by Bob Zurek
On May sixth, Jim Brown of Audio Systems Group gave a presentation to over 40 members of the AES Chicago Section on systems for stereophonic sound reinforcement. Mr. Brown began his presentation by bringing up the following points about stereophonic sound in large rooms. First he stated that stereo in large rooms is very different from stereo in small rooms. He also pointed out that two clusters of speakers do not necessarily make stereo, as well as that stereo does not necessarily mean 2 channels. Mr. Brown pointed out that things that are related to stereo are the directivity of human hearing, the difference between ears, time of arrival and phase differences, as well as pinnae transforms.
Mr. Brown continued by posing the question �Why use Stereo?� His answer to that that question began with: stereo allows acoustic summing of spaced microphones picking up the same sound source. He continued by stating that head directivity makes the acoustic summing better than electrical summing; acoustic summing is much less severe than electrical summing due to the lack of comb filtering in air. Mr. Brown went on to say that stereo in sound reinforcement provides a sense of spaciousness, audience satisfaction at a 6dB lower SPL, as well as localization. Jim then proceeded with a demonstration utilizing pink noise and a small sound reinforcement system to demonstrate comb filtering. He followed this up with demonstrations of music in mono and stereo played back over this small reinforcement system.
Mr. Brown then picked the presentation back up with a discussion of differences between large and small stereo systems. Some of the areas that he covered were: many listener seats versus few, the fact that arrival time differences dominate the large designs, controlling level difference is critical, and that visual source localization helps.
Jim then discussed some of the design criteria for stereo reinforcement systems, beginning with a discussion of work done at Bell Labs in 1936 that showed that reinforcement signal levels >10dB can break the precedence effect. He then stated that each channel must cover the entire audience with 3-6 dB amplitude shading to counteract precedence. He also noted the importance of uniform spectral balance and uniform loudness of the direct sound, as well as minimization of reverberant sound. He discussed that Snow of Bell Labs said to add an extra channel for each 25 feet of separation which is about 22ms. Jim also touched on delayed crossfeeds, and mono delayed sources to cover the far sides of rooms. He stated that an added mono delayed source should arrive just after the LCR signal and be at a level equal to sum of the LCR signals.
Jim then showed examples of stereo systems as some venues, and showed that stereo is not a lot more costly than mono reinforcement.Mr. Brown said that the additional costs for a stereo system over a mono system were: increased design complexity, a few good mixing decks for L-C-R, more signal processing, more power amps, and more speaker locations. Jim concluded his presentation with a discussion of how microphone techniques are different for large rooms as well. The micing in this case is done for many listeners, and he went on to state that most serious work since Snow on microphone technique has been geared towards small room listening. Jim discussed how spaced microphones enhance the system. He said that for large sources one spaced microphone should be used per channel. He also stated that coincident and near coincident stereo microphones do not work well due to the minimal time difference between them. He added that one should not try to make the mix too wide, and panning should be done only between adjacent channels. The presentation was followed by a question and answer session to augment the questions asked and answered throughout the presentation. The Chicago Section would like to thank Mr. Brown for a very informative presentation.
