The cable conundrum
- Thread starter DaveyF
- Start date
Not everyone, just all those who disagree... Life is like that.
Welcome aboard!
thanks Don, looks like an interesting site
disagree reminds me of this " Agree to Disagree"
The term "agree to disagree" or "agreeing to disagree" is a phrase in English referring to the resolution of a conflict (usually a debate or quarrel) whereby all parties tolerate but do not accept the opposing position(s). It generally occurs when all sides recognize that further conflict would be unnecessary, ineffective or otherwise undesirable. They may also remain on amicable terms while continuing to disagree about the unresolved issues.
Yes, it is more apt than "compromise", which implies everybody gave in a little. Instead, it just means we agree to quit talking about it and move on to the weather or other more benign topics of discussion, like tubes vs. SS or digital vs. analog sources.
Welcome to the WBF, cpp. While your post may have some relevance on other forum's, the intent of my OP was not what you surmised. Please re-read my OP. I happen to think the members here are a lot more introspective; they can and do understand the nuance of the issue in my OP. Which is one of the reasons that I hang out here
Not so certain, a more revealing system may require a certain type cable, perhaps a "better" cable ... but even within what I consider mid-fi systems, I've heard certain cables make a distinct difference.
The trick with cables, is how they sonically integrate within a particular system. Again, like most dependencies, it's often a synergistic matter. Certain cables sound better within certain systems for reasons that are beyond explanation. One needs to find that proper mix. My system specifically prefers "lively" cables, and I always find it funny when I claim that I therefore prefer Cardas. Audiophiles will always challenge that notion, claiming that Cardas are generally not known to be the "lively" type. Well, they're indeed correct, however, I use Cardas Golden Hexlink 5C, a long discontinued cable that stands sonically apart from the general Cardas "house" sound.
One could in fact refine or maximize a system towards a specific cable by maximizing connection quality. In my system, I've exchanged the majority of female plugs to Cardas to exactly match the cable. This, in my experience, will maximize a cables influence on any particular system, but at the same time, it may make a system "cable" dependent.
Choice of Cables is one thing, but connection quality is way too often over-looked. I've attended far too many system demo's in which the components were judged, unknowingly to many listeners, solely based on questionable connections within the entire system. One time, I attended a demo of a very expensive system in which the banana plugs on some "famous" hi-end cable came apart from the speaker during the audition. When I investigated, the tolerances between cable and speaker connection were not ideal, to the point that time alone would allow them to vibrate loose. When I looked at the cable structure of the entire system, it became obvious why this system could never justify it's cost. Funny, that night I hinted to all that the relatively underachieving sound quality was partially a result of this systems cable choice and poor connection quality, but instead of dealing with that reality, they continued to exchange components based evaluations.
I immediately left with a sigh!
tb1
I use cables to lash myself for my audiophile sins. I find that rough shielding and sharp terminations are superior for audiophile penance.
Wow. Where to begin?
What you hear on your headphones is very unlike what people hear when listening to speakers loading into a room vice headphones loading into your head. Everything is different. And how would you possibly know what the ambient noise floors of people’s listening room are?
Unbelievably bold statement from someone who currently does not own an LP playback system and never owned an LP playback system that had any claims to greatness. But never mind the slam at LPs, the detail you hear listening to digital through headphones is not the same as the detail resolution of listening to the same digital through a really good system loading into a room. Headphones are an alternate universe that some choose not to visit. It’s like if you are used to drinking regular coke and someone slips you a diet coke. Or if you are used to drinking whole milk and someone slips you powered milk. It leaves you with a sensation that you don’t find natural and absolutely don’t like.
What you hear on your headphones is very unlike what people hear when listening to speakers loading into a room vice headphones loading into your head. Everything is different. And how would you possibly know what the ambient noise floors of people’s listening room are?
Unbelievably bold statement from someone who currently does not own an LP playback system and never owned an LP playback system that had any claims to greatness. But never mind the slam at LPs, the detail you hear listening to digital through headphones is not the same as the detail resolution of listening to the same digital through a really good system loading into a room. Headphones are an alternate universe that some choose not to visit. It’s like if you are used to drinking regular coke and someone slips you a diet coke. Or if you are used to drinking whole milk and someone slips you powered milk. It leaves you with a sensation that you don’t find natural and absolutely don’t like.
Sorry, but just pointing to whole papers or basic electricity pages does not prove your point. The phases of the moon are scientific facts and, IMHO, do not add to the sound of cables. The only quantified effect in real operation conditions is that the difference in energy storage at two different frequencies was around a few nJ (10 e-9J ) when the signal was around 300 J. Most people will tell us that such difference is not audible.
Bruce Brisson cables seem to be optimized using the power phase criteria. Many people report they sound great. Excellent. But in no way this makes his cables more "scientific" than all others. And can not explain why his expensive cables seem to sound much better than the cheaper ones, according to those who have experience with them.
BTW, my experience with MIT cables was with a full set of the MH750 reference terminator with CVT and MH350 reference terminator with CVT. I really enjoyed them during some years.
Last edited:
You should avoid Valhalla's - their thin sharp cutting edges can be dangerous. And Transparent Opus can break a few bones easily.
Sure it does, you are apparently just not reading them. The latest PDF describes the whole phase issue very nicely and is based on science (electrical engineering) - and of course the scientific claims also match that university link I sent. It's fine to ignore them if you want, but I would rather have you give me concrete, scientific counterarguments than random ramblings.
It's this argument that makes me believe you are just being argumentative for the sake of being argumentative, and quite frankly, tiring. Because if you carefully review what has been said, no one is talking about the phases of the moon, I think. Again, ignore the arguments I am making, but let's not pollute this thread.
From my reading of the patents this is what they are all about: correcting impedance phase errors at various frequencies, given the scientifically and experimentally proven existence of phase errors at various frequencies due to the L/C characteristics of a cable. You claim said phase errors are not audible, hence I read into it that you think their attempt at "fixing" it insignificant (my words), but as you also say, people do claim their cables sound better than others. I find all my answers in the patents and white papers as posted, you apparently don't.
I'll just close with the following additional quote from the latest white paper PDF: An ideal capacitor has an impedance phase angle of -90 degrees that remains constant with frequency and applied voltage [...]-89.5° is a typical non-ideal capacitance phase-angle value for an audio cable when measured at audio frequencies. Contrast this theoretically optimal -90 degree figure with what he actually measured in the cable under test at various frequencies (as per previous posts), and then ponder whether a ~3-5 degree difference/offset from said optimal is significant or not, and what would that mean if similar variations exist all over the spectrum. I would like to see scientific counter arguments, formulas et al to prove your point.
Last edited:
Look up dispersion in transmission lines. I have never known it to apply to audio cables. Note treating a cable alone is insufficient; one must consider the source (amplifier) and load (speaker) when trying to correct phase errors (or anything else). It's a system...
Cables can be considered distributed RLC systems. I am not sure why having a purely capacitive reactance would be any sort of goal, but I have not really been following the arguments.
For the record, I find patents interesting but hardly definitive. Seen too many that were obviously hogwash that some poor overworked patent examiner passed anyway.
Cables can be considered distributed RLC systems. I am not sure why having a purely capacitive reactance would be any sort of goal, but I have not really been following the arguments.
For the record, I find patents interesting but hardly definitive. Seen too many that were obviously hogwash that some poor overworked patent examiner passed anyway.
Yes, it's more complicated than that, I agree.
I don't believe the goal is so much pure capacitive reactance as it is _uniform impedance_ across the spectrum, for uniform power transfer. So capacitance is only part of the picture; the white paper PDF spends a lot more time on inductance than capacitance. If you read it carefully, it also discusses specific wire winding techniques; e.g.:
By coiling the wire of a cable in a tight loop, MIT helps to overcome the problems of self-inductance by creating proper amounts of inductance in the cable through proprietary patents and patent-pending winding techniques. Additionally, MIT achieves better inductive properties through mutual inductance. Mutual inductance occurs when the flux of two independent, non connected coils interact [...] all audio cables inherently have shunt capacitance, and an inductive phase angle is therefore necessary to balance the shunt capacitor’s phase angle.
There are some important points here... a) He clearly does not want to minimize inductance; b) wrt mutual inductance, this would imply he's running independent same-winding-direction inductors close to each other within those boxes to increase inductance; c) the overall goal - taken in the larger context of the paper, the patents and the networks in the boxes - appears to be optimizing the relationship between capacitance and inductance across the spectrum at various chosen frequencies where he presumably measures significant phase shift deviation from optimal (i.e. what he calls the "articulation points"), in order to keep phase relationships between voltage and current as uniform as possible. Clever, if you ask me, assuming his solution works. The science behind the problem description is solid, but I cannot prove the solution in the patents, other than verify by ear. Perhaps this is where microstrip is caught up as well, although I didn't get that impression; and I don't think I ever offered the MIT solution as a scientific fact, unlike the problem description. Having said that, I've been wrong before, so solid counterarguments are welcome.
BTW, this is not the first time we hear about proprietary winding techniques to achieve a certain effect... Wilson has been running ads touting specific number of turns per inch (the angle of lay) in his speakers' wires, apparently in an attempt to optimize electrical characteristics in the entire wire-driver-crossover chain.
So you may ask why have we been discussing capacitance only thusfar... only because the referenced patent discusses it, as if it's easier to adjust that (with their proposed solutions) given specific inductance characteristics.
I shall have to read the paper at some point, I suppose... Patents are public, do not know what a "proprietary patent" might be. I do know a lot of companies prefer to not patent as it gives away their proprietary techniques.
Treated as a transmission line, the impedance of a cable is to first order:
Zo = sqrt[ (R + jwL) / (G + jwC) ], where
R = resistance
L = inductance
C = capacitance
G = conductance
j = sqrt(-1)
w = frequency in radians/s, note w = 2*pI*f
Assuming R is very small and G very large ( a lossless line), then Zo ~ sqrt(L/C). If you want a low-impedance cable to better match the low impedance of a speaker, you want to reduce L and increase C. Of course, too high capacitance can make an amplifier unstable, and there are always trades.
At high frequencies (well above audio) current tends to travel on the surface, using only the "skin" of the conductor and thus the term skin effect. That reduces the effective cross-sectional area, increasing resistance. I think I wrote a thread on skin effect in the technical area. Dispersion happens because the cable does not pass all frequencies equally, delaying (or phase shifting) some more than others. These and other effects can cause distortion, again usually at frequencies well beyond audio.
If the phase changes linearly with frequency, the derivative (rate of change wth frequency) of phase is constant, and that derivative is called the group delay. Constant group delay means all frequencies are delayed equally, and time information is preserved; that is, a pulse coming out looks like the same pulse going in, just delayed.
Enough for starters...
Treated as a transmission line, the impedance of a cable is to first order:
Zo = sqrt[ (R + jwL) / (G + jwC) ], where
R = resistance
L = inductance
C = capacitance
G = conductance
j = sqrt(-1)
w = frequency in radians/s, note w = 2*pI*f
Assuming R is very small and G very large ( a lossless line), then Zo ~ sqrt(L/C). If you want a low-impedance cable to better match the low impedance of a speaker, you want to reduce L and increase C. Of course, too high capacitance can make an amplifier unstable, and there are always trades.
At high frequencies (well above audio) current tends to travel on the surface, using only the "skin" of the conductor and thus the term skin effect. That reduces the effective cross-sectional area, increasing resistance. I think I wrote a thread on skin effect in the technical area. Dispersion happens because the cable does not pass all frequencies equally, delaying (or phase shifting) some more than others. These and other effects can cause distortion, again usually at frequencies well beyond audio.
If the phase changes linearly with frequency, the derivative (rate of change wth frequency) of phase is constant, and that derivative is called the group delay. Constant group delay means all frequencies are delayed equally, and time information is preserved; that is, a pulse coming out looks like the same pulse going in, just delayed.
Enough for starters...
Last edited:
Wow, I can remember enough of my math and calculus to make sense of this. One of the best combos of technical and non-technical descriptions I have read, thanks.
I might be missing something very fundamental here, but if we're talking about applying transmission line theory to speaker cables then I can't see how. Transmission lines are normally driven with impedances which match the TL so as to produce no reflections. The inevitable consequence of this is a relatively huge (in audio terms) power loss. Now perhaps the required source impedance could be electronically synthesized to eliminate this but power amps are in general voltage sources, meaning very low output impedance. So we'd need a customized design. Given that a speaker cannot be a conjugate match to the line (its impedance varies widely with frequency) what would be the point? Perhaps an impedance matching network could be designed so the speaker looks more uniformly resistive across the audio bandwidth though. The vast majority of speakers aren't designed to be driven by their nominal impedance, rather from voltage sources (with low output impedance).
As far as I can recall from my EE class, TL theory only needs to be invoked when the rise time of the signal is within an order of magnitude of the round trip propagation time. For typical 3m speaker cables the round-trip time might be 40nS worst case - very hard to imagine a power amp putting out a sub-microsecond pulse. So we're good just considering the cables' lumped parameters ISTM. Otherwise its an application of 'waveguide below cut-off' which theory I never quite did grasp. But as I said, I might be missing something significant.
I share Don's skepticism of this term 'proprietary patents' - its marketing weasel-speak to me. As is the word 'proper' in a supposedly scientific context. So if there are patents applied for, where are their numbers? Then we can look them up.
Yeeeeees, but all audio cables inherently have series inductance too (they are wires, after all) and that tends to counteract the shunt capacitance as a matter of course. But the inductance only becomes significant beyond the audio band.
Well, that's exactly the issue, he claims... he claims he measures different phase angles at different frequencies, thus different losses. If true, then his solution probably works; if not, then MIT cables are tone controls as I have claimed in the past (but I am willing to explore and change my mind). Start with the white paper from 1991 (it's easy layman's language with formulas), before getting to the patents.
BTW, this appears to be the basis of the Kubala Sosna patent; and the MIT literature, on the other hand, seems to indicate than this would create voltage/current phase relationships away from optimal.
Last edited:
Or he's missing something, so see if you can poke holes into his claims - I would love to see that. I have posted numerous patent numbers on this stuff in the various threads (see the new MIT Spectral cables thread), see what you think, some are seemingly silly (like a capacitor in parallel with the entire length of the positive length of the wire - what we would otherwise call a shorted capacitor - with as high a value as the amp could handle).
Of course it is, don't get hung up on that, that's not the point of his work. Focus on why he's intent on increasing inductance and what he does with capacitance...
This assumption is only true at radio frequencies!
At audio frequencies, you have to use the long form formula, which is very messy.
Jim Brown covers this in depth.
Transmission Lines at Audio Frequencies, and a Bit of History
by Jim Brown
Audio Systems Group, Inc.
http://audiosystemsgroup.com
http://www.audiosystemsgroup.com/TransLines-LowFreq.pdf
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |