Jump to content

What Do Thanksgiving and AR Have In Common?


kkantor

Recommended Posts

>http://auralization.blogspot.com/

"The Effect of Stuffing Type on Cabinet Resonance

John F. O’Hanlon

with considerable help from: David Blackstock

Herb Kuntz

Roy Champagne

Ken Kantor

Impedance measurements made by Roy Champagne (AR-3a) and the author (AR-4x and –3a) showed that cabinets stuffed with polyester fibers could not achieve as low a cabinet resonant frequency fc, as when the cabinet was stuffed with fiberglass (OEM stuffing). Experimental measurements are given below for the AR-4x.

[sorry I couldn't paste the graph here}

The fiberglass-filled cabinet has an fc of 54.5 Hz, whereas fc in the FiberLoft-filled cabinet was 58–60 Hz. Roy Champagne obtained similar measurements for fc in the AR-3a that showed an increase of about 2–3 Hz when fiberglass was replaced by Acousta-Stuf or other polyester stuffing. Ken Kantor commented that he has never been able to find a “drop-in” replacement for fiberglass in a speaker designed for its use, except for one Danish fiber that came close." ......

I excerpted the above from Ken's blog because it has the clearest conclusion of this most comprehensive stuffing study I have seen to date. Correct me if I'm wrong, but I read from this that fiberglass lowers Fc about 4-5 hz when compared to polyester fibers.

If that's all the fiberglass is doing for the reproduction of the lower bass range I have a problem with this conclusion. If, on the other hand there are other, unmeasurable sonic benefits like improved definition and articulation, then fine. Fiberglass is the preferred way to go.

That 4-5 hz seems pretty small in the grand scheme of the full audio spectrum in a dynamic, musical environment. I am a retired, ASQ certified quality enginner with some knowledge of statistics and am a bit sceptical about the true validity of the claimed improvement in the FC range reported in Mr. O'Hanlon's study. Is the Fc change truly siginificant? I have to ask:

1) How many data points led to the conclusion? 1, 5, 10, 20, 30?

2) What is the repeatability of the Fc measurement under constant conditions? I suspect if the st'd deviation of a repeated test is in the 2,3 or 4 hz range, then the claimed improvement may be absorbed in the normal 'background noise' of the test itself.

3) What is the confidence interval around the reported improvement?

Perhaps Ken can enlighten us a bit further regarding the testing protocol and any statistics behind the conclusion drawn.

IMO the best way to judge bass reproduction is by playing bass-rich music thru the woofer only. Only then can one truly listen to what the woofer is capable of using the most accurate instrument we have - our ears. I welcome any comments from posters regarding their experience with bass reproduction of AR's stuffed with fiberglass vs. PET fibers....

Remember, it's all about the music.

Carl

Link to comment
Share on other sites

Has a double blinfold test on fiberglass/polyester or acosta stuf ever been done?

1) Take 2 pair of AR3as and stuff poly & fiberglass.

2) Run a double blindfold test.

3) Switch the stuffings.

Re-run the tests.

The scientific method is by far the most accurate. However, as listeners our ears are where we ultimately set the bar.

Why do people avoid or ignore blindfold testing?

Take interconnects for example. Every blindfold test confirms high-end interconnects are pretty much snake oil and yet they flourish. Science also confirms "magic" interconnects are just that - magic.

One more thing. After 30 years I would think the deterioration in the drivers would amount to much more of a dfferential from the original sound than a change of stuffing. How about driver restoration as a serious topic and eventual project as a means for restoring the AR3/AR3a to its original glory.

Link to comment
Share on other sites

>Has a double blinfold test on fiberglass/polyester or acosta

>stuf ever been done?

>

>1) Take 2 pair of AR3as and stuff poly & fiberglass.

>

>2) Run a double blindfold test.

>

>3) Switch the stuffings.

>

>Re-run the tests.

>

>The scientific method is by far the most accurate. However, as

>listeners our ears are where we ultimately set the bar.

>

>Why do people avoid or ignore blindfold testing?

>

>Take interconnects for example. Every blindfold test confirms

>high-end interconnects are pretty much snake oil and yet they

>flourish. Science also confirms "magic"

>interconnects are just that - magic.

>

>One more thing. After 30 years I would think the deterioration

>in the drivers would amount to much more of a dfferential from

>the original sound than a change of stuffing. How about driver

>restoration as a serious topic and eventual project as a means

>for restoring the AR3/AR3a to its original glory.

There are arguments both for and against double blind testing. Much has been written in the audio literature regarding this. I also wonder why most, if not all, modern speakers do NOT have fiberglass stuffing. If it was that much better than PET, it would seem to me the loudspeaker industry would have found a way to continue using it in spite of the worker safety aspects.

Finally, it should be left up to the individual to judge what's best for their listening experience. Subtle changes like those aluded to in the above study reside in the realm of aesthetic nuance, wherein sonic nervana lies in the 'ear' of the beholder.

I think you make a valid point regarding driver deterioriation. Surely the compliance of surrounds and spiders changes over time. I'm not sure on this and don't know if there are any valid records of original AR manufacturing specifications left around. Perhaps Tom Tyson could comment on this. I do know that AB Tech has drivers they claim were manufactured in the far east to AR specifications. They may be the closest thing we currently have for replacing worn, old AR drivers as a restoration option.

Remember, it's all about the music

Carl

Carl's Custom Loudspeakers

Link to comment
Share on other sites

With Ken having championed this paper and Tom, Roy, Steve, and Bret having wholeheartedly endorsed it and presumably completely understood it and agreed with it, I feel I am lucky to have at least five people here to explain it to me. I have to confess that the more I read it, the more questions I have understanding it and the more problems I have agreeing with what little I do understand. So if anyone will be so kind as to explain this as though to someone who is a beginner, I'd appreciate your time and patience. Apparantly it's an important aspect of AS design or it wouldn't have been featured so prominently in Ken's blog. If you want to take this to a different forum since it will probably be long or offline altogether, I understand and have no problem with it. It may be boring to a lot of others here who also find it too elementary to bother with.

To start with, I've gotten through the first two sentences and accept them as empirical fact. Although it is a generalization, I assume we are talking about the same weight (not volume) of stuffing in identical cabinets and all other conditions are identical. So far so good. Now what does this have to do with input (I assume electrical speaker impedence) reactance as stated in the third sentence? And just dipping my toe in the second paragraph, what is meant by acoustic impedence? Clearly there is an explicit resistive and reactive component here. I understand electrical impedence being the sum of a resistive component and a component displaced 90 degrees in phase due to a time or phase delay as the result of magnetic storage and release of energy in an inductive circuit and electrostatic storage and release of charge as the result of a capacitive circuit but I don't follow what this means in a physical sense with a differences of air pressure and a complex impedence to air flow. Could someone please explain this to me. (BTW, you can see why I have a problem with the thermodynamic model used to analyze AS design since it would seem to me this is more suitably described by a fluid dynamics model which relates to Newton's laws of motion.)

Link to comment
Share on other sites

>>Has a double blinfold test on fiberglass/polyester or

>acosta

>>stuf ever been done?

>>

>>1) Take 2 pair of AR3as and stuff poly & fiberglass.

>>

>>2) Run a double blindfold test.

>>

>>3) Switch the stuffings.

>>

>>Re-run the tests.

>>

>>The scientific method is by far the most accurate.

I wish to voice my opinion in this discussion, Carl.

Using the AR-3A's only as a test medium.

The AR-3A has been around long enough for various examples to have shown up and be disected.

Perhaps the AR-1W would be the more appropriate speaker, with no mid and tweeter interference and an especially longer frequency path to control.

I am of the opinion, that the fibreglass in the enclosures was used to deaden cabinet resonances and rear driver resonances rather than be a restoring force.

Fibreglass is glass fibres, not tiny containers such as mini balloons.

The air inside the enclosure is at room pressure, not greater than or less than, when the woofer is at idle.

There is a slight infiltration around the dome and possibly the frame cabinet seal is not perfect, but for our purposes we can say a sealed system, but not 100% sealed, maybe 99.999%.

When the woofer receives a signal of controlled input and the cone recedes into the drivers frame, there is now some compression supporting the cone and breaking it, so to speak, acoustic suspension.

The fibreglas only sits there, motionless, inert, but touching the cabinet walls, crossover, rear of the woofers frame, the wires, and lastly, just occupying space.

Newly developed fibres such as, perhaps, Hollow-fill, I am not certain as to the constuction of fibres, but they may be like spaghetti or a drinking straw, but with partitions, or isolated areas controlled in manufacturing.

There would not be small, but authentic mini ballons that can be contracted by the back pressure of the woofer, and would re-expand upon forward movement of the woofer.

An interesting test might be to scientifically add 1 ounce pieces of fibreglass of a known quality and 1 ounce pieces of synthetics as see the differences.

I am afraid our hearing would not be a valid test here, but of course, of interest to hear as well.

For a AR-1W, the entire woofer range would be affected rather than the more restricted AR-3A 575hz cutoff.

If I remember right, the speakers originally had a fine short fibre fibreglas rather than the longer house fibreglass insulation.

When I used to change really old electric water heaters, they used the shorter fibreglass.

As always, I may be right or wrong, unfortunately I have to accept my opinion as my own, I can't blame it on my medication. lol

Maybe I shouldn't wear my AR 10" woofer basket on my head, after following my main man, Frank's, example. lol

Now that I have vented, I'm taking my grand daughter and her girl friend to dinner and to see X-men 3.

I am looking forward to your opinions, fellow members.

>However, as

>>listeners our ears are where we ultimately set the bar.

>>

>>Why do people avoid or ignore blindfold testing?

>>

>>Take interconnects for example. Every blindfold test

>confirms

>>high-end interconnects are pretty much snake oil and yet

>they

>>flourish. Science also confirms "magic"

>>interconnects are just that - magic.

>>

>>One more thing. After 30 years I would think the

>deterioration

>>in the drivers would amount to much more of a dfferential

>from

>>the original sound than a change of stuffing. How about

>driver

>>restoration as a serious topic and eventual project as a

>means

>>for restoring the AR3/AR3a to its original glory.

>

>

>There are arguments both for and against double blind testing.

>Much has been written in the audio literature regarding this.

>I also wonder why most, if not all, modern speakers do NOT

>have fiberglass stuffing. If it was that much better than PET,

>it would seem to me the loudspeaker industry would have found

>a way to continue using it in spite of the worker safety

>aspects.

>Finally, it should be left up to the individual to judge

>what's best for their listening experience. Subtle changes

>like those aluded to in the above study reside in the realm of

>aesthetic nuance, wherein sonic nervana lies in the 'ear' of

>the beholder.

>

>I think you make a valid point regarding driver

>deterioriation. Surely the compliance of surrounds and spiders

>changes over time. I'm not sure on this and don't know if

>there are any valid records of original AR manufacturing

>specifications left around. Perhaps Tom Tyson could comment on

>this. I do know that AB Tech has drivers they claim were

>manufactured in the far east to AR specifications. They may be

>the closest thing we currently have for replacing worn, old AR

>drivers as a restoration option.

>

>Remember, it's all about the music

>

>Carl

>Carl's Custom Loudspeakers

Link to comment
Share on other sites

The value of this work, such as it may be, is in developing both a qualitative and quantitative understanding of the acoustics of cabinet stuffing. This knowledge may then be applied to restoration, new product design, modelling, or moving on to greener pastures, at the discretion of the reader. Frankly, I have little personal interest in restoring vintage speakers. But, I do have a great curiosity about how speaker systems operate.

Some box/driver combinations are very sensitive to variations in cabinet size, stuffing, etc, while others are not. Clearly, the magnitude of the final effect that the choice of stuffing will have depends on many factors. For example, in an "infinite baffle," one might expect almost no effect from the stuffing. In a completely

"acoustically suspended" design, the opposite might be true.

I have not done statistical tests on the a large population of 3a's. However, I have done this in the course of quality control on the production line of various NHT products, that were migrated from fiberglass to polyester stuffing. Aside from theoretical considerations, the results recently stated are consistent with my real-world experiences with significant sample sizes.

Link to comment
Share on other sites

First of all, the point of the present work is analytic in nature. Few, if any, conclusions are drawn about audibility. Only one sentence about possible sonic consequences is even tentatively posited. Reports of listening impressions between types of stuffing are purely anecdotal, for the purposes of this discussion.

As big a fan of "double blind" protocols as I am, it is hard for me to grasp their role in such a purely physical investigation. In terms of your proposed methodology, I would suggest that an experimental protocol which includes both a control, and administrator blindness, would be more rigorous.

Link to comment
Share on other sites

>The value of this work, such as it may be, is in developing

>both a qualitative and quantitative understanding of the

>acoustics of cabinet stuffing. This knowledge may then be

>applied to restoration, new product design, modelling, or

>moving on to greener pastures, at the discretion of the

>reader. Frankly, I have little personal interest in restoring

>vintage speakers. But, I do have a great curiosity about how

>speaker systems operate.

>

>Some box/driver combinations are very sensitive to variations

>in cabinet size, stuffing, etc, while others are not.

I wonder why. As you note below, there are a myriad of variables that could affect these interactions.

>Clearly, the magnitude of the final effect that the choice of

>stuffing will have depends on many factors. For example, in

>an "infinite baffle," one might expect almost no

>effect from the stuffing. In a completely

>"acoustically suspended" design, the opposite might

>be true.

There may be some 'low lying fruit to pick here'. Perhaps sponsoring a study by a graduate student from a well recognized college or university versed in audio technology?

Transmision line subs are also worthy of study. There has been some studies done in this area already. Perhaps not a comprehensive as your AS study, however.

>

>I have not done statistical tests on the a large population of

>3a's. However, I have done this in the course of quality

>control on the production line of various NHT products, that

>were migrated from fiberglass to polyester stuffing. Aside

>from theoretical considerations, the results recently stated

>are consistent with my real-world experiences with significant

>sample sizes.

>

>

Remember, it's all about the music

Carl

Carl's Custom Loudspeakers

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...