This article originally appeared in the April, 2017 issue of Pro Sound News. Innovations is a monthly column where different pro audio manufacturers are invited to discuss the thought-process behind creating their products of note. Ken Weller is the product marketing manager at Celestion in the UK.
As anyone who has ever been part of a successful company will tell you, discovering or developing a market-leading advantage is a wonderful thing. But in today’s competitive world, the benefit you can derive from innovation can be short-lived. Nothing succeeds like success—but it will also attract competitors who copy your methods (bad) or (much worse) improve them to the point where your original innovation starts to look like old news. To stay on the cutting edge, especially in a technology-driven industry like ours, companies have to keep moving, developing and innovating.
It’s no secret that software-aided design has been a major player in that process of innovation over the last two decades, as computing power has grown and become more affordable. At GP Acoustics (Celestion’s parent company and also owner of consumer loudspeaker manufacturer KEF), computer-based Finite Element Analysis techniques and computer modeling have been widely used in product development since the 1990s. However, over the last decade, I think it’s fair to say we’ve created products that could not have enjoyed the success they have without the refinement that computer-aided design has brought them; our ability to rapidly create three-dimensional models that accurately emulate the electrical, magnetic and acoustic behavior of our design concepts has assisted us hugely at the research and development stage. That’s not to say you can do away with good old-fashioned engineering skills, but modeling techniques have been absolutely superb at wringing the maximum efficiency from our ideas and honing them to their bestpossible real-world implementation.
The Celestion Axi2050, launched in 2016, is the first product in what may well become a new range of ‘AxiPeriodic’ compression drivers, and provides a good example of what computers can do for product design when coupled with serious engineering innovation.
If you look at the history of compression driver designs, it was a slow process of incremental improvement throughout the 20th century: You’ve got the initial idea that a driver pushing air out through a flared horn narrower than the driver is more efficient than a cone speaker, and then you’ve got slow progress to improve the narrow high-frequency dispersion of the early systems, and the development and gradual refinement of phase plugs to address this and improve directivity and coherence.
Since the Millennium, a lot of effort in the PA industry has been focused on multiple-entry horns, where two or more drivers feed into a single horn, but all sorts of extremely complicated work then has to be done with waveguides to deal with the phase and resonance problems that can result when the output of the drivers combine in the horn, and of course, the more drivers you have, the more crossovers you need in the design, which increases manufacturing complexity and cost.
A cut-away diagram of the AXI2050 Axiperiodic Compression Driver.
We were conscious that if you’re designing mid- and high-frequency drivers, the crossover usually ends up sitting somewhere in the highly critical vocal frequency range, where our hearing is particularly sensitive to any deficiencies in the design. For the past few years, Celestion has been working on refinements of the compression driver principle from the opposite standpoint—trying to minimize the number of drivers and do away with crossovers, but still cover as much of the mid- and high-frequency audio range as possible while retaining fidelity and vocal intelligibility, and minimizing distortion.
During the same period, Jack Oclee-Brown joined our research department. For his PhD thesis, he published research showing just how the inherent modes of vibration in a compression driver diaphragm are prone to coupling with the acoustic modes in the driver cavity, producing all manner of unwanted resonances in the output—research that was also only possible using computer-based analysis of the underlying physics at work in compression drivers. The research concluded that an annular, or ring-shaped diaphragm, could suppress these resonances if carefully designed. Mark Dodd, our head of Group Research, came up with the idea of a periodically symmetrical, corrugated ring diaphragm, which would combine strength with modal suppression.
All these avenues of research combine in the Axi2050, which employs a single voice coil and the thin, annular, gently curved 175 mm-diameter titanium diaphragm designed by Dodd and Oclee-Brown, together with a multiple-aperture phase plug. After five years of research, the precise nature of the three-dimensional corrugation and its rotational symmetry about the newly named ‘axiperiodic’ diaphragm was finalized using modeling techniques to reduce the modal resonances in the driver to an absolute minimum, and the gentle curvature ensured that the first breakup mode is very high for a five-inch diaphragm.
Furthermore, because the titanium diaphragm is thin and very low in mass, it is sensitive enough to accurately reproduce frequencies up to 20 kHz, but at the same time sufficiently resilient and large in diameter to reproduce frequencies down to 300 Hz, at high SPLs if required. The large diameter also conferred another benefit: The compliance of the driver changes little with large excursions at low frequencies, which keeps low-frequency distortion (to which human hearing is particularly sensitive) at a minimum. Finally, the wideband response meant one driver could cover the entire MF/HF range, doing away with the need for a mid-to-high-frequency crossover—another of our design team’s aims.
Many companies claim that their new products are the result of many years of R&D; in the case of the Axi2050, that’s certainly true. Our teams worked for several years on mathematical tools and modeling software before any hardware was designed at all. But as a result, and thanks to the power of modern computing, we’ve been able to bring the performance of this wideband MF/HF driver to its absolute optimum, and gain a better insight into the physics at work in our products. I think that’s what’s known as a ‘win/win.’
Ken Weller is the product marketing manager at Celestion in the UK. He started playing with loudspeaker drivers at the age of 9, got his first job in a Hi-Fi store at 15 and has yet to escape the world of audio, having worked for both Tannoy and Bowers & Wilkins.