AR3a - Measurements

HF Response - Filter & Attenuation:

Indeed, the same "Response shaping" approach, apparently:

attachment.php

Attenuation per the mid driver study, same color key. Again, as previously, L-Pad || 25 Ohms at mid setting, Blue, overlays the stock pot @ 4 Ohms, Cyan. Response @ 2 Ohms also shown, Black:

attachment.php

Impedance:

attachment.php

Note: Deduct 0.5 Ohms from any specific impedance measurement posted here for use in design or calculations. CLIO employs coax for these measurements, and the long mic cable I'm using for these itself has a finite impedance of this value at these frequencies and measurement currents.... :yes:
 

Attachments

  • HF Response.jpg
    HF Response.jpg
    64.2 KB · Views: 284
  • HF Attenuation.jpg
    HF Attenuation.jpg
    69.3 KB · Views: 275
  • HF Impedance.jpg
    HF Impedance.jpg
    59.9 KB · Views: 264
Last edited:
Full-Range Axial Frequency Response

Much as others have determined over the years, HF at max and Mid at mid L-pad attenuation gives maximally flat on medial axis. Bass extends below 30 Hz, substantially as advertised:

attachment.php

Measures best from mid-driver vertical axis and above. ArcTan 6/42 = 8.13°, the max I measured:

attachment.php

Nearfield bass measurement reflects response with 2-Pi, one boundary reinforcement; it's more "neutral" in freespace.

The 1.8 kHz notch remains an issue, as does a broad 3+ dB peak between 500 Hz and 1.2 kHz.... :yes:
 

Attachments

  • Full Range #2 SIN.jpg
    Full Range #2 SIN.jpg
    66.7 KB · Views: 277
  • Full Range #2 MLS.jpg
    Full Range #2 MLS.jpg
    75 KB · Views: 276
Last edited:
I have been reading posts in a number of sites off and on for a couple of months because I have some speakers I need to do a bit of refurbishing on. A pair of Heath Kit AR3as and a pair of AR4xs. The 3as have been re-foamed but I note that neither tweeter seems to be doing anything (or my ears are simply not up to it). I will measure coils and see what is wrong. They have never really spent any time with real loud music, but they may have been run by an underpowered amp for a while (never knew that was an issue). I have no idea what condition things are in the 4s. I am the original owner of all of them.

I was thinking of at least cleaning pots and putting in new caps. If the 3a HF speakers are shot that will mean replacing them; probably not with originals.

But from my reading, and this thread particularly, I have developed a real headache over just what I should really do.

I know this is sort of heresy, but has anyone taken 3as and put in a new HF speaker, and redid the filter network with modern tools (and no pots) to see if the speakers can be made "better". Given the assorted graphs, it would seem that things could be made better with these changes.

I always bought the "accuracy" pitch for these speakers, and I always felt they were decent. I am going to use all of them in a home theater (without a sub woofer for now), so if I can get a better result I would be interested in trying it.
 
I know this is sort of heresy, but has anyone taken 3as and put in a new HF speaker, and redid the filter network with modern tools (and no pots) to see if the speakers can be made "better". Given the assorted graphs, it would seem that things could be made better with these changes.
We are just now discussing that over at CSP.

There are known options; I'm trying to encourage others to post their findings and recommendations so that they may be verified.... :yes:
 
They have never really spent any time with real loud music, but they may have been run by an underpowered amp for a while (never knew that was an issue).

I still need to catch up on this thread as I've been gone for a little while but I wanted to point out that driving speakers with an "underpowered" amp is not an issue. The issue is driving speakers with an underpowered amp that you are pushing into clipping. If you were reasonable with your volume and never heard any audible distortion then you did nothing wrong and that should not have caused a failure in your speakers.

Ray
 
We are just now discussing that over at CSP.

There are known options; I'm trying to encourage others to post their findings and recommendations so that they may be verified.... :yes:


Are there specific threads you are referring to? I looked but have not found what I would think of as an upgrade post.
 
Much as others have determined over the years, HF at max and Mid at mid L-pad attenuation gives maximally flat on medial axis. Bass extends below 30 Hz, substantially as advertised:

While waiting for your full-range measurements I've been going through all the 3a literature in search of an AR claim for system response, and it's damned hard to find. The closest I've come is a brochure in which AR quotes the 1970 Consumer Guide review, "extremely smooth frequency response from 30 Hz to 17 KHz." It's abundantly obvious from this that the folks at AR didn't have a very high opinion of this measurement even when they scored fairly well.

Is there any way to measure the performance of the pair in a typical listening environment, and from different listening postions? I'm wondering if there's any way to quantify what produces the experience that has always (at least, to me) distinguished these speakers from most others of its type: the ability to get up and walk around the room without the stereo effect changing. During the 70's and 80's my non-AR-owning friends often paced back and forth across my living room in disbelief, and one of them finally turned to me after about 10 minutes and said, "You get good stereo everywhere in this room. That's not fair."

I suppose it might require a binaural head to collect any usable data for something like this...
 
Are there specific threads you are referring to? I looked but have not found what I would think of as an upgrade post.


http://www.classicspeakerpages.net/IP.Board/index.php?showtopic=5109&hl=hivi

The discussion is actually more "acceptable replacement" for hard-to-obtain originals. There are no easily installable (as in, you don't have to replace the crossovers or make cabinet mods that are more complicated than drilling some additional mounting screw holes) replacement drivers that are considered "upgrades."
 
Last edited:
Is there any way to measure the performance of the pair in a typical listening environment, and from different listening postions? I'm wondering if there's any way to quantify what produces the experience that has always (at least, to me) distinguished these speakers from most others of its type: the ability to get up and walk around the room without the stereo effect changing. During the 70's and 80's my non-AR-owning friends often paced back and forth across my living room in disbelief, and one of them finally turned to me after about 10 minutes and said, "You get good stereo everywhere in this room. That's not fair."
You're talking about dispersion, directivity, and power response.

It's easy enough to spray the room with HF dispersion and generate a huge soundstage. What a listener hears at any location, however, is dependent upon what is being reflected back by the room, easily seen as being variably dependent upon the the reflectivity of the room itself and its contents. Drapes on one side and a blank wall on the other, and it's fruitless; you've got to construct and arrange the room as part of the speaker. I can measure the influence of the location of my coffee cup. It's the room we're listening to, not the program.

At the next level, and this is where AR's (and others') focus comes to the fore, is power response. Assuming uniform reflectivity, what the speaker is sending out must be uniform in two respects: first, the frequency response across the spectrum delivered by the several drivers must be smooth and flat, as a starting point, at least, and second, all of the drivers must exhibit the same dispersion pattern in their respective ranges, particularly in the mid and high frequencies, such that what is reflected back is similarly uniform at all angles. The Toole papers discuss this in detail; power response is easily quantifiable.

However, still relying primarily upon room reflections to generate the soundfield, we remain at the mercy of the room, and thus there is an enormous literature and industry devoted to the science of room treatments to "tune" the system, incorporating the room and speaker(s) in combination, to achieve the desired result. Another school teaches the opposite: forget all of that, deaden the room, design speakers with a very narrow dispersion pattern, aim them at the listener, and remove the influence of the room from the equation to the largest extent achievable, in effect, build big headphones, and listen from the precisely defined "sweet spot" between them.

There's another approach, and I hesitate to describe it as either a compromise or intermediate between these, because, though it incorporates elements of both extremes, it conceptually employs them differently: constant directivity. Originally developed and refined for use in performance spaces to control reflections and provide a uniform direct soundfield in indoor spaces like theaters and concert halls as well as outdoor venues, it may also be effectively deployed in our smaller listening areas to minimize the role of the room. Again, I refer you and other interested readers to the Toole papers, several of which may be found here:

http://www.harman.com/about_harman/technology_leadership.aspx

And to this Linkwitz lecture on the subject:

http://www.audiokarma.org/forums/showthread.php?p=2087949#post2087949

And Geddes, of course.

Many of AR's design objectives can be met with a $6 waveguide today without having to rationalize or conceal the facts.... :yes:
 
Last edited:
You're talking about dispersion, directivity, and power response.

It's easy enough to spray the room with HF dispersion and generate a huge soundstage. What a listener hears at any location, however, is dependent upon what is being reflected back by the room, easily seen as being variably dependent upon the the reflectivity of the room itself and its contents. Drapes on one side and a blank wall on the other, and it's fruitless; you've got to construct and arrange the room as part of the speaker. I can measure the influence of the location of my coffee cup. It's the room we're listening to, not the program.


I have a style distaste for heavy drapes or carpeting that has been applied to every home I've ever lived in, and has presumably resulted in most of my living rooms being pretty much the same acoustically. I guess I'll just have to continue to tell anyone who comes over that it's black magic. :)

So, now that you've completed all these measurements, how do they sound...?
 
Last edited:
There is another factor that will come to bear on the results for me. I have a receiver that includes advanced Audassy processing. This system balances the response at the listening position. It can adjust frequency response of the individual channel amps and corrects for the reflection characteristics of the room. It is supposed to do a pretty good job. This could obviate some of the issues of non-flat speakers. But dispersion is probably not fixed.
 
I'm thinking you may have fewer issues in HT, as it does not rely on the speakers to generate the image or soundstage; the center and surrounds do that.... :yes:
 
I have been reading posts in a number of sites off and on for a couple of months because I have some speakers I need to do a bit of refurbishing on. A pair of Heath Kit AR3as and a pair of AR4xs. The 3as have been re-foamed but I note that neither tweeter seems to be doing anything (or my ears are simply not up to it). I will measure coils and see what is wrong. They have never really spent any time with real loud music, but they may have been run by an underpowered amp for a while (never knew that was an issue). I have no idea what condition things are in the 4s. I am the original owner of all of them.

I was thinking of at least cleaning pots and putting in new caps. If the 3a HF speakers are shot that will mean replacing them; probably not with originals.

But from my reading, and this thread particularly, I have developed a real headache over just what I should really do.

I know this is sort of heresy, but has anyone taken 3as and put in a new HF speaker, and redid the filter network with modern tools (and no pots) to see if the speakers can be made "better". Given the assorted graphs, it would seem that things could be made better with these changes.

I always bought the "accuracy" pitch for these speakers, and I always felt they were decent. I am going to use all of them in a home theater (without a sub woofer for now), so if I can get a better result I would be interested in trying it.

I took a pair and replaced the mid and tweeter with KEF B110 and T27. I use the BBC FL6-23 LS3/5A crossover for the top and a 2nd order crossover for the lowpass, adding a 2nd roder highpass to the B110. It's complicated but it works. I'm listening to them now. I like the LS3/5a mids and highs, and the AR3 lows, so I combined them.
I've owned a few pairs of AR3 and AR3a speakers. I like the cabinets and woofers. The mids and tweets always sounded dull and lifeless to my ear in my rooms. YMMV, and probably will. http://www.audiokarma.org/forums/showthread.php?t=148334
<EDIT>

I'm not actually recommending that anyone DO this, merely observing that it's been done.
 
Last edited:
While waiting for your full-range measurements I've been going through all the 3a literature in search of an AR claim for system response, and it's damned hard to find.
Per Allison, 1970:

attachment.php


attachment.php


attachment.php

Fig. 9 is the on-axis system curve with molding, level controls at maximum. Some representative curves at other angles appear in Figs. 10 and 11.
 

Attachments

  • Allison Axial.jpg
    Allison Axial.jpg
    67.3 KB · Views: 207
  • Allison 30.jpg
    Allison 30.jpg
    68.4 KB · Views: 206
  • Allison 60.jpg
    Allison 60.jpg
    65.9 KB · Views: 206
Zilch,

Some fodder for you.

- A poor scan of my populist contribution to the debate. You might find the references at the end interesting.

http://www.kenkantor.com/publications/magic_speaker/magic_speaker.pdf

- A link to the AES paper covering the same topic. It appeared in the Journal a few years later, but I don't remember the issue.

http://www.aes.org/e-lib/browse.cfm?elib=11565

- An AES search that brings up various commentary about the subject.

http://www.google.com/search?q=kantor+site:aes.org


Having discussed, debated and brainstormed on the matter with the folks you mentioned and many others over the years, I am of the opinion that the problems remain unresolved. Perhaps some of them are intractable.

Also, I wouldn't underestimate the seriousness with which AR considered the matter, nor the engineering efforts that were directed towards it. But, hey, I'm biased!

-k



You're talking about dispersion, directivity, and power response.

It's easy enough to spray the room with HF dispersion and generate a huge soundstage. What a listener hears at any location, however, is dependent upon what is being reflected back by the room, easily seen as being variably dependent upon the the reflectivity of the room itself and its contents. Drapes on one side and a blank wall on the other, and it's fruitless; you've got to construct and arrange the room as part of the speaker. I can measure the influence of the location of my coffee cup. It's the room we're listening to, not the program.

At the next level, and this is where AR's (and others') focus comes to the fore, is power response. Assuming uniform reflectivity, what the speaker is sending out must be uniform in two respects: first, the frequency response across the spectrum delivered by the several drivers must be smooth and flat, as a starting point, at least, and second, all of the drivers must exhibit the same dispersion pattern in their respective ranges, particularly in the mid and high frequencies, such that what is reflected back is similarly uniform at all angles. The Toole papers discuss this in detail; power response is easily quantifiable.

However, still relying primarily upon room reflections to generate the soundfield, we remain at the mercy of the room, and thus there is an enormous literature and industry devoted to the science of room treatments to "tune" the system, incorporating the room and speaker(s) in combination, to achieve the desired result. Another school teaches the opposite: forget all of that, deaden the room, design speakers with a very narrow dispersion pattern, aim them at the listener, and remove the influence of the room from the equation to the largest extent achievable, in effect, build big headphones, and listen from the precisely defined "sweet spot" between them.

There's another approach, and I hesitate to describe it as either a compromise or intermediate between these, because, though it incorporates elements of both extremes, it conceptually employs them differently: constant directivity. Originally developed and refined for use in performance spaces to control reflections and provide a uniform direct soundfield in indoor spaces like theaters and concert halls as well as outdoor venues, it may also be effectively deployed in our smaller listening areas to minimize the role of the room. Again, I refer you and other interested readers to the Toole papers, several of which may be found here:

http://www.harman.com/about_harman/technology_leadership.aspx

And to this Linkwitz lecture on the subject:

http://www.audiokarma.org/forums/showthread.php?p=2087949#post2087949

And Geddes, of course.

Many of AR's design objectives can be met with a $6 waveguide today without having to rationalize or conceal the facts.... :yes:
 
Thanks, Ken!

I'm seeing a prevalence of what I consider to be magic phrases there, like "constant directivity" and "minimize early reflections...." :thmbsp:
 
Back
Top Bottom