Understanding Line Arrays, P A R T O N E
It has been said that there is nothing new under the sun. Most generations simply repeat the discoveries of the previous ones, the difference being that there are more technologies in place to bring the idea into existence. Even something as “modern” as the computer has lived in the human mind for hundreds, if not thousands of years. The search for a better mousetrap has given us many wonderful tools and innovations that improve our lives. The same process has driven sound reinforcement, and the last 100 years have shown a steady improvement in the quality of sound systems and listening rooms.
The latest rage in the loudspeaker industry is the line array. First, it should be noted that this is not a new technology. The line array has been around for as long as loudspeakers themselves and the current offerings are simply the modern incarnation of a very mature loudspeaker configuration.
Even the most advanced concepts used in modern line arrays were used routinely in sonar as early as the Second World War, and the ideas themselves are much older. There is nothing all that mysterious about the line array. Contrary to popular belief, it doesn’t defeat any of nature’s laws. In fact, it behaves exactly how it should given the way that waves combine with other waves in the physical world. Like any other tool, the key is to know when and how to use it. There are times when a line array might produce an advantage when radiating sound into a room. There are also times when its characteristics are a disadvantage. It’s up to the sound system designer to determine when and where a line array, or any loudspeaker type, should be used.
A line array is exactly what it sounds like: a group of loudspeakers assembled in a line. But before we talk about line arrays, let’s consider a more ideal version of this type of energy radiator - a line source. Some basics are in order.
Energy Sources
A “point source” radiates energy from approximately one point in space. A theoretical point source would be infi- nitely small, so it would appear to be a point from any observation distance. Real energy sources have a finite size, so point source-like behavior is dependent on the point from where you are observing. For example, our sun would appear to be a point when viewed from a distant galaxy, like the other stars appear to us. But if we observe it up close, it’s actually very large and not a point at all.
 |
fig.1 A LIGHT BULB SUSPENDED IN FREE SPACE APPEARS AS A POINT SOURCE FROM A DISTANT VANTAGE POINT |
 |
| fig. 2 A FLUORESCENT TUBE APPEARS AS A LINE SOURCE WHEN VIEWED FROM A CLOSE DISTANCE |
A line source radiates energy from a line. Going back to the lighting world, a four-foot fluorescent tube provides an example of a “finite” line source. The light radiates from all points on a line, not from a single point in space (Figure 2). Obviously, the length of the line source could be extended by placing more tubes end-to-end. Just as physicists like to consider how energy would radiate from an infinitely small point source, they also like to consider how energy would radiate from an infinitely long line source. That’s the neat thing about physics - it doesn’t have to be possible to be considered. An infinite line source in free space (no obstructions) would radiate energy cylindrically. A simple mathematical description of this phenomenon is P = 1/r where “P” is pressure (for sound) and “r” is the radius of the cylinder of radiation. When compared to the point source radiator, the energy does not fall off as quickly because it is only spreading in one plane - a pretty neat trick if you are trying to minimize energy attenuation over a long distance. The major advantage of the line source is less sound attenuation with increasing distance. In Figure 3, the SPL at point A and point B will be more similar if the sound is radiated from a line source rather than a point source.
Regarding loudness, the level from an ideal point source radiator in free space will attenuate at 6dB per doubling of distance. The level from a line source will attenuate at 3dB per doubling of distance. Since the goal of a sound system in an auditorium is to deliver the same sound level to all seats, a line source positioned on stage comes closer than the point source in achieving this.
 |
| fig. 3 THE SOUND LEVEL FROM A SINGLE LOUDSPEAKER WILL ATTENUATE AT A FASTER RATE THAN THE SOUND FROM A LINE OF LOUDSPEAKERS |
A continuous line source is easily realized in the electromagnetic domain. A stretched extension cord or overhead power lines passing electrical current are examples. It’s a little tougher with sound, where the energy is radiated by vibration. A line source can be approximated by placing a number of point sources in a line - hence the term “line array.” This works if the spacing between the sources is small when compared to the length of the sound waves that are radiated - a condition that effectively allows several individual sources to act like a single source. The wavelength is determined by the frequency of vibration and speed of propagation through the medium that is vibrating.
Wavelength can be understood by observing the waves that move away from an object tossed into a pool of water. It is the distance between identical points of successive waves. In sound, wavelength is inversely proportional to the frequency of the wave - the higher the frequency the shorter the wavelength. Sound waves range in size from 56 feet (20Hz) to about 1/4 inch (20kHz). The wavelengths that must be reproduced by a line array for speech range from about 4 feet (250Hz) to about 1 1/2 inches (8kHz). For a line of point sources to emulate a line source, they must be within one-half wavelength of each other at the frequency of interest. This means that they have to be smaller and more tightly packed with increasing frequency.
In practice, it’s usually easier to create low-frequency line arrays, since low-frequency waves are very long and allow for larger individual sources with a greater distance between them. As frequency increases, the need to “shrink and pack” the sources more closely produces some engineering challenges, and eventually it becomes impossible altogether due to limitations in how physically small a sound radiator can be and still emit a useful amount of energy.
 |
|
| fig. 4 THE RADIATION PATTERN OF AN IDEAL VERTICAL LINE ARRAY IS AN ELLIPSE. THIS IS USEFUL FOR REDUCING THE REFLECTED SOUND FROM THE FLOOR AND CEILING. |
This size/spacing relationship is vital to line array performance. If it is violated, the line array can have a very erratic radiation pattern, possibly defeating any benefits achieved. Since the performance of a line array changes with wavelength, we say that its response is “frequency dependent.” It is a non-trivial engineering task to create a broadband line array with useful radiation pattern control.
Of course, an infinite line array is a theoretical entity. Most real-world line arrays can only be considered infinite from a very close point of observation, like to a house fly perched on our florescent tube. If the line array is made a finite length, then an interesting thing happens. The radiation pattern will resemble a cylinder of finite height (I.e. a Frisbee). The radiation pattern for a vertical array will have a very broad horizontal coverage angle and a very narrow vertical one, which is a bit counterintuitive but nonetheless true (Figure 4).
Ironically, this is quite a useful pattern when we consider that it can be used to keep sound energy off of the ceiling and floor in an auditorium. Also, the vertical pattern becomes increasingly narrow with increasing distance from the source, like an ellipse. Again, this is not necessarily a bad thing if we use it right.
 |
|
| fig. 5 THE FINITE LINE ARRAY PRODUCES A NARROW VERTICAL RADIATION PATTERN. THE HORIZONTAL PATTERN IS THE SAME AS ANY SINGLE DEVICE IN THE ARRAY. |
Line arrays can be constructed from many different types of loudspeakers. Speech-optimized arrays are constructed from loudspeakers that are similar in size to the human mouth. This small size allows them to be packed very closely together. The array length (or height, if it is vertical) can be varied, but most designs are at least three feet tall or more. Greater lengths can produce greater working distances, along with some anomalies that we will address later.
Full-range line arrays are constructed from full-range loudspeakers. These loudspeakers are physically designed to allow close packing with minimal distance between adjacent units. Since the components are larger, the best listening positions will be farther from the array where the pathlength differences between elements are reduced on the listening axis.
Because of their radiation pattern, line arrays are best used at or just above ear height, and aimed at the last row of the audience. This placement takes the greatest advantage of the narrow vertical coverage pattern. If the array is tilted to place the on-axis position on the last row (i.e. the farthest seating position), then the closer seats will receive lower sound pressure level (SPL) than the farthest seats. But since they are closer to the array, they don’t need as much level. Playing the line array’s pattern against the listener distance can produce the same SPL at all listener positions, regardless of their distance from the array. Imagine an auditorium where the SPL is the same in the back row as in the front row and you can begin to appreciate the possibilities.
If the array is flown above a stage and aimed at the back row, much of the front of the audience will fall outside of the narrow vertical coverage pattern. One technique used to “correct” this is to bend the array - like a hockey stick or the letter “J.” J-arrays use the long part of the array to cover the rear of the audience and the bent part to cover closer seats. We might call this a hybrid array, since once bent the array is no longer linear and technically cannot be called a line array. J-arrays are useful in very large venues with distant and closeup seating positions and adequate ceiling height to house the array(s).
 |
|
| fig. 6 A PROCESSED LINE ARRAY DEDICATES A SIGNAL PROCESSING AND AMPLIFIER SECTION FOR EACH ELEMENT OF THE ARRAY. DELAY, LEVEL ADJUSTMENTS AND EQUALIZATION ALLOW THE RADIATION PATTERN TO BE TAILORED TO THE AUDIENCE AREA. |
The radiation pattern of a line array is ultimately controlled by the following parameters:
- The size of the individual components
- The response of the individual components
- The spacing of the individual components
- The relative timing of the individual components
- The length of the array
Items 2, 4, and 5 can be controlled electronically, and processed line arrays do just that. Each of the individual array elements is driven with its own amplifier and digital signal processor (Figure 6). This individual control allows three major parameters of the array to be adjusted: the focus distance, the vertical radiation angle, and the vertical coverage angle.
A processed line array can produce a radiation ellipse that focuses 100 feet from the stage, is aimed down at -3 degrees from perpendicular, and is confined to a 12-degree angle. Since this was obtained electronically, it can be adjusted to suit the venue. Think of it as “dial-a-speaker,” where the designer can create (within limits) the coverage pattern needed for the audience. This configuration can be taken to the next level by housing the loudspeakers, signal processing circuitry, and amplification in a single package, providing a truly versatile loudspeaker system that can be customized via signal processing per application.
Processed line arrays represent one of the most significant advancements in loudspeaker technology in the last half-century, and offer promise of future refinements and improvements. Some of the anomalies of a simple “stack of speakers” can be minimized or eliminated with the processed array.
In Part II, Pat Brown will discuss the practical benefits and drawbacks of line arrays, including what sorts of venues are best candidates for this technology.
