Although most medium and large events today use line arrays, we must not forget that they are not always the ideal system for every case. Using a line array requires more advanced knowledge of its theory, as well as of the manufacturer's user manual. Each brand has its own method of rigging and setup; this is not a drawback, since acquiring that knowledge takes up no space and costs no money.
It is also true that if the venue is small, if we don't have a flying system, or if the shape of the venue and other factors come into play, ground-stacked systems become the preferred option. Until the arrival of these new technologies, that was the way we worked. And the question I always ask myself is: did concerts sound bad before line arrays came along? No! Most of them sounded very good! So let's use those “techniques.”
Subwoofer arrays were already explained in detail in previous articles, and we already know what happens to their sound pressure and coverage depending on whether they are arranged in a line, an arc, cardioid…
We know that when we set up a classic system with more than one cabinet, there will be cancellations caused by the difference in the time it takes the sound to reach our ears. This is due to the difference in distance, so we have a phase difference. This will always exist, although to a greater extent in conventional systems. Remember that a 1ms difference between signals will produce a cancellation frequency of 1000Hz, whose first cancellation will be at 500Hz.
We need to be clear on some concepts and terms before continuing
Coupling: When there is maximum summation and minimal cancellation.
Combining: When there is a moderate amount of summation and somewhat less cancellation.
Combing: When there is a lot of cancellation, regardless of whether there is summation or not.
Echo: When discrete sources are perceived.
Reverberation: When the normal decay character of a room is perceived.
Isolation: When the second loudspeaker has little or no audible effect on the response of the first. As the time mismatch increases, a greater level mismatch is required to achieve isolation.
Which of these cases occurs depends on two key factors
Time mismatch: The difference in arrival time, or phase, of the sound sources at a given listening position. It depends on frequency.
Relative level mismatch: The difference in level of the sound sources at a given listening position.
Coupling occurs when both the time mismatch and the level mismatch approach zero. In this case, the signals can add up to 6 dB of sound pressure level. This is easier to achieve in low-frequency arrays, where the wavelength is long.
Combining occurs when the time mismatch is low and the level mismatch is high. To achieve this, the components must be very close together (low time mismatch), but there must be a way to obtain a high level mismatch. This can be achieved by using directional loudspeaker systems arranged as a point source, which widens the coverage.
Combing occurs when the time mismatch is large but the level mismatch is low. This happens when loudspeakers are arranged with redundant coverage patterns, in parallel arrays. While this can give us significant summation, it is highly position-dependent and the frequency response varies a great deal.
Echo occurs when the time difference is large and isolation is low, so that the systems sound like discrete sources.
Reverberation occurs when the time difference is large and isolation is large enough that the interaction sounds like the normal decay
of a room.
Isolation occurs when the level mismatch is large enough that the second loudspeaker has very little or no audible effect on the response of the first. As the time mismatch increases, a greater level mismatch will be required to achieve isolation. This is a highly recommended option.
Characteristics to consider when designing arrays
These are the main factors to keep in mind when building loudspeaker arrays:
Coverage: As the overlap zone increases, coverage narrows, and vice versa.
On-axis SPL: As the overlap zone increases, the on-axis sound pressure level increases.
Level distribution: As the overlap zone increases, the level distribution becomes highly variable.
Frequency response distribution: As the overlap zone increases, the frequency response distribution becomes highly variable. As the overlap decreases, the frequency response distribution becomes smoother.
Equalizability: Virtually any array can be equalized at a single point. But if we assume the goal is to provide an EQ curve that suits a large real portion of the coverage area, arrays with constant distribution patterns will respond better.
Classification of arrays
Horizontal arrays
There are several types of loudspeaker arrays. Each array has different pros and cons. When two loudspeakers are placed close to each other, we can classify them as:
- Point-source narrow:
When two loudspeakers are arranged in an arc in which the individual patterns of the cabinets are wider than the angle of the array.This results in a narrowing of the pattern (between the -6 dB points).
Maximum summation in the central area and less at the sides. Good low- and mid-frequency coupling. The large overlap area creates a wide area with a lot of "ripple" around the center and smoother response at the sides. It responds well to EQ except in the central overlap area. Applications: Long throw, and when the desired coverage angle is smaller than that of a single loudspeaker - Point-source wide arrays:
When two loudspeakers are arranged in an arc in which the individual patterns of the cabinets approach the angle of the array. This results in a widening of the pattern (between the -6 dB points).As for coverage, central buildup is minimal, so the pattern widens; on-axis sound pressure level is at its maximum. The small overlap area produces a narrow area with a lot of "ripple" around the center. It is very smooth except in the central area. It responds very well to EQ, except in the central overlap area. Applications: When the desired coverage angle is wider than that of one loudspeaker.
- Crossfire arrays:
- When two loudspeakers are arranged so that their patterns cross directly in front of the horn. This leads to significantly greater interference problems; it is not recommended.
The central buildup causes a narrowing of the area between the -6 dB points; on axis we get maximum summation. The frequency response and its equalizability are poor. Most of the central area has too much "ripple." It should never be used.
- Parallel arrays:
- When two loudspeakers are arranged on a parallel plane. The patterns overlap and create a highly inconsistent response; it is not recommended.
Coverage is the same as with a single loudspeaker, and on-axis sound pressure level delivers maximum summation. As for frequency response, each position has its own particular frequency response. High-frequency "ripple" is severe at all positions. It can be equalized for only one point; EQ will have a very hard time correcting the system. Applications: Subwoofers
-
If we have a single source on each side, we will get the familiar comb filtering we will see in the following figures. But in addition, whenever we use sound systems on a stage in the traditional way but with several cabinets per side, the classic "alleys" or cancellations will appear between the cabinets on one side. So if we add this effect to the one caused by the difference in path length from each side of the PA, the result will be very serious. We already know that if we are equidistant from each side we will get good listening with minimal cancellations.
These problems we used to have, and still have in many places, are largely solved by flying the systems, since the human ear has a harder time perceiving vertical differences than horizontal ones.
Line array
So let's see what happens with vertical arrays. Clearly, with this type of array we avoid the attenuation caused by the audience absorbing the higher frequencies. With the loudspeakers placed above the audience, we manage to reach farther.Vertical arrays
The different vertical arrays are as follows:
Long throw: -
This consists of placing the cabinets one on top of the other, with the upper cabinet inverted so that the high-frequency drivers are together, allowing us to throw farther.
Vertical coverage is equivalent to half a cabinet, that is, narrow. On-axis sound pressure is at its maximum, adding 6 dB; the level is uneven, and we will have more level at high and mid frequencies. If we look at the symmetrical placement of the cabinets, we can see that each component is at the same distance as the other cabinet, so on the array's axis we will have a good phase response.
Wide angle:
By placing one cabinet on top of the other and opening the vertical angle between them, we widen the angle and get more coverage. Vertical coverage is quite wide, on-axis sound pressure is normal, and so is the resulting level distribution. The frequency response is quite good.Parallel:
This consists of placing one cabinet on top of another, which gives more sound pressure in the center, although the frequency response is poor. Equalizing it is quite complicated, except in the case of a single source, so this array is not recommended. Remember that if we place one cabinet next to another we will get 6 dB more; adding a third will give us 9dB more, and so on, according to the formula: -
20log of the number of sources, so 20log2 sources=6dB, 20log3=9dB, 20 log4=12dB, …
To avoid too much sound pressure in the lower part of the audience, we will have to attenuate the lower cabinet by approximately 3dB for each tier down. This way, the sound pressure in distant areas can become similar to the sound pressure in nearby areas. This technique is known as “Amplitude Tapering” (gradual decrease in amplitude).
The drawback of this technique is that, by lowering the level of the bottom tier, the interaction area shifts into the coverage area of the lower tier's loudspeaker. As a result, the path difference will damage the frequency response of the new interaction area.
To solve this problem, electronic delay must be added to the lower tier, a technique called “Delay Tapering”.
For example, if we attenuate the lower tier by 3dB and the interaction zone therefore shifts, in that zone we will have a time difference and therefore comb filtering. To avoid this, we will have to adjust that time difference.
