Once upon a time, manufacturers such as Altec and JBL — along with organizations like SynAudCon — emphasized the proper design/engineering of center loudspeaker clusters, especially for speech P.A. system installations. They explained how the choice of a center cluster versus a left/right system could be thought of as a left vs. right brain hemisphere issue. As P.A. systems are primarily for communicating clear/intelligible speech information, they usually benefit from a coherent single-source center cluster, with less coverage overlap (causing signal delay issues) than L/R arrays.
In many typical, shoebox-shaped house of worship projects, when top professionals were hired to implement a P.A. system, center loudspeaker clusters were expertly designed with long- and short-throw horns and bass boxes, to precisely direct sound — uniformly as possible — across the sanctuary’s seating area, while minimizing energy onto the walls, to reduce the intensity of echoes (that reduce clear/intelligible speech) that commonly occurred. To simplify the installation of these optimized coverage (yet complex) center clusters, manufacturers began offering one-box center cluster solutions with a HF horn that projected a special/tapered beam angle, which along with a woofer, were lower in cost and simple to install.
Beyond the center cluster systems that were standard in P.A./speech systems in the 1970s, coverage was often supplemented with additional small loudspeakers in distant locations — such as under balconies — with the required delay. The use of a center cluster with large HF horns and bass-boxes could both improve intelligible speech in large-reverberant venues and provide natural localization for central sound sources.
Of course, there were countless low-cost P.A. installs with corner-located left/right speakers (mimicking a home stereo system), even when driven by a single mono amplifier, along with many split-mono (aimed in parallel) systems (with their inherent delay issues) with point-source speakers and L/R columns for auditoria/HOW venues with longer or higher- reverb times. Several P.A. engineering books published in the 1980s outlined mathematical models for predicting more intelligible speech in large venues. While very helpful for P.A./speech applications, the early mathematical models and sound measurements did NOT relate so well to the music entertainment industry applications and subjective impressions of music.
As good as those vintage Altec center loudspeaker clusters were for speech, they did NOT sound so good for large music groups and the bigger the band (more complex music/signal), the more distortion and ringing was apparent — even if the modest power handling was not exceeded. The power requirement to reproduce music is 100 times what speech requires. Depending on the type of music program, it’s not unreasonable to expect a music system to be able to generate long-term 110 dB SPLs. With the significant differences between music and speech, it’s important to determine your program requirements before looking at a speaker system.
A center cluster is typically a relatively small and mono point-source — and although great to mimic the sound of a talker — is not good sounding for reinforcing the size/space of a large music group. Community did some impressive work in the 1980s with better-sounding Fiberglass HF horns and the mighty (high-SPL) M4 mid-range horns and drivers. During the dominance of center clusters and large horns, horn design guru Don Keele designed increasingly superior performing HF horns for both JBL and E-V, providing both superior sound quality and excellent constant/uniform directivity; thus raising the state-of-the-art horn performance quality, for applications ranging from cinemas to large arenas and the better HOW sound projects.
Fig. 1 shows part of a flown 360° center cluster that has been working well for 30 years in the weight-limited FaithDome, seating over 10,000 people in the round, at Crenshaw Christian Center, in Los Angeles. The author assisted Michael Garrison and Jeff Long in the selection of cluster components, including six each: Community M4 midrange horns with [JBL 2365] HF horns and dual high-sensitivity 18” woofers [JBL 2240] in custom bass-boxes.
Live music entertainment systems typically have stereo arrays/speakers (I sold home stereos back in the mid-70s), so when I did sound for small touring groups in the late 70s, we set up our speakers like everyone else, near the front corners of the stage (hopefully over the audience heads); such stereo systems provided a big sound field, and was also the most practical solution to a quick set-up for a concert in a different venue each night; bigger rooms got bigger stacks. Live music entertainment systems excite the other hemisphere of our brains, where key factors include high sound levels, excitement/impact, imaging, spaciousness and wide bandwidth — along with ease of set-up.
Sound design issues include: budget, program material, ceiling/proscenium/set height, coverage angle, sound imaging, ease of use, sound level, spacing of arrays (venue size) and any variety of non-sound issues, such as sightlines and rigging weight limitations.
Enter the Trapezoidal Cluster Enclosures
The center clusters of the ‘60s and ‘70s (later known as flying junkyards) were composed of a grid, large HF horns and bass boxes; they were phased out in the 1980s in favor of trapezoidal (trap) cluster enclosures, including a mid-sized horn and at least one woofer in each trap enclosure/box. In 1980, John Meyer unveiled the Meyer Sound UPA-1, a compact, high-output, low-distortion two-way system. The UPA-1 was very different, as it was the first trapezoidal speaker (see Fig. 2). Offered by several manufacturers, most of these “trap boxes” were simple two-way designs, including higher-power drivers and horns that were better-sounding for music; very compact and scalable for cluster use — making them very popular for both the touring bands/companies and less ugly for auditoria/HOW installations.
While JBL created increasingly higher power drivers — due to larger and later vented 4” voice coils — and trapezoidal box speaker systems, EAW also became a force in powerful trap touring boxes, as did E-V and R-H. The live-music industry and HOW contractors snatched up these trap.-box speakers in great numbers! Other new high-tech start-up companies, such as Apogee, Meyer and TAD took more esoteric approaches (with modified drivers and closed-loop electronic processing) to providing better music quality and higher SPLs from their trap boxes.
While these trap boxes sounded much better for music, they were less efficient, so the amplifier manufacturers offered increasingly higher-power amplifiers. But the smaller-sized trap speakers compromised pattern control and did not provide the narrow dispersion needed for large venues. So later, bigger-sized trap speakers (see Fig. 3) with larger — and/or more — drivers, some with bass horns (yet far more compact than vintage bass horns) were offered to meet the narrow dispersion and higher-SPL needs of large venues. Ultra-compact trap boxes were popular for small venues and for down-fill. I helped a few contractors and factories, design compound-trapezoidal center cluster modules with the drivers and horns pitched down more optimally.
Not Your Grandfather’s Center Cluster
While the vast majority modern line arrays are used in stereo configurations for modern music concerts — many are also installed as exploded arrays in sports stadiums. As seen in Fig. 4, line-arrays are sometimes installed in a center cluster configuration (for modern arena projects).
And this year, 2019 may be known as the era that trapezoidal speakers advanced and made a comeback. New trap boxes are scalable and feature smoother inter-box summation for reduced comb-filtering between the modules in a cluster. Yorkville Sound’s new Synergy Array Series (see Fig. 5) offer some important new features, including asymmetrical vertical dispersion — to focus more sound down on the listeners — while minimizing energy onto the ceiling and walls, to reduce the intensity of echoes (that reduce clarity/intimacy). Modern trap boxes can be used in nearly any application: stand alone, or arrayed at Left, Center and/or Right of the stage. Next month, we will highlight some brand new takes on trapezoidal designs.
The Case for the Center Cluster Approach
I should emphasize that the use of a central array/cluster is very application-specific to the event, program material and the size of venue. In terms of program material and usage of a center speaker cluster, I am not referring to most modern live music events — i.e., bars, nightclubs, festivals, pop concerts and rock ‘n’ roll touring. And I’m not talking about a center-fill speaker for Left/Right arrays. I am talking about a center loudspeaker cluster/array that can cover the entire seating area, especially for sources of sound like speech and solo musicians that are central on the platform/stage. Here, the emphasis is on uniform sound coverage of a venue’s seating. I discussed the history and trends of computer modeling of uniform speaker cluster/array sound coverage, in the spring 2019 issues of FOH.
One-night concerts and temporary event setups can’t, and don’t need to, be as optimized for uniform sound coverage and have little to no need for localization of center sound sources such as speech (through a central array/cluster). In any application, the height of a center cluster is critical, as a too-high placement will cause a delay and lack of stage localization. Conversely, a low ceiling is also a constraint that can eliminate a center cluster from consideration. Array/cluster height and under-array lobing will determine feedback stability and how far out into the room you can cover, while still keeping within the ideal -6 dB drop in direct sound level.
As a sound consultant, I am troubled by the recent trend of sound companies installing split mono left and right arrays, aimed in parallel; while such a sound system install may be cost-effective compared to a true LCR system, the lower speech intelligibility from the inherent delay issues between the widely spaced arrays (along with the lack of localization for the center sound source) causes some compromises for speech and solo musicians.
Of the sound consultants I polled, most still see the argument for stereo vs. mono audio system installations as really an issue of: budget, ease of use, a large/stereo sound image and high-impact/sound-level for music vs. better intelligibility and center localization for the spoken word and solo musicians. Live theater systems almost always have both — in the form of LCR (Left-Center-Right) speakers/clusters. Cinemas always have LCR — a center speaker in addition to the Left/Right speakers. And home theater systems are the same, so why do fewer auditoria and HOWs have a center array/cluster now, while this was the norm decades ago? I feel that perhaps the biggest benefit of a center speaker cluster is the reduced cognitive-dissonance, as it relates to poor sound localization. Yet, due to the various factors we mentioned, it’s too bad that applications for center speaker clusters are limited.
Next month, we’ll speak with a number with speaker array/cluster experts/manufacturers to discover some better array design options and the introduction of new loudspeaker modules for left, center and/or right of the stage.
CASE STUDY: A Custom One-Box Array Solution
While trying to keep all of the previous design criteria in mind, I was recently hired to design a live sound system for a relatively contemporary church that seats about 450 people, in a venue that is almost square. The budget was tight, but seemed enough to provide not only left/right clusters to replace the failing line-arrays, but also a center speaker cluster for uniform sound coverage and localization for the central sound sources (speech and lead/solo musicians). There are some venues, with a high ceiling, where a line-array could be also used as a center cluster, the client venue did not have a ceiling nor a budget that was high enough for that option. So, I researched available center cluster solutions that were lower profile. Since then I did find a few new center cluster modules, with asymmetrical vertical and/or smoother inter-box dispersion, recently brought to market by innovative speaker manufacturers such as: d&b, L-Acoustics and Yorkville. But these were beyond the budget of the project and were not brought to market until after the job was completed.
Thus, I turned my attention to finding (three) one-box array solutions for LCR (left-center-right) locations that had sufficient drivers to provide enough SPL to replace the venue’s failing compact line arrays. A few manufacturers now offer one-box line arrays, and I have used some of them with good success on several SPL-limited projects; a few examples can be seen at line-arrays.com/one-box arrays, but none of them had quite the requisite combination of SPL, dispersion and low price for the project.
After some product research, I found that Danley, EAW and KV2 had point-source/one-box array solutions, with enough drivers to provide adequate both SPL and directivity, but these were also out of the modest budget (and had limited LF pattern control). So, after discussions with the client (New Hope Adventist Church, Fulton, MD) and the sound contractor, Frank Ward, we pursued a custom one-box array solution, with dispersion tailored specifically to the client/HOW venue that would also be a great option in other projects.
Tannoy’s modular point-source coaxial speakers could have also done the job, but were well outside of the modest budget. Coaxial loudspeakers enable sound from two (HF and LF) drivers to come from one source. This characteristic allows a wider field of listening to a synchronized summation of loudspeaker drivers than loudspeaker enclosures containing physically separated drivers. As well, the pattern of response from a co-axial is symmetric around the axis of the loudspeaker. In 1945, the Altec 604 was introduced, and it soon became an industry-standard coaxial studio monitor in the U.S. Soon after, Tannoy’s Dual Concentric design began dominating Europe from the 1950s onward. Stage monitors using co-axial drivers are on the rise, and are now made by several leading live-sound loudspeaker brands. We used a premium quality and very-high-level coaxial MF/HF driver to provide a symmetrical and more uniform acoustical dispersion pattern, with far less lobing.
Mid-Bass End-Fire Array
Our one-box full-range end-fire speaker array (see Fig. 6) has more directional mid-bass frequencies than any other loudspeaker array its size, providing higher SPL (LF-Q), with greater projection of direct/intimate sound into large/reverberant meeting spaces. Mid-bass polar plot simulations predict how our unique/tapered three-delay/pole main array design maintains off-axis directivity all the way down to 160 Hz, as compared to typical one-box full-range loudspeakers, that become almost omnidirectional at the lower mid-bass frequencies (causing distant sounding lower frequencies and potential feedback).
Typical brand-name loudspeakers provide only 1/3 of the off-axis attenuation (-6 dB) of our one-box full-range loudspeaker array (-18 dB attenuation of mid-bass frequencies). This is why most loudspeakers do not sound consistent across the full-frequency range and why lower frequencies commonly feed back in many sound reinforcement systems. Of course, the single 15” woofer in most full-range loudspeakers, has very limited maximum sound output level (or SPL), as compared to the five woofers in our array.
Why an End-Fire One-Box Main Array?
- The five woofers provide higher output and can match the output level of a compact line array.
- The end-fire LF array avoids the time smear that can be an issue with gradient-type cardioid LF arrays.
- The end-fire LF array improves low-frequency directivity, providing greater projection into reverberant meeting spaces, along with less feedback.
- The reduction of excessive LF energy onstage makes it easier for musicians to hear the monitor mix.
- The compact main array needs much less vertical space, as compared to a line array.
- The one-box design is pre-arrayed internally, so only one box needs to be rigged
and wired, greatly reducing the time for installation, reducing labor costs.
- The one-box array eliminates inter-box rigging and cabling, reducing component costs.
- Its point-source driver configuration provides symmetrical polar patterns and avoids the time smear that can be an issue with curved line-arrays. See Fig. 7.
- The coaxial/MTM (midbass-tweeter-midbass) speaker driver configuration provides much smoother polar plots as compared to J-shaped line-arrays (see Fig. 8).
David K. Kennedy, a 40-year live-sound veteran and author, operates David Kennedy Associates, consulting on the design of architectural acoustics and live-sound systems, along with contract applications engineering and market research for speaker manufacturers. He has designed hundreds of auditorium sound systems for churches, schools, performing arts centers and AV contractors. Reach him at david@hdsound.us.
