Jmmlc,
I'm sure the Le Cleac'h have very good sonic characteristics.
The Oblate Spheroid Waveguide is quite interesting to me considering the directivity characteristics, considering these are quite conical with proper throat and lip terminations, data another thread seemed quite interesting.
Recently I'm gerting ready to try an elliptic Oblate Spheroid Waveguide using a direct radiating driver, I think this will be interesting test as well.
I'm sure the Le Cleac'h have very good sonic characteristics.
The Oblate Spheroid Waveguide is quite interesting to me considering the directivity characteristics, considering these are quite conical with proper throat and lip terminations, data another thread seemed quite interesting.
Recently I'm gerting ready to try an elliptic Oblate Spheroid Waveguide using a direct radiating driver, I think this will be interesting test as well.
Horns are so revealing that if the driving equipment is not setup right, it's impossible to hear the true sound of the horns. I have not heard any horns that sound natural to me, but it may also be my perception about what is natural. If there are any Le Cleac'h horns in Taiwan, I'd sure like to get a chance to listen to them.serenechaos said:
Jmmlc said:
Hello Jean-Michel,
Thanks for the measurement. That is reassuring since as I said I plan on using your flair method for my next system. Do you remember what angle step size you used in this measurement? I can't quite tell from looking at the graph. I did want to compare my FEA to reality. There are definitely issues at high frequencies - if I do not use enough elements, the results are wrong, but using too many takes too long. Also there could be several other issues.
John
My reservation about equalization is the presence of time-domain reflections; these cannot be equalized, and are the result of internal reflections and diffraction within the horn and phaseplug assembly. The dispersion spikes seen in the conical horn examples could easily be the result of internal reflections and diffraction, although this cannot be known without direct examination of the time-domain data.
Broad-brush equalization (to compensate for the inherent mass rolloff of the driver) is benign, but narrowband equalization gets you in much deeper waters. Now you have to be careful not to smooth out a narrow dispersion spike (and getting it worse elsewhere), or smoothing out a frequency response ripple and making the time-domain worse. All of my instincts tell me to look at the time vs polar pattern, and note if there are specific narrow angles (that are close to the listening position) where time-domain artifacts appear.
This problem isn't exotic as it sounds; it also appears in conventional direct-radiator speakers as a result of edge diffraction, most importantly, off the left and right edges of the cabinet. Reducing diffraction off these edges has only a minor effect on the conventional on-axis frequency response (if it is visible at all), but has a significant effect on image quality and reduction of coloration on vocals.
The time, frequency, and spatial domains are different ways of looking at the same underlying physical mechanisms, and since loudspeakers are such complex devices, it's a good idea to use all three as extensively as possible. What we tediously measure in all three domains is audible as a single, unified perception to the ear.
Broad-brush equalization (to compensate for the inherent mass rolloff of the driver) is benign, but narrowband equalization gets you in much deeper waters. Now you have to be careful not to smooth out a narrow dispersion spike (and getting it worse elsewhere), or smoothing out a frequency response ripple and making the time-domain worse. All of my instincts tell me to look at the time vs polar pattern, and note if there are specific narrow angles (that are close to the listening position) where time-domain artifacts appear.
This problem isn't exotic as it sounds; it also appears in conventional direct-radiator speakers as a result of edge diffraction, most importantly, off the left and right edges of the cabinet. Reducing diffraction off these edges has only a minor effect on the conventional on-axis frequency response (if it is visible at all), but has a significant effect on image quality and reduction of coloration on vocals.
The time, frequency, and spatial domains are different ways of looking at the same underlying physical mechanisms, and since loudspeakers are such complex devices, it's a good idea to use all three as extensively as possible. What we tediously measure in all three domains is audible as a single, unified perception to the ear.
Re: Steve K's System
Just noticed that thread and well I have that
system too- Klipshorn with Azura 204 ,Lowther DX4 (ENABLed) with the same wooden dorknob phase plug , ,Yamamoto A-08 clone with Bud's transformers and some digital on Klipshorns. The room is difficult and I have them only 10' appart ( although the space is 21x21) but I think they are average at best and " world class " seems like a bad joke . I must be doing something wrong .Can I get some contact info to Steve and John ???
Oh , forgot to add that I also have EV T 350 tweeters.
Regards, L
PS Edit -actually, the system is so bad that I prefer to listen hiden behind the kitchen wall
Lynn Olson said:
Just noticed that thread and well I have that
system too- Klipshorn with Azura 204 ,Lowther DX4 (ENABLed) with the same wooden dorknob phase plug , ,Yamamoto A-08 clone with Bud's transformers and some digital on Klipshorns. The room is difficult and I have them only 10' appart ( although the space is 21x21) but I think they are average at best and " world class " seems like a bad joke . I must be doing something wrong .Can I get some contact info to Steve and John ???
Oh , forgot to add that I also have EV T 350 tweeters.
Regards, L
PS Edit -actually, the system is so bad that I prefer to listen hiden behind the kitchen wall
Hello John,
Those measurements were done at home in my living room and I could not rotate the enclosure. So I draw a circle on the floor with rubber tape and put my microphone stand at the vertical of that circle.
At every position I used a pulse mesurement deveopped by Angelo Farina (Aurora modules ) that uses a logarithmic sweep of sine.
I repeat this for several position of the microphone stand on the circle and so I recovered the complete response curve for every position.
The mean angle step between the positions was around 7° but may varies due to the simplistic method used.
Next time I'll do such mesurement I'll try to improve the measurement.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Those measurements were done at home in my living room and I could not rotate the enclosure. So I draw a circle on the floor with rubber tape and put my microphone stand at the vertical of that circle.
At every position I used a pulse mesurement deveopped by Angelo Farina (Aurora modules ) that uses a logarithmic sweep of sine.
I repeat this for several position of the microphone stand on the circle and so I recovered the complete response curve for every position.
The mean angle step between the positions was around 7° but may varies due to the simplistic method used.
Next time I'll do such mesurement I'll try to improve the measurement.
Best regards from Paris, France
Jean-Michel Le Cléac'h
John Sheerin said:
Hi Limono, which systems have you heard that you considered "world class"? I'd never go with a Lowther/AER based system for myself (more coloration than I can accept), but what I heard over at those two places had very special qualities I don't usually hear.
Mainstream high-end systems leave me cold these days - Acapella, Avante-Garde, Wilson, JMLab, assorted other big WMTMW boxes that retail over $100,000, and so on. The only systems that put a smile on my face are the exotic, reach-for-the-stars DIY systems and unusual small-production-run manufacturers that are unknown in the larger market.
Come to think of it, I felt the same way when I designed the Ariels 14 years ago - there certainly wasn't anything commercial that I wanted, which I why I had to get out of the reviewing business with Positive Feedback magazine.
Mainstream high-end systems leave me cold these days - Acapella, Avante-Garde, Wilson, JMLab, assorted other big WMTMW boxes that retail over $100,000, and so on. The only systems that put a smile on my face are the exotic, reach-for-the-stars DIY systems and unusual small-production-run manufacturers that are unknown in the larger market.
Come to think of it, I felt the same way when I designed the Ariels 14 years ago - there certainly wasn't anything commercial that I wanted, which I why I had to get out of the reviewing business with Positive Feedback magazine.
Lynn
I didn't doubt your taste or anything. Systems I liked were ,Avantgardes (back in Poland) and at RMAF (couple of years ago) - big Tannoys (with Japanese 211 amps) and expensive Audio Note system. Most of the big $100K I did not like as well.
Bud's enabled DX4's gave me a hard time but they (or I ) finally "broked in". The system has a nice qualites but in my room (which may be a culprit) as a whole is not good. That's why I'm curious about Steve and John implementations. Crossover details , etc.
Regards, L
I didn't doubt your taste or anything. Systems I liked were ,Avantgardes (back in Poland) and at RMAF (couple of years ago) - big Tannoys (with Japanese 211 amps) and expensive Audio Note system. Most of the big $100K I did not like as well.
Bud's enabled DX4's gave me a hard time but they (or I ) finally "broked in". The system has a nice qualites but in my room (which may be a culprit) as a whole is not good. That's why I'm curious about Steve and John implementations. Crossover details , etc.
Regards, L
Sounds like our tastes aren't that far apart - there's something about modern audiophile drivers I don't like. They seem to have lost their way - for example, the modern Scan-Speak tweeters aren't as good as the 9000 sticky-dome I was using when I designed the Ariels. They measure (a little) better, but don't sound as realistic.
Measurements don't show whether a speaker can sound beautiful or not; really poor measurements indicate that a given system is unlikely to sound that way, but speakers that measure good-to-excellent can sound grainy, dry, mechanical, and unmusical, while speakers with only just fair measurements can sound startlingly lifelike and shimmeringly musical, while still sounding noticeably colored.
The quality of "realism" and "beauty" seems to correlate with subtle construction details in the drivers (and horns), and the taste and skill of the system designer. Design-by-the-numbers doesn't usually cut it; the designer has to know which set of parameters to optimize, and have a good understanding of the inherent limitations of the drivers.
As for Steve and John's implementations, they did several things that were counter-intuitive; the speakers are right in the corners of quite wide rooms, with the big LeCleac'h/Azurahorn directly on top of the bass section of the Klipschorns. The stereo image is still stable despite what appears to be very wide spacing, a testament to the smooth dispersion of the horns.
I'm guessing the low-level crossover is around 220 Hz, so the Azurahorns are run right into cutoff; naughty naughty, I know, but it seems to work. I think they wanted to avoid the really bad region of the K-horn, and use the AER drivers over as wide a range as possible. I think the line-level crossover is either 6 or 12 dB/octave, nothing all that fancy, unlike the power amps, which are quite exotic. Both systems also use a single powered subwoofer below 40 Hz, but I don't think the integration is 100% successful - more subwoofers (3 or 4) would probably have better system integration.
I hear a HF rolloff starting somewhere around 12 kHz, not surprising for a Lowther or AER - but they don't sound dull, nor do they sound tipped-up and aggressive. Just not super-extended on the HF end.
The mids and most of the range? Wonderful, although at the same time moderately colored (too much for my own system, where I'm willing to trade off some "character" in return for as little coloration as possible). I should mention this is auditioned with symphonic and opera music on a top-quality phonograph, so we're talking about much lower levels of coloration than the usual PA-heritage high-efficiency horn system, as well as a noticeably lower level of coloration than any Avant-Garde or Acapella system I've heard to date.
I wouldn't characterize either Steve's or John's system as having the see-through-the-window sound of a Quad ESL57, which I'm not too sure is attainable with high-efficiency systems. On the other hand, ESL57's and BBC monitors can't begin to approach the dynamic realism, subtlety of tonal shadings, vividness or "you-are-there" quality of the best high-efficiency systems.
I should also add the Red Book CD's sound pretty flat and colorless on Steve and John's systems; the resolution is so high that the low resolution of 16-bit sources is all too apparent. Not that CD playback isn't enjoyable; it is, actually, but it kind of sounds like good AM radio in comparison to excellent FM. We usually all breathe a sigh of relief when we go back to all-analog sources. These systems are most certainly in the subjective 24-bit or higher range, while most audiophile speakers can hardly even distinguish between 16 and 20-bit sources. If it's in the recording, you WILL hear it.
I guess that's the downside of super-resolution systems. They can be built, but the demands on the rest of the system are considerable; my own crazy all-triode DHT amplifier is at a high enough level, and Steve and John's amps most certainly are, but nearly all commercially available amplifiers just aren't good enough.
Measurements don't show whether a speaker can sound beautiful or not; really poor measurements indicate that a given system is unlikely to sound that way, but speakers that measure good-to-excellent can sound grainy, dry, mechanical, and unmusical, while speakers with only just fair measurements can sound startlingly lifelike and shimmeringly musical, while still sounding noticeably colored.
The quality of "realism" and "beauty" seems to correlate with subtle construction details in the drivers (and horns), and the taste and skill of the system designer. Design-by-the-numbers doesn't usually cut it; the designer has to know which set of parameters to optimize, and have a good understanding of the inherent limitations of the drivers.
As for Steve and John's implementations, they did several things that were counter-intuitive; the speakers are right in the corners of quite wide rooms, with the big LeCleac'h/Azurahorn directly on top of the bass section of the Klipschorns. The stereo image is still stable despite what appears to be very wide spacing, a testament to the smooth dispersion of the horns.
I'm guessing the low-level crossover is around 220 Hz, so the Azurahorns are run right into cutoff; naughty naughty, I know, but it seems to work. I think they wanted to avoid the really bad region of the K-horn, and use the AER drivers over as wide a range as possible. I think the line-level crossover is either 6 or 12 dB/octave, nothing all that fancy, unlike the power amps, which are quite exotic. Both systems also use a single powered subwoofer below 40 Hz, but I don't think the integration is 100% successful - more subwoofers (3 or 4) would probably have better system integration.
I hear a HF rolloff starting somewhere around 12 kHz, not surprising for a Lowther or AER - but they don't sound dull, nor do they sound tipped-up and aggressive. Just not super-extended on the HF end.
The mids and most of the range? Wonderful, although at the same time moderately colored (too much for my own system, where I'm willing to trade off some "character" in return for as little coloration as possible). I should mention this is auditioned with symphonic and opera music on a top-quality phonograph, so we're talking about much lower levels of coloration than the usual PA-heritage high-efficiency horn system, as well as a noticeably lower level of coloration than any Avant-Garde or Acapella system I've heard to date.
I wouldn't characterize either Steve's or John's system as having the see-through-the-window sound of a Quad ESL57, which I'm not too sure is attainable with high-efficiency systems. On the other hand, ESL57's and BBC monitors can't begin to approach the dynamic realism, subtlety of tonal shadings, vividness or "you-are-there" quality of the best high-efficiency systems.
I should also add the Red Book CD's sound pretty flat and colorless on Steve and John's systems; the resolution is so high that the low resolution of 16-bit sources is all too apparent. Not that CD playback isn't enjoyable; it is, actually, but it kind of sounds like good AM radio in comparison to excellent FM. We usually all breathe a sigh of relief when we go back to all-analog sources. These systems are most certainly in the subjective 24-bit or higher range, while most audiophile speakers can hardly even distinguish between 16 and 20-bit sources. If it's in the recording, you WILL hear it.
I guess that's the downside of super-resolution systems. They can be built, but the demands on the rest of the system are considerable; my own crazy all-triode DHT amplifier is at a high enough level, and Steve and John's amps most certainly are, but nearly all commercially available amplifiers just aren't good enough.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.