Hello! This newsletter is based on a solution our CTO Charlie White designed and AVDomotics implemented at Northland Preparatory Academy in Flagstaff, AZ. By sharing this, we hope to communicate the proper design and implementation of sound systems.

Problem: NPA had a multipurpose Gym area which included a stage for the performing arts. Like many school gyms with hard basketball court floors, cinder block walls and corrugated metal ceilings, the acoustics were horrible. Coaches voices where nearly impossible to understand due to reverb times of over 20 seconds and band concerts sounded like a mash of indiscernible sound. Improper implementation of sound reinforcement (speakers) only added to the problem.

Solution: With room dimensions and through calculations, Charlie White CTO AVDomotics, designed a solution that mitigated nearly all of the acoustic problems. The following is a walk through of the design process.

The problem of speech intelligibility is two fold: The first, and most important, is the room acoustics. The second is the sound reinforcement system.

Room Acoustics:

In the world of acoustics, there are basically two types of rooms, large and small. Small rooms are those in which the wavelength of the lower frequencies are approaching or are larger then the dimensions of the room. A large room is one that the dimensions of the room are larger than the wavelength of the lower sound frequencies. The reason for this distinction is that acoustics behave quite differently depending on the environment. Our main concern in this space is speech. For the average person, the frequencies we are concerned with when it comes to speech are from 125Hz to 4000Hz. Using the lower end of this scale the length in feet of 125 Hz is:

The smallest dimension of the room under question is approximately 27ft, so we can consider this room acoustically large.

Now that we have determined that the room is large we can proceed to analyze how sound behaves in the room. The main controlling factor in large room acoustics as it applies to intelligibility is the reverberation time. Known as RT60, this is the time that it takes for a given sound in a room to decay in volume 60 decibels from its original level. As a sound is emitted from a source it moves throughout the open space until it encounters a boundary. Each time sound encounters an obstacle some of its energy is absorbed and some of it is reflected. The more reflective materials there are in the space the longer the sound will ‘bounce’ around before it is completely absorbed.

If the volume of the room, the surface materials and square footage of each material, as well as the absorption coefficient of each material are known, the RT60 time can be calculated. When this was done for the room in question, the RT60 time was on average 27 seconds. In an ideal situation, speech is best understood when the RT60 time is closer to 1 second! So, it is no wonder that the intelligibility in the room was so awful!

We then recalculated with the addition of absorption panels. As can be seen from these calculations, 15,000 sqft of additional material would be required to lower the RT60 to 5 sec. In light of the multipurpose nature of the room, this would be plenty. However, this would be very expensive and nearly physically impossible. It is our opinion that 5000 sqft would suffice, bringing the RT60 down to 10.5 sec. Although this is not perfect, it does not account for the room having people in it, which happen to be amazingly efficient absorbers.

Absorption panels come in a wide variety of shapes, sizes, colors, etc. The ones we think are most suitable for this application are from a company called Sound Seal. We have shown two types, one that is built to be more impact resistant, i.e. basketballs, volleyballs etc., and one that is their standard panel. These panels simply get screwed to the walls in a way that is both functional and as aesthetically pleasing as possible.

Another option is hanging baffles. These are 2’x4’ pieces of absorptive material covered in cloth that would be hung from the ceiling between the metal rafters. The advantage of these is that they are out of the way and not as susceptible to damage. The disadvantage is that the surfaces that cause most of the problems are the walls.

The most efficient way to tackle this is to use a combination of both absorption methods. For example, hang as many baffles as cost allows from the ceiling to raise the overall absorption of the room while placing panels on the walls in key areas to treat very specific problem areas.

Sound Reinforcement System (speakers and speaker placement):

A good, user friendly sound system can go a long way in helping with speech intelligibility issues. The main goal in designing a system is to only direct the sound where it needs to be as much as possible. This reduces the chances of reflection from adjacent walls. In order to do this, one must first determine where that area is going to be. As you can see from the images (not included), we defined the main listening area to start from twelve feet from the stage (first row) to the back of the room, and to a line on either side of the room leaving out the slanted side wall areas.

Once the listening area is determined, speaker placement must be considered. When speech is your main concern having a stereo left right setup is not a good idea. There are several problems. First, unless there is a very well defined ‘sweet spot’ where most listeners will be, then stereo has little effect. Second, if you are listening to a non-stereo source, like speech, from two speakers that are separated by a large distance, the sound from each speaker will reach you at slightly different times. This time delay can significantly reduce intelligibility. With this in mind, we propose putting a small cluster of speakers centered over the stage, as high as possible.

Now that we have a listening area defined and a place to put the speakers we must determine what horizontal and vertical angles the speakers must cover. Horizontal coverage angles are determined by measuring the angle from the speaker to the left and right most points of the front and back rows of the listening area. Vertical coverage angles are determined by measuring the angle from the speaker to the center of the first and back rows. As you can see the horizontal angle at the front is much wider than that of the back. For this reason two speakers will be used together to cover the front while one will suffice for the back. The vertical angles of the speakers will overlap such that the entire area is covered. Speakers are specified with horizontal and vertical coverage angles. Based on these angles the speakers are carefully aimed to achieve our goal.