"I do not see any reason that this has taken 7 years"
Lynn Olson was faced with "Sophie's Choice" during this journey, and relied on the kindness of strangers to keep his audio family together.
Lynn gave birth to the Karna and Amity tube amplifiers, and built his diminutive Ariel speakers for his kids to play with. Karna and Amity played well with Ariel, but Lynn knew Karna and Amity deserved higher resolution speakers that could also deliver a more natural sound in his room. We all want out kids to "be all they can be." Lynn became intrigued by the 4.77db increase in direct vs. indirect sound directivity index advantage of dipole speakers. Lynn wanted Karma and Amity to control the room through dipoles!
"The Ariel is now about 12 years old, and my tastes have changed a bit in the meantime. No, I'm still not a fan of horns, I still have reservations about tonality with classical music, especially the big-scale choral music I enjoy so much. The speaker designs I'm exploring these days are large-diaphragm dipoles - hearing both the Linkwitz Beethovens and the Bastani Apollo was a revelation about the dipole/open baffle sound."
Dipole technology advances in Linkwitz LX521 and JohnK NaNo Note2 demonstrated that the best dipole speakers are 4-way designs requiring complex equalization and massive power watts. State of the Art dipoles would not play well with Karna and Amity demands for simple gain stages, low power, and minimalistic signal paths. MiniDSP would never be allowed to feed Karna and Amity.
Lynn naturally made "Sophie's Choice" in favor of Karna and Amity. Dipoles had to DIE. Lynn continued wandering in the Audio Jungle seeking a better career for his kids. Some will say that Dr. Geddes, Jean-Michel Le Cléac'h and Altec Engineering were the three Wise Men leading a path to controlled directivity horns that Karna and Amity loved to play with. “It takes a village”, is closer to the truth, and it takes close friends like Gary Dahl for a happy journey.
1.4” compression driver + JMLC425 or JMLC350 horn + GPA 416 or Lambda TD15M midbass + BIG woofers
A 50 year old design formula using modern components and engineering details+details+details.
Lynn Olson was faced with "Sophie's Choice" during this journey, and relied on the kindness of strangers to keep his audio family together.
Lynn gave birth to the Karna and Amity tube amplifiers, and built his diminutive Ariel speakers for his kids to play with. Karna and Amity played well with Ariel, but Lynn knew Karna and Amity deserved higher resolution speakers that could also deliver a more natural sound in his room. We all want out kids to "be all they can be." Lynn became intrigued by the 4.77db increase in direct vs. indirect sound directivity index advantage of dipole speakers. Lynn wanted Karma and Amity to control the room through dipoles!
"The Ariel is now about 12 years old, and my tastes have changed a bit in the meantime. No, I'm still not a fan of horns, I still have reservations about tonality with classical music, especially the big-scale choral music I enjoy so much. The speaker designs I'm exploring these days are large-diaphragm dipoles - hearing both the Linkwitz Beethovens and the Bastani Apollo was a revelation about the dipole/open baffle sound."
Dipole technology advances in Linkwitz LX521 and JohnK NaNo Note2 demonstrated that the best dipole speakers are 4-way designs requiring complex equalization and massive power watts. State of the Art dipoles would not play well with Karna and Amity demands for simple gain stages, low power, and minimalistic signal paths. MiniDSP would never be allowed to feed Karna and Amity.
Lynn naturally made "Sophie's Choice" in favor of Karna and Amity. Dipoles had to DIE. Lynn continued wandering in the Audio Jungle seeking a better career for his kids. Some will say that Dr. Geddes, Jean-Michel Le Cléac'h and Altec Engineering were the three Wise Men leading a path to controlled directivity horns that Karna and Amity loved to play with. “It takes a village”, is closer to the truth, and it takes close friends like Gary Dahl for a happy journey.
1.4” compression driver + JMLC425 or JMLC350 horn + GPA 416 or Lambda TD15M midbass + BIG woofers
A 50 year old design formula using modern components and engineering details+details+details.
if you listen to CD or SACD chances are you listen to opamps
Far be it for me to be seen agreeing with POOH
The question is: which op-amps and for what, where in the production chain.
Nothing has undone Crowhurst's seminal work in '57 (JAES vol 4 no.5 p195 onwards) which points out all sorts of ways of "faking" good performance and the traps which arise when you clip a feedback amplifier. And we're all now aware of the issues associated with weird loads and feedback.
I'm trying to find a paper which looked at issues in recording.(it's on my desk somewhere - found it - Hamm, Russell O. "Tubes Versus Transistors-is there an Audible Difference." Journal of the audio engineering society 21.4 (1973): 267-273.)
It particularly looked at the impact of overload on input amplifiers - it clearly showed that an op-amp effectively cost 20db of headroom and discrete transistors 10db when compared with a triode due to the nature of the clipping.
The recent works of Lee and Geddes (eg. AES #115 2013) re-confirm this and much of Crowhurst.
So, if you've got a guarantee that your input voltage is bounded (so you won't clip) and a well-behaved load, an op-amp might provide a useful solution.
And the devil is in the details: while the overall architecture is old the design (especially) depends on quite recent work by Le Cléac'h and Geddes & Lee. All informed, as you point out, by Linkwitz's recent work.
BTW JMLC's dead links are to Bjørn Kolbrek's articles in 2008 AudioXpress "introduction to horns" articles.
My final reference to append to your rather fine summary of this thread is to the QuasiOptimal Crossover discussion, which deals with subwoofer issue
Last edited:
There is nothing we can do about the sound quality of the recordings.
What we can do is take care of the sound quality of the reproduction. The sound quality of the reproduction does matter, for all variants of recordings.
To my experience, in the quest of better sound quality in the reproduction system (though not necessarily better measurements), opamps in the signal path would better be omitted altogether.
I concur completely with the above quote.
It is applicable to each and every link in the reproduction (and recording) chain, not only to speakers, though, probably, with speakers the greatest trade offs are unavoidable.
Indeed it is.
One reason for it being so is that, as for today, a perfect sound reproduction is unattainable, at any cost. Thus, the quest for better sound reproduction is a never ending quest.
Indeed, it is so.
Again, when perfect sound reproduction is unattainable, different people give their own preferences to different aspects of sound reproduction; preferences as to what aspects they care about more than others, what trade offs they prefer.
This is before going into the diversity of various types of music; that is, some listen mainly to acoustic instruments and unamplified human voices, while others listen mainly to electronic instruments and amplified human voices.
To the first type of music, a sense of 'realism' is applicable, while it isn't applicable to second type. Hence, different requirements and preferences concerning the sound qualities of the reproduction chain.
Thank you for your beautiful summary.
It looks to me that all of us are facing a kind of "Sophie's Choice" during the journey on the quest of better sound quality in the reproduction chain. That is, since perfect sound reproduction is unattainable, we have to choose what aspects we prefer to be better, at the cost of lower quality in other aspects.
The moment for me was sitting down for some lunch at The Frank Gehry-designed Jay Pritzker Pavilion at the Millennium Park Chicago and listening to an orchestra practise.
The experience was quite enlightening in terms of hearing the real sounds of instruments in an open space. The purity or should I say the density of tone and spatial dynamics were a revelation.
It made me realise that largely, its all about the midrange and that if you are prepared to look past the complexity of execution the priorities can be simplified.
Hi sensitivity, dynamic range and low distortion.
Some time ago I had the opportunity to hear two JBL studio monitors side by side.
The relatively new small format LSR6632 and the vintage 4435 large format monitors.
With the LSR6332 you could say it offered a small scaled down version of the event.
But in comparison to the 4435 it sounded very confined like the sound was coming from inside a box. The 4435 while not having the nth degree of accuracy was far more life like in presentation.
The difference in sensitivity is 6db and the difference in cone area is 8 x times for the woofer and mid range. That = lower distortion and significant increase in dynamic range.
Each time you double x 2 the cone area you reduce cone excursion by a factor of four =1/4.
I take a few tips from a Red book I by Alex Bradmanoff that was full of Boazk, Jensen and Altec plans .
With 4435 the bass and highs are there but they are not a standout in the presentation like you get with a typical book shelf two way.
Given the 4435 used 1980s technology the mind boggles at what improvements could be made with current soa drivers.
But as we know the devil is in the details.
I think Lynn has done a great job and it does take time particularly with life's intervening issues.
I mean as audio amateurs we are not employed by someone like Kef and paid to design loudspeakers all day every day, day in day out.
Those guys might spend a year or so coming up with a prototype ready for market with lots of resources.
Too often, arm chair designs are implemented too early without trials to find the best set of complimentary compromises.
Previously I negatively engineered a 4way JBL 4345 monitor with the help of the designer.
I was surprised to learn the system was designed empirically according to Greg Timbers.
The drivers I plan to use are (AE 15M ), Tad 4003, Tad 703 and a Emilar 500 exponential cast aluminium horn.
At the moment I am building a test box of some 7 cu ft for the AE15m to take some measurements with LMS (ground plane) and in the proposed listening room to check the room gain.
I am also making a plane wave tube to measure the 4003.
So I will have some base line data to work with.
Once I have done that step I will trial some configurations using modular building blocks to sort out what works best in my open plan room which is some 5 x 8 metres with 2.9 metre ceiling.
As Lynn has done I expect I will need some LF reinforcement but how much and how its done will depend on the above trials.
A while ago I purchased Bass box software from the guy who designed Calsod.
This guy had done research on the audibility of LF sensitivity changes and basically said perceptual differences in level and distortion really come into play below 100 hertz.
What I have found is that doubling the cone area in the bass really increases the sense of authority while increasing extension below 40 hertz adds a new dimension.
So it will be really interesting to see what evolves as my project progresses.
I plan to also evaluate a Azura horn 425, the Emilar horn and possible a conical horn or an seos wave guide. Also look at some options for front loading the AE15M so as to enable a degree of latitude for the transition to the 4003.
Previously I trialled a 3 way tower system using dual 6 1/2 drivers and the room really soaked up all the speakers to throw at it so I knew then I had to scale up some what
City of Chicago :: Millennium Park Media Image Gallery
The experience was quite enlightening in terms of hearing the real sounds of instruments in an open space. The purity or should I say the density of tone and spatial dynamics were a revelation.
It made me realise that largely, its all about the midrange and that if you are prepared to look past the complexity of execution the priorities can be simplified.
Hi sensitivity, dynamic range and low distortion.
Some time ago I had the opportunity to hear two JBL studio monitors side by side.
The relatively new small format LSR6632 and the vintage 4435 large format monitors.
With the LSR6332 you could say it offered a small scaled down version of the event.
But in comparison to the 4435 it sounded very confined like the sound was coming from inside a box. The 4435 while not having the nth degree of accuracy was far more life like in presentation.
The difference in sensitivity is 6db and the difference in cone area is 8 x times for the woofer and mid range. That = lower distortion and significant increase in dynamic range.
Each time you double x 2 the cone area you reduce cone excursion by a factor of four =1/4.
I take a few tips from a Red book I by Alex Bradmanoff that was full of Boazk, Jensen and Altec plans .
With 4435 the bass and highs are there but they are not a standout in the presentation like you get with a typical book shelf two way.
Given the 4435 used 1980s technology the mind boggles at what improvements could be made with current soa drivers.
But as we know the devil is in the details.
I think Lynn has done a great job and it does take time particularly with life's intervening issues.
I mean as audio amateurs we are not employed by someone like Kef and paid to design loudspeakers all day every day, day in day out.
Those guys might spend a year or so coming up with a prototype ready for market with lots of resources.
Too often, arm chair designs are implemented too early without trials to find the best set of complimentary compromises.
Previously I negatively engineered a 4way JBL 4345 monitor with the help of the designer.
I was surprised to learn the system was designed empirically according to Greg Timbers.
The drivers I plan to use are (AE 15M ), Tad 4003, Tad 703 and a Emilar 500 exponential cast aluminium horn.
At the moment I am building a test box of some 7 cu ft for the AE15m to take some measurements with LMS (ground plane) and in the proposed listening room to check the room gain.
I am also making a plane wave tube to measure the 4003.
So I will have some base line data to work with.
Once I have done that step I will trial some configurations using modular building blocks to sort out what works best in my open plan room which is some 5 x 8 metres with 2.9 metre ceiling.
As Lynn has done I expect I will need some LF reinforcement but how much and how its done will depend on the above trials.
A while ago I purchased Bass box software from the guy who designed Calsod.
This guy had done research on the audibility of LF sensitivity changes and basically said perceptual differences in level and distortion really come into play below 100 hertz.
What I have found is that doubling the cone area in the bass really increases the sense of authority while increasing extension below 40 hertz adds a new dimension.
So it will be really interesting to see what evolves as my project progresses.
I plan to also evaluate a Azura horn 425, the Emilar horn and possible a conical horn or an seos wave guide. Also look at some options for front loading the AE15M so as to enable a degree of latitude for the transition to the 4003.
Previously I trialled a 3 way tower system using dual 6 1/2 drivers and the room really soaked up all the speakers to throw at it so I knew then I had to scale up some what
City of Chicago :: Millennium Park Media Image Gallery
Last edited:
I speak of my own of course.
A friend of mine yesterday related an experience he had recently with an ultra high-end system. The system consisted of the following:
Front-end:
dCS Scarlatti SACD/CD playback system
Uknown high-end turntable
Amplification:
Audio Research 40th Anniversary Edition Reference Preamplifier
Audio Research Reference power amplifiers (not sure which model, but the speakers were being bi-amped)
Speakers:
Vivid Audio Giya G1 loudspeakers
My friend is someone who has a high-end system himself, and he has listened to some of the best equipment available, but he says that this system was something special. He went with another friend who is also a knowledgeable audiophile. They took their own music to use in the demo. He said that they sat listening for an hour, and during that whole time they could not find one aspect to criticize. He said it was by far the best sound he had ever heard. One of the factors contributing to this was that the demo space was large- the speakers were 5m (16.5ft.) apart, and they sat 7m (23ft.) away from them. But that alone could not account for what they heard. He said that quite honestly there was nothing he could find that he could complain about. To say they left stunned by the experience is an understatement. So, yes, while I agree with the statement quoted above, I do think that with the right combination and a big room, we can get closer than most people think.
Deon
You are heading in the right direction.
1) Find out your own tastes. What really excites and turns you on? Enough so that you'll put thousands of dollars and hundreds of hours building your own hifi system to get there? Live acoustic symphonic or choral, live jazz club (with or without PA), live rock concert, a dance club, or audiophile goals for certain "reference" recordings?
2) Within reason, what technologies provide this kind of sound? It's obviously not possible to duplicate a $500,000 dance club system in a normal-sized house, but it can scaled down with the same sort of professional equipment. Getting close to the sound of acoustic ensembles is extremely difficult, but is a bit easier if you're willing to compromise on the scale and size of the sound. The audiophile goal is nebulous, since most of us have not heard the "reference" record in the room where it was mastered and had final approval from the musicians and the producer.
3) Now for the hard part. Get good PC-based instrumentation, an external sound card with 192/24 capability and 80 kHz bandwidth (I use the RME Babyface with USB 2.0 interface and ARTA software, and before that, the MLSSA card and software). Learn how to measure the impulse response with existing speakers that are reasonably flat, how to remove the floor bounce (I use 2 to 3 feet of soft pillows and blankets), set the start-stop window so other reflections are excluded (these will be easy to see if the speaker has decent transient response), and then run the FFT to see the frequency response (usable mostly above 300 Hz). Measure at 0 degrees, 15 degrees, 30 degrees (horizontal), and see if there are nulls in the vertical plane.
4) Start assembling drivers, cabinets, and prototype crossovers, or use digital crossovers and multi-amp so you can adjust on the fly. I usually assess drivers on a large IEC-or-larger flat baffle just to hear what they sound like without any crossover at all, although this cannot be done with tweeters or compression drivers.
All of you should take note of this. These are non saturable reactors. They force the coil to gain and loose inductance in a linear format. They are also much smaller than an equivalent air coil and if you must use autoformers this would be a superior way to implement them. For Zen, the effective perm is 45.
As stated by Mundorf.
Air coils
The ideal core material for coils is air. Air cored coils are, for physical reasons, superior to all metal core coils as far as accurate pulse reproduction and freedom from distortion are concerned. They can be used in all areas; either as highpass filter in the middle frequency range, as bass coil (with large conductor cross section) or in correcting components (with thin wire cross section).
Precision, dynamic, subtle tonal gradations, great detail and liveliness distinguish coils with air core from all others. In high-quality speakers, they are thus the basis for realistic and harmonic musical enjoyment.
Air coils
The ideal core material for coils is air. Air cored coils are, for physical reasons, superior to all metal core coils as far as accurate pulse reproduction and freedom from distortion are concerned. They can be used in all areas; either as highpass filter in the middle frequency range, as bass coil (with large conductor cross section) or in correcting components (with thin wire cross section).
Precision, dynamic, subtle tonal gradations, great detail and liveliness distinguish coils with air core from all others. In high-quality speakers, they are thus the basis for realistic and harmonic musical enjoyment.
Quick recap: I was feeling sorry for myself after breaking my leg, and Bud Purvine invited me to start a new audio project. The recovery was tediously long (six months before I could apply any weight on the leg, and another three to six months of learning to walk again), and life isn't much fun when your mobility is impaired. Bud had a similar experience in his senior year of high school, and knew that I needed something to keep me cheered up.
I had been interested in a high-efficiency (97 dB/meter or higher) speaker that had good time response for a long time, and started collecting data on what was out there. My primary criteria for "good time response" is a rapid decay-to-zero, without a lot of clutter from diaphragm resonance, diffraction artifacts, or internal cabinet modes. I was strongly influenced by meeting Laurie Fincham of KEF and the BBC research team in 1975, and as a result, my primary emphasis is a rapid decay-to-zero that is free of resonance, along with an overall flat response.
Traditional horns had rather poor time characteristics, thanks to diffraction in the throat (sometimes an intentional part of the design), diffraction at the mouth, and internal standing waves and/or high-order modes (Dr. Geddes' HOM's). When I started the project, there were no off-the-shelf OS horns, and I didn't want to infringe on Dr. Geddes' patents. So I started looking around at other horn profiles, and contacted Bjorn Kolbrek, who was researching LeCleac'h profiles.
Bjorn's BEM simulations looked promising, so I then contacted Martin Seddon in Australia for a custom (bespoke) horn using Bjorn's research. The AH425 was the result, and he sent the first pair to me here in Colorado.
I expected nearly all high-efficiency 15" drivers to sound the same in the flat-response piston band (700 Hz on down), and was rather surprised that they didn't. The Altec/GPA 416 was already being considered, and was my personal favorite; it turned out that Gary Dahl felt the same way, and made it part of his project.
I kept in contact with the project manager at Brush Wellman/Materion during the two years they were developing Beryllium diaphragms for 1" and 1.4" exit compression drivers, and Gary Dahl got one of the first versions of the Radian 745NeoBe drivers. I've encouraged Gary Dahl, Gary Pimm, and Martin to post their measurements and subjective listening results to this (and other) forums, while considering what LF format I'll use for my own system.
A bass horn is attractive in terms of lack of cabinet coloration, high efficiency, low IM distortion, and a directivity index similar to the MF/HF driver, but it just won't fit into the space available. So the most likely option at this time remains a pair of 15" drivers, with crossover optimization to maintain close phase match (within 5 degrees) at the listening position.
Last edited:
macka:
You are definitely heading in the right direction however you should seriously consider a front loaded mid bass horn if at all possible. You would have to pad down the midrange & tweeter to match the efficiency of the AE ( check Hornsrep to see if that driver is good for horn loading )and that would be a shame. The AE woofers are excellent but will cause you problems getting a smooth transition to the TAD driver at the crossover point. The polar response of the AE driver and the TAD/horn may be a mismatch at a 500-600hz crossover point.
You are definitely heading in the right direction however you should seriously consider a front loaded mid bass horn if at all possible. You would have to pad down the midrange & tweeter to match the efficiency of the AE ( check Hornsrep to see if that driver is good for horn loading )and that would be a shame. The AE woofers are excellent but will cause you problems getting a smooth transition to the TAD driver at the crossover point. The polar response of the AE driver and the TAD/horn may be a mismatch at a 500-600hz crossover point.
Wellll. IMO you are belaboring the Blatantly Obvious.
Geee. Who woulda thought Live often bears NO resemblance to Canned reproduction
Erm.. Pretty well ever Kid with ears in the 70's .
Did you Miss all that?
But then this interminable thread reads wayyy too much like an Open Mouth Radio talk show .. IMO 10,000 opinions, all vying for position/prominence.
Geee. Who woulda thought Live often bears NO resemblance to Canned reproduction
Erm.. Pretty well ever Kid with ears in the 70's .
Did you Miss all that?
But then this interminable thread reads wayyy too much like an Open Mouth Radio talk show .. IMO 10,000 opinions, all vying for position/prominence.