Unique System Approach

Both the size of the room and the desire for high quality stereo imaging of music programming led Breshears to employ his unique matrixed left-center-right (LCR) design, with distributed delay loudspeakers, for the house sound reinforcement system. It’s an approach that’s become quite practical with the advent of advanced digital signal processing systems (more on this later).

Breshears developed this novel method several years ago, and it’s proven quite effective in several previous projects. In fact, a paper detailing the theory and specifics of the approach was presented to the Audio Engineering Society (AES) in 1998.

Four primary sonic goals expected of all systems for applications of this type are realized, including consistent sound level coverage to all seats, consistent tonal response (again, at all seats), discrete sound arrivals and stereo imaging as accurate, in relation to the stage, as possible. In other words, various sources on stage are localized to their physical location.

The first step in the development of this system is specification of a single mono center cluster speaker system, with delay devices. Aiming and location of these loudspeakers is critical, with their selection (in this case from Renkus-Heinz) based upon good pattern control that will help direct sound to the audience area and off reflective surfaces.

Left and right arrays follow, under the same design criteria. With proper loudspeaker components in tandem with DSP, several center channel and opposite channel signal components can be used to supply distributed delay as well.

Sound Technology Consultants also did the video design, and the bottom of two loudspeaker clusters can also be seen in this photo.

For example, the far outside mid/high section of the right side speaker cluster will not only receive the primary right channel sound, but it will also receive a center channel feed that is delayed (with reference to the center cluster) and summed back in with the right channel sound feed. In this way, the right side speaker of the right cluster will act as a delayed side fill speaker for the center cluster. This approach is followed on several of the components of each of the main clusters.

“We’re using distributed delay devices to cover the entire room as sourced from the three primary speaker locations,” Breshears explains. “It happens that from the left cluster, for example, we’re using center channel components to cover the side areas that we don’t want to try to cover directly from the left cluster. So we’re using some of the sidefill loudspeakers from the center cluster and from that left cluster.

“In addition,” he continues, “all of the delay loudspeakers get a discreet delay feed that coincides with their location from the left cluster. The same thing applies for the center and right as well. So each of the delay speakers gets a delayed center, delayed right and delayed left feed. The result is very good distribution of all sonic information, good coverage, in other words, throughout the listening area.”

One of the challenges presented by the design is he challenge it’s not possible to mix and bus sound to all three clusters. However, this is not desired because it would create electronic phasing that would compromise clarity. The goal is to mix carefully, or discreetly route microphones to either left or center or right, but never all three channels.

“This LCR system approach, in implementation, is not unlike three separate single channel systems in principle,” Breshears says. “Individual sub-systems and components are adjusted and equalized as normal. Signal level balancing can be a little tricky. The key to system setup is to put a single input signal into one of the processor input channels, and set delays and levels as if it were a stand-alone system. The procedure is repeated for the other channel systems. Delays are measured for each individual signal chain (along with a precedence effect delay) and implemented for that signal. All in all, the processing is fairly complex and requires careful design and planning. But it it’s done right, the results can be pretty spectacular.”

LCR Arrays

The three primary clusters at Crossroads are made up of Renkus-Heinz (R-H) loudspeakers, assembled into a crafted package called RPA-4. These pre-configured arrays utilize mid-/high-frequency cabinets in the center, flanked above and below with quad 12-in-loaded low-frequency boxes, with R-H loudspeaker processing applied to increase directivity control, in particular toward the lower end of the frequency spectrum.

Specifically, each array includes four R-H CT7M/64 mid/high loudspeaker modules in four-wide configuration. The horns of these loudspeakers are rotated to supply 40 degrees (horizontal) by 60 degrees (vertical) coverage. The wider total horizontal coverage (4 x 40 degrees for 160 degrees total horizontal dispersion) played very well into the premise of each source array covering a good portion of the entire listening area.

CT7M Series loudspeakers, available with four different coverage patterns, incorporate several R-H proprietary technologies that combine to create arrays acting as a single source. CoEntrant topology integrates the acoustic output of a 10-in cone driver and 2-in compression driver into a true point source, while Complex Conic horns supply a more constant beamwidth and directivity, with circular mouths eliminating corners that can create distortion. Finally, TRAP (True Array Principle), another R-H technology, helps place the acoustic centers of each loudspeaker in near-optimal alignment when utilized in arrays.

Two pairs R-H CT9SVB low-frequency modules, each outfitted with four 12-in. cone drivers, sit side-by-side above and below the mid/high modules in each array. The drivers in these cabinets are frequency shaded to maximize directivity control to well below 100 Hz in the vertical plane.

“There is matched directivity at the crossover point between the CT Series mids and highs, because they use the same horn,” Breshears explains. “And then when you cross down to the low-frequency devices, each with their own set of specialized processing that adds tailored directionality, the whole array is then very well defined in terms of directivity, both horizontally and vertically. As an array there is very little deviation in tonality and also in levels in the coverage area of speakers.”