Tube Preamp Reviews

There is no negative loop feedback anywhere in the Seven; just the local feedback which occurs "naturally" in all cathode follower circuits. All stages but one are resistance/capacitance coupled, via fat (1½" diameter) 2µF/250V polystyrene caps, custom-made for C-J, and the associated resistors are large noninductive metal-foil types. Even in the power supply, there are no electrolytic capacitors in audio-rated circuits; the B+ filter caps are polystyrenes and polypropylenes. (B+ power-supply capacitors aren't exactly in the signal path, but they are in the circuit which supplies the voltages the tubes are modulating with audio signal. And they do affect the sound—as anyone can attest who has tried bridging small-value polypropylene caps across power-supply electrolytics.) The only large electrolytics in the Premier Seven are a pair of 4700µF ones which help to filter residual AC from the DC heater supplies.

Setting Up
Installation of the Premier Seven was not your usual lead-pipe cinch. You have to take the top off the upper preamp chassis and the bottom off the bottom chassis, then extract several foam-plastic blocks which serve to immobilize the floating inner circuit platforms during shipping. The screws on my sample were torqued down so tightly they were very difficult to remove. (In fact, one on my sample was unremovable; the screwdriver slots stripped before I got it to turn loose. I drilled it out.) Then two of those foam plastic blocks were found to be wedged under a turned-back lip of the chassis behind the front panel, making it necessary to squeeze them to half their thickness in order to get them out. Doing this was rather like trying to pinch mercury, and I was cursing a navy-blue streak before I finally tore them loose, in several jagged pieces, with a pair of long-nosed pliers. My advice is, if you buy this preamp, let your dealer prep it first.

Equipment used for my tests included the Ortofon MC-3000 cartridge in the Versa Dynamics 2.0 arm and turntable, Ortofon's T-3000 step-up transformer (for some of the tests), the Stax Quattro CD player, a Sony PCM-F1 digital tape system, a pair of VTL 300W monoblock power amps, and Sound Lab A-3 full-range loudspeakers. Audio interconnects were LiveWire Emeralds, while the audio cables were Monster M1, which I still like despite DO's put-down in Vol.11 No.7. The listening room is extensively treated with ASC Tube Traps, and program material was some of my own and others' PCM tapes, and CDs and analog discs from Sheffield, Opus 3, EMI, Bainbridge, and Reference Recordings.

Conrad-Johnson recommends "several minutes" of warmup before listening to the Seven. I allowed my sample to cook for 48 hours before auditioning, and found subsequently that, after being off for a day or two, its sound continued to improve (becoming smoother and sweeter, with greater depth) for almost half an hour after being turned back on.

The owner's manual is adequate, but not much more. Apart from some advice about absolute phase (the Seven's line section inverts it) and AC-plug polarity, there is little about how to use the preamp. There is no circuit description, and no suggestions about how to set the phono load switch, how to take advantage of the stepped level controls, or how to cross-copy between tape machines. There is, however, a warning about not setting the Record switch to the output of either tape recorder when it is recording. We are told this "may generate a howling sound." It may indeed! Or, it may generate a thunderous roar capable of doing serious amplifier or loudspeaker damage, depending on the recorder's gain-control setting.

There are two selector switches per channel on the Premier Seven. One of them (Source) selects one of the five inputs, the other (Record) allows you to feed any source to the tape recorders. This arrangement makes it possible to listen to one signal source while taping another, and to cross-copy from either tape deck to the other without changing the cable lashup. The only trouble is that, if you switch the Record selector to the input of a tape machine that is passing signal through it, its output gets routed back to its input, resulting in the possibility of fearsome feedback (see sidebar). Nor does the deck have to be in Record mode for this to happen. If its own monitor switch is set to Source and its playback level turned up, it will happen. You don't even need to have a tape deck in-circuit. Any signal processor plugged into a tape loop will do the same thing, and the feedback may be even stronger because most processing devices have no volume control; they are fixed at unity gain.

This is not sensible design. Consumer products should be bulletproof, not booby-trapped. That one-sentence warning message in C-J's manual can only postpone the day when a Premier Seven owner blithely flips the Record switch from Tape recorder 2 (turned off) to CD, passing over Tape recorder 1 (turned on and up) on the way.

What amazes me is that C-J has been using this Tape circuit in every one of their preamps since as far back as I can remember (I've been railing at it for just as long), but apparently no-one has ever complained. Or suffered from a coronary brought on by the ear-shattering roar that this design idiosyncracy can generate.

The separate volume controls were not too hard to get used to, but I'm just lazy enough to be irritated at how complicated they make the usually simple act of changing volume. In order to maintain channel balance, you must either count the detent steps as you rotate the first control, then repeat that number for the other, or you must peer at the LEDs to ensure that both rows are giving the same indication. The very small lettering next to each LED means you may have to peer quite closely, but level matching is made easier by the fact that the arrays are one above the other; when channel gains are matched, a vertical pair of the indicators will be lit.

I was not delighted with C-J's choice of attenuator increments. Not only were the 2dB steps quite unsubtle, they were large enough that there were many occasions when I felt that the "right" volume would have been halfway between two of the steps provided. The choice had to be between a bit too loud and a bit too soft. With a detented volume control, you can get intermediate levels by maneuvering the detent ball onto the top of the hump between two detent stops; the potentiometer wiper will then be at an intermediate position on the resistor element, and friction will usually keep it there. You can't do this with a switched attenuator.

The level-switching increment problem becomes worse when using high-gain power amplifiers, very sensitive loudspeakers, or unusually high-level signal sources. With my slightly-less-sensitive-than-average Sound Labs, and unusually sensitive VTL amps, much of the time I was operating the level controls dangerously close to the point where the level steps start incrementing by 3dB or more. The Stax Quattro pushed it over the edge, its 2.1V output (with no level control) forcing the controls back to the point where the steps were between 4 and 6dB in magnitude (footnote 2). This just won't cut the oleo. What the Premier Seven needs is another control in each preamp unit, perhaps a simple pushbutton like its present Mute switch, that would reduce line-section gain by 10 or 15dB. Or, provide 10dB more available (switchable) gain in the phono stage and 10dB less in the line section.

The Sound
The recent crop of top-priced tubed preamplifiers, from such firms as Audio Research and Counterpoint, has redefined the meaning of "tube sound." No longer can we generalize that tubes sound warm, fat, glassy-bright through the upper midrange, and soft (sweet) at the top. Lately, tubes have been sounding more and more like the best solid-state, and vice versa. The Premier Seven is another one.

It is, in fact, about as close as any tubed preamp has come to being perfectly neutral in sound—in nearly all respects. The high-level section has a gorgeous top end—effortlessly sweet, open, and airy, yet immensely detailed. Nothing I have heard has bettered its high end, although some have equalled it. Resolution of detail is excellent across the board, as are (almost naturally) all aspects of soundstage presentation. So phase-coherent is it that it can present a very specific, tightly bunched image, and also produce convincing images 'way out beyond the locational limits of the speakers when suitably phased signal is delivered to it.

There is no forwardness and no tubelike "glare;" the midrange is an essentially perfect replica of whatever you feed into the preamp. Only at the low end does the Premier Seven stray from Ultimate Truth, with a slight thinning-out of bottom which sounds like a gradual frequency-response rolloff setting in at 50Hz. It's a fairly subtle thing, but even a slight error in this direction will be less well received by most audiophiles than would an equal amount of low-bass exaggeration. (Of course, there was no point in running a frequency-response test; the high-level sections of all current high-end preamps measure perfectly flat down to the 20Hz limit of my signal generator. And please do not write to suggest that I could use CD test discs as a signal source, because my dB meter rolls off below 20Hz anyway.)

---

Footnote 2: The VTLs are approximately 6dB more sensitive than Conrad-Johnson's Motif MS100 power amplifier. However, according to Bascom King's measurements in the August 1986 issue of Audio, they are only 3dB or so more sensitive than the C-J Premier Five and, according to C-J's own figures, less than a dB more sensitive than the C-J MV50. It seems reasonable, therefore, to criticize the Premier Seven on this point.