Ohm Walsh 5 loudspeaker Dick Olsher Returns
The folks at Ohm would love to have us believe that there "is no place like Ohm." I know Dorothy believes that, but last time I listened to the Walsh 5 there were enough things wrong with it to spoil the Ohm-coming. To be sure, there was plenty of imaging magic; you might even say that, indeed, "there's no space like Ohm." But a slightly opaque and colored midrange seriously curtailed my enthusiasm for the speaker. All puns aside, I firmly believed back then that this was a speaker on the verge of greatness. So when Ohm let it be known late in 1987 that the Walsh 5 had been revised, I was naturally more than interested in a relisten.
I was unable to specifically pin down the nature or extent of the modifications, but even after a quick initial listen my impression was that the speaker had been significantly improved. The midrange was smoother and considerably more transparent. I suppose that transparency and disdain of boxy colorations are acquired tastes; years of intimacy with electrostatics have made me quite intolerant of muddled resonant boxes. The art has advanced greatly in the last 50 years, yet dynamic moving-coil manufacturers are still boxing box resonances. Heroic enclosures like the one used in the Celestion SL600 are a step in the right direction. Not surprisingly, the mini-monitors, the miniature boxes, are the least boxy-sounding because of the reduction in panel area: there is simply less radiating surface. Unfortunately, most little boxes are bass cripples. A good example of this is the ProAc Tablette. Its lack of bass and gratuitous brightness conspire to give it an artificial sense of transparency which many people like. I know I liked it. But J. Gordon Holt, Mr. Tonal Balance, was appalled: "violins sound like toy violins," he would complain. But Gordon, what do you expect from a castrato anyway?
The revised 5s impressed me as pushing closer to mini-monitor clarity and transparency, the best that box speakers have to offer. Another way to put this is that while they're not in the class of the best planar designs in these respects, they are quite respectable.
Shortly after the revised 5s arrived in Santa Fe, I was visited by John Strohbeen and Don Bouchard of Ohm. Mr. Strohbeen insisted that all of the absorptive panels be removed from my listening room so as to convert the dead-end/live-end arrangement to a live-end/live-end acoustic. He felt that this would help establish a more realistic soundstage, and, based on my experience with the old 5s, I was inclined to agree. The next day proved quite frustrating as we labored to eliminate a lower-midrange "ah" vowel coloration which was painfully obvious on the Lesley Test. We experimented with breaking up the room's standing-wave modes as well as with a variety of speaker placements. What helped the most was to position two large particle-board panels in the shape of a "V" (the point of the "V" facing the listening position) between and slightly behind the speakers. This alleviated the problem but did not entirely eliminate it.
I suspected that the culprit was the room itself, particularly the wall resonances—sheetrock panels are capable of significant resonant energy output in the lower mids. If you don't believe me, simply bang your fist on the wall. What you'll hear is the sound of the wall singing over a fairly narrow resonance. And wall resonances will routinely be excited by sound impulses from the speakers. It would really be nice to have a resonant-free listening room on a concrete slab and with concrete or stone walls. However, for most of us stuck with less than the best, the only solution is to control the resonances through sound absorption.
The first step was to reintroduce the fiberglass side panels into the room. The impact of this was quite dramatic: the lower-mid colorations were almost completely squelched. To be sure, the expansiveness of the soundstage suffered, but the spatial extension of instruments tightened up. The increased focus was actually more to my liking, compared with a wider but overblown stage.
Next, I removed the felt pads from the back of the speakers' "cans." The pads are provided to control rear dispersion and are probably useful in controlling undesirable reflections from wall placements. But when the speaker is removed from the rear wall, the pads serve no useful purpose and appear to hinder rather than help. In fact, without the pads the sound became more spacious and transparent—as though a muffler had been removed from the reproducing chain. Replaying Laudate! (Proprius 7800), it was also obvious that midrange resolution had improved. For example, it was now possible to better resolve hall reverberation.
Finally, the particle-board panels were removed and the rear absorptive panels were repositioned so that the room was configured as normal. The soundstage moved a few feet back, but I was left with very little in the way of timbral errors to criticize. Lesley's tonal quality was now right on. Cleo Laine's vocals (Live at Carnegie Hall, RCA LPL1-5015) had also markedly improved compared with the old 5s. Comparing the two versions in the same room and under the same conditions, the old 5s sounded opaque and recessed. The midrange textures of the revised 5s were simply purer and much more transparent. However, the upper mids were still somewhat laid-back—even with the presence switch all the way up or in the flat position.
At least with the speaker removed from the back wall, the region between about 1.5 and 3kHz measured and sounded 2dB down. This valley was even broader on the old 5s (I failed to call attention to this previously). With a suitable amp, however, this need not be a subjectively bothersome effect. The Cochran Delta Modes highlighted the tonal-balance deviation quite readily, but with the Wingate 2000A Series III the 5s exhibited plenty of midrange bravado and verve. The Persuasions (No Frills, Rounder 3083) virtually came alive in my listening room. So please note that the sonic impressions noted above and below are based on using the speaker with the Wingate amplifier.
At least once a year JGH gets on his soapbox about the importance of recommending amps and speakers as a combination because of the large potential for synergistic or antagonistic interactions. Class A speakers and a Class A amp chosen at random do not necessarily add up to Class A sound. It appears to me that many audiophiles have had a hard time accepting this. Here is a perfect example of this: the Wingate complements the Walsh 5 exceptionally well, while the Class A Cochran Delta Modes fare far worse.
Two additional minor problems require elucidation. The first one bothered JGH more than it did me. JGH, who had lived for a long while with the old 5s in his video room, also participated in the initial listening sessions with the revised 5s. He complained of a slight thinning out of the lower mids, as in tuba timbre lacking sufficient heft and body. John Strohbeen did not contest this point at the time, so it was not entirely surprising when Ohm sent along two new cans just as I was finishing up with the revised 5s. These cans were experimental to the extent that they contained two crossover networks. A switch on the back of the cans allowed me to change from the old to the new network, which supposedly provided more of a boost in the lower mids.
The extent of the boost turned out to be about 2dB from 150 to 300Hz. Certainly, with the new network the sound was, as expected, fuller and mellower. The surprise was that it was also a little smoother and cleaner through the mids. Because the new network is a step in the right direction, I presume that it will be standard equipment in all future production.
The final cavil has to do with the upper octaves. As good as the 5s had become, they still failed to match the sweetness and treble transparency of my electrostatic references, the old and new Quads. The treble of the 5s was consistently slightly grainy when compared with the best. But keep in mind that I've really never heard a dome tweeter that could hold a candle to the Plasmatronics helium tweeter, or, for that matter, even the Dukane Ionovac tweeter.
As was the case with the old 5s, I was unable to achieve very smooth deep-bass response from the new 5s. This is not due to any fault of the Ohm speakers, but to the generic problems of bass reproduction in a small room. In a concert hall there are hundreds of standing-wave patterns in the bass octaves. When added together spatially, the multitude of modes smooths out the frequency response. In contrast, there are very few bass modes in a small room. It is this sparsity of modes that yields serious dips and peaks in the room response; any full-range loudspeaker will have a tough time in a small room. And, other than liberal tinkering with an equalizer, which is a can of worms in itself, bass-heavy speakers and small rooms don't mix well.
In my room, the best bass extension for the Walsh 5s appeared to be associated with rear-wall placements. However, I chose not to do this because of the penalty to be paid in terms of imaging. In order to test the bass of the 5s under favorable room conditions, they were set up downstairs in JGH's listening room. Positioned against the rear wall and driven by the Threshold SA-1 monos, the Walsh 5s' bass was excellent in terms of extension, dynamic power, and definition. I remember JGH going gaga over the bass quality from several of his CDs, and declaring the bass reproduction as the best he's ever heard in his listening room. There was clean output into the low 30s, and our biggest problem was clipping the Thresholds on sustained low-frequency passages. Again, however, soundstage focus and depth did suffer from wall placement.
So the bottom line is quite favorable: the Walsh 5 is a full-range speaker that is quite clean and images very well. On the debit side, it is quite room-sensitive and will require considerable experimentation to achieve a proper balance between imaging specificity and bass extension. Midrange transparency and clarity are not quite in the class of the best planar designs; neither is the treble as smooth and transparent. But to their credit, the revised 5s do manage to achieve very low levels of boxy colorations, and compete very well with the best dynamic designs money can buy. Soundstaging is another strength and joy, and at their best the 5s can set up a very palpable illusion of the original performing space. With wall placement, the bass, as mentioned earlier, is deep and powerful.
Choice of amplifiers is critical. The best choices would be slightly forward designs that tend to complement the 5s' slightly recessed upper mids. Because the 5s need lots of current drive into a 4-ohm nominal load and a slightly forward-sounding amp, solid-state designs naturally come to mind. The Wingate 2000A certainly worked extremely well, but I would imagine that many other solid-state designs would also work well.
With the above caveats, the revised Walsh 5 with the new crossover network (I certainly hope that Ohm offers these mods as a retrofit to owners of old 5s) strikes me as a GAS: a Great American Speaker—one of the best American box speakers made, a clear Class B choice, and a clear vindication of the Lincoln Walsh driver technology. In its present incarnation, it is one of the few dynamic speakers that my jaded electrostatic taste buds could live with.—Dick Olsher