Interviews

Then there are the areas of electrostatics which people complain about. One, of course, is dispersion at high frequencies; the ribbon was the answer to that. The other thing is level; none of them go loud enough. In order to get level, you need to allow the diaphragms to move far enough. This is basically a question of having the plates far enough apart and getting them sensitive enough so you can get that level out of the speakers without having too big an amplifier. If you're going to have a big plate with big distances it's not going to be very much of a capacitance so you have to have pretty high bias. You have to insulate the things in novel ways without losing the efficiency, so that they don't arc under those conditions. You have to get the coating on the Mylar to stay there and not migrate all over the place, you have to be able to get exactly the right impedance of coating to suit what you're trying to do.

The next thing is the ability to move, which is how you get the bottom end. On the prototypes in the factory, you can play 20Hz at around 96dB sinewave, and you hear pretty much nothing. You feel it and you feel considerable pressure on your ears. They do go very, very low indeed.

Kessler: Doesn't this require massive bass drivers?

Vereker: No, they are about two foot by one foot, have large excursion, and pretty good sensitivity. The other thing which is different from other electrostatics is that the bass units are actually loaded. The speaker is about ten inches deep, and it has a series of tubes behind the bass panels which load the things and stop the excursion at resonance from going too far. It controls the whole thing, and you get the maximum levels instead of having them canceling.

With Naim 110 amplifiers, peak level at the moment is around 117dB—not at 20Hz, I hasten to add, but over most of the usable band. Sensitivity is about 88dB for 1W, or 2.8 volts, and the impedance is not unreasonable at all. It's about four or five ohms, and it drops a little bit here and there.

Kessler: As these are "dipole hybrids," are they as critical of placement as we expect of conventional electrostatics?

Vereker: You have to have them away from the wall, but since we've only got a couple of pairs, we haven't had the opportunity to go around in all the different rooms to see what happens. They don't seem to be particularly critical. They just need to have air behind them. In one of our listening rooms, which is very, very small, it was quite surprising how nicely they worked.

At the moment, we haven't got the passive crossover, so they're tri-amped using the Naim external crossover. But there will be two versions, and you can order it either way. Inside the bias box which lives inside the speaker, there's space to put an internal passive crossover so you can use it either way. An economy-minded consumer could start out with one small Naim amp and later change over from passive to active crossovers. People who are not into enormous levels will be pleasantly surprised.

Kessler: That sounds like a reaction to the early Quad. Are you still inspired by that design?

Vereker: I think so. I certainly still have a pair and I think that there are at least another two or three pairs among the people at Naim in the UK. Every so often I bring my pair up and stick them into our big demonstration room and we have a listen to remind ourselves of how good these things actually were. Whatever it is—27 years ago or something. It was a super piece of design.

One of the big differences between, say, Quad and ourselves is that Quad set out to make a speaker with all the advantages of electrostatics at a price which people could afford. We've not set out to do that, or in anyway clash with what they've done. We've aimed to make it as good as it can possibly be and to hell with the price. Whatever it costs, it costs. I can't give any idea as to how much it's going to be, but it will be a high-end item.

I look at the sorts of components that we're involved in. The plates, for example—two feet by one foot by one inch—have to be flat to better than 0.7 of a millimeter in that two feet because you want to keep the capacitance of all the units very, very close. That has involved our talking to experts at British Aerospace as to how to get materials, which they've suggested, and it's led us to particular alloys which have to be processed in a special way, with an awful lot of holes in them; when it's been drilled or punched, the whole plate has to have every single sharp edge removed. Precise radiuses on every single hole, inside and outside, otherwise you get all sorts of problems. The insulation then has to be put on. We've got to build clean-air rooms, two of them, one with one sort of chemicals and another without the chemicals because most of the chemicals actually eat bits and pieces until they've dried. The whole thing is an incredible ballgame.

One of the chemicals we asked for, the company said, "Sure we can supply you, but it only comes in 25 kilogram packs. Is that okay for you? Is that enough?" So we said that would be enough. Guy Lamotte, who is the designer of the speaker, came back and said, "Let's put in a regular order. For 10cc a year." Because that's about all we'll need of that particular chemical. In fact, 25 kilograms will keep us going until the end of the century.

Kessler: Did you consider using the technique employed by MartinLogan of curving the diaphragm for better dispersion?

Vereker: No. The dispersion has to do with the actual radiating areas, and how you cross over between one panel and the next, as much as it has to do with shape. The shapes are arranged in such a way that the sound is absolutely seamless, and you can wander around the room anywhere and the orchestra just sits there in the middle; you can move around the instrument. It's quite uncanny.

Kessler: How did you manage to find a ribbon with the same phase characteristics as an electrostatic?

Vereker: Aah, we didn't. We've actually taken other people's bits of ribbons and mounted them in a completely different way and given them a completely different type of phase-correcting throat or loading.

Kessler: As far as interfacing with a variety of amplifiers, we know that Quad's ESL-63, early on, had a problem with amplifiers without protection. When the speaker shut down—

Vereker: —it shorted the power amp. While that's probably okay, it actually did it for an awfully long time, like 200 milliseconds. Now that is a long time in the life of a transistor. It's not so bad because it wouldn't normally occur with modest-sized amplifiers under anything other than fault conditions, so that was probably OK. But the early ones were actually sensitive to arcs outside the speaker, and used to shut things down and could damage things quite badly. They fixed that; it's no longer a problem.

If you drove the original Quads too hard, you'd make a hole in the diaphragm and that would be that. The later Quads have very sophisticated protection systems which not only cut the low-frequency energy fed into the speaker at high levels—as you get louder, you actually get less at the bottom end—but it also shuts down when it senses an ionization inside the speaker. Ours has no protection at all. If you overdrive it, the diaphragm will stick itself to one of the plates, discharge, take a few moments before it charges up, then come back again—with no harm at all.

Kessler: What's the audible effect?

Vereker: It splats, and then silence. I don't know about damage to other people's amplifiers, but it doesn't damage Naim amps. So far as we know, Naim amplifiers didn't suffer from Quad shutdowns either, though I actually thought that it might occur.

Naim amplifiers are protected. First of all, there's the fuse; if any total disaster happens, the fuse will blow. The next thing is the thermal sensor in all the amps except the Nait. If the case exceeds 70°C, it cuts the mains, the whole thing turns off, and you have to wait for it to cool down. Also, the amplifiers have built-in power limitation. They constantly measure the current through the output devices, the voltage across the output devices, and integrate that with respect to time; it's a true power limitation. The 135 also has a fan to help dissipate the heat.

The 250 amplifier has four regulated power supplies, each of which will feed something like 16 or 17 amps continuously at ±40 volts, and give you a peak of around 30 amps for up to 10 microseconds. This would indicate a fault condition because the amplifier's rise time is less than 10 microseconds. Then the power supply will shut down, both sides exactly symmetrically, and it will take a minute or two for the amps to reset. You just turn the mains off.

That was the problem that Quads would cause when they shorted the amplifier out. Quad fixed the problem quite early on with an alternative earthing arrangement. Once that was fixed, there was no further problem.

Kessler: By the way, what are you calling the electrostatic?

Vereker: The FL1—Flat Loudspeaker 1.

Kessler: As far back as three years ago, you were talking about producing a Naim cassette deck. Has the possible demise of analog cassette due to DAT killed your plans?

Vereker: Not yet. The biggest difficulty has been to get Papst to make the transport we want. They've finally done it; it works really nicely, and it's beautifully made; very consistent. Unfortunately, they've been rather slow about it; samples due in January 1986 didn't arrive until September. There have been delays all along the way, so that hasn't enabled us to put it on the market before now. In fact, we would have if they'd actually produced the transport on time. The Papst transports are industrial ones, for hotel systems and that sort of thing, designed for unbreakability, to run for a year without hassle, 24 hours a day. The one we have is a dual-capstan design not used by any hi-fi company at the present time.

Our commercial problem is that we are likely to sell only 500 very expensive cassette recorders unless we can break into the professional market, and Papst insists that we buy 500 of these transports for roughly $l100,000. If DAT is better than the best analog cassette, then I'd have to say we'd end up with having three or four of these Naim analog cassette decks kept in the factory—or in my home—to be used for dubbing master tapes from friendly musicians. That sort of thing may arise.

The indications I have about DAT so far are that, while it may be better than CD in my terms, it doesn't compare with the very best analog.

But beyond cassette or DAT, there are so many other things which are dead certain to occur. There's the smaller dynamic speaker, the NAC52 preamplifier—a new high-end six-input preamp with user-configurable inputs—and a stereo television tuner. We've also got a small FM tuner coming, to match the Nait. This is nearly together, and the cassette deck is virtually finished.

Kessler: Would you consider making a DAT machine?

Vereker: Oh, if somebody would play with us with the transports. But we're not going to get into making helical-scan-type transports.