Art Dudley Listening

Domestic audio is based on two simple processes: transforming movement into electricity and electricity back into movement. Easy peasy.

Audio engineers have been doing those things for ages. Have they improved their craft to the same extent as the people who, over the same period of time, earned their livings making, say, automobiles and pharmaceuticals? I don't know. But if it were possible to spend an entire day driving a new car from 50 years ago, treating diabetes and erectile dysfunction with the treatments that were available 50 years ago, and listening to 50-year-old records on 50-year-old playback gear, the answer might seem more clear.

Actually, I was just kidding about the answer not being clear in the first place: Compared to the advancements achieved by their colleagues in other fields, audio engineers might as well have spent the past 50 years stripping the leaves off branches and dipping them into termite hills.

Certain new developments have been worthwhile. Hats off to everyone who makes low-friction tonearm bearings, transformers that resist saturation, and other marvels that eluded our elders. Modern capacitors work wonders in some applications, as do modern resistors. Silicon diodes, rectifiers, and regulators are useful. The KT120 is a nice tube. Modern plastics and adhesives have made possible some excellent panel-type loudspeakers. Through the miracle of science, we can now safely and effectively wash 80 years' worth of records.

But it seems the majority of engineers in today's audio industry put their greatest efforts behind the least worthy ideas. Complex cables and their integral "correction" systems. Systems that propagate soundwaves from behind and around the listener. Wireless this, remote that, gold-plated this, carbon-fiber that. Products in all categories that can scarcely be moved, let alone lifted, let alone afforded, simply because sheer bulk is the only way their inventors could imagine to make the things better. And, of course, ever-more-powerful amplifiers, as would be required to drive the industry's ever-less-drivable loudspeakers. God help us.

Remarkably, there remains, in the mainstream of perfectionist audio, a sticky film of reverence for Quad ESL speakers and Garrard 301 turntables that could nauseate at 20 paces—like hearing the members of Styx or Queensryche declare their love of Johnny Burnette and Gene Vincent. This seems especially true of the amplifier makers: "Marantz 8B? Greatest amp ever, and a profound influence on our work." Sure. That explains the Mercedes S-class prices, the ridiculously thick laser-cut faceplates, the complex, heavily regulated circuitry, and the output-power ratings that reach into the hundreds of watts and beyond.

That last one is especially hard for me to swallow.

The sounding board
For 18 years I've reported on that strange corner of the world where people insist on playing records through low-power amplifiers and high-efficiency loudspeakers—which, of course, is how the thing was done at the dawn of domestic audio. Ever the aspiring John Reed, I became a convert to the cause I covered: Thus I've not only spent a cat's age writing about scores of flea-watt amps and sensitive speakers, I've bought—and occasionally built—a goodly number of the things for myself.

Over time, I've become more unshakably convinced that this is the best approach for a record lover such as I, who values tone, touch, and musical flow over all else. (There are a lot of other all elses, from which you and every other listener are free to choose.) That conviction led to my purchase, last year, of a crazy, hulking pair of Altec 846A Valencia loudspeakers. Before their arrival, I had never enjoyed such a high and wild level of system responsiveness in my home.

This choice of words is not casual, but rather is inspired by my visit last year, while preparing an article for The Fretboard Journal, to the shop of renowned luthier Dana Bourgeois. He is among the handful of luthiers who spurred the return, to the steel-string guitar industry, of the voicing techniques once popular in factories before the 1940s and '50s: techniques that were abandoned in an effort to streamline production and to produce guitars that were more durable.

Bourgeois begins by considering the manner in which the instrument will be used—the player's touch and picking style, the gauge of strings that he or she prefers, the desired degree of loudness, and so forth—and selects for the top a pair of spruce boards of the precise degree of required stiffness. Bourgeois and his co-workers then brace the top; tap it at various different nodal points, listening for a particular tone; slightly thin the braces; then re-tap, re-listen, and re-thin until the desired tones are achieved. After the top is attached to the body but before its binding is attached, the luthier flexes the top and, if the desired flexibility is not observed, he or she gradually thins its periphery. Finally, after the top has been trimmed, a luthier trained in the procedure taps the bridge to ensure that the top is pushing back to just the right extent.

This method of matching the instrument—which is, of course, an acoustical source, amplifier, and loudspeaker all in one—to the player is so natural, so reliably right, that one wonders why it should be done any other way.

At roughly the time when the larger guitar companies abandoned the notion of voicing their instruments, the leaders of the domestic audio industry decided that, in their quest for flatter frequency response, greater frequency extension, and more "precise" stereo imaging, they would be better off designing their loudspeakers to be unresponsive—that is, to perform less efficiently at transforming electricity into movement, which is the single most important thing a loudspeaker does.

How then, you might wonder, would the consumer drive such an unresponsive loudspeaker? By buying a much more powerful amplifier, of course—because, thanks to Our Friend the Transistor, power is cheap. Besides, all competently designed amps sound the same. Right?

Parallel
Last October, on the day before the last leaves fell from the trees behind my house, a pair of Shindo Laboratory's newest mono amplifier, the D'Yquem ($24,995/pair), arrived at my house for a brief visit (footnote 1). The D'Yquem is named for Chateau d'Yquem, which produces the most expensive and universally well-regarded of sauternes. Novelist Thomas Harris had his star antagonist, Hannibal Lecter, buy a 1961 d'Yquem as a birthday present for protagonist Clarice Starling (it was never delivered), and Julia Child, Ernest Hemingway, and Thomas Jefferson were among the wine's most notable admirers.

Each D'Yquem amplifier produces up to 18W from a parallel pair of 300B directly heated triode tubes, operated in single-ended mode (and thus in pure class-A). Shindo Laboratory claims for the amp a signal/noise ratio of 110dB, an input impedance of 120k ohms, and harmonic distortion of 0.01%. Although an output-impedance specification is not offered, I assume that the D'Yquem's single secondary Lundahl output transformer is, in typical Shindo fashion, optimized for loads of 8 ohms and (especially) higher.

A glimpse inside reveals some interesting variations on the design and construction details I've seen in Shindo's other amplifiers during the eight years I've followed the brand. Like most amps in Shindo's current lineup, the D'Yquem doesn't use tube rectification, and while every semiconductor-rectified Shindo amp has an internally mounted EY88 diode tube between the AC transformer's high-voltage secondaries and ground—it slowly ramps up the rail voltage to help prolong the life of other parts—the D'Yquem has two EY88s, wired in parallel. Strange! Another unusual doubling-up is seen in the bias supply of this fixed-bias amplifier, where the two output triodes are served by no fewer than four potentiometers. The 300Bs are themselves a different sort for designer Ken Shindo: contemporary Russian-made tubes bearing the oft-traded Genelex label, instead of the Western Electric or Cetron tubes I've seen in all of Shindo's other 300B models.

As he has with all of his amps and preamps, Ken Shindo voiced the D'Yquem with a mixture of vintage parts from his reportedly vast stock of same—Sprague Vitamin Q and Black Beauty signal capacitors, a lovely old Mallory electrolytic cap for the bias supply, NOS Philips 6AW8A dual-triode/pentode tubes for input gain and buffering—and other parts that are decidedly modern. The latter include a smattering of French polypropylene-film capacitors from Solen, an uncharacteristically large (for Shindo) dry-electrolytic reservoir cap for the main power supply, a new type of Japanese ceramic-substrate resistor for select applications in the signal path, and the Swedish Lundahl transformers that Shindo has come to prefer for most of his products.

---

Footnote 1: Shindo Laboratory, Japan, Web: www.shindo-laboratory.co.jp. US distributor: Tone Imports LLC, New York, NY. Web: www.toneimports.com.