anatech said:
Hah! thought of the rebar issue too...but except for the foundation footings, the concrete was poured over compacted sand w/vapour barrier
😀
As for higher current...I may try that...but remember I was concerned about using these cables for test leads... used at these current levels. I'm squeamish about running my amp into .22ohms...call me chicken.
However, as a comparative test with flatline and ropelay, the current increase would affect those cables in a similar way. The main issue being heat.
jneutron said:
No interest in repeating school, thanks very much, especially since it's not going to increase my salary significantly.
Here's some info on the test equipment I used if that helps at all:
http://www.linearx.com/products/analyzers/LMS/LMS_01.htm
I can't say that I necessarily accept your assessment on your say so, however. Perhaps you can elighten?
phn said:
I'm confused, why would the speed of sound through air, and a solid have anything to do with electrical signals propagating through copper wire?
One is essentially mechanical energy, the other is EM...
I've always lived by he axiom or saying that:"judgement lies soley within the realm of fools and/or the ignorant. 'Judgement' only serves to eventually (or immediately) leave one stuck in the self-lies of one's own past."
There's my two cents. oh one more cent.
I believe, with full scientific understading (relative 🙄) of the issues involved, that cables do definitely make a difference. One simple point to remember, is that there is a frequency dependent issue around skin effect, that occurs in the 400 hz range, or begins to be one at that frequency..and is the same for air transfer issues as well..oddly enough. (at 400 hz, in air, it goes through the wave/particle transition). This combined with the understanding of human hearing function, makes cable dielectric, vibration, geometry and overall interactions...quite important when determining how a given cable 'sounds'.
If that does not sit well with you, dear reader, just remember that the specific capacity of your given senses does not equate with that of your neighbour. It's a cross between a learned skill/sensitivity and the basic physical capacity to hear such things, and the variations of these themes throughout the given crowd of users of different cables and the like. What I mean, is your particular ignorance does not equate to your neighbour's understandings and knowledge. Or the converse, your capacities does not equate to your neighbour's ignorance.
Just look at the condition of the voting public in the US right now for conformation of that.
The reality is that both sides of the arguement are correct. If you hear a difference, you do. If you don't..you don't. It's all about personal thresholds of perception and those facts that surround such things.
There's my two cents. oh one more cent.
I believe, with full scientific understading (relative 🙄) of the issues involved, that cables do definitely make a difference. One simple point to remember, is that there is a frequency dependent issue around skin effect, that occurs in the 400 hz range, or begins to be one at that frequency..and is the same for air transfer issues as well..oddly enough. (at 400 hz, in air, it goes through the wave/particle transition). This combined with the understanding of human hearing function, makes cable dielectric, vibration, geometry and overall interactions...quite important when determining how a given cable 'sounds'.
If that does not sit well with you, dear reader, just remember that the specific capacity of your given senses does not equate with that of your neighbour. It's a cross between a learned skill/sensitivity and the basic physical capacity to hear such things, and the variations of these themes throughout the given crowd of users of different cables and the like. What I mean, is your particular ignorance does not equate to your neighbour's understandings and knowledge. Or the converse, your capacities does not equate to your neighbour's ignorance.
Just look at the condition of the voting public in the US right now for conformation of that.
The reality is that both sides of the arguement are correct. If you hear a difference, you do. If you don't..you don't. It's all about personal thresholds of perception and those facts that surround such things.
macgyver10 said:
True. It was more a reflection on the obvious. If the speed of sound is 25 times higher... The reason the preceding sentence in my post is followed by a question mark is that I wasn't sure I was reading it right.
Edit: Perhaps that should be misguided reflection.
macgyver10 said:
Agreed. And in my case, I'm an old dog, so I cannot be expected to learn anything new.
macgyver10 said:
I make the assumption that the equipment you used was adequate for the task. It is not the equipment I am concerned about,. but rather, the load and technique.
macgyver10 said:
It is good that you do not simply accept my assessment. After all, I am just some clown on a forum spoutin some big words.😉
If you connect a scope across a load resistor, you form a loop. That loop will encompass some of the magnetic flux that is created by the load resistor when current is flowing through it. At DC, that is not an issue. But when the current is changing, there will be a voltage induced around that loop, proportional to the rate of change of the trapped flux. (Faraday's law of induction).
This error voltage must be understood and accounted for in any measurement of a rapidly changing current, especially if you are measuring across a low impedance. This error mechanism will not be eliminated via the use of bifilar or thick film resistors (aka caddock). It requires the careful design of a low inductance, low field coupling load and CVR.
About ten years ago, I went through this exercise with a current viewing resistor, a 250 micro-ohm one. At 6000 amps, the Faraday induction error was about 1.2 to 1.5 volts. At 20 Khz and 1 ampere, this error was 1.5 volts. Note that 1 ampere of drive across a 250 micro-ohm resistor should be 250 microvolts. Once this error source was understood, I eliminated it by a change in the measurement pickup geometry..simple wire dressing, that is all.
Why was .22 ohms chosen, anyway? The characteristic impedance of the cable is in the 100 plus ohms region, and typical loads are gonna be in the 4 to 30 ohm range..
macgyver10 said:
If you plow through the "article", you find he calculated the speed of a 50 hz signal as being 2.93 meters per second. While correct for planar wave TEM propagation driven normal to an infinite volume of copper, the "article" tends to lead one to the assumption that this velocity is along the cable, rather than radially. This is typically used to explain why skin effect causes a "smearing"...of course, this explanation is entirely bogus.
The main system reaction to skin effect is the reduction of the internal inductance of the conductors.. For a pair of cylindrical conductors, this is 30 nanoHenries per foot. Discussion of inductive energy shedding via storage combined with slew rate "rate of change" are beyond this discussion, but also cannot be ignored. The loads used to date are incapable of resolving such esoteric phenomena..
If you wish to measure low impedance systems accurately, you have to eliminate the magnetic energy errors, as they really confound the results when the slew rates are high.
Cheers, John
John,
As per our now ancient previous discussion regarding measuring these paramaters (which I still haven't done because I am a slacker), I have to ask you one question that I can't remember from before:
Regardless of all of the math, do you think speaker wire will make more of a difference with regard to sound than, say, a tweeter upgrade?
I'll posit that even if you can get a difference between cables, it will be very small compared to any other paramater of your system (especially when price😛erformance ratios are factored in). I say get a cheap run of thick, decent copper wire and be done with it.
Of course that's just me! 😀
David
As per our now ancient previous discussion regarding measuring these paramaters (which I still haven't done because I am a slacker), I have to ask you one question that I can't remember from before:
Regardless of all of the math, do you think speaker wire will make more of a difference with regard to sound than, say, a tweeter upgrade?
I'll posit that even if you can get a difference between cables, it will be very small compared to any other paramater of your system (especially when price😛erformance ratios are factored in). I say get a cheap run of thick, decent copper wire and be done with it.
Of course that's just me! 😀
David
dfdye said:
My expectation is the tweet will do more to change the sound. Based entirely on nothing but gut feeling. Changing out the tweets has the biggest potential to alter localization cues IMHO.
Followed by asymmetry in speaker toein/out. That wreaks havoc with image focus, as each driver disperses horizontally a bit different...any angular difference has the potential to shift the different frequencies of a single image to one side or another.
Personally....just run #12, sit in the "####" chair (my pound signs, not the software) and enjoy the music. My friend drives me nuts, he keeps listening to this same passage over and over and over, swapping this or that...while I listen for pleasure...sigh..to each his own..
Cheers, John
jneutron said:
I include the link because there is a diagram of the test circuit, which uses a 500ohm resistor as part of a voltage divider to derive the impedance of the DUT. Also, the device offers a "four-wire" method of measurement, which I used, to improve accuracy. Since this LMS device is designed for very low impedance measurement, I was surprised that something like what you mention would be missed. Especially since I have corresponded with the engineer of LEAP and LMS in the past...and he's no dummy. His work on mathematically modeling loudspeakers alone pretty much redefined loudspeaker CAD.
jneutron said:
It's just not in my nature. BTW, I'm a double-D, eighteen year old blonde who loves old guys, and trolls DIY audio forums...
jneutron said:
characteristic impedance at 100Mhz will be 100ohms, where reactive elements are more significant than the resistive, however just as Stereophile is applying "skin effect" where it doesn't belong, I assumed that these wires would be much closer in impedance to their DCR at audio frequencies. I expected to see them rise to their characteristic impedance at the high end (but it happened sooner than I expected). The choice of .22ohms was to match this DCR, and also just happened to be what I had on hand...I admit my reasoning here may not have been sound...but I'm sure I had a good reason at the time 🙂
However, I appreciate your input, and I may actually write to the LMS engineer and ask him specifically what is the lower limit impedance that his system can measure with reasonable accuracy (say 5%?)
macgyver10 said:
Interesting..I'll peruse it when I can. But I find the bulk of the errors is not in the equipment, but the user. (Well, at least when I am the user..)
Perhaps I should discuss test technique with that engineer. Who knows, something good may come of it..
macgyver10 said:
Where were you when I was in the market
???? (oh yah, not born..😀 ) Course, my son likes younger women..😉..but certainly not a cable guy with double-D's..ewwww Seriously, statements like that may cause some of us oldie geeks to have a heart attack..you wouldn't wan't that, now would you??🙁
macgyver10 said:
The cable Z is simply sqr(L/C). Zip Z is too high to worry bout, rather you should focus on what the cable does into a load more representative of the intended application.
macgyver10 said:
5% is ok for FR and phase, but may not be sufficient for localization parameters..for that, one needs less than .1db and better than 2 uSec, remaining accurate at those speeds is a requirement for localization..(actually, I'd go to .05dB and 100 nSec, but that's just me..
You need a load resistor that has less than a nanohenry inductance, with no loop pickup..A trivial task once the theory is understood.😉
Cheers, John
jneutron said:
Even when the actual results of the resistor measurement are going to be used to remove it from the DUT+resistor data?
This is additional functionality of the LMS software suite, which allows mathematical post processing of the data.
I was thinking of "common mode rejection" when I considered that the resistor measurement data didn't really matter (as long as it was consistent each time I swept) since I was going to invert it and remove it later anyway.
In theory I thought it would result in all data points not attributable to the DUT being removed from the data.
macgyver10 said:
When I work at this impedance and slew rates, one of the most difficult things to do is get repeatable results that are independent of the wiring to the test instrument.
When I perform the autozero cal with the HP 4263B, I have to go to great length to make sure the geometry is not altered after the cal cycle. Otherwise, the readings are just garbage.. With proper technique, I've been able to get repeatability at the 250 picohenry level, but no better.
While measure and subtract sounds good in theory, low Z measurements with high current slew rates does require a review of all the things that can bite ya. Without seeing exactly what you had setup, I can't recommend changes to technique.
Your technique however, does not have the ability to correct time based shifts. Localization is both amplitude and time based.
Cheers, John
A few years ago a speaker manufacturer suggested the following way to compare sp.cables.In a given system,for test purposes we consider the existing speaker cable as the ''reference''.On the cables under test,put female connectors on one end and male on the other.Take a good listen to the system and when you are ready start listening to the other cables by puting them between the ''reference''cable and the speaker.According to the manufacturer the difference to be heard is actually the damage that the cable under test causes to the audio signal.To see the damage of the''reference'' cable repeat same test with a second pair of this ''reference''cable.Sounds interesting?
phn said:
I'm not sure who you are accusing of lies here. Maybe all of the above.
The Pope I don't know much about (I left the church after Vatican-II), Stereophile - I take with a grain of salt (magic audio stones and such). But Hiraga I know, personally.
Jean Hiraga knows his stuff, and he doesn't care a bit about how much or how little something costs, or its brand name. What he does care about is how things sound, and he gets good results. As I stated earlier - better results than anything I've ever heard (and that's a lot). Closer to live music than anything else. And he does it over and over again. So I trust his judgment, knowledge and experience.
I suppose a lot of people cringe at the term "Audiophile" because they've been hit over the head with it too many times, and seen it so misused - mostly in advertising. That's a pity, as it has perhaps lost its original meaning.
OK, back to important stuff, like speaker cables.
panomaniac,you dont really need stereophile or the pope to tell any diy'er who's the real audiophile.Right?
Thanks guys for poking holes in the Stereophile article. That super slow speed in copper was bothering me all night! How could that be?? 
But here are a few questions;
We have gone on for post after post about the cable termination in Macgyver10's test. And the Stereophile article, with all its faults, does at least talk about impedance mismatch and reflections. So:
Why would impedance mismatch and ringing not exist in speaker cables? Or if it does exist, why would it not be audible?
Is the load impedance of the speaker too high - too low - too complex? Is it something to do with the amp?
I can tell you that bad termination and impedance mismatches are clearly visible on video signals at any scan rate - 15Khz and higher. Most people never notice it - unless it's pointed out to them. I see it all over the place because I know what it looks like and it bugs me.
Could this sort of ringing and echo exist on speaker cables? If not - why not?
But here are a few questions;
We have gone on for post after post about the cable termination in Macgyver10's test. And the Stereophile article, with all its faults, does at least talk about impedance mismatch and reflections. So:
Why would impedance mismatch and ringing not exist in speaker cables? Or if it does exist, why would it not be audible?
Is the load impedance of the speaker too high - too low - too complex? Is it something to do with the amp?
I can tell you that bad termination and impedance mismatches are clearly visible on video signals at any scan rate - 15Khz and higher. Most people never notice it - unless it's pointed out to them. I see it all over the place because I know what it looks like and it bugs me.
Could this sort of ringing and echo exist on speaker cables? If not - why not?
panomaniac,isn't it possible that the same could be true with audio?Perhaps people do not notice things unless it's pointed out to them?Perhaps the eye is more sensitive but in the end guidance could train the ear too.
Panicos K said:
Absolutely! And one would hope that the majority of people on this forum have "well trained", or at least attentive, ears. The ears are not any different, just the knowledge of how to use them.
The eye does seem to be harder to fool than the ear, but that's good for us music lovers, it makes recorded music enjoyable and often very realistic.
Wandering a bit OT - In color reproduction, were I work, we have "Metamers". Colors that appear the same to the eye but have different spectral content. That is good, but leads to problems. For Example - 2 images may look identical under light source A. But when seen under light source B, they are no longer identical. It's a huge problem. But without Metamerism, we would have a very hard time producing realistic images. Mixing 2 colors to achieve another wouldn't work.
I don't know if this applies to audio and audio cables, but perhaps it bears thinking about.
panomaniac said:
I'll address this using a high current solid state direct coupled amplifier and a "typical" dynamic loudspeaker with a large low Q impedance hump at the bass driver resonance and typical inductive rise to upper freq cutoff:
The amp is designed to have as close to zero output impedance as possible, largely the result of negative feedback. This results in a high Damping Factor. The higher the damping factor the less effect the impedance of the load (speaker) will have.
When you consider the impedance curve of the speaker, and add to that the nearly negligible influence of the cable (consider it part of the x-over) one would reasonably expect that there is far more in the elctro-mechanical loudspeaker system to affect the sound that could be detected in what kind of wire you're using.
Using the criticism of my test methods as an example: it's argued that my test procedure could result in wild inaccuracies in my data. This is largely due to my attempt to measure impedances that are this small. I see that as suggesting that the speaker system/amplifier system matching would overwhelm (swamp) any perceptable effects of the speaker cable.
Again I keep looking at this as an issue of micro vs. macro effects. Many audiophiles are consumed with eliminating "micro" deficiencies because of their supposed "cumulative effects", however they can be quite content to live with gross "macro" deficiencies in other parts of the system...including the limits of engineering and our own hearing apparatus. I've heard some really horrible sounding stereos in my time, and many of them had been "tweaked and tuned" to death. In some cases, the speakers looked and sounded much like what is put into acoustical tile ceilings and fed with 70V transformers...but the speaker cable was pure silver foil...single ended zero feedback triode..blah blah blah...
There are probably those of us who can afford such obsessive compulsion, but personally I'm still (after 25 years) knocking down the bigger obvious problems with my sound reproduction equipment. Not the least of which is an aging ear....
That kind of turned into a rant....sorry about that, but I feel better now!
Hi jneutron,
I find that cleaning the clips and a few short and open zeros help the readings converge, as long as the leads don't move too much. At the end of the day, the 100 KHz readings with 1 m leads are suspect.
-Chris
So true. I use the same meter (but the "A" version). The 1 meter length with the kelvin clips are at the ragged edge at 100 kHz (top of my range). The test fixture is much more useful here but useless for measuring cables.
I find that cleaning the clips and a few short and open zeros help the readings converge, as long as the leads don't move too much. At the end of the day, the 100 KHz readings with 1 m leads are suspect.
You are more patient than I sir. But then again, I haven't needed readings up there badly enough to work at it.
-Chris
- Status
- Not open for further replies.