Company Address: c/o Madrigal Audio Laboratories, P.O. Box 781, Middletown,
Conn. 06457.
Let me begin by confessing a few secrets about my reviews for Audio. I don't review for a living, I do it for fun.
As a result, I try to pick for review the kind of equipment that helps me find out how good high-end gear can sound. I could claim that I do this unselfishly and out of a noble desire to inform my fellow man. The fact is, I do it for my own pleasure. I review equipment simply to find out how close audio reproduction can come to live music, and to determine what would belong in an ideal reference system-if I could only afford it.
I have enjoyed every step in this search, but it is rare for any single piece of high-end equipment to stand out from the others, unless it performs a new function or introduces a new technology. Most high-end audio equipment has gotten so good that it is rare to find a piece of gear that is so much better than its competition that it is really exciting. This is particularly true of electronics such as preamplifiers and amplifiers. Almost all such high-end components are now at least very good, and many perform superbly in systems which are well set up.
With that said, let me go on to say that the Mark Levinson No. 26 dual monaural preamp still stands out from its competitors. I cannot promise that this is the world's best preamplifier. I'm still waiting to hear many contenders, and, given the rapid change in high-end electronics, it is impossible to predict how long any model can remain at the top. Still, I can say that the No. 26 is clearly the best preamplifier that I have ever listened to.
In fact, this Mark Levinson preamp has redefined my understanding of the state of the art in terms of both recording and home playback, and it has done so in a wide range of high-end systems and with a wide range of recorded music. It is simultaneously the most transparent and revealing preamp I have heard and the most musically convincing--at least in its ability to re-create how music sounds at the distances from which it is actually recorded. A few competing preamplifiers may produce more of a concert hall sound, but only at the price of changing the recording's original sonic perspective and some degree of euphonic coloration.
Before I heap too much praise on the No. 26's sound, I should tell you a little about its features, technology, and specifications.
To begin with, this preamp is available in a mix of configurations. You, can choose a unit with a phono board for either moving-coil input (58- or 64-dB gain) or moving-magnet/moving-coil input (38- or 44-dB gain) plus a power supply; this combination will cost you about $4,995. For about $4,750, you can choose a unit without a phono board but with a balanced-line input and a power supply. You can keep the balanced-line input, add a No. 25 external phono preamp, and use one power supply for both the main and phono preamps; this will cost about $6,350. You can buy separate power supplies for the main and phono preamps; if you also buy the balanced-line input card, you'll pay about $6,575.
If this sounds a bit complex, let me hasten to say that I found the "stripped down" No. 26 with a moving-coil phono board and a single power supply to offer a reference-quality package with all the features I really need but a mini mum of extra boxes and interconnects.
I have to admit, however, that the balanced-line input can make an audible improvement with those few CD players and tape decks that have balanced outputs. Further, a separate power supply for the phono unit does slightly improve the dynamics and S/N ratio with a moving-coil cartridge. You will have to make your selection based on your own needs.
In any case, the main preamp unit is nicely styled in black and muted silver, and has a distinctive enough sculptured front panel to rise above the standard black box, providing the kind of "feel" that its high price tag deserves. The front-panel features are particularly well chosen. The "Input Selector" has settings for "CD," "Tuner," "AUX 1," "Phono/AUX 2," "Tape 1," and "Tape 2." There are enough tape-switching features to allow you to easily use both analog and R-DAT recorders, plus recording switches for "Defeat," "Input," and "Monitor" and for selection between monitoring "Tape 1" and "Tape 2." That's an almost ideal set of switches for NB reviewing as well. There is an absolute-phase (polarity) switch, a stereo/mono switch, a Penny and Giles "Output Level" pot (custom-made, according to Madrigal, to track within 0.5 dB and to be good for at least 1 million rotations), and two switchable balance controls.
The interior features are equally impressive. For the moving-coil buff, there is an excellent set of phono loading options. The input switch uses high-quality relays in the signal path to preserve signal purity, and the circuit is designed so that only two switching contacts exist between the source input and the line output. Special attention is given to thermal compensation and stability. This preamp takes only about an hour to warm up, and its sound character does not change with time; drawing relatively little power, it can be left on permanently to further improve the consistency of its sound character.
Line-stage gain can be varied for each channel, which lets you set the output level for ideal compatibility with a given amp. There is no risk of having so much gain that your volume control has to stay at six o'clock or of losing apparent musical energy because the preamp does not drive the amplifier at the proper level. This feature can be of great benefit in a reference-quality preamp because many preamplifiers do not mate ideally with the gain of other manufacturers' amps. Further, the No. 26's output impedance is only 40 ohms, ensuring minimum line noise and preamp-to-amp interaction.
Self-shorting Camac input jacks are used to avoid circuit bangs and cable problems. Camac jacks are also used for the unbalanced outputs, and XLR jacks for balanced-line connections.
The XLR jacks are fed by circuitry designed to maximize common-mode rejection. I must say that I have never been able to detect any real reason for using Camac jacks instead of high-quality RCA connectors, but this seems to be a Mark Levinson "thing," and Camac jacks do help ensure that interconnects stay firmly in place. The problem is that the use of Camac jacks also forces the use of specially terminated interconnects, and these are anything but cheap.
Unlike many other high-end preamplifiers, the No. 26 is light and comparatively small. This may sound like a minor virtue, but if you really use the inputs and features of the preamp, it is very important indeed. Further, the power supply is small enough to leave some room in your cabinet or shelves and can be kept a good distance away from other signal sources and electronics.
I have rarely seen a unit which had as many technical specifications as the No. 26. Yet all the specifications have something in common: The figures are outstanding. Every distortion spec listed--all 23 of them!--is less than 0.006%, and the overload and clipping specifications should satisfy any audiophile. The S/N ratios and crosstalk specs are all excellent, and the RIAA accuracy is said to be ±0.3 dB. As for the circuit details, without reproducing the manufacturer's brochure in depth, there are several points of interest. The Mark Levinson No. 26 uses the same flat-ribbon conductors of silver-plated, high-purity, OFC copper used in Madrigal HPC interconnect cable. The circuit board is designed to place components in what the manufacturer feels are the best locations rather than in neat rows.
Like virtually all modern high-end preamps, this unit operates in the pure Class-A mode and has carefully chosen capacitors and resistors. Unlike some recent high-end preamps, however, the No. 26 places more emphasis on the power supply than on circuit features such as direct coupling (eliminating all capacitors from the signal path) or on trying to eliminate all feedback. Two toroidal transformers with Faraday shielding provide a separate power source for each of the two channels. The PLS-226 contains three power supplies: One for each channel and a third to feed the relay-control circuit. Spikes, r.f., and surges are eliminated at the a.c. inputs. Four Class-A regulators are inside the preamp chassis-one for each side of each rail. The goal behind all this sophistication: Eliminate any trace of noise in the circuitry.
Having made my obligatory bows toward features, technology, and specifications, let me return to the sound--or lack of it. One of the most striking things about the No. 26 is its combination of transparency and silence. No unit I have ever reviewed delivered the lack of perceptible noise with low-output moving-coil cartridges that this preamp achieves. Nor have I ever auditioned a preamp that provided so much musically natural information in low, medium, and loud passages and that was so totally silent and coloration-free in its high-level line stages.
In fact, it is the sound of silence that really describes this preamp. The No. 26 is almost totally neutral when inserted into the signal path. No preamp I have yet heard has been equally neutral-particularly one whose signal path goes through so many gain and control features. No preamp I have heard has provided similar gain in the phono or high-level stages without adding coloration. No preamp I know of provides so much detail and so natural a set of musical dynamics, delivers so much information in the upper octaves, and stays as musically sweet and natural. No preamp I have yet heard has provided as much detail in the bass, as neutral and musical a midrange, and so much information at the frequency extremes.
As nearly as I can determine, the No. 26 introduces less audible change in the sound than many straight-wire bypasses using relatively high-quality interconnects. In fact, this preamp was more neutral in its high-level stages than a very good passive preamplifier.
This makes it difficult to talk about the character of the bass, midrange, or treble or to describe the impact on depth, imaging, or the soundstage.
There are slight effects. Nothing inserted into a signal path is ever truly neutral, but the worst effect I found was a very slight thinning of the lower midrange and a slight shift in the soundstage which widened it at the expense of depth. Even then, I felt I was reaching to find flaws. These problems could just as easily have resulted from the fact that nothing is neutral.
All in all, after three months of extensive listening, the only warning I can give you about the No. 26 is that it can become awfully addictive. It has been a real pleasure to review. This preamp is one of those products that brings the excitement back to critical listening. It belongs on the short list for selecting the finest of reference systems.
--Anthony H. Cordesman
(Source: Audio magazine, Aug. 1989)
Also see: Mark Levinson No.23 Dual Monaural Preamp (Apr. 1988)
Mark Levinson No.20 Mono Amp (Aug. 1987)
= = = =