was about 200 m².

3.SIMULATIONS AND TESTINGS

3.1  Computer Simulations

Computer simulation using Odeon 4.0 software was used to evaluate the acoustics in the audience during the design processes.  The materials were set to obtain a similar value of the published mid-frequency reverberation times (1.52 sec. and 1.73 sec).  The impulse responses of 14 audience receivers were calculated. 

With all absorbent surfaces being changes to reflective ones and all other modifications being made, average mid-frequency reverberation time (occupied) was in the ranged of 1.7 to 1.9 sec.  Average lateral energy fraction was increased to 0.19.  The strength of all the balcony seats was raised from 0.8 dB to 2.6 dB.  The short reverberation time of the theater mode (1.2 to 1.4 secs occupied) was achieved by opening the retractable curtain on the rear wall. Nevertheless, due to the uncertainty of the existing conditions and computer simulation, acoustical measurements were being taken after all the absorbent ceiling and partial of rear side wall have being replaced to determine how much of the remained absorbent materials should be changed to reflective surfaces.

3.2 Scale model test

Scale model testing was used to measure early support (STearly) and late support (STlate) on stage comparing the conditions before and after renovation.  A 1/20 model of partial of the hall was made. Detailed designs of the orchestra shells were also analyzed.  The measurements were taken at 3 locations (1-m source-receiver distance). (FIG. 5) 

The values of early supports were similar before and after renovation (Table 1) (FIG. 6).   Late support was increased after renovation.

Table 1.  Summary of results based on the scale model test

Fig. 5  Locations of measurements based on the suggestions by Gade.

Fig. 6  Log-squared impulse responses of point 1 (left), 3 (center), and 6 (right) comparing the conditions before (upper) and after renovation (with MLS and curved ceiling on shell) (lower).

4. EVALUATIONS AFTER COMPLETION

Acoustical measurements were taken in the field approximately 2 weeks before completion to determine how much of the remained absorbent materials on side walls and on rear walls should be changed to reflective surfaces.  The absorbent materials on the rear wall of the main floor and on the first balcony were preserved.  

4.1 Physical measurements

Fig. 7  Average values of measured reverberation time as a function of frequency band comparing source positions and calculation methods.

Fig. 8  Measured strength factor as a function of frequency band comparing source positions and statistics.

The average value of early decay time (EDT) and reverberation time (T20) was 1.95 in mid frequencies while EDT was approximately 0.2 sec shorter than T20. (FIG. 7)  This would probably yielded an occupied reverberation time in the range between 1.7 to 1.75 secs.   The clarity index (C80) averaged over 0.5 to 2 kHz was about 0.6 dB, which was near the upper limit (1 dB) Beranek suggested for orchestral programs.  

The hall averaged mid-frequency strength factor (G) was 4.6 dB and 3.9 dB from the front source and the rear source respectively. (FIG. 8)  The minimum G values of the 14 locations were about 1.5 dB less than the average values.  The average mid-frequency LF from the front source was 0.173; slightly less than the expected value based on computer simulations.

4.2 Subjective Evaluation

Evaluations by 9 listeners were made at 2 locations on the main floor during a rehearsal of the Tainan Traditional Orchestra.  The questionnaire constructed of eleven 7-point semantic scales.  The programs varied from a full orchestra with 35 players, chamber groups, solo with accompaniment by the orchestra or by the chamber group, and solo.  During the evaluation, all of the retractable curtains were halfway opened to approximate the decreased reverberation decay of occupied audience.

The average value overall impression was 5.3.   The 5.3 average score of clarity showed that the original characteristic of the hall has being retained.  Except for the attributes associated with tonal balance and the balance among varies parts, the differences due to changes of programs were relatively small. 

Fig. 9  Results of subjective evaluations comparing the four types of programs.

5. CONLUDING REMARKS

Renovating a hall with good reputation was challenging, especially when the budget was low and the computer modeling technology was not very matured.   Fortunately, positive signs were shown both by physical measurements and subjective evaluations.   Continuing works will be conducted to measure the qualities on stage and the qualities of theater mode. Subjective evaluations will also be made for orchestral programs in western style.

6. REFERENCES

1.               Barron, M. Auditorium Acoustics (E& FN Spon, London, 1993)

2.               Beranek, L.L.  Concert Halls and Opera Houses: How They Sound?   (Acoustical Society of America, Woodbury, 1996)

3.               Chiang, W. “Subjective Evaluation of Acoustical Environments for Solo Performance”, Building Acoustics, 21, 18-36, (1999)

4.               Cremer, L. and Muller, H.A., Principles and Applications of Room Acoustics (Trans. Theodore Schultz, Applied Science Publishers, New York, 1982).

5.               Gade, A.C. “Acoustical Survey of Eleven European Concert Halls- a Basis for Discussion of Halls in Denmark”, The Acoustics Lab., Tech. Univ. of Denmark, Report No.44. (1989)

6.               Jaffe, C., “Selective reflection and acoustic coupling in concert hall design” chapter 9, Music and concert hall acoustics (edited by Yoichi Ando, Academic Press Inc, 1997)

7.               Lai, R. Post-Occupational Evaluation of Acoustical Environments of Concert Halls: Analysis of Acoustical Qualities of Concerts Halls in Taiwan. (in Chinese) (Wenshan, Taipei, 1987)

8.               Lin, K. Post-Occupational Evaluation of the Audience Section in the auditoriums of Cultural Centers in Taiwan. (in Chinese) A Master’s Thesis of Cheng-Kung University. (1997)

回首頁       回 論文 與 研究