William Zane Johnson Page 2
Messenger: I think there was a crucial element behind the changeover from vacuum tube to solid-state: at last you could get rid of the output transformer, which was a known limiting factor in tube design. With the transistor, you at last have a device that you can directly couple to your loudspeakers, to control them better, if that's what you want to do. Alternatively, by coupling the transistor directly to the speaker, you're giving it a much harder time than ever the valve had with a transformer in the way.
Johnson: There are trade-offs with all technologies. For example, with a typical dynamic loudspeaker that is intended to have an effective short circuit across its terminals for control, a direct-drive, solid-state amplifier will almost certainly provide better results in the bass region/ If you are talking about high currents, then the bipolar amplifier obviously has advantages, but there are a number of loudspeaker technologies where a quality voltage signal is required. Some of the electrostatic speakers will give far better results with a vacuum-tube amplifier than they will a solid-state amp.
The transformer problem is, of course, real, and while it has been a limiting factor in tube amp design, it really doesn't need to be. All of our vacuum-tube power amplifiers use partial cathode coupling in the output circuit and we can't, of course, take any credit for that; it was first used by Quad 30 or more years ago. The benefit of it is that there is some intrinsic cross-coupling in the push-pull output stage, so you can actually derive some benefit from the use of the transformer, as opposed to it being a total liability from the bandwidth standpoint. There actually isn't too much of a problem in designing transformers with a wide bandwidth. The power transformers that we use, for instance, are typically capable of a flat power response from around 15Hz to 40kHz or so.
Atkinson: There certainly is a financial problem . . .
Johnson: They're expensive, surely . . .
Atkinson: . . . which means that you have to operate at the very high end of the market . . .
Johnson: . . . but it has always been true that that quality costs moreeven with musical instruments, you don't buy a Stradivarius for the same price that you could buy an aluminum violin. You could take the position that the way Rolls-Royce builds their automobiles is both old-fashioned and too expensive, but on the other hand I would have to say very candidly that I don't know of a better automobile.
Atkinson: Have you tried designing a transformer-less valve power amplifier, like the Futtermans?
Johnson: No. I haven't had any desire to do that; the vacuum tube is just not a low enough impedance device, even when paralleled . . .
Atkinson: You can't get the current . . .
Johnson: . . . not with any degree of stability or longevity, and we just aren't interested in that.
One of the things we believe should be involved in high-quality, state-of-the-art product is longevity. The product should be serviceable or should be capable of being returned to a serviceable condition as long as someone might care to use it, and that implies not short-circuiting legitimate design. You won't find any direct-drive valve amplifiers from Audio Research.
Atkinson: HFN/RR had a letter of complaint following our SP8 review, the gist of which was that in this day and age we shouldn't' be reviewing "outdated" gear. The irony is that the SP8's very high overload margin on its auxiliary inputs makes it an ideal preamp for Compact Disc players [which had been launched in the UK at the time of this interview], valves or no valves.
Johnson: I don't know whether I consider that a compliment or not! The high output from CD players won't be a problem for any of our preamps because the input level control is at the beginning of the high-level circuitry; they're overload-proof! However, I listened to a Compact Disc player a couple of days ago in Scotland and it was terrible. I heard no music from it at all.
The entire envelope of sound had an electronic hiss, there was no depth. Plenty of width, yes, and plenty of dynamic range, but the sound was even tonally related to music!
Atkinson: Do you think the bad sound was the result of bad recording practice, or was it something more fundamental?
Johnson: I don't know, it was the first time I had heard CD. But I can best describe my reaction as that if I didn't already know what the sound of a violin was like, I wouldn't have recognized it from Compact Disc. Just because I know a given piece of music and know that that has to be a piano, or a violin, isn't enough. I want to actually hear what I should be hearing, not just reconstruct the sound mentally because I happen to know what it was originally like.
Atkinson: Linn Products' Ivor Tiefenbrun described it as the difference between "hearing" and "listening."
Johnson: Good for him. That's well said!
It's clear that some of the digitally mastered LPs are excellent. I have come to the conclusion that the problem with some of them is too high a level of cross-modulation products in the front end of the preamplifier used to play them back. With more linear equipment such as the SP10, for example, I find many of them to be more listenable.
Messenger: Is there any difficulty obtaining tubes these days?
Johnson: While it is true that there are a number of vacuum-tube vendors that no longer manufacture, there are probably nearly as many new manufacturers, particularly behind the Iron Curtain. For example, Siemens offers us a 6DJ8 that they get from Poland, and it's a very fine tube.
Atkinson: Do you still get tubes from the UK, from the M-O Valve Company?
Johnson: Most assuredly; they make the finest tubes for audio money can buy.
Let me make a comment here. It has been widely said that the perfect amplifier is "a straight wire with gain," but it common knowledge amongst a few manufacturers that unfortunately there is no such thing as a "straight" wire. Every kind of wire configuration introduces its own coloration and it is preposterous to claim that an active device is "perfect." Everything is relative, it's a matter of degree.
Messenger: An amplifier manufacturer told me that he knew how to improve his product, but felt that he couldn't because then it would only show up the quality of the ancillaries that are currently available. As a designer purely involved in amplifiers, you must be presented with difficulties in that direction?
Johnson: But Audio Research already has that kind of product, the D79C and the SP10 fit that profile. The other parts of the system must be chosen very carefully, or they will not give you the desired result. Many times you will end up with a more pleasant system using the SP8, which has a lower degree of resolution and definition than the SP10, for just that reason.
It's important that the system be matched. For instance, to have the speaker a more open window that the amplifier is to invite sonic disasteryou just hear its problems all the clearer. As a rule of thumb, it is far better that as you move backward in the chain each item be of a higher quality, ending up with the turntable being the very finest that can be put together.
Atkinson: Which is?
Johnson: I don't think I ought to get into that area, gentlemen.
But I would rather spend, say, $3000 or $4000 for the arm, cartridge, and turntable, whatever it takes for electronics, and end the chain with a small pair of Celestions or Rogers. I'd rather live with those than use a full-range electrostatics that will show me all the flaws in the system. I don't need thatI want to enjoy the music!