john curl said:
Though it doesn't seem to show itself in your measurements which you say is due to "microdiodes" and not dielectric effects.
By the way, how exactly did you go about ruling out dielectric effects being responsible for the distortion you're measuring?
I quoted you directly. Those who wish to read it in its original form can go here:
http://www.audioasylum.com/forums/prophead/messages/5991.html
se
SE, as usual, you have no idea what you are talking about, when to referring to my understanding of the situation. I would appreciate that the moderators take note of this.
For the record, I have done 100's of measurements on DA in caps, and this would extend to wire insulation as well, if I chose to run the tests, with wires, rather than caps. DA is a LINEAR DISTORTION so it does not show up on my measurements of NON-LINEARITY in metal wires. However, those of you who would actually like to look at some research in this area, might look at the work in 'Electronics World' over the last year and more by Cryril Bateman, regarding distortion in capacitors which goes deeper than my work of 15 and 25 years ago, respectively.
For the record, I have done 100's of measurements on DA in caps, and this would extend to wire insulation as well, if I chose to run the tests, with wires, rather than caps. DA is a LINEAR DISTORTION so it does not show up on my measurements of NON-LINEARITY in metal wires. However, those of you who would actually like to look at some research in this area, might look at the work in 'Electronics World' over the last year and more by Cryril Bateman, regarding distortion in capacitors which goes deeper than my work of 15 and 25 years ago, respectively.
John,
sorry for making a detour from the cable issue, but since you
brought up capacitors, I have a question. I have had endless
discussions before with some people about how DA could cause
distorsion (I am not discussing audibility of caps now, just
distorsion). Since DA is usually modelled by paralleling the cap
with a number of RC links accounting for the various types of
dipoles, and this seems to be a generally agreed on model,
it is clear that a cap with DA is a linear system. Maybe I am
missing something here, but I don't quite see how this could
show up as anything else than an a modified frequency
response. In my understanding that is not distorsion, which
I understand to be a non-linear phenomenon. However, you
are the first person I have seen to make a distinction between
linear and non-linear distorsion, which makes me curious how
you define the two. I have a feeling you might have an
explanation that others have failed to give.
sorry for making a detour from the cable issue, but since you
brought up capacitors, I have a question. I have had endless
discussions before with some people about how DA could cause
distorsion (I am not discussing audibility of caps now, just
distorsion). Since DA is usually modelled by paralleling the cap
with a number of RC links accounting for the various types of
dipoles, and this seems to be a generally agreed on model,
it is clear that a cap with DA is a linear system. Maybe I am
missing something here, but I don't quite see how this could
show up as anything else than an a modified frequency
response. In my understanding that is not distorsion, which
I understand to be a non-linear phenomenon. However, you
are the first person I have seen to make a distinction between
linear and non-linear distorsion, which makes me curious how
you define the two. I have a feeling you might have an
explanation that others have failed to give.
Too late for editing 🙁
Just wanted to add that for those interested there is an
excellent theoretical treaty on the causes and modelling
of DA in capacitors by Kundert, which is available online.
It has been a while since I read it, and I just briefly looked at
it again before posting, but he seems not have anything
to say on how this is related to distorsion, but that was
not within the scope of the article. However, we do learn,
for instance, that that little harmless real-valued constant
epsilon we all learnt about in school and read in books is
actually not at all a real-valued constant but a complex-valued
function of frequency.
http://www.designers-guide.com/Modeling/da.pdf
Just wanted to add that for those interested there is an
excellent theoretical treaty on the causes and modelling
of DA in capacitors by Kundert, which is available online.
It has been a while since I read it, and I just briefly looked at
it again before posting, but he seems not have anything
to say on how this is related to distorsion, but that was
not within the scope of the article. However, we do learn,
for instance, that that little harmless real-valued constant
epsilon we all learnt about in school and read in books is
actually not at all a real-valued constant but a complex-valued
function of frequency.
http://www.designers-guide.com/Modeling/da.pdf
john curl said:
Well, all I have to go by as to your understanding of the situation is what you yourself have said.
Um, John, you yourself said that there was a non-linear part to DA which wasn't part of the normal DA model and that a non-linear element would have to be added to the DA model to model the non-linear part, which you suggested could be done to a first order approximation by way of adding diodes in series with resistors.
To wit:
I would think that the model normally used for dielectric absorption was made for polystyrene caps back in the 1950's. Walt Jung and I have papers going back that far. It does NOT describe any non-linear distortion, and polystyrene usually doesn't have much non-linear distortion in any case. Of course, when you change over to mylar (polar?) or electrolytic caps, the model still works for the linear distortion, but I would agree that you would have to add a non-linear part to the model to get the non-linear part in the model. I have used diodes in series with resistors in the past, to model non-linear pots, etc. I can see that it could work for caps, to a first order at least.
Here you say there is a non-linear part to DA which you suggest could be modeled using diodes. And you say the distortion you've been measuring in interconnects is a diode-like effect.
So, I ask once again, how exactly did you go about ruling out dielectric effects being responsible for the distortion you're measuring?
se
For everyone: When you MODEL a component, you put in any and all potential effects that might effect performance. So, if I were to model a cap, I would put in series resistance, inductance, leakage, DA and any non-linear diode like effects to the cap model, in order to make it as complete as possible.
The IDEAL MODEL of DA does not have a non-linear component. However, it might be possible that a REAL MODEL of a cap has a non-linear component that is somehow related to DA. I don't know if this is true, but Cyril Bateman seems to think so in his articles on caps in 'Electronics World'
By the way, thanks to the individual who put Ken Kundert's paper up for everyone to look at. This should answer the most basic questions and perhaps more.
It might be noted that DA has been seriously researched for at least 50 years. It was very important when analog computers were popular, and many great papers were generated in the '50's on DA. It was somewhat forgotten during the '60's when digital computers superseded analog computers.
The IDEAL MODEL of DA does not have a non-linear component. However, it might be possible that a REAL MODEL of a cap has a non-linear component that is somehow related to DA. I don't know if this is true, but Cyril Bateman seems to think so in his articles on caps in 'Electronics World'
By the way, thanks to the individual who put Ken Kundert's paper up for everyone to look at. This should answer the most basic questions and perhaps more.
It might be noted that DA has been seriously researched for at least 50 years. It was very important when analog computers were popular, and many great papers were generated in the '50's on DA. It was somewhat forgotten during the '60's when digital computers superseded analog computers.
There is one seemingly obvious thing about capacitors that I
have never seen mentioned, although I am sure it must have
been studied. Unless I totally misremember my physics,
the charge in a capacitor will result in a mechanical force,
as explained by Coulombs law. No dielectricum or other material
in a capacitor is perfectly rigid, so it seems the distance
between the plates/foils/whatever will be dependent on the
charge, i.e. the voltage. Furhtermore, by elementary mechanics
we also know the distance will not vary linearly with the force.
In other words, the capacitance will be a non-linear function
of the voltage. I have not tried to make calculations how big
this effect might be. Probably it is so small that standard
theory has dismissed it and this is the reason it is never
mentioned. However, standard textbook dismisses other
things as negligible which are not considered so by many
in audio. Could this possibly be a non-linear effect in
capacitors that has been overlooked?
I am sorry if I have made some error in the assumptions or
reasoning above. It is a long time since I read about those
physical phenomena and I can't find the appropriate books
right now. Please tell me ASAP if I messed it up.
have never seen mentioned, although I am sure it must have
been studied. Unless I totally misremember my physics,
the charge in a capacitor will result in a mechanical force,
as explained by Coulombs law. No dielectricum or other material
in a capacitor is perfectly rigid, so it seems the distance
between the plates/foils/whatever will be dependent on the
charge, i.e. the voltage. Furhtermore, by elementary mechanics
we also know the distance will not vary linearly with the force.
In other words, the capacitance will be a non-linear function
of the voltage. I have not tried to make calculations how big
this effect might be. Probably it is so small that standard
theory has dismissed it and this is the reason it is never
mentioned. However, standard textbook dismisses other
things as negligible which are not considered so by many
in audio. Could this possibly be a non-linear effect in
capacitors that has been overlooked?
I am sorry if I have made some error in the assumptions or
reasoning above. It is a long time since I read about those
physical phenomena and I can't find the appropriate books
right now. Please tell me ASAP if I messed it up.
Hi,
IMHO, yes it should still work fine.
No need to tell you that those solid core powercords have no WAF whatsoever...
Cheers,😉
IMHO, yes it should still work fine.
No need to tell you that those solid core powercords have no WAF whatsoever...
Cheers,😉
Neither do my 280 cm line source ESLs, so I suspect it will be lost in the noise floor 😉
John, if anything degenerates into personal attacks or name-calling, be assured that I will step in. Disputes, questions, discussions, and disagreements about technical points are the whole raison d'etre for this forum. The closest thing I've seen to a personal attack here is someone saying "you don't know what you're talking about," and that falls somewhat short of the bar.
In my personal opinion, I think most of the comments about electrical properties of materials by most of the participants fall into that latter category
john curl said:
Yes, the Kundert paper is an excellent one, one which I've had for a while now. However it doesn't answer this twice asked and so far twice unanswered question:
How exactly did you go about ruling out dielectric effects being responsible for the distortion you're measuring?
se
Christer said:
That's basically just the reverse of what one would call "microphonics." It's a very real effect as anyone who's heard a photographic flash unit re-energizing can tell you.
And sure, if the capacitance changed in a non-linear fashion, it should result in a non-linear distortion. How signficiant it might be I've really no idea, but I'd imagine that for well-made film capacitors it would probably be exceedingly small.
One of John's colleagues has made a big issue about thise sort of motor/generator effect in interconnects and speaker cables. So here we have yet another possible distortion mechanism beside "microdiodes" in the wire itself.
se
At this time, I cannot exclude some subtle non-linear property of the dielectric of the wire material covering, as my reference cables both use teflon. However, I have not seen any direct correlation to this property to the dielectric used, at this time. Therefore, for safety sake, let's everyone use teflon in our interconnects🙂
Now when it comes to non-linear effects in caps. We have measured them for decades. By far, the most interesting distortion, comes from typical ceramic caps used in cost-effective situations. This is best brought out by modifying a TEK 577 curve tracer to show gross deviations from ideal.
Now when it comes to non-linear effects in caps. We have measured them for decades. By far, the most interesting distortion, comes from typical ceramic caps used in cost-effective situations. This is best brought out by modifying a TEK 577 curve tracer to show gross deviations from ideal.
For interest
The following is a post made by Julian Vereker some years ago the Naim Audio forum, after similar reactions to the directionality in their cables.
I've heard it myself, but I'm not going to fuel the bipolarity here by proffering an explanation as to why, those that don't think it's possible will never change and those that can hear it know they're right - this is another of those futile threads
The following is a post made by Julian Vereker some years ago the Naim Audio forum, after similar reactions to the directionality in their cables.
I've heard it myself, but I'm not going to fuel the bipolarity here by proffering an explanation as to why, those that don't think it's possible will never change and those that can hear it know they're right - this is another of those futile threads
Right on Julian. Julian's comments are appropriate to what most serious audio designers consider when making evaluations of audio components and their properties. This is how my audio associates and I interact.
We don't 'demand' DB tests or 'peer review' when making our evaluations. We trust our ears, our own test equipment when appropriate, and use everything that we can to make a better audio product.
For the record, Julian Vereker and I knew each other for more than 25 years. We once hung out at each other's houses, over periods of weeks. He was located in the UK and I was in Berkeley, CA USA. I learned a lot, in the 70's, from him. I hope he felt the same way about me. RIP, Julian.
We don't 'demand' DB tests or 'peer review' when making our evaluations. We trust our ears, our own test equipment when appropriate, and use everything that we can to make a better audio product.
For the record, Julian Vereker and I knew each other for more than 25 years. We once hung out at each other's houses, over periods of weeks. He was located in the UK and I was in Berkeley, CA USA. I learned a lot, in the 70's, from him. I hope he felt the same way about me. RIP, Julian.
john curl said:
Isn't that just like a Berkeley liberal, giving people a false sense of security.
NO insulation is 100% effective. Abstinance IS 100% effective and for the life of me I cannot figure out why these wacko liberals fight so hard to keep abstinance out of the cable education cirricula. We're not asking that ONLY abstinance be taught. Only that it be included.
What non-linear effects are you talking about specifically? Typical ceramic caps are made of ferroelectric materials which like ferromagnetic materials exhibit decidedly non-linear effects.
se
john curl said:
It's all well and good to trust your ears and go with whatever works for you. That's precisely what I do.
But when you then turn around and instead of simply saying "this sounds better to me" and make declarative statements of fact such as actual audibility or the cause of some particular distortion, etc., you should be prepared to substantiate those claims and not become indignant and insulting to those who may question or challenge such claims.
That's how it works for the REAL "big boys." The ones who are involved in serious research who don't feed people empty claims and wild speculation and expect it to be swallowed without question simply because they "trust their ears."
se
Unlikely in the extreme. The extrusion temperatures for plastic insulation are an order of magnitude too low for something like that. Add in heat capacity and thermal mass of the copper (we don't exactly let it come to an equilibrium temperature with the molten plastic!), and what you've got is uninformed speculation about an unproved phenomenon. All due respect to Mr. V as an electronic product designer, but he was no materials maven.
Microphonics in caps is a real issue, but is not normally related to changes in the dielectric thickness because of force, at least for plastic film caps. Usually, it's a sign of inadequate winding tension or poor termination. You might as well worry about intermodulation with mains frequency from photoacoustic effects.
I do have some horribly microphonic shielded cable here. I never traced down what the cause was, just pulled it out of my system. See, no-one claims that all cables sound alike all the time 😉
Re: For interest
Why is it that ONLY physical causes are suspected and NEVER psychological causes?
To suspect that the perceived differences may be due to well-established psychological phenomena is on just a firm ground as suspectiving hot plastic affecting the crystalline structure.
That's another trait of the REAL "big boys." They don't dismiss other possibilities out of hand because they threaten preconcieved beliefs.
By the way, if Mr. Vereker's claim is correct and directionality has nothing to do with how the wire is drawn, but rather is brought about by the application of hot plastic insulation, then cables made of bare wire simply inserted into plastic tubing would not exhibit any directional characteristics.
It would mean that enameled wire (i.e. magnet wire) using solvent-based coatings rather than extruded thermoplastics should also be immune to directional effects.
se
Why is it that ONLY physical causes are suspected and NEVER psychological causes?
To suspect that the perceived differences may be due to well-established psychological phenomena is on just a firm ground as suspectiving hot plastic affecting the crystalline structure.
That's another trait of the REAL "big boys." They don't dismiss other possibilities out of hand because they threaten preconcieved beliefs.
By the way, if Mr. Vereker's claim is correct and directionality has nothing to do with how the wire is drawn, but rather is brought about by the application of hot plastic insulation, then cables made of bare wire simply inserted into plastic tubing would not exhibit any directional characteristics.
It would mean that enameled wire (i.e. magnet wire) using solvent-based coatings rather than extruded thermoplastics should also be immune to directional effects.
se
Hi,
Unfortunately they do just the same...
I absolutely agree, a good filmcap would look like a block of metal when sawn in half.
Cheers,😉
Unfortunately they do just the same...
I absolutely agree, a good filmcap would look like a block of metal when sawn in half.
Cheers,😉
At this time, I cannot exclude some subtle non-linear property of the dielectric of the wire material covering, as my reference cables both use teflon. However, I have not seen any direct correlation to this property to the dielectric used, at this time. Therefore, for safety sake, let's everyone use teflon in our interconnects🙂
Now when it comes to non-linear effects in caps. We have measured them for decades. By far, the most interesting distortion, comes from typical ceramic caps used in cost-effective situations. This is best brought out by modifying a TEK 577 curve tracer to show gross deviations from ideal.
Now when it comes to non-linear effects in caps. We have measured them for decades. By far, the most interesting distortion, comes from typical ceramic caps used in cost-effective situations. This is best brought out by modifying a TEK 577 curve tracer to show gross deviations from ideal.
- Status
- Not open for further replies.