Beyond the Ariel

This forum is as a good a place as any for thinking out loud about the successor to the Ariel and ME2's. Since my fall in the terrible Colorado snowstorms of January, I've been recovering, and probably won't be doing any serious audio design until the summertime.

But I can still think and write, and have been making occasional posts in John Atwood's Clarisonus blog. One perennial question I always get is how to build an Ariel with XYZ drivers (and of course a different crossover and cabinet). As mentioned in my Web page, it take me about six months to design a speaker, and isn't something I do on a whim, or something to "use up" some drivers that are lying around.

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, even though the horns I've heard here in Colorado are some of the best I've ever heard anywhere. Fabulous micro and macro dynamics, but I still have reservations about tonality with classical music, especially the big-scale choral music I enjoy so much.

The speakers design 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. Interestingly, both the Linkwitz and Bastani designs get to similar place via completely different methods: Linkwitz with extensive equalization and multi-amping with high-power transistor amps, and the Bastani with extensively modified 12" near-full-range drivers, and a subwoofer and supertweeter to fill out the range.

Both the Linkwitz and Bastani have stunning dynamics - almost in horn territory without the colorations - and the spaciousness and 3-dimensionality of electrostatics. I can see why many Europeans are turning to open-baffle designs: they do things that never happen in conventional speaker boxes.

Of course, there's never a free lunch: dipole speakers have unique design challenges all their own. The most obvious is the tilted frequency response that starts a half-wavelength below baffle width. Linkwitz attacks this with multiple crossovers, multiple drivers, and complex low-level equalization, which straightens everything out again. Bastanis, by contrast, stretches a single 12" driver to its limits, and helps it out below 220 Hz and above 8 kHz with a subwoofer and tweeter.

What I'm contemplating is a bit of both and a little different: a prosound 12 or 15" coaxial driver, with efficiency of about 97 to 99 dB/metre, and a supplemental 12 or 15" driver that comes in below 160 to 250 Hz (switch-selectable to match room characteristics). The upper driver runs full-range, and is acceptable with any Qts. The lower driver has a Qts from a 0.6 to 0.8, and carries the bass in the region where the dipole 1/f effect starts to be significant. Below 80 Hz, bass is carried by stereo subwoofers in close proximity to the dipole array, and the subwoofers are independently powered by their own amplifiers.

The dipole array as I'm imagining it has either 2 or 3 drivers: one full-ranger, of very high quality, and one or two bass-fill drivers, with tapped air-core inductors adjusting the overlap region. Dipoles in particular need to compensated for room-response, due to their unusual radiation pattern, and the inability of the user to optimize imaging and distance from the wall/corner at the same time. By adjusting bass-fill driver overlap and subwoofer levels, the response in the 20 to 500 Hz region can be optimized for the room, without compromising image quality (which is controlled by the mid and HF response of the driver).

Yes, this is very different speaker than the Ariel, although I hope to retain the midrange spaciousness and natural voice quality in the new design, along with a 15 to 20 dB increase in dynamic range.

75For the current state of this project (known as the LTO) see here:
https://production.diyaudio.com/community/index.php?posts/3189501/
And here:
https://production.diyaudio.com/community/index.php?posts/3190604/

75
 
Lynn, I'm sorry to hear you've had a fall -I wasn't aware of that until this post, but you have my very best wishes for a speedy and complete recovery.

You're tuning toward dipoles I find fascinating: quick anecdote, which you might find of interest. I was recently at a local audio meet (including a couple of Ariel owners BTW) where an interesting variation on the dipole was lurking. Powered by vintage 300a drivers from a rather special SET amp, they sang.

Nothing fancy at all. Cheap Eminence Beta 15 bass driver mounted fairly close to the floor, XO to a Fostex FE167E at ~180Hz. Series XO, utilising the natural roll-off of the drivers. The baffle is relatively narrow, roughly 16in, triangular side-pieces forming a semi U frame, roughly 12in deep at the bass, tapering to nothing at the top. Sonics were downright staggering. For the megre price, I've heard little that gets close. Easy 40Hz, seemless transition. I went hunting for the XO, and I flat couldn't find it. Design was by JamesD, who frequents the World Designs forum, and has much fancier & most expensive acrylic versions -effectively a very highly modified version of the original Quasar using a 12in Supravox bass unit and AER MK1 up top. Bass isn't really dipole in radiation pattern in this layout, but it certainly sounds like it, and the things can be used close up to a wall or in corners will little degredation. Don't ask me how or why -it's beyond me! Looks like you're working in the same direction -I look forward to seeing what you come up with!

Best regards
Scott
 
diyAudio Editor
Joined 2001
Paid Member
Hey Lynn, welcome back!

I didn't realize that you are a member here.

I used to write you often about the Ariels with my newbie ideas and questions. You always answered and in fact took the time to consider my ideas and also explain why they might have problems or why a few of them were OK. Still have some of your e-mails. Thank you for that.

I had a lot to learn, so never built the Ariels, but probably cost you a fortune just for the number of times I visited your sites!

I ended up making the Basszilla speakers by Dick Olsher, with different 15" bass drivers, the Fostex 208SE in an open baffle, and the Aurum Cantus 2si ribbon tweeter. I managed to get him into discussions also, and I believe had some minor influence into the drivers he used and I built the tilting baffle he used as a testbed.

A lot of fun speaker projects have appeared recently and I 'm glad you are going to share your thought processes here.

Variac
 
More thoughts and comments on electrostats

SY said:
What advantages do you (or anyone else) see over large-panel ESLs?

Stupendous dynamics combined with big-as-the-outdoors spatial and out-of-the-box quality. Normally, electrostats have that wonderful free-and-easy open-air quality that is so wonderful and so characteristic of electrostats. The nearly-massless diaphragm is part of that, but the absence of box coloration is a bigger part. Try putting electrostats in a box - of any kind, any shape - and see just how much of the wonderfulness disappears.

Parenthetically, I've been informally test-auditioning mid and midbass drivers in an IEC-style baffle for years - in fact, that's how I chose the drivers for the Ariel, as well as using the IEC baffle for a measurement platform. It just never dawned on me the rolloff problems could be conquered in a successful way by "working the problem" as Linkwitz and Bastini did.

What Rob Bastini brought to the table were prosound efficiencies of 97 to 99 dB/metre, instead of the audiophile efficiencies of 87 to 89 dB/metre of the Linkwitz designs, or the even lower 82 to 85 dB/metre of electrostats. I can tell you, that changes everything, combined with the insane headroom of prosound drivers.

Typical audiophile drivers start to exhibit dynamic compression from voice-coil heating at levels as low as 95 ~ 98 dB - and of course, 98 dB is nearing the danger zone for most electrostats. By contrast, prosound drivers, especially those designed for studio-monitor use, are quite happy with 115 to 125 dB. Once you get used to horn dynamics, you become very aware of just how compressed most audiophile drivers sound in comparison. Horns, unfortunately, still sound like horns, although the very best have diminished the coloration quite considerably.

(Where I get into wrangles with horn enthusiasts is their assertion that horn coloration has been eliminated for all practical purposes, or doesn't matter to "true" audiophile. Uh, not to me, and I design speakers for my own enjoyment, not the marketing department of Harmon International.)

The truly memorable part of listening to the Bastani was the combination of wide-open spaciousness and sound that had physical impact. Usually, you get one or the other, but not both. Horns, after all, aren't known for wide-open spatial qualities, with the soundstage only rarely extending beyond the speakers. Dipoles with prosound drivers have both qualities.

Some people prefer a restricted soundstage, and consider extra-width and extra-deep effects artificial. I don't - I've worked in surround sound since 1973 (the Shadow Vector patent and prototype with Audionics) - and know the difference between artificial and genuine reproduction of spatial characteristics. A good dipole, whether electrostatic or dynamic, plays the recorded ambience as recorded, not in a fake-reverb kind of way.

This is because the brain/mind/hearing system assigns sound into different processing "slots": 0-1 mSec is used for localization, 1-25 mSec is used for assigning ambient qualities (how big is the space), and anything delayed more than 25mSec starts to sound like an echo. So measures to reduce diffraction and stored energy in any kind of speaker are worthwhile, since most of the stored energy is going to fall into the 0 to 1 mSec interval, where it damages the sense of localization. Room reflections, including the first one off the floor, are benign - in fact, the floor reflection, although troublesome to measure, actually assists localization.

This is why dipoles don't "create" ambience, the simply allow you to notice what's on the recording. If you're listening to a 1944 Toscanini recording, it's going to sound as dry as the Sahara Desert. Chamber music sounds just like musicians playing in a small, lively room. Symphonies sound like they're in big halls, as they should, although running the solo mikes up and down is really obvious and silly-sounding, since the lucky instrument zooms forward and then scurries away once it leaves the spotlight.

The most noticeable - and unique to prosound dipoles - is the sheer sense of presence of big-sounding instruments like piano and kettledrums. These things sound real, not hifi. Electrostats just don't have the gigantic dynamics these instruments require, and with horns, you don't get that sense of the sheer size of the instrument, although you get the loudness. And to my ears, the audiophile line arrays sound artificial, overhyped, and incoherent. (Not surprising; when you have a multitude of drivers delivering the sound at slightly different arrival times - think of the path-length differences - the sense of impact and coherence isn't going to be there, thanks to the arrivals being smeared over that critical fisrt millisecond.)

Anyway, if anyone reading this little thread is truly ambitious, try the dipole thing with a 15" paper-cone Tannoy, and prepare to be amazed. The less you hear the cabinet, the better a driver is going to sound. If the driver is truly great, you won't hear it's full potential in a box - any box, I don't care what wonder material it's made of.

P.S. I broke my leg on January 7th, and I'm finding out These Things Take Time. The really awful part lasted less than a day, and since the surgery on January 8th, I've been pain-free, gradually recovering, transitioning from a cast to a removable boot, and more and more proficient in gizmos like walkers and crutches. The X-rays on the February 20th and March 20th still showed a break, but one gradually decreasing in size. I'm hoping (fingers and toes crossed) that the next one on May 1st shows the break all gone, so I can start using a cane to get around.
 
Lynn,

I'll be very interested in reading anything you put up in the way of ideas... and have been since I first saw one of your Ariel articles in the mid-90's ('97?).

Lynn Olson said:
Both the Linkwitz and Bastani have stunning dynamics - almost in horn territory without the colorations - and the spaciousness and 3-dimensionality of electrostatics. I can see why many Europeans are turning to open-baffle designs: they do things that never happen in conventional speaker boxes.

Leaving the more complex boxes out of the equation here, with regards the dynamics, do you think this due to the open baffle design itself (the ease of moving the cone only against air in both directions) versus conventional boxes (where the cone moves against an enclosed volume of air and only produces air movement from one side of the cone)?

What co-ax have you looked at with regards the primary driver? Something along the lines of a Tannoy, possibly?

Anyhow, please recover swiftly and post more often :yes:!
 
Source of Colorations

Cloth Ears said:
Lynn,

Leaving the more complex boxes out of the equation here, with regards the dynamics, do you think this due to the open baffle design itself (the ease of moving the cone only against air in both directions) versus conventional boxes (where the cone moves against an enclosed volume of air and only produces air movement from one side of the cone)?

What co-ax have you looked at with regards the primary driver? Something along the lines of a Tannoy, possibly?



Stored energy grossly degrades the sense of realism in every way: it steals dynamics, compresses and distorts the soundstage (which is why I always audition in stereo), and almost flattens the sense of depth completely. It's like MP3 with extra colorations thrown in free.

Unfortunately, there are many many sources of stored energy, particularly in the first few critical milliseconds. Cabinets have standing waves and wall flexing, horns have significant diffraction at the horn mouth (which then travels back down the throat and reflects off the phase plug), and worst of all, cabinet damping materials don't really work well below 500 Hz. All this is quite audible, particularly when you hear the coloration go away. This kind of coloration is the kind of thing you notice in the first few seconds of listening - and we only tolerate it because that's how 99.999% of speakers sound, and have always sounded since the Twenties.

Electrostats store energy too, although far less than direct-radiators in boxes or any horn. Where the electrostat falls down are a series of small mylar-diaphragm resonances between 3 and 10 kHz, which is noticeable on a MLS waterfall measurement (although easily confused by different arrival times from different parts of the diaphragm).

All drivers fall down here, compared to what can be done with plasma drivers. In my experience - speaking for right now, anyway - cabinet colorations are considerably worse than driver colorations, at least with decent-quality modern drivers. With an old-school driver like the Altec 604 Duplex, OK, I doubt that cabinets make a lot of difference. But when coloration levels go down, cabinet coloration becomes much more noticeable, and is very difficult to reduce. Exotic cabinet materials really only help a little bit, and introduce odd new colorations of their own. (It takes magazine reviewers several years to realize a new type of coloration has been introduced.)

As for new drivers, well, I've tried getting OEM Tannoys for many years, with no luck. As far as I can tell, Tannoy only wants to sell complete systems, without the nuisance-factor of the DIY market. Can't say I blame them. So I'm looking at prosound folks like 18Sound, B&C, and Beyma.
 
  • Like
Reactions: 1 user
Check out the coax thread elsewhere

There are some good suggestions in the coax thread going on elsewhere at diyAudio: I spotted this driver from BMS of Germany, who make very serious prosound drivers:

12" BMS Coaxial

The complete list of BMS drivers is here

Now, the specs for coax drivers have to read very carefully. The combined curve is useless; what you want are the separate curves for the woofer and tweeter, paying particular attention to the rolloff region of the woofer. In a high-quality speaker, this needs to be as smooth as possible, both to decrease coloration and assure a smooth phase hand-off to the tweeter. Just because it's a coax doesn't mean you can get away with sloppy crossover design.

Unfortunately, the horn in front the woofer diaphargm creates standing waves, and these create reflections (stored energy, remember?) in that critical 0-1mSec timeframe. The frequency response curves for the woofer tell what you need to know about the potential candidate - can it be reasonably equalized, and will it sound good afterwards? Radical equalization has a way of sucking the life out of the driver, and leaving it alone is a prescription for an annoying coloration that ultimately becomes tiring to listen to.

I mention the BMS 12C362 above because the EQ for the woofer looks pretty simple: a notch filter at 2.2kHz, about one octave wide, and about 5 dB deep. Use a 2 kHz 12dB/oct or 18dB/oct highpass and a shelf filter on the tweeter, and there you go. Note this driver is 98dB/metre, and easily plays up to 124dB. Is that enough dynamic range?
 
Disabled Account
Joined 2007
Re: Source of Colorations

Lynn Olson said:


As for new drivers, well, I've tried getting OEM Tannoys for many years, with no luck. As far as I can tell, Tannoy only wants to sell complete systems, without the nuisance-factor of the DIY market. Can't say I blame them. So I'm looking at prosound folks like 18Sound, B&C, and Beyma.


Have you consiered Eminence? Also worth looking at:http://www.maxspeakers.com/products/fidelity/index.html

They have pro drivers up to 54cm, paper cone, 100+ db.
 
Those of you who are thinking seriously about this speaker format might want to read another thread also, assuming you haven't already.

I start you off where the ultra low storage termination information begins and there are two very different methods for perfectly terminating a big driver without a bunch of storage mass.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=30689&perpage=10&pagenumber=7


And yes this is a thread here in DIY

Bud
 
Re: Source of Colorations

Lynn Olson said:
Stored energy grossly degrades the sense of realism in every way: it steals dynamics, compresses and distorts the soundstage (which is why I always audition in stereo), and almost flattens the sense of depth completely.

A fellow audio enthusiast at the Melbourne Audio Club also has found this, but actually as a result of building his last pair of speakers. I don't think they were intended as such, but that is what he thinks now. He says that he has pretty much given up the idea of building anything else and finds box speakers are now far too colored to go back to. He likes to listen to them at the club meets but hasn't heard anything he prefers since the set-up was completed.

A small indication of them is here at Bob F's hybrid ribbon speakers. I'm sure I can get more details if you wish.

I do understand your intent is to use available cone drivers/co-axials, so I won't OT this thread anymore.
 
Ah Lynn.. my favorite in the HiFi "world".. :)

While I'm generally no fan of horns.. consider that a good bit of a cone driver's bandwidth at higher freq.s is in fact horn loaded. The larger the driver the lower the freq.. and the worse it is - bastini's in particular. Just about any coax has a similar reality. Really flat drivers (the manger in particular) of course "buck" this trend to some extent depending on their profile.

Worse still, the "horn" for these cone drivers is moving around.:smash:

Heck.. even a baffle acts as a horn. ;)

Also note that most pro drivers have increased distortion at lower spl's near the top of their passband (at higher freq.s) due to larger VC's (when compared to better hifi drivers). Both linear and non linear. Additionally, while they do exhibit less thermal compression.. they often suffer from mechanical compression that won't allow a lower freq. response anywhere near their max spl rating (..not that it should be a problem for anyone in a domestic setting - unless they are insane, or deaf and neighbor-less).

Despite the above.. I'm pretty much lock-step with your views. I just like hi eff. drivers better, more energetic/dynamic and IF they are of lower mass then they seem to offer greater clarity. Also, I've noted that linear decay is critical at least 9-10 db when factoring time - the shorter the better. Oddly though, HOW you achieve a cleaner decay is extremely important. Drivers with high internal loss and only moderate or worse ability to react to change, despite having *overall* cleaner decays (i.e. are "cleaner" further in time and level) sound significantly less clear. Likewise, drivers with acoustic resistance (particularly near time acoustic resistance), often via box stuffing and panel friction, kill ambiance and make the sound more monophonic (and "attached to the speaker").

I however specifically do NOT prefer a dipole pattern above the modal region - it creates a "sameness of sound" from recording to recording.. and in comparision to a really good enclosure falls short on depth. My preference is for a radial in the midrange and a wide dispersion pattern higher in freq. approaching 180 degrees (..when considering the front wall in a listening room). This does make several demands on both the "box" and the driver..
 
Great to see you here Lynn! I've been pestering you about the Karna design for months, going on a year. I too am working on OB speakers, namely the Arvo Pärt, or a variant thereof.

If I could, I'd suggest you take a brief read through the "Isiris" thread on htguide.com (under mission possible diy). This is an evolution parrallel to Linkwitz's own evolution from the Pheonix onward. This design, although not set in stone, is taking some tools from the pro audio world in its development.

There may be some useful info in that project that may help guide your own.

Hope you get better quickly!