Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Cable technology question (audio quality)

  1. Jan 16, 2013 #1
    hi, first post here

    if this is too long to parse please skip to the actual question in bold, at the end :)

    I wasn't able to find any other concise title for this. maybe some of you are aware of the fact that in audio some give great importance to cable selection. for some the audible differences are an accepted fact of life and the accepted wisdom is that they can't be explained by impedance (RLC) alone but that there are more "obscure" phenomena at play.

    I'm a computer egnineer (which includes a bit of EE) so I'm aware of the existence of phenomena like dielectric absorption, capacitance variance with voltage, EMI suspectibility (balanced cables), skin effect, transmission line behaviour (although the importance of the last two at audio frequencies can be questioned). also, note that I don't want to go into a discussion questioning the validity of the above claims, anyone considering this, please just ignore the thread, it is not its objective.

    so, here goes the actual question: is anyone aware of any field, any at all, where "special" cables are used and other phenomena except the obvious impedance need to be taken into account? by "special" I mean that unusual materials, geometries or construction techniques need to be used in order to ensure the least possible signal degradation. if it's so, I'd also like some study material to investigate further. I just want to look at this from an objective perspective, if there is any reason to consider it.

  2. jcsd
  3. Jan 16, 2013 #2
    Bear in mind that you have referred to R, L and C as discrete (also called lumped) properties.

    For some systems they are distributed. This applies particularly to the contribution to L and C.
  4. Jan 16, 2013 #3


    User Avatar
    Science Advisor

    There are no magic cables. Everybody buys generic cables designed to meet the needed specifications from the cable manufacturers. There are some "special" requirements for "RED" cable systems (compromising emanations) but it's not about audio signal fidelity.

    Last edited: Jan 16, 2013
  5. Jan 16, 2013 #4
    hi, thanks for answering

    this is why I mentioned transmission lines, at 20kHz (accepted max audible frequency) the wavelength is 15km. you'll hardly find any audio circuit of that size :) or am I missing something?

    what are the "RED" cable systems? your linked PDF mentions wireless systems so I guess they are high frequency cables.
    remember Arthur C. Clarke's law "any sufficiently advanced technology is indistinguishable from magic" :) so, I'm not asking about magic but rather about "advanced technologies", IF they indeed do exist (as in 'are needed').
    if I were to describe the "technologies" used by the audio guys: different copper types, silver, silver plated copper, different stranding patterns (twisted or straight), multiple interwoven isolated wires, exotic insulating materials etc. I was just curious if there's any field where at low frequencies a cable made of regular copper and insulated with PVC can prove to be unacceptable as long as it's not a matter of longevity or anything like that (UV exposure, fire hazard etc), there isn't any interference to pick and the length doesn't exceed few feet worst case. fact is that there are "special" looking audio cables but my guess is that at least some of them are sourced from generic manufacturers who make them for entirely different applications.

    for instance, this may sound very stupid, but would you think that a cable can be susceptible to vibration that leads to electrical signal degradation? or that maybe there is some sort of E/M interaction in the strands that can lead to loss? just asking, maybe this is already known and addressed in some super high tech NASA lab or whatever :)
    Last edited by a moderator: May 6, 2017
  6. Jan 16, 2013 #5
    RG-185/U is an interesting cable although it is intended for RF, not audio. It has an impedance of 2000 ohms and the conductor is constructed of magnet over polyethylene and the shield of magnet wire. It's primary use is for a delay line but I was unable to find its velocity factor.

  7. Jan 16, 2013 #6
    Some people hear above 20 kHz.
    I was working on some equipment at 18 kHz and the noise was driving me crazy.
    The owner walked by and he said there was no noise. (10 years older)
    The owner's son walked by and he said there was no noise. (Sportsman, shoots rifles frequently)
    Anyway I upped the frequency to 20 kHz. Everything was great. Could put my ear next to equipment and there was no noise.
    A young woman across the asle said the noise was driving her crazy, so I upped the frequency to 24 kHz and everyone was happy.

    On another subject:
    Was using an amplifier to drive 100 ft. of RG59 coax. RG59 was terminated at both ends with 75 ohm. Could view the input to the coax and the output of the coax and there was a change in output amplitude at 1 MHz. The output amplitude was constant up to 1 MHz and then was constant after 1 MHz.
    Have never heard of this effect.
    Don't believe it was the equipment being used.
  8. Jan 16, 2013 #7

    I know about what you're describing, old CRT TVs used to make a high pitch sound, the flyback transformer I think. I'm not sure it was the fundamental but some intermodulation effects (thus lower freq) I was hearing but certain thing is my grandmother did not hear it. but it was driving me crazy (I was little so presumably good hearing).
    thing is that CDs have a max output of 22.05kHz and many albums are sharply low-pass filtered in studio even at lower frequencies. of course there are the newer hi-resolution formats (SACD and DVD-A) that can reproduce higher frequencies but I'm not sure it's only that, many people reporting the differences use plain old CD players.
    not sure I get your RG59 example. you'll rarely see more than a few meters of cable in home audio equipment and at least with analog signals nothing gets even close to 1MHz. in fact, the higher you go in frequency, the lower the musical content with real music.
    no attenuation in the audio band has ever been reported with audio cables. it's said that normal people can't tell less than 1dB amplitude differences and trained professionals can reach down to 0.2dB.
  9. Jan 16, 2013 #8
    When my daughter was in middle school she told me that kids put a very high pitched ring tone on their phones that only kids could hear. She played it for me. I couldn't hear it no matter how hard I tried yet she could hear it easily. That resulted in a lot of text messages sent during tests.
  10. Jan 16, 2013 #9
    I know that hearing starts to be affected with the mere exposure to city ambient noise and kids are better than us.
    but that would hardly explain it as some of these people are not exactly young females (which are know to have better hearing). and a cable being able to convey say 18kHz is probably able to convey 22kHz with indiscernible attenuation. and, maybe more importantly, output at those frequencies is much lower with music produced by regular instruments, compared to say 4 or even 10kHz. I don't know of anyone who listens to sinusoidal tones when testing for differences, all relevant testing is done with music.
  11. Jan 16, 2013 #10


    User Avatar
    Science Advisor

    Sorry I gave you the wrong link: http://info.publicintelligence.net/NSA-CableInstallation.pdf Simple shielded cabling and grounding requirements for something a little more important than music.

    There is always an interaction between the signal and the cable. The question is if that interaction is at a level above the random noise level of human perception when our most sensitive electronic measurement devices see nothing that can't be accounted for using simple circuit theory at audio frequencies. If someone likes fancy cables and has the money to waste then let them be happy.
  12. Jan 16, 2013 #11
    I'm used to the imposed dichotomy: people are either delusional with regard to this or they know better. it's not black and white as some differences are said to be more than obvious to the least trained ear and this got me really curious. gaining this knowledge will not make me a better man but I still can't help from being curious.
    for instance, the ear is very sensitive to timing information as pitch is perceived by zero crossings within the internal ear. hence, the slightest error that occurs only near a tone's zero crossing (or not necessarily there) may be audible but at the same time appear negligible in a harmonic distortion or inter-modulation distortion plot. it took me a while to integrate all this information. I've seen time-domain measurements where an amplifier failed to perfectly follow the input during fast signal changes but distortion plots looked more than decent. what I mean is that an error may seem negligible if you choose the wrong way to look at it.
  13. Jan 16, 2013 #12


    User Avatar
    Science Advisor

    I thought this was about a passive element called a cable. Complex devices like amplifiers and speakers are a completely different subject. People spend stupid amounts of time fussing over wires and interconnects when the true source of high fidelity is in the speakers. Spend 80% of your budget on speakers and use a quality type generic cable.
  14. Jan 17, 2013 #13
    it is about cables.
    the fact about speakers is even more complex. an audio system has a weak link and it generally is the speaker. but as long as you solve that (and it does cost a bit), the other weak links are revealed. I have heard amplifiers sound the same with lesser speakers but completely different with better ones. at least in theory it can apply to wires. and since a capacitor, a relatively simple passive device is affected by nonlinear behavior, maybe there are similar phenomena in cables.
    I do indeed spend most of the budget on the speakers and I don't spend a dime based on what some audiophile says but still, my initial question can be valid. this is in absolutely no way a purchase decision thread nor do I expect definite, black and white answers from it.
  15. Jan 17, 2013 #14


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The thinnest 'bell wire' that could be used for loudspeaker leads has a resistance in the order of 0.05Ω/m which would introduce, 0.5Ω into a 5m pair. That could be relevant, at a pinch, to the frequency response of an 8Ω system (the overall resistive loss would hardly count). The capacity of the pair would be, perhaps 200pF, which could hardly be significant at audio frequencies in a low impedance system.
    Possibly, the worst effect of using thin wire would be high contact resistance at the terminals.
    I think the whole issue is to do with 'the HI FI THING' which has to involve spending as much money as possible before you can actually enjoy listening the to the actual music. People do need to pay for their 'magic', and gold plated has to sound a lot better than plain old copper.
  16. Jan 17, 2013 #15
    I agree, but you'll also have to agree that a relatively large number of people reporting major differences is at least intriguing. and if it's not exactly people with a tendency to be delusional, some of the actually graduated engineers, it's enough to get me curious. this, of course, doesn't eliminate the possibility that they indeed are deluding themselves (expectation bias and other psychological "noise" factors).
    to put it another way, if I found that there are applications where less than obvious factors that are usually unaccounted for become crucial, it would get me thinking.
  17. Jan 17, 2013 #16


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    We'd really need to see results of double blind tests before coming to any real conclusion. I'm sure that the simple effect of series resistance is finite and that non-linearity in contacts could possibly introduce measurable intermodulation products. But whether or not the effects at audible frequencies are really perceptible seems doubtful.
    The cynic in me has noted that the 'effect' happens to be in the direction involving the highest cost. Also, the comments that one reads from HiFi experts often seem to imply some truly fanciful connections between the processes in the signal path and the subjective experience.
    But I have to admit that my car seems quieter and faster just after I have checked the tyres, washed it or put more fuel in, so I am not immune.
  18. Jan 17, 2013 #17
    well, they have the perfect answer for that :D they say that DBT (ABX or other type) actually adds noise, thus the idea in itself is flawed. in a way they do have a point, a proper (note the emphasis) ABX test is hard to set up. they are generally carried in a noisy environment (room full of participants), subject has no control on listening levels, music choice, song sample duration, the audio system and the room acoustics are unknown (it actually takes a while to adapt to different sound) etc. also, listener experience is very relevant, upon first exposure to a good hifi set, first timers need a while to integrate what they are hearing as it is very different from the muffled mids, boomy lows and sizzling highs of a cheap stereo. unfortunately, many times young people listening to generic, badly recorded pop are used, for instance in trials carried on at by teachers at universities (students are the "cheapest" study material). ABX trials are rarely properly set-up, hence their bad reputation.
    your cost-related remark is very valid, it makes me skeptical too. but I did encounter reports where lower priced equipment was preferred, based on sound alone.
    as for the ubiquitous car analogy, I once suspected that the clutch on my bike started slipping but couldn't be sure. I gave it to a friend for a test ride, without telling him anything. first thing he said was "your clutch is slipping badly".
  19. Jan 17, 2013 #18


    User Avatar
    Science Advisor

    One of my most valuable job experiences was working in audio sales while attending school in the 1970s. I mainly worked as a technical guy who fixed equipment in store but also as a salesman.

    The first thing we learned was to size-up the amount of money in the costumers 'mark' pocket. The second thing we learned was not the laugh at the sucker until the check cleared.
  20. Jan 17, 2013 #19


    User Avatar
    Science Advisor

    True, but compared to the nonlinear device called an 'ear' audio frequency electronic devices are models of perfection.
  21. Jan 17, 2013 #20


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    HiFi electronics is crammed full of capacitors - (many electrolytics, in particular) both in the Power Amplifier and in the Crossover units (they're not all active these days, are they?). How many amplifiers are selected on the linearity of their capacitors, I wonder? I know that 'harmonic distortion' is the watchword of amplifier linearity but Intermodulation Distortion is a far more sensitive test because the products can be made to fall in-band (as they do in your ears).

    Everything is non-linear in the end and the weakest link is surely the one to go for first. The customer can see the connecting cable and both ends of it, at the terminals. I'm sure he (and it is mostly "he"'s involved in this) is very likely to be suckered by the salesman's "of course, Sir, you won't want to spoil it all by buying budget cable." and by the shiny wires and the sexy transparent sleeving. He's less likely to be checking all the internal solder joints with a lens, which is where you can definitely expect to get the occasional spot of non-linearity after a few months of running warm.

    I really am a grumpy old sod when it comes to wasting money - even other peoples'. :wink:
  22. Jan 17, 2013 #21


    User Avatar
    Science Advisor
    Homework Helper

  23. Jan 17, 2013 #22


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    What a great site. I see they actually offer 'Beeswax' capacitors!!! After a few years of 'warm runnings' under a thermionic valve they get an attractive coating of house dust (mostly discarded human skin cells, I understand).

    But I couldn't find any mention of Linearity ?? :wink:
  24. Jan 17, 2013 #23
    ok, since this deviated from the original subject (but please skip to the end, there's something interesting there), why not.

    one nonlinearity does not necessarily mask another. what counts is what the ear/brain system "thinks" about the combined result. a mere figure or even plot is not too telling without a proper understanding of the psychoacoustic phenomena.

    no, most speakers are still passive although there are a few full "high-end" active designs. some caps are better (but not necessarily expensive) than others, good speakers use the good types. my speakers used to cost $7k new (no, only paid a small fraction of that, bought used) and the caps are few dollars a piece but quality ones (Solen brand).

    you would not believe. many amplifiers/preamplifiers/etc use very expensive caps. some go to extreme lengths and eliminate all decoupling caps from the signal path, even giving up DC servos. I know about people who burned very expensive woofers because a defective CD player or whatever sent DC to the amp.

    harmonic/intermodulation distortion measurements are different ways of looking at nonlinearity and yes, IMD tests are more telling because they also tell about the source of nonlinearity (if high frequency tones are used the results are an indirect measure of slew-rate limiting too).

    valid points, but the "high-end" audio equipment is sold through distributors who set-up demo systems and many actually demo cables (including A/B comparisons).
    about damaged joints, you're right. these types would not take the lid off their beloved amp because they think it's sooo super high tech and irreparable damage will be done by non-audiophile air entering it, let alone look at solder joints.

    I think the real problem is that there's a small portion of extremists (euphemism for lunatics) among them, but the "non-enthusiast" views them all the same. I think they're not.

    returning to original question, I ran a generic Google query, excluding everything audio related. found this link: http://www.habia.com/MARKETS/IndustrialCables/VibrationSensors.aspx [Broken]
    looks like a generic industrial cable manufacturer. from website:

    Vibration and Acceleration Sensors
    Cables required are generally referred to as low noise cables. Suitable for use in the measurement of weak signals which may be susceptible to self induced noise within the cable. Mechanical disturbance of a cable (vibration, bending, twisting) can cause voltage spikes with magnitudes of tens of milli-volts. This noise is due to triboelectric charging of the insulator materials, which act as capacitors to store the charge.

    one thing I can tell you is that tens of millivolts is A LOT, even at speaker input.

    and they go on:
    The Habia solution to reduce the effect of the problem is to introduce a carbon layer between the dielectric and screen resulting in any charges formed from movement or pressure to be rapidly returned to and from the screen. This effectively traps any unwanted signals.

    the carbon thing reminded me of an audio cable manufacturer who uses a similar technology, it's called Van den Hul. they're among the more down-to-earth guys. Googled a bit more and found this guy who tested a few guitar cables: http://www.thegearpage.net/board/archive/index.php/t-276746.html

    he says:
    Cables with dullest thump in ranked order from dullest to less dull: (lowest microphonic characteristics)

    a. van den Hul Integration Hybrid (The unique Hulliflex outer jacket helped with this IMO - the inner layer of PVC is also the thickest of any of my cables)

    b. Sommer Stratos (very, very, thick outer jacket)

    c. Vovox Link Protect A

    obviously, not the most scientific procedure, but if looks like the VDH won, maybe not a coincidence?

    and I return to my suspicion... I personally think that at least some of the claims of audible differences are true. but I also think that many of these "high-end" manufacturers just use known, proven technologies from other fields in a fancy suit (with 50x profit margin) or even buy from generic makers like the one above.
    Last edited by a moderator: May 6, 2017
  25. Jan 17, 2013 #24
    have you spotted the 2k+ Audio Note cap? :)
    you will never find any objective data with these guys, that's a given.
  26. Jan 17, 2013 #25


    User Avatar
    Science Advisor

    At least capacitor construction and dielectric types have some correlation to signal transmission.

    These are the kind of 'cable' madness products made from pure 'snake-oil'.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook