I won't make any claims as to the audibility of cables, but one thing I find interesting is the bantering back and forth about the R,L,C components of the cable, and some asking if the R,L & C are known for a cable. If I remember my high frequency class correctly, L & C will vary depending upon the frequency, and from DC circuits, R will vary depending upon the room temp, current through the wire, etc. I'm not sure of the magnitude of the variation of each of these at different values of the parameters mentioned, but it would seem to me that if R,L, and C are to be important to the audibility of the cables, that the variation at different frequencies, temperatures and current load through the cable would be come important.
Peace,
Dave
Peace,
Dave
Re: The Carver Challenge
I welcome this, good on you. I hope you can do it even though I might have to eat my hat. 🙂
thetubeguy1954 said:
I welcome this, good on you. I hope you can do it even though I might have to eat my hat. 🙂
L and C are actually pretty much frequency independent, when expressed as L/meter and C/meter, the distributed parameters that they actually are. But if you just stick them on a C meter or an L meter what you read as lumped values will be frequency dependent as signals reflect around in the cable and come back and interfere. Though I'm pretty sure that frequency dependence would be unmeasurable at audio frequencies, since we're talking transmission line behavior then and a cable used by an audiophile is a really small fraction of a wavelength. 20kHz having about 10km wavelength in teflon dielectric.
The change with temperature, at least at temperatures humans are going to comfortably coexist within, will be less than the likely achievable matching accuracy. Cooper resistance increases about 1.004x for every degree celcius change (according to values found through googling). For a decent speaker cable that is maybe around a few tenths of an ohm total resistance, a 10C change in room temp would make about a milliohm change -- less change than you'd likely get by disconnecting and reconnecting a cable.
So, if these are somehow enough to matter, a matching test isn't practical. You can't match to that kind of precision unless you can measure that fine.
Hi Bill,
Not sure if you have seen these, but there's a bunch of speaker cable R, L and C measurements here and here. It's been a while since I read the article, but I think they measured L by shorting the opposite end of the cable, and C with an open at the opposite end. I don't know how much of the variation in these parameters is measurement error though.
Not sure if you have seen these, but there's a bunch of speaker cable R, L and C measurements here and here. It's been a while since I read the article, but I think they measured L by shorting the opposite end of the cable, and C with an open at the opposite end. I don't know how much of the variation in these parameters is measurement error though.
Re: Re: The Carver Challenge
Hello bwaslo!
My reference IC is made by Mike Rispoli, who also made my speaker wires and treated my Fostex FE206ES-R driver. I cannot send it for two reasons:
1) I don't know that Mike would want anyone examining how he made them ---{I don't know any other IC constructed like these are}--- or with what materials he used.
2) It's also the only one I have and quite simply don't want to live without it. Believe it or not it replaced a $1.8K IC and his speaker wires replaced Nordost Blue Heavens.
I'm not DIY oriented so please explain how I'd measure the R, L & C, ok? I'll make sure to include the length which is either 1 or 1.5m.
Hello Fredex!
I've always been amazed at those people who believe subjectivists have some vested interest in cables sounding different. As I've stated more than once here, I originally believed wires couldn't possible sound different. It was through attempting to prove what I believed was correct that I learned I was mistaken. So if I listen to two different ICs with different wires (copper, silver, silver-plated copper, gold, silver plated gold etc) different geometeries, different dielectrics and so on that measure the same in R, L & C I'll have no problem admitting I cannot hear a difference with these two ICs.
Truth be told many of the inexpensive ICs are so bland sounding that I struggle to hear a difference between them and have to really try to do so. Maybe it's because they're so uninvolving I don't care to try? It's almost like trying to taste the difference between two cheap stale peanut butters if you catch my drift.
Thetubeguy1954
bwaslo said:
Hello bwaslo!
My reference IC is made by Mike Rispoli, who also made my speaker wires and treated my Fostex FE206ES-R driver. I cannot send it for two reasons:
1) I don't know that Mike would want anyone examining how he made them ---{I don't know any other IC constructed like these are}--- or with what materials he used.
2) It's also the only one I have and quite simply don't want to live without it. Believe it or not it replaced a $1.8K IC and his speaker wires replaced Nordost Blue Heavens.
I'm not DIY oriented so please explain how I'd measure the R, L & C, ok? I'll make sure to include the length which is either 1 or 1.5m.
fredex said:
Hello Fredex!
I've always been amazed at those people who believe subjectivists have some vested interest in cables sounding different. As I've stated more than once here, I originally believed wires couldn't possible sound different. It was through attempting to prove what I believed was correct that I learned I was mistaken. So if I listen to two different ICs with different wires (copper, silver, silver-plated copper, gold, silver plated gold etc) different geometeries, different dielectrics and so on that measure the same in R, L & C I'll have no problem admitting I cannot hear a difference with these two ICs.
Truth be told many of the inexpensive ICs are so bland sounding that I struggle to hear a difference between them and have to really try to do so. Maybe it's because they're so uninvolving I don't care to try? It's almost like trying to taste the difference between two cheap stale peanut butters if you catch my drift.
Thetubeguy1954
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/part6/page1.html
This one is very nice (Java applet), accounting for skin effect also:
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/Java/twin.html
- Klaus
This one is very nice (Java applet), accounting for skin effect also:
http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/audio/Java/twin.html
- Klaus
Re: Re: Re: The Carver Challenge
There was guy here a while back in the solid-state forum who wrote he could measure different distortion of ICs, but those were not matched in their parameters and that guy didn't post any detailed data (only a screenshot thumbnail) and has not posted since (in fact that was his only post so far).
http://www.diyaudio.com/forums/showthread.php?postid=1875534#post1875534
- Klaus
In fact I was not able to measure any difference down to -140dB between ICs once these parameters are exactly matched, and any tiny residue at higher freq ranges (buried in the noise, even after massive averaging) that was there all seemed to be linear stuff. These tests were very sensitive, you can see the change in capacitance that comes when bending a cable slightly (microphonics), or smallest changes in contact resistance at the connenctors, or changing capacitance of the connectors (plugs and receptacles) themselves with shear forces etc. Because of the averaging (to get the noise down) of course loosely correlated (to the signal) spuriae would be damped (say "time-smear" from DA-effects), but without averaging te difference was plain noise, so I doubt there is something to it.thetubeguy1954 said:
There was guy here a while back in the solid-state forum who wrote he could measure different distortion of ICs, but those were not matched in their parameters and that guy didn't post any detailed data (only a screenshot thumbnail) and has not posted since (in fact that was his only post so far).
http://www.diyaudio.com/forums/showthread.php?postid=1875534#post1875534
- Klaus
I'm going to ask you to question my own philosophy.
Take two different cables, adjust them until the R & L & C match cable to cable.
Q1. Can we assume that both cables will exactly load the transmit end and the receive end identically?
Now assuming the affirmative to Q1:
Q2. Can these different cables, that load the send and receive ends identically, be different when we listen to the sound coming out of the final transducer?
stinius said:
Sure fredex
And it would be even more interesting if they switched the direction of the cable and could tell the difference. For what ever I know, they listened to the cable in the wrong direction.
😱
Cheers
That could well be the ultimate test.A diy interconnect with just a twisted pair of conductors and no shield,since many accept that the shield "determines"direction.
AndrewT said:
Of course. Sound perception involves a lot more than just the information picked up by the ears:
http://www.media.uio.no/personer/arntm/McGurk_english.html
Andrew,AndrewT said:
For Q1: The matching IMHO should only be done to equalize cables that are in the same ballpark right to begin with, especially when it comes to C. Which means that tests of ICs of the same length are futile, rather select them to have equal C (to about +-5% or so). An important factor (which I didn't test) migh be shield effectiveness, which depends on the shield resistance as it determines the lower cut-off frequency of the common-mode choke that coaxial cables represents. This influences the amount of rejection of ground-current effect between gear, the lower the shield resistance (and the higher the shield inductance) the lower is the cutoff where the common-mode action starts to fall apart. Mostly its in the kHz ranges
For Q2: When we have microphony and/or nonlinear effects (the latter I couldn't find, though) then two otherwise electrically identical ICs might sound different. Different HF behaviour for any number of reasens could also play a role.
Personally, for ICs... my quest is settled. My choice is the Sommer Polaris, which is a video triax (both together==low shield resistance and good separation of signal return and shield currents at lower freqs that a standard coax would have, from the skin effect. Or just any other good video/sat cable with a strong shield and low capacitance and low loss.
- Klaus
Hi TTG1954,
Don't you have any cables you can tell from line cord and that aren't in your system or of some proprietary construction? This need not be with your current favorites, right?
The only reason I mentioned length is so that whatever cable that was supplied would be long enough to reach between whatever it connects. It would be made long enough to approximate L and R, then tweaked with C. But it would need to reach to make a test with.
I wasn't planning to dig them apart, only to measure them. Why not ask Mike, I'd expect he'd welcome such a test, no? Or put them into an opaque cable jacket -- you should be able to trust that I won't look inside the jacket as everyone else is trusting to have you monitor and report on your own test proceedings.
I can't help with that one. The test would be of no use to me unless I could test with both cables afterward at least to find out what physical characteristics might be there to BE heard (assuming that were the result). The idea being to get some research in -- as has been said before, the first guy to unequivocally demonstrate (and ideally, find the basis of) an unanticipated form of audio distortion stands to make quite a lot of waves in the audio engineering world.
Well, some test gear will be needed, and depending on the tolerance people think would be valid, it won't likely be cheap. Most (including me) would want to see better than would be measured using basic service-grade or Radio Shack gear. I have quite a bit available to me on this end. Probably the best way to measure resistance would be to set up a Kelvin-bridge measurement (drive a high well-defined DC current of maybe 10A through the cable, and measure the voltage drop across the wires with a separate pair of leads into a millivoltmeter). If 200Hz and 20kHz sinewaves of high known current levels can be generated and driven through the cable, and a good accuracy AC voltmeter were used to measure the drop like for the resistance, series impedance could be measured to find the inductance. Capacitance is probably the easiest to measure using a garden grade DVM with a C scale that reads small values, though measuring capacitance within a pF will depend on meters used tracking each other and probe cables matching and maybe on how the cable lies relative to nearby conducting surfaces during measurement. Probably the same C meter should be used to measure both.
Don't you have any cables you can tell from line cord and that aren't in your system or of some proprietary construction? This need not be with your current favorites, right?
The only reason I mentioned length is so that whatever cable that was supplied would be long enough to reach between whatever it connects. It would be made long enough to approximate L and R, then tweaked with C. But it would need to reach to make a test with.
I stated why (how sound perception works) and gave you a link to a demonstration of why (two different perceptions of sound with the same sound impinging on the ears). You did follow the link and try the demonstration?AndrewT said:
Yes, I listened to the demo.
I hear DA with eyes open and I hear BA with eyes closed.
But, that has nothing to do with my questions.
I hear DA with eyes open and I hear BA with eyes closed.
But, that has nothing to do with my questions.
doug20 said:
With a ratio like that I can understand why differences can't be heard, not even talking about cables.
fredex said:
Jenga blocks, and when you finish with your Mancala game the stones make a good substitute for "brilliant pebbles".
They do twitch BTW especially when you accidentally short 8 220AH lead acid batteries across them. 🙁
- Status
- Not open for further replies.