[Digitalchkn]

 
Yes. The skin effect makes the cable's resistance increase with increasing frequency. but I'm pretty sure that it's been shown numerous times in this forum that the skin effect is negligible at audio frequencies for a reasonably sized cable.
 
Cheers

 
Probably true, but hard to say without knowing details of how the cable is constructed. What is clear, is that thinner cables with skinnier wires and narrower insulators raise all these values and consequently are more lossy. So there is your rule number 1. Try to avoid skinny cables.

 

[ab initio]

   
Probably true, but hard to say without knowing details of how the cable is constructed. What is clear, is that thinner cables with skinnier wires and narrower insulators raise all these values and consequently are more lossy. So there is your rule number 1. Try to avoid skinny cables.

I think we're going to have to make a check list for buying cables. This is a good start. Should we make number 2 to get the shortest cable that suits the application?
 
To make it truly useful, it would be nice to have a target thresholds for minimum gauge, maximum length, etc... Fortunately, this has already been complied for speaker cables. For interconnects, i think the rule is keep the length under 1 kilometer 

 
Cheers

 

[Digitalchkn]

 
I think we're going to have to make a check list for buying cables. This is a good start. Should we make number 2 to get the shortest cable that suits the application?
 
Cheers

 
That's a good number 2. For same kind of cable shorter is always going to be more "transparent".  That's generally true.
As another example, consider your 10M cable for use with headphones. Here your amp impedance may be say an ohm and you could be driving headphones of say 32ohms. Here cable resistance and more importantly inductance have much greater impact that the capacitance. This scenario gives us 0.6dB loss @ 20KHz.  How do we fix it? Same as before. Make conductors wider and length shorter.

 

[mikeaj]

  Even without specializing in a closely related topic within EE, don't all students take the fundamentals of E&M and circuits as part of their degree programs? Isn't part of the classwork justifying the simplified formulas by starting from first principles, estimating the order of magnitude of higher order terms, and setting criteria for when those terms can be safely neglected?

 
Yes, they should all take the classes. The quality of instruction and understanding is another matter though. 

  Okay, now to be real, sometimes discussions on "when to use this" and why and some practical considerations may be understressed.
 
 

  I'm pretty sure that most electrical engineers know immediately that transmission line effects are essentially non-existent  in audio, yet that is one of the fallacies purported by cable-believers to explain why the flashier expensive cable sounds more airy or whatever. is a few picofarads of cable capacitence or a couple microhenries of self inductance really going to cause more than 1 dB of rolloff in the audioband, or cause any properly designed amplifier to oscillate wildly? Zero people, including every single cable manufacture has been able to show any case study where those types of causes (which their esoteric designs try to address) have an audible effect.

 
More or less.
 
 

 
as far as pyscoaucoustics go, I'm not really concerned with what people think they hear. I'm worried about whether it is physically possible under any circumstance that the only possible audibly transparent cable costs $1000 (or even $100).

 
I'm saying how much psychoacoustics EEs know. Somebody might not realize how much variance and issues are in the transducers already or have no idea how much distortion is audible, etc. Actually, it seems like people here have been misunderstanding what you've meant with this line of questioning, and now I think I get it too. This point isn't really relevant to what you're really getting at.
 
 

  No, I'm not misunderstanding what an electrical engineer is. Electrical engineers apply the fundamentals of electricity and magnetism to solve real world problems. If they are going to apply the fundamentals of E&M, they had better learn something about it at some point. I'm pretty sure every EE has learned about Maxwell's equations at some point. I'm not saying all EE's know audio. I'm picking EE's because the transmission of electrical signals in cables falls into their jurisdiction.

 
How about those who work in such areas as controls (and say robotics, sensing, etc.), signal processing, computer networks, computer architecture, communications (thinking error-control coding, modulation, signals, not RF / antennas / etc.), and so on?
 
 
This is all kind of besides the point though.

 

[Digitalchkn]

   
How about those who work in such areas as controls (and say robotics, sensing, etc.), signal processing, computer networks, computer architecture, communications (thinking error-control coding, modulation, signals, not RF / antennas / etc.), and so on?
 
 
This is all kind of besides the point though.

 
Takes a fairly rudimentary course in electronics to understand what you would and wouldn't get with a $1000 cable -- no need to get into Maxwell's.
 
But if you are willing to spend a $1000 on a cable I've got a better suggestion. Send me the $1000 and I'll send you a cable that will get the job done.

 

[ab initio]

  How about those who work in such areas as controls (and say robotics, sensing, etc.), signal processing, computer networks, computer architecture, communications (thinking error-control coding, modulation, signals, not RF / antennas / etc.), and so on?
 

The controls bit I think is equally present in mechanical engineering as in EE. It's all differential equations anyway! The signal processing, computer networking, architecture, etc. was a separate degree from EE at my undergrad university. I call that computer engineering. Honestly, in the end I would hope that the overwhelming majority of engineers would have had some training in electronics and electronic circuits; however, it does seem that a lot of engineers slip through their degree programs with some gaping holes in some of their training. I was really upset when my EE roommate in college admitted he didn't know what a truss was and that he didn't care because "there's nothing new to learn in mechanics". At the same time, I was really mad when I found out that civil engineers were getting degrees without ever taking a course in fluid mechanics. I'll admit that I've never had a course in controls, although I think i've picked up quite a bit along the way.
 
 

   
Takes a fairly rudimentary course in electronics to understand what you would and wouldn't get with a $1000 cable -- no need to get into Maxwell's.
 
But if you are willing to spend a $1000 on a cable I've got a better suggestion. Send me the $1000 and I'll send you a cable that will get the job done.

Brilliant!
 
Cheers

 

[esldude]

   
That's a good number 2. For same kind of cable shorter is always going to be more "transparent".  That's generally true.
As another example, consider your 10M cable for use with headphones. Here your amp impedance may be say an ohm and you could be driving headphones of say 32ohms. Here cable resistance and more importantly inductance have much greater impact that the capacitance. This scenario gives us 0.6dB loss @ 20KHz.  How do we fix it? Same as before. Make conductors wider and length shorter.

Sorry, not quite.  Shorter isn't more transparent.  Transparent is transparent.  After that you couldn't perceive a difference.  So shorter isn't more transparent.

 

[Digitalchkn]

  Sorry, not quite.  Shorter isn't more transparent.  Transparent is transparent.  After that you couldn't perceive a difference.  So shorter isn't more transparent.

 
And now we're getting into psychoacoustics and subjective listening and so forth which I don't find to be the topic at hand.
 
Let's just say shorter cables are more transparent in their signal transfer characteristics. And yes cable inductances capacitances and resistances are easily measurable quantities that scale with length, so there I don't think there is a need to get into discussions whether length of conductors is irrelevant from standpoint of electrical characteristics.  In that sense shorter is more transparent, or opaque if you will.

 

[mikeaj]

 
Takes a fairly rudimentary course in electronics to understand what you would and wouldn't get with a $1000 cable -- no need to get into Maxwell's.
 
But if you are willing to spend a $1000 on a cable I've got a better suggestion. Send me the $1000 and I'll send you a cable that will get the job done.

 
I'm just saying that you wouldn't be a citable expert or even absolutely positive from a rudimentary course or even a degree. Knowing some can be more dangerous than knowing less. What if there were some effects -60 dB down or so that don't matter for the vast majority of applications and aren't mentioned in most courses, except they matter for some situation or two in audio because that can be relevant in audio? Or something to do with slight differences in timing, considering that a lot of the analysis is focused on steady state? (not that there are)
 
What I mean is that you can have the right answer with not necessarily the right reasoning or all the steps.
 
Would you know offhand how "off" lumped sum model is for 50 feet of 12 AWG speaker wire at 10 amps to a 4 ohm load? 
 
 
Kilobuck cables, exotic constructions, and the marketing crud are a different conversation (i.e. the one everybody else is having, I think)

 

[elmoe]

   
I'm just saying that you wouldn't be a citable expert or even absolutely positive from a rudimentary course or even a degree. Knowing some can be more dangerous than knowing less. What if there were some effects -60 dB down or so that don't matter for the vast majority of applications and aren't mentioned in most courses, except they matter for some situation or two in audio because that can be relevant in audio? Or something to do with slight differences in timing, considering that a lot of the analysis is focused on steady state? (not that there are)
 
What I mean is that you can have the right answer with not necessarily the right reasoning or all the steps.
 
Would you know offhand how "off" lumped sum model is for 50 feet of 12 AWG speaker wire at 10 amps to a 4 ohm load? 
 
 
Kilobuck cables, exotic constructions, and the marketing crud are a different conversation (i.e. the one everybody else is having, I think)

 
Which is why we have double blind testing. Nobody has ever yet been able to have any kind of significant results in telling cables apart in a DBT. I'm gonna quote bigshot for once 

, if you can't hear it then it doesn't matter.

 

[OddE]

To answer the OPs question:
 
I am an electrical engineer; I majored in RF design and know a thing or two about transmission lines. Audio interconnects are not transmission lines - unless your source is in California and your amp is in, say, Idaho.
 
The most I have ever spent on an interconnect was $50 or so - and that was simply because they were out of the $25 and $15 ones, and I was out of cable to just DIY what I needed.
 
Most of the time while in university, I DIYed cables - I made sure I always had a supply of Neutrik 373s* on hand, bringing out the soldering iron whenever someone needed an interconnect.
 
Audiophiles were pretty impressed when the cables came with a frequency response plot showing that they were perfectly flat in the audio passband. I used whatever well-shielded cable I had on hand - most popular was the Semi-Flex coax normally used for microwave lab interconnects. (Way, way, WAY overkill for this application, but it was available, it was easy to strip and solder - and it looked cool.)
 
*) Probably the cheapest high-quality RCA connectors out there - at least the cheapest I found locally...

 

[TheAttorney]

Also to answer the OP's question, I've got a degree in Electrical & Electronic Engineering (but then moved to computers for work).
 
There was nothing in that 3 years of education that would even get close to explaining the alleged advantages of a $1,000 cable.
As has been mentioned, one would need to specialise much further in order to claim to be a true expert.
 
As an aside, one of the topics was Wave & Transmission Theory (may have included Field Theory, but it's all a blur now) - the most impenetrable and tedious subject I've ever come across. Contained more maths than my Pure Maths education. The lecturer was so incapable of bringing this to life, that the university allowed us to bring all our books and notes into the final exam - and most of us still failed that topic. Had I paid more attention, followed by a few years of specialist research, I may of had something scientific to say on the subject of audiophile cables.
 
In my the early hifi days, I tried to make scientific sense of it all, but it was futile trying to match what I heard with any objective data, so I just gave up and became a subjectivist.

 

[elmoe]

  Also to answer the OP's question, I've got a degree in Electrical & Electronic Engineering (but then moved to computers for work).
 
There was nothing in that 3 years of education that would even get close to explaining the alleged advantages of a $1,000 cable.
As has been mentioned, one would need to specialise much further in order to claim to be a true expert.
 
As an aside, one of the topics was Wave & Transmission Theory (may have included Field Theory, but it's all a blur now) - the most impenetrable and tedious subject I've ever come across. Contained more maths than my Pure Maths education. The lecturer was so incapable of bringing this to life, that the university allowed us to bring all our books and notes into the final exam - and most of us still failed that topic. Had I paid more attention, followed by a few years of specialist research, I may of had something scientific to say on the subject of audiophile cables.
 
In my the early hifi days, I tried to make scientific sense of it all, but it was futile trying to match what I heard with any objective data, so I just gave up and became a subjectivist.

 
I would be interested to see any of this specialized data about cables that was so tedious and hard to understand that you gave up trying and started buying expensive cables.
 
Could you explain how you came to the conclusion that one would need much further specialization in order to know the real differences between "hi-end" cables and their standard counterparts?
 
When trying to match what you heard with any objective data, did you also double blind test these cables to make sure you actually heard what you think you heard? 

 

[superjawes]

Oh hey, I'm an EE, and I do wire harnesses for a living (in heavy trucks)!

...okay, it's not really audio...Still, the general practices can apply.

First off, make sure you have the right gauge. Larger wire (lower gauge number) has lower resistance. The line resistance will sap power as losses in a DC environment, but it can also mess with your reactive devices by altering the impedance relationship between amp and speaker. The Speaker Wire page was already linked if you want more info. With headphones, it is very easy to stay within acceptable ranges.

Second, harness (or cable) construction. This is really about convenience when it comes to headphones (at least IMO). With wire harnesses, you want to minimize strain on the wires, and that sometimes means allowing for a certain degree of movement, especially between suspensions. You also want to consider what kind of wear can happen to the wires and shield appropriately. Take a took at a heavy truck some time. You'll see the chassis harness surrounded by convoluted tube, and if it passes near exhaust, you'll see a heat wrap. You can apply this lesson to headphones by choosing a cable construction that meets your needs and environment. Certain outer materials are going to keep the cable "slick" relative to most surfaces (if you're moving it around), and certain materials will generally be more rugged, resistant to damage.

Finally, routing. This has a bit to do with cable/harness construction, but it's probably a better lesson for interconnects. Just be mindful of where the cable needs to go and how it's going to get there. Shorter is generally better so you don't accidentally get snagged. You might still need some length if your components are going to be far apart, and then you should consider ways to keep your cables from getting lost in the rat's nest of wires behind your components. Actually, you generally should try to prevent a rat's nest from forming.

So the primary reason, IMO, to use "better" cables in audio really come down to non-audio needs. The stock cable will probably sound just as good as a "better" one (assuming that designing engineer is earning his/her salary), but you can definitely put a price on convenience (not $1,000+, of course).

 

[Digitalchkn]

   
There was nothing in that 3 years of education that would even get close to explaining the alleged advantages of a $1,000 cable.

 
That's because there is no overwhelming advantage of a $1K cable at audio frequencies.