|   what is the role of capacitance in interconnects? | ibpolish
Sep 22, 2001 5:48 AM | | Mtry, in an earlier post you'd stated that capacacitance is an important criteria in interconnects. Could you and others of course, please elaborate on that statement? Is it ever dependent upon the nature of the equipment? Thanks in advance! |
| the role of capacitance in audio cables--AC electricity 101 | itellitlikeitis
Sep 22, 2001 6:38 AM | | Capacitance is characteristic of all cables. The amount depends on the way the cable is constructed (because that determines capacitance per unit length), and the length of the cable. In general, cable capacitance acts as what engineers call a parallel element which means that if it's great enough, it will attenuate high frequencies. How great is great enough? That depends on the rest of the circuit and can be calculated mathematically using well know electrical engineering formulas. Relatively short lengths of coax as inteconnects normally have no effect on audio signals except in some rare instances on the connection between magnetic phono cartridges and phono inputs on preamplifiers. Cable capacitance in speaker wires, especially those deliberately or inadvertently constructed with high capacitance "may" have a slight audible effect but much worse have been known in rare instances to create an electrically "resonant" load that can send some amplifiers into spontaneous oscillation. This can do severe damage to the amplifier and to the speakers. This is not some theoretical but never to be encountered in the real world occurance. Many repair technicians have reported the aftermath of these audiophile experiments. Normal non-audiophile cables do not present this risk. As for the expensive audiophile type--cavaet emptor. |
| You didn't mention where it might matter most. | Robot Czar
Sep 22, 2001 6:46 PM | | It has long be known that phono cartridges are sensitive to capacitance in interconnect cables (especially moving coil, I think). If you are still playing LPs you may want to pay attention to this as this is an example where cables CAN make an audible difference (in high frequency response). For line level signals you can generally ignore this, from what I have read. |
| re: what is the role of capacitance in interconnects? | Jon Risch
Sep 22, 2001 8:29 AM | | The idealized audio signal cable would have vanishingly low values of the three basic eletrical parameters: L, C and R. L is inductance, C is capacitance, and R is resistance.
Resistance is a fairly straight forward concept, and so I will not adress it in depth here. The more resistance, the more signal losses in a speaker cable due to voltage divider action. Resistance in an interconnect is usually of little consequence.
Inductance is usually only an issue for spaeker cables, which have a very low source impedance (the power amp), and a fairly low impedance load (the speaker). With these low impedances, what would ordinarily be minute amounts of inductance at higher circuit impedances, become enough to start rolling off the HF's in zip cords used for speaker cables. This is more likely to occur than HF roll-off due to IC coapacitance.
Capacitance comes into play when the source impedances start getting higher. For most typical audio electronics signal sources (CD player, preamp, DVD palyer etc.) the output impedance (think of impedance as a sort of AC resistance) is usually low enough that unless you tried to run an interconnect more than 6-10 feet, it wil not be an issue.
However, so-called passive line level "preamps", or passive level controls, can have a fairly high output Z, as can some tube equipment, and a few other very special cases, these can start to reach the levels of output Z that would interact with longer cables to cause a HF roll-off.
However, for most IC's, and most situations, the cable capacitance will not cause a HF roll-off.
Looking at JUST the simple LCR parameters though is not the whole story. There are secondary effects, such as insulation DA, motor-generator due to vibrations and EMF, and so on.
See:
http://www.nsc.com/rap/Application/0,1570,28,00.html
and
http://www.capacitors.com/pickcap/pickcap.htm
these talk about some of the secondary effects that cable capacitance can be a part of.
http://www.cardas.com/insights/measureresonance.html
which covers electrical resonances.
and this post, which covers a large number of references and handy URL's to check out on audio cable sonic causes and effects.
http://www.AudioAsylum.com/audio/cables/messages/27322.html
Jon Risch |
| re: what is the role of capacitance in interconnects? | ibpolish
Sep 22, 2001 10:29 AM | | does anyone have values that represent levels that one should not go over? and thanks for your kind and lengthy replies. tis appreciated. |
| re: what is the role of capacitance in interconnects? | itellitlikeitis
Sep 22, 2001 10:48 AM | | The value you should not exceed is a few dollars per cable. Anything more is a waste of money. |
| re: what is the role of capacitance in interconnects? | Jon Risch
Sep 22, 2001 10:59 AM | | Have any evidence for this claim at all, or are you also one of those who hide behind the "burden of proof" argument?
If you have nothing but what is considered BY SOME to be a lack of valid evidence, then you have essentially nothing to base this statement on. Wouldn't be the first time, nor will it be the last.
Jon Risch |
| Burden of Proof | itellitlikeitis
Sep 22, 2001 11:34 AM | | YES YES YES !!!!!!!!!!!!!!!!!!!!!!!!!!!!!
When someone introduces a product that performs the same function as others which have proven perfectly satisfactory over decades of use and charge substantially higher prices, THEN THERE IS A BURDEN OF PROOF ON THEIR PART TO PROVE THAT THEIR HIGHER PRICES ARE JUSTIFIED. And so far, they haven't. All they have are their salesmen and their toadies to help fleece the unsuspecting who are only trying to improve their enjoyment of their sound systems. |
| Burden of Proof | Jon Risch
Sep 23, 2001 8:19 AM | | I see.
You are going to hide.
In this post:
<a href="/crforum?14@@.eec3234/3">itellitlikeitis 9/22/01 11:48am</a>
you state:"The value you should not exceed is a few dollars per cable. Anything more is a waste of money."
This is in effect a claim that the cables are not sonically different, and that one should not pay more money because they are sonically the same.
So where is your evidence? All the objectivists have so far is: Nothing. No evidence. Very specifically stated by the objectivists a "lack of evidence". In other words, you are basing you claim on nothing at all. If all there is, is a lack of evidence (and I dispute that, it is more in the nature of "evidence we will not accept", rather than a true lack of evidence.) and nothing more, this is a most tenuous position, with very little room to be making claims that are in effect, depending on a null to equate to a negative.
Since there have not been concerted efforts to conduct professional grade tests on audio cables, at least not ones that are considered to be acceptable as evidence, then the few amateur listening test nulls are not even good expedient evidence, just more an absence of anything useful.
Then you decide that it is all a con:
"All they have are their salesmen and their toadies to help fleece the unsuspecting who are only trying to improve their enjoyment of their sound systems."
This claim is not supported either, with details on typical business margins, cost to manufacture and assemble, plus advertising and distribution costs, and a typical high mark-up percentage, just like clothes, appliances, jewelry, etc.
That point aside, their is an other option: build yourown for a fraction of the cost of retail cables.
I post this information, see:
http://www.geocities.com/jonrisch/i1.htm
and
http://www.geocities.com/jonrisch/s1.htm
I charge no money for, and make no profit from these recipes.
So your objection on some sort of psuedo-moral grounds
is a moot point.
The DIY cables can be built for less than the cheapest entry level aftermarket cables that are widely considered to have a basic minimum
level of performance (the $40 range), for approx. $25 or so. Four jacks with teflon insulation that are WBT clones for $11, 6-7 feet of cable for a stereo pair at a few dollars a foot, some heat shrink, maybe some silver solder. These are widely regarded as performing to the level of aftermarket cables in the price range of $300-600 dollars.
Speaker cables using the CC89259 design can be built for around $60-80 a terminated ten foot pair, which is about the same or less than the most recommended entry level speaker cables (AQ Type 4, for $80-100 the terminated ten foot pair). These compete sonically with speaker cables that range from $500 to $1000 a pair.
Jon Risch |
| Burden of Proof | mtrycrafts
Sep 23, 2001 12:40 PM | | "This is in effect a claim that the cables are not sonically different, and that one should not pay more money because they are sonically the same."
Is that what you read into it?
"So where is your evidence?"
Evidence for the negative, Jon? What would satisfy you and make you abandon your podium for cable differences which, of course you cannot prove, shoe credible evidence, or demonstrate yourself.
" All the objectivists have so far is: Nothing. No evidence. Very specifically stated by the objectivists a "lack of evidence".
Yes, Jon, lack of evidence for audible cable differneces. How did you know that? BUt, if he is mistaken, I am sure you can post your citations for audible differneces in cables?
" In other words, you are basing you claim on nothing at all."
No, Jon, based on nulls and no evidence for audible differences in all these decades of trying. Why is that? Is it so difficult?
"If all there is, is a lack of evidence (and I dispute that, it is more in the nature of "evidence we will not accept", rather than a true lack of evidence.)"
Oh? What evidence you have, Jon? Your tests that you refuse to diplicate and demonstrate? Ben Duncan's claims? Vandy's claims who also ran when asked to duplicate his work? HT mag? Which two cables did they compare for differneces, JOn?
" and nothing more, this is a most tenuous position, with very little room to be making claims that are in effect, depending on a null to equate to a negative."
He certainly cannot base it on positive outcomes as there don't seem to be any that can be duplicated. You keep hiding.
"Since there have not been concerted efforts to conduct professional grade tests on audio cables, at least not ones that are considered to be acceptable as evidence, then the few amateur listening test nulls are not even good expedient evidence, just more an absence of anything useful."
So, you have no evidence for audible cable differences then. Didn't think you'd have one after all these years. |
| re: what is the role of capacitance in interconnects? | mtrycrafts
Sep 22, 2001 5:21 PM | | "Have any evidence for this claim at all, or are you also one of those who hide behind the "burden of proof" argument? "
Well, for starters, Burden of proof is applicable regardelss. Then, hwat will you accept and what cards do you hold? Thus far you hold no cards, so you have nothing to talk about.
"If you have nothing but what is considered BY SOME to be a lack of valid evidence, then you have essentially nothing to base this statement on. Wouldn't be the first time, nor will it be the last. "
Lack of evidence? Why isn't there any evidence, JOn? Why haven't you been able to demonstrate it? And don't give me cost strawman bs.
So far ic have yet to be differentiated unless broken. |
| Burden of proof-- are you pleading the fifth? | itellitlikeitis
Sep 23, 2001 12:02 AM | | The burden of proof is on the marketers who have introduced the new high priced product, not on the established manufacturers whose products were not challenged for their performance qualities for decades when the industry made major performance improvement strides in every other area. The readers of this billboard should know that the author of the posting I am responding to is a frequent contributor to another billboard where the prevailing culture is one where audio cables are regarded as a kind of religion that worships high priced products, does not allow discussion of abx or dbt (a-b tests or double blind tests) and has as a frequent contributor, a manufacturer of high priced cables. All we get instead of real discussions from their side are the same references to other web sites with arcane theories and abstract arguements in unfathomable techno-babble. Don't be fooled by what these people say. This cottage industry is out there to take your money with no real proof that their expensive product is any better than the inexpensive products they aim to replace. This nonsense has been going on for at least 20 years and until the FTC puts a stop to it the way they did with outrageous claims for performance of audio amplifiers in the 1970s, this will continue on and on and on and on. Don't become one of the suckers. If you are unhappy with the sound of your equipment, spend your money on something that will really help---a better amplifier or better speakers. |
| re: what is the role of capacitance in interconnects? | Jon Risch
Sep 22, 2001 10:54 AM | | This is a hard one to answer. There is no simple answer.
For instance, typical cheapo cables can run from 30-50 pF per foot. That's picofarads per foot. RCA ends add another 10-15 pF or so. So a 1M interconnect made with cheapo cable could run from a low of around 115 pF to a high of around 195 pF.
Longer lengths would be commensurately higher in capacitance.
Premium or high performance cables can get as low as 15-20 pF per foot, or around 1/2 the capacitance of the cheapies.
However, the sheer level of the capacitance is not as much of a factor (in most cases, I will elaborate below) as is the quality of the capacitance. For instance, foamed polyurethane can have a very low capacitance per foot. It also sound pretty bad too. Years ago, the less expensive "premium" cables coming out of Asia used this insulation material, and the cables were a dissapointment to many. Finally, the demnd for these cables fell off so much, that they are virtually unavaliable today, the manufacturers quit making them. Most switched to foamed PE, which is a very decent insulation material. Not as good as teflon, but one of the better dielectrics.
What is more of a problem for long interconnect runs, and I will define long IC runs as anything over 2M, or a little over 6 feet, is that the source device may not have anything other than a simple/cheap op-amp output stage, and these can usually only drive a few hundred pF without sonic distress. See: http://www.AudioAsylum.com/audio/tweaks/messages/1080.html
Some will break into oscillation, others will just souns awful, yet others progressively loose sound quality in terms of dynamics, transient response, clarity and smoothness.
A simple perusal of the output Z will not tell you if you are OK driving longer cables, in fact, with op-amp based outputs, the common practice is to use a series resistor to provide some level of short circuit protection, usualy in therange of 47 ohms to 220 ohms or so.
The lower the resistor value, the more susceptible the op-amp is to being capacitively loaded! So just because a signal source may spec a 47 ohm output Z or 100 ohms, this does not assure long cable drive is possible, or that it will sound OK.
Obviously, the premium cables have the advantage of being able to run twice as far as some cheapo's before these problems will reartheir head, and this may be far enough to let you "get away" with running an IC that far. A run of 10-15 feet or so with premium low C cables is possible, anything longer, and you MUST know if your source can handle it.
Jon Risch |
| ML No.380: 680pF | Jon Risch
Sep 23, 2001 8:25 AM | | This level of equipment is seldom being driven by mass market sources using op-amps that do not have the drive capability.
I am sure the Madrigal preamp will operate just fine into this load.
As will most other high end preamps/sources.
Jon Risch |
| ML No.380: 680pF | mtrycrafts
Sep 23, 2001 12:46 PM | | Is there a warning anywhere by Madrigal as to the quality of components to use with theirs? No warning? If they designed it to operate with only certain components, that should be so stated and warned against using lesser components. Until that time, you have a bogus argument. |
| ANOTHER TESTABLE CLAIM, JON! | mtrycrafts
Sep 22, 2001 5:28 PM | | "as is the quality of the capacitance."
Data, evidence can be found where? Sheer bs, nonsense on your part, Jon.
"For instance, foamed polyurethane can have a very low capacitance per foot. It also sound pretty bad too. "
More claims, Evidence and data, Jon can be found where?
"is that the source device may not have anything other than a simple/cheap op-amp output stage, and these can usually only drive a few hundred pF without sonic distress. "
Need more evidence, Jon, not speculations. |
| re: what is the role of capacitance in interconnects? | FLZapped
Sep 23, 2001 3:31 AM | | [However, the sheer level of the capacitance is not as much of a factor (in most cases, I will elaborate below) as is the quality of the capacitance. For instance, foamed polyurethane can have a very low capacitance per foot. It also sound pretty bad too. ]
And of course, there is direct, tangible evidence for this claim, right?
[Finally, the demnd for these cables fell off so much, that they are virtually unavaliable today, the manufacturers quit making them. ]
And you have the proof to back up this claim I suppose? These companies just let you waltz in and look over their books.
-Bruce |
| re: what is the role of capacitance in interconnects? | FLZapped
Sep 24, 2001 7:43 AM | | [For instance, typical cheapo cables can run from 30-50 pF per foot. ...Premium or high performance cables can get as low as 15-20 pF per foot, or around 1/2 the capacitance of the cheapies. ]
Oh, so now capacitance is the determing factor as to whether a cable is cheap or not. Cute. And you cricized someone else for making a comment about "cost per foot"......
-Bruce |
| Here's a rule of thumb | Norm Strong
Sep 22, 2001 11:35 AM | | If the output impedance of the source times the capacitance, in pf, of the interconnect is less than a million. you're safe. If you don't know the capacitance of the interconnect, use 100pf/m. |
| re: what is the role of capacitance in interconnects? | mtrycrafts
Sep 22, 2001 5:16 PM | | "Looking at JUST the simple LCR parameters though is not the whole story. There are secondary effects, such as insulation DA, motor-generator due to vibrations and EMF, and so on. "
Put some GD numbers on these secondary effects, Jon, instead of insinuating that it is of any importance and audible. Data, Jon, not speculation. But you have yet to do this. Why?
Especially that motor generation to vibration. DATA, NOT SPECULATIONS. |
| other than LCR parameters | mtrycrafts
Sep 22, 2001 9:07 PM | | Flex life? You play with your cables? I don't, so it's life is much longer than mine ever will be:-) Ordinary use is nothing.
"- Belden is not specializing in high-end audio cable market -. "
The don't? Are you sure? Do you know which high end cable companies buy from Belden? Lots.
Besides, what does this have anuthing to do with it?
You wouldn;t have numbers for that noise, would you?
And, finally, audibility data will do. |
| other than LCR parameters | mtrycrafts
Sep 23, 2001 12:53 PM | | "Belden is not specializing in high-end audio cable market] "
Oh, ok, now I see your point. They just sell to them and not their only livelyhood. Ok. |
| other than LCR parameters | Jon Risch
Sep 23, 2001 8:44 AM | | Every guitar player on the planet has dealt with triboelectric noise, they often purchase cables that minimize this effect for the reason that is it audible when the move the cord coming from the guitar.
Just about every cable manufacturer offers part numbers that address triboelectric noise specifically, and many other designs incorporate factorsd to try and keep it down to a low roar.
So real world usage and the experience of literally tens or hundreds of thousands of guitarists makes triboelectric effect and the consequent noises and spurious signals a reality. It is a well known phenomenon, and does not need one of your ever demanded "audibility proofs".
I think even Bruce/FLZapped could tell you this one mtry. You should contact him, and set him straight on the need for training before listening tests!
Jon Risch |
| other than LCR parameters | Steve Eddy
Sep 23, 2001 10:09 AM | | You wrote:
****So real world usage and the experience of literally tens or hundreds of thousands of guitarists makes triboelectric effect and the consequent noises and spurious signals a reality. It is a well known phenomenon, and does not need one of your ever demanded "audibility proofs".****
As a guitarist I can certainly attest to the effects of triboelectric noise in guitar cords.
However I looked through this entire thread and didn't once see anyone talking about guitar cords. You seem to be supporting your argument through nothing more than inference drawn from a wildly different context than what's being discussed here.
I don't know about you, but when I'm listening to my system, my interconnects and speaker cables aren't being stomped on, flailed about smacking onto the floor, into amp stacks, microphone stands, etc.
Instead of supporting your argument with such specious inference, why not put a small amount of effort into obtaining some empirical evidence so that one might gauge the consequences of triboelectric effect in a context more in keeping with the one at hand?
Just get a one or two meter length of interconnect that you would suspect to have the worst performance with regard to triboelectric effects (perhaps some cheap spaghetti wire). For a sort of worst-case scenario, terminate one end with a 1K resistor and the other end with a 100K resistor. Suspend the interconnect a foot or so directly in front of a decent loudspeaker and using an impulse signal at the highest amplitude you can safely reproduce with said speaker, run an FFT taken across the 100K resistor.
Based on the result of that, perhaps the issue of triboelectric effect can then be discussed a bit more intelligently, productively and in a more proper context.
se |
| other than LCR parameters | Jon Risch
Sep 23, 2001 11:05 AM | | Steve,
Hi, long time no post!
Anyway, yes, the guitar situation has impedances that are much higher than typical audio system impedances.
However, I got that the thrust was that triboelectric effects were not even real or relevant to cables, and just wanted to show that they did exist, and that they can be an issue for cables.
As always, it is a matter of the levels that would manifest during audio system usage.
Your suggested experiment may or may not show what we are looking for,
but it might help quantify at least a level for that particular experiment.
Jon Risch |
| other than LCR parameters | mtrycrafts
Sep 23, 2001 12:58 PM | | "As always, it is a matter of the levels that would manifest during audio system usage. "
Yes, it is, isn't it, Jon, as not all levels are audible which you do not want to acknowledge but try to convince people that it is audible. You just don't have the evidence for this audibility. |
| other than LCR parameters | Steve Eddy
Sep 23, 2001 1:46 PM | | <i><b>Hi, long time no post!</i></b>
Well, you know. Just can't keep my nose out of things for long. :-)
When I'm confident I can ignore certain posts from a certain you know who, I'll return to posting you know where as I much prefer it over uh... well, 'nuff said. :-)
<i><b>Anyway, yes, the guitar situation has impedances that are much higher than typical audio system impedances.</i></b>
Yes, the impedances tend to be higher, particularly the output impedance of the pickups, but amplifier input impedances are fairly commonly around 100k ohms, which isn't terribly uncommon in audio gear. Mostly it's the fact that there's so much voltage gain involved between the pickups and the speakers. Even cables for microphones, which have output impedances more like 100 ohms are designed for low triboelectric noise again due to the considerable voltage gain involved.
<i><b>However, I got that the thrust was that triboelectric effects were not even real or relevant to cables, and just wanted to show that they did exist, and that they can be an issue for cables.</i></b>
I didn't get the impression that anyone was claiming triboelectric noise didn't exist or even that it couldn't possibly result in any audible effect. Rather there was a claim of its existence with the implication that it was significant enough to have audible effects, but the claim was not supported with any meaningful information with regard to the magnitude of the effects one might expect to find under typical conditions (i.e. cables not being stomped on or flayed about).
<i><b>As always, it is a matter of the levels that would manifest during audio system usage.</i></b>
Certainly. But without some idea of what those levels may actually be, the claim then amounts to little more than hand-waving and speculation. That may suffice for politics, closing arguments in jury trials, and marketing, but it does nothing to bring any useful knowledge to the subject.
That's one of the main reasons I had such problems with that certain you know who. A lot of hand-waving, a lot of inference (along with a lot of personal mischaracterizations and insults), but on many of these kinds of issues, <u>zero</u> illumination. In other words, a lot of choppin' but no chips flyin'.
<i><b>Your suggested experiment may or may not show what we are looking for, but it might help quantify at least a level for that particular experiment.</i></b>
Perhaps. Though it would, if nothing else, take us a bit farther down the road than we are at present and hopefully improve the signal-to-noise ratio of future posts on this subject.
se |
| re: what is the role of capacitance in interconnects? | audioengr
Sep 26, 2001 1:50 PM | | These "second-order" effects are not easily measured. I have found that dielectric absorption and the signal dispersion that it causes to be one of the most audible second-order effects. This is the difference you hear when you go from insulated conductors to non-insulated conductors. These differences are very difficult to hear in mono, but are quite obvious in stereo. However, if you have other performance limiters in your system, you will probably not hear any difference, even in stereo. Takes a lot of tweaking (and money)to weed out the limiters.
One thing that I was able to measure is the effect of damage to the conductor crystal lattice. Immersion in LN02 of an interconnect caused damage that did not change the L,R or C measurements between 10Hz and 100KHz. However, it sounded really bad. A high-speed step showed that the micro-cracks in the cristal were causing reflections that were not there in an un-treated cable. This lends credence to the anecdotal evidence that really pure and single-crystal conductors do sound better. See the study results at:
http://www.empiricalaudio.com/frAudio_FAQ's.html |
| re: what is the role of capacitance in interconnects? | audioengr
Sep 26, 2001 12:27 PM | | Capacitance is the number one factor in interconnect performance. All manufacturers worth their salt should publish numbers for this. The only other real equipment interactions come with the component source impedance. Lower source impedance is always better. This is different than "characteristic impedance" such as that quoted for digital interconnects. Destination or input impedance is less of a factor. This is fairly easy to simulate on a computer and I have performed thousands of such simulations. Some of my results can be viewed on my webpage:
http://www.empiricalaudio.com/frAudio_FAQ's.html
and related pages on http://www.empiricalaudio.com |
| |
