47 Laboratory 4706 Gaincard power amplifier
Perhaps as a backlash to all these increases in technological complexity, a number of audio designers have come to endorse the principle of "less is more." Balanced Audio Technology's Victor Khomenko argues that the Unistage design of their preamplifiers is superior to designs with two or more stages. Nelson Pass, in his more recent designs for Pass Labs, has been reducing the number of stages in his amplifiers, an approach that he claims yields better sound. It's been suggested that tubes retain their popularity at least in part because using them forces designers to keep things simple.
The pursuit of simplicity in the design of solid-state audio electronics is perhaps best exemplified by the products from 47 Laboratory. Designer Junji Kimura's motto is "Only the simplest can accommodate the most complex," and he has gone to great lengths to put this into practice. According to the 47 Laboratory literature, the Model 4706 Gaincard amplifier features the following:
• world's smallest number of parts: nine per channel (excluding attenuators)
• world's shortest signal-path length: 32mm (including length of parts)
• world's shortest negative feedback loop: 9mm
• dual-mono construction, with each channel in a separate chassis, all parts (except attenuators) mounted on a 45 by 30mm (1¾" by 1 1/6") circuit board
• rigid, compact chassis of machined aluminum plate
In addition, the associated Model 4700 Power Humpty power supply features:
• world's smallest power-supply filter capacitor: 1000µF
• high-capacity (170VA) cut-core transformer with individual coils
• an aluminum-tube chassis
• no damping materials in either structure, "to release vibrations smoothly"
Not everyone would agree that all of the Gaincard's design features are desirable. Its negative-feedback loop may be very short, but some designers feel that, for the most natural sound, loop negative feedback is best avoided entirely. Lower power-supply filter capacitance is not universally regarded as a good thing. Chassis damping is used to good effect in many amplifiers. Most of all, the claim of having the world's smallest number of parts is questionable, in that one of the nine parts is an op-amp—a highly complex "amplifier on a chip" that includes many transistors, resistors, and capacitors. Power amplifiers based on op-amps are widely regarded as suitable only for low- or mid-fi applications; high-end designs usually involve discrete components.
But regardless of claims and espoused design principles, the bottom line, always, is the product's sound.
My encounter with the first sample of the 4706 Gaincard provided for review was less than auspicious. I plugged in the interconnects, attached the speaker cables, plugged in the cables linking the Power Humpty to the amp, turned the volume controls to minimum, finally plugged in the AC power cable—and the amp emitted several sparks, accompanied by a loud thump from the speakers. I quickly unplugged the power cable and checked the connections to make sure that the outputs were not shorted, but everything seemed fine.
When the smoke cleared—literally—I got on the phone to Yoshi Segoshi, 47 Laboratory's North American distributor, and told him what had happened. He was surprised and, understandably, upset; apparently, the amp had been working fine before it was sent to me. He offered to send me a new sample, and I agreed to this. (I checked out the speakers with a different amplifier; fortunately, they weren't damaged, but I wasn't about to try connecting the same Gaincard sample again.)
A few days later, a package arrived; in it was a Gaincard, but no Power Humpty. I queried Segoshi about this, and he confirmed that that's all he'd sent; he didn't think it could be the power supply, which he said had a perfect record of reliability. I plugged the new Gaincard into the Power Humpty and was again treated to a spectacle of sparks and smoke. I called Segoshi again. This time, he said he would personally deliver a brand-new Gaincard/Power Humpty.
He showed up a few days later with just that combination. The new Gaincard/Power Humpty was connected, and everything worked perfectly, with no fireworks whatsoever. A voltage check of the two Power Humpty samples showed that the DC output of the first was about 50% higher than that of the second. (Segoshi informed me later that the first Humpty turned out to have a faulty diode.) No problems were encountered with the new review samples.
Connecting the Gaincard to the rest of the system was a bit tricky. The amplifier itself is very light; heavy cables can unbalance it or even pull it off an equipment rack. Speaker-cable connectors are small Phillips-head bolts (footnote 1), which will work only with bare wire or not-too-hefty spade lugs; banana plugs can't be used, and biwiring would be difficult to accommodate. Fortunately, the Nirvana spades worked fine, and I had no need to biwire. When the Gaincard was driven by the CAT SL-1 Ultimate preamp, the noise level with the Gaincard level attenuators turned all the way up was too high, so I turned them down three clicks, which resulted in a level of noise inaudible from my listening seat. (The Avantgarde Uno 2.0's +100dB sensitivity exposes any residual noise in my system.)
The Gaincard was not demanding in the choice of mechanical supports and damping; I found little benefit to using Aurios or Symposium Rollerblocks, and placing a VPI DB-5 "magic brick" or a Shakti stone atop the amplifier had no noticeable effect on the sound. (The Power Humpty's cylindrical shape prevented my using of any of these devices with it.) For most of my listening, the Gaincard rested on three Aurios, but I did this mostly out of habit. There was nothing placed atop the amplifier.
However, the Gaincard was extraordinarily demanding in the choice of associated equipment. The Nordost Quattro Fil interconnects and SPM speaker cables, which had worked well with several other amplifiers, proved too bright when combined with the Gaincard; the Nirvana SL interconnects and speaker cables provided a more pleasing tonal balance. 47 Lab's stock power cable looks like nothing special, and, in fact, the sound of the Gaincard improved significantly (less high-frequency hash) when the stock cable was replaced with the PS Audio Lab Cable, and even more when this was combined with the PS Audio High Current Ultimate Outlet.
Finally, the CAT SL-1 Ultimate, although in the very first rank of preamplifiers, obscured detail slightly when compared to the sound of the Gaincard driven directly by the Perpetual Technologies P-1A/ModWright P-3A. The latter arrangement, which I used for most of my critical listening, dictated using the Gaincard with just a single line-level source, and volume had to be adjusted with the Gaincard's passive attenuators (footnote 2). But the volume-control steps were too big; when I wanted to play music loud, I often had to choose between not quite loud enough and louder than comfortable. The Gaincard is available with attenuators that divide the loudness range into 12 steps, which provide finer level control. The cost of this option (which is retrofittable) is $150; I consider it essential for a system that doesn't have a preamplifier.
Footnote 1: 47 Lab could probably claim that the Gaincard has the World's Smallest Speaker Connectors as well—a dubious distinction.—Robert Deutsch
Footnote 2: The 4707 Input Chooser ($720) has two outputs and allows selection from among four sources.—Robert Deutsch