What is this?

[Guest Bret]

I'm attaching a picture of a speaker I bought on ebay. The speaker gives every indication of being an AR-2a, BUT it has the AR-2ax midrange / tweeter.

But the midrange is mounted strangely and I can't swear to it, but I think the tweeter might be mounted a little funny, too.

The label says it is a 2ax.

The SN is AX117825

It might just be an old, standard, 2ax but is the first that I've seen with the grills off that looks exactly like this.

Bret

[Steve F]

Tough to say from the picture because the cabinet details aren't clear. First generation 2ax's had a visible cabinet mod on the baffle where the 2a's twin mids had been replaced by the single 3 1/2" unit of the 2ax.

This looks like the 1 3/8" tweeter and the older 6-bolt woofer, which would indicate an old (first-generation) 2ax, although a very late one, since "new" 2ax's came after s/n 125,000.

Steve F.

[Guest Bret]

Thanks, Steve.

I'm attaching a better/bigger jpg of the speaker to this message so you can see the part of the cabinet that's in question.

Yes, it is the old 6-bolt woofer although the inner cone has long since disappeared, and that is the old 1 3/8" tweeter. These were damaged in packing/shipping so I can't listen to them (one of the tweeter pots was shoved into the cabinet, I have confirmed that the tweeter still works so it's a matter of replacing the pot). I am looking forward to hearing the difference in this speaker and the "new" 2ax.

Bret

[tysontom]

>

>Yes, it is the old 6-bolt woofer although the inner cone has

>long since disappeared, and that is the old 1 3/8" tweeter.

>These were damaged in packing/shipping so I can't listen to

>them (one of the tweeter pots was shoved into the cabinet, I

>have confirmed that the tweeter still works so it's a matter

>of replacing the pot). I am looking forward to hearing the

>difference in this speaker and the "new" 2ax.

>

>Bret

Bret,

SteveF is right, I think, that this is the first-generation AR-2ax with the 3-1/2-inch midrange and the 8-ohm version of the AR-3 1-3/8-inch super tweeter from the AR-2a. This is a pre-1970 AR-2ax (after 1970 AR went to the 4-bolt stamped-frame, foam-surround woofer and the 8-ohm 3/4-inch tweeter), and most of them did have the cabinet routed specifically for the 3-1/2-inch midrange. As you know, the earlier AR-2a had two 5-inch midrange drivers in a cast-plastic housing, and some very early AR-2ax versions, and retrofitted AR-2a's, had a masonite adapter plate for the new midrange.

Sound-wise, the original AR-2ax was superb with little coloration and excellent dispersion, etc. A hallmark of the original AR-2ax's sound was it neutral, uncolored, sound characteristic. The newer AR-2ax had improved dispersion, improved lower midrange and power-handling, etc., and was slightly brighter in sound with an even-more spacious sound. The bass response was almost identical.

--Tom Tyson

[tysontom]

>Bret,

>

>SteveF is right, I think, that this is the first-generation

>AR-2ax with the 3-1/2-inch midrange and the 8-ohm version of

>the AR-3 1-3/8-inch super tweeter from the AR-2a. This is a

>pre-1970 AR-2ax (after 1970 AR went to the 4-bolt

>stamped-frame, foam-surround woofer and the 8-ohm 3/4-inch

>tweeter), and most of them did have the cabinet routed

>specifically for the 3-1/2-inch midrange. As you know, the

>earlier AR-2a had two 5-inch midrange drivers in a

>cast-plastic housing, and some very early AR-2ax versions, and

>retrofitted AR-2a's, had a masonite adapter plate for the new

>midrange.

>

>Sound-wise, the original AR-2ax was superb with little

>coloration and excellent dispersion, etc. A hallmark of the

>original AR-2ax's sound was it neutral, uncolored, sound

>characteristic. The newer AR-2ax had improved dispersion,

>improved lower midrange and power-handling, etc., and was

>slightly brighter in sound with an even-more spacious sound.

>The bass response was almost identical.

>

>--Tom Tyson

>

One further comment I forgot to mention: the strange appearance of the tweeter and midrange mounting on the early AR-2axs is due to the 1-1/8-inch-thick speaker baffle board. All AR-1s, AR-2s and AR-2as of the period used this thick front baffle. Because of the cost of changing the cabinets, to my knowledge, AR did not change this design in the AR-2-series until the new version in 1970. The AR-3, first introduced in late-1958, was the first AR speaker to use a standard, braced, 3/4-inch speaker baffle board.

--Tom Tyson

[Guest Bret]

>One further comment I forgot to mention: the strange appearance of the tweeter and midrange mounting on the early AR-2axs is due to the 1-1/8-inch-thick speaker baffle board. All AR-1s, AR-2s and AR-2as of the period used this thick front baffle.<

Thank you for coming-back to mention this.

I found it weird that AR would use a flush-mounted hemispherical dome tweeter on a metal casing then recess the tweeter into the cabinet half an inch or so. Similarly, the 3.5" midrange driver is "in a hole" of sorts.

Have you ever tried foaming around the tweeter?

I have to admit to being a particular fan of the post-1970 version of the 2ax and am really looking forward to having just a minute or two to get these "original" 2ax's fired-up. They do some things so well, it's almost startling that this obviously inexpensive (comparitively speaking) midrange driver sounds as good as it does.

Bret

[tysontom]

>I found it weird that AR would use a flush-mounted

>hemispherical dome tweeter on a metal casing then recess the

>tweeter into the cabinet half an inch or so. Similarly, the

>3.5" midrange driver is "in a hole" of sorts.

Once again it was a matter of standardized parts, I think. The 1-1/8-inch front speaker panel would require special longer screws to tap into the "T-nuts" on the back side if the tweeter and midrange were flush mounted. The trade-off for the recessed drivers was probably justified to save the money just in "special" mounting hardware. Remember, the first AR-2s and early AR-2a did not make money -- it took awhile before AR began to get costs under control for the AR-2 series. Recessing the drivers somewhat does have a slight effect on dispersion, and probably contributes to diffraction problems, in addition to the grill edge molding. Nevertheless, once you are back in the far field the diffraction effect probably isn't audible.

>Have you ever tried foaming around the tweeter?

No, I have not. The "Acoustic Blanket" that AR first used on the AR-9 and AR-90 were designed to absorb reflections close to the midrange and tweeter off the front baffle, to improve definition, and there were supposedly some slight measurable differences, but the biggest gain probably was ad copy. I'm not trying to disparage the foam around the tweeter idea -- others have tried it also -- but its effect is probably neglible.

--Tom Tyson

[Steve F]

"...Recessing the drivers somewhat does have a slight effect on dispersion, and probably contributes to diffraction problems, in addition to the grill edge molding. Nevertheless, once you are back in the far field the diffraction effect probably isn't audible.

>Have you ever tried foaming around the tweeter?

No, I have not. The "Acoustic Blanket" that AR first used on the AR-9 and AR-90 were designed to absorb reflections close to the midrange and tweeter off the front baffle, to improve definition, and there were supposedly some slight measurable differences, but the biggest gain probably was ad copy. I'm not trying to disparage the foam around the tweeter idea -- others have tried it also -- but its effect is probably neglible.

--Tom Tyson"

There are a couple of interesting trains of thought going on here, which merit comment.

First is Tom's observation that diffraction is not particularly audible once the listener is in the reverberant ('far") field. This, of course, is at the very heart of speaker design, and the two camps are diametrically opposed. One side believes that the sound of a speaker is determined primarily by its first-arrival on-axis response. Many successful, highly regarded speaker designers subscribe to this theory. Henry Kloss and Andy Petite are in this camp, as are many of the well-known British designers. Many great-sounding speakers have come out of this approach.

On the other hand, Roy Allison and others have always believed that a speaker’s energy response, as measured in the reverberant field, is the primary determinant of a speaker’s sound. This thinking does indeed put somewhat less emphasis on diffraction and other near-field anomalies, since they tend to get "averaged out" in the total energy measurement. It should be noted, however, that paying close attention to near-field diffraction while relying on far-field energy measurements are NOT mutually exclusive goals. If a speaker is free of spurious intrusions on its response, then the designer is less restricted in the pursuit of his desired result, whether that result is a flat axial response or a smooth energy response.

This brings us to point no. 2—Sometimes we simply forget that in the 1950’s and 60’s, the understanding of the finer points of acoustics and speaker design that we take for granted today were just coming to be accepted as common knowledge. AR was an extremely advanced company, but the state of the art was not as far along at the time of the original 2ax’s introduction (1964) as it was in 1978 or today. If you look at the ADD-Truth in Listening literature that I gave to Mark to post on the site, you’ll see that on the AR-16 spec sheet in 1976 was the first mention ever by AR of the importance of minimizing diffraction to prevent interference from the cabinet. If the original AR was still in business today and they made an AR-3a with the same stated design goals (smooth, accurate energy response in the far field, low distortion, and deep bass extension from a relatively compact cabinet) its drivers would be vertically-aligned and there wouldn’t be any unnecessary cabinet protrusions. There is simply a greater understanding now than before. (Interesting, this is essentially a product definition for Ken’s 303.)

So, yes, the 2ax’s tweeter being recessed only hurts things a little, and is hardly noticeable in the far listening field, but if they had it to do all over again, they’d flush mount it and keep the drivers free of obvious interference.

Steve F.

[tysontom]

>

>There are a couple of interesting trains of thought going on

>here, which merit comment.

>

>First is Tom's observation that diffraction is not

>particularly audible once the listener is in the reverberant

>('far") field. This, of course, is at the very heart of

>speaker design, and the two camps are diametrically opposed.

>One side believes that the sound of a speaker is determined

>primarily by its first-arrival on-axis response. Many

>successful, highly regarded speaker designers subscribe to

>this theory. Henry Kloss and Andy Petite are in this camp, as

>are many of the well-known British designers. Many

>great-sounding speakers have come out of this approach.

>

>On the other hand, Roy Allison and others have always believed

>that a speaker’s energy response, as measured in the

>reverberant field, is the primary determinant of a speaker’s

>sound. This thinking does indeed put somewhat less emphasis on

>diffraction and other near-field anomalies, since they tend to

>get "averaged out" in the total energy measurement. It should

>be noted, however, that paying close attention to near-field

>diffraction while relying on far-field energy measurements are

>NOT mutually exclusive goals. If a speaker is free of spurious

>intrusions on its response, then the designer is less

>restricted in the pursuit of his desired result, whether that

>result is a flat axial response or a smooth energy response.

>

>This brings us to point no. 2—Sometimes we simply forget that

>in the 1950’s and 60’s, the understanding of the finer points

>of acoustics and speaker design that we take for granted today

>were just coming to be accepted as common knowledge. AR was an

>extremely advanced company, but the state of the art was not

>as far along at the time of the original 2ax’s introduction

>(1964) as it was in 1978 or today. If you look at the

>ADD-Truth in Listening literature that I gave to Mark to post

>on the site, you’ll see that on the AR-16 spec sheet in 1976

>was the first mention ever by AR of the importance of

>minimizing diffraction to prevent interference from the

>cabinet. If the original AR was still in business today and

>they made an AR-3a with the same stated design goals (smooth,

>accurate energy response in the far field, low distortion, and

>deep bass extension from a relatively compact cabinet) its

>drivers would be vertically-aligned and there wouldn’t be any

>unnecessary cabinet protrusions. There is simply a greater

>understanding now than before. (Interesting, this is

>essentially a product definition for Ken’s 303.)

>

>So, yes, the 2ax’s tweeter being recessed only hurts things a

>little, and is hardly noticeable in the far listening field,

>but if they had it to do all over again, they’d flush mount it

>and keep the drivers free of obvious interference.

>

>Steve F.

Wow, what an interesting subject: near-field vs. far-field speaker designs! The terms “imaging,” “sound stage,” “focus” and other descriptions came into vogue with audio writers somewhere in the 1970s, and had to do with intentionally reducing the off-axis response of a loudspeaker. By reducing off-axis output, a speaker would act more like a pair of headphones, and excite fewer interference reflections and minimize double-reverberation in a listening room. The idea was to minimize a speaker’s near-field (especially vertical) midrange and treble off-axis response (that tended to “smear” the image) in such a way as to avoid interference effects, yet maintain a certain degree of spaciousness in the sound. This could be accomplished through the use of hefty midrange drivers and the ubiquitous 1-inch tweeter. Good examples of these designs are the British loudspeakers such as KEFs and B&W, and in this country certainly many of the panel speakers such as Magnepans. Seated in the “sweet spot,” a listener could experience the realism and “imaging” of these speakers in an almost life-like way! At one time I owned a pair of KEF R107s and subsequently a pair of B&W 801s powered by Krell, and I know from experience how exciting pin-point “imaging” can be. It can give the sensation of a live performance. It literally brings the jazz quartet into your living room! It might not be necessarily accurate, but it is exciting and realistic!

Then there is the old AR “reverberant-field” approach. The image would seem to be smeared and the pin-point focusing missing. AR speakers (as well as Allisons and some others) seem to bounce sound all over the place, and the listener gets disparate localization from the speakers, etc. So what gives here? AR’s feeling was that listening should be in the far field, and that the speakers should reproduce sound much like the instruments in which they were reproducing, with nearly the same spectral pattern. Regardless of where the loudspeaker was intended to be used, its only job was to reproduce the sound accurately. Dispersion is part of that accuracy. Tests that AR had done, in conjunction with acoustical engineers at Bolt, Beranek and Newman, demonstrated convincingly that the sound heard in a concert hall (or even in a stage performance of a jazz group), is largely reflected energy, with only a small portion of direct-arrival sound giving any degree of localization. “Small” is the operative word, because if anyone has attended many live concerts it is clear that the sound from a large symphony orchestra is a spacious blend of sound, with little “localization” of individual instruments. There is no “sweet spot” in a symphony hall. So now the question might be, do you want to bring the orchestra into your living room or transplant yourself into concert hall?

Another argument for the wide-dispersion, far-field approach was validated during AR’s early-60s live-vs.-recorded concerts comparing the AR-3 to the Fine Arts Quartet and guitarist Gustavo Lopez. In order to convincingly reproduce the live sound of a live instrument, the loudspeaker must emulate the radiation pattern of that instrument as well. Listeners present during the roughly 95 public concerts staged around the country were unable to correctly identify the switchovers from “live” to “reproduced” sound. Many of these listeners were the so-called “golden ears” of high fidelity. It is easy to misjudge the significance of these concerts, but it should be noted that many audio experts consider the ultimate test of loudspeaker accuracy is direct, side-by-side, comparison to a live-music source. To put it into better perspective, no manufacturer in modern times has attempted -- let alone succeeded -- in publicly demonstrating live-vs.-recorded accuracy. There must be a reason for this: no manufacturer today would be willing to take the risk in attempting to validate their product’s accuracy; and more often than not, the radiation pattern of most modern loudspeakers is less than optimal to accurately reproduce the spectral balance of a live instrument. Ed Villchur was 100% confident of the accuracy of his AR-3 loudspeakers; this degree of confidence doesn’t exist today because many manufacturers don’t design so much for accuracy as for listener preference and what "sells."

So which approach is the correct one? It’s impossible to know which approach is right or wrong. Both have their supporters, but the near-field devotees have long-since won the battle in numbers. Gerald Landau, marketing director of AR during the heady 60s, once told a group of dealers, “the good thing about AR speakers is that once sold, they stay sold.” I think what he was saying is that an AR speaker’s natural, accurate-sound reproduction would be less irritating to a listener in the long run.

--Tom Tyson

[Guest Bret]

>If the original AR was still in business today and they made an AR-3a with the same stated design goals (smooth, accurate energy response in the far field, low distortion, and deep bass extension from a relatively compact cabinet) its drivers would be vertically-aligned and there wouldn't be any unnecessary cabinet protrusions. There is simply a greater understanding now than before. (Interesting, this is essentially a product definition for Ken's 303.)<

Thanks, Steve.

Sometimes I sound like German was my first language and I just cannot get the phrase-order corrected for English - It kinda reminds me of the scene in Life of Brian (Monty Python) where the Roman Centurian is correcting Brian's Latin as he paints "Romans Go Home" all over some Roman structure.

I've erased the first round of goop I wrote, and I'll try in English:

I expect to ask Ken a question about the design of the AR-303. I am postponing that question at this time as I have reason to believe he is currently too busy to answer. However, I would enjoy any and all speculation concerning a particular design element of the 303.

I notice that in building the AR-303 that Ken used an updated tweeter design and a superior, modern, woofer. So far we have a speaker that could sound like anything and did not have to bear any resemblance to the sound of the 3a. In fact, to this non-engineer, "forcing it" to sound like a 3a with different drivers seems like it should have been a real chore.

But then, there it is: The eyeball midrange. That was a long-time staple of AR designs, I suppose starting with the 3a, LST, 5, 10pi, LST-II, 11 (and 12?).

When AR went to the "vertical" series we lost that eyeball and picked-up a ring around the mids - I can only speculate that AR determined this was a better design. So, when Ken was designing the 303, what do you suppose/guess/fig'r was the motivation for going back to the eyeball-style midrange?

The part of our current 2ax discussion that brought this to mind was the mounting. I notice that on the 303 one could easily get the impression that Ken *needed* the cabinet front to be slick and reflective. He certainly had ample experience with foam all over the front of a cabinet (the MGC-1), so for some specific reason he incorporated this eyeball mid against a slick mounting surface.

Is that why a 3a will fill a room with a non-specific point-source "impression" better than an AR-9 will? The "far field" is closer to the speaker for some reason?

I'm not even dreaming of doing a speaker design myself, but I am interested in the relative merits of differing approaches.

By the way, what does vertical alignment of drivers do to imaging if the speakers are laid on their sides? I only have two-way "vertical" systems (and the 9s - which I'm disinclined to incline) that I can lay on their sides and that isn't telling me much.

Bret

[Guest Bret]

>There is no “sweet spot” in a symphony hall. So now the question might be, do you want to bring the orchestra into your living room or transplant yourself into concert hall? <

Tom, you put that so well; really getting to the heart of the matter.

You are talking about one of my pet peeves. If you are going to give me "near field" then I need binaural recordings. If you are going to give me far field, then I need appropriate microphone placement.

We can't solve or control that at all, so it's just "fun" to gripe about. By the way, a friend of mine with KEFs the model number of which I don't recall had a REALLY nice sounding setup. Too bad KEF went off the deep end with some strange, strange designs for a while. I see they are now back into the "co-axial" approach.

Anyway, I'm responding for a particular reason. A buddy of mine, on a budget, has gone into the world of experimentation. A couple of days ago he received an eBay-snagged Rega Planet CD player. This is one of the upsampling to 24-bit decks. He's using the onboard DAC and going to analog outputs connected to analog inputs on the preamp, connected to the amp-bone, connected to the AR-10pi's.

I am positively blown away by the difference in sound we get from this Rega compared to a Pioneer 1-bit LD, a Sony DVD, etc etc etc. The 10pi should be a "far field" design, too. But with this Rega spinning the CDs a sweet-spot appears (fairly large one), the back wall disappears, the speakers sonically become 6 feet wide, and you get the 3-D effect you'd expect from Carver's Sonic Holography or other wizardly manipulation. It's eerie to have a "sweet spot" that large and have so much "space" between instruments and layering of the depth of the image. And it is really strange to have a pair of 10pi's suddenly sound like a pair of speakers from the new design school.

The more things I try the less and less sure I am that I know anything.

Bret

[charger3834]

Bret,

Let us know what you think of the old vs the new 2ax in an a-b comparison. I have done this comparison and I will try it again, although I am not necessarily the most sophisticated listener. Also, Steve mentioned the 2a to 2ax adapter plate on early 2ax's and on modified 2a's. I have seen a picture of it in the 1969 catalog, however my early 2ax's with serial numbers in the 16000's, do not have them either. There could not have been too many produced with the plate.

Sorry for moving off topic, I arrived a little late.

[Guest Bret]

>Sorry for moving off topic, I arrived a little late. <

I think you are the only one still on topic!

Bret