You are reading the older HTML site
The Lowthers as Midranges: Audio Heresy
I saw Al Gore's movie recently, An Inconvenient Truth, in which he talks about things that are clearly true that no one wants to address, like "Global Warming." Sometimes the film aims at greater national political awareness, sometimes for an awareness of the entire planet as one geo-political unit, and sometimes it just makes jokes. He seems to challenge the idea, "My country, right or wrong." Similarly, in audio we have a few truisms: "Bigger is Better" is one, and many folks subscribe to the corollary of that axiom, "More Expensive is Better," both variants on the post-Einsteinian epistemological axiom, "More is More."
If you've read my befuddled rant about my grandmother's insistence that in her kitchen we were ankle-deep in electrons; or my little introduction to "Bullshit" in my Palladium cable review; you'll know that I'm sometimes up to taking on the "generally accepted truths," and the "best informed opinions" of audio. So, of course, I'm ready to take on the truism that Lowthers are full range loudspeakers. To begin with, they surely do not make the deepest lows below 40Hz (even in their large factory-built horns—even in Pass's larger than life KleinHorn), and they don't make high hi's very well, either. Max's first inconvenient audio truth is; "Lowthers are not full-range."
It is generally conceded that the Lowthers do have their own kind of special magic when it comes to reproducing music: they sometimes create a very good illusion of in-the-room musicians on the order of the older Quad electrostatic loudspeakers (similarly not full-range). They are praised by audio cognoscenti as having remarkable clarity, detail, dynamics, and the ability to capture small nuances that are "down in the mix." The well-respected Nelson Pass has designed at least two dedicated amps, his F-1 and his F-2 (the latter a "single-ended, current-source" amp), for the Lowthers. [See Art Dudley's piece; "Nelson Pass & the First Watt F2 amplifier;" Stereophile, Dec. 2005, p. 45, for more on the F-2.] The similarly well-respected Dick Olsher has designed a three-way loudspeaker named BassZilla around the Lowther as a midrange between a 15" woofer and a high output tweeter. [See http://store.hifiauthority.com/olsherkits.html for photos and description of BassZilla.] Nelson Pass reports he also gets his best results using the Lowther as a wide-band midrange. [See http://www.passlabs.com and find in his "articles" section his work on the Mount Rushmorean KleinHorn (parts I and II).] At least one other manufacturer that I know of has built a system using the Lowther as a midrange, the Hørning Perikles. [See Dudley's piece; "Buying Off the Rack;" Stereophile, Feb. 2006, p 41, for photo and info.] So there is a trend developing, a sea change towards Lowthers as midranges.
I seldom have an original idea, but when someone else has one I'm good at picking up the ball, to mix metaphors, and running with it. I must, therefore, 'fess up to using a Lowther as a midrange in the safe (3 ft. thick reinforced concrete walls) confines of my audio bunker beneath my little beach shack, and it is damn good—which is to say it comes close to my ideal of "great sound." Stereophile's columnist Art Dudley, and The Absolute Sound's editor, Robert Harley, speak to the ideal of "great sound" sometimes, and that ideal has many points in identity with, surprise, "The Lowther Sound"—ending with making a convincing facsimile of live musicians in the room.
"Let's see how the Lowther does as a midrange, in my system, with my music," I thought to myself one night in my insomnia. So my self popped the silky 6" Morel MW 164 midrange out of my big rig, and I popped in the more detailed 5" Lowther C45. It wasn't as much of an ordeal as I expected with the help of my audio buddy, Alan Shapiro, himself a Lowther owner. He did manage to fabricate a new adaptor plate, as it was not a direct replacement by size or bolt configuration, which isn't too tough for anyone with a table saw and a drill press, or the appropriate set of hand tools. He and I did have to tinker some with the crossover's midrange gain setting resistor, and also cludge my "notch filter" into my external crossover. By the by, Ejvind Skaaning freely admits using notch filters in crossovers in the FAQs section of his website. See http://www.audiotechnology.dk. But, in the end it turned out alright, as in "Awwll Raaght!" Maybe "AAWWLL RAAGHT!!!"
I chose the Lowther's C45 because it was smaller than my current midrange (not requiring a larger hole to be cut in my speaker baffle), not so high in output at 93 dB as the 8" Lowthers (requiring less attenuation), and considerably less expensive than the more muscular 5", DX 55 pair at 94dB/watt, or the 8" PM5A models at 98dB - 102 dB/watt (depending on enclosure) I had in the "lab." [To be most accurate: the Lowther C45 is 93dB efficient and costs $550/pr; the DX55 is 94dB efficient and costs $1095/pr; and the PM5A is 98dB efficient and costs $1995/pr.] In all, the swap was less than an afternoon's task, but tweaking the values took days.
I imagine many of you reading this are listening to three-way loudspeakers. If they're 8 ohms, and any good, they likely have a pretty good crossover with zobel networks on each driver to lower impedance peaks (to eliminate dips in output), and a phase compensator on each (to minimize time smearing). In such a 12dB network as mine, I could swap in the Lowther C45 without much fuss. In crossovers using 18dB or 24dB slopes it would be next to impossible to just change one resistor's value as there are many more parts, and each part is co-dependent on, and "talks to," the rest of the circuit. Using 6dB or 12dB slope crossovers, it would be possible but you'd have to devise a way to match the output of the woofer and tweeter by connecting extension wires to the internal crossover to try different value "gain resistors" externally. As there are no exact values to cover all existing speakers, you'd have to ear it in, trial and error. But the reward is worth it: "The juice is worth the squeeze," as some pre-Socratic philosopher once wrote. Or was that something a tweaker at the Head-Fi Meet told me?
What can anyone expect? Most of us realize that a certain class of woofers sound similar to one another. The differences between them, the difference between good, very good, and excellent is hard to describe, but we know it when we hear it. First, let's say a good woofer has to get down with authority to the lowest octaves, the region below 80 cycles is good, below 40 cycles is very good, and below 20 cycles is exceptional. Next, a good woofer has to be able to define the harmonics and textures; attack, sustain, decay ratios of the different low register instruments, to avoid muddy confusion of bass information as is likely with a lesser woofer. A very good woofer separates mud into the distinct sounds of a Fender bass, a bass drum, an electric guitar's lowest notes, a synthesizer, a Hammond B3 organ, etc., especially when all of them are jammin' their asses off together on one tune.
An exceptional woofer would help you hear enough harmonic detail and bass texture to tell which brand of synthesizer is playing, that maybe it isn't a Fender bass or a Hammond organ. On classical music, an excellent woofer will tell you at a glance if an organ has wooden or metal pipes. This information retrieval is due to control of the voice-coil, recovery speed to the neutral position, freedom from "ringing" or "hangover" of bass tones, freedom from cone "breakup," and freedom from intermodulation (IM) distortion as the vibrations ripple up and down the cone itself. You can train yourself to hear such distinctions.
There are many ways to skin a cat. A good woofer can be of the acoustic suspension design: low Q, pretty floppy, and (to keep it linear) using the air inside the sealed enclosure as a "shock absorber." These "direct radiators" are "inefficient," take more powerful amplifiers to drive them, and are most often found in systems designed for classical music. Less floppy speakers with stiffer suspensions (higher Q) often play louder, are happy in ported enclosures, perform well with moderately powered amps, and are often found in home systems designed for pop and rock. Woofers with stiffest suspensions (highest Q) are often used in various horns, can play very loudly, perform well with relatively low-powered amps, and are found in large numbers in Pro systems like those used at rock concerts. Usually, the stiffer the suspension, the larger the enclosure.
Match-making the pretty fairly efficient Lowther C45 to its best mate could be a lengthy task of taking measurements, and listening for long times to various combinations of woofers with the C45 as a midrange. Not being a speaker designer, I haven't time, lab equipment, or computer software enough to do scientifically precise readings. Being a tweaker I can say this: With Lowther C45s in my big rig (The Big Dude), I was able to achieve a nearly seamless blend (to my ears) of sound at the crossover from the woofer to the midrange.
I have known the 9" Scan-Speak woofer (21W-8555) for longer than I like to admit. It is the most recent iteration of a speaker on the same chassis, engineered by Ejvind Skaaning (or his staff), starting with the paper-coned Sen-Lab 21W-5408 which debuted in the '70s and had at least two or three iterations (I infer, since the ones I have hanging around are marked "model 2a."). Then there were the paper-coned and, later, polypropylene-coned models of the (Skaaning-founded) Dynaudio company's 21W54 that I used for a while during the '80s. And since the '90s the (Skaaning- founded) ScanSpeak company's 21W-8555 has been offered as a long-throw (8555.01) and regular-throw (8555.00) model with a great stiffness/weight ratio, carbon-fiber-impregnated, paper cone. With properly vented long-throw 8555.01 woofers, smaller enclosures can be used successfully.
I guess the ones I have are the sixth (or so) iteration over a period of three decades. They are quick and accurate, and they have appeared in many "high-end" systems. At $200 each, I like them a lot. They have one drawback: being rated at about 88dB/Watt/meter, they don't play very loudly—and that's why you often see them in pairs (at 91dB) in one system. Like the standard four-cylinder Toyota engine, they are tried and true, they've been incrementally improved over a generation, you know what you'll get with them, they do their job damn well, and with moderate use they'll last a long time.
Similarly the Vifa ring-radiator tweeter XT25TG30-04. It, too, appears in many high-end systems, sometimes all tarted up with cosmetic embellishments. It is known for its smooth response curve, its good dispersion pattern both on and off axis, and its ability to play pretty loudly with much less distortion than the standard fabric dome tweeters that came before. In an industry where every minor change is built up in the press to be a "revolutionary breakthrough," the ring-radiator, with its twin torus (doughnut) shape and wave-guide (often wrongly called a "phase plug"), was truly a major breakthrough. With a little tweaking of the rising high end of the Lowther C45 with a notch filter, and crossing over at a relatively low frequency, I was lucky enough to get the Vifa tweeter to blend pretty seamlessly into a fine three way without redesigning my crossover from the ground up.
In a way that is analogous to the woofer's being able to discriminate low register instruments, the Vifa tweeter can reproduce orchestral music. With violins soaring, it can still simultaneously capture and discriminate a cymbal, a tambourine, and a triangle working with each other as in Rachmaninoff's Symphonic Dances. With a lesser tweeter, the highs of such music get all mix-mastered together, homogenizing the distinct instruments into an almost indistinguishable spray of high frequencies.
With the Lowther C45 playing with the ScanSpeak 8555 below, and the Vifa ring-radiator above, I can say all of the parts of this speaker system combine into a synergistic result; that is, they each make the others sound better than when alone, or with other partners. The bass is very good if not huge, and what it lacks in depth, it makes up for with quickness, its ability to capture texture, and its surprising knack for discriminating the various harmonics of instruments in the lower register. The 91.5dB Vifa tweeter is likewise very quick and clean, capturing details without smearing harmonics that make up instrumental timbres. And that leaves the rest to the Lowther, the reproduction of the midrange wherein the vast proportion of music happens.
What we get in this particular system is an illusion that the speaker at hand is (as Tom Waits would say) "a bigger, better, new, and improved Lowther, with augmented bass" (that it needs, because at 5" in diameter the C45 bass starts to roll off at about 150Hz); and with silky smooth highs that are flattened out (that the C45 also needs, to overcome and compensate for its otherwise rising and peaky high end). But the characteristic sound is still pure Lowther. The defining qualities of the "Lowther Sound," what makes a Lowther, well, a Lowther, remain! Unchanged are the extraordinary clarity, the revealing details, the dynamic ability to go from very soft to very loud in an instant, the ability to capture nuance and innuendo of performance practices "down in the mix," the ability to play softly without losing its sonic character, the superior imaging, the space, the air, the delicacy, the power, and (when called upon, as in Wagner's "Siegfried's Funeral Music") the grandeur!!!
Max's second inconvenient truth is: "You can get a pretty good approximation of a single point source from a well designed three-way." That is to say, without laboring the point, using the Lowther C45 as a midrange doesn't compromise its sound. Within its most comfortable operating band-width, all the values of the single point-source remain in operation. When the world was young, and there were no computers to solve the problems of phase and gain in crossover networks, the Lowther promotional material stressed no crossover as a virtue of this speaker. I think this approach has frozen the Lowther, particularly the 5" models, into the role of a "full-range, single-point-source" driver. If a crossover is created with the skill and subtlety now available through computer-aided design, and the best premium parts are used (polypropylene capacitors, ribbon inductors, non-inductive resistors), you can get pretty fine multi-driver sound, as evidenced by all the best designs: the B&W's, the Wilson line, etc. After all, low frequencies are, in and of themselves, non-directional; and some tweeters are mounted on the same flanges as their midranges these days, which is as physically close to a single point source as you can get. Using the Lowther C45 as a midrange, mounted close as possible to its tweeter, brings it up to date.
Max's third inconvenient truth is: "Lowthers sound best as midranges." It doesn't get itself in trouble trying to make bass. Its high frequency peaks are less troublesome being crossed over beneath them, obviating the peaks in its 10dB rising high end. And, without having to reproduce lows and highs, the C45's dynamic range is expanded, enabling it to play louder. Letting separate drivers produce the high and low frequencies makes sense in this case, especially considering the frequency response curve of the C45. I find using the Lowther C45 as a midrange in my big rig actually improves its performance!!! Check out the articles cited in the beginning of this essay.
Standard three-way systems are the bread and butter of the industry. If the Lowthers are so damn good, but have problems in the extreme octaves of the frequency range (an inconvenient truth that anyone who's lived with Lowthers knows), why not design for them as midranges? Moreover, I think I also know of bigger, better, more efficient drivers that would make a propitious match for a 92dB system. More of that in the future.
Similarly, this reasoning is also applicable to the larger Lowthers, the 8" models. Dealing with them as midranges (though they might be wider-band than the crossover knees I'm currently employing) would mean finding even more efficient woofers and tweeters to match them with. Of course such drivers are expensive and have some unusual quirks of their own, like the woofer would need a much larger enclosure. High output woofs and tweets would work well with the larger Lowthers, if designers could use their ingenuity to design for the optimal crossover point knees and slope rates that accommodate the Lowther quirks. In a way that mirrors the ScanSpeak-Lowther-Vifa system, I feel certain now, a design could be worked out for a 12" or 15" woofer that would yield a much louder system, 96 or 98 dB/one watt, as well. Watch out Klipsch-horn lovers. We're out to getcha.
I'd like to know how many of you would enjoy building a Do It Yourself version of the 88dB speaker as I've outlined above. I think interested parties might be able to get the drivers from Lowther and Madisound, and we could get Dudious-88 crossovers built by Madisound if I send them a schematic. The only thing I have any doubts about is the cabinet. I have an enclosure in mind that would be great. I'll see if I can find the plans. Otherwise, I could specify rough parameters (size, bracing, hole cutout sizes), and you could take those specs to a cabinet maker and proceed from there with the furniture finish of your choice. Or, if you're on a tighter budget, you could build it out of MDF, using bungee cords to glue it together, and spray paint it matte black.
Any one interested in this project can contact me at email@example.com to see how many would be likely to take this on. Believe me, you won't be sorry. I'm as interested in the response to this speaker as I am interested in the response of this speaker. I know mine is first class. I've had pals come by with problematic CDs of music they never heard properly on their own system: Old And In The Way by a Grisman/Garcia Bluegrass band, for one. They were blown away finding out the CD wasn't terribly engineered, but their audio system might have been. That's a pretty nice, if double edged, testimonial.
For general ideas about kit building you might see http://www.madisound.com and click on "kits."
Approximate prices would be:
Lowther C45 $550/pair
SkanSpeak 8555 $400/pair
Vifa RingRadiator $110/pair
Dudious 3-way Xover?? Regular or premium parts?
2 cubic foot enclosures?? Pre-built or custom made?
Incidentals?? Binding posts, internal wire, stuffing material, silver solder, etc.
Are you audiophile hobbyist enough to really get into the nuts and bolts of speaker building? A factory version of a system like the above, if manufactured, packaged, shipped, advertised, and marketed in retail stores, would have to cost $10K or more. If you're interested, be in touch. When you get them up and running, I'm sure you'll do the Funkadelic Duck Walk as you haven't since you were a teenager. You'll be so turned on you might want to rush out and catch a giant wave. So be careful.