AR3a improved Xover capacitors

[Howard Ferstler]

All components the signal passes through will influence sound, I would (and have on my own) replace R2 (1 Ohm) to a MOX (Metal OXide) type resistor, sounds much better to my ear, than the standard wire wound used by AR.

http://www.classicspeakerpages.net/IP.Boar...-1204644282.jpg

If you use the B posistion on the switch, I would replace R4 as well, but the A position sounds far better to my ears.

BRgds Klaus

Obviously, any components other than wire will have some kind of impact on the sound of a crossover network. After all, it is the job of such components to do just that. Assuming equal resistance values, no resistor will sound different from any other. There is no esoteric characteristic that would make one sound different from another. What might matter is power-dissipation ability, however. I suggest using at least ten-watt rated versions for AR speakers of this vintage.

Howard Ferstler

[KlausDK]

Assuming equal resistance values, no resistor will sound different from any other.

This is your opinion, which you are of course entitled to have, but it is not a nature law. My ears has told me different.

I am not going into a lengthy discussion about this, just pointing out, that this is a matter of opinion.

BRgds Klaus

[KlausDK]

Well, opinion is not the same as fact.

Exactly my point, how nice we agree on this.

BRgds Klaus

[kkantor]

This is your opinion, which you are of course entitled to have, but it is not a nature law. My ears has told me different.

I am not going into a lengthy discussion about this, just pointing out, that this is a matter of opinion.

BRgds Klaus

Such things are not only matters of subjective customer opinion, they are important to product designers and brands looking for market advantages and sonic improvements. They are important to agencies who are accountable for monitoring "deceptive practices" in advertising. They are interesting to scientists looking for knowledge. They are important to suppliers looking to improve their parts.

Thousands of careful studies have been conducted over the years, run by different groups with sometimes opposite agendas. I'm talking about thoughtfully arranged experiments, done with real music, without stress, etc. Bottom line: not one test I know of has ever shown that resistors "sound different," aside from limit-case problems of thermal noise and power handling. In addition to these results, there is no know theoretical reason why resistors could or should sound different.

So, while nobody can stop you from holding your opinion, and repeatedly proving it yourself in your own system, it >is< an established fact, a fact that no experiment or study has managed to overturn. Many have tried. You might want to consider revising your opinion based on expert scientific work in the area.

-k

PS- evolutionarily, or for whatever reason you believe, the human ear has evolved with certain limits. We need to hear sounds close to use. Maybe miles away. We need to hear sounds occurring in the range of the creatures and events of importance to us. Awareness would be unbearable if we heard all the conversations going on in the next town. Or all the pounding of ant's feet on the ground. Or the echolocation of bats. Etc. Soooo.... our hearing has >thresholds<. Our hearing is not infinite. It has clear and well-known limits, and these may be applied to make components that cannot be heard when inserted in a circuit.

[speaker dave]

I've always thought of this as the "bicycle component mentality".

Back when I was really into bicycles (10 speed road bikes) they all had esentially the same design. Tied to the human anatomy, bikes had evolved for a century to the point where most bikes looked the same. Nobody had a different basic design that was superior to the others. Where they did vary was in the quality of the components. If you had the money and wanted the best you got Campagnolo. The quality and workmanship was great but the parts, cranksets, brakes, wheelsets, etc. were basically interchangeable. Your frame might be better crafted from a superior tubing grade but the dimensions and geometry would be well established.

I think audiophiles look at it this way and assume the difference between so-so, pretty good and great gear must be that the parts are more expensive and therefore the most expensive parts will give the best sound no matter what the system design. In truth there are good and bad designs. And good designs can give superior performance with just adequate components. Henry Kloss was the master of making great products that were built from budget parts. Conversely, today high-end-audio is littered with bad designs made with outrageously expensive components.

Thats not my definition of good engineering.

David

[KlausDK]

Thousands of careful studies have been conducted over the years, run by different groups with sometimes opposite agendas. I'm talking about thoughtfully arranged experiments, done with real music, without stress, etc. Bottom line: not one test I know of has ever shown that resistors "sound different," aside from limit-case problems of thermal noise and power handling. In addition to these results, there is no know theoretical reason why resistors could or should sound different.

Hi Ken

I have a set of home made 2 way speakers, where I wanted to correct the balance between the (Dynaco 28AF) tweeter and woofer (elevate the tweeter a few dB). I ordered new resistors to replace the ones in the speaker, installed them and I could not stand the sound, it was harsh, edgy and unbearable to my ears. Totally different speakers. I have had these speakers as my main speakers more or less since I build them in 1982, so I am pretty familiar with the sound. I put the old resistors back in one speaker, and sound quality was back to normal in that speaker. Ordered new set of resistors with the new values from audiophile source, MOX resistors as it turned out, not expensive 1½ $ each, and now the sound quality was as I was used to, with slightly elevated tweeter level off course.

I have a similar experience with capacitors, however, not from speakers but from an amplifier. I had a Swedish Sentec PA8 amplifier, that I've had since the late 70'ies. A few years back, I decided to replace all the electrolytic caps, since they were now 30 years plus. I carefully ordered the different caps from different sources, but the input capacitor, a bipolar cap of 10 uF I took what they had in the shop, which turned out to be a Monacor electrolytic cap, meant for speakers, but it was 10 uF. Same experience as with my speakers, sound was unbearable ugly, harsh torture to my ears, again I have listened to this amplifier 30 years plus, so I know what it sounds like. I got some film caps, I think it was MKC caps, and the amplifier was useable again.

I know this is not scientific, but I have no doubts that both resistors and capacitors in these two cases influenced the sound quality in a negative way.

BRgds Klaus

[kkantor]

Hi Ken

I have a set of home made 2 way speakers, where I wanted to correct the balance between the (Dynaco 28AF) tweeter and woofer (elevate the tweeter a few dB). I ordered new resistors to replace the ones in the speaker, installed them and I could not stand the sound, it was harsh, edgy and unbearable to my ears. Totally different speakers. I have had these speakers as my main speakers more or less since I build them in 1982, so I am pretty familiar with the sound. I put the old resistors back in one speaker, and sound quality was back to normal in that speaker. Ordered new set of resistors with the new values from audiophile source, MOX resistors as it turned out, not expensive 1½ $ each, and now the sound quality was as I was used to, with slightly elevated tweeter level off course.

I have a similar experience with capacitors, however, not from speakers but from an amplifier. I had a Swedish Sentec PA8 amplifier, that I've had since the late 70'ies. A few years back, I decided to replace all the electrolytic caps, since they were now 30 years plus. I carefully ordered the different caps from different sources, but the input capacitor, a bipolar cap of 10 uF I took what they had in the shop, which turned out to be a Monacor electrolytic cap, meant for speakers, but it was 10 uF. Same experience as with my speakers, sound was unbearable ugly, harsh torture to my ears, again I have listened to this amplifier 30 years plus, so I know what it sounds like. I got some film caps, I think it was MKC caps, and the amplifier was useable again.

I know this is not scientific, but I have no doubts that both resistors and capacitors in these two cases influenced the sound quality in a negative way.

BRgds Klaus

OK, thanks. I have had similar experiences myself. And, after more than 30 years as a designer with public contact, I have heard probably a few thousand. The issue that I have is that every time I have tried to track down what I heard, or what anyone else heard, one of these things happened:

1- It turned out that there were differences between the parts in terms of the actual resistance or capacitance. When this was matched, the difference went away.

or

2- Some other thing was being changed during the process, which was not discovered until later.

or

3- The hearing difference went away when the listener did not know which was which.

People see many things with "there own eyes" that are probably not true. There is a huge pile of money to be made by any company that can demonstrate that their resistor or capacitor really sounds better. Nobody can do this. They only publish advertising papers and print the words of happy customers or reviewers. I cannot believe that after all these years, since about 1950 when the idea first started, nobody has been able to put on a demonstration or give a scientific study.

-k

[KlausDK]

Hi Ken

It turned out that there were differences between the parts in terms of the actual resistance or capacitance. When this was matched, the difference went away.

I don't think that even a big variance in the amplifier input caps (lets say one is 11 uF and the other is 9 uF), will account for any differences in the mid and high range frequencies, this will may be cause a difference in the cut off frequency below 10 Hz, hardly an audible effect.

People see many things with "there own eyes" that are probably not true. There is a huge pile of money to be made by any company that can demonstrate that their resistor or capacitor really sounds better. Nobody can do this. They only publish advertising papers and print the words of happy customers or reviewers. I cannot believe that after all these years, since about 1950 when the idea first started, nobody has been able to put on a demonstration or give a scientific study.

I do not have your experience with things, neither do I have access to the same kind and amount of information, but I have seen and read many papers that claims that it is possible to detect differences between different components, also based on blinded listening tests.

As far as I remember, P.S. Dodds et al, has published scientific papers, where they were able to demonstrate audible differences between capacitors, they also presented at AES May 2008, a paper called "Audio Capacitors. Myth or Reality?"

BRgds Klaus

[genek]

It turned out that there were differences between the parts in terms of the actual resistance or capacitance. When this was matched, the difference went away.

Perhaps a tolerance stackup issue. Your components are all within spec, but lean toward the high or low end of range rather than the center, and if enough of them in a chain lean the same way, it skews the final result. The fix for this is to go to really tight tolerances and/or measure individual components and sort them according to which end of tolerance they favor, then alternate them to cause their variations to cancel each other out rather than become additive. PITA

[kkantor]

Thanks... I remember that paper. Rather controversial. Here is a quote from the paper:

The initial ABX tests involved a panel of fourteen

volunteer listeners comprising students and staff in

the University. The listeners generally found the tests

to be very difficult and quite tiring, and in general,

the results from the initial ABX tests gave no usable

outcome.

and here is the conclusion:

Taking the tests as a whole, the main developments

of the test methodology have been in the process of

listener selection, devising the training sessions, and

the relinquishing of control from the investigators to

the listener during the tests. Since the audio

differences between film capacitors are extremely

subtle, there is a crucial requirement that the tests

should cause the minimum of stress to the listener.

Situations where the listener is forced into making a

decision on the basis of unfamiliar programme

material within strict time limits are not conducive to

critical assessment of very subtle audio differences

I leave it to you to decide if this is good evidence.

-k

Hi Ken

I don't think that even a big variance in the amplifier input caps (lets say one is 11 uF and the other is 9 uF), will account for any differences in the mid and high range frequencies, this will may be cause a difference in the cut off frequency below 10 Hz, hardly an audible effect.

I do not have your experience with things, neither do I have access to the same kind and amount of information, but I have seen and read many papers that claims that it is possible to detect differences between different components, also based on blinded listening tests.

As far as I remember, P.S. Dodds et al, has published scientific papers, where they were able to demonstrate audible differences between capacitors, they also presented at AES May 2008, a paper called "Audio Capacitors. Myth or Reality?"

BRgds Klaus

[genek]

Situations where the listener is forced into making a

decision on the basis of unfamiliar programme

material within strict time limits are not conducive to

critical assessment of very subtle audio differences

I leave it to you to decide if this is good evidence.

From a "scientific method" POV, it's utter hogwash. "Unfamiliar programme material" and quick choices between randomized changes are the only way to discriminate between actual differences and listener expectations. The first thing a good tester should do is ask the listeners if any of the "programme material" is familiar to them and eliminate any that is. Also essential are some "do you prefer A or B?" decisions during testing in which no actual change in "programme material" is being made at all. It's just like taking an eye test in which the optometrist flips the diopters back and forth with lightning speed and occaisionally changes the line on the eye chart to ensure that you're not memorizing it during the test.

[kkantor]

From a "scientific method" POV, it's utter hogwash. "Unfamiliar programme material" and quick choices between randomized changes are the only way to discriminate between actual differences and listener expectations. The first thing a good tester should do is ask the listeners if any of the "programme material" is familiar to them and eliminate any that is. Also essential are some "do you prefer A or B?" decisions during testing in which no actual change in "programme material" is being made at all. It's just like taking an eye test in which the optometrist flips the diopters back and forth with lightning speed and occaisionally changes the line on the eye chart to ensure that you're not memorizing it during the test.

I totally agree.

It's also a case of "test until failure." Rather than defining the experiment a priori, then publishing a clear result, one can simply continue to move the goal posts, change the methods, pick the data, then exclaim, "Got It!" OK, if you got it, you can do again, right?

Anyway, I gotta really try and not go here. It's a belief system, I understand that. People do not like to be told that their primary senses are not 100% reliable. It's much more ego-syntonic to postulate weaknesses in the experiment or flaws in the whole rational method. I would be totally into accepting that idea if there was evidence.

http://www.kenkantor.com/publications/audi...ish_part_02.pdf

-k

[speaker dave]

I'm not sure why a quick choice with unfamiliar material should be a requirement. It seems that just offers a letout clause for those who are sceptical of ABX testing. ("I'm certain I could have heard a difference, given more time!") I would be impressed if anyone could hear a staticstically significant difference between, say, capacitors of different materials and the same measured values, even after a month of listening.

Those that believe would be quite willing to pay great sums for differences that required long listening sessions to familiar music. Even better if the differences are so subtle that only they can hear it. Proof of their superior acuity.

Of course, the differences are never subtle. How many time do we read that "I was blown away by the differnce that demagnetizing the album cover made to the speed, pace and dynamics of the LP" etc. etc.

David

[KlausDK]

I am not sure that I can follow here. Does that mean, that if a friend of mine has got a new amplifier or set of speakers, and claims that they are a great improvement to his previous gear, I will have to tell him, that unles he can refer to a double blind study that proves a difference (or has conducted one himself), I will have to assume that his new speakers or amplifier sounds the same as the old one(s)?

If not, where is the line where you can assume that this rule of sameness applies, where you have to assume that the use of different constructions using different materials will results in the same sound, unless proven otherwise) and where it will not nessecarily be so?

I must admit that my mind works the other way around, if things are different, I will have to assume that they sound different, unless someone has proven that it sounds the same, not the other way around.

To me it is a fact for me what I hear (and see). I know hearing it is different for other people, I have had guests that could not tell the difference between the sound form a getto blaster and my AR speakers, does that proove that there is no difference?

When asked for advice I always tell people to use their ears, if they can hear a difference between different components, they should choose what they like best and can afford, if they can hear no difference, go for the cheapest. To tell people that there for sure and in all cases, there is no difference and it does not matter what you put in your speakers, is in my mind not a good advice.

BRgds Klaus

[genek]

I'm not sure why a quick choice with unfamiliar material should be a requirement. It seems that just offers a letout clause for those who are sceptical of ABX testing. ("I'm certain I could have heard a difference, given more time!") I would be impressed if anyone could hear a staticstically significant difference between, say, capacitors of different materials and the same measured values, even after a month of listening.

Longer listening times give the ear/brain time to accommodate, and more time for imagination based on expectation to set in. To use my eye exam comparison, you can go a fairly long time with less than clear vision and "see through" a slight blur, and it's only when the diopter suddenly changes that you realize that what you were certain was an "O" is actually a "D."

[genek]

I am not sure that I can follow here. Does that mean, that if a friend of mine has got a new amplifier or set of speakers, and claims that they are a great improvement to his previous gear, I will have to tell him, that unles he can refer to a double blind study that proves a difference (or has conducted one himself), I will have to assume that his new speakers or amplifier sounds the same as the old one(s)?

If not, where is the line where you can assume that this rule of sameness applies, where you have to assume that the use of different constructions using different materials will results in the same sound, unless proven otherwise) and where it will not nessecarily be so?

It means that there is no way of being certain that that is not the case, yes. Every conclusion you draw about what you think you're hearing has an element of uncertainty in it that becomes higher as the perceived difference grows smaller. Where to draw the line that tells you that you can no longer trust your perception without measurements is the element of arbitrariness upon which the more fanciful segment of the high end audio industry relies.

When asked for advice I always tell people to use their ears, if they can hear a difference between different components, they should choose what they like best and can afford, if they can hear no difference, go for the cheapest.

This is actually pretty good advice. Even if better performance is real, why pay extra for if you can't hear it?

[speaker dave]

Longer listening times give the ear/brain time to accommodate, and more time for imagination based on expectation to set in. To use my eye exam comparison, you can go a fairly long time with less than clear vision and "see through" a slight blur, and it's only when the diopter suddenly changes that you realize that what you were certain was an "O" is actually a "D."

I agree that people are much more likely to hear real differences at the moment of switchover. I don't really believe people would detect a difference in a month they wouldn't hear in a minute. But if you force a choice in any period of time you fall open to charges that you prevented them from hearing real difference by forcing a speedy choice.

In the end the question is whether component changes are at all audible. Not whether they are quickly audible or dramatically different. I am sceptical that non-measurable component differences are discernable beyond random chance in a blind test and would give test subjects any opportunity to prove otherwise.

David

[speaker dave]

I am not sure that I can follow here. Does that mean, that if a friend of mine has got a new amplifier or set of speakers, and claims that they are a great improvement to his previous gear, I will have to tell him, that unles he can refer to a double blind study that proves a difference (or has conducted one himself), I will have to assume that his new speakers or amplifier sounds the same as the old one(s)?

BRgds Klaus

Yes, indeed. Now you are getting it.

(Okay, we'll assume new speakers sound different.)

David

[genek]

I agree that people are much more likely to hear real differences at the moment of switchover. I don't really believe people would detect a difference in a month they wouldn't hear in a minute. But if you force a choice in any period of time you fall open to charges that you prevented them from hearing real difference by forcing a speedy choice.

Yes, that's the fall back argument of someone who can't hear a difference in the "speedy choice" test and is determined not to accept the reality that the inability to do so proves that there is no discernible difference.

In the end the question is whether component changes are at all audible. Not whether they are quickly audible or dramatically different. I am sceptical that non-measurable component differences are discernable beyond random chance in a blind test and would give test subjects any opportunity to prove otherwise.

That's the whole point of the double-blind test, of course, to determine whether discernible differences exist at all. The conditions we have discussed are well established as maximizing the odds that any real difference will be detected.

[kkantor]

I am not sure that I can follow here. Does that mean, that if a friend of mine has got a new amplifier or set of speakers, and claims that they are a great improvement to his previous gear, I will have to tell him, that unles he can refer to a double blind study that proves a difference (or has conducted one himself), I will have to assume that his new speakers or amplifier sounds the same as the old one(s)?

If not, where is the line where you can assume that this rule of sameness applies, where you have to assume that the use of different constructions using different materials will results in the same sound, unless proven otherwise) and where it will not nessecarily be so?

I must admit that my mind works the other way around, if things are different, I will have to assume that they sound different, unless someone has proven that it sounds the same, not the other way around.

To me it is a fact for me what I hear (and see). I know hearing it is different for other people, I have had guests that could not tell the difference between the sound form a getto blaster and my AR speakers, does that proove that there is no difference?

When asked for advice I always tell people to use their ears, if they can hear a difference between different components, they should choose what they like best and can afford, if they can hear no difference, go for the cheapest. To tell people that there for sure and in all cases, there is no difference and it does not matter what you put in your speakers, is in my mind not a good advice.

BRgds Klaus

Klaus,

Truely, this is a very difficult issue. On one hand, the needs of the end user different from the designer or scientist. The end user wants to enjoy their system, their purchases, etc. On the other hand, it is hard for a professional to ignore some of the "facts" that get stated.

Think about all the pills at the drugstore ("chemist"). Some pills really work. Other pills make claims that can't be true, yet still have many happy customers willing to state how well they work. I don't want to deny a pill to someone who feels better, (of course if no harm is done.) Yet, I do not believe it is good thing for advertizers to promote whatever they wish, with no respect for testing. It is always possible to find a group of customers who will insist the pill works.

In general: speakers almost always alter the sound. Phono cartridges and turntables alter the sound. Larger amps, or amps with high output impedance, sometimes alter the sound. Capacitors sometimes alter the sound, if they are old and worn out, or if you change the type and specifications greatly, (eg- old electrolytic to new film). Wires, resistors, etc, almost never make any difference, unless defective. CD players, preamps, etc, only make a a difference in the case that the output level between different units changes.

-k

[kkantor]

Yes, that's the fall back argument of someone who can't hear a difference in the "speedy choice" test and is determined not to accept the reality that the inability to do so proves that there is no discernible difference.

That's the whole point of the double-blind test, of course, to determine whether discernible differences exist at all. The conditions we have discussed are well established as maximizing the odds that any real difference will be detected.

I'm going to back off from my earlier position, based on reading the latest comments between David and you. It's easy to think of proper, double-blind methodologies that would benefit from indefinitely long individual trials. Preference testing, as opposed to single-variable testing, probably should replicate the end-user experience in as many ways as is practical.

-k

[soundminded]

Klaus,

Truely, this is a very difficult issue. On one hand, the needs of the end user different from the designer or scientist. The end user wants to enjoy their system, their purchases, etc. On the other hand, it is hard for a professional to ignore some of the "facts" that get stated.

Think about all the pills at the drugstore ("chemist"). Some pills really work. Other pills make claims that can't be true, yet still have many happy customers willing to state how well they work. I don't want to deny a pill to someone who feels better, (of course if no harm is done.) Yet, I do not believe it is good thing for advertizers to promote whatever they wish, with no respect for testing. It is always possible to find a group of customers who will insist the pill works.

In general: speakers almost always alter the sound. Phono cartridges and turntables alter the sound. Larger amps, or amps with high output impedance, sometimes alter the sound. Capacitors sometimes alter the sound, if they are old and worn out, or if you change the type and specifications greatly, (eg- old electrolytic to new film). Wires, resistors, etc, almost never make any difference, unless defective. CD players, preamps, etc, only make a a difference in the case that the output level between different units changes.

-k

"On the other hand, it is hard for a professional to ignore some of the "facts" that get stated."

"Yet, I do not believe it is good thing for advertizers to promote whatever they wish, with no respect for testing."

What do the tests tell you professionals that we amateurs miss? It seems to me I've seen a lot of measurements of speakers that looked suspiciously similar, nearly textbook perfect and yet sounded nothing alike. I wouldn't have needed any double blind test to tell them apart. Exactly what are we missing?

BTW, I'm sure you know that taking medicine that doesn't work but whether by conicidence or by plecebo effect makes people "feel better" can have fatal consequences. Fortunately for them, fraudulent audio equipment manufacturers haven't been prosecuted by the government like medical quackery has been. If they were, a the prisons would be even more overcrowded than they are already.

[genek]

I'm going to back off from my earlier position, based on reading the latest comments between David and you. It's easy to think of proper, double-blind methodologies that would benefit from indefinitely long individual trials. Preference testing, as opposed to single-variable testing, probably should replicate the end-user experience in as many ways as is practical.

I wouldn't necessarily argue with that. Whether you like one thing better than another is probably not the same decision process as whether one thing is the same as another.

[genek]

What do the tests tell you professionals that we amateurs miss? It seems to me I've seen a lot of measurements of speakers that looked suspiciously similar, nearly textbook perfect and yet sounded nothing alike. I wouldn't have needed any double blind test to tell them apart.

I think I've missed those speakers entirely. I've never had the experience of expecting two speakers to sound the same based on their measurements and finding that they sounded different. OTOH, I have expected speakers to sound different based on their measurements and found that I couldn't tell them apart by their sound.

[speaker dave]

What do the tests tell you professionals that we amateurs miss? It seems to me I've seen a lot of measurements of speakers that looked suspiciously similar, nearly textbook perfect and yet sounded nothing alike. I wouldn't have needed any double blind test to tell them apart. Exactly what are we missing?

I've seen a lot of loudspeaker measurements but don't remember any of them looking similar enough that the speakers wouldn't be easily distinguishable. Speakers of different designs, and sometimes even mass produced speakers of the same design, just don't measure the same. Even if you could find two different systems that measure similarly on axis, their off axis curves would likely be highly different.

Of course, loudspeaker marketers, unsavory characters that they are, might be embelishing the curves to make them look more ideal. Are we talking reviewers curves or advertising curves? Do you have some examples? Please show us this textbook perfect speaker.

David