Acoustics of the Teatro Arcimboldi in Milano. Part 2: Scale model studies, final results

AUDITORIUM ACOUSTICS 2 Imperial College (London), 19-21 July 22 Acoustics of the Teatro Arcimboldi in Milano. Part 2: Scale model studies, final results D. Commins (1), R. Pompoli (2), A. Farina (3), P. Fausti (2), N. Prodi (2) (1) Commins Acoustics Workshop (2) University of Ferrara (3) University of Parma 1

Teatro degli Arcimboldi: shape and dimensions Volume N. seats RTocc. mid Paris, Opéra Garnier 1 2131 1.15 Milano, La Scala, 1125 2135 1.2 Wien, Staatsoper 116 1658 1.3 London, Royal Opera House 1225 212 1.1 Milano, Arcimboldi 195 2385 1.7 Buenos Aires, Teatro Colon 257 2487 1.8 Paris, Opéra Bastille 21 27 1.5 New York, Metropolitan 24724 3816 1.8 2

Comparison between Teatro La Scala and Teatro degli Arcimboldi La Scala Arcimboldi La Scala Arcimboldi Volume V 11252 m 3 195 m 3 N. of seats N 2282 2385 Ratio V/N 5 m 3 /seat 8 m 3 /seat Average height in stalls 19 m 15 m Lenght of stalls in plan 32 m 32 m distance of listener more far 37 m 39 m Total internal surface (in plan) 1635 m 2 198 m 2 Surface of the orchestra pit 125 m 2 13 m 2 3

Scale Model 1:16 wood scale model MLSSA system running at a sampling rate of 125 khz Piezoelectric omnidirectional source (freq, response 1-5 khz) Poliurethan foam dummies Binaural ¼ microphones Measurements Real Room Aurora system running at 48 khz, 24 bits (Digigram sound card) Wide-band dodechaedron loudspeaker (equalized to flat sound power between 6-16k Hz) Sennheiser MKE22 binaural microphone Soundfield ST-25 microphone 4

Scale Model of Teatro degli Arcimboldi Goals: Study of early reflections Research of echoes, focalizations and shadows Optimal alignment of reflective panels. Technical details: The model is in scale 1:16 Plywood, 2 mm thick Reflectors are built in plexiglass The seats are completely full of audience the dummies are made of poliurethan foam, with wooden spherical head. 5

Scale Model measurements Source in orchestra pit, curtain closed Source on the stage, with orchestra shell 7 binaural receivers, (blu dots) 6

Equipment employed sound sources First Sound Source: 6-ways piezoelectric tweeter 7

Equipment employed sound sources Second Sound Source: piezoelectric tweeter with pipe adaptor Copper Pipe Plexiglass adaptor Cavity Tweeter Plasticine 8

9 Directivity of the second sound source 4 Hz (25 Hz) -2-15 -1-5 5 1 15 + +15 +3 +45 +6 +75 +9 +15 +12 +135 +15 +165 ±18-165 -15-135 -12-15 -9-75 -6-45 -3-15 315 4 5 2 Hz (125 Hz) -2-15 -1-5 5 1 15 + +15 +3 +45 +6 +75 +9 +15 +12 +135 +15 +165 ±18-165 -15-135 -12-15 -9-75 -6-45 -3-15 16 2 25 8 Hz (5 Hz) -2-15 -1-5 5 1 15 + +15 +3 +45 +6 +75 +9 +15 +12 +135 +15 +165 ±18-165 -15-135 -12-15 -9-75 -6-45 -3-15 63 8 1 16 Hz (1 Hz) -2-15 -1-5 5 1 15 + +15 +3 +45 +6 +75 +9 +15 +12 +135 +15 +165 ±18-165 -15-135 -12-15 -9-75 -6-45 -3-15 125 16 2

Equipment employed binaural measurement system Binaural microphone (1/4 capsules) MLSSA board in an Epson portable PC Home-made software for computing IACC 1

Typical impulse responses in the scale model 11

Measurements in the real room Session Source Position n. receivers Audience and musicians Reflectin gpanels G map A Orchestra Pit, curtains closed 7 (62) no yes Si B Orchestra Pit, curtains closed 7 (62) no no si C Stage, orchestra shell 7 no yes no D Stage, orchestra shell 2 si yes no E Stage, opera scenery 7 no yes no 12

Measurements in the real room Signal generation and emission Notebook PC with PCMCIA sound board (Digigram VX Pocket V2, 24 bits, 48 khz) CoolEdit with Aurora plugins (log sine sweep signal) Dodechaedron loudspeaker with integrated power amplifier and wireless remote control (LookLine D3) Binaural recording 2 Sennheiser MKE 22 dummy heads 1 Tascam DAP1 DAT recorder 1 Sony TCD-D1 DAT recorder Waveterminal U2A SPDIF-to-USB interface B-format recording Soundfield ST25 microphone Notebook PC with PCMCIA sound board (Digigram VX Pocket 444 bits, 48 khz, 4 channels) Sound Pressure Level measurement Real-time, hand-held 1/3 octave spectrum analyzer (B&K 226 Investigator) Post processing software Cool Edit Pro with Aurora plugins Surfer for G maps 13

Sound Source Measurement of the sound power with the Sound Intensity method (ISO-9614- II) The loudspeaker on the stage Amplifier with remote control 14

11 15 1 95 9 85 8 75 7 65 6 Sound Power Spectrum Sound Power Level - Look Line D3 Dodechaedron loudspeaker 15 5 63 8 1 125 16 2 25 315 4 5 63 8 1 125 16 2 25 315 4 5 63 8 1 A L Frequency (Hz) Unequalized Equalized Sound Power Level (db re 1pW)

Microphones Soundfield ST-25 (B format) Sennheiser MKE-22 (binaural) 16

Measurement of the directivity of the Soundfield ST-25 microphone Microphone on the automated rotating table CoolEditPro generating the MLS test signal, the pulses for the rotating table, and simultaneously recording channels W (omni) and X (figure of eight) Deconvolved impulse responses 17

Polar plots of the ST-25 directivity 125 Hz 25 Hz 285 28 275 27 265 26 255 34 34535 355.14 325 33335 32.12 315 31.1 35 3.8 295 29 25 245.6.4.2 24 235 23 225 22 215 21 25 2 195 19 185 18 5 1 15 2 25 3 35 4 45 5 55 6 65 7 75 8 85 9 95 1 15 11 115 12 125 13 135 14 145 15 155 16 17517 165 Pressure Velocity 28 275 27 265 26 29 285 255 25 325 33335 32 315 31 35 3 295 245.4 34 34535 355.35.3.25.2.15.1.5 24 235 23 225 22 215 21 25 2 195 19 185 18 5 1 15 2 25 3 35 4 45 5 55 6 65 7 75 8 85 9 95 1 15 11 115 12 125 13 135 14 145 15 155 16 17517 165 Pressure Velocity 4 Hz 8 Hz 28 275 27 265 26 29 285 255 25 325 33335 32 315 31 35 3 295 245 3.5 34 34535 355 24 235 23 225 22 215 21 25 2 195 19 185 3 2.5 2 1.5 1.5 18 5 1 15 2 25 3 35 4 45 5 55 6 65 7 75 8 85 9 95 1 15 11 115 12 125 13 135 14 145 15 155 16 17517 165 Pressure Velocity 28 275 27 265 26 29 285 255 25 325 33335 32 315 31 35 3 295 245 1.2 34 34535 355 24 235 23 225 22 215 21 25 2 195 19 185 1.8.6.4.2 18 5 1 15 2 25 3 35 4 45 5 55 6 65 7 75 8 85 9 95 1 15 11 115 12 125 13 135 14 145 15 155 16 17517 165 Pressure Velocity 18

Measured acoustical parameters (averages) Teatro Degli Arcimboldi - Strenght - 1 Hz Acoustical parameter Measured Designed Rev. Time T3 1.7 s 1,8 s Early Decay Time EDT 1.75 s 1,7 1,9 s Clarity C8.5 db 1 to +3 db Strenght to 4 db 2 to + 4 db I.A.C.C..24 <.7 I.T.D.G. 25 ms 25 ms T3 [s] 2. 1.8 1.6 1.4 1.2 1..8.6.4.2. Reverberation Time T3 (empty) 63 125 25 5 1 2 4 8 [Hz] Max Avg Min 19

Other acoustical parameters EDT (without audience) Clarity C8 2.5 12 2. 6 EDT [s] 1.5 1. Max Med Min C8 [db] Max Med Min.5-6. 63 125 25 5 1 2 4 8 Frequency [Hz] -12 63 125 25 5 1 2 4 8 Frequency [Hz] Center Time ts Clarity C5 2 12 18 16 6 ts [s] 14 12 1 8 6 Max Med Min C5 [db] -6-12 Max Med Min 4 2-18 63 125 25 5 1 2 4 8-24 63 125 25 5 1 2 4 8 Frequency [Hz] Frequency [Hz] 2

Research of possible echoes Wall in back of the stalls Session B2, center of stalls (1) Barron s Echo criterion Session C. center of stalls (2) 21

Comparison between measurements performed with and without lateral reflective panels G with panels - G without panels T3 [s] 2. 1.8 1.6 1.4 1.2 1..8.6.4.2. Reverberation Time T3 63 125 25 5 1 2 4 8 Frequency [Hz] Panels No panels.9 IACC with and without reflecting panels.8.7.6 IACC.5.4 Panels No Panels.3.2.1. 25 5 1 2 4 8 Frequency [Hz] 22

Comparison between measurements with and without audience Reverberation Time T3 [s] 2, 1,8 1,6 1,4 1,2 1,,8,6,4,2, 125 25 5 1 2 4 [Hz] M4 senza pubblico M4 con pubblico Absorption of seats Variation of T3 in various theatres 1,4 1, coefficiente d'assorbimento alfa 1,2 1,,8,6,4,2, 125 25 5 1 2 4 [Hz] poltrone vuote poltrone occupate [s],8,6,4,2, -,2 125 25 5 1 2 4 [Hz] Teatro degli Arcimboldi Teatro alla Scala New National Opera House, Tokyo Staatsoper, Vienna Segerstrom hall, Costa Mesa Sydney Opera House Operà Bastille, Parigi 23

Comparison between scale model and real room Reflection from the ceiling in three seats, source in the orchestra pit Center of rear stalls Upper Balcony Center of front stalls Model Real Room 24

Comparison between scale model and real room T2 [s] 2.5 2.3 2.1 1.9 1.7 1.5 1.3 1.1.9.7.5 Source inside the orchestra pit, curtain closed: T2 25 5 1 2 Frequency [Hz] Real Room Model 25

Comparison between scale model and real room Source on stage, with orchestra shell: T2 2.5 2.3 2.1 1.9 T2 [s] 1.7 1.5 1.3 1.1.9.7.5 125 25 5 1 2 Real room Model Model (compensated for air absorption) 26

Conclusions The scale model was used mainly for studying the early reflections and for orienting properly the lateral reflectors Reverberation time and clarity measured in the model were coincident with the real room only in two octave bands (5 and 1 Hz) The measurements performed in the real hall resulted in acoustical parameters strictly close to the design goal The presence of the reflecting panels revealed to be strongly beneficial for the sound field in the stalls, even if they were not optimally oriented The effect of seat occupance revealed to be very little, confirming that the chosen seats give an absorption almost independent of the presence of the listener 27

28