Threshold FET 10 preamplifier Page 2
It is much easier to assess the accuracy of a preamplifier than it is to evaluate a power amp. As I mention in my report on the Mirror Image 1.1S power amp (also in this issue), the problem with a power amplifier is that you can't listen to it without using a loudspeaker. And since you can't listen to a loudspeaker without an amplifier, it is almost impossible to tell what either sounds like. You can make an educated guess, based on what a lot of loudspeakers sound like on a lot of amplifiers, but you can never be absolutely sure.
Not so with a preamp. Being a voltage amplifier, a preamplifier is supposed to be matched into a load high enough to have no effect at all on its sound. That is, typically, a load at least 10 times its output or source impedance. And since—except for volume and balance adjustments—a high-level controller like the FET-10 is intended to put out an exact replica of what goes into it, its success is easily gauged by comparing, by ear, the input with the output signals. That, the basis of the bypass tests which I have used for the last several years, is the reason I am able to report that a preamp's high-level section is either absolutely accurate, or deviates in this or that way from absolute perfection.
Bypass testing a phono preamp stage is a little more difficult though, partly because it is supposed to change the sound by impressing upon the signal the standard playback-equalization characteristic defined by the Record Industry Association of America, and partly because real-world phono cartridges emit certain spuriae peculiar to mechanical groove-trackers (and not easily replicated by another signal source). The first problem is easily circumvented by the use of an inverse-RIAA network, which gives the input response a characteristic complementary to the playback EQ.
The second requirement—the addition of small mistracking pulses to a cartridge's sound—while electronically simulatable in theory, has so far evaded such replication. For that reason, all my preamp tests include some listening with a real cartridge. The extent to which the preamp stages exaggerate or add roughness to the cartridge's slight mistracking pulses continues to be an important difference between competing preamp designs. (High-level stages have a similar effect on the garbage content in CD sound, but the basis of this problem has been diminishing steadily as players continue to improve.)
For bypassing, I derive my test signals from CDs and from original and 1:1-copied 15ips open-reel tapes. A special switch box enables me to adjust the bypassed signal to a comfortable listening level, while the preamp's volume control is then used to match its level to that. Other components used for my FET-10 tests included the Ortofon MC-2000 cartridge with its own transformer, the Well-Tempered Arm, SOTA Star Sapphire turntable, Threshold SA-1 power amplifiers, the Sony ES CD combo, and Sound Lab A-3 loudspeakers, with acoustical treatment of the room by ASC Tube Traps (footnote 1).
First, the high-level ('10/H) controller. The most easily assessible of the two, this came through with flying colors. Set for unity gain, a device like this should sound like a straight wire without gain. It did. Even under unpressured conditions of leisurely listening, I was unable to tell whether the FET-10/H was in or out of the circuit. I know some of you are going to say, "Well, AHC would have heard a difference!", or some such, but while I am as aware as anyone of my responsibility to be a nit-picker, I will not deceive you or myself by fabricating nonexistent differences just so I can demonstrate how fantastically well I hear. (I do not accuse AHC of having done that; I'm just saying that I won't.) I defy anyone, on a fair test, to reliably distinguish the FET-10's high-level unit from a straight-wire bypass.
Next, I turned my attention to the FET-10/P. Upon removal of the cover, this simple-looking sealed can becomes, like the Klyne SK-5A, a worm farm. The adjustments thus revealed allow you to optimize the matching of just about any phono cartridge in existence. Unlike Klyne, though, Threshold does not challenge you to do this by ear, but suggests that you follow the cartridge manufacturer's loading recommendations. I was happy to do so, but there's no reason you have to.
Also like the Klyne SK-5A (and, in fact, like every other preamp I have tried), the Threshold is not directly usable with the Ortofon MC-2000 cartridge. As usual, the problem isn't lack of gain, but excessive hiss. I have yet to find a substitute for Ortofon's own T-2000 step-up transformer for use with this cartridge.
Bypass tests on the FET-10/P preamp unit were no more rewarding than those on the control unit. The input and output were so alike that I could never be certain which I was listening to, without checking the setting on the bypass switch. Nothing seemed to change from one setting to the other, not spectral balance, texture, detail, or spatial characteristics. With a real phono cartridge instead of the bypass signal, there was not a trace of added texturing at the high end, where differences between preamplifiers most often show up.
That is not to say, however, that the FET-10/P sounds exactly like the front end of either the Klyne SK-5A or the Audio Research SP-11. It doesn't, even though those other preamps also did outstandingly well on bypass tests. But how can two preamps which sound perfect on a bypass test sound different from one another? Because individually subliminal imperfections may become "liminal" when the errors are in opposite directions and compared with one another. That, apparently, is what happens here.
By comparison with the SP-11, the phono stages of both the Klyne and the Threshold have noticeably better low end. Deep bass from discs is deeper, obviously more extended, and has much better heft and impact. However, where the Klyne's bass is tightly controlled almost to the point of being dried up, the Threshold's bass is somewhat rounder, giving bass instruments (basses, cellos, organ pedals) more the kind of bloom I hear at live concerts. (The SP-11's deep-bass deficiency, interestingly, is more noticeable from a cartridge than it is from the inverse-RIAA bypass source I use for evaluating a preamp's phono stage. I cannot explain why.)
Extreme highs from the Klyne and Threshold are virtually identical: crisp without being edgy, open without being tizzy. By comparison, the SP-11's extreme highs are very slightly soft, with an airy delicacy which may be less accurate than what I hear from the other two preamps but is, at least with electrostatic speakers, a little closer to the sound of live musical highs.
In terms of detail, the Threshold is perhaps a shade better than the SP-11 but not quite the equal of the Klyne. Since I do not value detail above all else, I did not find the differences between the three to be worth much weight, but others are welcome to differ with that. Soundstaging and depth from the Threshold were identical to that of the Klyne; both were excellent, although neither's depth rendition was quite in the class of the SP-11. I should mention, however, that there are those who feel that the SP-11 exaggerates depth. So be it—a little exaggeration does no harm at all, and sounds nice.
So, what did I think about the FET-10's sound? I loved it. It is exceedingly, persuasively musical, in the best sense of that word, and a delight to listen to. The fact that I find the SP-11 (with my associated equipment) just a shade more seductive, despite its very slight imperfections, does not in the slightest diminish my regard for the Threshold. Having, at the time of writing, listened to the FET-10 for about 30 hours, I still hear nothing about it that would bar it from a top Class-A recommendation. The Threshold FET-10 pre-trol is possibly the best such in the world; it is almost certainly the best buy in a pre-trol that you will find.
But is such a device cost-effective in the first place? A reasonable question: When any CD player with a volume control can be connected directly to a power amp, is it possible to justify $1800 for another volume control and source selector? It isn't, if CDs are all you will ever listen to. But if you want to listen to other signal sources, or do any serious tape recording, you will need a balance control and a way of juggling program sources. That means some active circuitry, with its potential for fouling the sound. If $1800 is what it costs to avoid the fouling, that's the way it is.
Congratulations, Threshold. You've got yourself a winner!
Footnote 1: I consider these standing-wave absorbers to be absolutely essential for decent LF reproduction in any room smaller than an auditorium. They are the only means for solving LF problems in a room not initially constructed with standing-wave traps in the walls. Any audio perfectionist who isn't using them is wasting a large percentage of his component-purchase outlays.—J. Gordon Holt