I think what's lacking is a "current recipe". It's fresh in Lynn and Gary's heads, but having spent thousands of dollars worth of my time on this thread, I can't state definitively what the most recent state is. My reference rig is underhung BD15 midwoofers crossed passively to large format diy waveguides, close approximations to OSWG geometry (with a long throat necessitated by the long throats inherent to screw-on drivers).
Lynn/Gary's rig is different tweets, LeCleach profile, and different woofs, from what I can tell, and diffraction is a key design focus- true of mine as well. Beyond that... I'd have to dig... and dig.... through post after post of off-topic drivel and long-winded alleyways that sometimes do and sometimes don't have meaningful information in the golden dumpster at the end.
We were warned.... but the extent of navelgazing is well beyond a couple-year development cycle. 12,000+ posts and 7 1/2 years.... the development has exceeded 50% of the initial tenure of the Ariel, we went from a 12 year old design to nearly 20 years old without an endpoint.
It's like playing the lottery, effort without return but with the occasional nugget that fools you into coming back.
Lynn/Gary's rig is different tweets, LeCleach profile, and different woofs, from what I can tell, and diffraction is a key design focus- true of mine as well. Beyond that... I'd have to dig... and dig.... through post after post of off-topic drivel and long-winded alleyways that sometimes do and sometimes don't have meaningful information in the golden dumpster at the end.
We were warned.... but the extent of navelgazing is well beyond a couple-year development cycle. 12,000+ posts and 7 1/2 years.... the development has exceeded 50% of the initial tenure of the Ariel, we went from a 12 year old design to nearly 20 years old without an endpoint.
It's like playing the lottery, effort without return but with the occasional nugget that fools you into coming back.
Last edited:
GPA 416 Alnico Classic Series in a 3 to 4 cubic foot closed box. Low diffraction in the horizontal plane. Optional subwoofer; Gary's using the Acoustic Elegance driver with 2 passive radiators.
Radian 745NeoBe combined with Azura AH425. Mounted in free air at least 2" above the bass cabinet.
Bessel/Gaussian 3rd-order HP and LP, with Zobel compensation on the LF section. About 15 dB of attenuation for the HP section, with an RC shunt to raise levels above 5 kHz.
This version exists, and is available for audition in the Pacific Northwest. There is another pair in Canada, built by another builder, along the same lines. There may be others we don't know about.
Badman, it sounds like your system has an overall similar topology, just with different drivers and horns. Since this topology goes back the Lansing Iconic of the Thirties, it's not surprising it turns up again and again in the studio-monitor world. There just aren't that many ways of combining 15" drivers with a MF/HF compression driver.
Radian 745NeoBe combined with Azura AH425. Mounted in free air at least 2" above the bass cabinet.
Bessel/Gaussian 3rd-order HP and LP, with Zobel compensation on the LF section. About 15 dB of attenuation for the HP section, with an RC shunt to raise levels above 5 kHz.
This version exists, and is available for audition in the Pacific Northwest. There is another pair in Canada, built by another builder, along the same lines. There may be others we don't know about.
Badman, it sounds like your system has an overall similar topology, just with different drivers and horns. Since this topology goes back the Lansing Iconic of the Thirties, it's not surprising it turns up again and again in the studio-monitor world. There just aren't that many ways of combining 15" drivers with a MF/HF compression driver.
Last edited:
Has it been 7 years! It's flown by! 
I've enjoyed this thread from afar. I've related all this to my little Sony E44 speaker:
Great bass driver. There's something special about high efficiency and low mechanical loss. Troels Gravesen is still interested too and is promising an update: TQWT-
The Duplex and the Iconic are siblings IMO. Take your pick.
IF you can get away with it, BW3 seems the way to integrate drivers IMO. BUt maybe a 15" driver is just too much of an ask. Where's the cone breakup, maybe 1.2-1.5kHz? It would all be a lot easier crossing at 2kHz, with a couple of those 8" Sony drivers in an MTM.
I've enjoyed this thread from afar. I've related all this to my little Sony E44 speaker:
An externally hosted image should be here but it was not working when we last tested it.
Great bass driver. There's something special about high efficiency and low mechanical loss. Troels Gravesen is still interested too and is promising an update: TQWT-
The Duplex and the Iconic are siblings IMO. Take your pick.
IF you can get away with it, BW3 seems the way to integrate drivers IMO. BUt maybe a 15" driver is just too much of an ask. Where's the cone breakup, maybe 1.2-1.5kHz? It would all be a lot easier crossing at 2kHz, with a couple of those 8" Sony drivers in an MTM.
Thank you very much. That is a lot to listen to and I am going to have a lot of fun exploring the new music.
If you don't mind, could you post a few more suggestions as the thread goes along? If you listen to a nice piece of music, please share it or add it as an addendum to a post. It would be highly appreciated.I must admit that the systems I have had would not pass muster with most guys here (old Pioneer transistor receivers with 2% rated distortion, currently self-powered speakers with 4" woofers, etc.), but they have allowed me to enjoy music. Right now I am busy slowly building up a good system. I have several parts and are in the process of building or acquiring others. For example I picked up a Denon DCD-1630G CD-player for a song the other day. It is a 10kg beauty, with dual PCM1701P mono DACs, and I plan to Lampizise the output. I am also talking to Lance Cochrane and he is going to guide me to build my own EL84 P-P amp (I managed to track down a nice chassis with custom transformers to use for the build). The speakers will probably be a variation of the Lampizator P17 idea. Later on I want to build a version of the Lampizator P24, but using field-coil drivers. But that will have to wait. Anyway, thanks for the great info. I am loving the thread.
Badman brought up a good point, and I'd like to develop it a bit more.
If you put a fully restored Lansing Iconic (from 1936), Badman's speaker, the Gedlee Summa, and Gary Dahl's system all side-by-side, they'd sound fairly similar, despite spanning nearly eighty years of technology.
The efficiency of all four speakers would match, mostly likely, within a couple of dB, since the 15" woofers pretty much all clock in around 97 dB/meter. They all use paper cones, and similar spiders and surrounds. The magnets would vary from field-coil (the only possible method in 1936), to Alnico to ceramic to neodymium, but the strength in the gap would be fairly close. Two bass drivers would be underhung (the Lansing and the Altec/GPA 416), and the other two would be overhung, with modern flux-modulation control.
The crossover frequencies would be close too, from 700 Hz to 1 kHz. That's not a big difference, and is dictated by the the first HF resonance of the bass diaphragm material and the acoustics of horns.
The compression driver diaphragms range from stamped duralumin (an aircraft alloy of aluminum) sheet to polyester to beryllium, so would have different onset of first breakup and different breakup patterns above that. All of them should be pistonic in the 1~5 kHz range, though. All four systems have circumferential phase plugs.
The horns are completely different, and the latter-day systems take advantage of modern simulation and analysis tools to deliver much smoother impulse and frequency response.
Similarly, the modern crossovers, particularly the Gedlee Summa crossover, are far more sophisticated than the very simple 12 dB/octave crossover of the Lansing Iconic. Unfortunately, Altec stayed with basic 12 dB/octave crossovers until the mid-Seventies, while the British charged ahead with the BBC's and KEF's Target Function Design, computer simulation on DEC minicomputers, and FFT measurements.
So all four systems are pretty similar up to 500 Hz, and then diverge above that. But I'd guess they all have similar characters in terms of dynamics and efficiency, and would sound quite different than the usual 6~8" cone-and-dome shown at the RMAF.
If you put a fully restored Lansing Iconic (from 1936), Badman's speaker, the Gedlee Summa, and Gary Dahl's system all side-by-side, they'd sound fairly similar, despite spanning nearly eighty years of technology.
The efficiency of all four speakers would match, mostly likely, within a couple of dB, since the 15" woofers pretty much all clock in around 97 dB/meter. They all use paper cones, and similar spiders and surrounds. The magnets would vary from field-coil (the only possible method in 1936), to Alnico to ceramic to neodymium, but the strength in the gap would be fairly close. Two bass drivers would be underhung (the Lansing and the Altec/GPA 416), and the other two would be overhung, with modern flux-modulation control.
The crossover frequencies would be close too, from 700 Hz to 1 kHz. That's not a big difference, and is dictated by the the first HF resonance of the bass diaphragm material and the acoustics of horns.
The compression driver diaphragms range from stamped duralumin (an aircraft alloy of aluminum) sheet to polyester to beryllium, so would have different onset of first breakup and different breakup patterns above that. All of them should be pistonic in the 1~5 kHz range, though. All four systems have circumferential phase plugs.
The horns are completely different, and the latter-day systems take advantage of modern simulation and analysis tools to deliver much smoother impulse and frequency response.
Similarly, the modern crossovers, particularly the Gedlee Summa crossover, are far more sophisticated than the very simple 12 dB/octave crossover of the Lansing Iconic. Unfortunately, Altec stayed with basic 12 dB/octave crossovers until the mid-Seventies, while the British charged ahead with the BBC's and KEF's Target Function Design, computer simulation on DEC minicomputers, and FFT measurements.
So all four systems are pretty similar up to 500 Hz, and then diverge above that. But I'd guess they all have similar characters in terms of dynamics and efficiency, and would sound quite different than the usual 6~8" cone-and-dome shown at the RMAF.
Last edited:
Lynn
All those speakers will certainly sound more similar to each other than they will sound to most other speakers. And I think that there is a very sound (pun intended) reason for this. IMO all speakers will converge to sounding the same as they get better and better - they will all begin to sound like the recording - nothing more and nothing less.
All those speakers will certainly sound more similar to each other than they will sound to most other speakers. And I think that there is a very sound (pun intended) reason for this. IMO all speakers will converge to sounding the same as they get better and better - they will all begin to sound like the recording - nothing more and nothing less.
And for some strange reason, , I always thought that was the goal - for the recording to sound like the recording. I have zero interest in what a speaker "sounds like" - its sole purpose, and that of the entire system, is to present what's on the recording with the least contamination possible. Decent playback systems should all sound almost exactly the same - because they're presenting the recording ... not themselves ...
, I always thought that was the goal - for the recording to sound like the recording. I have zero interest in what a speaker "sounds like" - its sole purpose, and that of the entire system, is to present what's on the recording with the least contamination possible. Decent playback systems should all sound almost exactly the same - because they're presenting the recording ... not themselves ...
RECIPE 10 23 2014
I got it! I have titled my post with "RECIPE 10 23 14", perhaps if we follow this convention as development continues, and title each post indicating a new iteration with a complete summary like yours above with "RECIPE" and the date, we can make finding the variants easier going forward.
I know my critique was harsh, but it's been a long time
BTW, here's a system pic for my rig as it stands today:
I got it! I have titled my post with "RECIPE 10 23 14", perhaps if we follow this convention as development continues, and title each post indicating a new iteration with a complete summary like yours above with "RECIPE" and the date, we can make finding the variants easier going forward.
I know my critique was harsh, but it's been a long time
BTW, here's a system pic for my rig as it stands today:
Metrification has only taken 150 years, so far...
XKCD has soemthing to say on this
Great idea but please use ISO 8601 dates -eg 2014-10-24 so that we're all going to be able to find them.
XKCD has soemthing to say on this
Badman, development info about Gary's system (and mine) may not be useful to you. I'm no expert on OSWG's, but I do know they require significant equalization, and the expert on that subject is Dr. Geddes. I have no idea what the correct measurement protocol for an OSWG is, nor how to do weighted spatial FR averaging over a measurement window in order to come up with appropriate equalization for the crossover.
The AH425 LeCleac'h with a T ratio of 0.7 is a very different animal. It is optimized for uniform diaphragm loading vs frequency, and flat response on-axis. Off-axis, it mimics a physical direct-radiator that's about 1.5~2X the throat size, so it gets pretty directional at 10 kHz. It is free of off-axis peaks and nulls, but the off-axis rolloff is certainly there. This results in a directivity index (DI) that rises steadily from the 700 Hz crossover point.
Although Gary is happy with the subjective results he is getting, I am looking for methods to offset the narrowing DI with frequency. A supertweeter with wide dispersion is an obvious choice, but it has to be time-aligned for satisfactory integration with the MF driver. A more offbeat method is taking a cue from AudioKinesis and use up-firing driver(s) to floodlight the ceiling with spectrally shaped HF content. It seemed to work for Duke at the Rocky Mountain show; there was no impairment of image quality that I could hear (at least in a show setting), and the spatial impression opened up quite substantially, so the system sounded more like an electrostat.
The design challenges of the two horn/waveguides are different. Dr. Geddes has made it clear in other posts there's more to equalizing an OSWG than taking a few measurements and hoping for the best. The OSWG is a proprietary technology, and not in the public domain; we shouldn't expect Dr. Geddes to disclose everything here at diyAudio.
The AH425 is close to flat on-axis, and only requires a minor HF lift for the Radian 745NeoBe driver. But the dispersion pattern, although smooth, is narrower than I would like at high frequencies. I want to retain the LeCleac'h profile, and will look into methods to increase dispersion while gradually taking over from the Radian driver at high frequencies.
The AH425 LeCleac'h with a T ratio of 0.7 is a very different animal. It is optimized for uniform diaphragm loading vs frequency, and flat response on-axis. Off-axis, it mimics a physical direct-radiator that's about 1.5~2X the throat size, so it gets pretty directional at 10 kHz. It is free of off-axis peaks and nulls, but the off-axis rolloff is certainly there. This results in a directivity index (DI) that rises steadily from the 700 Hz crossover point.
Although Gary is happy with the subjective results he is getting, I am looking for methods to offset the narrowing DI with frequency. A supertweeter with wide dispersion is an obvious choice, but it has to be time-aligned for satisfactory integration with the MF driver. A more offbeat method is taking a cue from AudioKinesis and use up-firing driver(s) to floodlight the ceiling with spectrally shaped HF content. It seemed to work for Duke at the Rocky Mountain show; there was no impairment of image quality that I could hear (at least in a show setting), and the spatial impression opened up quite substantially, so the system sounded more like an electrostat.
The design challenges of the two horn/waveguides are different. Dr. Geddes has made it clear in other posts there's more to equalizing an OSWG than taking a few measurements and hoping for the best. The OSWG is a proprietary technology, and not in the public domain; we shouldn't expect Dr. Geddes to disclose everything here at diyAudio.
The AH425 is close to flat on-axis, and only requires a minor HF lift for the Radian 745NeoBe driver. But the dispersion pattern, although smooth, is narrower than I would like at high frequencies. I want to retain the LeCleac'h profile, and will look into methods to increase dispersion while gradually taking over from the Radian driver at high frequencies.
Last edited:
I got it! I have titled my post with "RECIPE 10 23 14", perhaps if we follow this convention as development continues, and title each post indicating a new iteration with a complete summary like yours above with "RECIPE" and the date, we can make finding the variants easier going forward.
I know my critique was harsh, but it's been a long time
BTW, here's a system pic for my rig as it stands today:
An externally hosted image should be here but it was not working when we last tested it.
Dear Mr. Badperson,
I think the dates are a very good idea.
Could you also gives us a rundown on your system as it appears in that picture, or at least point us towards some recent coverage.
Thanks in Advance,
Terry
Good call! Going forward, if we can adopt this convention, let's use the ISO dates. If a mod is watching- can you edit my previous post title to reflect this? (Post 12048, http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-241.html#post4098781 )
Last edited:
Well, the nice thing about constant directivity devices is that the axial EQ is largely the same as the spatially averaged EQ- and the reflections within the room (or at least the excitation energy of the surfaces) are similarly balanced, so you only have to manage one, relatively consistent, response, as opposed to having the disconnect of high axial 10kHz energy but minimal reflected.
I'm interested in your system for the details, and the thread has value, I don't mean to imply otherwise, but those wanting to build a design with the elegance your name implies need to be able to lock things down and forge forward at some point, which is all I'm getting at. A companion Wiki would be welcome but I don't expect you want that trouble so the dates and recipe workaround may help make this thread more accessible.
Short rundown- diy waveguides of a smoothed conical profile, very close but not perfect OSWG a la geddes on top of BD15 neo underhung woofers in a low diffraction cab, vented with an overdamped tuning. Amp is a tripath deal, a modded motorola receiver of all things- it works nicely. Preamp is transformer volume control based in a breadboarded kind of configuration, and sources include a Linn LP12 with dynavector 17d2 into a stepup and Bel Canto phono 1 phonostage, and a diy dac run primarily off a squeezebox touch. Subs are passive radiator 15" with active 12"s, in about 3.5 cu ft.
As Lynn says, there are only so many ways down the street, and my development path was very different than B.T.A: I build and build and build, though individual projects are often on a long time scale, I usually have several in parallel and eventually have arrived at the current configuration.
During the time of this thread I've:
Moved cross country for a new job
Been promoted 3 times within that job, as well as later downsized
Had a child go from diapers to 3rd grade
Had 2 surgeries on my right leg (one ACL reconstruction, one hammertoe correction)
Built a lot of gear, from heil AMT waveguided dipoles to the present system and back again. I've done at least a dozen "complete" speaker designs, of varying degrees of quality, overlap, and execution.
The system shown actually has been adjusted several times just in the past couple weeks, with a rear firing section (basically a miniature version of the big dog with a 12" and smaller waveguide) being relegated to the garage due to wanting to play with the tripath amp (and connectivity/loading associated with that), and the horns have been elevated and shifted backwards somewhat from the shown picture to reduce diffraction of the midrange off the horn and better steer the vertical lobe. Additional absorption has been added to the foam ring around the perimeter of the waveguide as well, doubling its thickness and further preventing interference patterns from midrange diffraction and from the display.
It's a very different way of doing things than Lynns- I want to get things live, then refine from there, and I've found that for me that process works best. When I know I need something to be executed well, I can take my time and make nice cherry-wrapped rounded cabs as shown, but most of the time the process is more about learning than the end result. I believe that there are many valid system approaches, and while there's definitely an absolute quality scale for speakers, the priorities rearrange systems and system elements on this scale for each listener, within reason. Indeed, the most valid question most speaker salesmen should be asking (but don't do so much anymore) is "what are your favorite albums?" - not "What kind of music do you listen to?", which invariably leads to the "All kinds of stuff...."
I have found that I can find things to like about each well-executed concept I play with, and those strengths play to different preferences and different musical types. Like Lynn, I could never live with Lowthers, but I've also had some very enjoyable musical times with those and other wideband loudspeakers. There are things that they do exceptionally well and when working with such a loudspeaker, I tend to optimize it as much as I can, then enjoy the positive qualities while I dream of my next concepts to try.
This process has lead to a garage full of cabinets and drivers.... but it's very enriching.
The garage full of gear has also allowed me to share more with friends- just this very evening I threw together a sub based on on-hand parts, for my brother.
Earlier this year read a report on the emotional connection to music. Something like 30% showed no emotional connection to music when viewed from an MRI. The same response was noted in the prexam questionaire. These disconnected types see music more as noise and generally disruptive as volume increases.
I tend to think that those whom feel the music most pursue excellence in reproduction. Those that are generally ok with music are fine with just ok reproduction. The others simply could careless, dispite that $100k system that is only turned on when company is over or the house is empty and the maid is rockn it out
IMO. Music is not race. It is a collection of sounds/noise. Each person resonates with their particular collection. The similarity that music has with religion, IMO is the lack of tolerance. If it touch's the soul, so be it.
SMathews
Music
Another favorite: Shadowlands, by George Fenton and the Choir of Magdalen College, Oxford.
This one has quite an emotional tug for me because Bob Sickler and I listened to it together in his last months. I have a deep respect for Bob because he was my mentor back in my days at Audionics in the Seventies. This music is especially appropriate for somebody that's going to be making the transition and their friends; it's what the movie and the soundtrack are about.
Every time I listen this CD, it stirs memories of singing in the choir at KGV in Hong Kong, the zany things Bob and I got up to at Audionics (which ultimately got us fired), and his peaceful final months at his beautiful home overlooking the lush green valleys of Oregon, memories spanning the Sixties, Seventies, and early Nineties.
If you want to know why I listen to music and design hifi equipment, this is the answer. It's been a part of my life since I was ten years old.
Another favorite: Shadowlands, by George Fenton and the Choir of Magdalen College, Oxford.
This one has quite an emotional tug for me because Bob Sickler and I listened to it together in his last months. I have a deep respect for Bob because he was my mentor back in my days at Audionics in the Seventies. This music is especially appropriate for somebody that's going to be making the transition and their friends; it's what the movie and the soundtrack are about.
Every time I listen this CD, it stirs memories of singing in the choir at KGV in Hong Kong, the zany things Bob and I got up to at Audionics (which ultimately got us fired), and his peaceful final months at his beautiful home overlooking the lush green valleys of Oregon, memories spanning the Sixties, Seventies, and early Nineties.
If you want to know why I listen to music and design hifi equipment, this is the answer. It's been a part of my life since I was ten years old.
Last edited:
Quote from Lynn (highlite my edit)
Thank you for this. This is the true culmination point of this thread. A highly respected authority opens up and confesses that a flooder actually improves the sound.
This ceiling firing tweeter method is very old. Earliest I have came accross is from G.A.Briggs, the SFB/3.
Surely this should be well known, but extremely rarely executed in modern designs.
I only wonder why... Because it works very well in normal domestic rooms.
.
Attachments
This is something I'm on board with- and I'd add that secondary sources come in many flavors. I remember one eye-opening experience, when I was working with 2 channel system, with a display in between. After a casual power up and starting up a laserdisc, the system sounded really great one day, I realized that the tv speakers were left on, albeit at about 15 dB down from the mains. Filling in comb filter notches? Spreading modal behavior in the room, just a bit?
Since then, I've been a fan of ambience tweeters, flooders, bipoles, dipoles, and what have you- in systems like the BTA, where treble is so beamy, you can use it to get the balance of reflected energy closer to where the direct energy is, and simplify the balanced EQ process. Sure it's delayed and thus extra impulse noise, but if we respond to it well, then the improvement is worth the price.
I have been to Dr. Geddes' place. I have yet to hear a LeCleac'h horn. Dr.G's place was as good as I've ever heard and I've been to half million dollar rooms as part of the NYAR. In my mind that is some proof that room setup, acoustics and speaker design is way more important than amps, cables and other things audiophiles obsess about. At least in my mind, and I very much like tube amps and triode design.
I really hate beaming HFs, so I get your objectives Lynn. One of the things I like about oblate spheroidal waveguides. I am mimicking your and Dr. Geddes's designs with a whole lot of horns and waveguides at my disposal.
I get the loading argument of horns, but I don't think that is the absolute argument as I think that waveguides and the sensitivity of CDs should provide a leap over standard direct radiator.