Nelson Pass: Simple Sounds Better
Thomas J. Norton: You began your audio career at ESS...
Nelson Pass: I was at ESS from 1972 through 1973. Actually I already had an audio career when I showed up at ESS. I had a small company called PMA—which undoubtedly no one will have heard of—making loudspeakers at the time that I ran into ESS. And prior to that I had a partnership which constructed the Claw, a bass horn 9' deep and 7' by 7' at the mouth. It had a 50Hz taper on it, and was an astonishing piece of equipment—the police showed up every time we fired it up!
Norton: So you actually did a lot of your early work in loudspeaker design. Were you responsible for ESS's loudspeakers?
Pass: Yes...I had essentially nothing to do with ESS's electronics, which were designed primarily by Peter Werback, who taught me quite a bit about amplification (footnote 1). He didn't sit me down and tutor me; he handed me the National Semiconductor application books. Back then National Semiconductor was marketing op-amps and they had some rather large catalogs. For each op-amp they gave you the internal schematic (or something that resembled the internal schematic), and then they would give you pages of applications: how to make an AC to DC convertor with it, how to make phono stages, anything you wanted. I began becoming a topologist in electronics from studying all the internal schematics of the op-amps they were using. There was some incredible design talent going into those products; just about every trick in the book that you might ever want to learn could be found somewhere in the internal workings of some op-amp chip.
Norton: You got involved in designing a loudspeaker at least once at Threshold—a very unusual design that made the cover of an early '80s Stereophile [Vol.6 No.1].
Pass: [laughs] The "Ion Cloud" loudspeaker was another one of those ideas that had kicked around, and the day came when I decided that there was no reason not to build it. As you can see from the photos on the cover of Stereophile and elsewhere, it looked like a large barbecue grill. It had no moving parts and its construction was a grid of wires. We applied some very high voltages to what were, essentially, the tungsten filaments used in photocopy machines. It was really a push-pull electrostat that used ionized gas instead of a diaphragm. And it worked reasonably well. It took several kilowatts to get any sound out of it, but it was the most physically and sonically transparent loudspeaker I've ever run across. That is to say, you could see right through it, and it sounded like it wasn't there. It was quite remarkable in that regard.
We were feeding it with an Ampex tape deck running into our Threshold Model 4000 power amplifier, then driving the loudspeaker through a step-up transformer. It drew so much power at the display at CES that every time there was a loud passage or a transient, the AC line would drop. The tension arms on the tape deck would go slack, the sound would stop, the power would go back up, and it would start again; then the power would go, and it went into an oscillatory loop which included every element of the chain, including the AC line and the tensioning arms on the tape deck. We had a lot of fun doing that—a good demonstration of how much power it required. Fabulous device, but it put out ozone, and after some extensive exposure to the ozone I found myself lacking oxygen in my bloodstream...It was a year before I could go near a copy machine.
Norton: What led up to the formation of Threshold?
Pass: Threshold incorporated on December 5, 1974. Both René [Besne, until recently Threshold's vice president of marketing] and I had been employed at ESS; at the same time I was going to UC Davis taking a degree in physics. Operating in the ambience of ESS, we quickly came to the conclusion that if these guys could be successful, anyone could. René and I got along very well. We were very complementary types. We were both audiophiles—I worked in circuits, he worked in aesthetics and cosmetics and such. So we decided in late 1973 to start a company. Initially we didn't really have a good handle on what it was we wanted to do. We certainly had some expertise in loudspeakers, but we wanted to put a little distance between ourselves and ESS. In addition, doing electronics puts you a bump up in a field where hundreds of people are building loudspeakers in garages and such. I think at the time there were some 180 loudspeaker companies, but there were much fewer in the way of electronics. So that gave us a little clearer field.
The 800A amplifier was our first product. It was in development for about a year prior to its release, which was at the beginning of 1975. Actually it arose out of a trip that we made in René's bus down to Moaning Caves—
Norton: Moaning what?
Pass: Moaning Caves. It's south of Sacramento, going up into the foothills—a tourist attraction. It was a three-hour drive, and I was in the back of his bus with nothing to do. As is very often the case, creativity spins out of boredom. It was pretty clear from what was going on in the marketplace that amplifiers using class-A were starting to enjoy some attention. Some of the limitations in the products that were out there at the time were becoming recognized. The solid-state amplifiers of that time were poorly biased class-AB or even -B devices. And it was fairly clear that their sound suffered some serious limitations.
Norton: That's what caused the rejuvenation of some interest in tubes in the early to mid '70s; the early solid-state amps weren't that good.
Pass: I think it sustained the interest in tubes at that time. The amplifier I thought of as a reference subsequent to that was the Audio Research D-150—which I still wish I had a sample of. In any case, it was pretty clear that class-A was definitely the way to go in amplification, but it was also, from a physical standpoint, a very difficult proposition. People would want class-A performance but at the same time—this was in the heyday of Phase Linear—they wanted a lot of power.
Just sitting in the back of the bus, the brainstorm I had was of simply being able to dynamically vary the bias. Having decided that a variable bias scheme would be appropriate to tackling that problem, it became quite apparent that there were several ways of doing it. So with René's money we began researching that—building prototypes, listening to them, and playing around with them. By late 1974, essentially a year later, the design was done. That was the 800A.
There were a number of other things about the amplifier, actually, that were rather unique. It was probably the first triple-series amplifier, certainly the only one I've ever seen. Jim Bongiorno (footnote 2) will undoubtedly call up and dispute this. It was triple-series, triple-parallel output stage, and that was one of the methods that we used to allow it to dissipate so much energy with the semiconductors available at the time.
When we were working with this first prototype, Phil Coelho of ESS—we maintained good relations with ESS—suggested that we go down and see this audio fiend he knew named Joe Sammut. Sammut had Dayton-Wright XG8 electrostatics—very difficult to drive—a stacked pair running in parallel, so it was doubly difficult. The 800A had some really complex circuity to monitor conditions. We had set the current limiting where the 800A would deliver, I think, 30 to 40 amps for very brief periods of time—say, 10 milliseconds or so—and about 20 amps on a continuous basis. From all that we could see that was certainly more than anybody would ever need. So we hooked it up to Sammut's speakers and put on some music. It lasted about 15 seconds, then it shut down. We drove all the way back up to Sacramento, where I opened up the protection circuit, then we drove back down again.
Sammut was a really nice guy about all this. He was always available for us to test with the speakers—he was looking for something that would drive them. He had a bunch of old Mac tube amps at the time that were doing a pretty credible job, but he was looking for better. The second time around we probably lasted about a whole minute in this system. Finally, the third time around, I just simply clipped out the circuit that did the limiting, and it drove the speakers very well. Sammut was very pleased; very shortly after that he became the third partner of Threshold (footnote 3).
Norton: You stayed with the variable-bias design right up to the point where the Stasis products were introduced?
Pass: Yes, we stopped using the sliding bias—actually, dynamic bias is the phrase we've always used—until the Stasis amplifiers were introduced. That would have been, I think, late 1978.
Norton: I've noticed recently that you have upgrades available for most of your early amplifiers. What do these upgrades consist of?
Footnote 1: Peter Werback went on to found Linear Power, the outfit that makes auto sound amplifiers. He died flying his plane in 1984, up in the area where I live—got caught in the weather.—Nelson Pass
Footnote 2: Founder of GAS (now defunct) and Sumo (with whom he is no longer associated). He has an impressive track record of his own for innovative design, and is not known to be shy about defending his credits.—Thomas J. Norton
Footnote 3: In 1991 Joe Sammut was Chief of Operations for Krell Digital Inc.—John Atkinson