Jump to content

AR3 vs. AR3A


stan461

Recommended Posts

Probably been done before but......

What are the differences between an AR3 and an AR3a?

What I think is the same:

Cabinet (Same size?).

Midrange.

Woofer. If its the cloth surround AlNiCo type.

Known Differences:

The tweeter.

The crossover.

On the subjective side:

How would you describe the difference in how they sound?

Link to comment
Share on other sites

Guest bocoogto

The tweeter, midrange, and crossover are all different between the two models. The cabinet and woofer are the same for both.

The difference in the sound of an AR-3a vs. the AR-3 is very noticeable. The AR-3a sounds very much more "forward" in the mid and treble areas. The AR-3 sounds like an AR-3a with a blanket thrown over the grill cloth.

They are both very good speaker systems, but the AR-3a, in my opinion, is much more accurate than the AR-3. Others may have different opinions, but I believe my description above describes the differences.

Note that the mid and tweeters on later systems, including the AR-LST, AR9, were the same as the AR-3a.

Link to comment
Share on other sites

We've discussed this at some length before. The AR3a introduced new ferro fluid cooled tweeter and midrange drivers which had greater power handling capacity than AR3s. The crossover frequencies were changed as well. Especially notable was the decrease in the crossover between the woofer and dome midrange from about 1Khz to around 525 or was it 575 hz. This change altered the sound by making it brighter. Was this an improvement or was it pandering to market preferences and a step backwards in accuracy? It's hard to accept that in some ways the new improved version might have been a step backwards in at least some regards and not all steps forwards. This runs counter to our usual notion of ever onward and upward. Here's why I suggest that. AR3 acquitted itself admirably in many public live versus recorded demonstrations revealing itself to be a remarkably accurate speaker already. Many commercial recording studios made their recordings using the Altec A7 Voice of the Theater loudspeakers as monitors. Because these speakers had an inherent harsh high end, many recordings may have been made with deliberately rolled off highs making them sound dull when played through AR3s. Speakers with brighter sound like KLH6 were very popular doing quite well in A/B comparisons in stores. Even Tannoy modified its design in the mid 1960s making their speakers brighter to attract a larger segment of the US market (as told to me by a Tannoy salesman at a trade show.) Both AR3 and AR3a were examples of the difficulty of mating a 1 1/2" dome midrange with a 12" woofer. Even today, a midrange driver of this type rarely has a usable response below 750hz while the woofer has a somewhat irregular response at the top of its range. The problem wasn't effectively resolved until AR9 which introduced an 8" upper bass/lower mid driver handling the range between 200 hz and 1.5 khz. This freed up the other drivers to do what they do best. It's my hunch that skillfully using a graphic equalizer, either speaker could be made to sound very much like the other, but that's just a hunch.

Link to comment
Share on other sites

The AR-3 was introduced in 1958 and was the first speaker to use domed midrange and HF drivers. The smooth frequency response and wide dispersion, coupled with the 12-inch acoustic suspension woofer, gave the AR-3 a margin of performance ascendancy over its competitive rivals that has never been equaled in the audio industry since.

In Dec 1967, the 3a was introduced, like the AR-3 to great critical acclaim. The previous entries above detail some of the technical differences. However, there are some factual corrections needed:

- The 3a did not use ferro-fluid cooled drivers. It was not until the 10Pi/11 in 1975 that the 3/4" dome had ferro-fluid cooling.

- The 1 1/2" midrange of the 3a/LST was not the same as the 1 1/2" midrange of the 9. Likewise, the 3/4" tweeter of the 3a/LST was not the same as the 3/4" tweeter of the 9.

Steve F.

Link to comment
Share on other sites

Hi there

I guess I'll jump in and say a few words now.

The AR-3 midrange driver used the 12" woofer magnet system and the tweeter used the 10" woofer magnet system.

They both used aluminum wire voice coils without formers.

After the AR-3A came out, the main complaint, if you want to call it that, was midrange honkiness because of the higher crossover from the woofer to the midrange.

This was offset by lowering the crossover to 575hz on the 3A.

The AR-3A had much improved dispersion and power handling over the 3.

I believe the AR-3 had a more extended high end than the 3A.

The AR literature for several years after release of the AR-3A showed the original 12" woofer with the rings and the clothe surround, even after they changed to the new cone and foam surround.

I imagine this was to remove any visible issues with stores with old stock similar to not changing the 2AX version when they changed to the 3/4" tweeter and changed the woofer.

If anyone still owns AR-3's they obviously should not be driven hard because the aluminum voice coil wire does break.

Somewhere I read that AR pioneered the dome tweeter design and all of the domes today owe credit to AR's foresight.

You can't interchange AR-3 and 3A mids and tweeters because of their different crossover points.

To substitute an older cast aluminum framed clothe surround woofer with the later foam surround may or may not be audible.

I certainly would like to read more on this issue if anyone is so inclined please.

The original prototype woofer that Ed Vilchur used was, I don't remember what brand, he cut away some of the spider to allow it to flex better.

His wife was a draughtsperson, he suggested she cut out from a bed sheet a clothe surround for the woofer.

This showed that Ed Vilchur was a true gentleman.

Also if he had cut it out, you can imagine his wife rolling back the sheets that night and seeing a 12" hole in her nice sheets.

You can bet, he would have had to have gone out to buy a new pillow to sleep in the doghouse for a while.

Have a great day. HIFI what a nice hobby.

Vern

>The AR-3 was introduced in 1958 and was the first speaker to

>use domed midrange and HF drivers. The smooth frequency

>response and wide dispersion, coupled with the 12-inch

>acoustic suspension woofer, gave the AR-3 a margin of

>performance ascendancy over its competitive rivals that has

>never been equaled in the audio industry since.

>

>In Dec 1967, the 3a was introduced, like the AR-3 to great

>critical acclaim. The previous entries above detail some of

>the technical differences. However, there are some factual

>corrections needed:

>

>- The 3a did not use ferro-fluid cooled drivers. It was not

>until the 10Pi/11 in 1975 that the 3/4" dome had ferro-fluid

>cooling.

>

>- The 1 1/2" midrange of the 3a/LST was not the same as the 1

>1/2" midrange of the 9. Likewise, the 3/4" tweeter of the

>3a/LST was not the same as the 3/4" tweeter of the 9.

>

>Steve F.

Link to comment
Share on other sites

Hi there again

I guess I'll jump in and say a few words now and correct a point.

>After the AR-3A came out, the main complaint, if you want to

>call it that, was midrange honkiness because of the higher

>crossover from the woofer to the midrange.

The honkiness was reported in the AR-3 not the 3A.

>

>This was offset by lowering the crossover to 575hz on the 3A.

>

>The AR-3A had much improved dispersion and power handling over

>the 3.

>

>I believe the AR-3 had a more extended high end than the 3A.

Good day

Vern

Link to comment
Share on other sites

>- The 3a did not use ferro-fluid cooled drivers. It was not until the 10Pi/11 in 1975 that the 3/4" dome had ferro-fluid cooling.<

Steve - this is sorta "my era" and I have some doubts about what I think I know. Tom has on occasion likewise expressed a doubt.

When the 10pi and 11 were first introduced and incorporated the "white" tweeter, the literature makes no mention of ferrofluid. It wasn't until later that the tweeter went to the black fabric that ferrofluid was mentioned. I'm given to believe this would have been early in the 10/11's production.

Obviously the lack of mention does not mean that the white tweeters did not have ferrofluid.

Do you have any insight as to whether or not the "original" tweeters used in an 11 and 10pi originally had ferrofluid and an approximate date that the tweeter would have been switched from the "white" to the black?

I haven't had the opportunity to disassemble a non-working "white" tweeter to see for myself.

Bret

Link to comment
Share on other sites

Guest hilltroll67

"Both AR3 and AR3a were examples of the difficulty of mating a 1 1/2" dome midrange with a 12" woofer. Even today, a midrange driver of this type rarely has a usable response below 750hz while the woofer has a somewhat irregular response at the top of its range. The problem wasn't effectively resolved until AR9 which introduced an 8" upper bass/lower mid driver handling the range between 200 hz and 1.5 khz. This freed up the other drivers to do what they do best."

If there was/is a problem here, it was 'effectively resolved' in the AR5, long before the AR9.

Bob

Link to comment
Share on other sites

AR 5 had a smaller 10" woofer which did not have as much low end reach or power as the AR 12" woofer. Even so, you can see that the 10" counterpart of AR9, the AR90 took a similar approach. It is true that the 8" intermediary "mid bass" driver allowed for greater power handling in the 1.5 to 7 khz range by the 1 1/2" dome due to its not having to extend down to meet the upper end of the woofer's range and allowed the woofers to restrict their power to the lowest frequencies. But most important was that it allowed for a smoother overall response. Even the 10" woofer could be bettered at its upper end by using a smaller lighter driver specifically intended for that range. That's how I see it.

Link to comment
Share on other sites

>Even the 10" woofer could be bettered at its upper end by using a smaller lighter driver specifically intended for that range.<

Do you suppose that's why most modern designs, even fairly large ones, that claim "audiophile quality" status don't even have a woofer anymore?

A 6.5" woofer that reaches to 40Hz that also handles everything up-to the tweeter's range??

I'll believe physics can accomplish wonderful things with huge horns, etc, but a 6.5" driver that goes to 45Hz in a cabinet that sits attractively beside the TV?

Maybe Ken's array.

Speaking of which, I REALLY want to get a Heil AMT and stick it atop Ken's new woofer and see what we get.

Bret

Link to comment
Share on other sites

"Make it and they will pay!"

I'll be the first to admit that I don't understand the current high end audio market. It seems that as time goes on, you pay more and more and get less and less. Pretty soon you'll just go into a high end audio salon, sign over your house, and walk out with an empty box with reviews in it, maybe a few screws, connectors and other miscellaneous hardware. I thought people were nuts to buy a pair of two way 8" speaker systems for $1500 to $2000 ten or fifteen years ago. Now they're paying double that for a pair with 6 inch woofers. Maybe those Advent/2s sitting in my basement which need new foam and I paid $20 for at a garage sale some years back weren't such a bad deal after all.

Last weekend, for the first time in I can't remember how long, I went to a high end audio store because it had just opened in my neighborhood and I was curious. I heard a very expensive system based around the Martin Logan Summit electrostatic/hybrid loudspeakers ($10,000 a pair) a Krell 400 wpc amplifier which you'd need a fork lift to move, a monster McIntosh preamp and a Big Mac disc player. Total for the system must have been about $25K or more. I listened to a couple of their audiophile discs which I usually regard as crummy (one was a Telarc recording of Pictures at an Exhibition) and there was another piano/violin recording which I didn't recognize.) Having been tweaking my equipment for years I can practically tell you where every frequency response rise and dip is just by listening and although I'll go back in a few days with my own recordings, if what I heard holds true, there were at least five major deficiencies I picked out. I am not at all surprised. I think most of the people who design this stuff do not know what live unamplified music sounds like.

Link to comment
Share on other sites

  • 2 weeks later...

A great deal has been said about the differences in the AR-3 and AR-3a; and another thread mentioned the power needed to play “live” levels, or high playback levels of piano music through AR speakers, all of which are around 0.5% efficient. This message will address this well-worn thread once again.

In the mid 1960s, Robert Berkovitz, working for Mattes Corporation at the time, demonstrated in a study what was required of an amplifier if it were to reproduce “live” levels of a Steinway Grand (not a “D”) piano. He documented the power levels, as well as compared the waveforms from the piano and the loudspeaker, with oscilloscope wave-form traces. I am attaching a few images from that article with the following summary:

(1) Although not specifically mentioned, the loudspeaker in question was an AR-3;

(2) The power levels for peaks was well-over 100 watts rms;

(3) The waveform oscilloscope-trace comparisons between the output of the piano and the AR-3 were considered to be remarkably close.

http://www.classicspeakerpages.net/dc/user_files/569.jpg

http://www.classicspeakerpages.net/dc/user_files/570.jpg

This article demonstrated the AR-3’s remarkable accuracy and life-like sound reproduction in replicating live acoustic instruments. It also demonstrated that to reproduce such levels, an amplifier with at least 100 watts/channel rms would be necessary.

--Tom Tyson

Link to comment
Share on other sites

I think it's a bit like comparing apples and oranges. You could compare the total maximum instantaneous energy flux output(acoustic power) the average total flux, the total averange and instaneous flux as a function of frequency, the on axis energy distribution as a function of frequency of the speaker compared to the corresponding distribution at a given arbitrary point for the piano, and their average and maximum SPLs. IMO, the principal big difference between them is the spatial radiating pattern of the piano being very unlike the radiating pattern of the speaker. Generally, I think a piano like many instruments radiates it energy far more omnidirectionally at all frequencies than most loudspeakers including AR3. Given these differences, the fact that the speaker can be made to sound so similar to the piano at all is remarkable. BTW, I personally notice this difference with my own setup and comparison. A piano fills a small room engulfing you with sound the way few speakers can.

Link to comment
Share on other sites

Hi Tom

Thank you very much for this very valuable data.

It may be old news, it just re-affirms the capabilities of the AR-3 in a A - B comparison.

As much as some people might think the AR-3 was a clunker and ugly in appearance.

I feel the AR-3 revolutionized the speaker industry and gave other manufacturers a different direction to go in.

I love reading of old time comparisons.

It goes to show that Ed Vilchur was not just a smart man but a genius ahead of his time.

I wonder who our next genius will be and when he or she will be recognized.

Recently I found or actually re-found the Crown International web site.

It certainly has a wealth of discontinued documentation and also other reports.

For example.

I've quoted this without permission.

How much amplifier power do I need?

4 ohm speaker rated at 100 watts continuous and you are playing light dance music, the amplifier should be 1.6 x 100 w continuos per channel.

To handle heavy/metal grunge 2.5 x 100 w or 250watts per channel.

Typical loudspeaker sensitivity is 85dB SPL/W/m for home stereo.

The recommended power allows for signal peaks of 10 dB for folk, jazz and pop music. Actually the peaks might be as high as 25 dB, but we're allowing for some inaudible short-term clipping.

Nere field monitoring: 25 W for 85 dB SPL average (with 15 dB peaks), 250 W for 95 dB SPL average with 15 dB peaks.

Check out for more interesting information.

www.crownaudio.com

Have a great day and thank you again Tom for your very worthwhile information.

Knowledge is neverending.

>A great deal has been said about the differences in the AR-3

>and AR-3a; and another thread mentioned the power needed to

>play “live” levels, or high playback levels of piano music

>through AR speakers, all of which are around 0.5% efficient.

>This message will address this well-worn thread once again.

>

>In the mid 1960s, Robert Berkovitz, working for Mattes

>Corporation at the time, demonstrated in a study what was

>required of an amplifier if it were to reproduce “live” levels

>of a Steinway Grand (not a “D”) piano. He documented the

>power levels, as well as compared the waveforms from the piano

>and the loudspeaker, with oscilloscope wave-form traces. I am

>attaching a few images from that article with the following

>summary:

>

>(1) Although not specifically mentioned, the loudspeaker in

>question was an AR-3;

>(2) The power levels for peaks was well-over 100 watts rms;

>(3) The waveform oscilloscope-trace comparisons between the

>output of the piano and the AR-3 were considered to be

>remarkably close.

>

>

>

>

>This article demonstrated the AR-3’s remarkable accuracy and

>life-like sound reproduction in replicating live acoustic

>instruments. It also demonstrated that to reproduce such

>levels, an amplifier with at least 100 watts/channel rms would

>be necessary.

>

>--Tom Tyson

>

Link to comment
Share on other sites

>Hi again Tom

>

>Thank you very much for this very valuable data.

>

>It may be old news, it just re-affirms the capabilities of the

>AR-3 in a A - B comparison.

>

>As much as some people might think the AR-3 was a clunker and

>ugly in appearance.

>

>I feel the AR-3 revolutionized the speaker industry and gave

>other manufacturers a different direction to go in.

>

>I love reading of old time comparisons.

>

>It goes to show that Ed Vilchur was not just a smart man but a

>genius ahead of his time.

>

>I wonder who our next genius will be and when he or she will

>be recognized.

>

>Recently I found or actually re-found the Crown International

>web site.

>

>It certainly has a wealth of discontinued documentation and

>also other reports.

>

>For example.

>

>I've quoted this without permission.

>

>How much amplifier power do I need?

>

>4 ohm speaker rated at 100 watts continuous and you are

>playing light dance music, the amplifier should be 1.6 x 100 w

>continuos per channel.

>

>To handle heavy/metal grunge 2.5 x 100 w or 250watts per

>channel.

>

>Typical loudspeaker sensitivity is 85dB SPL/W/m for home

>stereo.

>

>The recommended power allows for signal peaks of 10 dB for

>folk, jazz and pop music. Actually the peaks might be as high

>as 25 dB, but we're allowing for some inaudible short-term

>clipping.

>

>Nere field monitoring: 25 W for 85 dB SPL average (with 15 dB

>peaks), 250 W for 95 dB SPL average with 15 dB peaks.

>

>Check out for more interesting information.

>

>www.crownaudio.com

>

>Have a great day and thank you again Tom for your very

>worthwhile information.

>

>Knowledge is neverending.

>

>

>

>>A great deal has been said about the differences in the AR-3

>>and AR-3a; and another thread mentioned the power needed to

>>play “live” levels, or high playback levels of piano music

>>through AR speakers, all of which are around 0.5% efficient.

>

>>This message will address this well-worn thread once again.

>>

>>In the mid 1960s, Robert Berkovitz, working for Mattes

>>Corporation at the time, demonstrated in a study what was

>>required of an amplifier if it were to reproduce “live”

>levels

>>of a Steinway Grand (not a “D”) piano. He documented the

>>power levels, as well as compared the waveforms from the

>piano

>>and the loudspeaker, with oscilloscope wave-form traces. I

>am

>>attaching a few images from that article with the following

>>summary:

>>

>>(1) Although not specifically mentioned, the loudspeaker in

>>question was an AR-3;

>>(2) The power levels for peaks was well-over 100 watts rms;

>>(3) The waveform oscilloscope-trace comparisons between the

>>output of the piano and the AR-3 were considered to be

>>remarkably close.

>>

>

>>

>

>>

>>

>>This article demonstrated the AR-3’s remarkable accuracy and

>>life-like sound reproduction in replicating live acoustic

>>instruments. It also demonstrated that to reproduce such

>>levels, an amplifier with at least 100 watts/channel rms

>would

>>be necessary.

>>

>>--Tom Tyson

>>

As always I think of something after I save the file.

Actually if I rattle too long I guess the program times out and everything gets lost.

To continue on briefly.

Crown and 4 ohm speaker still.

Home stereo; 150 W for 85 dB SPL average ( with 15 dB peaks ), 1500 Watts for 95 dB SPL average (with 15 dB peaks).

You read correctly, "1500 watts per channel" continuous, "( DC/or 20 - 20,000 Hz @ a specified almost non-existant distortion. "My words". )"

I think that we may need a little more oomph for our 3's, 3A's and LST's than we have been lead to believe.

Your comments please.

Thank again Tom for getting this topic going again.

Have a great day.

Link to comment
Share on other sites

>Tom,

>

>Thank you. Very interesting post/info.

>

>BTW. Do you have a preference (AR3 or AR3A)? I have the AR3a's

>and am considering purchasing a pair of AR3's when the

>opportunity arises. That's what prompted me to start this

>thread.

>

I really don't have a preference. I have such great respect for both speakers that I enjoy listening to either system. Clearly, the AR-3a was a improvement over the AR-3 in several respects, such as off-axis response and power-handling. The AR-3, however, received higher critical acclaim when it was introduced, and it set the standard of performance, but it came out (technically) nine years prior to the AR-3a. It has become a great classic in every since of the word. The AR-3a certainly improved upon the same formula that made the AR-3 such a good speaker, but the refinements were quite subtle. For example, there are no pronounced differences in the measured response of the two speakers -- there are more similarities than differences. I heard a "new" AR-3 and a "new" AR-3a compared in the AR Music Room in 1967, and the "new" speakers sounded much more alike than have been described by some people on these pages.

The biggest problem I see is that it is difficult to find AR-3s today, after so many years, that perform as they did when new. This is also applies to the AR-3a's 3/4-inch tweeter, as well. The glues become brittle; the foam-rubber damping rings harden; the suspensions on the tweeters change through time so that the performance of this speaker has likely deteriorated slightly through time. When AR was fabricating these tweeters, the rejection-rate on drivers was very high; today I doubt that any of them would pass with flying colors, but perhaps that is to be expected with anything that is thirty- or forty-years old.

I wanted to share the Berkovitz study on the piano-vs.-AR-3 because it showed that the AR-3 was (once again) capable of very accurate replication of the signal source. I have always felt this to be the real definition of "high fidelity." It is true that it could not duplicate the radpat of a piano -- no speaker could -- but the AR-3 was clearly able to accurately reproduce the electrical signal it received, as graphically shown in the oscilloscope images. This was especially noteworthy in that the test was originally performed to determine the output requirements of an amplifier to reproduce "live" levels of a piano using a typical, low-efficiency speaker (without regard to the speaker model), and the accuracy results were strictly coincidental to the power-requirement studies.

--Tom Tyson

Link to comment
Share on other sites

>I think it's a bit like comparing apples and oranges. You

>could compare the total maximum instantaneous energy flux

>output(acoustic power) the average total flux, the total

>averange and instaneous flux as a function of frequency, the

>on axis energy distribution as a function of frequency of the

>speaker compared to the corresponding distribution at a given

>arbitrary point for the piano, and their average and maximum

>SPLs. IMO, the principal big difference between them is the

>spatial radiating pattern of the piano being very unlike the

>radiating pattern of the speaker. Generally, I think a piano

>like many instruments radiates it energy far more

>omnidirectionally at all frequencies than most loudspeakers

>including AR3. Given these differences, the fact that the

>speaker can be made to sound so similar to the piano at all is

>remarkable. BTW, I personally notice this difference with my

>own setup and comparison. A piano fills a small room

>engulfing you with sound the way few speakers can.

I’m not completely with you on the “total maximum instantaneous energy flux output.” What is that all about? In any event, I don’t think the piano, or any other acoustic instrument, radiates energy evenly at all frequencies. Sound is definitely omnidirectional in the lower bass frequencies; but at higher frequencies the radiating surfaces do not have a completely open, unobstructed path to radiate that energy. Sound radiates out of the bottom, diffracts off the floor, and out of the top of the piano -- deflection angles depending to some degree on whether the top is closed, partially or completely open. This characteristic was also true of the violins and cellos in the Fine Arts Quartet: some sound was blocked by the musicians themselves, and so forth, thus preventing complete omnidirectional sound radiation, yet the acoustic-power response of the AR-3s was able to replicate the radiation pattern of the instruments pretty closely. Obviously the key to accurate reproduction of live musical instruments is uniform frequency response both on- and off-axis, and low distortion.

http://www.classicspeakerpages.net/dc/user_files/571.jpg

http://www.classicspeakerpages.net/dc/user_files/572.jpg

I have lost the caption to these quasi-scientific images; they were sent to me by David Moran or Alvin Foster or someone in The Boston Audio Society Speaker, and were pertinent to a audio demonstration in someone's home in the Boston area. These were given to describe the approximate radiation pattern of a piano of that member's piano. Steve F might remember the details, but the general pattern is probably close to the piano's output.

--Tom Tyson

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...