Music and speech are both essential to effective communication in houses of worship. Both have divergent needs with respect to the loudspeaker systems supporting them, and oftentimes a system that supports one requirement is not likely optimal or even suitable to support the other. What some may refer to as stereo in a large room very often is not, and stereo is very difficult to implement in a typical HOW space. Line arrays, if they truly are a real line array, are not necessarily the best answer for all applications. Sorry folks.
Effective communication, both in music and speech, requires a connection to the people presenting that is coordinated and impacting, both visually and aurally. Naturalness contributes to good connection and comprehension. It has been said, “the mind can only absorb as much as the rear can endure” referring to the length of a service. The same is true for the rest of the listener’s environment. We covered these concepts in-depth with the December 2006 issue. Read Speech Intelligibility In Modern Houses of Worship on page 20 to get more information.
We will now take a look at some of the possible loudspeaker arrangements that can be used to satisfy the needs for both speech and music programs. As we do, we will discuss the inevitable compromises that arise when one is attempting to satisfy requirements that have divergent needs. The goal is to help you come away with a general understanding of the needs and possible strategies that you may seek to balance your particular needs and perhaps financial abilities.
In all cases, it is assumed that the systems will be designed to evenly cover the intended seating areas and avoid covering the wall surfaces in a way that causes even greater problems. (see Collateral Coverage Friend or Foe Part-1 & 2 in TFWM September and October 2005)
So, let’s discuss some of the various loudspeaker array topologies, their characteristics and where their strengths and weaknesses are.
THE CENTER POINT SOURCE
A center point source as it implies, is in the center, usually near the front edge of the platform area. It may be comprised of traditional horn loaded, point and shoot, loudspeakers or line array elements, but will be located and arrayed from near the center of the platform; see (Figure-1). The point source will typically have great intelligibility and high localization to the center of the platform.
Since most humans have a hearing system (two ears on the sides of their heads) that localizes better in the horizontal plain rather than the vertical, the stronger perceptual sense, (eyesight) will provide us with an aural perception that the sound is located in the middle of the platform. As a point source will be arrayed from a focal point, it is a great choice to cover those ever-so-popular fan-shaped seating areas.
Their weakness is that they are so coherent that they have a sterile, one-dimensional sound to them. This is fine, actually preferred, for a speech program, but lacks the breadth and width needed for music. For these reasons a point source is preferred for speech, but has fallen out of favor where high impact music is required.
In general terms the Center Point Source approach is one of the more cost effective means of providing sound coverage in a room.
SPLIT SOURCE, L/R SYSTEMS
Often referred to as stereo, and seeing as we have already discussed the issues with calling these systems stereo, we’ll call them L/R systems for the purpose of this discussion. In general, these systems are comprised of loudspeaker systems, either of point and shoot loudspeakers or line arrays, which are placed to the Left and Right of the platform area; see (Figure-1). The intent is to provide an acoustic image that is approximately as wide as the platform. It often does not actually work out that way, as each loudspeaker system ends up basically covering approximately half of the room. So let’s say hypothetically that each covers the majority of the room well, or at least 2/3 of it. By definition, most folks in the room will receive sound from two different sources at different arrival times and levels.
Generally speaking, this will introduce areas of frequency cancellations and reinforcements that will vary as one moves about in the listening area. Given that the two sources are usually fairly far apart, this pattern of cancellations (called comb filtering) is fairly dense, and tends to get integrated back into, what is perceived to be, a reasonable smooth frequency response.
Usually music will not suffer greatly from modestly different time arrivals, due to its melodic nature, and so the feeling of size and width will often be preferred over a point source for music. However, if the time arrivals from the two sources are far enough apart, discrete echoes and intelligibility problems will begin to become apparent, even in music. Making uniform L/R loudspeaker coverage in wide fan-shaped rooms is particularly difficult, requiring careful effort to make coverage for the lateral, opposite side of the room.
Despite the difficulties and deficiencies, L/R systems, fueled by current line array popularity, have gained traction with the music folks who have little regard for impact on speech program. These L/R systems do generally sound better for music and provide the power and width to support high impact music— but are commonly deficient for speech. There are two installations of loudspeakers in this configuration, one each Left and Right, and if the coverage has any hope of being binaural in much of the seating area, the costs involved in implementing a L/R system is approximately double that of a Center Point Source.
L/R PLUS CENTER SYSTEMS (AND CROSS MATRIXING)
This category of loudspeaker systems is a combination, and in some instances an integration of the two previously described systems. While often called LCR systems, it is difficult to come up with a scenario that actually works when a source is fed to all three loudspeaker systems at the same time. Thus, I have chosen to refer to these systems as L/R+C. The L/R+C has a center point source, used for speech and important lead vocals and instruments, with L/R system elements, used for music programs only. Sounds simple, right? Not so in a real application.
If you had a ‘narrow’ versus ‘deep’ room, it is fairly easy to come up with a L/R system that covers most of the room with binaural coverage. The same is true for a center point source; L/R+C made easy.
The trouble arises when the room becomes wider than deep, or worse— fan shaped. In these instances, the L/R elements usually have a tough time covering the space effectively and often do not work in these sorts of spaces. The problem goes back to the ability to cover a room with relatively even coverage from either the Left or Right side alone.
However, like the center point source, each of the L/R loudspeaker systems have to cover the whole room, or as much of it as possible, but at further distances side to side than a center point source, due to their positions off to the sides of the platform.
Take another look at (Figure-1) where you will see the plan view of an L/R+C arrangement. The sound coverage in the seating areas from the Center Point Source is shown in (Figure-2) where we can see that the majority of the main floor coverage is within +/-2dB; good coverage. The coverage of the Right L/R source is shown in (Figure-3) where we see that the coverage to the lateral side of the room is 11-12dB down from the right side.
Could another loudspeaker be added to the Right side to cover that area? Sure it could, but that loudspeaker would also cover the front part of the platform as well, potentially causing feedback, and still not cover the left side of the room adequately.
OK, so this is where we get clever and look at the tools available. Enter Cross Matrix, where we use elements of the Center Point source as a delay element to the Right L/R source. “What?” You say…
Ok, here’s how it works. The loudspeaker in the Center Point source is aimed at the Left side of the room, and on a vector that is approximately in line from the Left seating to the Right L/R source. Whoever said delays have to be front-to-back in a room?
(Figure-4, previous page) shows the result on the coverage when the Center Point source Left coverage component is used as a delay element for the Right L/R source loudspeakers. That’s right, we are back to the +/-2dB coverage across the room. Stick with me here. The L/R loudspeaker systems have their coverage expanded by sending their signals through the lateral elements of the Center Point source loudspeakers through a signal delay; see (Figure-5).
For instance, the Left Outer loudspeaker of the Center point source has Center AND Delayed Right signals processed through it. The reciprocal is true for the Right Outer loudspeaker in the Center Point source; the Center loudspeaker in the Center Point source only. The processing to accomplish this is shown in (Figure-6) where the Center Point source Outer loudspeakers are doing dual duty as delays for the L/R source loudspeakers and un-delayed Center channel. This concept can be implemented with either Point Sources OR Line Arrays for the L/R elements, provided they have adequate horizontal coverage in the first place.
This example is grossly simplified and actual implementations are considerably more involved. The signal processing for one of these systems is considerable and can usually only be accomplished in a processor like a BSS SoundWeb London that has a drag-n-drop fully configurable architecture. Another consideration is the capability of the sound operator, as these systems require that care be taken to make sure that source signals are routed carefully and deliberately to the L/R+C mix busses. The concept is fairly simple, but the implementation is difficult and worthy of a complete article unto itself.
Despite the complexities and burden of implementing such a system, the results can be very impressive, providing the benefits of a discrete L/R+C system at a reduced cost. With regard to overall cost the L/R+C system incorporates both a Center Point Source and a L/R system and the costs are, as you would expect, fairly high. Essentially, there are two systems in the room— one primarily for speech and the other for music.
DISTRIBUTED OR ARCHED ARRAYS
This is basically an adaptation of a Center Point Source where the center source elements are expanded, distributed or exploded from a center point near the platform. They are typically arrayed in an arc across the platform front edge. The number of loudspeakers is determined by the total angle to be covered and the coverage pattern angle of each loudspeaker element.
The rotational center of the arc is often actually outside of the room and it is important to know where that point is, as all of the loudspeaker elements will generally array from that point. In (Figure-7, page 74) we see the arrangement of an Arced Array on our prototypical room. Our center point of the arc is 54’-0” behind the front edge of the platform. It works out that three loudspeaker elements cover the room fairly well, as shown in (Figure-8). The trade off with an arced array is that there are going to be inevitable interference zones in the seating area that will cause the frequency response to comb filter a bit, and the localization will not be as definitive as with a Point Source.
An Arced Array provides a wider overall apparent image to the platform, from the audience perspective, and will give some sense of width and size to the sound for music. The speech is less compromised than in a L/R arrangement, as the time differentials from each of the Arced Array’s individual elements is kept shorter and has less potential for degradation of speech intelligibility. Each element in the array is essentially on a line that focuses back onto the platform area, relative to the audience, so the image is generally maintained as coming from the platform to the audience.
With some coordination with the architect the interference zones may be placed into the aisles, or more practically, the aisles moved to coordinate with the interference zones which can mitigate the worst of the overlap zones; see (Figure-9).
An Arced Array is also economical as we see the room was reasonably covered with only three loudspeaker elements. Arced Arrays may be comprised of point source elements as well as line array elements. The proper use of line array elements may negate the need for delays if the room geometry is such that delays may be called for with traditional approaches.
Another consideration with Arced Arrays will be coordination with other production and architectural elements in the building. Things like projectors, production lighting, air ducts and sight lines to projection screens need to be looked at to ensure effective integration. Yes, the Arced Array is a compromise, but in budget constrained projects, it may be a good way to provide performance that reasonably satisfies the needs of both music and speech.
It is a fact that many HOW projects have budgets that may be inadequate to provide an optimal solution for both music and speech in their initial implementation. One effective strategy is to initially go with an Arced Array, comprised of loudspeaker elements that can later be repurposed into an L/R+C system as the budgets become available. Another is to plan to repurpose the initial system elements into one of the ancillary areas, such as youth or children’s meeting spaces, and then upgrade the main room with a more optimal solution later.
In all cases, make sure not to compromise the main meeting space sound too greatly. The loudspeaker system is the top priority in terms of the single greatest aspect that influences the overall success of a system’s implementation. The room geometry, size, layout and the performance level of the program— for both music AND speech— should drive the loudspeaker choices above all else. Vigorously avoid shortchanging the loudspeaker system in your initial process as it is much easier to upgrade a mixing console, or in-ear monitors, than the loudspeaker system. In general, cut features not quality; it is much better to do less things with excellence than do more things with mediocrity.
There are a many choices with respect to the approach to implementing an effective loudspeaker system in a worship space. The room geometry, size, surface treatments, aesthetic considerations, musical styles, programming emphasis, technical coordination and, not the least, available funds will have to be considered.
Each of the general deployment schemes listed above have their strengths and weaknesses. When you go to another facility and see or hear something you like, try not to buy into the urge of “I gotta have that!” it is highly likely your room is quite different than the one you are listening in.
Another thing to avoid is going to a mega-church and getting a bad case of the wants; it is very likely that they have considerably more funds to work with than your organization. At the core it becomes a matter of first determining the real system requirements, and then seeking an implementation of technology that satisfies your needs.
Resist the urge to bend the requirements around the system you may think you want. Consider that the spoken word part of the service should be very high on your priority list, and that effective interpersonal communications is what a HOW is primarily about.
Lastly, seek guidance from a professional that has a history of fulfilling these diverse requirements and GO SEE THEIR WORK! Listen to both music and speech; a competent designer/contractor will be happy to make this happen for you. It is a common charge to be the best stewards of the resources applied in a HOW setting. Please help us eliminate the sad statistic that a House Of Worship usually goes through three sound systems before they get one that works.