AR2-AX: Just got them singing again and..

[fairchance]

I love them!!!Mine are 1971 vintage and I just did a refoam and replaced a tweeter with one dated 1973. I never thought the bass would be this good coming from a relatively light weight woofer and a smallish cabinet. Beautiful balanced sound. Just ordered new 18 count oatmeal linen for the grills. When I get it done, I'll post some pictures. I can't stop listening to my favorite female vocalists. How do the 3a's sound compare to these? Is the huge price difference worth it or is it just a collector,rarity thing.

thanks,

Dave

Chalk Hill Pa

[Mexicomike]

I had 2AXs back in the day (bought them in the mid 60's); in '71 I bought 3As. The 3a, as one would expect and as published by AR, has more top, more bottom, and better dispersion. HOWEVER, depending on placement, the 3As can be bass-heavy and I had to change the placement of the 3as as opposed to the 2axs for that reason. IMO ARs recommended placement for the 3As could make the bass boomy - at least in the room they occupied at the time. Once placement was sorted out, the 3As were clearly superior to me and, as AR used to advertise at the time, the 3A was the best speaker they knew how to make.

But the 2ax was famous for being very well balanced - it didn't go as deep or as high as the 3A but it sounded very natural because it didn't sacrifice high at the expense of low or visa versa. In fact, I did a lot of a/b testing and there were many pop recordings that sounded virtually the same on both speakers. At that time, it was primarily in classical music recordings where the 3A showed its extension on both ends and better presence (dispersion).

For a few years I ran the 3As as primary speakers with the 2AXs wired as a "phantom" 4 channel setup that was semi-popular at the time. It consisted of adding two additional speakers at the rear of the listening room, pointed at the listening position. They were wired with a single cable from one main speaker, through both "rear" speakers, and then to the other main. If I remember right, it was done this way: One wire went from the + terminal of the right main speaker to the +terminal of the right rear speaker; a wire then went from the - terminal of the right rear speaker to the - terminal of the left rear speaker; then a wire from the + terminal of the left rear speaker to the + terminal of the left main speaker (I could have the +/- thing remembered wrongly). Depending on the recording, and the difference signal in the L/R channel, this setup added a nice bit of natural-sounding "ambience" that, at times could be quite impressive.

I had the 3As and 2AXs until the mid 70's when I went on to "better" speakers...that weren't. But that's a long, stupid and boring story. I currently have/use LSTs and 3As and all those "better" speakers I bought regularly for nearly 35 years are dust...

However, re the original question - if you have a properly operating pair of 2AXs and 3As, the 3As will demonstrate their extended frequency range and dispersion improvement over the 2Ax. Whether this is worth the price depends on the use to which you will put the speakers. BUT as has often been expressed, there is probably no such thing as an AR3A or 2ax that can perform as it did when it left AR back then because there are no new AR mids/tweeters and the old ones are unlikely to perform as they did 30+ years ago, even when re-capped to specs. I think woofers are not an issue and that the originals can work now as well as they did new once refoamed.

[genek]

For a few years I ran the 3As as primary speakers with the 2AXs wired as a "phantom" 4 channel setup that was semi-popular at the time.

You'd probably feel right at home in my current HT setup. 3a's in the front and 2ax's in back.

[Mexicomike]

"You'd probably feel right at home in my current HT setup. 3a's in the front and 2ax's in back. "

Yes, I'm sure I would!

At the risk of hijacking this thread, your avatar make me decide to post a quick blurb re records. Yesterday I decided to put on some vinyl (I also have an AR TT, current Ortofon cartridge). It took about 30 seconds for me to AGAIN think, "what the heck happened?" - In many cases, vinyl from 40 years ago is clearly superior to the CD version and FAR superior to the best Itunes-available versions of the same material.

Now, I will readily admit that the convenience of digital format make it my choice 99% of the time but on those occasions where I put on a record, I can't help but think that somehow we've lost the plot as far as music reproduction. In theory, digital should be superior to analog, at least as far as dynamic range is concerned, but in practice it doesn't seem to actually happen. Though not in every case, the vinyl often has more "depth," presence, and extended frequency response. I don't know why that should be but it's pretty obvious, even to my wife who loves music but is not any sort of "audiophile."

Again, I'm sorry for this hijack but it's actually your fault, Gen, because of your avatar!

[RoyC]

I can't stop listening to my favorite female vocalists. How do the 3a's sound compare to these? Is the huge price difference worth it or is it just a collector,rarity thing.

thanks,

Dave

Chalk Hill Pa

Dave,

I actually prefer female vocals through the 10 inch woofer (3-way) models of that era (AR-2ax and AR-5) over the AR-3a. I have had many opportunities to compare many specimens, and always come to the same conclusion. The price difference is not justified in that regard, imo.

Roy

[fairchance]

Thanks guys. I'll probably be on a quest for a pair of 3a's in need of some love.

[Luzifull22]

yess OK Thankss! :) :)

[Steve F]

I love them!!!Mine are 1971 vintage and I just did a refoam and replaced a tweeter with one dated 1973. I never thought the bass would be this good coming from a relatively light weight woofer and a smallish cabinet. Beautiful balanced sound. Just ordered new 18 count oatmeal linen for the grills. When I get it done, I'll post some pictures. I can't stop listening to my favorite female vocalists. How do the 3a's sound compare to these? Is the huge price difference worth it or is it just a collector,rarity thing.

thanks,

Dave

Chalk Hill Pa

‘New’ 2ax (post 1970) vs. 3a

This is a great question.

Although the 3a’s bass, according to AR, extended approximately “1/3 of an octave lower,” it sounded even more dramatic than that. While the 2x/2ax/5 all had solid, tight, full bass, the 3a’s bass was magisterial, especially for its day. To be listening to a good recording and all of a sudden, without warning, to hear and feel a really low tone come across the room and fill the pit of your stomach was really something. The 10-inchers never shocked you like that. The 3a did.

However……

Many people will tell you that the 3a had a definite “heavy’ or slightly “sloggy” character to its sound. For whatever reason, the lower mid region seemed to project a kind of “thickness” that at times made the 3a sound slow on its feet and lacking in detail—especially when it was compared in a retail showroom A-B against the Large Advent.

In addition, the 3a’s mid is rightly characterized as having a bit of a honky or “woody” coloration. The 2ax didn’t seem to suffer from these traits, and despite its less extended bass, the 2ax could often sound more natural and less colored than the 3a.

BTW, there was no difference in their “highs.” They both used the same ¾” dome tweeter, both crossed over at 5000 Hz. The 3a did have superior midrange dispersion, with its smaller-diameter 1 ½” dome mid driver covering a wider frequency range than the 2ax’s 3 ½” midrange cone.

The 3a was almost twice the money of the 2ax: $250 ea. vs. $128 ea. Value was the 2ax’s strong suit.

But now, 40+ years after the fact, the 3a might have greater appeal if just for its historical position alone. I’d take the 3a today. And no amount of low-frequency compensation, no matter how judicious or frequency-specific, could make the 2ax’s bass sound like the 3a’s.

Steve F.

[tysontom]

‘New’ 2ax (post 1970) vs. 3a

This is a great question.

Although the 3a’s bass, according to AR, extended approximately “1/3 of an octave lower,” it sounded even more dramatic than that. While the 2x/2ax/5 all had solid, tight, full bass, the 3a’s bass was magisterial, especially for its day. To be listening to a good recording and all of a sudden, without warning, to hear and feel a really low tone come across the room and fill the pit of your stomach was really something. The 10-inchers never shocked you like that. The 3a did.

However……

Many people will tell you that the 3a had a definite “heavy’ or slightly “sloggy” character to its sound. For whatever reason, the lower mid region seemed to project a kind of “thickness” that at times made the 3a sound slow on its feet and lacking in detail—especially when it was compared in a retail showroom A-B against the Large Advent.

In addition, the 3a’s mid is rightly characterized as having a bit of a honky or “woody” coloration. The 2ax didn’t seem to suffer from these traits, and despite its less extended bass, the 2ax could often sound more natural and less colored than the 3a.

BTW, there was no difference in their “highs.” They both used the same ¾” dome tweeter, both crossed over at 5000 Hz. The 3a did have superior midrange dispersion, with its smaller-diameter 1 ½” dome mid driver covering a wider frequency range than the 2ax’s 3 ½” midrange cone.

The 3a was almost twice the money of the 2ax: $250 ea. vs. $128 ea. Value was the 2ax’s strong suit.

But now, 40+ years after the fact, the 3a might have greater appeal if just for its historical position alone. I’d take the 3a today. And no amount of low-frequency compensation, no matter how judicious or frequency-specific, could make the 2ax’s bass sound like the 3a’s.

Steve F.

Steve gives an interesting, and I believe accurate, accounting of the apparent sonic differences between a speaker like the AR-2a^x and the AR-3a of the 1970-vintage period.

As for bass, the AR-3a's 12-inch woofer definitely had an advantage over the 2ax's 10-inch woofer, but the differences were fairly subtle until you put on music with sub-sonic frequencies, pipe-organ bass or perhaps a jazz recording with a (sometimes rare) well-recorded kick-drum bass line. The beauty of AR's 12-inch woofer was its very low distortion. Depending on the setting of the level controls and the music content, the bass could sound "heavy" at times.

Yet the AR-3a did have some perceived issues, some of which were mentioned in a 1970 issue of Consumer Reports magazine. Prior to that, however, the organization tested some of the earlier speakers, and the AR-3 (the famous AR speaker model that preceded the AR-3a) did indeed receive a top rating in the April 1965 issue. It was rated along with the ADC 303A, AR-2x, AR-2ax, KLH Model Six, Fisher XP-7, Sherwood SR-3, EMI DLS-529 and the KLH Model Four. In this test, CU noted: "The AR-3 produced the deepest bass of any loudspeaker tested. Sound in highs particularly smooth. Within the check-rated group were the two most expensive loudspeakers tested, the KLH Four and the AR-3. CU found that their prices buy you profoundly clear and deep bass response -- the lowest octave in the musical spectrum. In this they clearly stood a notch above the others tested, with the AR-3 having a slight edge on the KLH Four. In the vital higher frequencies, however, neither speaker was considered particularly better than the others in their group. For instance, the KLH Four was slightly peaky in the treble, lending its sound a mild 'presence.' And the 'smoothness' that CU noted in the AR-3 was in reality a small defect, the result of a slight, but measurable lack of energy in the region near 8000 Hz -- the region of much significant music as well as hiss and record scratch."

In May 1970, CU again looked at loudspeakers, and although they did not specifically test the more-expensive models such as the relatively new AR-3a and Bose 901, the organization decided to look at the new unit even though it was outside of the scope of the speakers being tested in this issue. It was in this issue that CU opened a can of worms with its comments on the Bose 901 (resulting in a major lawsuit) and almost unnoticed, there was a paragraph on the AR-3a, which stated, "In April 1965, CU check-rated the Acoustic Research AR-3 loudspeaker and noted that it had the deepest bass response of any speaker we had ever tested. Naturally, we were curious about the quality of its successor, the AR-3a. So we decided to test it even though its price ($250) put it well outside the scope of the accompanying report. The AR-3a delivered just as good bass response as it predecessor, but we heard and measured a distinct exaggeration in the middle frequency register that made the speaker sound thick and heavy. (On investigation, we traced the peak to the AR-3a's new mid-range unit, combined with a similar problem in the woofer). We found a similar problem in the treble, too. Even with the tweeter control fully advanced, the tweeter could not match the relative loudness of the woofer. That made the treble in the AR-3a sound a bit distant, although the tweeter frequency response was very smooth and extended. Had we ranked the AR-3a with the other speakers in this report, it would have fallen between the high- and medium-accuracy groups. Thus, in CU's view, the AR-2ax is a better speaker than the AR-3a -- and the AR-2ax costs only half as much."

Interestingly, all of the audio-magazine test reports of the AR-3a were simply stellar, with no mention of the problems that CU encountered. Many magazines stated that the AR-3a was for them "the reference standard," and so forth. This is not to say that CU was wrong, but there has always been a lingering question about the listening environment of the CU laboratories at the time of the 1970 test report. It should be noted, too, that AR did modify the crossover in the AR-3a around the time of this test -- an inductor value was changed slightly, but probably mostly to accommodate the new ferrite-magnet 200003-0 woofer used after late 1969 in the AR-3a.

I have owned and used several versions of the AR-3, AR-3a and AR-2ax. My impression is that you must get well back in the room when listening to the 3 and and 3a -- more so than with the AR-2ax. In fact, the AR-2ax images much better than an AR-3a, simply because the midrange is much more directional. Moreover, the very wide dispersion of both the AR-3 and 3a cause considerably more destructive interference, lobing and so forth when heard up close in the near field. These anomalies are almost impossible to detect well back in the sound field, and the superior dispersion in the midrange pays big dividends when listening in the reverberant field. What it really says is that the AR-3a is somewhat more finicky about placement in a listening room than the AR-2ax, but once it is placed properly, it can produce a much more spacious and natural sound.

--Tom Tyson

[Mexicomike]

"What it really says is that the AR-3a is somewhat more finicky about placement in a listening room than the AR-2ax, but once it is placed properly, it can produce a much more spacious and natural sound."

Totally agree with that; it precisely mirrors my experience back in the day with the two speakers. As an aside, CU was never considered a valid tester of loudspeakers by serious audiophiles during that age. Whether that was appropriate or not, I don't really know. I know that I seldom agreed with their opinions when I listened to "my next possible speaker purchase." Which I did quite frequently - I was very much into constant "improvement" (as I saw it then) from around 1964 to around 1990.

[genek]

CU's camera tests were a source of similar discontent among "serious" photographers as well. But I suspect that they were probably pretty good simulations of how the majority of not-so-serious consumers used the stuff they bought.

[soundminded]

Steve gives an interesting, and I believe accurate, accounting of the apparent sonic differences between a speaker like the AR-2a^x and the AR-3a of the 1970-vintage period.

As for bass, the AR-3a's 12-inch woofer definitely had an advantage over the 2ax's 10-inch woofer, but the differences were fairly subtle until you put on music with sub-sonic frequencies, pipe-organ bass or perhaps a jazz recording with a (sometimes rare) well-recorded kick-drum bass line. The beauty of AR's 12-inch woofer was its very low distortion. Depending on the setting of the level controls and the music content, the bass could sound "heavy" at times.

Yet the AR-3a did have some perceived issues, some of which were mentioned in a 1970 issue of Consumer Reports magazine. Prior to that, however, the organization tested some of the earlier speakers, and the AR-3 (the famous AR speaker model that preceded the AR-3a) did indeed receive a top rating in the April 1965 issue. It was rated along with the ADC 303A, AR-2x, AR-2ax, KLH Model Six, Fisher XP-7, Sherwood SR-3, EMI DLS-529 and the KLH Model Four. In this test, CU noted: "The AR-3 produced the deepest bass of any loudspeaker tested. Sound in highs particularly smooth. Within the check-rated group were the two most expensive loudspeakers tested, the KLH Four and the AR-3. CU found that their prices buy you profoundly clear and deep bass response -- the lowest octave in the musical spectrum. In this they clearly stood a notch above the others tested, with the AR-3 having a slight edge on the KLH Four. In the vital higher frequencies, however, neither speaker was considered particularly better than the others in their group. For instance, the KLH Four was slightly peaky in the treble, lending its sound a mild 'presence.' And the 'smoothness' that CU noted in the AR-3 was in reality a small defect, the result of a slight, but measurable lack of energy in the region near 8000 Hz -- the region of much significant music as well as hiss and record scratch."

In May 1970, CU again looked at loudspeakers, and although they did not specifically test the more-expensive models such as the relatively new AR-3a and Bose 901, the organization decided to look at the new unit even though it was outside of the scope of the speakers being tested in this issue. It was in this issue that CU opened a can of worms with its comments on the Bose 901 (resulting in a major lawsuit) and almost unnoticed, there was a paragraph on the AR-3a, which stated, "In April 1965, CU check-rated the Acoustic Research AR-3 loudspeaker and noted that it had the deepest bass response of any speaker we had ever tested. Naturally, we were curious about the quality of its successor, the AR-3a. So we decided to test it even though its price ($250) put it well outside the scope of the accompanying report. The AR-3a delivered just as good bass response as it predecessor, but we heard and measured a distinct exaggeration in the middle frequency register that made the speaker sound thick and heavy. (On investigation, we traced the peak to the AR-3a's new mid-range unit, combined with a similar problem in the woofer). We found a similar problem in the treble, too. Even with the tweeter control fully advanced, the tweeter could not match the relative loudness of the woofer. That made the treble in the AR-3a sound a bit distant, although the tweeter frequency response was very smooth and extended. Had we ranked the AR-3a with the other speakers in this report, it would have fallen between the high- and medium-accuracy groups. Thus, in CU's view, the AR-2ax is a better speaker than the AR-3a -- and the AR-2ax costs only half as much."

Interestingly, all of the audio-magazine test reports of the AR-3a were simply stellar, with no mention of the problems that CU encountered. Many magazines stated that the AR-3a was for them "the reference standard," and so forth. This is not to say that CU was wrong, but there has always been a lingering question about the listening environment of the CU laboratories at the time of the 1970 test report. It should be noted, too, that AR did modify the crossover in the AR-3a around the time of this test -- an inductor value was changed slightly, but probably mostly to accommodate the new ferrite-magnet 200003-0 woofer used after late 1969 in the AR-3a.

I have owned and used several versions of the AR-3, AR-3a and AR-2ax. My impression is that you must get well back in the room when listening to the 3 and and 3a -- more so than with the AR-2ax. In fact, the AR-2ax images much better than an AR-3a, simply because the midrange is much more directional. Moreover, the very wide dispersion of both the AR-3 and 3a cause considerably more destructive interference, lobing and so forth when heard up close in the near field. These anomalies are almost impossible to detect well back in the sound field, and the superior dispersion in the midrange pays big dividends when listening in the reverberant field. What it really says is that the AR-3a is somewhat more finicky about placement in a listening room than the AR-2ax, but once it is placed properly, it can produce a much more spacious and natural sound.

--Tom Tyson

This thread and the thread comparing AR3 to AR3a point up the difficulty of designing a 3 way loudspeaker system to cover the entire audible range as though they were a single unit whose combined output is the equivalent of one which exhibits the characteristics of a device that is not resonant over ten audible octaves of sound. This arises because most loudspeaker drivers have a range of about 2 1/2 to 3 octaves over which they are not particularly resonant. Something's got to give. Fix a problem in one region and a new one pops up somewhere else. One of AR3's problems is in the upper region of the woofer between 500hz and 1 khz. In this range, the woofer becomes more directional, acts less like a piston, and has an irregular response. Curves for a 12" driver in the same vein, say Peerless 830500 and even others of that type (XLS and XXSL) exhibit irregularity typical for them in that range. Move the crossover frequency down an octave and install an improved midrange dome and the problem is at the low end of the dome's range. This was a problem with AR3. Make the woofer smaller to improve its upper range as in AR2ax and the deepest bass suffers by comparison. Replace the small dome midrange with a 3 1/2 inch cone driver and dispersion in the upper end of the midrange is reduced.

AR 9 blew the woofer midrange interface problem away. Adding a forth driver, the 8" LMR between the 12" woofer and the dome midrange allowed those drivers to be restricted to the ranges they work at best. The execution was excellent, very clever, very effective. Adding the second 12" woofer was the frosting on the cake. When properly functioning (restored with a suitable amplifier) program material with deep bass demonstrates that in this regard AR9 simply outclasses AR3/3a in the same way AR3a outclasses AR2ax.

It should be kept in mind that AR's goal was to create speakers that would sound like acoustic musical instruments. To this end it was part of a system which included not only all of the playback equipment but the recording equipment and recording techniques and variables as well. But the available program material and the range and capabilities of other playback equipment available to consumers was far more limited in the 1960s and 1970s than it is today. The capabilities of AR speaker systems under optimal conditions could not be fully realized due to those limitations. For example, few vinyl phonograph records had the low end content of a compact disc and if they did, acoustic feedback was a constant problem. AR9 demonstrated that it could cause even Denon 1520 CD player to lose tracking on heavy bass passages. Thankfully other players I've used are better isolated. The use of a graphic equalizer to smooth out the response of classic speakers brings new life to them and allows them to exhibit far better (more tonally accurate) sound than was possible for consumers to obtain from them at the time they were manufactured and marketed. I recommend using two, a 10 band for equalizing the speaker in the room and a five or seven band for equalizing individual recordings. Among the most surprising results to me is just how really accurate and excellent AR2ax can sound. When properly adjusted in this way it is a superb speaker, far better than I ever heard it decades ago when it was marketed. The equalizer overcomes the inadequate tweeter output when compared to the midrange and woofer, compensates for its high frequency rolloff, eliminates a small notch in its midrange, and extends its deep bass. Within its power handling capabilities a pair could produce very deep undistorted bass in a 3500 square foot basement. In direct comparison with KLH model 6, with no equalization applied AR2ax comes off as a muffled poor performer while KLH model 6 has its own shortcomings. When adequately corrected not only do the differences sharply diminish, they can be made to sound almost identical.

Anyone familiar with Bose 901 and is perfectly honest about them knows that when used as manufactured and recommended, they cannot reproduce the highest octave of sound. This was confirmed in the newest version by listening to series 6 very briefly about 3 years ago. Series I and II also exhibit a clearly audible peak in the 200 to 500 hz range of about 7 or 8 db and the lowest bass falls off much too steeply. (I'm not sure what the characteristics in this regard are for later versions but in all likelihood, whatever the peak or lack of it in the upper bass lower midrange, at the very least the lowest octave may have been sacrificed for efficiency in the ported designs starting with series III.) This gives credence to the old saying "no highs, no lows, it's Bose. The Series I and II ability to produce the deepest bass when all controls are flat are only evident when they are played at very loud volume with a lot of amplifier power available. If a 3 way speaker is a difficult thing to design, a one way system is just about impossible. It would therefore come as a surprise that this speaker can not only be rescued, it can be re-engineered to be the most accurate speaker I own or have ever heard, in fact except for its deep bass capabilities, beating out AR9. However, my first attempt was an abject failure and the second attempt after much more experience with other speakers and understanding where Bose 901s problems were took nearly four years before I considered it entirely successful.

[Mexicomike]

As has often been noted, much of this is about how you will actually use the speakers. And to me, part of the whole classic speaker thing is about "what they were" at the time. For example, the 3 and the LST were considered major milestones in speaker design at the time; later AR speakers did not hold the same level of distinction. THAT IS NOT to say that the 9, for example, wasn't a better speaker, just that by that time AR was no longer considered to be at the same level relative to others that they were in, say 1973. IMO Nestorovic 5AS speakers outperformed anything AR made in the mid 80s which is why I bought them after auditioning them in my home along side numerous other speakers, including the AR 9. So to me a lot of this is just wanting to have some old classic speakers that perform pretty well. I guess I'd compare the AR 3/9 discussion to comparing a 63 split window Corvette Sting Ray to an 1985 model. The 85 will outperform it but the '63 is an icon. But neither will perform as well as a current Hyundai sedan. So if I wanted REAL speaker performance, I wouldn't try to accomplish it using a 30-40 year old speaker. Then again, since I can no longer hear beyond 13khz, I'm not sure that any speaker, regardless of how how good its response, would sound better to me.

And I DO like the LSTs imaging and soundstage...

[tysontom]

"...One of AR3's problems is in the upper region of the woofer between 500hz and 1 khz. In this range, the woofer becomes more directional, acts less like a piston, and has an irregular response. ...."

And then there is the issue of the AR-3's 1 kHz woofer crossover. There has been a lot of conjecture on the pitfalls of a 1 kHz crossover for the AR-3, and it's true that 1 kHz is a high x-over for a 12-inch woofer, but there are two important things to consider: (1) the AR-3 woofer's "moving" parts are actually more like 8½-9-inches in diameter, and the anechoic-measured response of this woofer at 60° off-axis at 1 kHz is -5dB. At 1000 Hz at 30° off axis, the AR-3 woofer is down only 2 dB! This is not terrible. On axis, the AR woofer is ±1.5dB from 38-1000 Hz -- excellent by any standard; (2) The 2-inch AR-3 dome midrange driver overlaps the woofer's response at the crossover frequency down to around 700-800 Hz, though rolling off to around 8-10 dB at that point, at least it assists the woofer in off-axis energy at that frequency. Therefore, there's not this "falling-off-the-cliff" decline in off-axis woofer performance or response "roughness" characterized by some here regarding the AR-3. Certainly, the AR-3a's lower 575 Hz crossover was a significant improvement over the AR-3's 1 kHz crossover, but the differences are much more subtle than some might think. More importantly might be the smoothness of all these drivers and their integrated fr in the far field -- and in this respect, both speakers do exceedingly well!

--Tom Tyson