Any sound heard indoors is a combination of the original sound traveling directly from the source to the ear, plus many thousands of reflections from every surface in the room. Together they form a complex sound field around the listener.
When the room surfaces are "hard", the reflected sounds reach the ear at almost the same intensity as the original direct sound. But the reflections arrive slightly later because they travel longer paths. The ear gets confused by hundreds of similar sound images arriving at slightly different times, and perception becomes difficult (similar to a TV signal with multiple "ghost" images).
In a reverberant space such as a restaurant, people naturally exert more vocal effort to overcome the background noise, escalating the problem until everyone is shouting. Crowd noise can easily exceed 80-85 decibels.
Hard (reflective) surfaces are things like concrete, glass, gypsum board, or steel deck. Soft (absorptive) surfaces include acoustical tile, fabric-wrapped fiberglass panels, open-cell foam, unpainted concrete block, carpet, padded furnishings, etc.
Reverberation is the uncontrolled buildup of sound bouncing randomly within the room, and is best suppressed by the sound absorbing materials distributed throughout the room. When a sound is suddenly stopped, the reflections will continue for a time until they decay to inaudibility. This is called the Reverberation Time, or RT. RT is the most common measure of the ”liveness” of a room.
Echoes are distinct reflections, arriving late enough to be perceived as a separate signal. Echoes are best controlled by the room dimension ratios (a cube is particularly bad), and by the shape, angle and finishes on key surfaces.
Too much reverberation creates an unpleasant atmosphere and degrades speech clarity. Amplified speech will quickly become incomprehensible in a very "live" room. For example, a typical school gymnasium has very little absorption, and the room is highly reverberant. When speaking to a person five feet away, speech is easily understood. Backing further away, you quickly reach a point where it is virtually impossible to understand speech content. This happens because the intensity of direct sound (straight line path) diminishes naturally with distance. But the reflections are essentially unchanged, and soon overwhelm the direct signal.
In general, plan to add acoustical treatment on:
In most cases, 50-70% coverage of the surface is adequate. There is a point of diminishing returns, where extra material does not translate into further perceptible improvements.
The sound absorption of a material are measured in terms of the Noise Reduction Coefficient (NRC), which is simply the average absorption for the material for the audible speech frequencies (250 Hz – 2000 Hz). A material rated at NRC 0.70 material will absorb about 70% of the incoming sound energy, leaving the remaining 30% to either pass through the material or to be reflected back.
Sound passes through the porous absorptive material, which is why it cannot function as a barrier. If 50% of the energy is absorbed by the material, the remaining 50% of the energy transmits through. In the logarithmic realm of the decibel, a 50% decrease in energy is only three dB -- a generally unnoticeable amount of noise reduction.
Thicker absorptive materials perform better than thin materials, particularly for low frequency sounds. For example, a 1/2" thick fiberglass panel absorbs about 45% of the incoming sound energy, while the same material in 1" thickness will absorb about 80%.
Similar results can also be achieved by using airspace behind the material. A 1" thick fiber glass board mounted with 3" of airspace behind it performs almost as well as a 4" thick fiber glass board. The same thing applies for acoustical tile. In a direct mount against a hard backing, standard mineral fiber acoustical tile absorbs roughly 40% of the incoming sound energy. But in a suspended grid with several inches of backing airspace, the same tile is about 70% absorptive.
An acoustical material should not be painted
or sealed, because the paint seals the pores, which degrades the absorption
significantly. Painting a standard mineral fiber acoustical tile is
roughly equivalent to removing two-thirds of the material. Many rooms
have been fine acoustically, until someone decided to paint the ceiling tile.
Then the room suddenly becomes very reverberant.