Tim de Paravicini Page 3
De Paravicini: It sounded tolerably acceptable; it had a pleasing quality. It was smooth and easy to listen to, but it measured horribly. Any reviewer that measured it would have shot it down in flames. It needed to be polished up. And so, within the confines of the mechanical layout that they had chosen I decided to overhaul the whole thing, while keeping the overall sound quality similar to what they wanted. I wasn't going to try and sell them my viewpoint on the sound quality.
Stereophile: Which was what?
De Paravicini: Well, my viewpoint is that tubes and transistors should sound alike.
Stereophile: You mean tubes and transistors should be capable of equal accuracy.
De Paravicini: In essence, yes.
Stereophile: All of your work seems to have been done with tubes. Is that mainly because the people you became associated with were using vacuum tubes, or did you choose vacuum tubes because you felt they might have had more promise?
De Paravicini: I felt basically that tubed power amplifiers just have fewer problems to be overcome.
Stereophile: What are the problems with transistors?
De Paravicini: Well, feedback is one. Some of it is always necessary to get distortion down to a reasonable level, but transistors take less well to it. Transistors are deficient in speed, overload capability, ruggedness and so on.
Stereophile: You feel that tubes are faster than transistors?
De Paravicini: Well, they are developing transistors that are the equal of a tube. But these are still very costly.
Stereophile: Could you comment on T.I.M. problems in tubes vs. transistors?
De Paravicini: In Japan I was doing a lot of research on the problems of slewing distortion in amplifiers. I was able to detect the fact that the amplifiers, when fed with music which contained a lot of transients, were being overloaded by them. That's transient intermodulation distortion or T.I.M.
Stereophile: Why does this happen?
De Paravicini: Because there is a slight time delay through an amplifier. The feedback that should prevent the overload doesn't get back to the front of the amplifier until after the transient has already overloaded it.
Stereophile: So it's a tube's superior overload characteristics that make it less susceptible to T.I.M.?
De Paravicini: That and its higher speed. The second reason I felt tubes would be easier to work with was that transistors had to be run in class-B or class-AB in order to keep them from overheating. Transistors did not like heat, and consumers didn't want large heatsinks. This. meant lots of feedback, to try and cover up the crossover distortion at the point where operation switched from one output transistor to the other. This was not a problem with tubes, because for home audio they had traditionally been run class-A.
Stereophile: Getting back to transformers. What's so unusual about your designs?
De Paravicini: The transformers are conventional inasmuch as I use copper wire in the winding.
Stereophile: What about winding techniques?
De Paravicini: I prefer to keep that proprietary.
Stereophile: How do you address problems such as phase and hysteresis distortion?
De Paravicini: Phase characteristics are basically a matter of how you wind the coil structure, which is what I don't wish to discuss. The hysteresis is a function of how much magnetizable material there is in the core, so the larger you make that, the less significant hysteresis will be. And the lower the frequency, the more significant hysteresis becomes. In other words, the heavier the output transformer, the better it is likely to be. As there is no substitute for [cubic inches] in a car, there's no substitute for iron in an audio transformer. But the expense goes up very markedly, so there is a practical limit.
Stereophile: Well, we're sort of getting up to the present. Let's go back to the TVA-1. Was Michaelson & Austin's TVA-10 as successful as the TVA-1 in terms of sales?
De Paravicini: In many ways it was. The point is that in Europe people didn't need the higher power, so the 40Wpc TVA-10 was more than sufficient. But for those who wanted a no-holds-barred design, the next step was to be the M-200. M & A wanted to have a nice, big, flashy, impressive-looking amplifier and that was the M-200.
Stereophile: What's the rated power?
De Paravicini: Two hundred watts. That's 200 watts right across the audio band. It embodies topology similar to the 10 but with a lot more output and a bigger and better output transformer.
Stereophile: This was designed in what year?
De Paravicini: Around 1977 or '78.
Stereophile: Okay. So now M&A is out of business? Then who was showing the amplifier at the 1984 Winter CES? I saw it there.
De Paravicini: It was Audio Access out of New Jersey. A couple of Chinese fellows, one of them is named Andy Liu. They apparently had gotten an import arrangement with M&A. In the meantime, M&A filed for bankruptcy. At least that's what I hear. I haven't seen word of it published, though.
Stereophile: I see, this is an unconfirmed rumor then.
De Paravicini: Well, yes, but it's come from at least half a dozen different sources.
Stereophile: Now we're in 1978? What next?
De Paravicini: I had already designed the prototype of the EAR 509 while I was consulting for M&A. And when I decided to quit M&A, I was reluctant to part with the design.
De Paravicini: Because I had done some unique things in it. And because I had for many years promoted the PL509 as an output tube, and I did not wish to give the idea to someone else.
Stereophile: Well, the 509 in some ways does sound like no other amplifier I've listened to. Tremendous depth.
De Paravicini: Well, as I say, I felt that I had an amplifier which met practically all my criteria. I'm not saying it's the ultimate amplifier, but I admit that it is very good. I felt I had something that was good, compact, efficient. I had a design that had all the qualities of a solid-state amplifier, the ideal bases, clean lower and mid range, very tight, with lots of current capability.