Listening #71

Here's how God makes audiophiles: He starts with several million blank brain cells, then programs each one, individually, to function as either a love for one single aspect of music reproduction or a hatred for another. There are over a thousand such cells—far too many to list here—but theologians and audio reviewers have worked together to compile this list of the Top 20, which, just like real life, contains a little more love than hate:

Love of lushly textured midrange
Love of laid-back treble
Love of good rhythm and pacing
Hatred of surface noise and tape hiss
Love of movement, flow, and momentum in sound (ie, "music")
Love of stereo imaging effects
Love of big sounds
Love of little sounds
Love of sibilants and plosives
Hatred of wow and flutter
Hatred of stereo
Hatred of surround-sound
Love of flat frequency response
Love of stirringly dramatic music
Love of deep bass noises
Love of perfect pitch
Love of accurate pitch relationships
Love of the flute
Love of the banjo
Love of Leroy Anderson

Two or three times a week, God makes an enormous new batch of those and other cells. Then He scoops them up in a jar, shakes the jar a little (or a little too much, judging from the limited social skills of some audiophiles), and pours them into a special chamber in each baby's brain, more or less at random.

Thus some babies grow up to be audiophiles who take "high fidelity" to mean "flat frequency response" and to hell with everything else. Some others take "high fidelity" to mean "good rhythm and pacing" and to hell with everything else. Some cling to CDs and inefficient loudspeakers with dense, complex crossover networks, while others declare that 78rpm records and horn loudspeakers are the highest of high fidelity. It doesn't really matter. God loves them all equally well.

By the way, there are no such things as "brain cells of indifference." If you meet someone who is indifferent to, say, the sound of the banjo, it simply means he wasn't blessed with the cell for loving it. Too bad!

As an audiophile, your own unique set of loves and hates is, quite literally, a gift from God. If you meet someone who accuses you of hating "high fidelity" just because your list is different from his, you should pity him. If he persists in haranguing you, then you should shun him: That person is clearly an apostate of hell (or perhaps just an overly puritanical audio reviewer—but there's no sense taking chances).

Crazy things
Men and women set out to accomplish all sorts of crazy things, but we seldom achieve every one of the results we had in mind. So it goes with people who build performance spaces: No matter what an architect has in mind—or an acoustician, or a contractor—rooms and halls and theaters that are designed for music virtually never sound exactly as expected. Music in a given space will always sound like music in that space. Which is perfectly fine with me.

There's a big old barn nearby that people use for chamber-music performances every July and August, even though the place was designed for something else (storing and drying hay, I would imagine). Music in that space always has a certain distinctive sound: When I listen there, I'm compelled to listen in a certain way, simply because my attention is drawn more to some aspects of the sound and less to others. Which is also fine with me.

The same can be said of literally every place I've ever listened—or performed, for that matter. They all sound different from one another, and they all serve to make every performance unique in my memory—aided, in that respect, by a unique combination of sights and smells and emotions, with which that memory will forever be associated: another gift from God. There's nothing wrong with trying, time and time again, to duplicate a very good musical experience—say, for instance, the Kinks at the Fillmore East in 1971, heard from the front row of the loge—as long as you're prepared to get something else instead.

Good experiences with live music are just as common as you allow them to be. Some day you'll hear a sound you thought you didn't like—whether a Mendelssohn trio or a Sousa march or even a mob of banjo pickers—and the experience of listening to it in that space will bring out some quality you missed before. You might love it!

Here's how people make amplifiers: Unlike God, they begin with a bias that compels them to do things a certain way—like the bias that compels people to go left or right or straight at every intersection between one place and another—and then they go about soldering parts together, favoring the ones that seem to work well and rejecting the ones that don't. Some commercial builders are likelier to choose parts based on their cost—cheap ones for the cheap customers, rich ones for the rich buyers—but that's another matter altogether.

There are literally hundreds of different performance aspects to worry about when creating an audio-frequency amplifier, and different builders emphasize them in very different proportions to one another. So, too, might an individual builder create different amplifiers to satisfy the tastes of different listeners. That approach isn't necessarily wrong, and may be righter than most.

Today I'm listening to an amplifier that someone left with me just a few weeks ago, more or less for the fun of it. After enjoying it thoroughly—almost giddily—for several evenings in a row, I took the amp out of my system and drove it to my friend Neal Newman's house, where we measured its technical performance using Neal's signal generator, distortion meter, dummy load, and oscilloscope. I'll refrain from naming the amplifier for now, if only to avoid triggering any associations or prejudices in the minds of readers who think they know what it sounds like. In any event, for the purposes of trying to describe the relationship between measuring amps and listening to them, I don't think the identity of this particular sample matters all that much: There are probably a dozen different amplifier models out there, old and new, about which I could write the following.

Neal and I began by trying to determine the basic circuit types used in the amplifier's various stages: All we knew ahead of time was that this tube amplifier—let's call it the Verdi—had a push-pull output stage. Seeing that the cathodes of the output tubes were tied directly to ground, we surmised that the Verdi was a fixed-bias design (footnote 1). The lack of adjustment potentiometers in the bias circuit suggested that the DC on the grid was set by just two or three resistors, presumably hand-selected in order to balance and stabilize the output section as a whole. Output impedance was determined by an off-the-shelf transformer wound with 40% taps on both ends of the primary: The Verdi's output section appeared to operate in ultralinear mode, and there were no global feedback taps whatsoever.

The power supply was straightforward—a simple pi filter for smoothing, tube rectification for the rail voltage, and solid-state rectification and regulation for the bias voltage—but the input section was inscrutable, and remains so. The first half of the first dual-triode was apparently a simple voltage amplifier, but from there the trail grew faint, although we were able to identify one particular byway as a differential feedback loop for the input section (see below). The Verdi's custom-built mains transformer had one pair of 6.3V secondary leads for the rectifier tube filaments, and another for all of the other tubes, the filaments of which were connected in parallel.

The first measurements we performed on the Verdi required little more than a good digital multitester. We observed a rail voltage of 373VDC on the output pentodes, this dropping to just over 300V for the input dual-triodes: sensibly lowish numbers, as anyone who's concerned about tube longevity would agree. Bias on the signal grids of the output tubes was in the neighborhood of –27V.

The plate current measurements for the output section threw us a curve: Even after warming up the Verdi for four or five minutes, we measured 88 milliamps on one tube in a complementary pair, 101 milliamps on the other. Those numbers continued to shift for at least 10 minutes more, after which they were virtually equal and remarkably stable. Could that differential feedback circuit—which itself depended on the stabilization of the Verdi's input tubes—actually have been a means of regulating plate current under real-world conditions? (In which case, and in light of the fact that fixed-bias amps are by definition slightly more complex than their auto-bias counterparts, was this an especially ingenious example of more-is-less engineering?) We didn't test the manufacturer's claim of class-A operation: That would have required us to unsolder some parts to facilitate active current measurements, which we weren't comfortable doing to a product that hadn't been formally submitted for review. But given both the highish plate current at idle and various other elements of the Verdi's observed performance, detailed below, we assumed that it could honestly be called class-A under most conditions.

Footnote 1: For further reading in this interesting subgenre, check out John Atkinson's review and measurements of the VTL Compact 100 power amplifier.