October 2013

SET And SIT Amplifiers 
Nelson Pass on his exciting new designs.
Article By A. Colin Flood

  The designer of monster power amplifiers with thick front-face plates and a big blue Cyclops eye, Nelson Pass designed a new style of extremely low power, solid-state amplifier using a special kind of transistor. The Pass' SIT amplifiers are the first of a new generation of audio amplifiers using 50 watt Static Induction Transistors (SIT) in a single-stage, single-ended Class A circuit without feedback or degeneration.

Like tubes, SITs have soft overload clipping. When brief bursts of musical energy occur, SITs react with rounded waveform tops instead of sharp and hard clipping of solid-state. SITs have a curve which looks a lot like a triode vacuum tube; low at first and climbs steadily. The distortion curve is similar, a steady rise instead of a valley with high distortion at both ends.

Ten Times The Dynamic Range In The First Watt
It is hard to believe that even a quality one-watt tube amplifier, such as a Single-End Triode (SET), could sound as good as a quality solid-state amplifier on anything except über-efficient big ole horn loudspeakers. Yet Steve Deckert of Decware explains why the first watt matters so much. In "Why The Set & High Efficiency Speaker Approach Works," he says "Aside from the amplifier's superiority by simplicity, there is a more profound reason for using SET amplifiers. The magic predominately lies in the first watt."

By magic, Deckert means the "inner detail and most of the dynamics". He gives an example of a pair of very high efficiency speakers (96dB/W/m) playing with one watt against the average noise floor in a listening room (55dB). The amplifier and speaker combination has a dynamic range of 41dB (96 – 55 = 41dB). Adding a second watt of power increases the dynamic range by 3dB; it sounds about twice as loud. For every additional 3dB increase in loudness, you need to double the amplifier power. "This should clearly illustrate," Dekert says, "that there is over 10 times the dynamic range in the first watt as there is in the second."

Sonic Lessons Of Triodes
"I don't intentionally seek the triode sound" of the first watt, Pass said. "I seek to apply the sonic lessons taught by amplifiers which don't necessarily measure spectacularly well but sound very good. Triodes just happen to fall into that category. I would not see to duplicate triodes per se, except possibly as an exercise."  

This author reviewed the Pass Labs X250 abut 11 years ago. Pass is known as the master designer of the biggest, baddest and possibly best sounding audio amplifiers around! The X250, for example, while it can operate in clean Class A mode, is also a 100-pound monster capable of brief 1800 watt bursts! Yet now Pass seems to be going in another direction. He started with the First Watt papers and product, and now these "less power is more enjoyment" designs. So I asked him, "watt" gives?

"Pass Labs still builds large and powerful amplifiers," he says, "I created First Watt at home with an eye toward making great little amplifiers, taking the opportunity to explore what can be done at low power. I get to do anything I want, release them in limited quantities (or not at all) and I don't have to care about commercial success. Of the things I release I observe the impact on the audio scene and factor that data into my vision of the commercial audio landscape. When I uncover commercial potential I have the option of incorporating it into future Pass Labs product, as with the Xs amplifiers."

Pentode-Like Curve Calls For Feedback
"A SIT is merely a type of device," Nelson Pass said, "You could build an X250 topology with SITs, but it is not likely to play to the strengths of the device. The strength of SITs lays in their remarkable ability to deliver good performance in common-Source mode with a single device and no degeneration or loop feedback.  No other part does this trick as well, but the approach is limited in wattage and some of the measured characteristics."

"The X250 is a design that caters to the strengths of more conventional JFETs and MOSFETs" says Nelson Pass. A junction gate chip, the JFET is the simplest type of field-effect transistor. JFETs and the metal–oxide–semiconductor (MOSFET) transistors are used for amplifying or switching electronic signals. "Unlike the Triode characteristic of the SIT, they have a Pentode-like curve." This usually means more than one stage and a multiplicity of devices and often calls for the use of some form of feedback, but we work hard to keep it simple and very linear, and the result is still pretty sweet sounding but with the desired power and authority."

Dick Olsher writes excellent reviews. He walks through the interior of an electronic component like a tour guide in a museum; Dick Olsher leaves no feature unexplained. He toured the guts of the SIT-1 monoblock for Enjoy the Music.com last year. Dick Olsher concluded the new Pass design is "a sonic home run".

He wrote the SIT-1 sounds "like a superb 300B based SET but with far more transient speed and extension at the frequency extremes. There's plenty of soundstage transparency and dimensionality, as well as superb layering of the depth perspective." The powerful 300B tube is popular with tweaking audiophiles trying to drive average efficiency speakers, such as radiator cones, with tubes.  "The 300B, Wikipedia says, "has good linearity, low noise and good reliability; it is often used in single-ended triode (SET) audio amplifiers of about eight watts output; a push-pull pair can output 20 watts". Reading this, one might conclude that the SIT design is better than tubes, and better than solid-state.

Long-Term Listening Enjoyment Of Single-Ended Circuits
In Dick Olsher's excellent article, he also quoted Nelson Pass at length, about Total Harmonic Distortion (THD) of SIT and tubes, which bears repeating. Pass says a practical reason to explore simple single-ended triode-like (SET) circuits is that they generally create a harmonic distortion spectrum that is predominantly second order in nature with the residual being essentially third order.

If you are going to have distortion, you will want it in a low order harmonic form, kept to only second and third harmonic if possible. Low-order distortion products are consonant with the music and are certainly imperceptible at total levels below about 1%. The SIT-1's THD is specified at 0.7% at 1-watt for both 4 and 8 Ohm loads. Measured distortion does rise to over 2% above 5-watts, though the character of the amp does not appear to change subjectively.

Pass confirms a popular argument amongst tube lovers: "Most SET amplifiers measure at 2% to 3% harmonic distortion at low power levels but the distortion spectrum is characterized by the near absence of higher order distortion." It is that higher order distortion that tube lovers, such as this author, argue "detracts from long-term listening enjoyment". This is the hard edge of solid-state, this author feels, which wears your ears out.

Pass goes on to say that while feedback is useful for the reduction of THD, it does not generally improve the relative distribution of distortion products. This means, he said, "that you can't make a pentode sound like a triode by merely cranking up feedback levels".

This author looks for amplifier output wattage to double as Ohms decrease. To me, this indicates the amplifier will control the woofer, not the other way around. Bob Carver's KT88-based Cherry 180 tube monoblock amplifier, for example, has 200 watts at 8 Ohms, but only 230 watts at 4 Ohms, though it is still 215 watts at 2 Ohms.

"You mention Bob Carver?" Nelson Pass asks. "I hold him in the highest regard, and if I needed a good tube amp, I wouldn't hesitate to buy one from him." Woofer control though, Pass said "is a subjectively misleading topic. The SIT is comparable in woofer control to a typical SET, that is to say not much. If you need to control a woofer, look elsewhere for a higher damping factor". Pass has examples of low power/low damping factor amplifiers, which play more realistically into some woofers than amplifiers with fantastically better specifications. It is, he said, strictly a case of "horses for courses," meaning it is important to choose suitable amplifiers for particular speakers.

Only Five Watts
Depending on the amplifier design, Pass says these single-ended models decline in output with impedance, so his SIT models are nominally only 5 watts into 4 Ohms but decrease to only 2.5 watts into 2 Ohms. "Other designs could differ," Pass said, "and there is no technical reason why they couldn't double as the impedance halves. I don't see this sort of design as playing to the strength of the parts, though, and the market looking for amplifiers like this does not tend to be low impedance."

The Total Harmonic Distortion (THD) of the SIT amplifiers seems high (5%) compared to solid-state amplifiers. Because it is mostly second-order harmonics, high THD is acceptable in tube amplifiers, but is high THD acceptable in a SIT ones too?

"Perhaps to many it is not," Pass said. "I am not here to make everyone happy. I have something with a unique character, and I say "Here, take a listen to this." You might not care for it, but it's different, but if you are interested in sonic qualities, it has something to say to you. The 5% figure is up toward the top of the power scale where the device's character is doing a soft clip, much like a tube. Most of the listening of this audience is down at 1 watt or less, where distortion is more respectable. Of course I have many products where the distortion is dramatically lower. People are welcome to buy those instead, as well as those of my fine competitors."

Dick Olsher said Pass did indeed accomplish the powerful (for a tube) 300B triode sound. What is next design objective for the god of amplifier designs?

Pass says he is currently exploring ways to apply the lessons of the SIT devices to other, more powerful and maybe cheaper amplifiers. The Xs 150 and 300 are the first beneficiaries of some of these techniques. Though the amplifiers "clearly they do not fall under the 'cheaper' category".

You had a wonderful white paper a decade ago describing how and why quality Class A amplifiers had to weigh a lot to sound great, in fact I even remember a price per pound ratio! What happened to that ratio? Why doesn't it apply to SITs?

"There is no real rule of thumb, but weight and lots of heats sinking are still reasonable things to look for. It still applies pretty well, and SITs even more so. SITs are just devices, but they shine in single-ended Class A applications -- observe the amount of hardware the SIT-1 uses to deliver 10 watts. You could easily think 3 pounds per watt. Inflation has crept in though, and the price of things like Aluminum reflect the cost of energy, and I don't need to explain that to you. Probably more with tubes... you want to run pure single-ended Class A without feedback, you will still need maybe five pounds of metal for every watt."