The Carver "Silver Seven" Clones: Who Needs Tubes for the Tube Sound? (Issue No. 14 Summer through Winter 1989-90)

Carver M-4.0t ; Carver "Silver Seven-t"

Whether the shamans and vested interests of the High End like it or not, these solid-state Carver "t-mods accurately (and without magic) replicate the sound of the $17,500/pair "Silver Seven" tube amp.

The subject of Bob Carver's controversial t-mods, and of his amplifier design philosophy in general, is far too involved to be rehashed and reargued every time we write a new Carver amp review. We refer you to Issue No. 10, pp.

32-44, and Issue No. 11, pp. 28-29 (the premiere review of the Silver Seven), where all the necessary background in formation is presented without the knee-jerk anti-Carverism of the ultrahigh-end establishment.

Carver M-4.0t Carver Corporation, P.O. Box 1237, Lynnwood, WA 98046.

M-4.0t Magnetic Field Power Amplifier, $799.00. Tested sample on loan from manufacturer.

This review should have appeared in the last issue, where we ended up running only a capsule summary on the M-4.0t for lack of space. (See Issue No. 13, p. 56.) Mean while Carver's clock has been ticking faster than ours, and theM-4.0t is about to be phased out in favor of the virtually identical but cosmetically slicker TFM-42 and TFM-45. No matter; the M-4.0t is the original Silver Seven t-mod, and all qualitative statements about it are true of the entire family it engendered.

As we explained in our original Silver Seven review, the whole point of that somewhat grotesque though brilliantly successful exercise in vacuum-tube cultism was to provide a model for a widely marketable solid-state clone like the M-4.0t. According to Bob Carver, the earliest Sil ver Seven (the one we had reviewed) and the final proto type of the M-4.0t produced a null deeper than --60 dB when bridged together; since then, however, the Silver Seven has undergone certain minor modifications, whereas the M-4.0t has remained unchanged, so that the null today is almost surely not as deep. That, of course, does not make theM-4.0t a less "good" amplifier, since the null by itself is not a figure of merit; the important thing is that the Silver Seven, in both its earliest and current versions, and the M-4.0t are different executions of the same conceptual power-amp architecture and sound like fraternal, if not identical, twins.

Being among the very few who have heard the carly and the current Silver Seven, as well as the M-4.0t and the latest-and-greatest Silver Seven-t (see below), we can unhesitatingly report that they all sound essentially like one and the same amplifier, although we had no opportunity to run a round-robin of ABX double-blind listening tests, which might conceivably have revealed differences in sonic freckles and eyelashes. Those who understand the principle of the Carver t-mod (as Stereophile appararently does not or would rather not) realize that there is no trick or magic to the creation of such soundalikes; the surprising thing would be if they sounded different because after the t-mod there is no physical mechanism whereby a difference could occur.

That said, we must state that we disagree with Bob Carver's choice of the Silver Seven's transfer function as the model for his new generation of amplifiers. The Silver Seven probably-indeed, almost certainly-represents the ultimate tube sound, but the ultimate tube sound is not the ultimate sound (i.c., absence of amplifier sound), at least not in our opinion. Our theoretical block-diagram ideal is a block with near-infinite input impedance, near-zero output impedance, and wideband gain with near-zero distortion.

The transfer function of the Silver Seven, or of any other audiophile tube amplifier, differs from that ideal in several respects. The output impedance is typically well over an ohm; there is a light sprinkling of benign second-harmonic distortion (say, 55 to -60 dB) throughout the audio spec trum; and the distortion at the extremes of the spectrum tends to rise somewhat abruptly with level. The higher impedance drive can be expected to result in frequency response ripples at the amplifier/speaker interface, some of them in excess of 1 dB; the tweeter range will tend to be slightly rolled off; the Q will rise at the woofer resonance;

the entire sonic presentation will have a marginally softer, rounder, warmer quality than with near-zero source impedance. We feel that this kind of sonic formatting should take place at the discretion of the record producer, whenever he considers it desirable; however, those who like to have it permanently built into their power amplifier are entitled to their preferences-chacun a son godt, as Prince Orlofsky says. We, 100, are seduced by it from time to time.

Once you accept its very slight leaning toward the vacuum-tube personality-still far short of the C-J type of sound, for example-the M-4.0t is a super amplifier, and what makes it so is primarily the Carver magnetic-field

power supply. We never cease to marvel at this design, one of the very few genuine inventions in amplifiers, which multiplies voltage/current output per cubic inch-and per dollar-by a factor of 4 or 5 or better, in comparison with conventional power supplies. In this admittedly extreme case, the Silver Seven is 22 times the price and 13 times the weight of the M-4.0t, yet the latter is very nearly as powerful, especially when the load is in the normal range of 4 to 8 ohms. (The output transformer of the Silver Seven gives it a tremendous advantage, of course, in the range of 1 to 2 ohms.) So much clean power in a small, lightweight, low cost package will make anyone question the value of what ever alternative choices the marketplace offers. We must also point out the complete absence in the M-4.0t of the slight commutator noise that used to be a marginal flaw of earlier versions of the power supply. The amplifier is dead quiet, mechanically as well as through the speakers.

On the test bench, the M-4.0t exceeded by a consider able margin its basic 375/37S5-watt specification. Into 8 ohms, with both channels driven, clipping occurs in the upper 400's at most frequencies except the very highest. At 1 kHz you can nudge 500 watts per channel. With only one channel driven, these numbers go way up. Into 4 ohms, with both channels driven, the numbers go up but do not double; the power supply has its limits (and most houses have 20-ampere circuit breakers). The small-signal band width of the amplifier is 0.78 Hz to 81 kHz (-3 dB points);

noise measures -110 dB relative to 375 watts into 8 ohms (shorted inputs, A-weighted)-right on specs. The output impedance is 1.2 ohms, probably the most revealing number of them all.

We found absolutely nothing to fault in the sound of the M-4.0t, as long as we judged it as a tube amplifier. We heard nothing unpleasant or unmusical, ever, regardless of the speaker load; we never ran out of power; in fact, there was never even a minor problem of any kind. With a few speakers, such as the first-generation Carver "Amazing" (not the one reviewed in this issue), the amplifier's small deviations from utter neutrality actually proved to be synergistic. The Carver speaker had a wee bit more snap and sparkle with the Carver amplifier; obviously one zigged where the other zagged. Bottom line, though-what other clean, musical, almost 500/500-watt stereo amplifier is available out there for $799? (All right, for $839, which is the price of the new and no different TFM-42.) Carver "Silver Seven-t"' Carver Corporation, P.O. Box 1237, Lynnwood, WA 98046. Silver Seven-t Magnetic Field Power Amplifier (monophonic), $1000.00 ($2000.00 the pair). Tested samples on loan from manufacturer.

There are two ways to look at this new departure in Carver electronics. One is to say that it is basically nothing more than a double M-4.0t. It uses the same circuit boards, slightly modified to fit; it has the same power supply, only double strength; it has the same output configuration, but with twice the number of output transistors per side. It could be argued that a mono-bridged M-4.0t is the same amplifier, except for the upward shift in the impedance matching characteristics of the bridged configuration. All that is perfectly true-and a good thing to remember should some exquisitely subjective know-nothing assert in print that the two amplifiers sound totally different. But the other way to look at the Silver Seven-t is to recognize that it is Carver's first all-out bid for the high-end religionist's dollar and as such far more threatening to certain vested interests.

The difference is in the packaging. The M-4.0t looks like- any other Carver amplifier. The Silver Seven-t echoes the sexy tube look of the big Silver Seven, with an old fashioned round analog meter dominating the distinctive slanty-blocky mono chassis. The meter measures output level in dB (0 dB = 575 watts) and has four different scales for those who like to play Mr. Spock. The metalwork of the chassis is very nice; the $2000 seduction of the high-end customer is very convincingly orchestrated. We could name half a dozen very high-end amplifier manufacturers who may not be sleeping well as a result.

The Silver Seven-t is the most powerful amplifier we have ever reviewed. It delivers even more clean power into the load than the early Silver Seven, although Bob Carver claims that the present version of the latter, which we have not measured, puts out a few more volts than the t-mod but not more current. The Silver Seven-t is capable of over 50 amperes peak current into low-impedance reactive loads.

To mess with such voltage and current levels for more than a few seconds without a specially equipped lab bench is a bit hairy, so we must be somewhat general in reporting measurement figures. Clipping level into 8 ohms is in the vicinity of 600 watts at most frequencies (well over that at 1 kHz); into 4 ohms we are talking about the 900-watt neighborhood (over 1000 watts at 1 kHz); into 2 ohms there is still an increase in power, with readings in excess of 1100 watts at all but the highest frequencies; into 1 ohm the clipping level drops back to the paltry 600's. With a clean signal just short of clipping, a load of 0.61 ohm will not trip the current limiter switch but 0.57 ohm will. Whew! THD at levels approaching clipping into either 8 or 4 ohms is in the 0.1% to 0.15% bracket at 1 kHz, rising to 0.5% or so at the frequency extremes, a very typical tube-amp profile (mostly 2nd harmonic, of course). Gain is 29.0 dB; signal to-noise is comparable to that of the M-4.0t, as is the small signal bandwidth of 0.8 Hz to 80 kHz (-3 dB points). The output impedance is a tubey 1.1 ohms.

The surprising thing is that none of the muscle of the Silver Seven-t is wasted when playing orchestral or organ CD's through big speakers of ordinary efficiency. We were quite disarmed by the happy musicality that goes with such immense reserves of clean power, tube-like character or no.

What a nice sound-not even a small premonition of strain, ever!--despite the not quite tightly damped bass and some what polite highs. Go ahead, Bob, be a tube freak.

-------

[adapted from TAC 14]

---------

Also see:

Philips Gives Some Special Fillips to Its High-End Audio and Video Line -- Philips LHH1000 ; Philips CDV488

Various audio and high-fidelity magazines