Your browser is out-of-date!

Update your browser to view this website correctly. Update my browser now


Real Coverage, Part 1

How to Design Better Sounding Business Music Systems. The first in a two-part series focusing on designing business sound systems that meet the needs of an increasingly audio-savvy public

BUSINESSES ARE RECOGNIZING MORE and more that top-quality sound is an integral part of a customer’s experience. Studies have shown that sound quality affects how customers feel about a company’s products and services. It even affects customer perception of food quality and staff friendliness in a restaurant.

Part of this trend stems from the fact that customers are more sophisticated about music quality than in the past. They have CD or DVD players at home and are accustomed to hearing good sound quality in their living rooms, in their cars and at the movie theater, making the demands on business venues higher than ever. Therefore, AM-radio-quality music is a business sound system blunder — unless you sell antiques, perhaps.

How do you put together a top-quality business music system? It may seem easier than it really is. We can’t design systems the way we used to, using obsolete rules of thumb or just substituting a louder, low-quality sound for a quieter one and calling it a job well done.


We have to approach each design analytically. The first step is to determine what the customer wants and expects. The questions you ask up front help you decide what kind of sound system to propose. See the sidebar “What Does the Customer Want?” for some of the questions you should ask before designing the system.

As we start thinking about the design, we need to translate our client’s requirements into terms of coverage, adequate sound levels and bandwidth. Once these requirements are identified, we can start thinking about loudspeaker and component selection, speaker layout patterns and speaker density.

Unless it’s a simple “point-and-shoot” self-contained system (which is often a good choice), we need to be able to correctly commission the system after it’s installed, ensuring proper settings and optimal performance.

This article, continued in a later issue of S&VC, will touch on concepts I wrote about in the JBL Ceiling Speaker Technical Application Guide, as well as a number of new concepts I’ve been thinking about since finishing the book. The intent is to assist you in designing and setting up superior business music sound systems whether you are a new systems contractor or an experienced one. To go into more depth about these topics, download the guide from


We need to look at the functionality of a system to ensure that it has all the control functions and features required: Does it have the performance features to make the system sound really good? Does it sound equally good when the music level goes down as when the music level goes up? Are there features to keep the music at a pleasant operating volume? Is it easy to add a subwoofer either now or later for additional bandwidth? The system also needs to be easy to install and set up, and, possibly most important, it needs to be difficult to accidentally mess up.

Moving on to power amplification, we won’t know about the amplifier power requirement until we do the speaker design; but we can identify our preferred performance standard. In order to get excellent sound quality, you must use a good power amplifier. These days, it’s becoming less acceptable to use the “commodity-grade” amplifiers of yesteryear. We’re talking about the weakest-link phenomenon. If your amp can’t hold up its part of the sound quality requirements, then performance and quality in the rest of the system will be reduced.

Accessories. Before we start talking about loudspeakers, let’s touch on volume control accessories. I’m not a fan of speaker-level volume controls. It’s much better, when possible, to control the volume at line level, before the amplifier. Speaker level controls typically use tapped auto-transformers (sometime called autoformers). Autoformers can do some strange things to audio quality. They can cause severe distortion. They can change the frequency response as they’re turned up or down. When they saturate (which happens when more low frequency voltage is sent to them than they handle), they tend to really saturate. Even though you might not ever hear the low-frequency content coming out of the speakers, the saturation it causes can trigger the current limiting circuitry on the power amplifier, which in turn causes distortion throughout the audio spectrum.

Your best option when using autoformers is to high-pass the system, which then at least partially negates your objective of high quality sound. In addition, it is difficult to find autoformer controls that handle as much power as a typical zone in a high quality system requires. If at all possible, use a controller that allows you to adjust the volume at line level, eliminating all these problems.

Now, let’s talk about loudspeakers!


Have you ever heard, or followed, rules of thumb like “space ceiling speakers as far apart as the ceiling is high” or “as far apart as the distance from the listener’s ear to the ceiling?” So have I. But as I researched more about how well these “rules” worked, I realized that while they were simple to follow, they weren’t very practical. The conditions upon which these rules were supposedly based — the needs of commodity-grade systems — just aren’t what’s required today. Again, no hard-and-fast rule is going to apply in all cases. You need to find out what the customer needs before you start making plans around speakers.

The main objectives in deciding about the placement pattern and density of loudspeakers in a distributed system are covering the area effectively, providing sound that is audible and intelligible over the entire area, and making sure the system is capable of sustaining whatever sound pressure level the application requires.

A Common Misunderstanding. A misunderstanding about the coverage angle specification of loudspeakers can easily result in system design mistakes. It is very common to see a “polar coverage” spec and assume that the speaker will actually cover this angle. Loudspeakers actually cover less area than their spec sheets would imply. (Let me clarify that the coverage angle is typically the angle at which the sound level is 6 dB down from the on-axis sound level.)

Polar vs. Listening-Plane Coverage. There are two different types of coverage measurements that often get confused for one another. It is standard in the loudspeaker industry to state the coverage in a polar pattern — in a sphere that is 1 meter from the speaker in all directions. The angle where the sound level is down 6 dB from the on-axis level is called the edge of the polar coverage pattern. This is what appears on spec sheets.

It’s a legitimate specification, but it does not represent what the coverage will be over a flat listening plane, as in any room, because it doesn’t take into account the difference in distances that people are from the speaker. For speakers projecting from a ceiling onto a flat listening plane, the sound has to travel farther off-axis (to the sides) than it travels on-axis (directly below the speaker) resulting in a much greater drop-off of sound level off-axis. The result is that the actual coverage angle (at -6 dB) on the listening plane is more narrow than the polar spec. Some ceiling speaker manufacturers use their polar measurement to claim extraordinarily wide coverage. Do not use this specification to lay out coverage patterns of ceiling speakers!

To Illustrate. Imagine a loudspeaker with a 180° polar spec. If you were to incorrectly interpret this as 180° coverage on the listening plane, then one speaker would be all you would ever need for any application. But imagine a single speaker trying to cover an entire department store or restaurant. In fact, you will see that unless a speaker can send more sound to the sides than it does directly on-axis, it never covers more than 120°.

The sound system designer needs to work with the actual coverage over a flat listening plane because that is the plane in which we live, listening at a height of 3 to 6 feet above the floor, depending on how tall we are and whether we’re standing or seated. This is called the listening-plane coverage specification of the speaker. The listening-plane spec represents the reality of the speaker’s coverage for the listeners. Laws of physics dictate that the listening-plane coverage is always more narrow than the polar coverage pattern.

Figure 1 shows a speaker that has a 140° polar coverage (i.e., its 6dB down points). We can see that it would be a mistake to assume that this speaker can cover 140° over the listening plane. In fact, the level at the edges of a 140° pattern is actually more than 15 dB down compared to on-axis — not 6 dB down. It’s interesting to note that the same proportions hold true for any ceiling height: No matter how high the ceiling is, the off-axis distance is even farther away by the same proportion. So for the loudspeaker in this example, whether the ceiling height is 8 feet or 20 feet, the listener who is at the edge of the 140° pattern, who you might think is at the 6dB down point is really 15 dB down.

The actual listening-plane coverage depends on the polar plot of each speaker. On average, the coverage of the listening plane from a speaker with a 140° polar coverage is usually between 90° and 110° (see Figure 2).


How do you convert polar coverage to listening-plane coverage as you design sound systems? There are two ways. One is to use a computer program that does the conversion for you. If you have a copy of EASE, place the speaker in the ceiling, set the listening plane height to the typical application height and see how much area it covers (at the 6dB down point). Usually that’s enough, but you can also do a little bit of math and figure out what the real listening plane coverage angle of the speaker is. I created a simple program, “Distributed System Design,” to do this same conversion for the speakers I sell. It is available at no charge on our Web site, but it only includes our speakers. I think there is a program or two available from Syn-Aud-Con that has a more generic database of speakers.

The second way to compute the listening-plane coverage is to start with the exact polar plot of the speaker and use a conversion table. (Real polar plots directly from test equipment are more accurate than an artists’ redrawings.) Polar plots are usually normalized to the on-axis value, which is usually labeled “0 dB.” For every angle off-axis, there is a “difference-figure” between this normalized on-axis value and the volume at that angle.

To convert to listening-plane coverage, add the 3dB Correction Factor figure from Table 1 for that angle off-axis to the figure from the polar plot. If you’re doing this correctly, the coverage pattern is getting more narrow than the original polar plot.

By using the actual polar plot of the speaker and applying these correction factors from the chart, the angle that results in a figure of -6 dB is the angle of coverage for the speaker. This angle is the real 6dB-down angle for that loudspeaker when it is projected onto the listening plane. Remember that this coverage angle is valid regardless of the ceiling height.

Example 1. If we look at the polar plot of a hypothetical speaker with 140° coverage, we see that at 70° off-axis (140° total for both sides) the level is down 6 dB compared to the on-axis level. By looking at the polar-to-listening-plane conversion chart, we need to add -9.3 dB to this -6dB figure to find the actual level on the listening plane at this off-axis angle. We find that the level of this 140° speaker (as specified by the polar coverage) is actually -15.3 dB, not -6 dB, down at 70° off-axis. Therefore, listeners located at this off-axis angle will hear sound that is more than 15 dB down from the level they hear when they pass directly underneath the speaker. This is a very large difference.

To find the actual 6dB down point of the speaker for the listening plane, take the actual polar plot of the speaker and at every increment of 5° off-axis, apply the correction factors from the polar-to-listening-plane conversion chart. The 6dB-down angle is that angle at which the new figure reads -6dB (polar dB down plus the additional dB down from the correction factor). While the final resulting angle depends on the actual polar plot of the speaker, it can generally be said that most speakers with a nominal polar coverage of 140° can be expected to reach -6dB between 45° and 55° off-axis, resulting in an actual listening-plane coverage between 90° and 110°.

Example 2. Let’s look at a speaker that has a 180° polar coverage. Let’s further assume that it is a mythical “perfect” speaker where the volume doesn’t go down at all at any angle. Its polar pattern when placed in a ceiling is a perfect half-circle. To find the real 6dB-down point, we apply the correction factors and find that at 60° off axis (120° coverage), the sound is 6 dB down. Therefore, the real listening-plane coverage of a perfect 180° speaker is only 120°. Now, let’s realize that a speaker with a 180° polar coverage “spec” can actually be down 6 dB at full off-axis and still have a spec of 180°. In this case, its coverage is going to be even less than 120°.


A couple of interesting points follow from this discussion. First, let me point out that subwoofers resemble Example 2. They have omnidirectional coverage of 180° (when in the ceiling). And they, too, have listening-plane coverage of only 120°.

Second, while it is not within the scope of this article, you can see that this principle applies to all loudspeakers projected anywhere. The fact that the listening plane in any venue is hardly ever a perfect sphere around the speaker means that you will always project a speaker onto a listening plane; and the coverage is almost always going to be more narrow than what the polar spec would indicate.

A student once mentioned to me that he had done some church designs taking a side view, positioning the speakers, then taking a protractor and scribing on the vertical coverage pattern. When the systems were installed, they had more holes than expected. But, after a class on coverage, he realized that it is a matter of projecting a speaker onto a listening plane so that the actual coverage is going to be narrower than he plotted.


Let me complicate things just a little more with a discussion on broadband vs. single-frequency coverage specifications. But let me assure you that there are some solutions. You may have done all the conversion work above, and it might apply only to one frequency! If the speaker does not have similar coverage at all frequencies, then it covers greater or lesser angles at different frequencies.

Some ceiling speakers specify their coverage only at a particular frequency, at 2 kHz for example, because it used to be believed that 2 kHz was all you needed to be concerned about. But the coverage can be vastly narrower at higher frequencies and broader at lower frequencies. These types of speakers do not provide even coverage throughout the audio spectrum. Every spot within the listening area ends up with a different frequency response and different sound level (see Figure 3). In addition to this making it impossible to set a coverage angle, it also defeats attempts to equalize because whatever spot you choose to set your EQ, it’s different everywhere else.

But I promised you a solution, and here it is: Whenever possible, it’s a good idea to use speakers that do cover similar angles throughout a broad frequency range. This is one advantage of having speakers with very small diameters (so they don’t start beaming until a very high frequency), or having multiway speakers where higher frequencies are reproduced by small drivers (tweeters or compression drivers for high-power speakers). Horns on the high-frequency drivers further help in providing even coverage at all frequencies. See Figure 4 for an example of even coverage across frequencies.


Let’s take one more look at the old rule of thumb saying to space the speakers as far apart as the distance from the listener to the speaker. This rule was based on an assumption that the speaker covered 90°. I hope you now understand that even for speakers where this coverage is true for some frequencies, that the polar spec is only valid at one frequency and that, even at that frequency, the speaker never actually covers a full 90°. In today’s high-quality sound systems, you need to consider real coverage, not polar or assumed coverage.

To be continued…

Rick Kamlet is the director of commercial sound for JBL Professional. His prior roles included senior product manager, director of engineering, director of technical services and national sales manager for a number of professional audio manufacturers.

What Does the Customer Want?

BEFORE YOU START DESIGNING A business sound system, find out:

Fidelity Expectations. Do the clients want a good, basic system; something a little better than average; or maximum bandwidth and maximum fidelity?

Sound-Level Requirements. Will the system be used strictly for background music, or does it really need to rock?

Usage. What kind of music will be played through it? For example, if it’s urban funk, then you’re going to have to think about the bass quality and SPL capability.

Form. Do they want in-ceiling or on-wall speakers?

Coverage Requirements. How even do they want the coverage to be? Is it okay for it to perceptibly vary in volume within the space, or do they want it even throughout? Are there areas that don’t need to be covered at all, or where they might want lower SPL, such as the cash register area in a store?

Low-Frequency Coverage. How even does low-frequency coverage need to be? If you’re using subwoofers, is it okay to use a small number, meaning the sound will be loudest close to the sub(s) and softer elsewhere, or do the subs need to cover evenly?

Paging. Do they need paging? If so, how important is paging intelligibility?

Zones. How many zones need separate volume control, different source control or paging assignment?

Benchmarking. Do they want this system in order to keep up with a key competitor? If so, this can give you a benchmark.

Cost. Does the system they want fit into their cost requirements? If not, which functions can be adjusted to meet the budget?

Collect this information and confirm your understanding with your customer before you start designing. You’ll have a much clearer idea of how to proceed when you know what your customer needs.

Featured Articles