This review page is supported in part by the sponsors whose ad banners are displayed below

Before the sound commentary commences, a few parting words from Nelson Pass. They shed a bit more light on his current JFET experiments; what inspires his consistent efforts to pare down circuit complexity; and what he'd consider the perfect amplification part. He also penned a paper entitled The Sweet Spot.

In it he explains methods for running various types of output devices in the sweet spot of their load lines. By manipulating specific operational values without resort to negative feedback, significant distortion reduction is possible. It makes fascinating reading. It also confirms that much which used to be known during the heydays of tube circuits has been forgotten again. And it explains how there's rather more behind the apparent simplicity of FirstWatt amps than meets the eye.

"Things continue to heat up on the R&D front as I have received some stocks of different JFETs, all depletion-mode types. These parts self-bias like triodes and they have even more of the 'plate characteristic', that is to say that the transconductance is heavily dependent on the drain-source voltage. 

"Among these are the vastly parallel 'Beast with a Thousand JFETs', a tribute to Roger Corman, which uses 1176 x Toshiba JFETs in parallel complementary-follower pairs of 2SJ74 and 2SK170 which of course are no longer available.  Preliminary results indicate that the performance will be simply stunning.

"Also unobtainium are the complementary Sony 2SJ28 and 2SK82, a tray of which is sitting on my bench glowing in the dark. And eight pieces of another part that doesn't exist in anyone's catalog, the most powerful depletion-mode JFET in the world.  This one will probably be an amplifier that has only one semiconductor in the circuit, period. So it's going to be particularly busy this fall, working up some incredibly simple little amplifiers from these exotic parts. I expect to turn these into some "F" product and spin them out to those DIYers up for a real parts challenge.

"I recall the David Gelernter book 'Machine Beauty' in which the first chapter goes on for many pages describing the variations of the proper mating of power and simplicity to create Beauty in machines, prose, art or logic etc. I find myself consistently drawn to the concept with amplifiers. When I was a teenager, I wandered into an exhibition that off to the side had a Picasso pencil drawing entitled 'Femme'. It was a single pencil stroke that described most of what you might want to know about a woman and I was permanently imprinted by its power and simplicity.  Many years later I wandered into the Picasso dealership in LA where a very gracious proprietor spent a morning accompanying me through stacks of Picasso.

"It turns out that there are at least a couple of hundred 'Femme' and none of the ones he had were a single stroke. This notion of Beauty extends to amplifiers although the reference to power is not to watts but to the ability to express music in a satisfying way. The first Zen in 1994 was one of the simplest amplifiers imaginable at the time. It's not that people couldn't conceive of such a thing. They just didn't have the nerve to build it. Some critics commented that it was intended as a joke. It was not but it did need more power.

"A recent advertisement reads 'An amplifier should be as simple as possible - but no simpler!'. Looking at the product, I smile and conclude that it could have been simpler. There is still progress to be made. Newer parts are responsible for some of that but there is still plenty of room for improvement in their arrangement. People who discuss new FETs keep coming back to a particular Holy Grail - the triode.

"They seem to want a gain part that has a square law dependence on control voltage and also on 'plate' voltage. For those that know what they are talking about, the presumption is that the 2nd harmonic distortion cancellation available with a triode solves 'the problem'. 

"While this is a good technique, I don't think it is truly a final solution. I am not looking for solid-state tubes. I'm looking for parts and topologies that result in simpler amplifiers with better performance. That better performance is perhaps difficult to define but having played with a few hundred amplifiers over the last 40 years, I trust that I'll know it when I see it."
For those believing in the ultimate part materializing in the near or far future, Nelson had some useful parameters:

"The Holy Grail would most likely have very high voltage, current and wattage ratings. It would have very low and linear capacitance and very high and very linear transconductance both with respect to Ids and Vds and temperature (for a FET - substitute other pin designations for tubes and such). Complementary versions would be convenient as well as depletion mode self-biasing."
Commenting on parts as used in the J2: "I would like to see better parts yet. The improvements we have seen lately are accidental byproducts of parts designed for better switching. The SITs and the old Sony V-FETs parts are interesting for their dependence on Vds, which allows some cancellation of second harmonics *under ideal conditions* but I'm afraid that this technique has some limitations if you are looking for the Grail."

Enlarge Enlarge Enlarge Enlarge