Ayre Acoustics AX-5 integrated amplifier
Stereophile 08/2013 - Art Dudley
The only thing better than a review that writes itself is a product with a compelling story. Although the latter asks a little more of us here, it's usually the more enduring pleasure.
So it goes with the new AX-5 amplifier ($9950) from Ayre Acoustics, in which designer Charles Hansen has both revived an overlooked technology from a half-century ago and brought to market a more affordable embodiment of one of his own most well-received products. All this comes in a package that requires no more room than the average electric typewriter, and that takes the place of everything you might normally put between your digital source components and your loudspeakers. It breaks a few rules—and almost breaks the dreaded five-figure price barrier—but the AX-5 is, in fact, an integrated amplifier, if one whose like you may not have seen till now.
That all-but-forgotten technology is the diamond circuit: a gateable bridge network of four bipolar transistors first described in 1964. Imagine, on the left-hand side—the signal-input side—a PNP transistor, and, below it, an NPN transistor, tied together at their bases; and, on the right-hand or signal-output side, an NPN transistor and, below that, a PNP transistor, this pair having their emitters tied together. The emitters of the left-hand transistors are tied to the bases of the adjacent right-hand transistors, and the collectors are biased all around: negatively in the case of the PNPs, positively for the NPNs. When drawn in the manner I described, with connecting nodes at north, south, east, and west, the schematic assumes a diamond shape, hence the name.
Richard Baker of MIT, to whom the patent for the diamond circuit was assigned in 1967 (he applied in 1964), described it as having a number of strengths: It can operate effectively in a floating- or above-ground condition; it can produce considerable power gain; it's reliable; it's fast; and, perhaps best of all, the diamond circuit is simple. (Baker's invention is also uniquely adaptable to modular construction, a distinction that, while not germane to this review, makes his bridge network especially well suited for use in integrated circuits, including some contemporary D-to-A converter chips.)
Charles Hansen, a self-described amateur historian of audio technology, acknowledges those advantages and adds one of his own estimation: that the diamond circuit, used as an output section, simply sounds better. The reason? Hansen suggests that, when compared with other solid-state push-pull topologies—in which two phases of a signal are recombined to form a full wave—the diamond is the only one in which the two half-signals are joined at a single point in the circuit, with no intervening circuitry. Thus, the diamond circuit creates an output that's more faithful to the shape of the input (footnote 1). Ayre biases the diamond output section in the AX-5 to operate in class-A/B, whereby it delivers 125Wpc into 8 ohms or 250Wpc into 4 ohms.
Hansen's other remarkable idea for the AX-5 is a circuit innovation called variable-gain transconductance or VGT, first seen in Ayre's top-of-the-line KX-R preamplifier of 2008 ($18,500). As Hansen explains, most active preamplifiers work by applying to the input signal a certain amount of voltage gain, so the signal can effectively drive a power amplifier. But in order for there to be a reasonable volume range—and to simply keep the playback level from being too loud—the voltage-gain stage is preceded by a potentiometer, which attenuates the signal. The drawback of this is that such a preamp will exhibit its maximal signal/noise ratio only at its maximal (unattenuated) volume. As Hansen puts it, "Since most preamps are used anywhere between –10dB and –40dB for an average listening level, this means the S/N ratio in actual use will be 10–40dB worse than on the spec sheet."
As implemented in the AX-5—which doesn't incorporate a preamplifier stage per se—Ayre's VGT circuit allows the user to determine how much gain is generated by the amplifier's input stage, which itself comprises a total of four complementary-differential JFETs. The volume knob on the AX-5's front panel controls a pair of enormous, motor-driven, Shallco silver-contact rotary switches, each of which contains dozens of hand-selected, low-noise resistors. Every volume-level adjustment made by the user has the effect of switching into the AX-5's input-stage circuit a different set of resistors, the values of which alter the transconductance of those JFETs—and thus calls into play a specific level of gain corresponding with that setting. The volume system has 46 steps of 1.5dB each, over a range of 67.5dB. (I'm told that, by changing a single resistor in each of the AX-5's channels, one can adjust the overall gain range to accommodate, say, speakers that are significantly more or less sensitive than average.) Thus the AX-5 doesn't use signal attenuation at all, but rather creates variable input-circuit gain, on demand, to suit the desired volume level.
In keeping with Ayre's long-standing practice, the AX-5 is not only a zero-feedback design; its circuitry is fully balanced from input to output. Interestingly, Charles Hansen endorses balanced technology not for its ability to reject hum and noise that might enter the circuit through cables, but because it (similarly) rejects the hum and noise that could enter the signal path from the power supply. "When the day comes that a totally perfect power supply is developed, there will be no more need for using balanced circuitry in home-use applications," he has stated.
The circuitry for the AX-5's fully analog power supply is located near the front of the chassis: ie, as far as possible from the input circuitry, much of which is snugged against the rear panel. Input selection is accomplished with FET switches, implemented with Ayre's own supporting circuitry, a technology the company has employed for a number of years, and which Hansen praises as being superior to all others in noiselessness and reliability. Between the left- and right-channel input boards one finds the big Shallco rotary controls, fronted with a very high-torque Lin step motor and the custom mechanicals used to work the switches: an impressive dual-mono system, synchronized with a toothed polymer belt. A fairly enormous EI-frame power transformer occupies the very center of the chassis, straddled by the left- and right-channel output sections and their generously sized heatsinks.
Ayre says they designed the AX-5 so that its dimensions would correspond with the infamous golden ratio, in common with the other 5 series in their line. I'm baffled as to how it qualifies for such a distinction—according to my ruler, and ignoring jacks, knobs, and other protrusions, the AX-5's case is almost perfectly square. That said, I was impressed by an observation from Ayre's Alex Brinkman, who says that, after their ideal dimensions were selected, "we got the transformer we wanted, with the specs we wanted—and it was too big to fit in that chassis. So rather than make it even bigger, we cut an opening in the bottom of the chassis, to accommodate the transformer's height." Indeed they did, and rather neatly, I'd say. The finished box is made of aluminum, with a brushed exterior. Most of the fasteners are stainless steel, and the quality of fit between various panels was all right; but the panel edges were sharper, and their overall feel less smooth, than I would wish to see in a product that sells for nearly five figures.
Installation and Setup, Part One
The AX-5 comes equipped with four balanced inputs (XLR jacks) and two single-ended inputs (RCA jacks): Charles Hansen reasons that it's easier and less expensive to adapt a balanced input for single-ended use (footnote 2) than the other way around. (And the other way around will always be somewhat compromised.) In any event, given that fully balanced operation is one of Ayre's calling cards, it's reasonable to expect that the prospective buyer of an AX-R would not be troubled by the two-to-one ratio in favor of balanced inputs. Indeed, even though I don't keep a balanced system—which, I gather, is only slightly less challenging than keeping a kosher kitchen—I spent slightly more time using the amp with a borrowed sample of Ayre's balanced QB-9 D/A converter than my own single-ended Sony SCD-777ES SACD/CD player.
Prior to use, either the buyer or the buyer's dealer must access the AX-5's Setup mode in order to configure and activate one or more of its six inputs, none of which is usable straight out of the box. In Setup mode, both the input-selector and volume knobs—on the left and right sides of the front panel, respectively—are used to scroll through and select various bits of information, typically with the aid of two front-mounted button switches. Configuration didn't take terribly long—it's a software thing, done from outside the amp, no tools required—and would seem to be within the capabilities of anyone with the patience to read the manual and to follow the steps as written. (None of the AX-5's setup regimens was sufficiently intuitive that I could perform it without manual in hand, even after weeks of use.)
Once configured, the AX-5 was simple to use. With AC power applied by means of a hefty rear-panel rocker switch, the Ayre is roused from its low-power-consumption mode—in which bias current is removed from the collectors of the output transistors—with a push of the right-hand button. (Subsequent to that, a brief press of the same switch effectively mutes all inputs; a long press returns the AX-5 to low-power mode.) From there, things are straightforward: Use the left-hand knob to select the desired input (only inputs that have been configured will show up as choices: a boon, I think, for those of us who dislike the cognitive clutter of choices in which we have no interest), the right-hand knob to adjust the volume. As to the latter, the electromechanical noise produced by each step up or down the ladder of loud takes some getting used to—sounding, as it does, like a roof leak dripping into an empty metal pail. After a day or two, I found it rather charming.
One could also rely on the included remote handset: a big, hefty, confidence-inspiring piece of metal that I came to enjoy more than most such devices. I especially enjoyed using its illumination button—the only soft-touch button on the handset that's set aside from all the others, allowing the user to more easily find it in the dark—which back-lights, for approximately eight seconds, all of the handset's soft-touch pictographs, most of which make sense.
Two miscellaneous setup observations:
- With the AX-5, Ayre endures in using Cardas's proprietary single-knob speaker-terminal pairs, which favor spade lugs above all else. In past encounters with these terminals, which I admire for both their mechanical positivity and the quality of their construction, I managed to nonetheless fit my (preferred) banana plugs with a little effort. Yet on my review sample, the folks at Ayre had drilled, very precisely, a pair of holes on each Cardas terminal block, sized and placed perfectly for an average banana. I loved them. It still isn't clear whether this was done just for me, or if it's a modification that Ayre will apply to all units, but I hope it's the latter.
- The Ayre AX-5 ran hot. Not the-curtains-are-on-fire hot, but hot enough. Compared to my Shindo combo of Masseto preamplifier and Corton-Charlemagne monoblocks—representing exactly 20 vacuum tubes in all—the Ayre AX-5, by itself, made the room warmer.
Installation and Setup, Part Two
A day or two after my review sample of the AX-5 arrived, Ayre's Alex Brinkman visited my home to make sure that all was well. His helpful observations included: "This amp takes a long time to break in," "Remember to break it in," and "Remember to take a long time breaking it in." Brinkman is cool, so I said that I would.
When he got back to Ayre headquarters, in Boulder, Colorado, Brinkman sent me an e-mail: "Even though your unit was broken-in at the factory, with the AX-5, break-in is an even bigger deal than usual. So please . . ." You know where he was going with that. Yes, I made all the more sure to do a good job of breaking it in.
Brinkman's concern was fueled by his observation that, on at least one AX-5 in the field, pairs of inputs that had been used the most sounded considerably better than those that had been used the least; break-in time was assumed to be the cause of the disparity. To the dismay of all, that later proved untrue: It turned out that certain diodes used on the AX-5's input boards—used in protection circuits for the above-mentioned FET switches—were defective, and exhibiting leakage current. It was just a coincidence that a well-worn input sounded drastically better than an untouched one: The untouched one proved to be hobbled with leaky diodes.
Ayre got in touch with me—and with dealers who'd already sold AX-5s—regarding the problem and the fix, the latter requiring the field replacement of the old left- and right-channel input boards with ones that have newer, better, and altogether less leaky diodes. As per Stereophile's policy, John Atkinson and I agreed to apply the fix to the review sample, albeit not before noting the pre-fix sound of that unit.
That sound was decidedly ordinary. During its first weeks in my home, before the board-swap, the AX-5 sounded good, but not $10,000 worth of good. Not by a long shot.
That was about to change.
What a good tube amplifier sounds like and what it plays music like are two different things. A good tube amp usually sounds warm and colorful and well textured, often with a bottom end that's a bit fat and a treble range that isn't terribly trebly. A good tube amp plays music in such a way that lines of notes have a sense of flow and momentum, with a distinctly forceful tactile quality: an unmistakably human sense of touch, from the subtlest sound to the bangiest.
When most audio salespeople describe a given solid-state amplifier as being just the thing for tube lovers, here's what they usually mean: It sounds warm and has a fat bottom end and not a whole lot of highs. That's all well and good, but that isn't what I necessarily want to hear: What I want to hear is flow and momentum and a human sense of touch.
I have never heard a solid-state amplifier that could both sound like and play< music like the best tube amplifiers of my experience, but the warmed-up, broken-in, post-diode-fix Ayre AX-5 got a huge portion of the second part right. Through it, I found it easy to enjoy the records in my collection more as music than as sound, and while that may seem like faint praise, it is anything but.
Through the Ayre, "Renée Falconetti of Orléans," from Jenny Hval's Innocence Is Kinky (CD, Rune Grammofon RCD 2142), sounded rather different than through my own combination of Shindo Masseto preamp and Shindo Corton-Charlemagne amplifiers: the latter imbued the music with much fuller, more colorful bass; the former was much more explicit about not only the lyrics being sung but also details in the arrangement. (Stacking of voices, especially, was far easier to hear and appreciate through the Ayre.) But the Ayre allowed the music almost as much force, from the stringy-sounding keyboard bass that opens the song through the compellingly sung litany at the end. And the Ayre gave it the same relentlessness, the same hypnotic flow. The two different experiences were differently enjoyable and insightful.
During the AX-5's time in my home, the experience of listening to David Grier's I've Got the House to Myself (AIFF file ripped from CD, Dreadnought 0201) was equal parts pleasant, exciting, and very insightful, the last quality more than I'm used to. Apart from three original tunes, this is a collection of mostly flatpicking standards—a sort of bluegrass version of Joe Pass's Virtuoso album—in which the guitarist begins each number by hewing to the melody, followed by a series of variations. Through the Ayre, the balance of incisive detail, perfect musical timing and momentum, and naturally warm timbral colors allowed me to follow and understand each variation with fresh ears. And when, in the alternate-tuning number "Evening Prayer Blues," Grier moves from softly picked lines to a series of forceful bendings of the bottom string, the effect was magnificent—and remarkably lifelike.
In what may, to some enthusiasts, be its most compelling strength, the AX-5's clarity of detail and clarity of line seemed allied to the amp's clarity of space: The sense of openness with which the Ayre presented spatial elements contained in good stereo recordings—details of size, of wholeness, of position and perspective—was exceptional. With recordings that strive for a natural presentation, such as André Previn and the London Symphony Orchestra's recording of Tchaikovsky's Manfred Symphony (LP, EMI/Alto ASD 3018), the Ayre communicated a natural perspective with believable hall sound and an exceptionally good (I'm tempted to say "tube-caliber") sense of scale. On the other hand, with those stereo recordings whose imaginary stages are more an artificial creation, as on Joanna Newsom's Ys (AIFF ripped from CD, Drag City DC303CD), the Ayre held my attention with its ability to make unlikely sonic juxtapositions sound convincing.
Of course, that least appealing of all transistor-amp qualities—that they reliably allow horrid recordings to sound as horrid as they are—was also present and accounted for: eg, the screechy and grainy Barber Adagio for Strings, by Leonard Bernstein and the New York Philharmonic (CD, Columbia S3K 90578). The awfulness of Columbia's classical sound has been greatly exaggerated with regard to some of their recordings, but not this one. Similarly, though the Ayre was itself never bright per se, the sound of Peter Rowan's eponymous debut album (LP, Flying Fish 071) certainly was, as heard through the AX-5.
Although never sounding quite tube-like in the most obvious ways, the AX-5 provided good if not generous amounts of timbral color, and very good levels of natural instrumental texture: The sounds of antique wind instruments in particular, as throughout Musica Antiqua Vienna's Le Jardin Musical (LP, Supraphon 1 11 2126), were pleasantly tactile. Overall balance was quite good: decidedly modern, but not the least bit bright. Electric basses sounded appropriately full and no more—as with Colin Moulding's excellent lines throughout "Books Are Burning," from XTC's Nonsuch (CD, Geffen GEFD-24474). And the sound of a low D-flat from a digital bass keyboard in "Amen," from Leonard Cohen's Old Ideas (AIFF ripped from CD, Columbia 88697986712), showed that the Ayre's bass performance could be perfectly scary when called for. Nice.
At the start, I was dubious (as one should always be, I suppose): With its lack of a phono preamp, its lack of traditional controls, and its lack of tubes, the Ayre AX-5 is a few steps to the right of my comfort zone. By the end of its time here, I was extremely impressed—especially with the warmth, color, and clarity I heard while this amp was busy driving the hell out of DeVore Fidelity's Orangutan O/96 speakers.
On the minus side, the AX-5 is more difficult to configure than it needs to be, not as luxuriously finished as it ought to be, and not quite as flexible as I would want it to be. I would gladly trade the ability to set input-specific levels, and even the ability to dim or extinguish its lighted panel, for a balance control or a mono switch.
On the plus side, it's one of the three best, most musical, most human-sounding solid-state amps I've ever heard, in which regards it rivals the DNM PA3-S and the darTZeel NHB-108. Heck, it's one of the finest amps I've lived with, period. As such—and as something that takes the place of both preamp and power amp, and to which one need add only a good phono preamp—the AX-5 is a respectably good value. I can imagine this product becoming something of a standard, and deservedly so.
Ayre AX-5 specifications
Ayre AX-5 associated equipment
Ayre AX-5 measurements