Dome mids vs. cones--meaningful difference?

[Steve F]

Not entirely "subjective." The BA MR 3-inch piston was exactly twice the diameter of the 3a's 1 1/2-in dome MR. so the BA would be non-directional to exactly one-half the frequency that the 3a's mid was. The 3a and 11 went to 5kHz, the BA went to 2500Hz, so their respective mids' directionality in their operating passbands were identical. Identical, in theory. In the actual system (not as a single driver on a flat baffle in an anechoic chamber), the 3a had a bit of a "horse blinder" from that intrusive picture-frame cabinet molding. The 11--benefiting from experience-- corrected that. The BA, of course, like any good modern speaker (once AR showed the way with the ADDs) had no horse blinders. So the VR-M90's near-field mid dispersion was a tick better (as a complete speaker system) in practice than the 3a's and identical to the 11's. This is objective, not subjective. Factual. The numbers are the numbers. Do I have graphs? No. Is this still a fact? Yes.

The BA mid had a FAR of 175Hz, the AR mid 400Hz. Subjective? No Factual. 

The BA mid crossed over more than 2x over its FAR point, which meant it wasn't pushed into excursion-related distortion like the AR mid. Not "subjective." Simply better design practice.

The BA mid could handle more power and never failed. We never replaced a single 3 1/2-inch driver while I was there, and several systems used it. That's sort of subjective, because I don't have AR's failure rate pct, but whatever it was, it was more than BA's 0%.

I wish I had the BA raw driver FR curves. You'll just have to take my word for it, I'm afraid. Ruler flat. Perfectly uniform off-axis fall-off. Textbook driver, bullet proof, no failures, articulate and effortless. A driver that benefitted from 30 years' of hindsight and experience that the AR MR dome had paved the way. That the AR MR dome is still even in this discussion is an amazing tribute to its longevity and excellence. As long as diaphragm/coil/magnet speakers are still used, the Classic AR drivers will never be outclassed.

But on a case-by-case basis, they can be outpointed. This is one of those cases.

Steve F.

[tysontom]

Ha, ha!  Both contenders have begun "bouncing" off the ropes a bit, slugging it out.   AR-3a 3½-inch dome midrange vs. Boston Acoustics 3½-inch neodymium-magnet cone midrange.   No winner, no loser, but now it has become an almost purely academic argument.  To a degree, the debate has come down to a subjective-versus-objective, "philosophical" debate.  

The Boston Acoustics 3½-inch midrange was likely an excellent driver, and no one questions that assumption; in fact, it may be the finest transducer the world has ever known, but what is not known is its actual quantitative performance, how smooth or rough its output, warts and all.  However, I do believe what my friend Steve tells me, and therefore I must have faith that this BA driver is/was a stellar performer.  Faith!  But I am also a skeptic when it comes to electro-mechanical devices, and I like to see the proof of performance.  I fall into the old school on measurements.  I am conflicted. 

With quantitative loudspeaker measurements, a company can back up its performance claims to some extent with lab results, thus reinforcing its claims for "performance excellence," but this trend was never very popular (and grew worse) in the consumer high-fidelity loudspeaker industry, showing precious little correlation with speaker sales.  Alternatively, speaker measurement tended to open a can of worms.  It cost a lot of money and required additional engineers, and it exposed vulnerabilities—someone would have to answer for that; therefore, most consumer speaker companies steered clear of big anechoic chambers and such.  Avoid such testing... "like the plague."  I am reasonably confident that the Boston Acoustics speaker company also did not place much credence in advanced-measurement equipment and research; certainly if it had, the public would have seen more of the output from the labs, or we would have test results here on the little 3½-inch cone midrange driver.  Of course, there were tests done and measurements made, but no in-depth, detailed acoustics measurements in the AR manner (the pro-audio folks did these measurements).  BA had very innovative products and commercial success; research, not so much.  Had there been, the company would likely have capitalized on it and published even the most rudimentary response graphs, distortion curves, power curves and so forth, making a strong case for a family of fine reproducers.  They would say, "Not only do our speakers sound great, they also measure well!"

The lack of this type of testing was steeped with tradition going back to the KLH and Advent days under the guidance of creative-genius Henry Kloss.  There were virtually no lab measurements from those companies, either.  No anechoic chambers, no holes in the ground to measure woofer performance, etc.  Kloss was not a trained acoustics engineer, and he believed not in quantitative acoustical measurements beyond very basic in-room measurements.  He was, however, a creative, very bright "tinkerer" with a exceptional ear for "voicing."  He knew a good thing when he heard it, and usually he was right.  He therefore taught his disciples the art of speaker "voicing."  This philosophical lineage dribbled downhill from KLH to Advent and on to Boston Acoustics, founded by the former Advent team of Andy Kotsatos* and Frank Reed.  Nevertheless, this company was very successful, and a lot of this success can also be attributed to the marketing and design thinking of our friend Steve! 

On the flip side, what you did see from the old Acoustic Research was the good-and-the-bad measurements from AR's anechoic, free-field and reverberant measurements.  On top of that, all of the measuring instruments at Acoustic Research were calibrated regularly to assure compliance with ASA and US Bureau of Standards (now known as the National Institute of Standards and Technology).  Rigorous test methods all the way.  So, at this point we know exactly what is going on with the AR 1½-inch dome driver.  We know whether or not it has any peaks and dips along the way, both on- and off-axis, and we know its strengths and weaknesses.  We don't know this about the BA 3½-inch midrange driver.  We assume—and have faith—that it performs superbly, but there is no hard-core evidence that it does.

Now, the AR-3a edge-molding "window-frame, horse-blinder" effect is real in one sense: near-field measured response.  But diffraction occurs at nearly 90˚ off axis from the drivers, so it is not a major problem, even in the near field; importantly, diffraction is a major issue if one listens to the speakers from 3-5 feet away.  Once back in a listening room, edge-molding diffraction, lobing, crossover interaction and other anomalies are literally swamped and cannot be measured or even identified, as all of the reflections from the output of these drivers is blended into a seamless acoustic-power response output.  It is certainly true that the AR-10π and AR-11 improved on the diffraction issue, but it is imperceptible back in the reverberant field.  What might have been an improvement in reduced diffraction is also offset slightly with less-stellar dispersion of the newer cloth-dome tweeter.

So, where does this leave us.  These are two top-notch midrange drivers, but we don't have all the data we need to know the true performance of the BA driver.

* Andy Kotsatos was "Andy Petite," a name that evolved from his father's nickname, "Le Petit," acquired as a restaurant owner in New York.

—Tom

[Steve F]

"I am reasonably confident that the Boston Acoustics speaker company also did not place much credence in advanced-measurement equipment and research; certainly if it had, the public would have seen more of the output from the labs, or we would have test results here on the little 3½-inch cone midrange driver."

This is incorrect; BA's developmental individual driver measurements and full system measurements were complete and rigorous. They just chose not to create marketing materials out of them like AR did in the 1960's-70's. BA's measurement techniques and completeness were the generation more advanced than AR's archaic pen-and-graph traces of their time that you'd expect them to be. BA could also perform advanced magnetic field analysis and breakup mode analysis that simply didn't exist in the 3a's or 11's time. There were mounds of curves and data. I think Tom is referring to Andy/Henry's "mostly by ear, with a quick check by looking at the curve" approach at Advent in the 1970's. Not at BA. Not with all the advanced equipment that was at the disposal of all those young, modern, highly-motivated engineers. Curves and files and data up the ying-yang. More analytical driver and complete system data than AR of 1971 could ever dream of. F-22 Raptor vs. F-100 Super Sabre kind of difference in loudspeaker measurement and analysis technology. The 30 years between 1970 and 2000 were a universe of difference. Think of medicine, television, phones of 1970 vs. 30 years later. The same differences are true for loudspeakers. Technology moves forward.

Just because BA didn't create marketing materials for their raw driver data (the 1990's-2000's mass-audio market had no interest in those curves) shouldn't be construed to mean such data didn't exist. The BA 3 1/2-in mid was an incredibly advanced, modern driver, with all the advantages that I've outlined in previous posts. I will not waste everyone's time repeating myself. The numbers are the numbers: a 3-in piston has the same directionality to 2500Hz that a 1 1/2-in driver has to 5000. The BA had a lower FAR, 175 to 400 Hz. It crossed over >2x above its FAR point, which is far better design practice than AR used on the 1 1/2-in dome.  (And AR knew it, which is why the 9 and 90 and 98LS were 4-ways.)

What are we talking about? What's the issue? That I don't have the raw driver curves to show you? That's the issue? Like I said, You'll just have to take my word for it. Sorry. But is the AR 1 1/2-in dome mid really being presented here as being a "superior" driver to the BA 3 1/2-in mid? Is that what is really being said?

The AR dome did not have superior dispersion within its operating passband than the BA did in its. The BA handled more power and had less distortion at its lower end. The BA had a zero % failure rate. What's the question? The AR dome was a superior midrange? No sale. It may have been the best of its day, and is still remarkably good even by today's standards. But time moves on.

Steve F.

[Aadams]

This was the original topic of the thread. 

I want to keep this discussion to the importance of mid-dispersion for forward-facing single-driver-per-band designs, and whether there is truly an audible/meaningful difference between a 1 ½-inch dome and a 4- or 5-inch cone when the listener is sitting in a normal position (less than 10 or 15 degrees off axis from the left and right speakers of a stereo pair).

What you have been saying is enjoyable and informative but what is the answer to the original question or has it been answered? 

Adams

 

[Steve F]

Fair enough.

 The answer is that for a normal listener, sitting in the normal position about 10-15 degrees of-axis of the L-R speakers in a stereo pair, perhaps 6-8 feet away from the speakers, he’ll be very close to “on axis” of the midrange drivers, and no, there won’t be any appreciable difference. If the speakers are “toed in,” which they often are, then the listener will be directly on-axis of the midrange drivers and if he’s in the near-to-critical listening field, there will not be any appreciable deficiency due to the mids being 5-inch cones.

 

If the listener is farther back in the reverberant field, then the smaller mid driver’s superior power response will likely prove more spacious and natural sounding, since the speaker with the larger-diameter mid driver will show a droop in its power response at the top of the mid driver’s band where it’s getting really directional. You can’t take a 5-inch driver past about 2700Hz before it gets pretty beamy.

 

But......first-arrival sound still dominates the tonal signature, according to many speaker designers. Even 10 or 15 feet back you’re still going to hear the mids’ “first arrival” sound before anything else. AR felt that power response swamped first-arrival sound and mashed it all together. The time differential between first arrival and reflected room power response was too close in time for the listener to distinguish, they said.

 

Hogwash, say the first arrival people. You can tell, and that’s precisely why “theoretically incorrect, narrow-dispersion” speakers manage to still sound great, just great. Look at the various best speakers of today: the B&W’s, the Aerials, the Revels, etc. All have large cone mids, all “wrong” according to AR of the 1960’s. Yet, they are all flat-out fabulous-sounding speakers, aren’t they?

 

Pretty strong argument for those who feel that first-arrival on-axis FR is the predominant determinant of perceived tonal quality.

 

Steve F.

[Aadams]

What I get from this is, if one is dealing with well designed speakers then what sounds best becomes very personal. After many years I have found a good groove with AR 12 inchers.  I could duplicate the sound I like with other equipment but what a pain and with so little time left.

  

[tysontom]

2 hours ago, Aadams said:

This was the original topic of the thread. 

I want to keep this discussion to the importance of mid-dispersion for forward-facing single-driver-per-band designs, and whether there is truly an audible/meaningful difference between a 1 ½-inch dome and a 4- or 5-inch cone when the listener is sitting in a normal position (less than 10 or 15 degrees off axis from the left and right speakers of a stereo pair).

What you have been saying is enjoyable and informative but what is the answer to the original question or has it been answered? 

Adams

 

Dear Adams,

As we've said previously, all things being equal, the 1½-inch dome midrange driver will have wider dispersion in the treble frequencies than the 4-5-inch cone driver, simply by virtue of the smaller radiating diameter.  Domes typically are better behaved in the higher frequencies with fewer break-up modes, but it also depends on how high the cone driver is taken in the system before being crossed over to the tweeter.  Subjectively, if you sit close to the speaker and the speakers are aimed at you, the 4-5-inch cone will have a more focused sound (the so-called "image") than the wider-dispersion dome driver; however, further back into the room, the 4-5-inch driver will have lost the sense of spaciousness due to the lack of off-axis energy and the diminishing direct output (as a function of distance to the speaker from the listener).  In effect, unless the cone driver is over-bright in direct output, it will lose total acoustic output the farther back you get from the speaker and can actually begin to sound dull due to the lack of total acoustic-power output in the room.

Therefore:

1. ·        Dome has a more diffuse, less-focused sound close to the speaker;
2. ·        Cone has better "image" but less spaciousness back in the room.

Which is better is up to you.  Do you like spaciousness or intimate detail sound?

—Tom Tyson

[Aadams]

Which is better is up to you.  Do you like spaciousness or intimate detail sound?

—Tom Tyson

Not sure if your question was rhetorical but on the chance you are polling me, I prefer spaciousness.

Adams

[tysontom]

2 hours ago, Aadams said:

Which is better is up to you.  Do you like spaciousness or intimate detail sound?

—Tom Tyson

Not sure if your question was rhetorical but on the chance you are polling me, I prefer spaciousness.

Adams

No, not meant to be rhetorical.  If you prefer spaciousness (and sit back within the listening room and not right in front of the speakers), then the wide-dispersion speaker will be better for you.  You will get a better sense of ambiance of the recording space, much akin to sitting back somewhat in a symphony hall or large musical stage.  I've always felt that if you want the best of both worlds, spaciousness and "imaging," headphones (which are "binaural")  provide a good solution, as you hear what the microphones recorded without the second, or double, reverberation of the playback environment..

[Aadams]

26 minutes ago, tysontom said:

No, not meant to be rhetorical.  If you prefer spaciousness (and sit back within the listening room and not right in front of the speakers), then the wide-dispersion speaker will be better for you.  You will get a better sense of ambiance of the recording space, much akin to sitting back somewhat in a symphony hall or large musical stage.  I've always felt that if you want the best of both worlds, spaciousness and "imaging," headphones (which are "binaural")  provide a good solution, as you hear what the microphones recorded without the second, or double, reverberation of the playback environment..

I doesn't matter why, I will just skip the verbiage and say for me, your statement is exactly correct.

Adams

[ar_pro]

You will get a better sense of ambiance of the recording space, much akin to sitting back somewhat in a symphony hall or large musical stage. 

Would that this was practically true. In a concert hall, there's usually some back to move to; in a home listening environment, most people will be encroaching on a back wall - if they hadn't already started out that way, that is. Far enough back, and you're possibly changing/reinforcing a different room resonant effect as well as acquiring "spaciousness". If I move far enough back in our big listening space, it's like sitting inside a Klipschorn.

[Aadams]

5 hours ago, ar_pro said:

You will get a better sense of ambiance of the recording space, much akin to sitting back somewhat in a symphony hall or large musical stage. 

Would that this was practically true. In a concert hall, there's usually some back to move to; in a home listening environment, most people will be encroaching on a back wall - if they hadn't already started out that way, that is. Far enough back, and you're possibly changing/reinforcing a different room resonant effect as well as acquiring "spaciousness". If I move far enough back in our big listening space, it's like sitting inside a Klipschorn.

I think you are correct and what a can of worms this leads to.  Now we are getting into the elusive and subjective "you are there" experience which I personally only ever sense with certain brands and era classical recordings in a specific room with low ambient noise and a system with sufficient dynamic and frequency range.  If you don't listen to large scale classical I am not sure the spaciousness factor matters that much.  Same with headphones which really help defeat the ambient noise problem but even then  a crap recording is still crap.  For non classical music OTOH I don't really care about spaciousness or soundstage or whatever, it just has to have a message I like.  Again highly subjective.

[Jeff_C]

11 hours ago, tysontom said:

I've always felt that if you want the best of both worlds, spaciousness and "imaging," headphones (which are "binaural")  provide a good solution, as you hear what the microphones recorded without the second, or double, reverberation of the playback environment..

For me the recording has to be "binaural" to get a good sense of image and spaciousness. With a standard stereo (non-binaural) recording, headphone listening for me is not that satisfying, although they do improve listening in poor acoustic environments as you say.

[iso]

11 minutes ago, Jeff_C said:

For me the recording has to be "binaural" to get a good sense of image and spaciousness. With a standard stereo (non-binaural) recording, headphone listening for me is not that satisfying, although they do improve listening in poor acoustic environments as you say.

Wide dispersion does have some advantages in listening, but more recent trend of constant but narrow dispersion does offer as better imagining capability in room acoustics. This way reflections in the room do not smear signal as much as when using wider dispersion. Listening dummy head "binaural" recordings with headphones is farthest we can go this way. This way room acoustics do not smear imagining like when using narrow dispersion speakers in the room and even less than when using wide dispersion speakers in the room. 

Standard recordings are meant to be listened with speakers... ie not usually with headphones. I suppose that this is one reason for cross feed techniques...

Kimmo

[Aadams]

On 2/15/2017 at 2:31 PM, Steve F said:

I want to keep this discussion to the importance of mid-dispersion for forward-facing single-driver-per-band designs, and whether there is truly an audible/meaningful difference between a 1 ½-inch dome and a 4- or 5-inch cone when the listener is sitting in a normal position (less than 10 or 15 degrees off axis from the left and right speakers of a stereo pair).

If the criteria are, near on axis listening at six to eight feet a away, the answer is:

A modern, high quality, small cone driver is superior to any AR dome mid implementation between 200 and 1000hz in any 3 way AR up to the 58s.  I have been listening very closely to BA 5" mids for the last year, just after this thread began, and I am certain that, at my house, the only thing that comes close is the 8'' mid in an AR9. 

Aadams

[AR surround]

2 hours ago, Aadams said:

If the criteria are, near on axis listening at six to eight feet a away, the answer is:

A modern, high quality, small cone driver is superior to any AR dome mid implementation between 200 and 1000hz in any 3 way AR up to the 58s.  I have been listening very closely to BA 5" mids for the last year, just after this thread began, and I am certain that, at my house, the only thing that comes close is the 8'' mid in an AR9. 

Aadams

I'm curious regarding the differences you are hearing which makes the modern cone (BA 5") superior to the AR dome mid.   I.e. are voices too squeeky, disjointed, crackly etc in the AR 3-way dome application compared to the BA cone?

How close is the AR9 8" cone in performance to the BA cone?

Before I started using an AR915 as a center channel speaker between the AR9's, I used an AR48 with it's cone midrange.   The 915 with its dome midrange was much superior to the cone midrange 48...but there could be numerous other reasons for that advantage include better matching with the 9's.

[genek]

I am seldom sitting in the "normal" listening position when my audio system is in use. More than half the time, I'm not even sitting at all. I put something on to listen to and then go about whatever it is I'm doing. So for me, the most critical thing is sound that doesn't change radically when I move across the room.

I have both AR-3a and AR-2ax, and both have worked well for me like this. The AR-6 I used to have before we downsized homes a few years back, not so much until I put a pair of Microstatic tweeters on them. From this, I intuit that the dome vs cone makes a difference for tweeters, but probably not so much for mids.

[Aadams]

1 hour ago, AR surround said:

I'm curious regarding the differences you are hearing which makes the modern cone (BA 5") superior to the AR dome mid.   I.e. are voices too squeeky, disjointed, crackly etc in the AR 3-way dome application compared to the BA cone?

How close is the AR9 8" cone in performance to the BA cone?

Before I started using an AR915 as a center channel speaker between the AR9's, I used an AR48 with it's cone midrange.   The 915 with its dome midrange was much superior to the cone midrange 48...but there could be numerous other reasons for that advantage include better matching with the 9's.

I had a long answer but realized a short one would be equally effective.

The difference is most apparent listening to human voice.  IMO the reason the AR 12 " 3 way dome implementations are inferior by comparison is because the entire musical voice range from 82 to 1000hz is shared by both the woofer and the mid range, with the 12inch speaker carrying most of the load except when it was crossed at about 500hz in the LST.

If you don’t listen to voice or try to hear lyrics you may never notice the difference. 

The AR dome midrange doesn’t really shine in its role until well above 1000hz where a single cone cannot compete except on axis.

I cannot account for the poor performance of the 4” mid cone of the AR48.  I have never heard one but being crossed at 400hz means all spoken voice and 90% of musical voice was coming from the woofer not the mid.

RE: the 8inch LMR it is a close call but I still think the small BA driver gets the edge down to the 250ish range where mine are cut.

There is nothing wrong with the AR dome it was just never built to perform across the entire human voice range.  And there is emphatically nothing wrong with the AR9.

Adams

[AR55]

I can’t honestly say whether AR’s 1½” dome midrange is better or worse than Boston Acoustic’s 3½” cone midrange in a 3-way speaker configuration.  I own AR-11’s, and I have listened to various Boston Acoustics speakers over the years including the VR-M90, but I have never directly compared the two.  I remember liking the sound of the VR-M90, but not enough to consider replacing my 11’s. 

It is true that dome midranges have all but disappeared from 3-way speakers, but so have ¾” dome tweeters and 12” woofers.  You are also unlikely to find a 3½” cone midrange in a 3-way setup.  Your typical conventional 3-way speaker manufactured today has a 1” dome tweeter, 5” or larger cone midrange(s) and 6½” to 10” cone woofer(s).  By the late 80’s AR had switched to this configuration with the TSW series, as did many other manufacturers including at some point Boston Acoustics.

Lastly, I agree that AR improved their high end speakers (AR-9/AR-90 & AR-9Lsi/AR-98Lsi) by adding an 8” lower midrange cone.  I own both the 98Lsi and 9Lsi.  There is better clarity/detail on and off axis.  This is most noticeable in the midrange frequencies.  Male vocals in particular seem more lifelike.  I have yet to find any modern speakers that retail for less than $4,500 a pair (9Lsi’s in 2018 dollars) that would entice me to replace my 9Lsi’s. 

In case anyone is wondering, I have 2 music/home theater rooms in my 1,400-sf house and I am very fortunate to have a wife that enjoys music & movies more than having her home showcased in Better Homes and Gardens.  One room is geared more toward watching movies, and the other with the 9Lsi’s is geared more toward listening to music, primarily stereo.

[Pete B]

I wrote this back on March 16, 2005 from this thread, the last post on page 1:

http://www.classicspeakerpages.net/IP.Board/index.php?/topic/1189-fora-9-speaker-concept-based-on-the-ar-9-by-pete-b/&

 

"I explained in Kris' UMD thread that small cone drivers provide much more VD and often have as good if not better dispersion because the dust cap acts as the radiator at high frequencies. This is supported by the fact that there are not many mid dome drivers available, only one 1.5" that I know of. There is little demand for them."

Pete B.

 

[Aadams]

3 hours ago, Pete B said:

explained in Kris' UMD thread that small cone drivers provide much more VD

 

What is VD?  I searched the linked thread and could not find it.  I can think of three things that it could be and one of them is not good.  And, to what sized small cone were you referring? In your opinion can a modern 5" cone equal the dispersion of an AR dome at 4000hZ?  If not can multiple 5" cones do it? 

Adams

[Pete B]

VD = volume displacement, surface are times Xmax.

What do you view as the -3dB down angles for the AR-9 1.5" dome at 4KHz?

I would not plan to use it to 4K but that is the frequency you used.

 

Steve F's simplistic analysis assumes that both drives are perfectly pistonic, which they are not,

something I noticed a very long time ago.

[Aadams]

1 hour ago, Pete B said:

VD = volume displacement, surface are times Xmax.

What do you view as the -3dB down angles for the AR-9 1.5" dome at 4KHz?

I had the 3a dome in mind but AR9 will do.  The 9 dome mid seems to be much less diffuse. I would say maybe 45 degrees off axis.

I would not plan to use it to 4K but that is the frequency you used.

My choices are a 5.25 inch at  4200hz or a 6.5 inch at 2500hz.   Which would provide better dispersion?  In your Opinion.

 

 

Adams

[Pete B]

Do you have polars for the AR-9 1.5" midrange?  I've not looked for them in many years.

I can't imagine using a 6.5" UMD, in fact I'd probably go smaller to a 4.5".

[Aadams]

2 hours ago, Pete B said:

Do you have polars for the AR-9 1.5" midrange?  I've not looked for them in many years.

I can't imagine using a 6.5" UMD, in fact I'd probably go smaller to a 4.5".

I think I should have been clearer.  I was asking with a 3 way in mind.  If, using a cone mid driver, you had to span the range between 200hz at the 12" woofer to the tweeter then given the following choices would you use

a 6.5' crossing at 2500hz or

a 5.25" crossing a 4200hz or perhaps

a 4.5" crossing a 4200hz? 

Is it OK if I PM you?

Aadams