In the last issue, we reviewed the basic physics concepts behind performance-grade line arrays. As we explore the upcoming crop of new technologies, please keep in mind that the basic concepts – driver spacing, inter-element spacing, coupling criteria etc. – still apply. In this article, we’ll explore how various companies approach working within these criteria to create new levels of performance. In particular, we’ll investigate the EAW Digitally Steerable Array (DSA) series, the Duran Audio Target series, and the Magnetic Audio Devices (MAD) Surface Array technology.

As before, the systems provide bandwidth (within some restrictions) that supports both music and speech. Each also represents years of development, and foreshadow the direction the professional audio industry will take. At this writing, the EAW DSA should start shipping within the month, the Duran Audio Target series should be available by the end of the year, and the MAD systems are available now.

As we discussed earlier, line array theory has been explored for decades, though physical limitations in driver spacing and the resultant interaction within the coverage area limited bandwidth to speech reinforcement. The advent of affordable signal processing in the form of digital processors, however, opened new avenues of performance for small line arrays. Introduced in the mid-nineties, Duran Audio’s Intellivox system is probably the best known initial example of the potential benefits of DSP in line arrays. Designed as a speech bandwidth system, the Intellivox system provides individual DSP processing and amplification for each driver within the columnar array. By manipulating the respective phase and amplitude between respective drivers, the system provided the ability to steer the primary and secondary coverage lobes in the vertical domain, and control of their angle opening and focus distance. This allows the system to be mounted flush to a wall, providing a downward-aimed coverage pattern without physically tilting the enclosure itself.

EAW expanded on this approach with their new DSA series of self-powered systems. By providing individual amplification and on-board DSP, the DSA series is capable of impressive vertical steering of the primary coverage lobe, as well as internal high-pass and low-pass filters, parametric EQ, delay, and limiting functions. EAW employed a modular approach, providing a broad-bandwidth system (DSA250) with a sister low-frequency unit (DSA230, essentially the LF section of a DSA250 in a separate enclosure), packed in an extruded aluminum column. The aluminum case also acts as a heat sink for the sixteen amplifiers in the DSA250 (eight amps in the DSA230). [It should also be noted that JBL used a similar form and function for their EVO system though the EVO is not a line array. The EVO system provides other interesting functionality, but does not address beam steering]. The packaging allows the DSA-series units to be mounted either flush to a wall or flown, with the vertical coverage angle dictated by software programming via EAW’s DSAPilot software.

EAW states that the intent for the DSA250 was to provide a stand-alone system with the ability to provide steerable pattern control down to about 300Hz, suitable for high output, high definition speech and music reinforcement (the DSA230 is used to provide additional low frequency power and pattern-control extension, or as a stand-alone speech-only system). Based on research developed for the very large-scale EAW KF900-series, the DSA uses similar principles to optimize system performance and provide vertical lobe steering. The DSA250 consists of a total of eight 4″ drivers and a tightly -packed array of eight soft-dome tweeters loaded on a common multi-cell horn providing 120-degrees of horizontal coverage. EAW states that the system provides the ability to both shape and steer the primary coverage lobe, varying the vertical coverage pattern anywhere from 15o to 120o,within a frequency range of 200Hz to 15kHz. The tight-pack arrangement of both high frequency and low-frequency drivers minimizes inter-element lobing at the upper-end of their respective frequency ranges.

The DSA250 has a stated output of between 120 and 125dB, depending on the shape of the vertical lobe. As with other arrays, if the total surface area being covered is reduced, the available SPL is higher, and if the coverage area is larger, the available SPL is reduced. Again, the vertical coverage angle and aiming is dictated by the DSAPilot software, which also provides system monitoring and multi-level password protection. It should be noted that there are limits to the amount of vertical beam steering that should be implemented, though a more extreme angle can be obtained if the upper frequency ranges are bandwidth-limited.

As mentioned above, the DSA250 can be supplemented with the DSA230 low-frequency unit if additional low-frequency bandwidth, power, and pattern control are desired. Again, as with other arrays, the longer the array, the more control is applied to low frequencies. For example, a single DSA250 with two DSA230s (effectively three DSA230 arrays) provides frequency response down to about 100Hz(-3dB), and pattern control down to approximately 100-200Hz.

The DSAPilot software provides a number of design and operational features. After entering the coverage area parameters and enclosure mounting height, the program can pre-configure the necessary coverage and steering angles, and upload the parameters via RS485 link (Cobranet is also available as an option). System performance parameters can also be manually configured, and exported to EASE 4.0 for further study. In addition, the program provides the ability to select and highlight individual units via LED indicators on each enclosure, a nice feature when troubleshooting the installation.

The result is an elegant solution for problem environments where pattern control and intelligibility is required. While not a concert-grade line array, the DSA system is an interesting answer, utilizing the latest developments in self-contained systems.

Since Duran Audio was one of the first major manufacturers to utilize integrated DSP and amplification for every driver in a system, it was an obvious extension to develop their Target series of concert-grade line arrays. Rather than being restricted to voice-band reinforcement, the Target series is essentially a full-size line array with onboard DSP and amplification. Consisting of the T-2820 (dual 1.4″-exit neodymium drivers on individual constant-directivity horns and a horn-loaded pair of 10″ cone transducers) mid-high unit and the B-215 (two direct-radiating 15’s) low-frequency unit, the array intended to be primarily configured as a straight-hung array with the coverage lobe shaped and steered to cover the designated seating area. The system utilizes Duran’s Digital Directivity Synthesis (DDS) processing and Digital Directivity Analysis (DDA) software to create the coverage pattern required for a given venue. The software allows the coverage pattern to be asymmetrical, providing a varying coverage pattern provided that the array configuration (horizontal and vertical, curved or straight) ands length are appropriate to the task.

The DDA software automates the system implementation process to a great degree. Room specifications (dimensions and reverberation data) are entered, along with the desired speaker mounting location. The software then calculates the phase and amplitude settings required for each driver in the array, and uploads them to the system. Duran Audio states that the resulting waveform optimizes both near and far-field coverage, and can cover complex asymmetrically shaped listening areas, while minimizing energy on undesirable areas, such as walls and balcony fascia.

The on-board amplification consists of hot-swappable four-channel (mid-high) and two-channel (woofer) units. In addition to amplitude and delay control, the on-board DSP also provides equalization and compression/limiting functions, as well as amplifier monitoring via an RS485 link. Please note that while the DSP functions allow manipulation of the coverage area, the array elements themselves are still governed by standard physics coupling criteria, as we explored in Part 1. (July/August ’03 issue -KRC)

The Magnetic Audio Devices (MAD) Surface Array technology is a different animal altogether. The heart of this technology lies in the application and marriage of line physics with planar magnetic transducers. For those unfamiliar with the terminology, planar magnetic transducers are essentially large-format ribbon drivers, which traditionally have been precluded from professional audio applications because of inherent limitations in power handling, efficiency, and repeatable manufacturing procedures. However, the inherent advantages of the technology led HPV Technologies of Costa Mesa, CA to pursue methods of overcoming these obstacles.

Marketed as MAD Surface Array systems, the planar magnetic drivers used in these systems were specifically engineered to address the traditional limitations. The result is a very lightweight transducer with excellent transient and phase response, which, when arranged in an appropriate array, yields impressive power handling, efficiency, and frequency response. Since these transducers are by nature full-range, a number of other benefits are obtained. First, the only crossover in the system is for optional subwoofers. Therefore, traditional issues of crossover-related phase anomalies and intelligibility problems throughout the vocal range are eliminated. This also means that the system requires far fewer amplifiers than a large three-way line array. Next, the transducers replace all the horns, compression drivers, and low frequency cone transducers (except for subwoofers, if desired), which eliminates all the traditional issues of throat distortion, compression-driver distortion, and power compression. In addition, since the array starts as a dipole, no cabinets are needed. The result is a very lightweight, full-bandwidth system with excellent intelligibility. Frequency response for the basic array is approximately 80Hz to 20kHz, and can achieve up to approximately 145dB (1W/1m).

Beyond essential system performance, an intriguing feature of the MAD array is its performance as an array. Since the transducers can be arranged to adhere to the surface area coupling criteria (see previous article for further details), the array acts as a single device. In addition, and this is a key point, the array acts to provide pattern control behavior in both the horizontal and vertical dimensions. This means that the primary coverage lobe can be designed to keep the energy only on the seats, and off of the sidewalls, stage, and ceiling. As Colonel Klink would say, “Very interesting”.

The array is available as a standard 90o horizontal “rib”, but can be ordered to fit specific custom applications. The number of ribs required is defined by both bandwidth and power output requirements, since the total surface area of the array dictates overall system performance. Minimum requirements are typically eight ribs per side to achieve concert-level performance. However, MAD also offers a portable self-contained (self-powered/processed) system called the MuzikBox that consists of four rows of six transducers with a series of four 12″ cone transducers providing low-frequency extension. The MuzikBox is designed to provide 60o of horizontal coverage, and at the time of this writing, a pair is being used to provide reinforcement for a large jazz festival in New York’s Lincoln Center.

In each of the MAD systems, the rows of transducers can be articulated to change the vertical coverage angle. The system was designed to quickly rig and fly, and so is appropriate for both fixed and touring systems. The first installation of the technology took place late last year in Sacramento’s ARCO Arena for the NBA season. ARCO Arena has been notorious for the high levels of crowd noise, and reports are that the four new arrays that were installed are quite capable of overcoming the highest ambient levels. Side-by-side comparisons with other standard concert-grade line array systems have been conducted throughout the country, generating a considerable amount of discussion in the pro-audio industry.

An interesting side note on the MAD systems is system pricing. While the initial array costs are apparently high, the overall cost for a complete system are competitive. First, less cable and fewer amplifiers are required, since the transducers are full-range. Next, because the arrays are so light, addition savings can be realized in infrastructure. For example, because each of the four arrays weighed less than 350 pounds, ARCO Arena saved approximately $70,000 by being able to delete the steel support sub structure that would have been required by any other system.

The MAD arrays represent a paradigm shift in array technology, and the Duran and EAW systems represent milestones in loudspeaker development in their own arenas. Because the MAD system uses specifically engineered drivers that can be manufactured consistently, though, other manufacturers may find it difficult to generate a competitive system. The DSP-based answer is more easily appropriated, so expect to see versions of large-format DSP-based self-powered systems like the Duran Target system from each of the major manufacturers in the near future. As other manufacturers head in this direction, the apparent advantages (and disadvantages) of the various line array elements will remain a part of the equation, so again, pay attention to the basics, and don’t get sidetracked by sexy software technology. DSP-based systems are a cool tool and have definite performance advantages in many applications, but the bottom line is that any properly designed system should sound as consistent as possible from seat to seat. It still depends on both the array element design and system designer’s skills, and how physics are utilized to accomplish specific results.