Results of the acoustical measurements Performed at L’Auditorium
de La Citè de La musique (La Villette) 20 May 2002
Introduction
This report contains the most relevant results of a set of acoustical measurements performed on 20th May 2002 in the Concert Hall of La Citè de La Musique, La Villete, Paris.
The measurement were performed employing an omnidirectional sound source (dodechaedron, LookLine D-300) and two different microphones: a Soundfield ST-250 (4 channels) and a binaural dummy head (Sennheiser MKE2002).
33 receiver positions were employed, with two different positions of the loudspeaker.
The sound source had been preliminarly calibrated, employing a Sound Intensity apparatus (B&K2260), in the Acoustics Laboratory of the University of Parma, ITALY.
This allowed for the measurement also of the relative sound level G (Strength). In practice, the test signal was the periodic repetition of a logarithmic sine sweep and of a sample of pink noise. Both were equalized for a perfectly flat 1/3 octave spectrum.
The sine sweep was employed for recovering the impulse responses, the pink noise for measuring accurately the Strength, by means of real-time spectrual measurements performed by means of a B&k type 2260 analyzer.
The sine sweeps were re-recorded by means of two different recording systems: the 4 signals coming from the Soundfield microphone were sampled through a Digigram VX-pocket 440 sound card, inerted in a notebook (Dell Inspiron 8000). The signals coming from the binaural microphone were recorded on a Sony TCD-D100 DAT recorder, and subsequently transferred on the computer by means of an optical digital link (through an USB adaptor, Waveterminal U2A). In both cases, the recordings were performed at 48 kHz, 16 bits.
The deconvolution of the impulse responses and the computation of the acoustical parameters was performed by means of CoolEditPro and of the Aurora plugins (www.ramsete.com/aurora)
The graphical representation of the results was obtained by means of an Excel spreadsheet (for the spectra) and by means of the mapping software Surfer (for contour line maps).
Some brief comments are added at some key points in the following, after the corresponding numerical parameters are shown. However, this report does not pretend to contain an exhaustive discussion of the results: it is simply a very aseptic presentation of the raw data, with minimal post-processing required for providing easy inspection of the huge amount of data collected during the measurements.
Photographs
Plan of the room and positions of the source and measurement points
Spectra of the spatially-averaged values
Citè de la Musique - Clarity C50
3.42 1.88
-0.75 -0.78 -0.83
-1.31 -2.20
-1.04 1.21
6.00
-0.56 -0.76
-6.00 -4.00 -2.00 0.00 2.00 4.00 6.00
31.5 63 125 250 500 1000 2000 4000 8000 16000 Lin A
Frequency (Hz)
C50 (dB)
Citè de la Musique - Clarity C80
5.62
3.73
2.09 1.63 1.46
0.88 0.13
1.46 3.88
9.71
1.84 1.65
-6.00 -4.00 -2.00 0.00 2.00 4.00 6.00
31.5 63 125 250 500 1000 2000 4000 8000 16000 Lin A
Frequency (Hz)
C80 (dB)
Comment #1:
The clarity at central frequencies (500, 1000, 2000 Hz) is just slightly too low for speech, and it is perfect for symphonic music.
Citè de la Musique - Strength (S1)
6.52
2.98
2.64
5.30
5.60 5.81
6.19
6.58
5.15
5.84
0 1 2 3 4 5 6 7
31.5 63 125 250 500 1000 2000 4000 8000 16000 Lin A
Frequency (Hz)
Strength (dB)
Comment #2:
The Strength at central frequencies (500, 1000, 2000 Hz) is perfect for symphonic music.
At 250 Hz the Strength, instead, is much lower, probably due to the grazing-incidence effect over the seats. At very low frequency the Strength increases again, but the value shown at 63 Hz appears to be suspect of some measurement errors (probably during the measurement of the sound power of the loudspeaker, which was made with ISO-9614-II, but employing a 12mm spacing for the sound intensity probe, which limits the lower useful frequency band to 125 Hz).
Citè de la Musique - Strength (S2)
7.15
2.77
1.64 3.00
3.80 3.95
4.21
4.58
3.86 3.94
0 1 2 3 4 5 6 7 8
31.5 63 125 250 500 1000 2000 4000 8000 16000 Lin A
Frequency (Hz)
Strength (dB)
Citè de la Musique - Reverberation Times
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
31.5 63 125 250 500 1000 2000 4000 8000 16000 Lin A
Frequency (Hz)
T60 (dB) T30
T20 T10 EDT
Comment #3:
The Reverberation Time at central frequencies (500, 1000, 2000 Hz) is perfect for symphonic music.
However, the rev.time falls significantly towards lower frequencies, with a dip at 125 Hz, which causes the sound to be significantly “cold”.
The optimal spectrum of the reverberation time for a symphonic concert hall should be inherently flat around 1.8 s, or should present a slight increase towards the lower frequencies, where values around 2s should be obtained.
The fact that EDT and T10 exhibit a much deeper dip at 125 Hz than T30 indicates that the surfaces causing the selective absorption at this frequency are close to either the source or the receiver positions. The most probable candidates for this effect are the seats.
In the following sheets, the spatial maps of the most relevant parameters are shown.
