AR-3a Response Curves and Test Reports

[Zilch]

Look again at the anechoic on- and off-axis response of the 3/4-inch tweeter, same as used in the later AR-2ax. For entertainment, I've compared the performance of the waveguide tweeter, which is not generally rated up above 15kHz, to that of the AR 3/4-inch dome tweeter. Granted, these were not measured in the same venue under the same conditions, but the off-axis characteristics are clearly evident, and the dome tweeter is far superior off axis than the waveguide, which is really not rated off-axis beyond 15kHz, as I said. Note too, that the waveguide is charted at 10dB/division vs. 5dB/division for the AR tweeter, so the waveguid looks smoother in that presentation, and more linear in its off-axis decline. It's a less-rigorous method of measuring, for that matter.

Apples and oranges, Tom, classic AR "Bait and Switch" subterfuge, which Howard has already tried to pull off three or four times ahead of you here without success. You're comparing the in-system performance of a 90° beamwidth waveguide to the AR3a tweeter measured on a test baffle. Alas, that's NOT how it performs in AR3a:

http://www.classicspeakerpages.net/IP.Boar...ost&id=4533

I post my measurements because they are in color, and you can see what's what, but if you compare to those presented in Allison & Berkovitz, measured 40 years ago, you'll see that they are substantially the same, so much so that you can easily tell if Allison's left and right are inboard or outboard measurements.

You're calling for higher resolution and measurement to 20 kHz, here it is, the EconoWave driver and waveguide, also a 90° design, as measured by Geddes, every 7.5° from on-axis out to 90°, 180° total beamwidth:

http://www.classicspeakerpages.net/IP.Boar...ost&id=4574

That's unsmoothed, as I recall, but you can confirm it on the Gedlee website. I originally posted that here on CSP to demonstrate that the AR3a tweeter's highly touted wide dispersion was rendered all but moot by mounting it in a sub-optimal cabinet design where it can't even match the performance of a contemporary 90° waveguide. AR-303 clearly corrected some of this (when you post the polars, we will see the result of those efforts), as did the AR3a Limited Editions. The actual AR3a system directivity may be calculated from the Allison and Berkovitz data, and it's not at all what everyone has been led to suppose....

[Steve F]

Apples and oranges, Tom, classic AR "Bait and Switch" subterfuge, which Howard has already tried to pull off three or four times ahead of you here without success. You're comparing the in-system performance of a 90° beamwidth waveguide to the AR3a tweeter measured on a test baffle. Alas, that's NOT how it performs in AR3a:

Ah, but it is EXACTLY how it performs in the 10 Pi-11-12, which have essentially NONE of the recessed baffle/obstruction issues of the 3a-5-2ax.

Along with their new ferro-fluid cooled tweeters that allowed a significantly higher voltage drive through the crossover and consequently greater available HF output, the 10 Pi was able to do extraordinarily well in those 1976-77 Neil Grover jazz drummer L v R demos, far more demanding demos than the string quartet L v R demos over a decade earlier. The 10pi and 11 were "max dispersion" designs; one can draw whatever conclusions one wishes.

BTW, the 10 Pi in that demo was used in the '2 Pi' setting, making it acoustically identical to the 11. Victor Campos told me that they used the 10 Pi so that any favorable PR would accrue to the top-of-the-line product.

People can like what they like and believe what they believe.

Steve F.

[tysontom]

Apples and oranges, Tom, classic AR "Bait and Switch" subterfuge, which Howard has already tried to pull off three or four times ahead of you here without success. You're comparing the in-system performance of a 90° beamwidth waveguide to the AR3a tweeter measured on a test baffle. Alas, that's NOT how it performs in AR3a:

http://www.classicspeakerpages.net/IP.Boar...ost&id=4533

I post my measurements because they are in color, and you can see what's what, but if you compare to those presented in Allison & Berkovitz, measured 40 years ago, you'll see that they are substantially the same, so much so that you can easily tell if Allison's left and right are inboard or outboard measurements.

You're calling for higher resolution and measurement to 20 kHz, here it is, the EconoWave driver and waveguide, also a 90° design, as measured by Geddes, every 7.5° from on-axis out to 90°, 180° total beamwidth:

http://www.classicspeakerpages.net/IP.Boar...ost&id=4574

Zilch, thanks for your reply and for the links. By the way, there is a pattern of resorting to personal snipes when you debate. Why don’t you keep it above board. Your reference to AR’s “Bait and Switch subterfuge” and the reference to “max dispersion” shows a negative, emotional side to your arguments. Why don’t you keep it scientific and keep it clean so it can stay out of the Kitchen. I understand where you are coming from, but lets just debate it without insulting one another.

I took a close look at the Econowave driver plot you referenced and its off-axis performance, the one in which you said was measured by Geddes at 7.5° intervals out to 20kHz. I compared that data to the 2π anechoic data on the AR-3a ¾-inch tweeter and the Allison:One tweeter. Granted, the AR and Allison drivers mounted in the cabinet will cause some near-field destructive interference effects, but it has been clearly shown that those early reflections are largely swamped in the far field (move the measurement microphone around to different places, and the interference and diffraction dips and peaks change frequency); so in essence, the dispersion of the AR-3a tweeter/Allison tweeter in anechoic measurements and in-system measurements, overall, are going to be similar. Therefore, using the data you supplied in your link to the Geddes measurement (my table would not tranlate to the CSP site properly, but you get the idea):

Econowave: 20kHz/30°-4dB; 20kHz/60°-15dB; 20kHz/75°-22dB

AR-3a: 20kHz/30°-4dB; 20kHz/60°-4dB; 20kHz/75°-8dB

Allison Tweeter: 20kHz/30°-3dB; 20kHz/60°-4dB; 20kHz/75°-6dB

The interpretation of this data shows the Econowave thing to be more than -20dB down at 20kHz at 75 degrees off-axis compared to the AR tweeter, and much more so when compared with the Allison tweeter. Could there be a mistake with that curve? The off-axis performance isn't great. The vertical scaling still favors the Econowave with its 10dB/division vs. the more critical 5dB/division as used by AR, so the Econowave "looks" smoother than the others.

It only takes a second to realize that the AR/Allison tweeters are infinitely superior in off-axis dispersion to the Econowave tweeter, and these direct-radiator tweeters don’t suffer from the typical coloration you get from a waveguide or horn-type tweeters. If you look at JBL’s literature on the waveguide tweeter, the applications never mention home-type sound reproduction or monitoring purposes; the references in the data sheets I’ve seen show:

• Performing Arts Facilities

• Houses of Worship

• Sports Facilities

• Multi Media Spaces

• Retail Spaces

• Presentation Suites

• Educational Facilities

• Portable Audio-Visual Systems

Therefore, why would we want to use that type of device in lieu of the better direct-radiator tweeters?

--Tom Tyson

[genek]

Ah, but it is EXACTLY how it performs in the 10 Pi-11-12, which have essentially NONE of the recessed baffle/obstruction issues of the 3a-5-2ax.

When the 10pi was first introduced my local audio dealer still had a dozen 3a's in his warehouse, so he set up a side-by-side A/B of the first 10pi's he received and a pair of late model 3a's, which he was marking down to clear out. He had both speakers on a waist-high shelf (more or less 2pi) and had set their level control pots/switches to match as closely as he could. After spending about an hour listening to different recordings on both speakers from various locations in his demo room, I could hear no difference between the two models. Which, of course, was his intent, since he had six pairs of 3a's in stock and only one demo pair of 10pi's. I think he sold out all six pairs of 3a's that weekend. Did the ADD cabinet/grill redesign priduce any audible (as opposed to measurable) change in sound? Not to my ears.

[Zilch]

Thank you, Steve.

I have no data on 10 Pi, but it would certainly be interesting to measure them.

Clearly, AR recognized the "issues" revealed in Allison & Berkovitz and set about mitigating those; the implications are clear.

I agree, people can like what they like and believe what they want -- these are the facts as best I have been able to ascertain them, and they certainly should not diminish anyone's enjoyment of AR3a.

Therefore, why would we want to use that type of device in lieu of the better direct-radiator tweeters?

There is nothing pejorative in the phrase "max dispersion." I am sorry you perceive it that way, as nothing derisive is intended.

1) JBL Pro uses PT waveguides in professional products for professional applications such as those enumerated in their literature; they do not sell to consumers. We have adapted it for DIY use at home.

2) It is a 90° constant-directivity waveguide with a fundamental design criterion of -6 dB @ +/- 45°, and produces very smooth and well-controlled response within those beamwidth limits. However, despite having no aspirations of delivering wide dispersion, even at 75° (150° beamwidth), its worst case response through 20 kHz is +/- 6 dB; reflections from that angle retain the spectral balance of the direct source far more accurately than those generated by AR3a.

3) 2-Pi as AR used it was flush-mounted, masking cabinet edge effects, which are clearly revealed in the 4-Pi measurements of the same speaker by Allison & Berkovitz. This is a replay of prior discussion; if most users built their AR3a's into the wall, it might be relevant, but, as shown in the paper, they were/are rarely deployed in this fashion.

4) While max dispersion may have been an AR and Allison design objective, in retrospect, the appropriate question is "To what purpose?" Our answer thus far is spaciousness, and the consequences of accomplishing enhanced apparent source width via this means are known and acknowledged. Toole teaches that wide dispersion is not required to produce the perception of spaciousness in listening rooms; we may be ascribing far more importance to wide dispersion than it warrants....

[Carlspeak]

When the 10pi was first introduced my local audio dealer still had a dozen 3a's in his warehouse, so he set up a side-by-side A/B of the first 10pi's he received and a pair of late model 3a's, which he was marking down to clear out. He had both speakers on a waist-high shelf (more or less 2pi) and had set their level control pots/switches to match as closely as he could. After spending about an hour listening to different recordings on both speakers from various locations in his demo room, I could hear no difference between the two models. Which, of course, was his intent, since he had six pairs of 3a's in stock and only one demo pair of 10pi's. I think he sold out all six pairs of 3a's that weekend. Did the ADD cabinet/grill redesign priduce any audible (as opposed to measurable) change in sound? Not to my ears.

Thanks for the segway to this post Gene. I couldn't help but think of the first cartoon in Rodrigues' book brought up in one of my earlier posts in this thread. I post it here at the risk of getting my hand slapped for copyrite infringement.

Hey, doesn't that look like an LST at the top of the pile and a JBL (block foam pattern grille) at the bottom?

Anyway, I hope it brings a smile to the faces of a few viewers who visit here.

I know it will to Soundminded's!

[Steve F]

Did the ADD cabinet/grill redesign priduce any audible (as opposed to measurable) change in sound? Not to my ears.

My post on the 10 Pi/11 vs. the 3a was a little tongue-in-cheek ironic, if anyone read way, Way, WAY between the lines.

Here’s the thing: The 3a has been accused in many quarters (and I’m paraphrasing here, so don’t jump all over me) for having a near-field measured response that is a catastrophe of phase cancellations and diffraction effects.

AR supposedly eventually “recognized” this and set out to correct all those ills with the ADD line (10 Pi, 11, 12, etc.)—very small cabinet moldings, truly flush-mounted drivers, absolutely acoustically-transparent foam grilles, etc.

All good stuff, right?

One problem—they (the ADD three-ways) all had their mids and tweeters mounted in a dead horizontal line as viewed when the cabinets were vertical, as they would be on floorstands. Now, every speaker designer “knows” that you’ll get near-field interference and cancellation effects between horizontally-oriented drivers. That’s the reason that AR’s next series was the 9 and the subsequent Verticals (90, 91, 92), with their M-T drivers all nicely vertically-aligned.

So the 10Pi/11/12 are all “wrong.” Lots of destructive near-field interference will result from the H orientation of their M-T drivers. High Fidelity magazine measured this with the 2ax, whose M-T was dead horizontal. If you look at their on-axis FR curve, there is a cancellation dip centered right at 5kHz, the M-T x-over. Neither High Fidelity nor AR recognized this phenomenon at the time, so the test report doesn’t say anything. But there it is, clear as a bell, just as theory “predicts” it will be.

I asked AR why the 3-way ADDs were done that way, and AR gave me the weak answer that they were ‘bookshelf’ speakers and when they were mounted horizontally, the M-T were vertical.

Balderdash.

Most of the time, people stand-mounted them or put them in entertainment units vertically, at least as often as people put these 55 Lb boat anchors on bookshelves. So, 50% of the time, the speakers were mounted incorrectly.

But….funny thing: When AR did the jazz drummer Grover L v R demo, the 10 Pi’s were on stands just behind the drummer’s seating position mounted……VERTICALLY, with the midrange and tweeter HORIZONTALLY-ORIENTED. All wrong. A veritable “catastrophe” of near-field cancellations, poised to ruin its direct first-arrival response and wreck the demo.

Funny thing happened along the way, though. The demo was great, very impressive, very convincing, to a group of skeptical, golden-eared BAS audiophiles. Me? At that time, I was 22, my hearing was the best it would ever be, I sat in the first row (as near-field as could be), and I was a very active (and quite accomplished, if I do say so) JAZZ DRUMMER at that time. I was certainly a well-qualified listener.

So, near-field more important, or far-field more dominant? Pick your own little bits of evidence that support your position, discount the opposing viewpoint, and have fun. I’m simply reporting what happened and how the speakers were designed and positioned. The rest is up to you.

BTW, the 10 Pi had M-T switches that could be set either 0, -3dB, or -6 dB. I’d be willing to bet that a 3a in MAX position and a 10 Pi in -3 or -6 would sound pretty similar. But both max’ed out, they did not sound similar, from a spectral balance standpoint. Did the reduction in cabinet diffraction make any difference? I’m not going there…..

Steve F.

[Zilch]

I will not continue to debate the impact of the edge moldings on the speaker at a fixed angle off axis (Allison's own curves showed that the response would vary considerably at different angles, with the cumulative effect being what matters), but I will point out that (after looking at your posted curves) that any speaker that is 20 dB down at 10 kHz at 75 degrees off axis and 15 dB down at a rather tame 7 kHz at that angle, and is over 10 dB down at 7 kHz at 60 degrees off axis, is anything but a wide-dispersing design, no matter how it is measured.

I will say it again, Howard: It is NOT a wide-dispersing design, and neither is AR3a; it took four AR3a tweeters to accomplish the wide dispersion objective in LST....

[genek]

I can feel the cold hand of banishment starting to pull this thread toward the kitchen. Can we please try to avoid getting personal in our discussions?

I suggest a little exercise. Everybody try to avoid mentioning another person by name in their posts for an entire day.

[genek]

I asked AR why the 3-way ADDs were done that way, and AR gave me the weak answer that they were ‘bookshelf’ speakers and when they were mounted horizontally, the M-T were vertical.

I've always thought the ADD series speakers with their wide band of wood at the base looked really weird turned on their sides.

[Zilch]

And you want to know how much time he spent in the book actually analyzing and dealing with the import of the live-vs-recorded concerts Ed Villchur orchestrated back in the 1960's?

I'll tell you how much: zero.

The best he could do is dismiss them out of hand, as you have done, which is a cop out.

I'm not going there again, Howard; I have devoted post after post and page after page substantiating Toole's perspective on this matter: the LVR demos were all but meaningless from a scientific perspective, and such attention as he DOES devote to them is more than they deserve when viewed in the context of his book.

That horse is dead. You need to hitch your cart to somewhat more modern means of propulsion....

[tysontom]

I can feel the cold hand of banishment starting to pull this thread toward the kitchen. Can we please try to avoid getting personal in our discussions?

I suggest a little exercise. Everybody try to avoid mentioning another person by name in their posts for an entire day.

I agree: this debate is on the verge of getting into the ditch. We all have our opinions, and if we need to get ugly let's go to the Kitchen.

--Tom Tyson

[Steve F]

I've always thought the ADD series speakers with their wide band of wood at the base looked really weird turned on their sides.

Yes, they looked really strange. It's quite obvious that they were designed, from an ID standpoint, to be vertical. Maybe Engineering never spoke to the ID firm. Maybe Mktg and Eng were at different meetings. Maybe Eng didn't think that vertical drivers mattered at all in the far field and the next gen Verticals were a just cynical marketing exercise that they (Eng) knew didn't affect the actual far-field sound.

I'll be quiet now. Yes, the ADD's looked much better vertical. For that matter, I always though the 10 Pi looked 'top heavy' with it's band of wood (the door covering the controls) at the top. I always thought it should have been at the bottom like the others.

Steve F.

[speaker dave]

You want to know just how many ultra-wide and smooth dispersing main-channel speakers Toole discusses, listens to, and measures in his book? You want to know how many super-wide dispersing speakers were involved in his panel listener comparisons, at least as discussed in the book?

I'll tell you how many: zero.

Howard Ferstler

Lets look at just one.

Page 394 Fig. 18.14, speaker "R" shows a d.i. (directivity index) that climbs very smoothly to just below 5db and stays there all the way to 20kHz. Response curves, both on and off axis, are wonderfully smooth and flat. For those not used to d.i. numbers an omnidirectional speaker has a a d.i. of 0 dB. A hemispherical radiation speaker has a d.i. of 3dB. A d.i. of 5 is very little more directional, and maintaining it as a constant directivity to 20kHz is very impressive.

An AR3a, by the published curves, has hemispherical radiation (d.i. of 3) to about 12 or 13kHz.

I'm not sure where this "ultra-wide" dispersion came to be a unique AR feature compared to modern speakers. Plenty of systems out there have dome tweeters and dome mids. Most dome tweeters are 1" but there are some 3/4" units as well. No AR system would have any advantage over any other system also using 3/4" tweeters. Over a system with a 1" dome (common as mud) the AR advantage would be a 1/3 wider frequency range with the same dispersion (say a dispersion at 13.3k that another system could only match to 10k)

Also, please define "smooth dispersing" if interference effects and edge reflections have been deemed to be inconsequential.

Note that the stacked Allison tweeters in your home system will add another 3dB of directivity (less dispersion) and significant lobbing that I would think was at odds with "smooth dispersing".

The AR3a was a fine speaker and well ahead of the competition in its day. Do we really have to proclaim it better than any speaker ever made to feel good about it? (Seems like an insult to the AR designers and their products that came later.)

David

[Zilch]

I think your primary beef with the sound of the AR-3a has nothing to do with the supposed diffraction effects or conflicted radiation pattern. What probably bothers you most is the treble rolloff, which cuts into the speaker's ability to deliver a razmatazz treble thrill with the kind of recordings you prefer. The problem is your musical taste.

Did I say I had a beef with the sound of AR3a?

Or are you just making that up like you so obviously have all the rest of this?

It's an outmoded, "old school" song you sing here, Howard, much as Corey Greenberg and John Atkinson said nearly 20 years ago:

http://www.stereophile.com/reference/101/

[Zilch]

Page 394 Fig. 18.14, speaker "R" shows a d.i. (directivity index) that climbs very smoothly to just below 5db and stays there all the way to 20kHz. Response curves, both on and off axis, are wonderfully smooth and flat. For those not used to d.i. numbers an omnidirectional speaker has a a d.i. of 0 dB. A hemispherical radiation speaker has a d.i. of 3dB. A d.i. of 5 is very little more directional, and maintaining it as a constant directivity to 20kHz is very impressive.

And not coincidentally, we might suppose, it ranked highest among the four TOTL speakers shown with respect to subjective listening preference.

The AR3a was a fine speaker and well ahead of the competition in its day. Do we really have to proclaim it better than any speaker ever made to feel good about it? (Seems like an insult to the AR designers and their products that came later.)

Nah, we do that for other reasons....

[kkantor]

Who the heck here mentions Zilch by name? We do not know his name.

Howard Ferstler

Howard,

I don't appreciate the deliberate attempts at provocation that appear here from time to time. Your statement above fits that category. Please knock it off.

-k

[kkantor]

I agree, too.

The interesting thing is that as time goes on most of the really exciting and dynamic posts on this website will be in the kitchen.

Howard Ferstler

Yes, and that will leave the AR section to be about information, restoration and other brand-specific things of use and value to many visitors here. As it should be.

-k

[kkantor]

Did I say I had a beef with the sound of AR3a?

Or are you just making that up like you so obviously have all the rest of this?

It's an outmoded, "old school" song you sing here, Howard, much as Corey Greenberg and John Atkinson said nearly 20 years ago:

http://www.stereophile.com/reference/101/

Zilch,

You deliberately stirred up things here. Interesting discussions have ensued. Bravo. Please don't ruin their value with potshots and insults. Everyone here already knows about Howard's book, its value, its limitations and its reviews. One link a year would be plenty. Sheesh.

-k

[kkantor]

Yes, they looked really strange. It's quite obvious that they were designed, from an ID standpoint, to be vertical. Maybe Engineering never spoke to the ID firm. Maybe Mktg and Eng were at different meetings. Maybe Eng didn't think that vertical drivers mattered at all in the far field and the next gen Verticals were a just cynical marketing exercise that they (Eng) knew didn't affect the actual far-field sound.

I'll be quiet now. Yes, the ADD's looked much better vertical. For that matter, I always though the 10 Pi looked 'top heavy' with it's band of wood (the door covering the controls) at the top. I always thought it should have been at the bottom like the others.

Steve F.

In my opinion, many of the final product decisions at AR during this period of time were compromises (or worse) between the oft-warring factions of "engineering" and "research." No one person had unilateral control over designs, and it would be a mistake to analyze any particular product as being a perfectly pure expression of AR design philosophy. I'm sure this kind of situation is common at many larger companies, in many industries.

-k

[tysontom]

I'm not sure where this "ultra-wide" dispersion came to be a unique AR feature compared to modern speakers. Plenty of systems out there have dome tweeters and dome mids. Most dome tweeters are 1" but there are some 3/4" units as well. No AR system would have any advantage over any other system also using 3/4" tweeters. Over a system with a 1" dome (common as mud) the AR advantage would be a 1/3 wider frequency range with the same dispersion (say a dispersion at 13.3k that another system could only match to 10k)

David

David,

Wow, this link is heating up. Incidentally, thanks very much for your insightful and highly knowledgeable comments!

Theoretically you are certainly right and in general there is little argument with your point. There isn’t anything unique about the dispersion of the AR-3a tweeter; however, there are two characteristics of the 3a ¾-inch dome that separate it from most of the contemporary domes: (1) the dome is flush-mounted to the top plate, and (2) the dome is a rigid material rather than a flexible, soft-dome type construction typical of most current tweeters.

A good example of the first characteristic would be the excellent Hiquphon Model OW 1 ¾-inch dome tweeter. To my knowledge this hand-made tweeter is one of the premiere tweeter designs currently available -- it certainly qualifies from the standpoint of cost. It has the requisite “semi-horn” recessed top plate, standard to so many contemporary tweeters, a soft-fabric come material, and I think the off-axis performance suffers slightly due to these things. It is as smooth as it gets on-axis, but it does not compare with the AR-3a tweeter off-axis. For example, at 20kHz, the Huquphon is 10-12dB down, 30°off axis while the AR ¾-inch dome is only 3-4 dB down at that same point. Overally, I’m sure the OW 1 beats the AR tweeter in linearity, distortion and on-axis smoothness.

A good example of the second item is the contrast between the AR-3a ¾-inch dome and the newer, but nearly identical AR-10Pi-AR-11 ¾-inch soft dome. The 10Pi soft-dome tweeter has greater sensitivity and more extended output, but it does not quite match the off axis performance of the AR-3a dome.

--Tom Tyson

[tysontom]

David,

Wow, this link is heating up. Incidentally, thanks very much for your insightful and highly knowledgeable comments!

Theoretically you are certainly right and in general there is little argument with your point. There isn’t anything unique about the dispersion of the AR-3a tweeter; however, there are two characteristics of the 3a ¾-inch dome that separate it from most of the contemporary domes: (1) the dome is flush-mounted to the top plate, and (2) the dome is a rigid material rather than a flexible, soft-dome type construction typical of most current tweeters.

A good example of the first characteristic would be the excellent Hiquphon Model OW 1 ¾-inch dome tweeter. To my knowledge this hand-made tweeter is one of the premiere tweeter designs currently available -- it certainly qualifies from the standpoint of cost. It has the requisite “semi-horn” recessed top plate, standard to so many contemporary tweeters, a soft-fabric come material, and I think the off-axis performance suffers slightly due to these things. It is as smooth as it gets on-axis, but it does not compare with the AR-3a tweeter off-axis. For example, at 20kHz, the Huquphon is 10-12dB down, 30°off axis while the AR ¾-inch dome is only 3-4 dB down at that same point. Overally, I’m sure the OW 1 beats the AR tweeter in linearity, distortion and on-axis smoothness.

A good example of the second item is the contrast between the AR-3a ¾-inch dome and the newer, but nearly identical AR-10Pi-AR-11 ¾-inch soft dome. The 10Pi soft-dome tweeter has greater sensitivity and more extended output, but it does not quite match the off axis performance of the AR-3a dome.

--Tom Tyson

The Hiquphon has an aluminized-metal dome, not fabric as I mentioned. I does have a significant "semi-horn" top plate, however.

[speaker dave]

The Hiquphon has an aluminized-metal dome, not fabric as I mentioned. I does have a significant "semi-horn" top plate, however.

Hey Tom,

I'll have to look a little more carefully at the various off-axis curves that have been put on the site for various domes. As far as I know the first order effect is size, but if secondary factors (dome material, etc.) have a strong effect, I'll defer to your knowledge on this.

Agreed on flares and waveguides. I've designed a number of wide angle flares (120 to 150 degrees, typically) and you are always increasing directivity (that is, making the unit more directional). Generally you achieve constant directivity by adding directivity at the bottom of a driver's range to equal its inherent directivity at the top of its range.

Regards,

David

[Zilch]

Agreed on flares and waveguides. I've designed a number of wide angle flares (120 to 150 degrees, typically) and you are always increasing directivity (that is, making the unit more directional). Generally you achieve constant directivity by adding directivity at the bottom of a driver's range to equal its inherent directivity at the top of its range.

I'm no horn/waveguide theorist, but it appears we cannot get there. On the other hand, the evidence from examples of alternatives which DO achieve it indicates that there may be more and better mileage to be accomplished via cue extraction.

(BTW- Howard and Zilch, my choice was a 10pi, plus an add-on uAcoustics tweeter array.... guess what I was thinking about.....)

Uhmm, "max dispersion," maybe?

[Zilch]

Well, I am glad you read here that you apparently really do like the system. No use debating any more, since the bottom line should be just that: do you like the system - and you apparently do.

Assume whatever suits your pleasure or purpose at the moment, Howard, but I didn't say that, either.

For the purpose of these discussions, I wear my unbiased objectivist hat....