Scientific protocol calls for lots of testing before a product is finally assembled. However, in the highly subjective and opinionated realm of loudspeakers, testing has become not so much an afterthought as an after-the-fact proposition in some instances. The transducer and cabinet assembly of each component in a line array are subjected to plenty of testing as each element is developed and connected, but the dynamic and incrementally minute nature of the line array itself tends to get its final checks in situ, flying above the crowd.
Routine Systems Testing
JBL has historically been a major proponent of systems testing — during the development of the VerTec line array system at the turn-of-the-century rigs were routinely set up in the parking lot and on the roof to check ground planes and other splays, and in one instance, an entire system was set up in an airplane hangar at Van Nuys airport near the Harman Industries’ headquarters. “This enabled JBL engineers to validate correlations of predictive software tools with actual measurements of large-scale arrays, which is especially important with line-array type system formats,” recalls David Scheirman, JBL’s vice president of tour sound.
“The issue with very large sound systems is that the bigger they get, the farther away you have to measure them from and it can reach the point where it becomes less and less practical to test physically,” comments Paul Bauman, JBL’s director of tour sound engineering. It’s a pragmatic and logistical problem, but one rooted in ideology as much as science. “Back in the early days of EASE there was a debate over angular resolution,” says Bauman. “The analogy to that today is whether you can accurately extrapolate far-field data from near field measurements.”
A Small and Interesting Niche
This situation, along with the proliferation of line arrays being used in applications ranging from massive arenas to small clubs, has created a small and interesting niche, one that more and more systems designers and component manufacturers are coming to rely on. In the 1960s and ‘70s, Ron Sauro used to work at NASA, measuring trajectories that would help put Voyager, Pioneer and Apollo spacecraft into space. One of the human manifestations of the peace dividend, Sauro now parses data about speakers in his lab in Santa Clarita, Calif., where his company NWAA Labs tries to bring some order to a combination of tricky measurements and marketing-speak. “It used to be that you could measure key aspects, like magnitude and phase, of a speaker in a system fairly easily when most AP systems took the form of clusters,” he says. “Once the line array became popular, it was slowly realized that it couldn’t be measured accurately the same way. With a cluster, you could take all of the far-field information and trace it back to a single point in space. But with line arrays, where the point sources are multiple and highly variable, you have to measure more and from farther away to create a meaningful set of data.”
NWAA accomplishes this with its MACH (Multi-Angle Computerized High-Speed) testing system, which uses a proprietary array comprised of 19 matched Earthworks M-30 microphones and a Linear X precision turntable capable of 0.1-degree accuracy. It simultaneously samples impulse responses within an anechoic chamber and uses these to create loudspeaker radiation “balloons” (a graph of a sound radiation pattern) in a fraction of the time used by other testing methods. This data is gathered by a multi-channel version of EASERA and is then converted to EASE V4, EASE V3, CLF 1 and 2 formats for transmission to the client or user. If you were to apply the older methodology of taking 2,500 [impulse response] measurements to a line array, it would take you a week,” he says. Using MACH, he says they can do it in less than an hour, yet still measuring for every possible configuration of LF, MF and HF boxes at any and all possible angles, by rotating the box around the axis of the driver.
Cutting Through the Clutter
While some of his clients are systems designers seeking to cut through the clutter of marketing materials masquerading as spec sheets, the majority are actually the speaker makers themselves, including JBL, Meyer and EAW, seeking to provide their customers a more valid way to compare performance characteristics before a system is hung. What becomes an enormous amount of measurement data is codified into a program called Generic Listening Library (GLL), that seeks to document and archive performance records of all types of speakers.
The message, though, should be less about the data than how it is used, says Pat Brown, president of Synergetic Audio Concepts (Syn-Aud-Con) in Greenville, Ind., and its testing division, Electro-Acoustic Testing Co., one of a handful of independent testing facilities in the sector. “There is good agreement on the required data set for simple loudspeakers with modeling as the stated objective, [but]… it is more important that the data be appropriate for the device being tested,” Brown cautions. “Understanding loudspeaker data is not trivial, and comparing different loudspeakers with a few ‘one number’ metrics such as sensitivity or power handling can be completely meaningless. Unfortunately, this is often how the buying decision is made, and this fact can affect how loudspeaker specifications are determined and published. Collecting the data is a much easier task than conveying to the end user what it means and how to use it in a meaningful way.”
Testing at this level is likely to remain a niche, albeit an intense one. “There’s something to be said for having an independent, impartial laboratory do these kinds of measurements,” says Bauman. “It’s reassuring to clients like sound designers who deal with issues like allocating the resources and time needed to do it right, which is often not practical even for large manufacturers anymore.”
Dan Daley can be reached at ddaley@fohonline.com
