Sutherland Engineering 20/20 phono preamplifier
It seemed a simple and obvious question, but I couldn't get an answer out of Ron Sutherland. Why did his new 20/20 phono preamp use an AC power supply instead of batteries? I asked directly, I asked repeatedly, I tried framing the question in different ways, all to no avail. Did the AC supply make it sound better? Was it less expensive to build? Were potential customers turned off by having to replace batteries once every year or two?
After all, his other phono preampsthe less expensive Ph3D ($1000), his more expensive Hubble ($3800), and his instant classic, the PhD ($3000 when last available)all share Sutherland's supercool power supply: a bank of D-cell batteries. All three also share a supremely quiet, natural purity that Michael Fremer has described as being "free from electronic detritus." That purity, and the underlying battery power, was what made Sutherland's phono preamps special. They were why people bought them instead of competing products. Some might say they were the reasons the Sutherland models existed at all. Their combination of sonic purity and battery power supplies was Ron Sutherland's signature, and arguably maintained his place at the forefront of today's audio designers. So why no batteries?
Sutherland politely acknowledged that mine were "good, simple questions," and that he'd "have to think about how to answer them." We discussed various aspects of the 20/20's design, potential future Sutherland products, even manufacturing techniquesbut as for the battery question, all I got was "It's a different design . . . a different product, and balances a number of design parameters. It is what it is."
So what is what it is?
Batteries or no, the Sutherland 20/20 is a gorgeous piece of audio gear. It shares a family resemblance with Sutherland's earlier models, the PhD and Direct Line Stage, but its slim proportions make it look much more graceful. Even Sutherland's top model, the Hubble, seems a bit chunky in comparison. Sometimes all the elements of a designsize, shape, finish, detailingcome together just so. The 20/20 is one of those designs.
Inside, too, it's sleek and simple. Under the top cover, which is removed by loosening four thumbscrews (neat!), are two identical, completely separate subchassis, connected only by the front and rear panels. This approach completely separates the two channels and, as Sutherland said, "Optimizing a circuit layout for performance and manufacturability is exacting, iterative, and very time consuming . . . and it's easier to do with a mono circuit than a stereo one. So why not spend all the time and attention you have on just one board, make it as simple as possible (mono), then use it twice, and separate them so they don't interact?"
The 20/20's boards themselves are typical Sutherland: beautifully laid out, lavishly executed, with top-quality components used throughout. The gain and cartridge loading are set by jumpers, a less costlythough less versatile and not as "purist"alternative to using plug-in resistors,
The "Why no batteries?" question aside, the 20/20's power supplies are another example of clever optimization, albeit one that Sutherland is a bit uneasy about. His concern isn't technical, but rather Will audiophiles allow me to do this? The this is using an inexpensive, outboard power supply designed for laptop computers for each channel. "The goal is to feed perfect, ripple-free, noise-free DC power to the boards. I think of it as putting distance between the circuit and the wall receptacle. Rather than do it all myself, I divide the process and let these power supplies do the first part of it for me. They do what they do extremely well, which is the rectification and bulk of the filtering, and the elimination of the noise and ripple associated with the rectification. Plus, now all that's done in a box that's 6' away from my audio circuits."
The outboard supply puts out 48V DC with about 17mV of ripple. The 20/20 then runs that through 11 cascaded stages of low-pass RC filtering based on super-premium electrolytic capsto eventually deliver about 30V to the boards, with residual noise and ripple that he describes as "way less than 1µVtoo low for me to measure and really know what it really is that I'm measuring." This over-the-top filtering approach is inefficient, he notes, citing the drop from 48 to 30V, but "because I'm using AC power, it can afford to be." He then reiterated that "I just need audiophiles to allow me to do this: to use these supplies to do their part incredibly well, and then to do my part incredibly well, to optimize the overall system."