Jon,
Thanks for a most complete review of these cable issues. I realise that a lot of work got into composing this message, and a lot of time of which you probably have too little to begin with 😉
I'll get a copy of your AES paper and will read the other refs. But before that, two remarks if you allow me.
Somewhere in your post you mention that 'nobody has shown yet that these effects do not cause distortion above -90dB' or words to that effect. That I find is not really fair. If someone proposes the audibility of an effect (that in itself is real, no question about that), it is up to him/her to prove that this effect is of such magnetude that it is above -90dB and thus (lets assume that) audible.
Secondly, I regret that you found it necessary to portray DBT as necessarily done by 'amateurs'. Firstly, I resent that being an amateur in this, I would somehow be less smart or less thorough than the professionals. There are lots of interesting and usefull audio issues that have been brought forward by amateurs, those not professionally involved in audio. Secondly, does that mean that you DO accept DBT's by audio professionals? Those DBT's done for instance by Lipshitz/Vanderkooy with Ivor Tiefenbrunn?
[That particular test failed to show a reliable detection of the inclusion of a vintage Sony F1 digital processor into an otherwise analog replay chain. Until one of the participants (Lipshitz) noticed a change in hiss level with or without the F1 and from then on scored 100% correct.]
Jan Didden
Thanks for a most complete review of these cable issues. I realise that a lot of work got into composing this message, and a lot of time of which you probably have too little to begin with 😉
I'll get a copy of your AES paper and will read the other refs. But before that, two remarks if you allow me.
Somewhere in your post you mention that 'nobody has shown yet that these effects do not cause distortion above -90dB' or words to that effect. That I find is not really fair. If someone proposes the audibility of an effect (that in itself is real, no question about that), it is up to him/her to prove that this effect is of such magnetude that it is above -90dB and thus (lets assume that) audible.
Secondly, I regret that you found it necessary to portray DBT as necessarily done by 'amateurs'. Firstly, I resent that being an amateur in this, I would somehow be less smart or less thorough than the professionals. There are lots of interesting and usefull audio issues that have been brought forward by amateurs, those not professionally involved in audio. Secondly, does that mean that you DO accept DBT's by audio professionals? Those DBT's done for instance by Lipshitz/Vanderkooy with Ivor Tiefenbrunn?
[That particular test failed to show a reliable detection of the inclusion of a vintage Sony F1 digital processor into an otherwise analog replay chain. Until one of the participants (Lipshitz) noticed a change in hiss level with or without the F1 and from then on scored 100% correct.]
Jan Didden
Konnichiwa,
Perhaps the correct term would have been "non-statisticians".
Without forestalling any of Jon's reply, the vast majority of published DB Listening Tests, ESPECIALLY the ubiquotous (or should that be infamous) Audio "ABX" Test fail to show any relevance or significance of any kind in any context where fairly small differences are concerned. The problem is systematic and hence to include the results of such fundamentally flawed tests as any form of evidence is simply inadmissible.
It matters not what sort of qualification in whatever area of expertise the individual has that carries out a 16 Trial ABX Test.
Sayonara
PS, your example of the PCM-1 test is an good illustration of what is wrong with small sample size ABX testing.
janneman said:
Perhaps the correct term would have been "non-statisticians".
Without forestalling any of Jon's reply, the vast majority of published DB Listening Tests, ESPECIALLY the ubiquotous (or should that be infamous) Audio "ABX" Test fail to show any relevance or significance of any kind in any context where fairly small differences are concerned. The problem is systematic and hence to include the results of such fundamentally flawed tests as any form of evidence is simply inadmissible.
It matters not what sort of qualification in whatever area of expertise the individual has that carries out a 16 Trial ABX Test.
Sayonara
PS, your example of the PCM-1 test is an good illustration of what is wrong with small sample size ABX testing.
Jon was kind enough to lump all this information together. Although most we do not have any quantified analysis yet, but with all the considerations together, it makes it easier to reference when one is trying to quantify which are the real factors.
The capacitor tests seem very ineresting because I had conducted similar test probably 20 years ago, and could not find such differences, I think it's time to take a closer look.
I can also positively say that differences in cables are audible. But in some situation, the quality of the equipment used to conduct listening test need to be very good, and certain types of musice usually reveal differences more so than others. So a standard selection of material is also important.
I think it is not in the interest of the audio industry to solve all the engineering issues that associate with differences among cables. If they did that, then the only competition would be price, and this is the last situation they want to get into.
The capacitor tests seem very ineresting because I had conducted similar test probably 20 years ago, and could not find such differences, I think it's time to take a closer look.
I can also positively say that differences in cables are audible. But in some situation, the quality of the equipment used to conduct listening test need to be very good, and certain types of musice usually reveal differences more so than others. So a standard selection of material is also important.
I think it is not in the interest of the audio industry to solve all the engineering issues that associate with differences among cables. If they did that, then the only competition would be price, and this is the last situation they want to get into.
Re: Reply to Original Posted Question, Part 3
What does Pease have to say about your assertions of DA and speaker wires?? If you're gonna use him as a reference, why not explain to all what he thinks about all your assertions..as I recall, it ain't pretty...
Again, nice..but, the article never did test apples to apples, but showed only a range of parts. Had they changed only one parameter at a time and tested, it would have provided more useful information..Requests from JC for more information was met, how to say it, rather inelegantly on his part..
What you don't do is point out that your entire DA misconception as it applies to speaker wires is based on testing of capacitors that are designed to operate at voltage gradients at between 10 to 75% of the dielectric flashover rating..while the insulation of speaker wires, with say 20 mils thick insulation and 500 to 1500 volts per mil dielectric strength, runs in the kilovolt to tens of kilovolts regime...far less stress on the dielectric than a typical capacitor.
Has anyone ever, ever, measured this? Do you have anything along these lines, or is it just conjecture?
If you ever decide to actually try to measure it, do so with a cable Z closer to speaker impedances...that way, the capacitive energy storage is equivalent to the inductive storage. It'll be in the 300 to 1K pf per foot range, but certainly would bring any effect that exists up to a measureable level..course, your Dale NI 250's are useless as a load, you need a sub nH load.
The decision as to the viability of wood as a construction material should not be based on bending a toothpick...apples to apples...
There are two forces that occur with an energised conductor..the first is the pinch effect, this is the wire attempting to shrink in radius as a result of the current..the second is the force between the conductors..the relevant equation is:
Force (newtons/meter) = (u0 * I2)/(2 * PI * r).
At cables post 107711, I did this calc for a #12 wire pair, carrying a kilowatt into 8 ohms.. the result was .0018 ounces of force per inch of conductor..you consider that to be an issue?
(sorry, just hadta do that...)
Uh oh, be careful...Al may be listening...😀 good text, no comment.
Cheers, John
Jon Risch said:
What does Pease have to say about your assertions of DA and speaker wires?? If you're gonna use him as a reference, why not explain to all what he thinks about all your assertions..as I recall, it ain't pretty...
An excellent article, even though they got ESL incorrect for jelly roll capacitors..Jon Risch said:
Jon Risch said:
Again, nice..but, the article never did test apples to apples, but showed only a range of parts. Had they changed only one parameter at a time and tested, it would have provided more useful information..Requests from JC for more information was met, how to say it, rather inelegantly on his part..
What you don't do is point out that your entire DA misconception as it applies to speaker wires is based on testing of capacitors that are designed to operate at voltage gradients at between 10 to 75% of the dielectric flashover rating..while the insulation of speaker wires, with say 20 mils thick insulation and 500 to 1500 volts per mil dielectric strength, runs in the kilovolt to tens of kilovolts regime...far less stress on the dielectric than a typical capacitor.
Has anyone ever, ever, measured this? Do you have anything along these lines, or is it just conjecture?
If you ever decide to actually try to measure it, do so with a cable Z closer to speaker impedances...that way, the capacitive energy storage is equivalent to the inductive storage. It'll be in the 300 to 1K pf per foot range, but certainly would bring any effect that exists up to a measureable level..course, your Dale NI 250's are useless as a load, you need a sub nH load.
The decision as to the viability of wood as a construction material should not be based on bending a toothpick...apples to apples...
Jon Risch said:
There are two forces that occur with an energised conductor..the first is the pinch effect, this is the wire attempting to shrink in radius as a result of the current..the second is the force between the conductors..the relevant equation is:
Force (newtons/meter) = (u0 * I2)/(2 * PI * r).
At cables post 107711, I did this calc for a #12 wire pair, carrying a kilowatt into 8 ohms.. the result was .0018 ounces of force per inch of conductor..you consider that to be an issue?
Requests for actual measured wire movement have met with diversion..do you have any numbers for wire motion during music?Jon Risch said:
As far as I'm concerned, your -90 dB number is pure hogwash. And, it's not because I disagree with the number, per se, but, because you still limit yourself to using a measurement technique which does not measure localization..you are measuring weight with a yardstick...wrong tool...Jon Risch said:
Oh stop waffling, for goodness sakes!! it is, or it isn't!!😀 😀 😀Jon Risch said:
(sorry, just hadta do that...)
Jon Risch said:
Uh oh, be careful...Al may be listening...😀 good text, no comment.
No comments beyond your nice treatment here..Jon Risch said:
Cheers, John
Re: Reply to Original Posted Question, Part 4
The external magnetic field of a shield is directly proportional to the inhomogeneity of the briad pattern..a 100 mil braid error will result in the magnetic field extending beyond the braid on the order of that error, 100 mils..at 5 times that, it's less than 4% of the error on the outside surface of the braid..that's the dipole component, which drops off as 1/R. Higher order harmonics fall of as 1/RN, for N=2,3,4...
Your explanations for justification of bi-wiring are pretty useless, you've not demonstrated viability in any way, either by test or equation..in fact, your arguments along this line are easy fodder for people to claim you're full of hogwash...You are attempting to invoke magical hearing qualities, when really, you should be pushing the accepted limits of what we can hear in a stereo soundfield..as I am fully confident the old "what we can hear" arguments are in fact, the actual hogwash..and that 5 uSec/.06 dB level differences are closer to what should be addressed.
Until such time as everybody understands that humans require a non-zero timeframe to re-establish localization cue interpretation, fast switch DBT's are useless for localization..geeze whiz...it's a learned thing..
Although you never state such, you do tend to talk about long term tests, which, IMO, are required for localization..
If it includes long term listening adjustment to accomodate localization readjustment, then it is worth perusing..
Is that allowed?
Cheers, John
Wire pairs extend the mag field regardless of the presence of a shield. The shield materials do not react sufficiently to distort the signal within the pair, especially at low impedance. External magnetic surfaces that are within about 2 to 5 conductor spacing distances will affect the magnetic storage, however the extent of the effect w/r to localization is unknown...for distortion, that is easy enough to measure...have you done so, or is this also just speculation?Jon Risch said:
Jon Risch said:
The external magnetic field of a shield is directly proportional to the inhomogeneity of the briad pattern..a 100 mil braid error will result in the magnetic field extending beyond the braid on the order of that error, 100 mils..at 5 times that, it's less than 4% of the error on the outside surface of the braid..that's the dipole component, which drops off as 1/R. Higher order harmonics fall of as 1/RN, for N=2,3,4...
Correct..but there ya goes with the wrong tool again..😀Jon Risch said:
Jon Risch said:
Your explanations for justification of bi-wiring are pretty useless, you've not demonstrated viability in any way, either by test or equation..in fact, your arguments along this line are easy fodder for people to claim you're full of hogwash...You are attempting to invoke magical hearing qualities, when really, you should be pushing the accepted limits of what we can hear in a stereo soundfield..as I am fully confident the old "what we can hear" arguments are in fact, the actual hogwash..and that 5 uSec/.06 dB level differences are closer to what should be addressed.
Oh geeze, here we go again...😉 😉 😉 😉Jon Risch said:
Until such time as everybody understands that humans require a non-zero timeframe to re-establish localization cue interpretation, fast switch DBT's are useless for localization..geeze whiz...it's a learned thing..
Although you never state such, you do tend to talk about long term tests, which, IMO, are required for localization..
Useless as evidence of fact..useful perhaps as anecdotal evidence of something to be examined...Jon Risch said:
But rejected, if I recall correctly.Jon Risch said:
If it includes long term listening adjustment to accomodate localization readjustment, then it is worth perusing..
Jon Risch said:
Is that allowed?
Cheers, John
soongsc said:
That is indeed the case. Line cord susceptibility can be eliminated, by mods to line cord technology or amp input coupling..but, will aftermarket guys kill the goose? No, silly to think so..
Cheers, John
pinkmouse said:
Oh...thanks...
I was concerned that something had violated forum rules.
In that case, please delete #229 on..thanks.
John
I usually stay out of these threads but I was peripherally involved with the Jung/Curl article. I critiqued the
instrumentation since it stemmed from an article that Walt and I did on uses of instrumentation amplifiers.
I also built and tested my own version of the 'bridge null' instrument. First of all the bridge null comparator
and all its variants is a brilliant 19th century invention which easily showed the published results. I took
some abuse for these results too by people who believed that measuring any difference was impossible
(read threatening).
The real problem was with the interpretations of these results. As Pease showed in the first place a simple
_LINEAR_ RC ladder can be used to model soakage to an almost arbitrary level. Pease then went on to
show how to compensate the then popular dual slope AtoD's to better than 18 bits. It is my personal
opinion that attaching the term 'memory' to this effect was a most unfortunate event. Saving anymore
than one dimension of a signal (i.e. a single voltage at a given point in time) in a capacitor would violate
some of the prime laws of information theory. The capacitor quality thing quickly ballooned into 'the
music out depends on all the previous music in'. Yes there are nonlinear dielectrics that have measurable
distortion but there are also many 'good' dielectrics where the perfectly linear model gets you down to
-100+ dB levels very easily.
instrumentation since it stemmed from an article that Walt and I did on uses of instrumentation amplifiers.
I also built and tested my own version of the 'bridge null' instrument. First of all the bridge null comparator
and all its variants is a brilliant 19th century invention which easily showed the published results. I took
some abuse for these results too by people who believed that measuring any difference was impossible
(read threatening).
The real problem was with the interpretations of these results. As Pease showed in the first place a simple
_LINEAR_ RC ladder can be used to model soakage to an almost arbitrary level. Pease then went on to
show how to compensate the then popular dual slope AtoD's to better than 18 bits. It is my personal
opinion that attaching the term 'memory' to this effect was a most unfortunate event. Saving anymore
than one dimension of a signal (i.e. a single voltage at a given point in time) in a capacitor would violate
some of the prime laws of information theory. The capacitor quality thing quickly ballooned into 'the
music out depends on all the previous music in'. Yes there are nonlinear dielectrics that have measurable
distortion but there are also many 'good' dielectrics where the perfectly linear model gets you down to
-100+ dB levels very easily.
Oh, I forgot to say that I have shorted 2 4/0 welding cables across 1000 amp-hour batteries and you can begin to feel the repulsive force. It's a large leap of faith to go from this to the current levels in a normal system and think the effects have any meaningful effect. Afterall these basic physical effects tend to have very low orders of nonlinearity anyway.
So say I can modulate the capacitance of my cables by 10ppm when immersing them in a 160dB SPL sound field. I say even then there would be only slight thirds and maybe seconds added because thats the nature of small perturbations of 'bulk' things.
So say I can modulate the capacitance of my cables by 10ppm when immersing them in a 160dB SPL sound field. I say even then there would be only slight thirds and maybe seconds added because thats the nature of small perturbations of 'bulk' things.
Hi Scott
It's unfortunate that many people try to extend a dielectric working at 10 to 75% of it's flashover voltage, in the microfarads, to something working in the .1 % flashover range, in the low to sub nanofarads in a low impedance setup.
Cheers, John
You are a master of understatement..😀scott wurcer said:
Yah, he did nice work.scott wurcer said:
It's unfortunate that many people try to extend a dielectric working at 10 to 75% of it's flashover voltage, in the microfarads, to something working in the .1 % flashover range, in the low to sub nanofarads in a low impedance setup.
Yah, the words were a bit unfortunate..but, remember the audience they were playing to..it doesn't pay to be all technical and dry..it puts people to sleep..or, worse..scott wurcer said:
Cheers, John
scott wurcer said:
Scott...you are indeed one sick individual.
Yah, I think believing that .0018 ounces per inch with a kilowatt drive signal is of any concern, is certainly one huge leap..scott wurcer said:
Wow...your listening environment is a lot louder than mine...😉scott wurcer said:
Then again, you short kilo-amp hour batteries with 4/0 wires for fun...
Cheers, John
Re: Re: Reply to Original Posted Question, Part 3
I don't think that dragging Mr. Pease's misinformed personal opinion of me into the technical discussion is truly relevant, nor is it helpful for you take the attitude that you have on this subject. What happened to all those platitudes and urges to have honest technical discussions?
Is it relevant if the person who listened to your dual coaxial speaker cables thought you were a XXXXXX!!!!!?????
Not really, strictly from a technical standpoint.
I think that the data presented at Mr. Pease's web page is relevant, and can stand on it's own. That Mr. Pease is upset about my referencing it is another matter entirely.
I hope that you can refrain from the snide comments and covert attacks in future posted replies to me and others.
And as I have stated before, I think that you are mis-stating the level of stress necesary for capacitor distortions to show up.
The measurements in the "Picking Capacitors" article were not taken near the dielectric breakdown point, they were taken at more or less normal line level voltages.
In point of fact, DA manifests at low voltages, and is not limited to a high voltage situation.
Speaker cables in the home can be exposed to voltage levels of close to a hundred volts, modern SS power amps are now readily available in that range of output.
The other factor is, we are not talking about nice, neat, tidy normal capacitor dielectrics, we are in fact, including PVC, which is a terrible quality dielectric when compared to PE or teflon.
We have the measured levels from the various articles I cite, including the "A Real-Time Signal Test for Capacitor Quality" article in Audio Amatuer. You may not like the test procedures, but there IS some data there, as long as you don't deliberately close your eyes and discount it completely.
Was this entirely neccessary, with the nasty comments?
First, few retail speaker cables have a characteristic Z close to that of the load, and while there are some DIY speaker cable designs that come a bit closer, including my own CC89259, the braided CAT5, and your double coax (which is conceptually the same as the patented Wire World speaker cables), very very few of them really get down all the way where they would need to be.
Secondly, if the sheer amount of capacitance can be avoided, then why not avoid it? Your criteria is limited to that of your own theories, which I have not yet begun to discuss, and I keep reminding you of this, and you keep ignoring it and insisting on forcing those theories into the discussion, even when they are at the moment, out of place.
Do you honestly think that only YOUR theories are worthy of discussion, and all else must be ignored or the subject turned back to only your theories?
I was not satisfied with your figures at that time, and have not verified that they are correct.
In any case, it does not address the second portion of the vibration issue, that of external vibration from the sound in the room.
This is still in the nature of insisting that only your theories are relevant, and all else is so much drek. Measuring what you are calling localization may not be the only factor we need to be concerned with. If you focus ONLY on that, and ignore all else, then what you are doing is not necessarily still within the realm of classic science, but has perhaps become something else?
However, if we take one moment to look at the basic numbers you have been posting, then it would seem that you are advocating a very small amount of signal distortion as an issue, when you reference 0.03 dB differences, that is roughly equivalent to about -50 dB down or more. Some of the other numbers you have posted are also fairly far down there if converted to how many dB down that would be.
I find it odd that when I mention -90 dB, and have given an example of why I think that this is relevant, you call it pure hogwash, yet when you mention some very small amounts of signal distortion, those are perfectly relevant.
Jon Risch
jneutron said:
What does Pease have to say about your assertions of DA and speaker wires?? If you're gonna use him as a reference, why not explain to all what he thinks about all your assertions..as I recall, it ain't pretty...
I don't think that dragging Mr. Pease's misinformed personal opinion of me into the technical discussion is truly relevant, nor is it helpful for you take the attitude that you have on this subject. What happened to all those platitudes and urges to have honest technical discussions?
Is it relevant if the person who listened to your dual coaxial speaker cables thought you were a XXXXXX!!!!!?????
Not really, strictly from a technical standpoint.
I think that the data presented at Mr. Pease's web page is relevant, and can stand on it's own. That Mr. Pease is upset about my referencing it is another matter entirely.
I hope that you can refrain from the snide comments and covert attacks in future posted replies to me and others.
And as I have stated before, I think that you are mis-stating the level of stress necesary for capacitor distortions to show up.
The measurements in the "Picking Capacitors" article were not taken near the dielectric breakdown point, they were taken at more or less normal line level voltages.
In point of fact, DA manifests at low voltages, and is not limited to a high voltage situation.
Speaker cables in the home can be exposed to voltage levels of close to a hundred volts, modern SS power amps are now readily available in that range of output.
The other factor is, we are not talking about nice, neat, tidy normal capacitor dielectrics, we are in fact, including PVC, which is a terrible quality dielectric when compared to PE or teflon.
We have the measured levels from the various articles I cite, including the "A Real-Time Signal Test for Capacitor Quality" article in Audio Amatuer. You may not like the test procedures, but there IS some data there, as long as you don't deliberately close your eyes and discount it completely.
Was this entirely neccessary, with the nasty comments?
First, few retail speaker cables have a characteristic Z close to that of the load, and while there are some DIY speaker cable designs that come a bit closer, including my own CC89259, the braided CAT5, and your double coax (which is conceptually the same as the patented Wire World speaker cables), very very few of them really get down all the way where they would need to be.
Secondly, if the sheer amount of capacitance can be avoided, then why not avoid it? Your criteria is limited to that of your own theories, which I have not yet begun to discuss, and I keep reminding you of this, and you keep ignoring it and insisting on forcing those theories into the discussion, even when they are at the moment, out of place.
Do you honestly think that only YOUR theories are worthy of discussion, and all else must be ignored or the subject turned back to only your theories?
I was not satisfied with your figures at that time, and have not verified that they are correct.
In any case, it does not address the second portion of the vibration issue, that of external vibration from the sound in the room.
This is still in the nature of insisting that only your theories are relevant, and all else is so much drek. Measuring what you are calling localization may not be the only factor we need to be concerned with. If you focus ONLY on that, and ignore all else, then what you are doing is not necessarily still within the realm of classic science, but has perhaps become something else?
However, if we take one moment to look at the basic numbers you have been posting, then it would seem that you are advocating a very small amount of signal distortion as an issue, when you reference 0.03 dB differences, that is roughly equivalent to about -50 dB down or more. Some of the other numbers you have posted are also fairly far down there if converted to how many dB down that would be.
I find it odd that when I mention -90 dB, and have given an example of why I think that this is relevant, you call it pure hogwash, yet when you mention some very small amounts of signal distortion, those are perfectly relevant.
Jon Risch
- Status
- Not open for further replies.