Tube Power Amp Reviews

Had: "The choke takes the place of a conventional resistor as used in most amplifiers. It allows the 300B to swing a greater amplified AC audio signal to the output tube. And that's coupled through an oil-filled 0.22µF, 600VDC capacitor to the grid of a dual-section low-mu 6BL7 triode, which performs two functions in this design. First it provides the low-impedance bias voltage to the T-1610 output tube. And it also couples the amplified AC audio signal directly to the control grid of the T-1610. And that's fed B-plus voltage through the air-gapped output transformer!" He laughed one o' them good-ol'-boy belly-shakers of his. (Don't wanna pop your bubble or anything, but Had is originally from Ohio...) Gotta hand it to him, he always sounds cheerful, even when describing his air gap! [slap!] Missed me that time!

The CAD-1610-SE's power supply actually consists of eight different supplies. The power transformers are rated at 200% continuous commercial service at full-rated output power. The high-voltage power-supply section for the T-1610 is a full-wave center-tap configuration, "not some cheap voltage doubler as found in many of our competitors' amps," sniffed the ever-feisty Had. That's connected to a simple pi network, and all filament voltages in the amplifier are DC. He explains that DC supplies prevent AC ripple from capacitively coupling to the electrodes of the input and output tubes.

Had touted highly the CAD-1610-SE's overload recovery: "The ability of an amplifier to instantly recover from clipping is much more important than is commonly believed," he averred. "In the amplifier power war, everyone's focused on higher and even higher power output to solve the clipping problem. But in reality, Jonathan, the most critical aspect is how fast a recovery an amplifier can make after overload! With the incredible dynamic range of live and recorded music, even 2000W of power isn't enough!"

He continued, saying that most of the music being enjoyed in the average audiophile's home these days is the product of the first 3W of power. "It's on the transients of loud low-frequency material that tremendous signal voltage may appear at the input of the amplifier. And that's when overload recovery ability becomes of critical concern. The CAD-1610-SE instantaneously recovers from brief—or even extended—overloads! And it'll overload symmetrically at any frequency in the audio bandpass. If you monitor the high-voltage rail—550V DC—at the anode to the cathode of the T-1610 output tube during soft and loud musical passages, you'd find there's no more than a volt or so change from soft to loud passages!" Because of all this, he says that the big Cary CAD-1610-SE yields "faithful reproduction of extremely low frequencies at full output levels."

And you know I gave 'em a good workout!

Setup
I always clunked the 300B on for about five minutes before firing up the T-1610. Just respecting the rocket: the manual suggests 15–30 seconds. Ten minutes after lighting up the T-1610, I checked the bias to make sure it was running at around 200 milliamps of bias. I tried and preferred 220mA, where Had likes to run 'em. At initial installation, I did a final check after about half an hour. It never hurts to make sure, but I rarely had to fuss with bias. I did check the meters whenever I turned the amps on to make sure everything was on the up'n'up. Let's say that 1500V on the plate engenders a certain amount of respect.

Turning the great beasts off means throwing the knobs in the opposite direction. Mute the preamplifier (advisable in any case), turn off the T-1610 output tubes, and shut down the rest of the waterworks by finally clunking the 300B knobs to Off. Funnily enough, on shutdown there was none of the tinkle of contracting glass I'm used to hearing when powering down big 6550 or KT88 amps.

I don't know if this was an impedance issue, but I listened mostly with the input volume controls wide open, as encouraged to by Cary & Co. However, I have to admit, when I turned them both back a hair to, say, balance the image...I sometimes slightly preferred the sound! It was a tad darker and less quick, but somehow sexier. As I say, I left them mostly wide open, sometimes seasoning to taste when playing bright, chaffy, modern pop recordings by lowering both volume knobs a tad.

Another small but important detail: Coming in balanced on an XLR may not be optimal for a single-ended design such as this. While input-balanced transformers are of the highest quality—silver wire and nickel core, in this case, according to Had—a balanced input does require that additional bit of wire and circuitry. Ah, but most of the preamps I have around here are balanced. Ah, but the Audio Research Reference 2 sounded great with the Carys. Ah, but it's optimized for balanced operation even though it has single-ended inputs and outputs. Tried both ins'n'outs with the same cable, actually, and the Ref 2 balanced with the '1610 was best. Way best!

I wouldn't say the stern flagship BAT VK-50SE was a perfect match for the CAD-1610-SEs either. The big Cary monoblocks seemed to need something of the very slight, highly entertaining, even seductive plushness imparted by the Ref 2—the kind that makes your toes wriggle. This was a very slight effect, understand, and even less pronounced with balanced connections, but nevertheless a marriage made in heaven. I don't think the Cary cared for the Mark Levinson No.32 Reference preamplifier, and I'm not sure it ought to have—even though that full-featured solid-state preamplifier, which I'm still full of reverence for, did allow me to compare balanced and single-ended running.

But then, late in the review period, a Conrad-Johnson 16LS preamplifier showed up. Ooh, that was nice too—and the C-J is single-ended, of course. In fact, this brings up something of a conundrum. It's totally ironic that the newly moneyed Wall Streeters and dot-com billionaires with the cash to buy these monster amps are sure to be the worst at setting them up. These are tools for connoisseurs—a dealer who knows how to match 'em up is essential. I had to work at getting them just so with changes of wire, speaker cables, power cords, associated components...everything had to be perfect. In that way, they had something of the character of the Lamm M1 amps I reviewed in the May 2000 issue: very Nagra VPA in their transparency, very demanding of components upstream and downstream. This was made so by their extraordinary, almost 845-on-steroids sound, single-ended and supercharged, honey!

So I wound up relying on the great sound of the AR Ref 2 fed by all the great digital gear I have on hand—except when getting great results with the C-J 16LS, which, as a single-ended device, worked very well with the big Carys too. And all this hooked up with TARA Labs The One, a few with the new ISM Onboard modules. The Carys sounded best with the new PS Audio Lab Cable power cords, run "Poor Man's Balanced" on two separate legs of positive phase, as detailed in previous episodes of "Fine Tunes."

But the CAD-1610-SEs hummed unless I plugged both Cary-supplied Powercon Powerkords from their input IECs to the same 30-amp circuit via a Cardas Power Extender. Ah, but the Lab Cable has a little clip that allows you to disconnect the ground without losing the shield. With the clip slipped on one cable and these heavyweight power cables in-circuit, Poor Man's Balanced ran without a hitch, and added dynamics and power handling via balanced operation from both sides of our friendly neighborhood power transformer. Synergistic Research Designer's Reference² connected literally everything else.

See what I mean? Every detail hadda be perfect.

Single-ended on Steroids!
Belying their rocket-to-the-moon output tubes and huge, gantry-by-Michael-Graves presence, the 55W, single-ended CAD-1610-SEs sounded lithe and quick, like a perfectly tweaked his'n'hers pair of 289 AC Cobras. Yes, that would be the 289, I think, not the articulated, tire-squealin', 427/428, Big Boy monster truck. The CAD-1610-SE is, after all, only 55Wpc RMS. But what 55Wpc!

I sat down for a long listening session late one evening, beginning with Kind of Blue in both SACD and CD formats (Columbia/Legacy CD 64935 and CK 52861, respectively), then "Secretly Happy" from John Hassell's Fascinoma in both formats (Water Lily WLA-CS-70-SACD, WLA-CS-70-CD) for ultimate atmosphere and acoustic presence, then a couple of tracks from Dave's True Story's Sex Without Bodies (Chesky JD164), and winding up with "Run On," from Moby's great Play (V2 63881-27049-2)—female and male vocals, jazz, Moby's madness (try to classify him), all at realistically high volumes.

First things first: As far as power was concerned, I could freely push the big CAD-1610-SEs; they played much louder than any single-ended amp you or I are ever likely to encounter, speaker impedance be damned! These big, powerful amps took charge of the JMlab Utopias' 12" lightweight sandwich-technology woofers with ease, punching out a perfectly righteous bottom end with no excuses, except with the most complex of material at lease-busting levels. Even then, the sound got a little fat and woolly only at the very bottom, as you'd expect, and a bit shelved-back down there when really big bass saturated the balance. But nothing objectionable, and always hugely impressive for what it did pump out. No wonder Had had been so delighted that I'd be using the CAD-1610-SEs to drive the Utopias.