is mounted above a flat floor is with the top of the enclosure parallel to the floor. In this orientation, the system will provide even SPL over a floorplan that is approximately twice as wide as the mounting height, and five times as long. Tilting the enclosure down by approximately 10- to 15° relative to the slope of the floor will produce a floorplan the same as an EVI-12 or EVI-15. At the standard aiming, the 45° nearfield operational angle defines an ap- proximate offset to the first useable row of one-half the height of the speaker system above the listening plane.
In a typical installation, the top surface of the loud- speaker will point slightly above the head height of the furthest targeted seating or standing area. This will ensure the minimum amount of slap echo from the back wall. In an under-balcony situation, the sharp cutoff above the zero degree axis prevents early ceiling reflections from causing interference patterns in the listening area. Since an EVI-28 has a very smooth and rapid drop-off towards directly below the cabinet, you can actually stand right in front of the speaker (see Figure 12) without ear strain or heavy microphone feedback. The remark- able absence of lobes to the rear allows the system to be mounted directly overhead to target a par- ticular area without disturbing the audience below or behind the cabinet.
For example, FIGURE 11 shows a typical under- balcony application that has a floor with an upward slope of 5°. The speaker is mounted 10 feet above the seated head height, so the horizontal width is fixed at approximately 20 feet. The enclosure is tilted back by 5° to provide a 50-foot throw, with the outskirts of the pattern filling in the rear aisle area with tonally accurate but reduced overall SPL. If the under-balcony seating area is only 35 feet deep, then the enclosure should be tilted down by about 10° relative to the floor in order to prevent excess slap echo and preserve intelligibility.
Figure 12 shows a typical small-room application, perfect for a 20-foot by 30-foot boardroom or meeting hall. In this case, the head height is actually defined by a standing height of approximately 6 feet, so the long-throw axis should be very close to hori- zontal. Then the included 40 degree angle points directly towards the entire listening area, minimizing slap echo while retaining a full width, high intelligi- bility and even SPL throughout the listening area.
Q vs. Intelligibility:
The “Q” of a system is a good measure of the system’s directivity, and in some ways a good mea- sure of whether the system’s in-room response will be consistent across the frequency range. A typi- cal 12-inch two-way system with a 60o x 40o horn will maintain a fairly constant Q from 16-30 or 12-15 dB (normally about 26, or 14dB) from 3,000–20,000 Hz, and a 90 o x 40o a Q of 13-15 (11-12dB). Very-high-directivity horns such as a 40o x 20o will have an average Q in the range of 45 (16.5 dB), and very-low-directivity horns like a 120o x 40o will have a Q of 7-9 (8.5-9.5 dB). It is generally thought that a high-directivity horn will have greater intelligibility across a given pattern area than a lower-Q device, and this is true in most cases. Also true in most cases is that a standard horn pat- tern will not fill a typical floorplan, thus drastically reducing the effectiveness of the higher Q. In direct comparison, the VI horn pattern will fill a majority of floorplans with direct-field sound rather than re- lying on reverberant energy to “fill in the gaps” in overall sound quality and quantity. The Q of the VI systems are very consistent from 1,000–20,000 Hz, with a range of 10-18 (10-12.5 dB) and an aver- age Q of 13.5 (11.5dB) or consistent with the overall directivity of a 90o x 40o system. This might lead you to believe that the VI systems are “low-Q” and inappropriate for highly reverberant rooms, but the plots of the direct-field SPL and its accompanying C50 ratio show a significant advantage to the VI system over a conventional 60o x 40o system. As we all know, high Q only helps when the polar response fits into the listening space as closely as possible. Please keep this in mind when using Q as a bench- mark for system intelligibility. For those unfamiliar with the term C50, it is a measure of intelligibility defined by the acoustic power in the room in the first 50 msec divided by the power from 50 msec to infinity, expressed in dB. The minimum recom- mended level of intelligibility is 0 dB, roughly equiva- lent to 10% Alcons.
A Note on Simulation Software:
Data files for AcoustaCADDTM and EASETM are available from Electro-Voice’s BBS, by special request or on Electro-Voice’s Website (http:// www.electrovoice.com). The figures have all been produced in EASE, but with the current 10° resolu- tion of the software, much of the 2° resolution data we collect has been lost in the required averaging translation to the 10-degree format. The net result
