At this year's CES, we had the pleasure of interviewing Andrew Jones, audio engineer and legendary speaker designer who has worked at KEF and Pioneer. Now the Vice President of Engineering at ELAC America, Jones is redefining the home speaker market with the ELAC Uni-Fi UB5 series. We were completely enamored with these bookshelf speakers, which are priced at just $500 for a pair. Patrick Norton chatted with Mr. Jones about the challenges of designing speakers, which we've transcribed below.
Tested: This is kind of a serious geek out treat for me. We're here with Vice President of Engineering at ELAC, Andrew Jones. I've enjoyed your work for a long time. It turns out you're kind of big into maker culture. I asked you when you started designing speakers and you said forever. How young when you started building speakers?
Andrew Jones: I guess it was around twelve, thirteen I got interested in hi-fi and I have a identical twin brother. We'd both got interested in hi-fi together. His interest veered towards electronics and mine to the speakers, but I like to say we started off at birth, because with being identical twins, it's not only that, we're mirror twins. He's left handed and I'm right handed, so we were born in stereo.
That's crazy. He's building amplifiers, you're building speakers. What was the challenge when you first started?
The challenge was understanding, first of all. It's fine to go and buy something but it's knowing how it works, so when you're young you start taking things apart, realize you don't know how to put them back together again, so you're going to have to learn that process. All through school I studied maths, physics, and chemistry. I went to university to do physics with acoustics, because I knew that's what I was going to need for speakers, and then I did a few years' research in both speaker techniques but also anti-noise. You know all the modern day noise-cancelling headphones? I was working on big speakers on ships to cancel the noise from the engines, that kind of thing, but my real interest was hi-fi so I joined KEF.
KEF was, at the time in England, the speaker university, and my mentor, Laurie Fincham, was the technical director there. I learned everything there, and we got to know everybody in the industry that was important and knew things, so you could just ask questions of anybody. If I was stuck on something, I could ask Peter Walker from Quad. He'd just give me a call, "Andrew, I was thinking about what you said the other day," and lay out a beautifully simple explanation. It was a wonderful training that set me up for everything I've done since then.
At this point were you starting to design tweeters and mid-range and woofers? How do you go from physics to physically designing drivers that are functioning in the real world?
KEF always built their own drivers, so I started off in research there developing measurement techniques and that stood me for everything I've done since then, because, yes, you listen to speakers, but you got to tie it into how they measure, and if you don't measure accurately, you're not going to make that connection. That was my first part of my life at KEF, but then I ended up being Chief Engineer. I was in product development, and since KEF have always built their own drivers, I learned along with the engineers in the factory. If I needed to build a new driver, I'd just walk into the factory and pull some cones off the line, or have a new profile tooled up, and so I learned how to do that there.
Then it just turned out that every company I joined after that, when I left KEF I joined Infinity here in the States, and then I joined Pioneer, and then I started up the TAD division for home audio, and always with the capability in-house of designing and building drive units, so I learned a lot that way. Even with ELAC now, ELAC builds all their own drivers in Germany. The ones I'm doing here are not built in Germany, but I design them, and then I sit in the lab at the factory and can make sure it works.
Somebody once said to me like, "Well, you know? It's audio. There should be one design then you should be done with buying speakers." Why do we have everything from giant planar speakers to floor standing speakers with seven sub woofers? Is everybody just trying to figure out a different way to skin the particular audio playback cat?
There are some requirements that might be application specific, or model specific, and so within that space you have to design specifically. For example, I'm doing the Debut series. Those speakers that I developed, I make sure that I planned ahead what products will incorporate drivers in the Debut series, and so make sure that the drivers will work in all the applications I envisage for that series, so I'm not developing a different driver for every single speaker in that series because you can't. If it's a series, it's supposed to sound similar, but just give you a bit more. If you're having to redesign the drivers for every speaker, how are they going to sound the same?
In the Debut series there's a small bookshelf speaker, a larger bookshelf speaker, a floor standing speaker, two sub woofers, and a center channel.
And now an extra bookshelf and an extra floor stander, so you can see it quickly gets insurmountable if you're designing a different driver for literally every speaker.
What's your perfect speaker sound like?
I'd have to say real music, but how do you define what real music sounds like? In the high end audio world the idea was the closest approach to the original sound. You could go to, let's say, a classical concert, hear unamplified instruments in an acoustic venue and go, "Yes. That's what it should sound like."
From time-to-time there's been live versus recorded demos. The problem is most of the music that most people listen to wasn't made like that. It was created in the studio, it was never a live event, it was, let's say, manufactured. We really should say created if we want to be nice.
We were talking about that on Tested a few weeks ago. If you look at Pet Sounds, glorious sonic achievement, but it's like seventeen thousand pieces pulled together into this glorious construction.
Yes, so there's a producer deciding what they want to incorporate. There's recording engineers involved, and at the moment you start making the decisions as to how you capture those instruments, some of which don't have an acoustic counterpart. An electric guitar needs an amplifier. It needs the amplifier that you like the sound of. Even a real instrument, you got to put a microphone on. Which microphone? There are as many choices in microphone as there are in speakers to listen to, and as many different opinions, so those engineers and producers decide what they are creating, so that's what we're trying to recreate. It's not this notion of an original event, it's the notion of what did they capture and want us to hear, and so that's what I'm trying to get close to.
That's very metaphysical. It's very like an ineffable substance. I want to capture the magic.
Yes. That's difficult, because I used to have this cute little slogan. Because I'm a physicist and an engineer I like to say, "It's not metaphysics, it's better physics." Ultimately it is metaphysical in the sense of that capturing the imagination that they put in, and when you're listening, when you know you got the speaker right is when you sit down and go, "I'm enjoying that. What comes next? What's this other music going to sound like?" Rather than go, "Yeah. Okay. What's next?"
That's definitely a Johnny Cash song. Instead of having the hair stand up on the back of your neck and be like completely immersed in it. Do you spend a lot of time, do you change drivers over and over again, or does it pretty much come together on the design process or is it very iterative?
Iterative. It is a iterative process, but not necessarily from listening and then changing your mind. A lot of what I do is simulation based, measurement based, because I've learned through the years what contributes to getting good sound, and so all of the early part of the development is, I shouldn't be admitting this, done without listening. I really shouldn't be admitting that I didn't listen to these really until I got them to the show.
Were you excited when you played them? Because I was excited.
I walked into a room and they went from like ... Was it a quartet? I remember it as being like the strings were really, really perfect, and then you're like, "Oh." Then you walk up and bomb a deadmau5 song in my ears... Like I said before, it was just this fantastic, beautiful, wall of sound, in the Phil Spector sense. Is it exciting the first time you play a pair of speakers?
If it's good, yes.
Have you ever gotten a pair out of the box and been disappointed?
It's never been catastrophic, as I say, because I've learned how to get close to the end result through engineering. It is always exciting really, the first time you listen, and especially if you start to hear that hint of magic, and I think other people hear that also, and even if I am bringing a prototype to the show, it has to be at a stage where I am proud of what it's doing, and know that I can make it even better, and I don't use the cop out of if someone likes it I go, "Oh, not bad for a first attempt, is it?" If someone doesn't like it I go, "I know, it's only a first attempt." That's really cheating.
I do get excited about showing the new product to people, and I get excited about the reaction to it. If I do get that excitement then I know I've done a good job, and that's what I'm trying to do really.
Thank you so much.
Ladies and gentlemen, Andrew Jones. That was so cool.