Elektroacoustic systems

Transcription

1 Emplacement réservé pour une photo Elektroacoustic systems Isabelle Schmich and its installation in performance halls ACOUSTIC AND LIGHTING DEPARTMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 1

2 Reverberation Enhancement Systems - What For? Many auditoria suffer defects of various types, which are not always possible to correct architecturally. Too small volume, large overhang balconies, acoustically unsuitable shape, excess of sound absorbing materials, which h create : dry acoustics, dead areas, remote and weak sound, lack of homogeneity, disturbed sound images Other halls need variable acoustics because of their performance program ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 2

3 Reverberation Enhancement Systems - What For? Reverberation time for different types of performances Program Reverberation time (s) Popular music, Jazz 0,8-1,2 Lecture, Conference 0,7-1 Speech drama 1-1,2 Chamber music 1,3-1,5 Opera 1.1-1,6 Symphonic Concert 1,8-2,5 Looking for control and variability of : > Reverberation Time (RT) > Sound Pressure Level (SPL) and Early Energy Distribution > Intelligibility, Envelopment, Clarity > Space-time impression ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 3

4 Changing the Acoustics of a hall PASSIVE SOLUTIONS Movable absorbing/reflecting elements (curtains, panels, banners ) Movable ceiling or walls (change of the volume) Stage acoustic fittings (orchestra shell, panels, reflectors, orchestra risers) Cumbersome, limited efficiency, e cy, high cost ACTIVE SOLUTIONS Installing a Reverberation Enhancement System (increases RT and SPL) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 4

5 Development of electroacoustic systems ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 5

6 Basic equations Without system: S r = H sr. S s With one active cell : S r = H sr. S s + H hr. S h Open Loop Gain : G mh. H hm Signals: S h = G mh. S m = G mh. (H sm. S s + H hm. S h ) S m H sm Gmh S h H hm H hr S h = (I - G mh.h hm) ) -1. G mh. H sm. S s (I - G mh.h hm ) <==> rebouclage S s H sr S r Source -> Receiver transfer function : H tot = S r / S s = H sr + H hr. (I - G mh. H hm ) -1. G mh. H sm = H sr +H ea ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 6

7 System Families Feedback systems (MCR, CARMEN, AS, AFC): acoustic feedback is used to enhance the reverberation In-line systems (VIVACE, SIAP, ACS, ASR): acoustic feedback minimised electronic reverberation Hybrid systems (LARES, Constellation/VRAS): ti acoustic feedback used electronic reverberation - Absorption + Volume ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 7

8 Specifications for choosing a R.E.S. Undetectable by listeners, natural sounding No change of the acoustic qualities of the walls and the geometry of the hall Large variation of RT (> +50%) over the full frequency range Control of early energy (clarity, intelligibility) Inaudible self-background noise No space-time disturbance of sound images and orchestra balance Long term reliability Easy to use Must not conflict with other equipment (lighting, ) g, ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 8

9 CARMEN is a French Acronym of : Contrôle Actif de la Réverbération par Murs virtuels à Effet Naturel (Active control of reverberation by natural effect virtual walls) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 9

10 Active Virtual Wall Principle A small number of active cells, each composed of a microphone close to the associated loudspeaker, spread all around the hall. Acoustic and electronic filters suited for controlling cells stability. Quasi-locally reacting cells behave as active reflectors/diffusers placed on walls and ceiling Reverberation builds up by natural acoustic interaction between the active virtual walls cells, just as natural reverberation. ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 10

11 Active Virtual Wall Principle Each cell is composed of : a loudspeaker an electronic processing unit a microphone insensitive to the loudspeaker acoustic radiation Pi A/D Pc DIGITAL SIGNAL PROCESSING Pr = r.(pi - Pc) D/A Several independent cells behave as a virtual wall ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 11

12 CARMEN - Block Diagram ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 12

13 CARMEN Main Points Virtual positioning i of the ceiling and walls Reverberation enhancement produced by natural energy exchange between the virtual walls elements (no artificial reverberation). ΔRT: [50%,100 %] for 125 Hz Hz Space-time coherence of the sound field is preserved Reinforced early reflections Fully natural sound picking up from stage (no source image shift). Good subjective perception of the volume Public Address possible ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 13

14 Some examples of working CARMEN Systems ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 14

15 Grimaldi Forum, Monaco Multipurpose auditorium with: - Theatre - Chamber music - Ballet -Opera - Symphonic Concert Salle des princes: V = m 3 N = 1900 seats T R = [ 1.2 s s] ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 15

16 Grimaldi Forum, Monaco 29 cells 14 lateral cells 15 cells in the ceiling 3 forestage microphones ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 16

17 Grimaldi Forum, Monaco ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 17

18 Grimaldi Forum, Monaco ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 18

19 Acoustic Results in the Salle des princes, Monaco Reverberation Time TR20 (s) 4 RT (s) Preset 6 3 Preset 5 sans conque Preset 4 avec conque 2 Preset 3 Preset 2 1 Preset 1 Inactif Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 19

20 Salle des princes, Monaco Config0=Passif if Config1=Théatre Config2=Opéra Config3=Concerto Config4=Symphonique Config5=Grand Symphonique 1 Config6=Grand Symphonique 2 ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 20

21 Salle des princes, Monaco ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 21

22 Brighton Dome, UK V = m 3 N = 1900 seats T R = [1.2s - 2.1s] 30 cells : 6 in the tier front 12 in the columns below the balconies (2x6) 8 in the dome 4 over the stage ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 22

23 Acoustic Results in the Brighton Dome, UK Reverberation Time Tr 20 (s) Conf1: Chamber music Tr (s) conf 0 conf 1 conf 2 conf 3 conf 4 Conf2: Concerto Conf3: Symphonic Conf4: Opera Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 23

24 Norwich Theatre Royal, UK V = 3800 m 3 N = 1280 seats T R = [ 1.1 s s] cells : 14 cells on the lateral wall 3 cells below the balcony 6 cells in the ceiling 4 cells in the dome 2 forestage microphones ' ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 24

25 Norwich Theatre Royal, UK Measurement points ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 25

26 Acoustic Results in the Norwich Theatre Royal, UK Echograms 0 Chorus Symphonic -10 Concerto -20 Chamber Music System off Gain (db) Time (s) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 26

27 Acoustic Results in the Norwich Theatre Royal, UK Reverberation Time Passive Chamber Music Opera Concerto Symphonic Chorus RT T30 (s) Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 27

28 Palace of Arts, Miskolc TCS TCS TCS TCS TCS TCS cells : V = 3500 m 3 N = 580 seats T R = [ 1.3 s s] 8 cells at the lateral walls 2 cells under the balcony 9 cells at the ceiling over the audience 5 cells at the stage reflectors ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 28

29 Palace of Arts, Miskolc ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 29

30 Acoustic Results in the Palace of Arts Reverberation Time 4 3 Preset 9 Preset 8 Preset 7 Oratorio Chorus Roman. Symphony RT20 (s) 2 Preset 6 Preset 5 Preset 4 Preset 3 Classic Symphony Concerto + Concerto Opera 1 Preset 2 Chambre music Preset 1 Speech Drama Passive Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 30

31 Waterside Theatre, Aylesbury V 7500 m 3 N = 1206 seats T R = [ 1.1 s 2.0 s] cells : 12 cells at the lateral walls 6 cells under the balconies ceilings cells on the ceiling over the audience 3 cells on the forestage reflector ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 31

32 Waterside Theatre, Aylesbury ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 32

33 Acoustic results Waterside Theatre, Aylesbury ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 33

34 Acoustic results Waterside Theatre, Aylesbury ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 34

35 Théâtre Mogador, Paris 31 cells : V =7000 m 3 18 lateral cells (2x9) N =1800 seats T R =[1.3 s s] 7 cells in the dome and balcony ceiling 6 double cells in balconies soffit 2 forestage microphones ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 35

36 Acoustic Results in the Théâtre Mogador Reverberation Time RT RT (s) Conf 4 Conf 5 Conf 3 Conf 2 Conf 1 Passif Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 36

37 CARMEN at Berlioz Festival (La Côte Saint-André, France) V = 7500 m 3 N = 1100 seats T R = [1.2 s s]? 24 cells : 8 lateral 11 in the ceiling 5 on stage ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 37

38 Reverberation time Acoustic Results in la Côte Saint-André RT20 (s) 4.00 RT (s s) Preset 1 Preset 2 Preset 3 Inactif Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 38

39 Echogram in la Côte Saint-André ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 39

40 A NEW REVERBERATION ENHANCEMENT SYSTEM CARMENCITA ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 40

41 Why a new system? Existing CARMEN system: for major hall between 800 and 2500 seats where classical music performances are given on a regular basis Carmencita system: for small multi-purpose halls between 200 and 800 seats where classical music performances are occasionally given ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 41

42 Carmencita Principle 1 Based on the CARMEN system principle but: DSP DSP Hybrid system Less cells (8-12) Network connection between the cells Less expensive microphones DSP Remote-control Network Computer DSP Carmencita Bloc Diagram ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 42

43 Carmencita Principle 2 crophones Mi Carmencita Signal Matrix Loudspeakers x x x x x x 2. 0 x x x x x x 3 x x 0. x x x x 4 x x. 0 x x x x 5 x x x x x x x x x x x x x x x x... 0 Some signal transfers added Introducing delays and gain Discrete elecronic reveberation 0 = equalization filter X = reverberation filter. = optional reverberation filter ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 43

44 Carmencita in La Rampe Multi-purpose hall Metallic structure Absorbingbi walls and ceiling Movable tier for public 700 to 800 seats V = 4500 m3 RT off = 1.2 s CARMEN system with 24 cells installed ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 44

45 Carmencita Test layout 4 cells on the side walls 4 cells in the ceiling No cells on stage ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 45

46 Carmencita Test Results 1-10 Echogram -20 Carmencita Passive -30 Gain (db) Time (s) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 46

47 Carmencita Test Results 2 RT20 (s) Reverberation Time of Carmen and Carmencita Pas siv e 3 Carmen 8 cells Carmencita 8 cells Carmen 24 cells Preset 1 Carmen 24 cells Preset 2 Carmen 24 cells Preset Frequency (Hz) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 47

48 Conclusion Preliminary tests have been successfully done at La Rampe. The measurements allow the evaluation of the efficiency of the system. Listening tests for assessing the good acoustic quality of the system have been carried out. This electroacoustic system is totally discrete and will be easy to install into multi-purpose halls. Further tests and development need to be done. The final version of a Carmencita cell will be an integrated system composed of a microphone, a specially designed loudspeaker, DSP and temperature sensor. Each cell will be very easy to mount into the hall. ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 48

49 Questions? ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 49

50 Regenerative / In-line systems ADVANTAGES Regenerative systems > Natural sound > Space and time respect of sound source images > No use of electronic reverberation In-Line Systems > First reflections control > SPL / RT uncoupled > Very long RT possible DISADVANTAGES Regenerative Systems > Diffuse action on 1 st reflections > SPL / RT Coupling > Background HVAC noise increased In-Line Systems > Disturbance of sound images > Uneven stage covering > Frequent ent coloration due to electronic reverberation > Disturbing time variance effects ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 50

51 What is colouration? Each hall has its own tonal quality that can be characterized through the colour of the hall Colouration is due to a different time-decay of the sound energy at specific frequencies caused by room resonance effects. ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 51

52 CARMEN Tuning Parameters 6 parametric filters 22 tuning parameters for one wall Cell 1 G τ G,F,W,S G,F,W,S Cell 2 G τ G,F,W,S G,F,W,S Wall 1 5 parametric filters G τ G,F,W,S G,F,W,S Cell m G τ G,F,W,S G,F,W,S Cell n G τ G,F,W,S G,F,W,S Wall n G τ G,F,W,S G,F,W,S Cell p G τ G,F,W,S G,F,W,S 26 tuning parameters for one cell ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 52

53 Tuning Systematic procedure: Checking of the installed hardware Tuning of each cell one by one Basic stability of the system: using an automatic stabilization algorithm Cells grouped into virtual walls Adjustment of each virtual wall parameters Slowly increase of reverberation until specifications are achieved Preset n Preset Fire curtain Reflector Forestage Stage drapery 1 Speech Drama open open + up seat level + seats Yes 2 Ballet (orchestra in the pit) open open + up orchestra pit Yes 3 Opera (orchestra in the pit) open open + up orchestra pit Yes 4 Symphony No closed closed + down Forestage at stage level (orchestra on forestage) 5 Chorus (choir on forestage) closed closed + down Forestage at stage level No Creation of different presets in order to achieve the best results for a specific performance type. Each preset can provide a different sound color of the hall (eg. adjusting the frequency response). Speech Drama + Opera preset use forestage microphones to increase intelligibility. Manually preset tuning: subjective listening is more precise than existing measurable parameters. ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 53

54 Carmencita Tuning Tests with a ProTools system and a specially designed Plug-In Equalisation for each individually cell = diagonal of the signal matrix. Addition of the non-diagonal terms of the signal matrix. Delay and gain applied according to the shape of the hall so that the signal keeps the complete physical meaning when introduced into another cell. Some very discrete electronic reverberators with low gain added to increase the efficiency of the cell. ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 54

55 Carmencita Test Results 3 RT (s) 125 Hz 250 Hz 500 Hz 1000 Hz 2000 Hz 4000 Hz Passive Carmen 8 cells Carmencita 8 cells Carmen 24 cells (preset 1) Carmen 24 cells (preset 2) Carmen 24 cells (preset 3) Carmencita globally increases the RT by 29% (34% at 500 Hz) (only 8 cells) Each cell produces almost 4% of RT enhancement (CARMEN 2% / cell with 24 cells) RT increases 10% compared to the efficiency of a CARMEN system with 8 cells Because the hall is at the upper limit for the use of the Carmencita system it seems logical that no more reverberation enhancement is achieved in this volume (12 cells?) ACOUSTIC AND LIGHTING DEPARTEMENT CARMEN Microtech Gefell - 27 th March 2012 PAGE 55