Harbeth P3ESR loudspeaker Page 2

Rather than reaching for Mahler symphonies, I found myself listening to recordings of small numbers of instruments and with a fairly restricted dynamic range—such as our December 2009 "Recording of the Month," Anouar Brahem's The Astounding Eyes of Rita (CD, ECM 2075). Even so, the grumbly bottom-octave notes of the double bass in the middle of the slow movement of Brahms' Symphony 4, in the recording by Carlos Kleiber and the Vienna Philharmonic (SACD, Deutsche Grammophon/Esoteric ESSG90018) were satisfyingly audible, if somewhat miniature in scale. Similarly, while the softly struck bass drum in the title track of Daniel Lanois's Shine (CD, Anti- 86661-2) didn't have its full weight, the Harbeths reproduced enough of it to carry the musical load. The opener on Shine, "I Love You," with Emmylou Harris, begins with a synth bass line that rocks between two notes an octave apart. While these notes weren't as well differentiated as they are with full-range speakers, they still had a sufficiently satisfying "phat" quality through the Harbeth, even if the distinction between the octaves was a little blurred.

In this respect the P3ESRs benefited from the tighter low-frequency control of the Classé CTM-600 monoblocks, the Simaudio Moon Evolution W-7 sounding less well defined. But the Harbeth's upper bass retained its definition even with the tubed Balanced Audio Technology VK-55SE. By comparison, my 1978 pair of original Rogers LS3/5As sounded overripe even with the Classés. Those vintage speakers also sounded significantly more nasal in the upper mids, and more spitty and coarse-grained in the treble than the Harbeths. My review samples of the earlier HL-P3ES-2 have long since been returned to the distributor, but I did note in my 2007 review that that model's presence region was "a touch exaggerated, and its lower midrange was not quite as transparent as it was at higher frequencies."

I didn't get that impression with this new Harbeth. Even without the grille, the highs were smoothly balanced, with no region sticking out. The balance was a touch on the warm side, but without any obvious midrange coloration; both the male voices on my 2008 Cantus recording, While You Are Alive (CD, Cantus CTS-1208), and Mary Chapin Carpenter's contralto on her new album, The Age of Miracles (CD, Zoë 01143-01133-2PE), sounded superbly natural. And the characters of the bass clarinet and bass guitar, when the two instruments were played in similar registers, were well differentiated on The Astounding Eyes of Rita.

Stereo imaging was stable and accurate. Ysaÿe-Kreisler-Bach, Arturo Delmoni's 1989 excursion into music for unaccompanied violin (John Marks JMR14, available as a premium gold CD from our e-commerce page), was reproduced with the violin placed within a warm, supportive, expansive acoustic. Changing from the solid-state Simaudio W-7 to the tubed BAT VK-55SE amplifier gave even more space behind and around the violin, but the top octaves now tilted up a little, which actually worked better with naturally made recordings such as this.

The obvious rival to the Harbeth is the almost identically sized and priced Spendor SA1, which I reviewed in August 2009. I no longer had these speakers to hand, but from my listening notes I would venture that they are somewhat more mellow-balanced than the Harbeths, within a similar performance envelope. The SA1's low frequencies also had a little less weight. A system that is somewhat on the bright side, or that doesn't allow the speakers to be placed well away from room boundaries, might work better with the Spendor.

I used my high-priced reference system for much of the preliminary auditioning of the Harbeths—it's important, when introducing something new, to change no other components. But eventually, when I'd gotten a handle on the P3ESR's balance, I switched to a more real-world rig. This comprised Peachtree's iDecco 50Wpc integrated amplifier ($999; review underway) hooked up to the speakers with Cardas Neutral Reference cables, and with my 160GB iPod Classic sitting in its dock.

This is the system I'm listening to as I write this review, with Richie Havens performing The Who's "Won't Get Fooled Again" in his inimitable fashion, via an Apple Lossless file ripped from his Nobody Left to Crown (CD, Verve Forecast B0011631-02). Yes, indeed! Haven's slightly lispy delivery, frantic, bar-chorded strumming, and the obbligato cello are all reproduced in full measure, forcing home the point of Pete Townsend's lyrics. Harbeth's P3ESR is one those speakers that gets the overall balance of what it does very right. If this system were to be placed behind an acoustically transparent curtain, few who heard it would believe that they were listening to a system costing less than $5000, or to speakers as small as these. Those listeners definitely would get fooled again—and to the benefit of their enjoyment of their music collections.

Oops
Saturn must be ascendant or something: Of late, Stereophile has had reliability problems with several review samples, and the Harbeth was no exception. Toward the end of the review period, I was giving the speaker a workout with some high-level rock when the soundstage lurched to the right and the balance became bright and brassy. "What the . . . ?"

I turned down the volume, checked all the wiring, and played the track again. All was again as it should be. Again I turned the volume up, and again it all went wrong. "What the . . . ?"

I remeasured the quasi-anechoic response of both samples on their tweeter axes. At continuous levels below about 7V RMS (equivalent to 8W into the Harbeth's 6 ohm impedance), the responses of both speakers were the same as they'd been when I'd performed the formal measurements. But at sustained levels above that, the right-hand speaker (serial no. 0472R) developed a severe peak in its low-treble region. This peak disappeared when I backed off the level. I described this in an e-mail to John Marks, who had auditioned these samples before shipping them to me, but he had heard nothing such as this amiss with them.

All I could conclude was that, at the high power level, a crossover component had either opened or shorted, detuning the speaker's balance. Reducing the level apparently brought things back to normal, both with listening and measuring, but this problem suggests that the crossover may be inadequately specified.

Summing up
The Harbeth P3ESR is one of those rare audio components that, within its obvious limitations, gives no other indication that it has been compromised. Yes, I did much of my auditioning with the budget-priced Peachtree iDecco integrated amplifier, but before the accident happened with the right speaker, I was using the Harbeths with the dCS Puccini SACD player and DAC ($22,000), the Simaudio Moon Evolution P-8 preamp ($12,000), and Classé CTM-600 monoblocks ($16,000/pair), all connected with expensive AudioQuest Wild cables. The P3ESRs did not sound outclassed in this system, merely restricted in loudness and bass extension.

I love the Harbeth P3ESR. I think it's the best iteration yet from any manufacturer of the BBC LS3/5A minimonitor concept.

COMPANY INFO
Harbeth Audio Ltd.
US distributor: Fidelis AV
14 E. Broadway (Route 102)
Derry, NH 03038
(603) 437-4769
ARTICLE CONTENTS

COMMENTS
claud's picture

Might the apparently upgraded P3ESR-XD merit remeasuring of its off-axis response?
https://harbeth.co.uk/p3esr-xd-loudspeaker/

hemingway's picture

I would also be very curious to see how the XD update has affected measured and subjective performance (especially measured).

X