Without proper evaluation criteria, it is easy to be confused about what separates a good sound masking system from a poor one. The fact is, the criteria is fairly simple. A quality sound masking system needs to do two things. It must provide speech privacy and it must provide comfort. If it fails to do both, it will not work properly.
For over 50 years, an industry of acoustical academics, engineers and manufacturers in the speech privacy world have worked hard at defining privacy. Indeed, there certainly has been no shortage of ways to measure privacy. In fact, in many ways, the industry as a whole has agreed to disagree on the choice of one rating system and instead has allowed for a variety of standards, testing methods and measurements. Tests such as ASTM E-1130, E1374-02, E1110-06 and ANSI S-35-1969, and rating systems such as RASTI, STI, SII, AI and PI, all work within their own set of parameters and metrics for securing privacy, speech intelligibility and speech unintelligibility.
But what about a metric for comfort?
Some say, comfort is subjective and perhaps if you are talking about the temperature of your shower water, or considering the feel of a new mattress, comfort is indeed subjective. But when it comes to sound masking there are viable metrics that can be identified and measured and that can be used to define whether comfort has been achieved or not.
Acoustical Comfort for sound masking can be measured by the following five qualities: the audio distribution range of the sound that is initially generated produced and distributed; the uniformity of the sound masking throughout the entire space; the wrap around of the generated sound; the target dB level; and finally the privacy measurement achieved.
Therefore, let’s break each of these down further.
1. Audio Spectrum Distribution Range
In much the same way as with light, a quality sound masking system must have a full broad band spectrum. This quality is a driving factor when it comes to comfort. To provide acoustical comfort, a sound masking system must be able to produce and distribute a minimum spectrum range of 65hz to 16khz Many sound masking system noise generators cannot produce audio below 100hz and have speakers attached that are unable to distribute sound below 200hz.
A sound masking system must have the ability to produce a true quality sound. A system without the ability to create and distribute a full spectrum sound cannot produce a quaility environment. This would be the equivalent of installing dimly lit or narrow tonal light bulbs in an environment and similarly, would not be comfortable to “live” under. Remember, the ultimate consumable of any sound masking system is the sound that it creates and the final ambient sound people “hear”.
2. Uniformity
Next, the sound’s uniformity has a huge impact on comfort in two primary ways.
Here’s the first. Since the smallest variance that a normal human ear can perceive is approximately 1 dB of sound, a sound masking system with a tuning capability of +/- ? dB will give you a tighter tolerance and therefore a more uniform field of sound. A successful and comfortable sound masking system must be able to achieve a +/- 1 dB overall level for uniformity.
Unfortunately, many systems today are at best +/- 3dB. This condition of low uniform tolerances creates too wide a tolerance and creates a window at 6dB for overall sound level. This condition in any environment is distracting and uncomfortable.
Decibels don’t perform in a linear fashion, and every additional 10 dB of sound is perceived as a doubling of the sound volume level. This makes that 6dB window, a much more intrusive change. Quality sound masking must achieve tight tolerances to provide greater comfort.
Next, comfort is impacted by the design and layout of the speakers used to distribute the sound masking. When sound is distributed evenly throughout an environment, the space exists in acoustical harmony. If the sound distribution is spotty, or there is an uneven level of sound, the space becomes distracting. This happens because your ear continually has to adjust to the variances of the sound levels due to uneven levels of the frequency dispersion. To be comfortable, speakers must be uniformly distributed throughout an environment.
In addition, speaker orientation must be carefully considered. Typically, in-plenum speakers are used in most instances as they have been proven to be more effective in creating uniform sound fields across the full spectrum of sound. When an installation requires a downward firing speaker, these should be used with accordance with specific and proper design recommendations. It is also worthy to note that small downward speakers are unable to produce a full broadband noise and because they are much more directional that in-plenum speakers they can be quite distracting and on a whole are not recommended for masking
So when considering uniformity and comfort, tighter tolerances and a proper design layout is critical to achieve a smooth distribution of sound.
3. Wrap Around
How often a sound repeats or “wraps around,” affect comfort levels. For example, consider the large circulation fans that used to be in some classrooms. The “whirr” of the fan contained a cycle. A cycle which at first was almost imperceptible, but when subjected to over the course of the day, you would hear over and over again. This cycle, or repeat, became imprinted in your brain and as you continued to listen to the sound, you began to expect it, and the repetition became uncomfortable because your brain processes cycles as actual information. Fatigue would always follow as listening to a pattern or repeat over a few hours can become very tiresome.
Fact is, all audio masking systems have a repeat. However, the differentiator here, and what we need to strive for, are sound repeats that are spaced as far apart as possible, so that the brain will not be able to “tune in” to the repeat and eventually fatigue. To ensure a comfortable sound masking system choose one with a long repeat
For example, a masking system should have at least a 10 hour wrap around to ensure comfort. Many typical systems repeat every two to five minutes – which is way too often (remember that fan in the class room?). Easy enough here, the longer the wrap around, the more comfortable and less distracted the listener will be.
4. Target dB Level
A sound masking system must maintain a target ambient sound level and have a proper and a measurable sound level to speech level ratio. When done right, this creates acoustical comfort because it crushes (minimizes) the acoustical dynamic range and improves privacy (by raising the ambient background sound levels). For reference, a typical masking system in an office should have ambient sound levels at 47dBA. If it provides less than 45dBA, privacy won't be achieved. Greater than 49dBA and masking is typically intrusive. It’s a delicate balance. Don't be fooled by marketing campaigns that suggest your masking levels will achieve privacy in typical office space at lower dB levels. No privacy equals little comfort.
5. Privacy Measurement
For acoustical comfort, let's not forget that we need to have privacy in our space. A masking system intended to be used for privacy must meet standards for privacy such as ASTM E1130. By using a standard, we can ensure that the masking system levels are appropriately tuned and set.
Jonathan Leonard, president of Lencore Acoustics, has been involved in the commercial real estate, acoustics, construction, architectural and facility management industries for over 20 years.
