Room acoustics
Measurement of Directional Impulse Response
Measurement of Directional Impulse Response
Spherical Microphone & Loudspeakers
Steerable Parabolic Loudspeaker
I was fortunate to work at CNMAT for several years under the direction of Dr. David Wessel (1942-2014), famous for his work in timbre spaces and many other contributions to computing and music.
He also had a personal mission to "fix" Hertz Hall, UC Berkeley Music Department's main concert auditorium.
Ironically named, Hertz Hall was built with terrible acoustical problems: flutter echos, poor speech intelligibility, and bird-like "chirps" that could be triggered from specific locations on the stage.
Occasionally David would send the CNMAT team and several car-loads of experimental gear into Hertz Hall to attempt to diagnose and solve the problem.
In one session, we deployed our spherical loudspeaker array and its companion spherical microphone array (nicknamed "Sputnik") to attempt to measure the directional impulse response from the stage.
However, due to geometry spherical arrays are not great at forming tight beams. To test an idea, I built a servo-steered parabolic loudspeaker with a single tweeter at the focal point and a co-axial laser pointer, which was much more effective at stimulating a directional impulse response.
Using the parabolic loudspeaker, I was able to isolate the cause of the bird-like "chirp": the architect had designed the auditorium's side walls using periodic vertical slats approximately 2 inches wide. The slatted structure operates as an acoustic diffraction grating, causing the angle of sound deflection by the wall to depend upon the frequency. For an observer at a fixed point within the room, they experience a frequency-dependent delay as the path length of a reflection from the grating wall varies as a function of the reflection angle of sound incident to the wall.
Finally, we were able to diagnose the flutter echo problem: the curved back wall of the auditorium had a focal point on the stage, causing a periodic repetition of sound originating from the focal point. The flutter echo was ultimately fixed by adding absorptive panels to the center of the back wall.
Schwenke, Roger, et al. "Room acoustics measurements with an approximately spherical source of 120 drivers." The Journal of the Acoustical Society of America 123.5 (2008): 3771-3771.