[ab initio]

Hi all,
I was browsing about trying to understand why anybody needs $40k interconnects to hear Aretha Franklin's farts (http://www.amazon.com/review/R33X3TA6E7RIYT) when the scientific grade Ghz bandwidth bnc cables cost in the $10s.

It seems to me the weak link should be in the connectors... namely in the physical connection itself. Does anyone know of a good source of technical information regarding the electrical caracteristics of the physical connection between mating surfaces? Im curious how the impedance (and uncertainty) of this interface compares to that of the cables themselves.

In the lab we can measure really high speed events using standard cables. I think the trick is having the appropriate termination on the scope side in relation to the source. No silver 6 sigma cables required, although a loose connector can transmit imformation faster than the speed of light....(http://www.foxnews.com/scitech/2012/02/22/loose-wire-led-to-stunning-faster-than-light-particle-finding/)

Cheers

 

[jaddie]

Quote:

Hi all,
I was browsing about trying to understand why anybody needs $40k interconnects to hear Aretha Franklin's farts (http://www.amazon.com/review/R33X3TA6E7RIYT) when the scientific grade Ghz bandwidth bnc cables cost in the $10s.

It seems to me the weak link should be in the connectors... namely in the physical connection itself. Does anyone know of a good source of technical information regarding the electrical caracteristics of the physical connection between mating surfaces? Im curious how the impedance (and uncertainty) of this interface compares to that of the cables themselves.

The properties of connection surfaces are well known, but get down to mostly issues of reliability.  The main problem is non-condutive oxidation, and there are two ways around it: higher contact pressure, and more stable materials that oxidize more slowly.  The contact pressure aspect has several dimensions to it, pressure in the contact measured in pressure per unit area, the type of service the contact is used for as a function of voltage and current applied, and the surface materials.  High contact pressure reduces the rate of contact failure caused by oxidation.  Higher voltage and current applications cause small amounts of oxidation to be "burned through" helping to maintain contact reliability. Certain metals oxidize slowly, or not at all.  Gold is the most popular, though low karat gold contains enough non-gold material that it does oxidize, so 24K is most desired for low signal applications.  High pressure point contacts are also used.  Extremely low resistance contacts require silver plating.  The common nickel over brass combination we find in audio jacks turns out to be more than adequate for most audio applications, with gold over brass or copper being slightly more reliable, but only slightly.  
 
Contact resistance is specified in milli-ohms, and often is stated as a minimum and maximum over a certain number of contact cycles. 30 to 60 mOhms is fairly typical with a doubling of the initial contact resistance after 10,000 to 50,000 cycles. 
 
Considering all of that, contact resistance has no effect on the audio it carries.  Connector impedance is a function of its construction, and also a non-factor in audio, though it's perfectly fine to use a 50 Ohm RG-58 cable with the appropriate BNC on it if you like, but of no advantage at audio frequencies.
 
 
Quote:

In the lab we can measure really high speed events using standard cables. I think the trick is having the appropriate termination on the scope side in relation to the source. No silver 6 sigma cables required, although a loose connector can transmit imformation faster than the speed of light....(http://www.foxnews.com/scitech/2012/02/22/loose-wire-led-to-stunning-faster-than-light-particle-finding/)

Cheers

 
High speed events are measured by taking cable delay into account first, or measuring differentially, in which case propagation delay is not a factor.
 
The story you reference showed how a loose optical connection caused an error in measurement of the speed of neutrinos.  The loose connector did not transmit faster than light speed.  
 
All electrical cables have propagation delay, with typical figures in the 60% to 80% (of the speed of light) range.  In most cases the amount of delay is much less important than knowing what it is.

 

[ab initio]

Hi Jaddie, thank you for your reply.
 
As per the thread title, I was hoping for some links or references to scientific literature. Do you have references or links for the discussion on materials, oxidation effects, connector design (preferably for audio applications), etc...? It would be really great if you could share a reference where you found all that info on contact resistances. I'd also be interested to learn how they measure those figures.
 
I would assume that the contact area would play an important role in the contact resistance.
 
Quote:

The contact pressure aspect has several dimensions to it, pressure in the contact measured in pressure per unit area

I think you mean force per unit area

What is actually important? Is it the contact pressure or the  total contact force?
 
 
Quote:

Considering all of that, contact resistance has no effect on the audio it carries.  Connector impedance is a function of its construction, and also a non-factor in audio, though it's perfectly fine to use a 50 Ohm RG-58 cable with the appropriate BNC on it if you like, but of no advantage at audio frequencies.

 
This is exactly what I'm looking for! How is the impedance dependent on construction? How does this affect transmission of audio bandwidth signals? Is this something I'd find in Horowitz and Hill?
One of the pro-expensive-cable arguments is that expensive cables might somehow provide superior impedance matching along the signal chain that leads to minimized signal reflections, and consequently minimize problems arising from signals reflecting back into the source. Obviously, the ability of the source to handle these back reflections will vary from component to component, but the electrical characteristics of cables + connectors to audio band signals should be easy to specify, right? Is there literature that outlines this analysis?
 
You also only mention contact resistance. Is there an appreciative amount of connector capacitive that would contribute to the complex impedance of a cable? Where can I find this answers?
 
 
Quote:

The story you reference showed how a loose optical connection caused an error in measurement of the speed of neutrinos.  The loose connector did not transmit faster than light speed.

 
Pardon my sense of humor, I forgot about Poe's law!

 
Cheers

 

[jaddie]

Quote:

Hi Jaddie, thank you for your reply.
 
As per the thread title, I was hoping for some links or references to scientific literature. Do you have references or links for the discussion on materials, oxidation effects, connector design (preferably for audio applications), etc...? It would be really great if you could share a reference where you found all that info on contact resistances. I'd also be interested to learn how they measure those figures.

Yes, sorry for no links.  My references are all print, hard copy, and from quite a while ago.  I researched contact parameters and materials for a design project quite a few years ago, and the information has stuck with me.   I'm not in my office now, and they are, so it'll have to wait a while.  I could google it, but then, so can you.
 
Contact resistance measurements are done with milli-ohm meters, which are a bit special, and have a way to compensate-out the test leads and their own contacts.  Fairly expensive gadget unless you need it all the time.  
 
Quote:

 
I would assume that the contact area would play an important role in the contact resistance.
 
I think you mean force per unit area

What is actually important? Is it the contact pressure or the  total contact force?

Yes, area is important, but for small signals contact pressure is more important.  Probably force per unit area is more correct, though it's measured in something like pounds per square inch. 
 
Quote:

This is exactly what I'm looking for! How is the impedance dependent on construction? How does this affect transmission of audio bandwidth signals? Is this something I'd find in Horowitz and Hill?

Cable and connector impedance is totally a construction, and mostly dimensional issue, conductor spacing, dielectric type, etc.  In practical terms, it has absolutely NO effect on audio signals, however. It's only a factor in impedance matched transmission line systems where the cable length is longer than 1/4 of the wavelength of the signal being carried.  Audio interconnects are not an impedance matched,  power transmission-based system.  The source is low, then load/termination is high, its a voltage transmission system.   I don't own H &H. 
Quote:

One of the pro-expensive-cable arguments is that expensive cables might somehow provide superior impedance matching along the signal chain that leads to minimized signal reflections, and consequently minimize problems arising from signals reflecting back into the source. Obviously, the ability of the source to handle these back reflections will vary from component to component, but the electrical characteristics of cables + connectors to audio band signals should be easy to specify, right? Is there literature that outlines this analysis?

The analysis would be correct for RF signals, incorrectly applied for audio signals over short distances (meaning under 10 miles or so).  In an impedance matched transmission line system, where the wavelengths being carried are shorter than the line itself, matching the source, cable, and load impedance maximizes energy transfer, and minimizes reflections. Reflections become an issue in a system like that because the reflected wave produces standing waves along the line, and power is also reflected back to the source.  If transmission lines are long enough with respect to the wavelength carried the timing of the reflection is also an issue.  However, this has nothing to do with audio.  Audio wavelengths are huge compared to the length of the wire. We aren't working in an impedance matched system, and the total propagation delay of an interconnect is too tiny to be even considered a fraction of an audio wavelength.  No power is reflected from the load at all.  And this is again a voltage transmission system, so very little actual power is transferred to the load anyway.  So the electrical characteristics of the cable and connectors are way out of the picture, and apart from a small group of cases involving the complex load of a speaker and some rather odd cable, the cable and connectors typically have no effect on the signal carried.  We could include in "odd cable" that with extremely high inductance, impedance or resistance.  Normal cable would have low R, L and minimal C, all of which would be below the point of affecting the signal.

You also only mention contact resistance. Is there an appreciative amount of connector capacitive that would contribute to the complex impedance of a cable? Where can I find this answers?

I think that's covered.  I don't mention connector capacitance because it's not a factor, and neither is the complex impedance of a cable until it gets to be either unusually high or the cable is extremely long.  

 

[ab initio]

Again, thank you for your reply! This is good stuff. I'm interested in putting numbers on all of these effects (unless it's already been done before, in which case I think it would be useful to document the data here). In my opinion, it would be extremely useful to quantify these various effects and I don't think it should be difficult to do. There are a lot of fanatical claims, both here on HF and abroad, regarding good cables vs bad cables and the possible physical explinations as to how differences between cables arise.
 
It seems like it should be easy to quantify the effects on a transmitted signal given source with specified properties, a cable with specified properties, and a destination with specified properties. I am interested in seeing exactly what sort of effects on the signal is possible given typical parameters. That said, I need some papers and/or texts as a starting point so all of the appropriate effects can be addressed in a scientifically rigorous way.
 
Cheers

 

[esldude]

jaddie has given you excellent information here.  The short answer is there are no effects worth bothering about in typical audio cables.
 
Unfortunately, I can tell you from experience, your chances of refuting fanatical claims, at least to change the fanatics mind are somewhere well below 1%.  People want to believe.  You can refute them simply or in detail.  No level of additional understanding or debunking is ever good enough to convince those not wishing to be convinced. 
 
On the other hand, as a one time fanatic myself it can happen.  Would happen more often if there wasn't an industry making money and entertaining people on false premises.  Even here such views are mostly relegated to one little sub-forum which is more than you get in nearly all other audio hobbiest forums.  Mostly you get shouted down en masse or banned by the owner of the forum.
 
If you wish to learn for your own satisfaction, then it all depends on how far you want to go.  How much effort you are willing to expend.  Just understanding LCR effects is a good start.  And you may have that already as I don't know your level of knowledge obviously.  Won't be hard to perform searches and find some basic articles on that.
 
Another thing worth knowing is a bit about how humans commonly get fooled hearing differences in audio.  Then when something looks technically unlikely and psycho-acoustically fits a perception bias you will usually see if something has much chance to be real or not.  With nearly all audio cables the answer is not. 

 

[mikeaj]

I'm forgetting the names and links I should be remembering, except for one:
 
nick_charles did some practical tests on interconnects here:
http://www.head-fi.org/t/405217/my-cable-test-enterprise
 
Of course, that's comparing entirely different cables to each other and not breaking down things as cleanly as you want, but given the differences seen there (and in others' tests), it's hard to find motivation to rigorously address those concerns on a piece-by-piece basis.

 

[ab initio]

Quote:

jaddie has given you excellent information here.  The short answer is there are no effects worth bothering about in typical audio cables.
 
Unfortunately, I can tell you from experience, your chances of refuting fanatical claims, at least to change the fanatics mind are somewhere well below 1%.  People want to believe.  You can refute them simply or in detail.  No level of additional understanding or debunking is ever good enough to convince those not wishing to be convinced. 
 
On the other hand, as a one time fanatic myself it can happen.  Would happen more often if there wasn't an industry making money and entertaining people on false premises.  Even here such views are mostly relegated to one little sub-forum which is more than you get in nearly all other audio hobbiest forums.  Mostly you get shouted down en masse or banned by the owner of the forum.
 
If you wish to learn for your own satisfaction, then it all depends on how far you want to go.  How much effort you are willing to expend.  Just understanding LCR effects is a good start.  And you may have that already as I don't know your level of knowledge obviously.  Won't be hard to perform searches and find some basic articles on that.
 
Another thing worth knowing is a bit about how humans commonly get fooled hearing differences in audio.  Then when something looks technically unlikely and psycho-acoustically fits a perception bias you will usually see if something has much chance to be real or not.  With nearly all audio cables the answer is not. 

I esldude, thanks for your reply!
 
I'm all about learning---I'm a professional student. I have a very very modest background in electrical engineering, gained mostly through problems encountered along the way in my work. I'm vaguely familiar with the reception that the scientific method receives here, which is why I've restricted myself to this corner of the forums.
 
I welcome any level of explanation from LCR circuit analysis to EM field theory. Wikipedia is a fantastic source of internet information. Texas instruments, Burr Brown, Analog Devices and others also have some nice technical reports available that discuss various design aspects for filters, amplifiers, etc...
 
I'm looking for a complete understanding what is happening along the audio chain. This forum is replete with contradictory opinions on just about everything, which is why I've asked about citations or links to literature on these topics. I agree that jaddie is providing honest, good information, as well as yourself. However, for every "The short answer is there are no effects..." there is another "I hear XXX differences between these cables". Just because I believe your opinion, doesn't make it sound science

I want rigorously establish, from first principles, what happens to Mr. Electron as he travels down Monoprice Avenue compared to when he strolls down AudioQuest lane. Will there be a difference? Of course. Will it matter for audio? Doubtful. But, I want to understand the effect and the relationship it has with the various materials, connectors, braid, etc... for my own satisfaction. Generalized statements and opinions from (possibly very well informed) strangers on the internet are certainly informative for me to focus my attention on certain aspects, but I still need scientific literature to be rigorous.
 
Maybe there are other nerds out there that would be interested in rigorous explanations of the principles of operation of their audio gear.

 
Cheers!

 

[jcx]

interestingly the net "current carrying" electron drift speed in Cu, Ag, other good metallic conductors is way slower than a stroll at common audio transducer currents, cable wire size: order of 1 meter / hr

 

[ab initio]

Quote:

I'm forgetting the names and links I should be remembering, except for one:
 
nick_charles did some practical tests on interconnects here:
http://www.head-fi.org/t/405217/my-cable-test-enterprise
 
Of course, that's comparing entirely different cables to each other and not breaking down things as cleanly as you want, but given the differences seen there (and in others' tests), it's hard to find motivation to rigorously address those concerns on a piece-by-piece basis.

Thanks Mikeaj,
 
This is a great place to start to give a general idea how these this behave in real life!
 
Does anybody have this text? It seems relevant:
Kaiser, Kenneth L. Transmission lines, matching, and crosstalk. CRC PressI Llc, 2006. [clicky]
He begins a discussion on hi-fidelity speaker wire on page 2-3
 
[edit]: here's another interesting google scholar find with an available PDF:
Brown, Jim. "The Effects of Cable on Signal Quality." Sound & Video Contractor 8.9 (1990): 22-33.  [clicky]
 
[edit #2]: Ha ha jcx, trivia fact of the day

I admit I was a bit skeptical---so here's the wikipedia article [clicky]. Now that I've seen it, I vaguely recall having heard it before.
 
Cheers

 

[jaddie]

Cables as related to audio applications:
 
"Amplifier - Loudspeaker Interfacing" by R. A. Greiner, J-AES, May, 1980
I consider this the premier work on the subject.  Deals with the question of transmission line effects as well as all other electrical parameters in cable
 
"Cables and the Amp/Speaker Interface", by R. A Greiner, Audio, August 1989
Sort of a re-hash and expansion on his 1980 AES paper, but worth the read
 
Try googling "transmission line theory" for a few references on that.  Also, check the ARRL Handbook for articles about transmission line, impedance matching and the concept of reflected power from a load mismatch.  The articles will all be about radio frequencies, but it's good to understand, so you can see why it doesn't apply to audio frequencies.
 
I'm sorry, I'll have to post contact materials references later.  Like I said, they are articles from fairly old technical journals, either the J-IEEE or EDN.
 
You should also be aware that most of us here are not professional students, and while we may have facts available for recall from experience, someone in your position has far more time and facility to research these principles.  Any good technical library would have all the references you'd ever need about pretty much everything you've asked about.  For some of us, the accumulation of useful facts has been a life-long course study, and it's not something that is always easily conveyed in a few short posts in a forum.  Part of appreciating knowledge is appreciating the effort it takes to obtain.  Demanding a complete bibliography here doesn't seem quite fair when, as a student, you have access to libraries and resources that others don't. 

 

[ab initio]

Quote:

Cables as related to audio applications:
 
"Amplifier - Loudspeaker Interfacing" by R. A. Greiner, J-AES, May, 1980
I consider this the premier work on the subject.  Deals with the question of transmission line effects as well as all other electrical parameters in cable
 
"Cables and the Amp/Speaker Interface", by R. A Greiner, Audio, August 1989
Sort of a re-hash and expansion on his 1980 AES paper, but worth the read
 
Try googling "transmission line theory" for a few references on that.  Also, check the ARRL Handbook for articles about transmission line, impedance matching and the concept of reflected power from a load mismatch.  The articles will all be about radio frequencies, but it's good to understand, so you can see why it doesn't apply to audio frequencies.
 
I'm sorry, I'll have to post contact materials references later.  Like I said, they are articles from fairly old technical journals, either the J-IEEE or EDN.
 
You should also be aware that most of us here are not professional students, and while we may have facts available for recall from experience, someone in your position has far more time and facility to research these principles.  Any good technical library would have all the references you'd ever need about pretty much everything you've asked about.  For some of us, the accumulation of useful facts has been a life-long course study, and it's not something that is always easily conveyed in a few short posts in a forum.  Part of appreciating knowledge is appreciating the effort it takes to obtain.  Demanding a complete bibliography here doesn't seem quite fair when, as a student, you have access to libraries and resources that others don't. 

This is great info and exactly the kind of direction I was hoping to find.

I did a bit of googling, which is how I found the items I listed above---hooray google

! The JAES papers give me a nice place to start exploring web of science and google scholar.
 
I apologize if I come off as demanding---it is only because I am curious! I am not familiar with the EE literature, which is why I came here and asked about literature references. I deeply appreciate your patience with me, sharing both your experience and these references, which are a big help to me and anybody else who is taking the initiative to learn the science behind our hobby.
 
Thanks again for another helpful response

 
Cheers