Robin Marshall: A Modicum of Genius Page 3

Atkinson: . . . with a plastic-cone 8" woofer and 1" soft-dome tweeter in a relatively small box.

Marshall: I decided to go in a different way all together. And, you know, I think it was good that Mo gave me the complete freedom to do that. He didn't put any constraints on me doing things. He didn't say, "Hmm, I'm not sure about that." He was prepared to let me just have a completely free hand, finish the design, and then he would go out and sell it. Fortunately it all worked.

It lasted two years. I left Mo to start Epos (though I went back to work for Mo for a week after my own company had been floated; I'd already agreed to do the SIM show in Milan for him in June '83). There was no animosity of any kind in that split. Mo and I were then, and we remain, I hope, good friends.

I wanted to design more than simply mass-market hi-fi product. I wanted to get involved in other aspects of design. I'm very interested in transducers as a whole, particularly in drive-units for musical instrument use. That's an area I'm very interested in. So I set up, or I had my partner set up, the company really to do engineering consultancy. We didn't really set up to make hi-fi loudspeakers. That came later. I did some work for a company on some drive-units for active noise control, as it happens.

Atkinson: This is where you blast a sound source with antiphase noise from a computer, and the computer then continually remodels the noise spectrum to try to reduce the overall level?

Marshall: Yes. It puts considerable restraints on the drive-unit technology because the drive-unit has to have phenomenal bandwidth and phase-response control to work. If the phase response starts to go out the window, then you've no longer got anti-noise.

Atkinson: You did come out with a domestic loudspeaker design fairly quickly, I believe, the Epos ES-20 being launched at the 1984 Heathrow Penta Show. Almost uniquely for a new loudspeaker company, you manufactured your own drive-units, including a metal-dome tweeter.

Marshall: Up till that point, I don't think anybody else other than Celestion used one. And it was the first of the current generation of metal-dome tweeters to make the dome out of aluminum. I remember quite distinctly that somebody from Celestion told me that it would never work because the aluminum would suffer metal fatigue inside 10 seconds, and would fall to pieces. I always find it slightly amusing that not very long afterwards, Celestion made an aluminum-dome tweeter—I went and told them the same thing.

Atkinson: I assume that maybe they'd had some disastrous early experiments.

Marshall: I think it was more because when they launched the SL6, they didn't know whether metal domes were going to be commercially acceptable. The tooling costs to make an aluminum-dome tweeter would have been perhaps too high for a speaker that was just going to tread the waters. The SL6 copper-dome tweeter was electro-deposited copper. There were no tooling costs really, you just make some cheap mandrels to do it on; it's a zero-investment technique. But they did start people like me thinking, "Hmm, maybe this is the way to go." Celestion should take credit for that.

Atkinson: You're probably aware that metal-dome tweeters are controversial here in the USA, with many designers feeling that their advantages are only obtained at the expense of problems elsewhere. What are the specific reasons you chose to use them?

Marshall: Let me first of all say that I'm beginning increasingly to think that dome tweeters might not be the best way of doing things. I'm beginning to think more and more that, should we not be looking at cone rather than dome tweeters? But having decided that you're going to make a dome tweeter, you want at least to keep the diaphragm bending modes out of the audio passband. And there is no other material, no usable material, other than a metal, where that is possible. You could probably make a tweeter diaphragm out of ceramic, which would be, perhaps, quite wonderful, but the prospect of making it is daunting (footnote 1).

Footnote 1: I remember Kenwood launching a range of speakers in Japan that used sintered alumina tweeters, but I have no idea what happened to them.—John Atkinson