by Jeremy Krug
Privacy has all but vanished from the modern glass conference room and from much of the open-plan commercial office space. While additional frosting, static films, or vertical blinds can return some small measure of visual privacy, restoring speech privacy through acoustic treatment takes more knowledge, finesse, and careful specification. Acoustic treatment, after all, is not really ‘sound-proofing’ or ‘noise-cancelling.’
Architectural acoustics is the science of controlling sound energy, but there are a few terms that inevitably jump into the discussion because they are common household phrases, although they are widely misunderstood. Without going into too much detail, one should accept ‘sound-proofing,’ in the traditional sense, is really hard to do. In scientific terms, it is close to impossible—acoustical energy is measured to detect earthquakes half a world away, and its energy travels with no regard for fiberglass panels on the walls of 10,000 office suites along the way.
Even controlling weak forces—such as human voices between rooms in purpose-built recording studios, sound stages, or broadcast facilities—is a painstaking process where it is difficult to keep sounds in or out of adjacent spaces. This level of construction is far more specialized and expensive than a typical office space would warrant. While sound-proofing for human voices is not impossible, it is an expensive proposition and the required construction practices frequently run counter to modern office design.
‘Noise-cancelling,’ on the other hand, is somewhat possible in a very small and controllable space—one only needs to know where all the sound is coming from, where it will be heard, and then get in between. For instance, in the small gap between an ear and a set of headphones it is possible to perform some noise cancellation of the world outside. Incoming sounds captured through a microphone must be inverted and processed, and then played back through loudspeakers quickly enough to cancel the original sounds. The mathematics involved in accomplishing this in even a small room is far too chaotic since the sounds can arrive from nearly any direction and the listener can be in almost any position. Therefore, short of requiring everyone in the office to wear isolating headphones (not a very practical solution) true sound cancellation is impossible in this setting.
To understand why these concepts are not quite as easily achieved as might be hoped, and to figure out what options still remain, it is helpful to become a little more familiar with what sound really is and how it interacts with the physical world.
What is sound anyway?
Sound is energy that moves in waves through the air, but not in the simple wavy lines displayed on computer screens. Instead, the movement is a compression wave—a very soft and flexible spring. Air molecules are pushed away from the source and collide with other air molecules (compression)—increasing air pressure at the front of the wave and decreasing pressure behind (rarefaction). For example, the cones of a loudspeaker move back and forth thousands of times a second to literally push and pull the air, creating waves of acoustical energy or sound. The momentum is transferred to the next air molecules in line, while the first group falls back, and so on. This continues back and forth, with the whole process moving away at the source at the speed of sound—about 340 m (1120 ft) per second.
Sound energy travels out in all directions from the source three-dimensionally. Even when a door is closed and a barrier is created, the sound energy that passes through the gap between the door and floor will not travel out like a laser beam, rather it will spread out spherically again from the gap. Sound energy can transfer from the air through solid materials and back to air again. Its strength is weakened, but some energy still travels through.
Putting a number to it
Sound energy is measured in strength or intensity using decibels (dB), and its frequencies are measured in hertz (Hz). The decibel scale can be complicated because it is a logarithmic scale rather than a linear scale—meaning a 10-dB increase in signal is 10 times more energy, not 10 more units of energy. (Further, 3 dB is roughly double the power, 20 dB is 100 times more energy, and so on.) There are also many different decibel scales such as dB-sound power level (SPL), which is used to describe real acoustic energy moving in air. This scale will measure 0 dB-SPL as absolute silence. Typically quiet office ambient noise is around 30 to 35 dB-SPL, conversational speech is around 60 to 65 dB-SPL, and prolonged exposure above 85 dB-SPL is where the Occupational Safety and Health Administration (OSHA) steps in to regulate exposure and protection rules (Figure 1).
Frequencies in Hz or cycles-per-second are also measured on a logarithmic scale, although for most of architectural acoustics, the ranges are most important. Conveniently, there are just a few key numbers to know. The total range of human hearing runs from about 20 to 20,000 Hz. The equivalent of which would be the rumbling of earthquakes on the low end, up to dog whistles on the high end. The human vocal range runs from about 100 to 8000 Hz, while the critical range actually determines intelligibility or what we understand of human speech runs from about 200 to 5000 Hz. This number is important because covering sound in the critical 200 to 5000 Hz range is what speech privacy is all about.