AR-11 The Good, The Bad, and The Ugly ...

[Pete B]

I was going to post a fairly detailed review of the AR-11 based on essentially reverse engineering them, however, it's not all good news and given the undue hostility directed at me for anything negative said about AR designs, I'm not going to bother. I do have a few comments, some are my opinions, others are engineering facts:

The Good:

Nice build quality on the midrange.

Air core inductors to eliminate saturation distortion.

4 ohms to get the most out of high current solid state amps.

A long (linear) throw woofer, at least in it's day.

The woofer and crossover were designed as an economical system, lower than usual woofer DCR allowing higher (cheaper) DCR inductor.

The Bad:

All of the drivers are primitive, without the use of extended pole pieces, copper or Al shorting rings, or even any simple methods used to improve motor linearity besides the original innovation of an over hung voice coil.

Voicing seems to be for the mass market, using high Q crossovers to make them loud when shown in show rooms.

I believe that the high currents and high Q in the crossover causes failure of some of the electrolytic caps which leads to driver failure, especially on the tweeter.

The 200003 woofer used in the AR-11 has a much lower moving mass than the later AR-3a type woofer, 68 g as compared to ~105g. This was a major mistake in my opinion, the low Fc is what made the AR-3a have deep extension. I was in fact surprised to find this change as it was key to their performance. It was done to make them more efficient(loud) and with more 50-60 Hz thumping bass for mass market appeal.

The woofer does not have enough mechanical Xmax and bottoms with a loud crack.

The Ugly:

The tweeter build quality is pathetic. A $15-20 SEAS or Vifa 3/4" tweeter has far superior build quality.

The low pass rolloff of the woofer XO has a fairly high Q, the high pass of the midrange and tweeter XOs also have high Q when the level switches are at max. The only reason I can see for this is to get that extra 1-2 dB of level and make them loud in showroom comparisons for mass market sales. I was also informed that this is why the Advent has a large peak at 1 kHz, to make them sell to the mass market. As I've stated before this high Q makes the system much more sensitive to inductor DCR and capacitor ESR and not in a good way.

Many here say they run AR-11s with the level switches full up. I found that -6 dB on the mid and -3 dB on the tweeter was as close to a natural sound as I could get. I have a feeling that 4 to 5 on the mid, and 2 on the tweeter would be closer. These settings approximately agree with the center settings of the AR-3a. Even with these settings the system just did not sound "right" it was diffuse as if out of focus, what I believe is poor driver integration.

I plan to do a direct comparison but based on my notes the Advents with BSC sound far better than the stock properly restored AR-11.

The good news is that it's not too difficult to fix the voicing of the system.

Pete B.

[frankmarsi]

Peter B., if I may inject my two cents? Back in 1978-79 I had a friend who always tried to emulate the sound of my AR-3as. He bought AR-11's because at that time 3a's were gone. And to add to your findings in plain terms they never had the guts or slam as my 3a's did, even though he bought a Phase Linear4000 and Phase Linear 400 as I had to drive them. His seemed to go louder with-out fear of burning the tweeters as I always did with my 3a's, but no balls in the sound although a louder high-end, but lacking somewhat in the bass. Whenever I brought over my vinyl his speakers never sounded as robust as mine. So seemingly you are correct in your findings.

My question here is how come no one has responded to you, or am I incorrect to say that? There seems to be a 'capacitor' growd, a 'graph' growd and others with opinions, even an AR-LST growd, but nonetheless an obvious divide amongst members. A scientific area, and a mixed area,but no clear unity.Why doesn't anyone speak of the latest great quality album that may be a good listening treat and test? Sounds like a Ster--file growd versus a Stereo Review and Audio Magazine separation going on here. Perhaps that's O.K., but I thought there'd be more sharing of information on this site? Any response?

Frank Marsi

[kkantor]

Pete,

How do you feel the 11 compares to a 10pi?

-k

www.aural.org

www.tymphany.com

[Pete B]

Hi Ken,

I wish I could comment but I've not heard the 10pi, or simulated it.

My friend who owns these 11s said he wished he'd bought 10pi's.

I did comment in another thread that the autotransformer and cap probably act as a high pass filter to reduce bottoming, it also probably adds some loss between the amp and XO to lower the Q of the filters. So it's possible for reasons other than or in addition to the level function of the auto transformer, it helps in other ways.

I'd like to hear them side by side.

Pete B.

>Pete,

>

>How do you feel the 11 compares to a 10pi?

>

>-k

>

>

>www.aural.org

>www.tymphany.com

[Pete B]

Hi Frank,

Nice to hear that you confirm my findings Frank.

As far as people responding, well I've talked on internet groups for over 10 years, took a 7 year break busy with family, and there's a lot of politics and emotions, which I did not expect. I'm not the type for brand loyalty, I like or am curious about many different brands/systems/designs.

I find that lay-experts, not you by the way, come to these boards wanting their unresearched, and usually untested in a controlled environment, beliefs validated by "experts" and attack when they are not. There's also a thing called group think, the insecure group has common beliefs or a culture - well I think that's right, Joe says it's right, Mike says it's right, Jack says it's right, it must be right how could we all be wrong? They are often wrong.

But overall this place is mild and rather good compared to others. The internet is famous for flame wars as you probably know.

People probably don't like to talk to others who do not share common beliefs, and we do have differing beliefs here as you noted.

That's life, to each their own, it's just a hobby for some so it's all fine.

You make an interesting point about sharing of information, I believe that I'm fairly open and giving with my findings, some are less so.

I'm surprised that more of the old AR engineers are not here, or are they and not identifying themselves? Who designed the AR-11? That person would probably not be happy with my findings, however having worked in large companies I know that the orders come down from the bean counters and the suits. MAKE IT LOUD, HYPE UP THE BASS, MAKE IT CHEAP we have to sell product. We don't have funding to redesign that tweeter, woofer, etc. the old parts are good enough use them. Are you done yet, what's taking so long? Are you done yet, what's taking so long? This is what real life engineering is about, not people in lab coats.

Sounds like your making a lot of progress on your LST's, Frank, how's your Buick?

I'm surprised no one commented about the beautiful ladies on your photography web site. I think most men wish they had that job photographing gorgeous ladies like you do.

Pete B.

>Peter B., if I may inject my two cents? Back in 1978-79 I had

>a friend who always tried to emulate the sound of my AR-3as.

>He bought AR-11's because at that time 3a's were gone. And to

>add to your findings in plain terms they never had the guts or

>slam as my 3a's did, even though he bought a Phase Linear4000

>and Phase Linear 400 as I had to drive them. His seemed to go

>louder with-out fear of burning the tweeters as I always did

>with my 3a's, but no balls in the sound although a louder

>high-end, but lacking somewhat in the bass. Whenever I brought

>over my vinyl his speakers never sounded as robust as mine. So

>seemingly you are correct in your findings.

>My question here is how come no one has responded to you, or

>am I incorrect to say that? There seems to be a 'capacitor'

>growd, a 'graph' growd and others with opinions, even an

>AR-LST growd, but nonetheless an obvious divide amongst

>members. A scientific area, and a mixed area,but no clear

>unity.Why doesn't anyone speak of the latest great quality

>album that may be a good listening treat and test? Sounds like

>a Ster--file growd versus a Stereo Review and Audio Magazine

>separation going on here. Perhaps that's O.K., but I thought

>there'd be more sharing of information on this site? Any

>response?

>Frank Marsi

>

[dynaco_dan]

Beautiful women?

What photography website?

I think I'll check it out and see how well you focus.

[Pete B]

>Beautiful women?

>

>What photography website?

>

>I think I'll check it out and see how well you focus.

Frank posted this in the first post I saw from him, I sure wish I took up photography as a profession:

http://www.frankmarsi.citymax.com/page/page/403443.htm

Now weren't you all paying attention? LOL!

[dynaco_dan]

>>Beautiful women?

>>

>>What photography website?

>>

>>I think I'll check it out and see how well you focus.

>

>Frank posted this in the first post I saw from him, I sure

>wish I took up photography as a profession:

> http://www.frankmarsi.citymax.com/page/page/403443.htm

>

>Now weren't you all paying attention? LOL!

Thank you for the link.

Nice clean lense.

I think my new pacemaker kicked in a few times.

Some people have all the luck, poor Frank.

But, you know what.

It's looks like a tough job, but, you know, someone has to do it.

I can see, dragging myself to work every day.

Beautiful women, to be sure.

[Pete B]

>Peter B., if I may inject my two cents? Back in 1978-79 I had

>a friend who always tried to emulate the sound of my AR-3as.

>He bought AR-11's because at that time 3a's were gone. And to

>add to your findings in plain terms they never had the guts or

>slam as my 3a's did, even though he bought a Phase Linear4000

>and Phase Linear 400 as I had to drive them. His seemed to go

>louder with-out fear of burning the tweeters as I always did

>with my 3a's, but no balls in the sound although a louder

>high-end, but lacking somewhat in the bass. Whenever I brought

>over my vinyl his speakers never sounded as robust as mine. So

>seemingly you are correct in your findings.

>My question here is how come no one has responded to you, or

>am I incorrect to say that? There seems to be a 'capacitor'

>growd, a 'graph' growd and others with opinions, even an

>AR-LST growd, but nonetheless an obvious divide amongst

>members. A scientific area, and a mixed area,but no clear

>unity.Why doesn't anyone speak of the latest great quality

>album that may be a good listening treat and test? Sounds like

>a Ster--file growd versus a Stereo Review and Audio Magazine

>separation going on here. Perhaps that's O.K., but I thought

>there'd be more sharing of information on this site? Any

>response?

>Frank Marsi

Something interesting Frank is that I've supplied a lot of information about the observations and measurements that I've made on the 200003 woofers that I've worked on. I do this to help people who are complaining about not getting the AR bass performance that their used to.

Yet when I ask a few simple questions, I get no response:

"It would help if someone measured the piston diameter of the AR-303 woofer as a sanity check."

"I'd like to hear from others with woofers with the screen on the back as to how many corrugations the spider has, and the part number on the back of the cone.

The date code, any part number markings, and DC resistance would also be helpful."

Not directed at you Frank, but do you think I'll continue giving information so freely?

Pete B.

[Guest Dargo]

Chair Right !!

[Guest Bret]

Pete,

I've been doing some reading (forensics) that doesn't square with your conclusions. Obviously I can't quibble about your data or the way you arrived at it, because you didn't elaborate.

>Voicing seems to be for the mass market, using high Q crossovers to make them loud when shown in show rooms.<

The "system" Q is targeted at .7 and from all I can read .7 (or more precisely .707) is the target to hit flattest woofer response. The literature says that both the 10pi and 11 were targeted at .7 and Ken's measurement of a 3a put the Q at .71. This would be perfectly in keeping with AR's history and design philosophy. Flat frequency response.

If crossover Q was raised, it was done in order to leave the system Q alone.

What I can understand is someone's wanting to change the woofer's response (in the crossover region) in order to better match a 12" to a 1.5". Evidently that was so difficult that AR engineers abandoned it for top-of-the-line models thereafter. This would have been improved by raising system Q, which they didn't do.

Looking at the design from a Q standpoint, I wonder why the ADD guys (in replacing the 3a) didn't do what Ken later did in the 303: Increase the cabinet size and get a woofer that's "better" in the high-end of the frequencies in the crossover range, and raise Q.

My guess is that you (read: one) can't properly damp a 12" acoustic suspension woofer in a cabinet of less than 1.48cu ft and get a Q as high as 0.7 while maintaining anything like "deep bass".

Raising Q in the 1.48cu ft cabinet would have helped on the top end, but there would have been a large penalty on the low-end as the Fc would have been raised considerably and the -3db point in the low bass would have followed suit. (in other words: the bass would have disappeared using a 12" woofer in a 1.48cu ft cabinet with a target system Q of 1.)

Again, I'm talking forensics. There is little use in saying that they could have used a larger cabinet or used a smaller woofer. They didn't.

"Okay, so why didn't they stick with the AR3a's crossover?" The tweeter in the 11 and 10pi are, uh. . . gosh, I don't know what they are other than "apparently louder" than the 3a had been to-this-point. They had higher power handling ability so maybe the crossover was re-designed to take advantage of that?

The "tonal balance" of the speaker would have changed and considering the 3a was pretty balanced you'd want to "put it back."

>The 200003 woofer used in the AR-11 has a much lower moving mass than the later AR-3a type woofer, 68 g as compared to ~105g. This was a major mistake in my opinion, the low Fc is what made the AR-3a have deep extension.<

Unless you meant Fs and accidently typed Fc, this is contradictory to anything I can read about the 3a vs 11. The 3a and 11 have the same "targeted" Fc of 42Hz.

The 3a info chiefly comes from http://www.classicspeakerpages.net/ar/ar-303/303v3a.pdf

What changed was the drivers' Fs from 13-14Hz up to 18Hz. That would have changed the -3db point in the low bass, just as you said (and as I can also testify seems apparent with pipe organ music) but perhaps was done to improve the behavior of the 12" as it crossed-over to the 1.5" midrange (which I theorize is why I prefer the 10pi overall).

This change would have resulted in less bass extension, as we've noted, but I don't see how it could have simultaneously resulted in more bass close above Fc, as you said, given that Q did not change.

I'm just confused because if Q was raised beyond .7 it would result in an earlier roll-off (above Fc) and so I don't understand how a bass peak above Fc could be gained even if they were trying. I'm guessing you did something that moved Fc up.

Is there some interdependence I haven't taken into account?

It was predictable that the low Fs on your rebuilt driver wouldn't affect Fc too much (as the driver and cabinet's sizes seem to be chief determiners), but it is obvious that the 2 ohms you used for equivalent inductor impedence is way off the mark as you ended-up with a Qtc of above .9. Adding the inductor's resistence should have dropped Fc a little while raising Q quite a bit, but you got a tiny, tiny change in Fc and raised Qtc by 50%. I'm thinking, "Something about this just looks wrong." (as a model, I mean)

Maybe the reason you experienced this 50-60Hz hump and dramatic loss of low bass was that something you did raised the Q way beyond .7 ??? Could it be that you, somehow, raised the Fc to 50-60Hz accidently and were hearing system resonance? (Fc does seem to be more audible in these than the 3a, although I have no explanation) This would agree with my observations about what happens when you use a Tonegen replacement in these.

>I was in fact surprised to find this change as it was key to their performance. It was done to make them more efficient(loud) and with more 50-60 Hz thumping bass for mass market appeal.<

I hope you won't take exception to my saying this, but while I can follow your logic on deep bass extension, I guess I don't know enough to see any reason to believe that this 50-60Hz bass peak even happened unless you unwittingly raised Fc to 50-60Hz. If you observed it I don't doubt it, but I'd like to know what you did about cabinet damping. Again - the system Q of the 3a and 11 were the same, only the Fs of the woofer was raised by 3-4Hz.

It's that statement about mass-market appeal that I found curious, so I bothered to take the time to try to understand what was going-on here. As you may remember from personal experience, or surely you have seen reported; giving an AR speaker any additional bass, or a bass peak, would have been exactly the wrong thing to do for mass-market appeal.

The great problem AR had when quickly demo-ed against anything else was the "laid-back" nature of the midrange and tweeter. Almost anything else "sounded better" A/Bed quickly because most other speakers of the 11s day were way, way too "forward" and bright.

Were you expecting to hear a problem after measuring Qtc at >.9?

Bret

[Pete B]

There are a few things that I cannot explain because I do not have enough data/design information, and I believe that I should also clarify a few things here.

Your drawing some incorrect conclusions about my statements because you seem to be interpreting my use of Q as for only the woofer system Qtc. There are actually several that I'm referring to:

Woofer closed box Q - Qtc

Woofer XO lowpass Qwlp - 2.85 mH/120uF (approximately 2nd order)

Midrange XO highpass Qmhp - 40uF/.88 mH (approximately 2nd order)

Tweeter XO highpass Qthp - 10uF/.101 mH (approximately 2nd order)

Woofer XO contour for diffraction compensation, there is no Q associated with this.

I've used these terms Qwlp, Qmhp, Qthp to be more specific, they're not normally used in crossover design.

One reason that I was surprised to find such a difference in moving mass is that the advertised Fc for the 3a and 11 did not change significantly, yet it seems that the mass did and with the same enclosure volume (and high compliance) this does not make sense. I have many questions, is the mass measured on your drivers correct, had they been worked on before, what was the intended moving mass for the early 3a, and the 11? It will take time to collect all this data.

My best read of this situation is that the early 11 was prototyped with the higher moving mass driver, thus the same advertised Fc as the 3a. All 11's might not be the same, if people would read the part number on the back of the cone and tell us we would have a better idea. Note that Frank in his post above confirms my point about deep bass. I would not be surprised if they all shipped with the lighter cone driver but again no one is offering input to confirm this.

My comments about high Q were mainly referring to what I called above Qwlp, Qmhp, Qthp and I really do not have a problem with the Qtc of the system. What brings up the 50 to 60 Hz range is due to too much baffle step, and what hurts deep base is the higher closed box 12 dB/oct rolloff due to lower moving mass.

You mention me wanting to hear something, actually this AR-11 sounds probably much better than most 11s as far as deep bass goes, because I used a highly compliant edge, and I believe that the spider is beyond broken in given the 15 Hz Fs. I believe based on Fs and what others are finding that this driver is an anomaly providing an Fc of 44 Hz. I stated that I was refoaming this driver as an experiment to measure an early driver. I would replace the spider if I was to call this a complete rebuild and then Fs/Fc would go up. My comments were based more on the amount of baffle step, what others are reporting about deep bass response (including you), that were confirmed by Unibox simulations. I did not have the higher moving mass driver or an AR-3a for comparison purposes.

You state:

>Maybe the reason you experienced this 50-60Hz hump and

>dramatic loss of low bass was that something you did raised

>the Q way beyond .7 ??? Could it be that you, somehow, raised

>the Fc to 50-60Hz accidently and were hearing system

>resonance? (Fc does seem to be more audible in these than the

>3a, although I have no explanation) This would agree with my

>observations about what happens when you use a Tonegen

>replacement in these."

I went back and checked I did not say anything about "dramatic loss of low bass" please be accurate in your statements. I *do* think that the situation would be worse with a Tonegen driver or a fully rebuilt 200003 which has a stiffer suspension. This 11 has fairly good extension, I was referring to what many are reporting here.

Your guess about a 12" driver in a 1.48 cu ft box is incorrect, it would take some time to explain this.

You state:

"The great problem AR had when quickly demo-ed against anything else was the "laid-back" nature of the midrange and tweeter.

Almost anything else "sounded better" A/Bed quickly because

most other speakers of the 11s day were way, way too "forward"

and bright."

Where did you have the mid and tweeter switches when you made these AR-11 comparisons? Given all the permutations of those switches the 11 can be many different speakers, there is no single 11 sound. My comments are by todays standards since so many here state that these vintage systems are better than modern designs.

I'll have to comment on the rest of your post at a later time.

I'm actually loosing interest in AR products, my goal was to understand the AR-3a/11, and improve it if possible.

Pete B.

>Pete,

>

>I've been doing some reading (forensics) that doesn't square

>with your conclusions. Obviously I can't quibble about your

>data or the way you arrived at it, because you didn't

>elaborate.

>

>>Voicing seems to be for the mass market, using high Q

>crossovers to make them loud when shown in show rooms.<

>

>The "system" Q is targeted at .7 and from all I can read .7

>(or more precisely .707) is the target to hit flattest woofer

>response. The literature says that both the 10pi and 11 were

>targeted at .7 and Ken's measurement of a 3a put the Q at .71.

> This would be perfectly in keeping with AR's history and

>design philosophy. Flat frequency response.

>

>If crossover Q was raised, it was done in order to leave the

>system Q alone.

>

>What I can understand is someone's wanting to change the

>woofer's response (in the crossover region) in order to better

>match a 12" to a 1.5". Evidently that was so difficult that

>AR engineers abandoned it for top-of-the-line models

>thereafter. This would have been improved by raising system

>Q, which they didn't do.

>

>Looking at the design from a Q standpoint, I wonder why the

>ADD guys (in replacing the 3a) didn't do what Ken later did in

>the 303: Increase the cabinet size and get a woofer that's

>"better" in the high-end of the frequencies in the crossover

>range, and raise Q.

>

>My guess is that you (read: one) can't properly damp a 12"

>acoustic suspension woofer in a cabinet of less than 1.48cu ft

>and get a Q as high as 0.7 while maintaining anything like

>"deep bass".

>

>Raising Q in the 1.48cu ft cabinet would have helped on the

>top end, but there would have been a large penalty on the

>low-end as the Fc would have been raised considerably and the

>-3db point in the low bass would have followed suit. (in

>other words: the bass would have disappeared using a 12"

>woofer in a 1.48cu ft cabinet with a target system Q of 1.)

>

>Again, I'm talking forensics. There is little use in saying

>that they could have used a larger cabinet or used a smaller

>woofer. They didn't.

>

>"Okay, so why didn't they stick with the AR3a's crossover?"

>The tweeter in the 11 and 10pi are, uh. . . gosh, I don't

>know what they are other than "apparently louder" than the 3a

>had been to-this-point. They had higher power handling ability

>so maybe the crossover was re-designed to take advantage of

>that?

>

>The "tonal balance" of the speaker would have changed and

>considering the 3a was pretty balanced you'd want to "put it

>back."

>

>>The 200003 woofer used in the AR-11 has a much lower moving

>mass than the later AR-3a type woofer, 68 g as compared to

>~105g. This was a major mistake in my opinion, the low Fc is

>what made the AR-3a have deep extension.<

>

>Unless you meant Fs and accidently typed Fc, this is

>contradictory to anything I can read about the 3a vs 11. The

>3a and 11 have the same "targeted" Fc of 42Hz.

>

>The 3a info chiefly comes from

>http://www.classicspeakerpages.net/ar/ar-303/303v3a.pdf

>

>What changed was the drivers' Fs from 13-14Hz up to 18Hz.

>That would have changed the -3db point in the low bass, just

>as you said (and as I can also testify seems apparent with

>pipe organ music) but perhaps was done to improve the behavior

>of the 12" as it crossed-over to the 1.5" midrange (which I

>theorize is why I prefer the 10pi overall).

>

>This change would have resulted in less bass extension, as

>we've noted, but I don't see how it could have simultaneously

>resulted in more bass close above Fc, as you said, given that

>Q did not change.

>

>I'm just confused because if Q was raised beyond .7 it would

>result in an earlier roll-off (above Fc) and so I don't

>understand how a bass peak above Fc could be gained even if

>they were trying. I'm guessing you did something that moved

>Fc up.

>

>Is there some interdependence I haven't taken into account?

>

>It was predictable that the low Fs on your rebuilt driver

>wouldn't affect Fc too much (as the driver and cabinet's sizes

>seem to be chief determiners), but it is obvious that the 2

>ohms you used for equivalent inductor impedence is way off the

>mark as you ended-up with a Qtc of above .9. Adding the

>inductor's resistence should have dropped Fc a little while

>raising Q quite a bit, but you got a tiny, tiny change in Fc

>and raised Qtc by 50%. I'm thinking, "Something about this

>just looks wrong." (as a model, I mean)

>

>Maybe the reason you experienced this 50-60Hz hump and

>dramatic loss of low bass was that something you did raised

>the Q way beyond .7 ??? Could it be that you, somehow, raised

>the Fc to 50-60Hz accidently and were hearing system

>resonance? (Fc does seem to be more audible in these than the

>3a, although I have no explanation) This would agree with my

>observations about what happens when you use a Tonegen

>replacement in these.

>

>>I was in fact surprised to find this change as it was key to

>their performance. It was done to make them more

>efficient(loud) and with more 50-60 Hz thumping bass for mass

>market appeal.<

>

>I hope you won't take exception to my saying this, but while I

>can follow your logic on deep bass extension, I guess I don't

>know enough to see any reason to believe that this 50-60Hz

>bass peak even happened unless you unwittingly raised Fc to

>50-60Hz. If you observed it I don't doubt it, but I'd like to

>know what you did about cabinet damping. Again - the system Q

>of the 3a and 11 were the same, only the Fs of the woofer was

>raised by 3-4Hz.

>

>It's that statement about mass-market appeal that I found

>curious, so I bothered to take the time to try to understand

>what was going-on here. As you may remember from personal

>experience, or surely you have seen reported; giving an AR

>speaker any additional bass, or a bass peak, would have been

>exactly the wrong thing to do for mass-market appeal.

>

>The great problem AR had when quickly demo-ed against anything

>else was the "laid-back" nature of the midrange and tweeter.

>Almost anything else "sounded better" A/Bed quickly because

>most other speakers of the 11s day were way, way too "forward"

>and bright.

>

>Were you expecting to hear a problem after measuring Qtc at

>>.9?

>

>Bret

[dynaco_dan]

Hi there;

Well, fellas, you both talk above my head but I still enjoy reading what you both have say.

You both are technical people and it is obvious you both enjoy your views on the subject at hand.

I do remember my very first experience with a hifi system demo back in a stereo store in 1965.

When the salesman switched back and forth between Brand X and AR-4X's, he also did a sneeky, and pressed the loudness button when he switched over to Brand X.

He disengaged the loudness button when he went back to the AR-4X's.

I was not wise to what he was doing at that time.

He had me convinced that the Consumer Reports article about the AR-4X that I had in my hand, was bogus.

He also, after looking around, as if he was going to tell us a national secret, and told my Big Bros and I that, south of the border the label on the equipment was called McIntosh.

I was sold, signed and delivered.

A week later, Big Bros and I went into another stereo store, "BIG MISTAKE", and we heard a very large speaker system, amongst a wall of small speaker systems, on display.

That monstrous sytem was the AR-4X's.

By themselves they were unbelieveably wonderful.

With improper A-B'ing the AR-4X's sound flat, no bass, no mids, and no highs.

This tends to be the same for any quality hifi speaker system I have heard.

Also back then, they were ugly, they were walnut veneered, yes, but no fretwork, no shiny aluminum frames and only non-colourful beige linen grill cloths.

The salesmen all really had a field day, with the AR grill cloths off, ugly again, and poor sounding also.

One really had to have a lot of experience, to have ended up with a pair of old classic speakers in their home.

There was a lot of things going against a person purchasing any classic speakers made by AR in particular.

It took me only one early on buying mistake for me to see the light.

I have read here, where others have bought a system way back in their college days, or similar and still have them.

Please don't be scared to offend us AR owners of something negative you may wish to write about, you are at least 25 years too late to offend us anyways.

You can however, assist others like myself, who may wish to change out for example, from an OEM capacitor to something more readily available today.

For those of us, who need a replacement driver, your advice is really appreciated.

Which will not take an arm and a leg to buy, but will still help us maintain our speaker systems for a few more decades at least.

Thank you fellas for your very welcome contributions.

[soundminded]

The bass performance (for all loudspeaker systems) is accurately predicted by Newton's second law of motion as applied to forced oscillation. It is a staple of every college freshman engineering curriculum and is seen again and again in physics, calculus and sophomore mechanics-dynamics. It is not an approximation, it is an exact model which has been time proven for centuries now only yielding at near light speed to Einstein's modifications. It is a second order ordinary differential equation relating mass m, spring constant k, and viscosity b to displacement as a function of time.

The solution of this equation for the system mechanical resonance frequency is;

f©=(1/2PI)* SQ ROOT[(k/m)-(b/2m)squared]

k is the springiness of the suspension, hopefully mostly air trapped in the box in an acoustic suspension system. This is entirely linear as defined by the gas laws P1*V1=P2*V2 where P and V are volumes before and after slight compression or rarifaction. F, is the restoring force and is the pressure/cone area. k is delta F/delta x where x is the displacement.

m is the total moving mass

b is the velocity related factor which is the aerodynamic drag the speaker must overcome to push or pull air between the fibers of the stuffing. The amount, nature, and distribution of the stuffing is therefore critical to tuning the system. Any change in the density, packing, amount, or type of fibers or how it is distributed in the box will change b and therefore the tuning of the mechanical system. If b were to decrease by say compressing the stuffing making it impossible to pull air through it, it would increase f©. It would also remove the restoring force of any air trapped behind the fibers reducing the effective volume of the remaining air further increasing f©. m, k, and b must be adjusted for critical damping of .707 to obtain the flattest possible frequency response. Less damping (higher Q) will result in a peak at resonance which translates into spurious damped oscillation at the resonance frequency every time the system is excited at any frequency. More damping meaning a Q of less than .707 will decrease the low frequency response but at least there will be no spurious resonance. (This is how the suspension of a car is tuned and the solution of many other mechanical problems.)

One thing you can clearly see is the disadvantage of any ported system because k resulting from air pressure in such a system is strongly a function of frequency, being very low at resonance frequency of the air column and multiples of resonance frequency and very high at the midpoints between them. That is why ported systems have an inherently high Q which must be overcome if the system is to have anything like a smooth bass frequency response.

The problem for the upper end of the frequency response of any driver including the AR woofer is that at some point as frequency increases, its inertial mass becomes too great to respond to forced oscillation no matter how much exciting force is used. Reducing the mass increases this usable upper limit frequency but the penalty is increasing f© as well.

It is interesting that the second order ordinary differential equation is exactly the same equation (with different letters) used to describe and solve the LCR problem for tuned electrical circuits. The system repsonse is the superposition of the electrical and mechanical resonances. It doesn't matter either whether the elecrical resonances are at the loudspeaker output level in a passive crossover network or at the low signal preamplifier level controlled by equalizers or active crossover networks. Given how inexpensive these lower level circuits and audio amplifiers have become, it is surprising (to me) that they haven't mostly replaced passive crossover networks yet.

[Guest Bret]

Soundminded,

Are you then postulating that the original design of the AR-11 used a somewhat heavier cone material than did later models AND therefore when the driver was changed to a lighter material that Fc was increased but the literature wasn't?

I can buy that.

I went looking for Ken's measurements of the various DUTs, but I can't find a link that works. I think we're going to find that the 10pi drivers (and therefore early 11 drivers as well) were heavier than the later drivers - just as Pete says. The cone materials were different.

Bret

[soundminded]

>Soundminded,

>

>Are you then postulating that the original design of the AR-11

>used a somewhat heavier cone material than did later models

>AND therefore when the driver was changed to a lighter

>material that Fc was increased but the literature wasn't?

>

>I can buy that.

>

>I went looking for Ken's measurements of the various DUTs, but

>I can't find a link that works. I think we're going to find

>that the 10pi drivers (and therefore early 11 drivers as well)

>were heavier than the later drivers - just as Pete says. The

>cone materials were different.

>

>Bret

I can't say but it's clear there is a tradeoff. Where do you want to compromise the design? At the high end of the woofer response with a heavier cone maker a poorer match with the dome midrange or at the lower end with a ligher cone reducing the lowest bass output somewhat. AR faced with the same problem insofar as their best effort was concerned answered "neither" taking the bull by the horns and blasting away at both ends. By adding the 8 inch lower midrange in AR9 the transition problem to the dome went away and by doubling the size of the enclosure, adding a second woofer, restricting the stuffing to the top of the box, and reducing the woofer crossover to 200 hz they reduced f© by nearly a full octave, doubled the power handling capacity of the low frequency drivers, improved efficiency, and by side mounting the woofers coupled them to the room to better take advantage of its acoustic bass reinforcement. Now that's what I call an improvement. They more than kicked it up a notch.

[Pete B]

Are you ready to offer a reference to those power amp designs that run on 70.7 V supply that you claimed?:

http://www.classicspeakerpages.net/dc/dcbo...id=&page=3#6914

Pete B.

>The bass performance (for all loudspeaker systems) is

>accurately predicted by Newton's second law of motion as

>applied to forced oscillation. It is a staple of every

>college freshman engineering curriculum and is seen again and

>again in physics, calculus and sophomore mechanics-dynamics.

>It is not an approximation, it is an exact model which has

>been time proven for centuries now only yielding at near light

>speed to Einstein's modifications. It is a second order

>ordinary differential equation relating mass m, spring

>constant k, and viscosity b to displacement as a function of

>time.

>

>The solution of this equation for the system mechanical

>resonance frequency is;

>

>f©=(1/2PI)* SQ ROOT[(k/m)-(b/2m)squared]

>

>k is the springiness of the suspension, hopefully mostly air

>trapped in the box in an acoustic suspension system. This is

>entirely linear as defined by the gas laws P1*V1=P2*V2 where P

>and V are volumes before and after slight compression or

>rarifaction. F, is the restoring force and is the

>pressure/cone area. k is delta F/delta x where x is the

>displacement.

>

>m is the total moving mass

>

>b is the velocity related factor which is the aerodynamic drag

>the speaker must overcome to push or pull air between the

>fibers of the stuffing. The amount, nature, and distribution

>of the stuffing is therefore critical to tuning the system.

>Any change in the density, packing, amount, or type of fibers

>or how it is distributed in the box will change b and

>therefore the tuning of the mechanical system. If b were to

>decrease by say compressing the stuffing making it impossible

>to pull air through it, it would increase f©. It would also

>remove the restoring force of any air trapped behind the

>fibers reducing the effective volume of the remaining air

>further increasing f©. m, k, and b must be adjusted for

>critical damping of .707 to obtain the flattest possible

>frequency response. Less damping (higher Q) will result in a

>peak at resonance which translates into spurious damped

>oscillation at the resonance frequency every time the system

>is excited at any frequency. More damping meaning a Q of less

>than .707 will decrease the low frequency response but at

>least there will be no spurious resonance. (This is how the

>suspension of a car is tuned and the solution of many other

>mechanical problems.)

>

>One thing you can clearly see is the disadvantage of any

>ported system because k resulting from air pressure in such a

>system is strongly a function of frequency, being very low at

>resonance frequency of the air column and multiples of

>resonance frequency and very high at the midpoints between

>them. That is why ported systems have an inherently high Q

>which must be overcome if the system is to have anything like

>a smooth bass frequency response.

>

>The problem for the upper end of the frequency response of any

>driver including the AR woofer is that at some point as

>frequency increases, its inertial mass becomes too great to

>respond to forced oscillation no matter how much exciting

>force is used. Reducing the mass increases this usable upper

>limit frequency but the penalty is increasing f© as well.

>

>It is interesting that the second order ordinary differential

>equation is exactly the same equation (with different letters)

>used to describe and solve the LCR problem for tuned

>electrical circuits. The system repsonse is the superposition

>of the electrical and mechanical resonances. It doesn't

>matter either whether the elecrical resonances are at the

>loudspeaker output level in a passive crossover network or at

>the low signal preamplifier level controlled by equalizers or

>active crossover networks. Given how inexpensive these lower

>level circuits and audio amplifiers have become, it is

>surprising (to me) that they haven't mostly replaced passive

>crossover networks yet.

[Guest Bret]

>Your drawing some incorrect conclusions about my statements because you seem to be interpreting my use of Q as for only the woofer system Qtc. There are actually several that I'm referring to:

Woofer closed box Q - Qtc<

Woofer XO lowpass Qwlp - 2.85 mH/120uF (approximately 2nd order)

Midrange XO highpass Qmhp - 40uF/.88 mH (approximately 2nd order)

Tweeter XO highpass Qthp - 10uF/.101 mH (approximately 2nd order)

Woofer XO contour for diffraction compensation, there is no Q associated with this.

I've used these terms Qwlp, Qmhp, Qthp to be more specific, they're not normally used in crossover design.<

Why on earth would you refer to the “Q” of crossover elements? I don’t understand. What is to be gained, or learned, from that with the possible exception of sensitivity to capacitor ESR, and I’d have never thought of that. Is that how you determined the sensitivity to cap ESR? That’s what I’m taking away from this discussion, so if I’m wrong, please straighten me out.

I guess I’m confused because in your earlier post about rebuilding the woofers you did mention getting a Qtc of > 0.9 .

>One reason that I was surprised to find such a difference in moving mass is that the advertised Fc for the 3a and 11 did not change significantly, yet it seems that the mass did and with the same enclosure volume (and high compliance) this does not make sense.<

I guess I don’t understand, again. In a closed system the Fs and volume of the enclosure are the two greatest and most elemental driving forces for Fc. You can “goof around with it” a little by also fiddling with damping and inductor resistence, but not much if you have a target Q to meet. Without getting into true exotica I don’t see how you could get this performance out of a box much smaller than this. I’m talking 35Hz-1kHz here, not sub-woofer performance.

> I have many questions, is the mass measured on your drivers correct, had they been worked on before, what was the intended moving mass for the early 3a, and the 11? It will take time to collect all this data.<

I think Ken’s given most of it to us in the DUTs done that I can’t find a link-to. Remember there was one speaker with a new cone, spider, the whole sheebang from a 9. But we had at least two from 10pi’s that were original to the speakers and I maintain that we had at least two unchanged from 9s. As for the measurements being correct, that would have been done by Ken’s colleagues on their expensive equipment. I would think it would be right, but I certainly can’t swear it.

>My best read of this situation is that the early 11 was prototyped with the higher moving mass driver, thus the same advertised Fc as the 3a. All 11's might not be the same, if people would read the part number on the back of the cone and tell us we would have a better idea. Note that Frank in his post above confirms my point about deep bass. I would not be surprised if they all shipped with the lighter cone driver but again no one is offering input to confirm this.<

I don’t have the information you need, and I don’t have access to it.

>My comments about high Q were mainly referring to what I called above Qwlp, Qmhp, Qthp and I really do not have a problem with the Qtc of the system.<

What do you think the end-resultant Qtc of the system you had for reference was?

> What brings up the 50 to 60 Hz range is due to too much baffle step<

I’m not really following you, Pete. I thought any desirable baffle-step compensation was almost entirely a function of system placement and probably had less effect on “flatness” than would room modes. Are you suggesting that if the speaker were of different proportions the optimum baffle-step compensation for any installation could have been minimized?

I’m the thing around here that’s most baffled. I still need you to explain how you can conclude that AR purposefully raised output at 50-60Hz in order to make the speakers thumpier for consumer delight. I wouldn’t have thought that the necessary correction to baffle-step problems would have changed one iota from the 3a to the 11.

I’m not doubting, at all, that some improvement could have been made. But the way you said it conveyed to me that AR purposefully put a hump in the 50-60Hz output to gain consumer acceptance. I simply don’t see how they could have done this on purpose with the driver, crossover, and cabinet in question; **unless** as I previously said, they raised the system’s Fc and you were hearing and measuring the cabinet’s resonance, which as you’ve explained can’t possibly be the case with the pair you have/had.

If there’s another way to do it using the same air-wound, 17gauge, 2.85mH coil and the same cabinet and damping, without moving Fc, while increasing Qtc, I’d really like to know what it is if you have time to explain it. (because I want to avoid accidently causing it)

I can’t work-through soundminded’s calculus equation because I have no idea what the drag-coefficient of 20 ounces of *that* fiberglass is. I doubt seriously that AR engineers knew, either, except as a measurement taken experimentally after-the-fact. How springy is air in a 1.48 cu ft box at standard temp, humidity, and pressure? I guess you can experiment to that, too, but wouldn’t the speakers sound different in Denver and New Orleans if that were a major factor? And how about when a front comes through. I think we’re starting down the path of increasingly insignificant variables.

>You mention me wanting to hear something, actually this AR-11 sounds probably much better than most 11s as far as deep bass goes, because I used a highly compliant edge, and I believe that the spider is beyond broken in given the 15 Hz Fs.<

Well, the target was 18Hz, at least the design was based on 18Hz Fs. So, yeah, your pair probably got a few Hz lower at the -3db point, but I just don’t see Fc changing more than twice the difference in Fs. Can you?

If you were forcing the Fc down to 39Hz, I’m still blinded to what produced a hump at 50-60Hz that wouldn’t have been true from the very first AR-1.

>I believe based on Fs and what others are finding that this driver is an anomaly providing an Fc of 44 Hz.<

I’m with you. Something’s not right. Sounds like Q was off. Damping?

>I stated that I was refoaming this driver as an experiment to measure an early driver. I would replace the spider if I was to call this a complete rebuild and then Fs/Fc would go up.<

Right. Got it.

So in your pair we’ve got a hump for consumer acceptance at 50-60Hz which in this cabinet is 6-16Hz above Fc of 44Hz which you believe is low for higher Fs newer drivers but remains 2Hz higher than the target of 42Hz for the original drivers with higher moving mass. Your happy with the Qtc of this cabinet, I suppose meaning it is .707 or below in actual practice, but there’s a designed-in hump at 50-60Hz put there by a problem with baffle-step to impress consumers which I don’t understand how changed from an original 3a.

You didn’t say any of that, like that. I’m just trying to get from point A to point B and understand how we can make the intermediate stops along the way.

I thought I understood this, at least primitively.

> My comments were based more on the amount of baffle step, what others are reporting about deep bass response (including you), that were confirmed by Unibox simulations. I did not have the higher moving mass driver or an AR-3a for comparison purposes.<

I really wish you had both that and the original tweeter designed for that speaker. One of the interesting questions this raises is that later 3a’s were being outfitted with the same woofers as 11s, I would think.

>I went back and checked I did not say anything about "dramatic loss of low bass" please be accurate in your statements.<

Right, you didn’t. Frank said his friend’s 11s had no . . . well to paraphrase, guts. I used too close to a superlative.

>Your guess about a 12" driver in a 1.48 cu ft box is incorrect, it would take some time to explain this.<

I’ve got all night to be taught. Sometime when you have the time I’d love the lesson. Because from my relative ignorance, it looks like AR did an amazing job of balancing trade-offs with that woofer in that box.

>Where did you have the mid and tweeter switches when you made these AR-11 comparisons? Given all the permutations of those switches the 11 can be many different speakers, there is no single 11 sound.<

Always all the way up. To demo an AR speaker against a JBL, Phase Linear, Pioneer, Koss, Cerwin-Vega (sometimes unavoidable), ESS, or other brands the store carried with those switches set any other way was to instantly have them dismissed. Didn’t matter if they were on the floor or at ear level. But, they were the only thing in the room that could convincingly reproduce piano.

>My comments are by todays standards since so many here state that these vintage systems are better than modern designs.<

I suppose that depends on what’s in your experiential universe of “modern designs.” I can see how someone can say and think it considering what’s readily available to hear these days. Can’t you? But I know that compared to the best, it isn’t really true. On the other hand, being a bass-head myself, it’s hard to find a speaker that does what these do without using a subwoofer.

>I'll have to comment on the rest of your post at a later time.<

I look forward to it.

Bret

[Guest Bret]

>it is surprising (to me) that they haven't mostly replaced passive crossover networks yet.<

I think we're just waiting for digital amps to get to the point of delivering more current.

It makes perfect sense for speakers to evolve to being active, all of them. Imagine how much control the designer would have. . . notch-filters, and changing output to perfectly offset physics. . .

Bret

[Pete B]

Let me say that I strongly disagree with nearly everything in your post, we've discussed this before. Not interested in further discussion, please start a new thread for your "theories".

Pete B.

>The bass performance (for all loudspeaker systems) is

>accurately predicted by Newton's second law of motion as

>applied to forced oscillation. It is a staple of every

>college freshman engineering curriculum and is seen again and

>again in physics, calculus and sophomore mechanics-dynamics.

>It is not an approximation, it is an exact model which has

>been time proven for centuries now only yielding at near light

>speed to Einstein's modifications. It is a second order

>ordinary differential equation relating mass m, spring

>constant k, and viscosity b to displacement as a function of

>time.

>

>The solution of this equation for the system mechanical

>resonance frequency is;

>

>f©=(1/2PI)* SQ ROOT[(k/m)-(b/2m)squared]

>

>k is the springiness of the suspension, hopefully mostly air

>trapped in the box in an acoustic suspension system. This is

>entirely linear as defined by the gas laws P1*V1=P2*V2 where P

>and V are volumes before and after slight compression or

>rarifaction. F, is the restoring force and is the

>pressure/cone area. k is delta F/delta x where x is the

>displacement.

>

>m is the total moving mass

>

>b is the velocity related factor which is the aerodynamic drag

>the speaker must overcome to push or pull air between the

>fibers of the stuffing. The amount, nature, and distribution

>of the stuffing is therefore critical to tuning the system.

>Any change in the density, packing, amount, or type of fibers

>or how it is distributed in the box will change b and

>therefore the tuning of the mechanical system. If b were to

>decrease by say compressing the stuffing making it impossible

>to pull air through it, it would increase f©. It would also

>remove the restoring force of any air trapped behind the

>fibers reducing the effective volume of the remaining air

>further increasing f©. m, k, and b must be adjusted for

>critical damping of .707 to obtain the flattest possible

>frequency response. Less damping (higher Q) will result in a

>peak at resonance which translates into spurious damped

>oscillation at the resonance frequency every time the system

>is excited at any frequency. More damping meaning a Q of less

>than .707 will decrease the low frequency response but at

>least there will be no spurious resonance. (This is how the

>suspension of a car is tuned and the solution of many other

>mechanical problems.)

>

>One thing you can clearly see is the disadvantage of any

>ported system because k resulting from air pressure in such a

>system is strongly a function of frequency, being very low at

>resonance frequency of the air column and multiples of

>resonance frequency and very high at the midpoints between

>them. That is why ported systems have an inherently high Q

>which must be overcome if the system is to have anything like

>a smooth bass frequency response.

>

>The problem for the upper end of the frequency response of any

>driver including the AR woofer is that at some point as

>frequency increases, its inertial mass becomes too great to

>respond to forced oscillation no matter how much exciting

>force is used. Reducing the mass increases this usable upper

>limit frequency but the penalty is increasing f© as well.

>

>It is interesting that the second order ordinary differential

>equation is exactly the same equation (with different letters)

>used to describe and solve the LCR problem for tuned

>electrical circuits. The system repsonse is the superposition

>of the electrical and mechanical resonances. It doesn't

>matter either whether the elecrical resonances are at the

>loudspeaker output level in a passive crossover network or at

>the low signal preamplifier level controlled by equalizers or

>active crossover networks. Given how inexpensive these lower

>level circuits and audio amplifiers have become, it is

>surprising (to me) that they haven't mostly replaced passive

>crossover networks yet.

[ar_pro]

Great point! And there's also the potential for strict adherence to the loudspeaker's *design standard* through the active system's adjustment capabilites, too. An access port could provide immediate PC analysis to the owner or manufacturer, and with the right measurement system, make the loudspeaker eminently adaptable to different listening environments - much more so than the fixed-component crossover systems (relentlessly-changing their value in capacitance and ESR, according to some ;-) ) to which we have all become so accustomed.

Aside from the obvious technical obstacles, there would be some serious psychological barriers to be dealt with - a whole lot of audiophiles are married to their amplifiers, and many are possibly more committed to them than they are their spouses. Talk about an uphill fight!

[Pete B]

How about a new thread on active and/or digital crossovers, rather than take this one off topic?

Pete B.

>>it is surprising (to me) that they haven't mostly replaced

>passive crossover networks yet.<

>

>I think we're just waiting for digital amps to get to the

>point of delivering more current.

>

>It makes perfect sense for speakers to evolve to being active,

>all of them. Imagine how much control the designer would

>have. . . notch-filters, and changing output to perfectly

>offset physics. . .

>

>Bret

[Guest linedrive]

Has any one here ever replaced an AR crossover with an electronic one, or done any experiments with active crossovers and their AR speakers?

Ted

[soundminded]

>Let me say that I strongly disagree with nearly everything in

>your post, we've discussed this before. Not interested in

>further discussion, please start a new thread for your

>"theories".

>

>Pete B.

Newton's laws of motion are not theories, they have been demonstrated so many times by so many people for so long that they are accepted by scientists and engineers universally as facts. That's why they are called laws. Learn them and how to apply them and they will be your friends. Ignore them, challenge them, fight them and you will lose every single time. Don't argue with me, I didn't discover them, argue with Sir Isaac...only under the circumstances, you will have to meet him at a seance. If you don't like it, find some other universe to live in, one where they don't apply. In this one, there is no escaping them.

[Pete B]

I took a good amount of time to try to answer your last questions, now you have new ones, partly based on misinterpreting what I said. Your question about crossover Q for example, there are many papers in the AES about crossover design and yes Q is one design parameter concerning 2nd order filters. Your questioning many different aspects of loudspeaker design I just can't cover all of them. Let me say that what I wrote is based on my engineering judgement, my particular system did not have a serious 50 to 60 Hz bump, it was better than most due to the loose suspension, but based on the engineering, and what others have stated such as Frank, this is how I see it by reading the design. Take it for what it's worth.

Pete B.

>>Your drawing some incorrect conclusions about my statements

>because you seem to be interpreting my use of Q as for only

>the woofer system Qtc. There are actually several that I'm

>referring to:

>Woofer closed box Q - Qtc<

>Woofer XO lowpass Qwlp - 2.85 mH/120uF (approximately 2nd

>order)

>Midrange XO highpass Qmhp - 40uF/.88 mH (approximately 2nd

>order)

>Tweeter XO highpass Qthp - 10uF/.101 mH (approximately 2nd

>order)

>Woofer XO contour for diffraction compensation, there is no Q

>associated with this.

>I've used these terms Qwlp, Qmhp, Qthp to be more specific,

>they're not normally used in crossover design.<

>

>Why on earth would you refer to the “Q” of crossover elements?

> I don’t understand. What is to be gained, or learned, from

>that with the possible exception of sensitivity to capacitor

>ESR, and I’d have never thought of that. Is that how you

>determined the sensitivity to cap ESR? That’s what I’m taking

>away from this discussion, so if I’m wrong, please straighten

>me out.

>

>I guess I’m confused because in your earlier post about

>rebuilding the woofers you did mention getting a Qtc of > 0.9

>.

>

>>One reason that I was surprised to find such a difference in

>moving mass is that the advertised Fc for the 3a and 11 did

>not change significantly, yet it seems that the mass did and

>with the same enclosure volume (and high compliance) this does

>not make sense.<

>

>I guess I don’t understand, again. In a closed system the Fs

>and volume of the enclosure are the two greatest and most

>elemental driving forces for Fc. You can “goof around with

>it” a little by also fiddling with damping and inductor

>resistence, but not much if you have a target Q to meet.

>Without getting into true exotica I don’t see how you could

>get this performance out of a box much smaller than this. I’m

>talking 35Hz-1kHz here, not sub-woofer performance.

>

>> I have many questions, is the mass measured on your drivers

>correct, had they been worked on before, what was the intended

>moving mass for the early 3a, and the 11? It will take time to

>collect all this data.<

>

>I think Ken’s given most of it to us in the DUTs done that I

>can’t find a link-to. Remember there was one speaker with a

>new cone, spider, the whole sheebang from a 9. But we had at

>least two from 10pi’s that were original to the speakers and I

>maintain that we had at least two unchanged from 9s. As for

>the measurements being correct, that would have been done by

>Ken’s colleagues on their expensive equipment. I would think

>it would be right, but I certainly can’t swear it.

>

>>My best read of this situation is that the early 11 was

>prototyped with the higher moving mass driver, thus the same

>advertised Fc as the 3a. All 11's might not be the same, if

>people would read the part number on the back of the cone and

>tell us we would have a better idea. Note that Frank in his

>post above confirms my point about deep bass. I would not be

>surprised if they all shipped with the lighter cone driver but

>again no one is offering input to confirm this.<

>

>I don’t have the information you need, and I don’t have access

>to it.

>

>>My comments about high Q were mainly referring to what I

>called above Qwlp, Qmhp, Qthp and I really do not have a

>problem with the Qtc of the system.<

>

>What do you think the end-resultant Qtc of the system you had

>for reference was?

>

>> What brings up the 50 to 60 Hz range is due to too much

>baffle step<

>

>I’m not really following you, Pete. I thought any desirable

>baffle-step compensation was almost entirely a function of

>system placement and probably had less effect on “flatness”

>than would room modes. Are you suggesting that if the speaker

>were of different proportions the optimum baffle-step

>compensation for any installation could have been minimized?

>

>

>I’m the thing around here that’s most baffled. I still need

>you to explain how you can conclude that AR purposefully

>raised output at 50-60Hz in order to make the speakers

>thumpier for consumer delight. I wouldn’t have thought that

>the necessary correction to baffle-step problems would have

>changed one iota from the 3a to the 11.

>

>I’m not doubting, at all, that some improvement could have

>been made. But the way you said it conveyed to me that AR

>purposefully put a hump in the 50-60Hz output to gain consumer

>acceptance. I simply don’t see how they could have done this

>on purpose with the driver, crossover, and cabinet in

>question; **unless** as I previously said, they raised the

>system’s Fc and you were hearing and measuring the cabinet’s

>resonance, which as you’ve explained can’t possibly be the

>case with the pair you have/had.

>

>If there’s another way to do it using the same air-wound,

>17gauge, 2.85mH coil and the same cabinet and damping, without

>moving Fc, while increasing Qtc, I’d really like to know what

>it is if you have time to explain it. (because I want to

>avoid accidently causing it)

>

>I can’t work-through soundminded’s calculus equation because I

>have no idea what the drag-coefficient of 20 ounces of *that*

>fiberglass is. I doubt seriously that AR engineers knew,

>either, except as a measurement taken experimentally

>after-the-fact. How springy is air in a 1.48 cu ft box at

>standard temp, humidity, and pressure? I guess you can

>experiment to that, too, but wouldn’t the speakers sound

>different in Denver and New Orleans if that were a major

>factor? And how about when a front comes through. I think

>we’re starting down the path of increasingly insignificant

>variables.

>

>>You mention me wanting to hear something, actually this AR-11

>sounds probably much better than most 11s as far as deep bass

>goes, because I used a highly compliant edge, and I believe

>that the spider is beyond broken in given the 15 Hz Fs.<

>

>Well, the target was 18Hz, at least the design was based on

>18Hz Fs. So, yeah, your pair probably got a few Hz lower at

>the -3db point, but I just don’t see Fc changing more than

>twice the difference in Fs. Can you?

>

>If you were forcing the Fc down to 39Hz, I’m still blinded to

>what produced a hump at 50-60Hz that wouldn’t have been true

>from the very first AR-1.

>

> >I believe based on Fs and what others are finding that this

>driver is an anomaly providing an Fc of 44 Hz.<

>

>I’m with you. Something’s not right. Sounds like Q was off.

>Damping?

>

>>I stated that I was refoaming this driver as an experiment to

>measure an early driver. I would replace the spider if I was

>to call this a complete rebuild and then Fs/Fc would go up.<

>

>Right. Got it.

>

>So in your pair we’ve got a hump for consumer acceptance at

>50-60Hz which in this cabinet is 6-16Hz above Fc of 44Hz which

>you believe is low for higher Fs newer drivers but remains 2Hz

>higher than the target of 42Hz for the original drivers with

>higher moving mass. Your happy with the Qtc of this cabinet,

>I suppose meaning it is .707 or below in actual practice, but

>there’s a designed-in hump at 50-60Hz put there by a problem

>with baffle-step to impress consumers which I don’t understand

>how changed from an original 3a.

>

>You didn’t say any of that, like that. I’m just trying to get

>from point A to point B and understand how we can make the

>intermediate stops along the way.

>

>I thought I understood this, at least primitively.

>

>> My comments were based more on the amount of baffle step,

>what others are reporting about deep bass response (including

>you), that were confirmed by Unibox simulations. I did not

>have the higher moving mass driver or an AR-3a for comparison

>purposes.<

>

>I really wish you had both that and the original tweeter

>designed for that speaker. One of the interesting questions

>this raises is that later 3a’s were being outfitted with the

>same woofers as 11s, I would think.

>

>>I went back and checked I did not say anything about

>"dramatic loss of low bass" please be accurate in your

>statements.<

>

>Right, you didn’t. Frank said his friend’s 11s had no . . .

>well to paraphrase, guts. I used too close to a superlative.

>

>>Your guess about a 12" driver in a 1.48 cu ft box is

>incorrect, it would take some time to explain this.<

>

>I’ve got all night to be taught. Sometime when you have the

>time I’d love the lesson. Because from my relative ignorance,

>it looks like AR did an amazing job of balancing trade-offs

>with that woofer in that box.

>

>>Where did you have the mid and tweeter switches when you made

>these AR-11 comparisons? Given all the permutations of those

>switches the 11 can be many different speakers, there is no

>single 11 sound.<

>

>Always all the way up. To demo an AR speaker against a JBL,

>Phase Linear, Pioneer, Koss, Cerwin-Vega (sometimes

>unavoidable), ESS, or other brands the store carried with

>those switches set any other way was to instantly have them

>dismissed. Didn’t matter if they were on the floor or at ear

>level. But, they were the only thing in the room that could

>convincingly reproduce piano.

>

>>My comments are by todays standards since so many here state

>that these vintage systems are better than modern designs.<

>

>I suppose that depends on what’s in your experiential universe

>of “modern designs.” I can see how someone can say and think

>it considering what’s readily available to hear these days.

>Can’t you? But I know that compared to the best, it isn’t

>really true. On the other hand, being a bass-head myself,

>it’s hard to find a speaker that does what these do without

>using a subwoofer.

>

>>I'll have to comment on the rest of your post at a later

>time.<

>

>I look forward to it.

>

>Bret