On the one hand we have Walt Jung's article:
http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
http://waltjung.org/PDFs/Picking_Capacitors_2.pdf
However, it is at great odds with Rod Elliot's article in many points, such as the magnitude of audible effects caused by dielectric absorption, paralleling electrolytics with films, and so on:
http://sound.westhost.com/articles/capacitors.htm
So, which one is right?
Then we have this capacitor distortion page, and I'm trying to figure out what the graphs are actually showing, as I can't read the Japanese:
http://fnt-www.ss.titech.ac.jp/~hajime/uec/distortion/cap/
Any ideas?
http://waltjung.org/PDFs/Picking_Capacitors_1.pdf
http://waltjung.org/PDFs/Picking_Capacitors_2.pdf
However, it is at great odds with Rod Elliot's article in many points, such as the magnitude of audible effects caused by dielectric absorption, paralleling electrolytics with films, and so on:
http://sound.westhost.com/articles/capacitors.htm
So, which one is right?
Then we have this capacitor distortion page, and I'm trying to figure out what the graphs are actually showing, as I can't read the Japanese:
http://fnt-www.ss.titech.ac.jp/~hajime/uec/distortion/cap/
Any ideas?
But you have to consider the company Walt was keeping for that article
Rod has occasionally gone overboard and made simply wrong statements on his site in his enthusiasm for "debunking" audiophile claims
But I found Rod's capacitor page overall very good at sticking to clear engineering based explanations
Rod's presentation of the linear model of dielectric adsorption is consistent with Bob Pease' "capacitor Soakage" article
http://www.national.com/rap/Application/0,1570,28,00.html
Cyril Bateman's "Capacitor Sound" series in Electronics and Wireless World made high resolution measurements of distortion in many capacitor types and should supercede the Marsh/Jung article - if you believe that measurable distortion explains subjective capacitor "sound"
Rod has occasionally gone overboard and made simply wrong statements on his site in his enthusiasm for "debunking" audiophile claims
But I found Rod's capacitor page overall very good at sticking to clear engineering based explanations
Rod's presentation of the linear model of dielectric adsorption is consistent with Bob Pease' "capacitor Soakage" article
http://www.national.com/rap/Application/0,1570,28,00.html
Cyril Bateman's "Capacitor Sound" series in Electronics and Wireless World made high resolution measurements of distortion in many capacitor types and should supercede the Marsh/Jung article - if you believe that measurable distortion explains subjective capacitor "sound"
juergenk said:
''You can fool some people for now but not all people for ever''. An old saying that proves caps and wire are far too old a story not to be having any base since people all over the world celebrate picking favorites and casting vote with their wallets.
Snake oil merchant sneaks where there is real oil trade. And hardcore engineers fail to see there is actually oil around too, when chasing the snake oil merchant for our benefit. Thats what is wrong.
even in really science, not only in diy audio, there are always at least two opinions. 
I don't mind about that.
But was puzzles me, is that the popular model of DA seems not to explain the subjective(?) effects.
Even more as the model is used for explanation and repeated every time.
So I'm asking to prove my logic.
I don't mind about that.
But was puzzles me, is that the popular model of DA seems not to explain the subjective(?) effects.
Even more as the model is used for explanation and repeated every time.
So I'm asking to prove my logic.
salas said:
Why should you get a generic cap when you can get a high-quality Rifa or Electronic Concepts cap for a couple of $?
salas said:
There's a billion people around the world who believe Jesus was resurrected. And if we believe Audioquest, wire is all about faith.
"Of course, you can choose not to believe, but you'd be missing out on so much." --Alasdair Patrick of Audioquest.
It seems there's no hope for us unbelievers.
I think part of it is the misuse of the term "dielectric adsorption"
to have any meaning "dielectric adsorption" must exclude some other dielectric related phenomena - like (nonlinear) varying dielectric constant, but I think I see people using the term to mean any deviation from ideal dielectric behavior
for an engineer there is a dielectric adsorption test procedure that gives a # that other people can verify by making the same measurement
there is a natural extension to a linear model and measurement over larger frequency range that Bob Pease shows
Then there are examples of unsupported attribution of capacitor sound to "dielectric adsorption"
Cyril Bateman seems to have a odd usage in that he refers to dielectric adsorption as a cause of "capacitor sound" without one iota of logical linkage - his measurements are all of harmonic distortion of sine wave excitation of capacitors with varying DC bias - surely a nonlinear dielectric response - but not at all linked by any data to dielectric adsorption
to have any meaning "dielectric adsorption" must exclude some other dielectric related phenomena - like (nonlinear) varying dielectric constant, but I think I see people using the term to mean any deviation from ideal dielectric behavior
for an engineer there is a dielectric adsorption test procedure that gives a # that other people can verify by making the same measurement
there is a natural extension to a linear model and measurement over larger frequency range that Bob Pease shows
Then there are examples of unsupported attribution of capacitor sound to "dielectric adsorption"
Cyril Bateman seems to have a odd usage in that he refers to dielectric adsorption as a cause of "capacitor sound" without one iota of logical linkage - his measurements are all of harmonic distortion of sine wave excitation of capacitors with varying DC bias - surely a nonlinear dielectric response - but not at all linked by any data to dielectric adsorption
phn said:
Because a generic cap is adequate as much as a Mundorf Supreme for audioband according to hardcore engineers. So its about testing the concept.
Audioquest is humoring plus making money. Leave the traders aside.
Just pick up wire and caps from your used parts bin, logos faded, crappy cheap old stuff. Substitute and swap some in a circuit.
Guess what? They have a sound of their own.
juergenk said:even in really science, not only in diy audio, there are always at least two opinions.
I don't mind about that.
But was puzzles me, is that the popular model of DA seems not to explain the subjective(?) effects.
Even more as the model is used for explanation and repeated every time.
So I'm asking to prove my logic.
Scientists have proven terribly inadequate in researching the phenomena and proving the models, or there are no true scientists interested in audio. I am leaning to the second assumption. Just not a market big enough for real research.
Are we talking signal caps, or all caps?
I think the best way to decide which pudding you like best is with a spoon.
I am haveing problems in believeing my amp sound so horrible using $10 signal caps, that it will be significantly "fixed" by using $100 caps... I agree many non audio-phool priced caps, out there.
I simple Cmoy amp is all you need to hear the diffirence between caps, I am not saying, I will ever be able to identify a specific cap by ear, but sure anyday, you can tell which cmoy has the input caps when you AB it, it sounds less clear, regardless of price - well... up to my $10 magic ceiling.
As for PSU your average amp with cheap PSU caps, compared to the same amp, with same value caps, in nice low ESR version models, garden varieties like Vishay etc...There will BE a diffirence in the sound of the amp. Less haze more textured bass... etc...
As I said get a spoon and try some, if it turns out you have expensive taste, I'm sure you can find something else you can do without.
I think the best way to decide which pudding you like best is with a spoon.
I am haveing problems in believeing my amp sound so horrible using $10 signal caps, that it will be significantly "fixed" by using $100 caps... I agree many non audio-phool priced caps, out there.
I simple Cmoy amp is all you need to hear the diffirence between caps, I am not saying, I will ever be able to identify a specific cap by ear, but sure anyday, you can tell which cmoy has the input caps when you AB it, it sounds less clear, regardless of price - well... up to my $10 magic ceiling.
As for PSU your average amp with cheap PSU caps, compared to the same amp, with same value caps, in nice low ESR version models, garden varieties like Vishay etc...There will BE a diffirence in the sound of the amp. Less haze more textured bass... etc...
As I said get a spoon and try some, if it turns out you have expensive taste, I'm sure you can find something else you can do without.
I didn't really want this to turn into a discussion about blind testing. I agree with what Nelson Pass wrote a number of times, that a difference tends to be most clear after one is used to a reference system for a long time, then something is changed. Of course, there's no reason the change cannot be blind, made by someone else, or a computer.
My signal chain is DC coupled, so this was not an issue, until now I reached the point where I need to make a passive line-level crossover for a biamping setup, so capacitors are unavoidable. I am thinking of Kreskov's transient perfect 2nd order. Now, I need to choose whether, after the trouble of making everything DC coupled, inserting capacitors in the signal path is going to degrade sound, or I should go to the significant expense of using two DACs and doing the crossover in DSP before them. On the one hand, I want to approach the most undistorted sound, but on the other, I don't want to spend more than necessary to do that.
The ear's dynamic range is very large, but say in a realistic setup the noise floor will be the limiting factor, and is likely to be no lower than -115 dB even with everything shielded and good layout etc. So, can I be certain that a choice of capacitors is possible for the crossover that will guarantee capacitor-attributable distortion will be significantly below that figure (noise is not as objectionable as some signal-correlated distortions, and thus the 'significantly lower')?
My signal chain is DC coupled, so this was not an issue, until now I reached the point where I need to make a passive line-level crossover for a biamping setup, so capacitors are unavoidable. I am thinking of Kreskov's transient perfect 2nd order. Now, I need to choose whether, after the trouble of making everything DC coupled, inserting capacitors in the signal path is going to degrade sound, or I should go to the significant expense of using two DACs and doing the crossover in DSP before them. On the one hand, I want to approach the most undistorted sound, but on the other, I don't want to spend more than necessary to do that.
The ear's dynamic range is very large, but say in a realistic setup the noise floor will be the limiting factor, and is likely to be no lower than -115 dB even with everything shielded and good layout etc. So, can I be certain that a choice of capacitors is possible for the crossover that will guarantee capacitor-attributable distortion will be significantly below that figure (noise is not as objectionable as some signal-correlated distortions, and thus the 'significantly lower')?
Quince, discussions about capacitor sound always round that corner to detour into blind testing and ABX and all that stuff. Really tired stuff.
By the way, though your signal chain, so called, is DC coupled, can you think of how your PSU capacitors (electrolytics, surely?) affect the signal? From the perspective of this thought, there exists no "signal chain" independent of everything else in your amp.
By the way, though your signal chain, so called, is DC coupled, can you think of how your PSU capacitors (electrolytics, surely?) affect the signal? From the perspective of this thought, there exists no "signal chain" independent of everything else in your amp.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.