I think the autotech guys will have something to say about LeCleach horns.
horns.pl
I've sampled the plastic SEOS and the quality is top notch. The fiberglass ones are reportedly much better. Anyway, it is a cheap alternative for quality horns, something that the diy community did not have a few years ago.
horns.pl
I've sampled the plastic SEOS and the quality is top notch. The fiberglass ones are reportedly much better. Anyway, it is a cheap alternative for quality horns, something that the diy community did not have a few years ago.
Pano, have you ever investigated the non saturable reactor style of inductor? These are the weird "slammed" looking E/I ones you get from the crossover cans of, I think Altec, but it could just as easily have been JBL horn systems? Mille' Nestorovic claimed they were the best sounding he knew of, though we never used them in the speakers. I have been experimenting on my Lowther crossovers with this type and to my ears Mille' was right.
Bud
Bud
Sorry about the delay in posting. I've just returned from a 3600-mile round-trip to the Oregon Coast (three days in Cannon Beach), seeing my old audio-pals in Portland, and visiting my 8-month-old grandson in Seattle. Great trip, but the driving was plenty tiring, and driving through the smoke-filled expanses of Eastern Washington, Idaho, and across Montana wasn't much fun. (Extensive grass fires covering a huge area, and resulting in air quality worse than a Los Angeles or Calcutta smog alert.) Bad enough to drive through it, but it must have been far worse for the poor folks living there.
As a result of the driving stress and filthy air quality, I got a nasty sinus infection when I returned to Colorado, but by taking several grams of D-Mannose and Vitamin C for three days, have been able to chase it away. The first arrival of autumn rains here in Colorado have also cleared the air, so breathing is all better again. Nice to be able to see the snow-capped mountains again - Colorado and much of the Mountain West had a very unpleasant and smoky summer thanks to the extreme heat and dryness. (The folks on the west side of the Cascades escaped this, and apparently had a rather nice summer.)
Anyway, I'm feeling better again, so here comes the postings!
Thanks for the thoughtful posting, Pano. It got me thinking about my own journey through speaker design. I was fairly mainstream when I started this gig back in 1975 for Audionics, with a strong Brit flavor to what I designed, following Laurie Fincham's guidelines on Target Function Filter Design and carefully designing the phase angles between drivers. That was a natural outcome of the work on the Shadow Vector Quadraphonic Decoder that I patented and developed for Audionics from 1972 to 1975. You're very aware of the phase angles between channels and how they affect imaging - both left to right and front-to-back.
It's a very short jump from phase angles between channels - which will blur and shift the phantom image - to paying attention to phase angles between pairs of drivers through the crossover region. This is ignored by most speaker designers, for some reason. Perhaps they just don't hear it - not everyone is sensitive to "phasiness", which is major defect of bad quad or bad stereo.
What most speaker designers are unaware of - and just about every reviewer I've met - is that many crossover topologies are inherently phasey-sounding. For example, even ideal 1st and 3rd-order crossovers have 90 degree phase angles between drivers. Add even minor response deviations to the ideal crossover, and phase angles go up to 120 degrees, which is immediately audible, even across the room, as a phasey, colored, unstable quality, with very poor tonality. It is most audible on vocals as a disjointed, artificial, nearly "autotune" quality, with strong and annoying sibilants.
It's what I hear when I go to shows. Drivers that sound out of phase, unstable imaging, and pretty poor tone quality. I'd offer to fix their speakers - it would only take a few hours - but most speaker designers are so arrogant that they hit the ceiling when I make even the slightest suggestion that their babies might have just a touch of crossover issues. And then I find out that the famous-name "genius" designer hired out the crossover design to an outside consultant, which is another way of saying all the "genius" did was select the drivers and sketch out a pretty-looking cabinet with lots of cool-looking sharp edges.
What's wrong with most of the speakers at shows is pretty easy to fix - but I'm not going to do it for free, which is what a lot of companies making $50,000 to $150,000 speakers seem to expect from their designers.
I still pay a lot of attention to phase angles between drivers - that's just basic crossover design and good practice. If you don't pay attention to that, and just trust to luck, well, you'll be spending a lot of time chasing out will-o-the-wisp colorations, every time you make a small crossover adjustment. The speaker will always have something wrong with it that is hard to pin down - an elusive coloration that changes with every recording you play.
After I left Audionics in 1979, I was reasonably happy with the low-efficiency (86 dB/metre) linear-phase speakers I had designed as my last project there. I was using the quite good Audionics CC-2, one of the better transistor amps of the day, with a 70V/uSec slew rate and unconditional stability into nearly any load.
What started the long journey into the kingdom of Tone was hearing a beat-up Dynaco Stereo-70, with the trivial "triode conversion" (move one wire), completely annihilate the latest $3000 Audio Research and every transistor amp on hand. This was in 1990, at the second meeting of the soon-to-be-notorious Oregon Triode Society. No, the dealer didn't ask us back, after the laughably ugly ST-70, complete with plenty of vintage rust, demolished and humiliated every product he carried. It was an amazing eye-opener for me; I set aside the 200-watt MOSFET amp project my two Tektronix pals had been working on for three years and took vacuum tubes seriously.
It was the tone that got me. That miserable-looking ST-70 had TONE; it sounded like real music, like the players were right there in the room, or at least somewhere close, like hearing them play in a nearby club while you ate a great lunch at a sidewalk cafe.
I never did buy the whole "additive coloration" argument that was popular at the time. Sure, you can add 2nd-harmonic distortion to a transistor amplifier, but that does not add tone. It gets richer sounding, sure, but no, it never makes the transistor amp sound like real music. Extra output resistance, like Bob Carver said? Wrong. Adding series resistance to the output of a transistor amp just makes it sound even worse, something transistor amps don't need.
I'd been doing speaker design long enough to know that there aren't any colorations that add "realness" to the sound. Colorations detract from quality; it's a stupid audiophile myth that adding second-harmonic distortion, cabinet coloration, or any other defect improves the sound. It just doesn't. It sounds fake, it sound "hifi", it sounds bad.
It's counterintuitive, but removing defects, one by one, lets more through. You hear more. The defects mask realism, they don't enhance it at all.
So vacuum-tube amps, rather than adding colorations (the conventional wisdom at the time and still mainstream opinion today) are free of an unknown coloration that afflicts transistor amps. This coloration mutes the tone colors and flattens the dynamics of real music. The impression of "TONE" that makes music come alive is dulled, muted, or not there at all.
The edges and surfaces of the music are all there, but the body is missing. It's like edge-enhanced VHS video - almost cartoon-like. And this flat, empty cartoon is what most audiophiles call "accurate"! Switch off the hifi, though, and play real music, and there's no correspondence at all. A typical audiophile hifi system sounds almost nothing like real music.
So began the long journey through direct-heated-triode amplification, and steadily raising the efficiency of the loudspeaker. I had no intention of giving up what I knew about speaker design - if anything, it made me more determined to find out what was going on in the technical domain.
One very significant difference with direct-heated-triodes is the extremely low inherent distortion of the device itself. I am convinced the ear, unlike instrumentation, is not fooled by feedback or other circuit tricks and can somehow detect what the device is doing at the physical level. The most linear devices keep sounding the best, and the task of the circuitry is to get out of the way and simply linearize/optimize the operating conditions of the device. Not fix it after the fact, even if it measures better.
I'm finding the same things with speaker drivers. The more linear field flux of an underhung voice coil, combined with a near-saturated gap, along with an Alnico magnet, is audible. More importantly, it is audible in the 200 to 800 Hz range where the driver is in the piston range and should be operating with no coloration at all. The fact that drivers operating in the piston range can sound different, even at levels far below full power, is quite significant - since the cone is de facto "perfect", the only remaining source of coloration is the magnetic system.
The differences I'm hearing, both in amplifiers and loudspeakers, seems to come down to the underlying physics. The more linear systems sound better. It's not a surprise that the designers of the Thirties focussed on the basic physical linearity of the systems, since feedback was not yet a universal panacea to all ills, and speaker equalization had not been invented. They were forced to focus on the basics; get distortion down, make it as flat as possible, and use every watt possible.
This is when Bell Labs/Western Electric (later spun off as ERPI, then Altec) designed the zero-feedback Class A PP Western Electric 86 and 92A amplifiers (the single-ended WE 91A actually used feedback), the predecessor to the 515 and 416 woofer, and the first versions of the 288, along with first multicells (designed by Stephens).
I'm not a millionaire collector living in a luxury suburb of Tokyo. I have no interest at all in collecting a bunch of super-expensive WE, Altec, and JBL equipment in mint condition, complete with vintage wire (the latest fad). No thanks, not my thing.
I want to find out why it sounds that way, setting aside the antique faults, and re-discovering the long-lost virtues. I can say that amplifiers from the Thirties do not sound like "Golden Age" amplifiers of the Fifties - they do many things better. Some people like to copy, I like to understand, which involves re-creating the mindset of the original designers, and discovering the world they worked in. What were their technical and subjective goals? What were the limitations of the source materials of the day? What was practical and what was not?
Last edited:
Some minor historical notes: ERPI was formed as a WE subsidiary to take over WE's sound picture and non-telephone business from Jan 1, 1927.
Stephens may have designed the multicells for the Shearer and later Altec systems, but E.C. Wente of WE held the 1933 patent, and it was first used in the Fletcher system in 1933/34.
It's interesting to look at the curves for the tubes designed prior to the use of feedback; they are extremely linear compared to newer tubes. They had to be, as they were used in telephone repeaters, where very often one amplifier was used for multiple channels, and for both directions. Any distortion would result in crosstalk between the telephone channels. When feedback arrived, they also quickly noticed that they had to have increased headroom, because the clipping characteristics of the NFB amps would produce severe crosstalk even with very little clipping.
I think I have the same approach as you to these vintage designs, Lynn. I'm no collector, want to know why. There are things to learn from these designs.
-Bjørn
Stephens may have designed the multicells for the Shearer and later Altec systems, but E.C. Wente of WE held the 1933 patent, and it was first used in the Fletcher system in 1933/34.
It's interesting to look at the curves for the tubes designed prior to the use of feedback; they are extremely linear compared to newer tubes. They had to be, as they were used in telephone repeaters, where very often one amplifier was used for multiple channels, and for both directions. Any distortion would result in crosstalk between the telephone channels. When feedback arrived, they also quickly noticed that they had to have increased headroom, because the clipping characteristics of the NFB amps would produce severe crosstalk even with very little clipping.
I think I have the same approach as you to these vintage designs, Lynn. I'm no collector, want to know why. There are things to learn from these designs.
-Bjørn
Repeater duty, back in the days before Black's invention of negative feedback (he also worked at Bell Labs), was extremely demanding of amplifiers, since a long-distance intercity connection had so many repeaters (every twenty miles or so). Bell had to minimize buildup of group-delay errors, in order to keep speech intelligible, as well as buildup of high-order distortion products as the signal passed through multiple full-duplex repeater amplifiers (the original "line amplifiers").
This is non-trivial for a complex system, and failure to accomplish these goals had a direct impact on the Bell System's revenues (poor intelligibility, customer complaints and loss of revenue). Long-distance use of RF modulation on the same twisted-pair wire wasn't introduced until the mid Thirties, and coaxial cable was an outgrowth of WWII radar research. So until RF modulation, and later, coax cable, came into widespread use in the Bell System, long-distance telephone communication was carried out on twisted-pair wires at baseband frequencies.
The Bell Long Lines system required minimum distortion buildup over a link as long as New York to Los Angeles, minimum group-delay buildup despite multiple transformers from coast-to-coast, and sophisticated crosstalk rejection from the adjacent wire pairs. Bell Labs had their work cut out for them; no wonder they were the world's leading research lab for audio in nearly all areas, including the earliest movie-sound (the Vitaphone synchronized-disk system).
Moving on to the present, there are interesting issues that arise with unintended full-duplex cross-modulation in audio power amplifiers; distortion products and narrowband driver resonances (from the loudspeaker) appear on the plate or emitter of the power device in the final stage of the power amplifier, and will crossmodulate with forward-path audio, creating new IM distortion terms that do not appear in amplifier tests that use simple resistive loads.
As mathematical understanding of chaotic systems improves, I suspect we will discover that simple-appearing feedback systems are susceptible to surprisingly complex behavior when you look at low-level distortions. In particular, feedback systems probably have quite complex and unpredictable behavior when the forward path is time-modulated (thanks to nonlinear Miller capacitance in amplifying devices), as well as incorporating switching elements in the forward path (Class AB transitions with associated charge-storage delays). Right now, we use simple models that predict overall system behavior (in order to predict stability and distortion), but I think there's a lot more going on in the real system than the models account for.
Look at how grossly inaccurate the Webster horn model turned out to be - yet that was in use for more than 70 years. I suspect we're going to find our amplifier models have turned out to miss some very important parameters, while measuring the wrong things. It took more than 20 years for the audio community to discover the importance of slewing distortion, and the requirement for correctly shaped dither signals and reduction of jitter in PCM systems took nearly as long.
Last edited:
I have a pair here, also the slightly bigger JMLC-350. They both work very well on both the GPA 288 and the 399. In fact TAD TH-4001 as well.
Hey Lynn,
Sounds like a fun filled, if smokey, trip. Glad you're feeling better. Did you get by to hear Gary's system this trip? I last heard it in 2009 when he went to the new woofers and single midrange. It's a beautiful system, super clean and so "right."
Sounds like a fun filled, if smokey, trip. Glad you're feeling better. Did you get by to hear Gary's system this trip? I last heard it in 2009 when he went to the new woofers and single midrange. It's a beautiful system, super clean and so "right."
Ahhh.... that makes some sense of what I've experienced at shows. When I hear good crossover work, I compliment the designer. All of them will offer thanks, but few will talk about the crossovers. I thought "trade secrets" but maybe they don't elaborate because they didn't do the work!
Based on what I've heard from the first prototypes, to my surprise, you can listen as close as two meters with surprisingly good integration between the 416 and the AH425. Personally, I would recommend 3 to 6 meters as optimum, with the arc between the speakers, as seen from the listening position, from 40 to 55 degrees. Put them too close together, there will be too much center information (crowded soundstage), put them too far apart, the center gets unstable and swimmy. Space them so the soundstage has a smooth and even energy distribution across the stereo panorama.
Even close-up, the new speakers sound like a very large electrostat, with the difference being a more "in-the-room" presentation than electrostats. If a typical electrostat sounds like the music extends behind the panel, say from the panel to 20 feet behind it, these speakers have sounds extending from the middle of the room (yes, nearly to your knees) to far behind the loudspeaker, subjectively 20 to 50 feet, depending on how 3D the recording is. But all recordings have a very noticeable sense of depth, even historical mono recordings. If there is studio-generated out-of-phase information on the recording, the out-of-phase sounds will extend in a 180-degree arc, directly to the right and left of the listener.
The level of the horn needs to be slightly re-adjusted for listening distance, since it has a higher directivity index (translation to English: it doesn't fall off with distance as fast as the 15" woofer).
Even close-up, the new speakers sound like a very large electrostat, with the difference being a more "in-the-room" presentation than electrostats. If a typical electrostat sounds like the music extends behind the panel, say from the panel to 20 feet behind it, these speakers have sounds extending from the middle of the room (yes, nearly to your knees) to far behind the loudspeaker, subjectively 20 to 50 feet, depending on how 3D the recording is. But all recordings have a very noticeable sense of depth, even historical mono recordings. If there is studio-generated out-of-phase information on the recording, the out-of-phase sounds will extend in a 180-degree arc, directly to the right and left of the listener.
The level of the horn needs to be slightly re-adjusted for listening distance, since it has a higher directivity index (translation to English: it doesn't fall off with distance as fast as the 15" woofer).
Last edited:
Maxfield Parrish
Before I moved to Colorado, I thought the skies in Maxfield Parrish paintings were gorgeous, but not something you'd ever see in real life. Well, maybe you don't see skies like that in the Pacific Northwest, but here in Colorado on certain special days, yes you do. And here's an example, taken with my Olympus OM-D about an hour ago:
Before I moved to Colorado, I thought the skies in Maxfield Parrish paintings were gorgeous, but not something you'd ever see in real life. Well, maybe you don't see skies like that in the Pacific Northwest, but here in Colorado on certain special days, yes you do. And here's an example, taken with my Olympus OM-D about an hour ago:
Attachments
More Maxfield Parrish
A little bit of Photoshop adjusting to the RAW file, but not much, and to align it with the visual impression. Here's another photo taken a few minutes earlier:
To answer the previous question, sure, a 15 x 15 foot room would be fine - that would be about 20% larger than the room where I heard the prototypes.
A little bit of Photoshop adjusting to the RAW file, but not much, and to align it with the visual impression. Here's another photo taken a few minutes earlier:
To answer the previous question, sure, a 15 x 15 foot room would be fine - that would be about 20% larger than the room where I heard the prototypes.
Attachments
Last edited:
What do Altecs sound like? Well, they're the auditory equivalent of a brilliant Colorado day minutes before a sudden storm. Sparkling. Vivid. Dynamic.
Yes, I heard not one, but TWO great systems on my trip. David Robinson's, editor of Positive Feedback Online, who had the all-up MBL system (preamp, amp, and speakers) with an astonishing front end: a USB DSD DAC that played both 1X DSD and even more remarkable, 2X DSD, all from a laptop with the most astonishing copies of mastertapes I've ever heard. All of your favorites, going back to the Seventies and Sixties, but in fidelity that was indistinguishable from an Ampex ATR-100 playing the original mastertape - not a second-generation copy. Many recordings that I had thought had screwed-up mastertapes were crystal clear, and revealed a level of artistry from the performers that I didn't know they had. More to be shown by David at the Rocky Mountain Festival in a couple of weeks.
Gary Pimm had a DAW-based system with J River software playing the 44.1/16 and 88.2/24 files, routed through a software rack that had third-party VST plug-ins for crossover and equalization functions. An Echo Layla with modified analog electronics took care of the DAC conversion. As usual, the sound was great, but I told Gary, "Go Alnico, man, you won't go back, trust me on this". Since he's got a pair of Altec 604's and a pair of Electro-Voice 15's in his basement, he can just try a pair of each and see which he likes best. Or better yet, swap the 604's, which are very very difficult to design crossovers for, for a pair of 416's with Alnico magnets. He'd probably come out ahead in the trade, and the 416's are way better than anything from Eminence.
The low and infra-bass from Gary's system is the best I've ever heard, bar none. We were playing the soundtrack from the London Olympics opening ceremony, and the sound of that giant bell shook the whole house. Crystal clear stop-starts, not like subwoofer bass at all, and an unlimited sense of power. Pretty amazing.
Yes, I heard not one, but TWO great systems on my trip. David Robinson's, editor of Positive Feedback Online, who had the all-up MBL system (preamp, amp, and speakers) with an astonishing front end: a USB DSD DAC that played both 1X DSD and even more remarkable, 2X DSD, all from a laptop with the most astonishing copies of mastertapes I've ever heard. All of your favorites, going back to the Seventies and Sixties, but in fidelity that was indistinguishable from an Ampex ATR-100 playing the original mastertape - not a second-generation copy. Many recordings that I had thought had screwed-up mastertapes were crystal clear, and revealed a level of artistry from the performers that I didn't know they had. More to be shown by David at the Rocky Mountain Festival in a couple of weeks.
Gary Pimm had a DAW-based system with J River software playing the 44.1/16 and 88.2/24 files, routed through a software rack that had third-party VST plug-ins for crossover and equalization functions. An Echo Layla with modified analog electronics took care of the DAC conversion. As usual, the sound was great, but I told Gary, "Go Alnico, man, you won't go back, trust me on this". Since he's got a pair of Altec 604's and a pair of Electro-Voice 15's in his basement, he can just try a pair of each and see which he likes best. Or better yet, swap the 604's, which are very very difficult to design crossovers for, for a pair of 416's with Alnico magnets. He'd probably come out ahead in the trade, and the 416's are way better than anything from Eminence.
The low and infra-bass from Gary's system is the best I've ever heard, bar none. We were playing the soundtrack from the London Olympics opening ceremony, and the sound of that giant bell shook the whole house. Crystal clear stop-starts, not like subwoofer bass at all, and an unlimited sense of power. Pretty amazing.
Last edited:
Thanks for the report, Lynn. Yep, those subs of Gary's are amazing.
Speaking of amazing - I spent an amazing day today on the company of Don Keele and Tom Danley. Both of whom had brought their speaker to listen to. Never thought I 'd get the chance to hear the Keele CBT and the Danley Synergy horns side by side. But I did. It was a real treat. Wish you could have been there, Lynn.
Speaking of amazing - I spent an amazing day today on the company of Don Keele and Tom Danley. Both of whom had brought their speaker to listen to. Never thought I 'd get the chance to hear the Keele CBT and the Danley Synergy horns side by side. But I did. It was a real treat.
Wish you could have been there, Lynn.
Thanks for the report, Lynn. Yep, those subs of Gary's are amazing.
Speaking of amazing - I spent an amazing day today on the company of Don Keele and Tom Danley. Both of whom had brought their speaker to listen to. Never thought I 'd get the chance to hear the Keele CBT and the Danley Synergy horns side by side. But I did. It was a real treat.
Well played, Sir! How'd ya pull that one off?
Greg
PS, you can't throw something like that out in the wind and not expect a million questions about how it sounded....jeez.