Steve F

The 'best' AR 12" bookshelf speaker

71 posts in this topic

OK all, here’s a random thought. A friend of mine recently joined the Classic Speakers discussion forum and has, after all these years, “discovered” AR speakers. He’s heard my 9’s and my 3a’s and has decided to seek out a pair of 303’s or restored 11/10 Pi’s.

It got me to thinking. What was the ‘best’ AR 12” bookshelf speaker? Then it came to me:

The ‘best’ 12” bookshelf speaker was probably the original AR-58. The very first incarnation of that model, only in production for a very short time. The first 58 had a 1 ½” dome midrange and a ¾” dome tweeter. It was essentially an AR-91 in bookshelf—not floorstanding—guise.

What could arguably be said to make this “better” than the AR-11/10 Pi was the fact that the 58 had its drivers vertically-arrayed (when the cabinet was vertical, in the most-used orientation), as opposed to the 11/10 Pi, which had the mid-tweeter “incorrectly” horizontally-arrayed.

For those sharp-eyed members who are going to jump in with the suggestion that the AR-78LS was the “best” 12” 3-way bookshelf, I say this:

1. You may be right, since the 78 had the “dual-dome” mid-tweeter assembly, for uniform dispersion both horizontally AND vertically

2. The 78 may have been more of a ‘compact floorstander’ than ‘bookshelf’ speaker.

So I present the original, short-lived 58 as the ultimate AR 12” bookshelf speaker. The very latest AR-9-generation midrange and tweeter, vertically-arrayed, in a bookshelf, 3a-sized cabinet.

It’s so rare that even I don’t have literature on it, and virtually no remembers it. But I do, and it was a great speaker.

Steve F.

Share this post


Link to post
Share on other sites

Maybe the best 10" system?

There has been some discussion of the AR-12 and some have called it a "newer, better AR-5." I've never seen them, but based on some discussion such as the following thread, it appears they were 3-way with a 10" woofer.

Kent

Share this post


Link to post
Share on other sites

Maybe the best 10" system?

There has been some discussion of the AR-12 and some have called it a "newer, better AR-5." I've never seen them, but based on some discussion such as the following thread, it appears they were 3-way with a 10" woofer.

Kent

Yes, the 12 was an excellent 10" bookshelf speaker. However, it was more of a 'combination' 2ax-5 than a direct replacement to either one. The 12 used a 2 1/4" cone midrange with a spiderless design, immersed in ferrofluid. AR claims that this was the first use of ferrofluid in the industry, although that was always disputed by EPI. (I asked the founder of Ferrofluidics--who supplied the substance--and they said AR placed the first order. Appearance on the market in a finished product may be another matter, however.)

Interestingly enough, AR said the ferrofluid in the 12's midrange was as beneficial for coil centering as it was for cooling, hence the spiderless design.

But fans of the 5 will point out that the 5 used a dome midrange which had infinitesimally better dispersion than the 12's 2 1/4" unit, so it wasn't a clear-cut "win" for the 12. Also the fact that the 5 was a "3a with a 10-inch woofer" means that the 12 "should" have been a 10" version of the 11.

Did you ever see or hear the original 58?

Steve F.

Share this post


Link to post
Share on other sites

I'd love to hear any of them, as while I am an AR fan, my experience is limited to my 18's, and my 312HO's....My question is what gives the 58 an advantage over the 90's 303/303a? while all 3 drivers are not vertically aligned (mid and tweet are offset from the woofer, which may have advantages in baffle step issues?), they are mirrored. IIRC the 303a's can be bi-amped, which to me would also be an advantage. and when taking age into consideration, the fact that the 303/303a's are 15 years newer, and IIRC rubber surround woofers, age deterioration when considering which is best "right now" might put the 58's/78's etc at at disadvantage.

Share this post


Link to post
Share on other sites

I'd love to hear any of them, as while I am an AR fan, my experience is limited to my 18's, and my 312HO's....My question is what gives the 58 an advantage over the 90's 303/303a? while all 3 drivers are not vertically aligned (mid and tweet are offset from the woofer, which may have advantages in baffle step issues?), they are mirrored. IIRC the 303a's can be bi-amped, which to me would also be an advantage. and when taking age into consideration, the fact that the 303/303a's are 15 years newer, and IIRC rubber surround woofers, age deterioration when considering which is best "right now" might put the 58's/78's etc at at disadvantage.

Your comments about the 303 are absolutely valid and I agree with every one of them.

However...., the 58 could legitimately be considered as part of the 'original' AR lineage, coming from the company while it was still based in MA and being intro'd just after the time of the 9/91 (1980-ish).

The 303 series was designed by Ken Kantor around 1995 after AR had moved to CA.

My comments about the 58 refer to the 'best' of the so-called "original" 12" ARs, of which the 58 could still be considered one.

Steve F.

Share this post


Link to post
Share on other sites

Hi Steve

Was it the AR-58S you are referring to?

Minh Luong

Share this post


Link to post
Share on other sites

Hi Steve

Was it the AR-58S you are referring to?

Minh Luong

No, the straight 58, no letter suffix. If the "S" version had the 1 1/2" dome mid, then yes, that one also. But most of the various 58's were L, LS, LSi, etc, and they all had the 4" cone midrange. The very early 58's were direct descendants of the 3-3a-11. A very nice speaker, indeed.

Steve F.

Share this post


Link to post
Share on other sites

Hi Steve

Did the Original AR-58 use AR-11 midrange or AR-9 midrange? I used to own 2 pairs of AR-58S(real walnut veneer), a pair of later AR-58S(walnut vinyl), a pair of AR-58Bx and a pair of AR-58Bxi(both walnut vinyl). I love the AR-58S with real walnut veneer over others. The AR-58S cabinet actually feel feel more solid compared to the AR-11 cabinet and I do like the vertical drivers layout as well.

Minh Luong

Share this post


Link to post
Share on other sites

Most AR speakers starting with the ADD series were designed to be positioned vertically, and looked downright "wrong" when turned on their sides. I'm not exactly sure what a speaker that requires a 20" or higher space between shelves would be called, but I know I don't have any bookshelves that tall in my library.

Share this post


Link to post
Share on other sites

Hi Steve

Did the Original AR-58 use AR-11 midrange or AR-9 midrange? I used to own 2 pairs of AR-58S(real walnut veneer), a pair of later AR-58S(walnut vinyl), a pair of AR-58Bx and a pair of AR-58Bxi(both walnut vinyl). I love the AR-58S with real walnut veneer over others. The AR-58S cabinet actually feel feel more solid compared to the AR-11 cabinet and I do like the vertical drivers layout as well.

Minh Luong

It used the AR-91 midrange, which was very similar to the AR-9 midrange, except that it had a slightly lower resonance that allowed it to cross over from the woofer in a 3-way design. It wasn't the mesh-covered midrange from the 3a/11.

Yes, you are quite right in noting that the later 58's were B, BX, BXi, not L, LS, LSi as I said earlier. The "B's" were AR's designation for their lower-end speakers in the early-mid '80's; the "L's" were AR's designation for their higher-end speakers in the early-mid '80's.

Quite honestly, the 58B, BX, and BXi (the 4" mid models) didn't interest me all that much, as they were clearly indicative of a company that had lost its edge and was just filling market price points with arbitrary models.

Steve F.

Share this post


Link to post
Share on other sites

I have a pair of 58LS with the later 4" cone mid. I also have pairs of 3a and 2ax.

The 58LS is an outstanding speaker in it's own right and one of my favourites to listen to. However, it doesn't have quite the same level of detail in the mid range that the 3a and 2ax both have in spades. I can imagine that the 58 with the dome tweeter could well be up there amongst the very best. The LS has better top end than either the 3a/2ax(perhaps the hf tweeter has deteriorated less obviously than the older speakers but I suspect that this was intentional as the evolution of speakers seems to continue with ever brighter sounding tweeters) but the mid range is just not quite as well defined. I have found with the 3a/2ax that they are much less prone to sounding different with different amps than the 58LS, which can be become bizarrely bright with modern amps. I have a pair of 98LS which I have changed to LSI (crossover mod) which was definitely needed to calm down to the upper mid. I can't imagine who let the original through QA sounding like that! In fact, I still have it on my to do list to tame the upper mid slightly more and bring the lower mid up slightly to balance things further.

For sheer accuracy, transparency and balance, the 3a and the 2ax win hands down, each having their own "best bits". The 2ax renders woodwind superbly for instance, and on the 3a strings are excellent. I'm guessing the 58 with dome would perhaps have the best of all worlds with the detail of the older midrange.

Share this post


Link to post
Share on other sites

No, the straight 58, no letter suffix. If the "S" version had the 1 1/2" dome mid, then yes, that one also. But most of the various 58's were L, LS, LSi, etc, and they all had the 4" cone midrange. The very early 58's were direct descendants of the 3-3a-11. A very nice speaker, indeed.

Steve F.

Steve,

This is certainly an interesting comparison! I am embarrassed to say that I have no information whatsoever -- either for parts information or literature -- on a "straight" AR58. It may well exist, but the model you have described that I know is the AR58s, and it used the 12-inch woofer, the 1½" dome and the ¾" tweeter from the AR9. Is it the best of the single-woofer 12-inch AR line? That's debatable, of course, but it was a fine speaker. The AR58s was originally described as a "bookshelf" speaker (it would have to have been placed vertically on a bookshelf for optimum performance), but with the advent of AR "Verticals," it was clearly intended as a floor-or-stand-mounted speaker, and it was "tuned" (highly damped) more for floor and stand use because of its fairly low "Q" of 0.5 or 0.6 -- designed to attenuate the bass at resonance slightly to keep it from sounding bass-heavy when placed on the floor. Besides, people were getting away from "bookshelf" speaker placement by this time and usually placing the speakers on the floor or on a stand/shelf. However, a minor problem for the AR58s was its crossover frequencies of 700 Hz and 7.5 kHz (the usual half-section LC AR crossover type), which placed a slight "strain" on the off-axis performance of the 58s' 12-inch woofer. This was also true for the AR78s, which was the single-woofer counterpart to the AR9LS-series. Well, the original AR-3 also had a high woofer crossover, but it was not a serious problem despite what many people thought, mainly because the midrange's crossover slope was so gradual that it extended well down into the upper-bass range. Nevertheless, it is always better to keep the woofer out of the upper frequencies.

How about the AR98LS? It was by no means a "bookshelf" speaker -- no less optimal than the AR58s in a bookshelf -- yet with its 200 Hz woofer crossover, it was clearly superior to any of the other 12-inch speakers of this genre in terms of power-handling capability, off-axis performance in the midrange and midrange and upper-bass smoothness, due to its 4-way design. With the Lambda tweeter, I believe this design was probably superior to the discrete midrange/tweeter design of the AR58 -- much in the manner in which the AR98LS is arguably smoother and flatter than the original AR9.

--Tom Tyson

post-100160-0-56948600-1294337352_thumb.

Share this post


Link to post
Share on other sites

The 98LS should be a great choice for the best 12" system, but I've found the floor-standing AR-91 to be an overall better-integrated speaker in a moderately-sized room - it's not a booksehlf model, though, so it's not in the running.

I'm not sure if it was the LS's integrated dual-dome approach, but my preference has been for the discrete mid/tweeter setup of the earlier Vertical systems, and much prefer the AR-9 and AR-90 systems in a large room, where they clearly outclass the later LS series in bass extension & impact, as well as providing a very smooth transition between the multiple drivers.

In a large listening space, with prodigious amplification, the 9 and 90 come very close to presenting a seamless image - especially on grand piano, which is an excellent test.

Share this post


Link to post
Share on other sites

Steve,

This is certainly an interesting comparison! I am embarrassed to say that I have no information whatsoever -- either for parts information or literature -- on a "straight" AR58. It may well exist, but the model you have described that I know is the AR58s, and it used the 12-inch woofer, the 1½" dome and the ¾" tweeter from the AR9. Is it the best of the single-woofer 12-inch AR line? That's debatable, of course, but it was a fine speaker. The AR58s was originally described as a "bookshelf" speaker (it would have to have been placed vertically on a bookshelf for optimum performance), but with the advent of AR "Verticals," it was clearly intended as a floor-or-stand-mounted speaker, and it was "tuned" (highly damped) more for floor and stand use because of its fairly low "Q" of 0.5 or 0.6 -- designed to attenuate the bass at resonance slightly to keep it from sounding bass-heavy when placed on the floor. Besides, people were getting away from "bookshelf" speaker placement by this time and usually placing the speakers on the floor or on a stand/shelf. However, a minor problem for the AR58s was its crossover frequencies of 700 Hz and 7.5 kHz (the usual half-section LC AR crossover type), which placed a slight "strain" on the off-axis performance of the 58s' 12-inch woofer. This was also true for the AR78s, which was the single-woofer counterpart to the AR9LS-series. Well, the original AR-3 also had a high woofer crossover, but it was not a serious problem despite what many people thought, mainly because the midrange's crossover slope was so gradual that it extended well down into the upper-bass range. Nevertheless, it is always better to keep the woofer out of the upper frequencies.

How about the AR98LS? It was by no means a "bookshelf" speaker -- no less optimal than the AR58s in a bookshelf -- yet with its 200 Hz woofer crossover, it was clearly superior to any of the other 12-inch speakers of this genre in terms of power-handling capability, off-axis performance in the midrange and midrange and upper-bass smoothness, due to its 4-way design. With the Lambda tweeter, I believe this design was probably superior to the discrete midrange/tweeter design of the AR58 -- much in the manner in which the AR98LS is arguably smoother and flatter than the original AR9.

--Tom Tyson

The 58s may well be the model I'm thinking of--as I said, I have no literature on it.

I'm keeping this discussion very narrowly-focused: 12" 3-ways with dome mid-tweeters, built before AR's move to CA and having cabinets that could be (loosely) described as "bookshelf."

Under my admittedly narrow, arbitrary definition, I think the 58s could be the so-called 'best' AR 12" 3-way bookshelf speaker.

As far as the midrange dispersion of the AR 12" woofer is concerned, simple math calculations of piston diameter vs. wavelength indicate that the 'small' AR 12" woofer (11 inches with a 1/2" surround, for an effective piston diameter of 10 inches, max) would not become seriously directional until after 1000 Hz. Some people characterize the AR 12" woofer as having a piston of only 9.5 inches, which strengthens the 'it's not directional below 1000 Hz' argument even further. In any event, if a listener could reliably and repeatedly distinguish between the midrange dispersion characteristics of the AR 12" woofer as used in the 3a/11 (575 or 525 Hz x-o) and the midrange dispersion of the 58s (700 Hz x-o), hats off to them.

Would never happen. Seriously, the standard 5% x-o component tolerance would create slightly shifting actual x-o frequencies from unit to unit (even though they might easily pass FR QC) such that one end of the tolerance spectrum of an 11 and the other end of the tolerance spectrum of a 58s could result in woofer-midrange crossover frequencies that are nearly identical between the two speakers. That's the real world, and let's not kid ourselves.

I just think the 58s with its very-latest-generation domes was one heckuva speaker--and very much unheralded, too.

(Undoubtedly, at least the equal of the 303, no?)

Steve F.

Share this post


Link to post
Share on other sites

The 58s may well be the model I'm thinking of--as I said, I have no literature on it.

I'm keeping this discussion very narrowly-focused: 12" 3-ways with dome mid-tweeters, built before AR's move to CA and having cabinets that could be (loosely) described as "bookshelf."

Under my admittedly narrow, arbitrary definition, I think the 58s could be the so-called 'best' AR 12" 3-way bookshelf speaker.

As far as the midrange dispersion of the AR 12" woofer is concerned, simple math calculations of piston diameter vs. wavelength indicate that the 'small' AR 12" woofer (11 inches with a 1/2" surround, for an effective piston diameter of 10 inches, max) would not become seriously directional until after 1000 Hz. Some people characterize the AR 12" woofer as having a piston of only 9.5 inches, which strengthens the 'it's not directional below 1000 Hz' argument even further. In any event, if a listener could reliably and repeatedly distinguish between the midrange dispersion characteristics of the AR 12" woofer as used in the 3a/11 (575 or 525 Hz x-o) and the midrange dispersion of the 58s (700 Hz x-o), hats off to them.

Would never happen. Seriously, the standard 5% x-o component tolerance would create slightly shifting actual x-o frequencies from unit to unit (even though they might easily pass FR QC) such that one end of the tolerance spectrum of an 11 and the other end of the tolerance spectrum of a 58s could result in woofer-midrange crossover frequencies that are nearly identical between the two speakers. That's the real world, and let's not kid ourselves.

I just think the 58s with its very-latest-generation domes was one heckuva speaker--and very much unheralded, too.

(Undoubtedly, at least the equal of the 303, no?)

Steve F.

Steve,

Well, this whole thing is "nitpicking" to the max, but it's fun to debate it!

Therefore… I think the biggest challenge to your argument has to do with the configuration of the AR58s: it's two or three inches taller on the long dimension than an AR-3a, for example, and it has a vertical midrange-tweeter alignment orientation. As you know, if you mount a "vertical" speaker horizontally -- as though on a bookshelf -- it will have truncated horizontal dispersion, and the speaker will not sound quite as good as it would mounted vertically. No big deal, but there is a difference. By pure happenstance, the AR-3/AR-3a midrange/tweeter orientation is staggered somewhat, such that it works about as well horizontally or vertically, but The AR58s was intended to be placed vertically, regardless of its mounting location, and there are precious-few "bookshelves" with 28-29-inch-high shelves. This type of speaker will almost always end up on the floor, on the top of some shelf or on stands -- probably not in a bookshelf in the classic way.

AR called it a "bookshelf" speaker, because it was the only thing with a 12-inch woofer configuration that could be labeled as such, and it was a sort of interim speaker design, but it's probably not a true bookshelf speaker in the true sense of the word. I believe that AR was afraid they might lose some traction with the folks who owned AR-3s, AR-3as, AR-10s and so forth if they didn't mention "bookshelf" in the description. Strangely, the AR-10 and AR-11 were nearly always shown on the speaker stands perched vertically, whereas those speakers sound somewhat better mounted horizontally!

I'm glad you make the argument about the piston diameter and the 1000 Hz crossover for a 12-inch woofer! This reinforces the argument about the crossover in the AR-3! Nonetheless, the lower the crossover for a heavy moving system, probably the better it is -- not just for off-axis response, but impulse response as well.

Since you can mount the AR98LS vertically in a cabinet just about as easily as an AR58s, I'd pick the 98 first. I seem to remember that it was "cleaner" sounding and much more "effortless" than most of the 3-way designs. The AR78LS would come in second and possibly the AR58s third -- perhaps tied with the AR-10/AR-11 -- and not counting the 303 in any of this as you mentioned the "original" designs and not the west-coast designs.

The next question might be: which tower speaker was better, the AR-9 or the AR-9LS (and LSi)?

--Tom Tyson

AR_A-New-Chapter-Bookshelf-Design_AR58s_AR48s_AR38s_AR28s.pdf

Share this post


Link to post
Share on other sites

Steve,

Well, this whole thing is "nitpicking" to the max, but it's fun to debate it!

Therefore… I think the biggest challenge to your argument has to do with the configuration of the AR58s: it's two or three inches taller on the long dimension than an AR-3a, for example, and it has a vertical midrange-tweeter alignment orientation. As you know, if you mount a "vertical" speaker horizontally -- as though on a bookshelf -- it will have truncated horizontal dispersion, and the speaker will not sound quite as good as it would mounted vertically. No big deal, but there is a difference. By pure happenstance, the AR-3/AR-3a midrange/tweeter orientation is staggered somewhat, such that it works about as well horizontally or vertically, but The AR58s was intended to be placed vertically, regardless of its mounting location, and there are precious-few "bookshelves" with 28-29-inch-high shelves. This type of speaker will almost always end up on the floor, on the top of some shelf or on stands -- probably not in a bookshelf in the classic way.

AR called it a "bookshelf" speaker, because it was the only thing with a 12-inch woofer configuration that could be labeled as such, and it was a sort of interim speaker design, but it's probably not a true bookshelf speaker in the true sense of the word. I believe that AR was afraid they might lose some traction with the folks who owned AR-3s, AR-3as, AR-10s and so forth if they didn't mention "bookshelf" in the description. Strangely, the AR-10 and AR-11 were nearly always shown on the speaker stands perched vertically, whereas those speakers sound somewhat better mounted horizontally!

I'm glad you make the argument about the piston diameter and the 1000 Hz crossover for a 12-inch woofer! This reinforces the argument about the crossover in the AR-3! Nonetheless, the lower the crossover for a heavy moving system, probably the better it is -- not just for off-axis response, but impulse response as well.

Since you can mount the AR98LS vertically in a cabinet just about as easily as an AR58s, I'd pick the 98 first. I seem to remember that it was "cleaner" sounding and much more "effortless" than most of the 3-way designs. The AR78LS would come in second and possibly the AR58s third -- perhaps tied with the AR-10/AR-11 -- and not counting the 303 in any of this as you mentioned the "original" designs and not the west-coast designs.

The next question might be: which tower speaker was better, the AR-9 or the AR-9LS (and LSi)?

--Tom Tyson

Every point you make is valid, of course. And in re-checking the specs, the 78LS was actually 3/8" shorter than the 58s, so if anything, the 78 was more of a 'bookshelf' speaker than the 58!

Of course, neither was a bookshelf speaker. I think I meant 'bookshelf' more as a size category than as a mounting configuration. 'Bookshelf' as in well under 30", as opposed to 'tower' with specific feet or a base and over 35" tall.

I guess my main point was this:

When we think of AR 3-way bookshelf speakers (regardless of how they were actually used most often) with separate dome mids and dome tweeters, we think of the 3, 3a, 11, and 10Pi. We never think of the 58s. To my knowledge, this is the very first thread ever about the 58s. Yet a very good case can be made that the 58s was, in fact, the best performer of all of them. It was a walnut cabinet. It was intro'd at a price of $325 ea. The 11 was discontinued at a price of $350 ea, having gone from $295 to 325 to 350. So not only was the 58s a terrific performer--arguably better than the 11--it was a better value as well. When the 78LS was tested by High Fidelity only a few scant years later (April 1984), it was $430 ea--a full $105 ea. more than the 58s, with only very marginal--if any--true audible improvement at all.

I'll stand by my original point: the AR-58s, with the AR-9 generation dome mid and tweeter, was an incredibly good speaker, a great value and perhaps the most overlooked AR speaker ever.

Until now, that is.

Steve F.

Share this post


Link to post
Share on other sites

By pure happenstance, the AR-3/AR-3a midrange/tweeter orientation is staggered somewhat, such that it works about as well horizontally or vertically

AR's literature of the classic period has photos of its speakers in both horizontal and vertical installations. Why was it "by pure happenstance" that the drivers were arranged so that the speakers would work well both ways? Couldn't they tell from the measurements and listening tests they conducted how different driver arrangements worked?

Share this post


Link to post
Share on other sites

I have never heard a pair of AR-58's, so I have absolutely no reason to disagree with the premise that it was very fine, nor that it is the most overlooked speaker of the AR-3 family tree. The term "best" seems a bit strong to me, however.

Some thoughts...

Simply because it used AR-9 drivers does not mean the 58 was inherently superior to its predecessors. Original AR-3a/11/10pi type drivers were altered for use in the AR-9. I would argue that the earlier, more compliant AR-3a/early 11/10pi woofer was a better choice for the single woofer, small cabinet 3-way. Further, Tom's comment about the low Q of the 58 series suggests many folks might prefer the "bloom" of the earlier designs sporting higher Q.

Imo, crossover design is more important than the use of modified drivers, or the vertical placement thereof.

I have heard AR 303's next to AR-11's, and preferred the 303's. The 303 was designed from the ground up to be a modern AR-3a. The AR-3 and AR-3a were designed to be at the top of the heap. The AR-58 used off the shelf parts, and was designed to fill a niche. I would, therefore, NOT assume that the 58 was "at least the equal of the 303". I would be very interested to see how the 58 would fare against the AR-3a Limited of the early 90's.

Of course, it is in the eye of the beholder, but if visual appeal is thrown into the mix, the AR-58 could never "best" the 3a cabinet in my jaded opinion. B)

Steve, Your points are well taken. I've been wanting to get my hands on early 58's for some time now, but they are very scarce. I will keep trying, and promise to report back when I do.

Did the AR-58 series have level controls?

Roy

Share this post


Link to post
Share on other sites

I have never heard a pair of AR-58's, so I have absolutely no reason to disagree with the premise that it was very fine, nor that it is the most overlooked speaker of the AR-3 family tree. The term "best" seems a bit strong to me, however.

Some thoughts...

Simply because it used AR-9 drivers does not mean the 58 was inherently superior to its predecessors. Original AR-3a/11/10pi type drivers were altered for use in the AR-9. I would argue that the earlier, more compliant AR-3a/early 11/10pi woofer was a better choice for the single woofer, small cabinet 3-way. Further, Tom's comment about the low Q of the 58 series suggests many folks might prefer the "bloom" of the earlier designs sporting higher Q.

Imo, crossover design is more important than the use of modified drivers, or the vertical placement thereof.

I have heard AR 303's next to AR-11's, and preferred the 303's. The 303 was designed from the ground up to be a modern AR-3a. The AR-3 and AR-3a were designed to be at the top of the heap. The AR-58 used off the shelf parts, and was designed to fill a niche. I would, therefore, NOT assume that the 58 was "at least the equal of the 303". I would be very interested to see how the 58 would fare against the AR-3a Limited of the early 90's.

Of course, it is in the eye of the beholder, but if visual appeal is thrown into the mix, the AR-58 could never "best" the 3a cabinet in my jaded opinion. B)

Steve, Your points are well taken. I've been wanting to get my hands on early 58's for some time now, but they are very scarce. I will keep trying, and promise to report back when I do.

Did the AR-58 series have level controls?

Roy

Good points, Roy. An interesting discussion, to be sure.

Without having seen the actual schematics, I'd be willing to venture a guess and say that the 58s's x-o was exactly the same as the 91's, given both speakers' use of the exact same drivers and having the same x-o frequencies. It's my impression that the 91 represented AR's very best efforts at a 12" 3-way crossover in 1979, so I have no reason to think that the 58s's x-o would be anything less. "Off the shelf" does not necessarily mean "inferior."

Regarding the Q, I'd say two things:

1. I think by 1980 there was a somewhat greater awareness of how the relative 'tightness' of bass appealed to listeners, and technically-adept manufacturers tended to shade their products towards the perceived tastes of their market, one way or the other.

2. I have a hunch that Q was a somewhat less understood and less-intentionally manipulated variable in the 3a's day (designed in 1966, intro'd in 1967) than in the 58s's day. Those 12 years were a virtual lifetime of speaker knowledge and design experience. The original 3a lit never even mentioned Q as a specification, yet by 1978 AR was routinely listing Q in all its published specs. It was simply a more understood concept by then.

The 3a was a classic-looking speaker, with its iconic picture-frame molding, linen grille cloth and that wonderful brass logo w/red lettering. I'm just pointing out that the 58s had a real walnut veneer cabinet, in stark contrast to the sad-sack vinyl 58B, BX, and BXi that followed.

I actually only heard the 58s briefly, on one occasion. It sounded like a 91. I owned 11's and 91's at the same time, and A-B'd them for months on end in my living room. They were very close to each other. The 91 may have been ever-so-slightly smoother and warmer through the lower mid. 58s's would be similar, I'm quite sure.

Your comment about the 303 vs. the 11 is interesting. In what ways did you prefer the 303?

Steve F.

Share this post


Link to post
Share on other sites

AR's literature of the classic period has photos of its speakers in both horizontal and vertical installations. Why was it "by pure happenstance" that the drivers were arranged so that the speakers would work well both ways? Couldn't they tell from the measurements and listening tests they conducted how different driver arrangements worked?

Perhaps I should have said it was by "design" that the staggered midrange/tweeter arrangement was used on the AR-3 and AR-3a (and later versions of the AR-5). It was by happenstance that the speakers could be mounted equally well -- horizontally or vertically. If you examine the speaker baffle on the AR-1/AR-3/AR-3a, there is simply insufficient room to mount the midrange and tweeter in a vertical or horizontal alignment.

With the exception of steep-crossover-slope networks, such as the 24 dB four-pole designs, there is usually quite a bit of interaction between midrange and tweeter drivers, particularly in frequencies of the crossover region, as both drivers will produce output over a shared frequency range. This is particularly noticeable in the near field as one moves around off-axis in front of a speaker (especially noticeable with midrange and tweeter mounted side-by-side), and it results in interference effects, a sort of "lobing." This is significant if you listen in the near field, and imaging is important to you. Again, it is *not* particularly audible in the reverberant field any more than "flat" on-axis 1-meter system anechoic frequency response is audible in the far field.

"…Couldn't they tell from the measurement and listening test they conducted how different driver arrangements worked…?" AR understood this phenomenon, for sure -- as it was well understood in the field of acoustics -- but it is not audible in the reverberant field, so it was not significant enough at that time to be considered important. AR during the classic period was obsessed with flat power response in the reverberant listening field, so "imaging" and near-field frequency response was of secondary importance.

This design philosophy changed in 1978 with the advent of the AR9 and subsequent tower designs and vertical-tweeter arrangement. By placing the drivers in a vertical alignment, AR was able to get most of these interference effects -- lobing -- into a vertical plane rather than in a horizontal plane. Therefore, in the near field, these affects are less audible on the AR9 and subsequent tower speakers than with the old conventional side-by-side arrangement, particularly the AR-10/AR-11 when mounted vertically. The AR-3/AR-3a, however, represented a sort of compromise with their staggered arrangement, but interference effects -- in the near field -- are very much present. Again, it's not what you hear when you get the sum of all the reflected surfaces back in the reverberant field. This is the reason that most of the older AR's -- with their sometimes jagged near-field frequency response -- always sound natural and life-like when heard in the reverberant field in a normal listening room.

--Tom Tyson

Share this post


Link to post
Share on other sites

Although I basically agree with Tom here, I’d like to look at things from a slightly different perspective.

It’s easy to fall into the trap of thinking all loudspeaker/acoustic knowledge/experience has existed all along, and to the same degree. So it’s very natural to simply pose the question “Couldn’t AR just tell from their measurements?” and equally natural to simply answer, “AR understood this phenomenon but had a different design philosophy at that time….”

I would offer for consideration the distinct possibility that AR did not fully appreciate the significance of driver interference and cabinet diffraction in the early-to-mid ‘60’s. (That was when the 2ax-3a-5 were designed. They may have appeared on the market in the late 60’s, but they were designed in the mid ‘60’s.)

I would say that the 3a’s drivers were mounted as they were—diagonally staggered—as much for mechanical/fitting convenience as any other reason. That the interference/radiation pattern was nearly the same horizontally as vertically was a coincidence, not a design goal. If it were a design goal, then the 2ax would have been that way also. It wasn’t.

AR continued to demonstrate that they didn’t consider it to be of significance right through the 10 Pi-11-12 ADD models. Those had their mids and tweets mounted all “wrong”—side by side, a la the 2ax—even though their cabinets were cosmetically intended, unambiguously, to be vertical, with that walnut trim strip at the bottom of the 11-12 (the door at the top of the 10 Pi) going side-to-side.

I submit that the 9 and the Vertical Series was not as much a simple “change of heart” by AR as it was a new awareness of that vertical mounting/interference design aspect—an awareness that simply didn’t exist to anywhere near that degree (if at all) in the mid-‘60’s. With the Verticals, AR finally realized that, yes, you could have wide dispersion, excellent far-field power response AND clear, almost interference-free near-field response, all at the same time.

That knowledge didn’t exist in 1965 the way it did in 1977. The collective understanding in any field advances with time, and as observers after the fact, we all tend to compress that in retrospect.

Steve F.

Share this post


Link to post
Share on other sites

could be very true, that the knowledge wasn't there, and it might not have been there just because at the time imaging wasn't a big deal or even a consideration; soundstage and spaciousness were, which the driver placement of the 2 & 3 series, from what I've read at least, may have had some advantages...heck, through the mid 60's (when the classic AR's were designed), the bulk of popular recordings were still primarily mixed in mono..reading interviews with George Martin & Paul McCartney, they spent a lot more time mixing the mono mix of the beatles early albums, and the stereo mix was a real afterthought (obvious in listening to a lot of the stereo mix of early recordings, where you'll hear stuff like paul's voice out of the right speaker, and his bass out of the left ;)

Share this post


Link to post
Share on other sites

Your comment about the 303 vs. the 11 is interesting. In what ways did you prefer the 303?

Steve F.

Well, it was only a brief comparison, and by no means scientific. The unanimous impression among 3 of us was that the 303 had a smoother, more open, response through the midrange, especially noticeable with vocals. We preferred the 303's deep bass response slightly as well. The 303 woofer has a rubber surround, and is a close cousin to the wonderful NHT 1259 12 inch woofer. It should be noted that its woofer has the advantage of operating in a slightly larger cabinet. The AR-11 did have the advantage of level controls, and the 303 had none...which brings me back to the question: Did the AR-58 series have level controls?

Perhaps the 58 could match the smoother midrange of the 303, but I would be very surprised if it could match the bass response. I have mixed, matched, and measured all of the various versions of the AR 12 inch woofer, and have never found the AR-9, nor the later Tonegen woofer, to be the equal of the earlier iterations with regard to bass response (in a 3a size cabinet). The AR-9 and Tonegen woofer, however, both reach further into the midrange, which may contribute to an improvement in that regard. Whether by design or not, Q of .5 or .6 is quite low, and in my experience will not provide the subjective reverberant bass response associated with earlier AR speakers. In the AR-9 era, AR had much larger speakers with greater bass response, so could it be that this was not a priority for the AR-58?

It is interesting that the mirror-imaged AR 303 had the tweeter and midrange vertically placed, but they were offset relative to the woofer. There is a school of thought that believes this arrangement is preferable to having the high range drivers equidistant from the baffle edges.

Roy

Share this post


Link to post
Share on other sites

Hi Steve

The reason for AR-58s crossover point set at 700Hz was due to the limitation of the AR-58s midrange which has Fs of 604Hz in comparison to the AR-3a/AR-11 midrange that has Fs of 485Hz so you can push it down to 525Hz crossover point still be OK.

AR-3a Improved and AR-11 sound much closer to the original AR-3a with foam surround woofer than AR-58s. So don't rush out to find a pair of AR-58s and hope for it to out do the AR-3a sound. You will be disappointed! Personally I love the AR-58s for it represents the end era of the AR-3a family. Its vertical driver arrangement made more sense to me than the classic AR-3a or AR-10Pie/AR-11 look. I never listen to my AR speakers placed horizontally and certainly not shove them into bookshelf by any mean. I also like the 58Bx and 58Bxi because these two models sound very good with Japanese receivers from Marantz, Sansui and Pioneer. 100w/ch of Japanese receiver power will made these speakers sound wonderful and for how much they cost as AR speakers, I don't mind the walnut vinyl covered MDF cabinets! The 12" woofer AR bass was there up close... These are my personal feeling and listening experience and by the way I am a real AR fanatic so I may be bias without realizing it! To me AR speakers carry a very unique sound quality of its own. Once you learn to love it then you will have a difficult time to pull yourself off from that addiction. I enjoy all of the AR speakers I owned no matter they are the most expensive AR Limited 3 speaker system or the tiny little AR-18s that sound wonderful with 65w/ch Sony receiver...

AR-58s has different crossover from AR-91. AR-58s crossover has no switches while AR-91 has switches for mid and high range drivers.

Minh Luong

post-101112-0-85741800-1294441676_thumb.

Share this post


Link to post
Share on other sites

The reason for AR-58s crossover point set at 700Hz was due to the limitation of the AR-58s midrange which has Fs of 604Hz in comparison to the AR-3a/AR-11 midrange that has Fs of 485Hz so you can push it down to 525Hz crossover point still be OK.

Minh Luong

Hi Minh,

Good info! It dovetails with my observation that the unfiltered response of the later AR 12 inch woofer had more extension into the midrange. This would make the higher crossover point of the AR-58 a bit easier to achieve.

It would certainly be fun to line them all up next to each other for us to listen and discuss. :)

Roy

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now