Wilson WATT Series 3-Puppy 2 loudspeaker Page 2

The final configuration David chose for the Puppy was to use two 7.5" Dynaudio woofers—units with large magnets and large-diameter voice-coils—in a ported enclosure, the small, 2.75"-deep, oval vent being on the unit's rear. Like the WATT, the Puppy's front baffle is covered with a rubber blanket, to minimize reflections of the drive-unit outputs from the cabinet edges, with a cellular-foam grille held on by Velcro. The cabinet is made from fiberboard, with a complex internal construction using multiple crossbraces to maximize rigidity in the Puppy's intended passband. The 24" height of the Puppy is intended to place the WATT at the optimum listening height, and three small adhesive-backed pads on the Puppy's top surface provide appropriate midrange decoupling. (Care should be taken in moving the WATT when placed atop the Puppy, as these pads tend to slide around.)

Four sets of 5-way binding posts can be found on the Puppy's behind: one pair for normal connection to the amplifier; one pair for the necessary high-pass filtered output to drive the WATT; a third for direct connection to the woofers if the owner wants to bi-amp the system with an external electronic crossover (Wilson Audio prefers the normal passive mode); and a final pair to hold a jumper which must be in place for normal use. A short length of MIT cable—the Puppy "Tail"—is supplied to connect the WATT to the Puppy and passes through an internal channel to emerge on the Puppy's top surface.

The Unger quotient
This review actually began last Fall, at that time of the earlier WATT 2 and Puppy system. Wilson Audio had supplied both a pair of Puppies, finished in black laminate, and a pair of WATT 2s (serial numbers 963/964), finished in gloss black paint. For a number of reasons, I was late starting the review and didn't get around to setting the speakers up until December 1991. My first impressions were favorable. But then at the 1991 WCES in January, Wilson Audio announced the Series 3 WATT, which was said to feature an improved midrange/woofer and a revised crossover. I therefore put the review on hold while I waited for the WATT 3.

Finally, not only did a review pair of WATT 3s arrive (serial numbers 1243/1244), beautifully finished in a gloss gray-painted finish (footnote 5), but so did Wilson Audio's David Wilson and Tierry Budge, to set them up in my listening room. I had already approximately determined the optimum positions for the Puppies in my room, playing with the distance to the rear- and sidewalls to balance the ultimate bass extension against mid- and upper-bass definition. But to watch David and Tierry find the ultimate positions for the two WATT/Puppy combinations was to witness an exercise in attention to detail. They stayed with my approximate 33" placement with respect to the rear wall, but moved each speaker laterally until they were happier with the overall bass/midrange balance of each. This gave slightly asymmetrical sidewall-speaker distances of 58" (left) and 51" (right), with each speaker toed-in a little toward the listening seat so that the listener could just see each WATT's inner sidewall. A tape measure was then produced to move the seat until left and right ear-to-speaker distances were identical.

Now came that part of setup that even the Odd Couple's Felix Unger could not fail to be impressed by. Having determined where the speakers should be to within a matter of inches, David and Tierry affixed the spikes to the Puppies, then laid out masking tape on the rug around each Puppy's perimeter, marking the tape at the front with ½" spaced lines. For left and right speakers in turn, David would then listen to a track in mono from one of his piano recordings (Ragtime Razzmatazz) a number of times, with that speaker moved by ½" between each audition. This recording has a rich left-hand register, giving a full spectrum in the frequency region where the distance between the speaker and the sidewall results in a characteristic comb filtering. By listening to the manner in which this comb filtering interacted with the sound of the piano's lower midrange, David was able to zero-in on the exact position giving the most musical midrange balance, a subjective process he calls "voweling." The final setup stage for each speaker was to compare the two best positions, the marks on the tape being essential in order to repeat identical placements in this process.

The entire setup took about an hour of intense concentration on the part of David and Tierry, and at the end of it I was keen to hear recordings I knew well. But I have to say that, even before it began, the WATTs and Puppies were producing a superbly musical sound. The iterative "voweling" process merely determines the apex of the speaker's performance curve. But the broadness of that curve means that this is a speaker system that will always sound good, even when not perfectly set up. But that extra care is well worthwhile. Isn't the excellent the enemy of the good?

A word is in order on the WATT handbook. Comprehensively illustrated with the subjective changes to be expected from different room placements and treatments, the book offers excellent advice for any speaker, not just the WATT. Unusually, contour maps for both horizontal and vertical planes (in gauss) are given for the WATT's stray magnetic field, something I have never seen before.

dB or gb?
Before I get down to describing the WATT's and Puppy's musical merits, pink noise reproduced by just one loudspeaker sounded extremely smooth throughout the midrange and treble at the 2.5m listening position, with just a slight degree of HF emphasis and a rather loose upper-bass region. Vertically, the WATT was perhaps best-balanced with the listener's ears at or just above the tweeter axis; ie, some 37" from the floor. Above that, a peak in the mid-treble can be heard; sit so you are at WATT woofer level and the midrange becomes very peaky.

The WATT sitting on the Puppy passed the knuckle-rap test with flying colors—it was like nothing so much as tapping a rock. Nevertheless, a check with a stethoscope revealed that the region of the WATT's side-panels level with the rear tubular alloy handle were not as nonresonant as the rest of the enclosure, giving some output in the 300–400Hz region. Subjectively, however, this is very likely irrelevant. Though the Puppy cabinet is more lively than that of the WATT, what resonances it does have are well above the region where it is handling any signal, making it unlikely that they will be excited to any significant extent.

But no matter how useful pink noise and test tones are in checking out what a speaker may be doing wrong, it is its performance on music that is all-important.

The first record I reached for was a test pressing of Stereophile's Intermezzo LP. Canadian pianist Robert Silverman turned in a world-class, risk-taking, goosebump-raising performance of this Brahms sonata, particularly of the second Andante movement, where he makes time stand still. The Steinway in the Santa Barbara church had had a majestic character to its sound, coupled with a superbly even note-to-note balance. Both aspects came over superbly well on the Wilson system. The piano sounded correct, both tonally and dynamically.

From my experience of similar units in other speakers, I had worried that the WATT's use of a Focal tweeter would have rendered the upper register of the piano too bright. But this wasn't the case; the sound was smooth without being mellow, detailed without being thrust forward at the listener. Evidently the extensive modifications that Wilson performs on this unit have a significant subjective effect. There was a slight emphasis of LP spits and pops, and tape hiss was a little more audible than usual, but this excess of top-octave energy was not musically important.

I said that the piano sounded tonally correct. Well, that wasn't entirely accurate, as a thickening in its tenor register was present. Now, my room does have a problem itself in this region—I couldn't get the Thiel CS5's mid- to upper-balance correct either—but Sam Tellig also felt in our April issue that the WATT/Puppy had an audible "hump" in this area. While this hump wasn't musically unpleasant—and there certainly wasn't any tendency for notes to overhang—the slight excess of energy in this region does represent a departure from strict neutrality. (I understand that some audiophiles have reversed the Puppy's polarity in an attempt to control the hump. David Wilson points out, however, that this severely disturbs the coherency of the crossover between Puppy and WATT.)

Lower down in frequency, the midbass was clean, with extremely good articulation. The track "Love Letter" on Bonnie Raitt's Nick of Time (Capitol CDP 7 91268 2) starts with exposed kick-drum beats and back-beat snare-drum. While the kick drum had good weight, the manner in which its sound started and stopped was excellent. And when the nasal-sounding bass guitar joined in, the similarly pitched sounds of the two instruments did not interfere. This "quick" ability extended to timpani and classical bass drum. On a recording I made of Elgar's Dream of Gerontius in England's Ely Cathedral back in 1984, there is a cataclysmic single bass-drum stroke accompanied by a double timpani beat—the score (figure 120) is marked fff sforzando!—just before the penitent Gerontius utters "Take me away" after being ushered into the presence of God by the Angel of the Agony. When I made the recording, the impact of this climax was such that it felt like a pressure front passing your body, sucking the air from your lungs (footnote 6). I rarely hear this from loudspeakers, the recorded impulse seemingly exciting everything they do wrong in the bass. The WATTs/Puppies hung on to that wavefront, the bass drum sounding like a sharp thunderclap but with the timpani double beat still distinguishable. I have to say that I have never heard this degree of bass-transient coherence from a reflex design before.

Down in the low bass, the Puppies did run out of steam. Despite their authority in presenting the midbass power of orchestral music, they really can't deliver the full measure of the organ's lowest registers. But they do try. And coupled with the system's excellent dynamics, it was only rarely that I felt the need for that extra half-octave. But then, during the Dorian Pictures at an Exhibition transcription for organ (DOR-90117), it would have been nice to have had it.

Perhaps it was in the region of articulation, of musical dynamics, that this system impressed the most. Quiet passages were not veiled, but as this absence of smearing was not due to detail being artificially thrown forward, when the music got loud it remained unfatiguing. With the Silverman Intermezzo, for example, the big chords at the start of the sonata reached 100dB spl flat at the listening position, but without making me feel like turning down the level.

Footnote 5: The speakers are supplied with their finish protected by clear acetate film.

Footnote 6: Purists may shudder, but to capture the full dynamic range of this immense work, I used dbx noise reduction with my Revox A77. There don't seem to have been any ill effects due to this lack of high-end purity; in fact, comparison with commercial recordings, such as the in-many-ways-excellent Richard Hickox/LSO performance on Chandos (CHAN 86441-42), reveals just how much gain-riding professional engineers use to fit such a work within the CD's dynamic window.

Wilson Audio Specialties
2233 Mountain Vista Lane
Provo, UT 84606
(801) 377-2233

fkrausz's picture

fkrausz's picture


Great fun reading this post.  Seems that you haven't let yourself be this satirical in recent years.  Let me encourage you to do so whenever the muse inspires you.

John Atkinson's picture

Great fun reading this post.  Seems that you haven't let yourself be this satirical in recent years.  Let me encourage you to do so whenever the muse inspires you.

This was a fun review to write, as well as being the longest review of a single product I have ever written. Can't believe it was more than 20 years ago!

John Atkinson

Editor, Stereophile