|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
96.25.74.167
In Reply to: RE: Cable article link posted by jrlaudio on October 16, 2013 at 05:43:07
"Contrary to what the audiophile world thinks, skin effect doesn't come into play till you reach about 50kHz or so. Regardless of the metal copper, silver whatever. At 200Hz there is no "skin depth". "
Hmmmm. Bullshit!
Take a gander at the field of electromagnetic non-destructive testing, you will likely be shocked to discover that it's typically done at audio frequencies. The very reason is that most of the interesting electrical things that happen are in that region. Essentially the illuminated object can be modeled as an LR circuit producing a return phase shift that's a function of frequency. If you want a reference try to find a copy of Libby: Introduction to electromagnetic nondestructive test methods.
The rub is it that the signal couples back into the inducing circuit and you end up with varying phase shifts and losses as a function of frequency through the audible spectrum. Put another way, it smears transients.
It's simply a factor like many others and I'm not arguing that it's always a dominant one but on the other hand it might be. The audio bandwidth (>3 decades) is huge and we have very temporally sensitive receivers in our heads...
Rick
Follow Ups:
While it doesn’t make an interesting magazine read, in the reactive and resistive domain, actually every aspect of the cable as a system can be measured by a network analyzer like this one.
This one, an HP-3562a is one I have used to have and will give a precise measurement from 100KHz down to tiny fractions of a Hz.
http://www.et06.dk/HP3562A/images/HP3562A.jpg
Once you have the true R/L/C components and the network they are arranged in, you have an electrical model of what you’re dealing with. That model then allows one to predict (design) what one will get under a signal condition of your choosing. Once one gets much higher in frequency, it is the effect of all the R’s L’s and C’s in a circuit path, cell phones and computer circuitry wouldn’t be possible without ways of modeling an arbitrary signal path and all its inductive and capacitive elements.
Another cool thing you can do is look at a cable while it’s carrying a signal, you can examine the signals at each end and subtract them so that you’re looking at the difference between the two ends of the cable.
If the cable is altering the signal in any way, that must show up as a difference between the input and output end.
If you want to take REALLY LOW distortion measurements, you can subtract the magnitude and phase of the distortion in your input signal from the test signal so source distortion is null’d out.
If you wanted to make an ultra low alteration speaker cable, what you would want is a low loss R-8 family like 9913. Per foot, I don’t think it is possible to have an easier path to a speaker cable with lower series R, lower series L and lower parallel Capacitance and lower energy storage in terms of reactance. I made a 125 foot loudspeaker cable for testing loudspeakers made of a pair of these cross coupled (center of one and shield of other = once conductor.). So far as alteration of signal, it was equal to about 10 feet of two Kimber cables in parallel. Here, if interested, find some of this stuff, compared to high end hifi speaker cable, it’s cheap.
http://www.theantennafarm.com/catalog/times-mic-lmr-400uf-1472.html
9096 coax here;
http://www.radioworks.com/ccoaxstd.html
There are others too all in the old RG-8 family / size.
Also while this doesn’t make for an intriguing speak cable add, a great deal of effort has gone into making ultra low loss / stray component coax cable because “the hidden stuff” in the cable is with increasing frequency is what limits the cables effectiveness. As a result, these kinds of cables in every way are about the best behaved wire it’s possible to make in that gage.
Tom
This technique measures many interesting characteristics, but alas, it is incomplete.
The effects of crystal lattice are totally overlooked. One needs a high-speed stimulus and real-time multi-GHz scope to see these things. A TDR can also be used.
Just put your cable into a bath of liquid nitrogen and remove it and see if any of these measurements change. They won't. However, the SQ will suck due to the broken crystal lattice. The only way to see this change is the equipment I mentioned. Been there, done that.
Steve N.
"The effects of crystal lattice are totally overlooked. One needs a high-speed stimulus and real-time multi-GHz scope to see these things. A TDR can also be used."
Love to see those pictures, if you still have the necessary test equipment.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
The graph above shows the same interconnect cable before and after dipping in liquid nitrogen. Red is after. Notice all of the new reflections.Steve N.
Edits: 11/18/13 11/18/13 11/18/13
Very interesting. Thanks.
What was the scale? I had trouble making out the lettering on the graph.
There appears to be a DC offset between the two cases, unless this was an artifact of the plot to keep the curves separate. Could you explain what you think is happening?
Have you done these tests in a controlled way, e.g. with two cables that started out with identical measurements and then one was treated? This would make it possible to run repeated (even random) tests to show definitively that the differences weren't due to some other factor, e.g. power line voltage, noise, temperature, warmup of test equipment or whatever...
Or maybe these measured differences is old hat to cable experts. I just haven't come across it before, but then I haven't gone looking either. Thanks.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
The two cables had identical measurements, before and after the TDT measurement.If I remember correctly I moved the two graphs so you could see the detail.
I ran the test multiple times at once on both cables. It is definitely not due to power or anything else. No power variations will affect this equipment.
Edits: 11/18/13
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
It was posted on my website for many years, but I discontinued selling cables a few years ago and it was removed from the website.It just demonstrates that even at audio frequencies, there are things that the typical measurements do not detect, but the human ear does. My brother is a metallurgical engineer, so he educated me a bit on metal crystal lattice. Metallurgy is the aspect that most cable manufacturers ignore or don't understand. This is why there are good silver cables out there and poor ones. They are not all the same by any means.
Edits: 11/19/13
How do you know that the differences are due to changes in the metal and not to changes in the dialectric?
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Because I have designed cables with poor metallurgy and good metallurgy with the same conductor material, gauge, dielectrics and construction. The good metallurgy trumps the poor. Been there, done that.
It's quite possible the strange spectra shown in graph explains why practically noone employs direct immersion, at least for audio related items.
Actually I'm using Kimber Kable which both measures and sounds fine to me. I no longer have access to a network analyzer so just measured the swept magnitude and phase.
I have a roll of smallish diameter Heliax in the garage that I got years ago for a VHF Ham antenna when I lived in the sticks but never used and I've always wondered if that might not make a good, albeit awkward, speaker cable. If RG-8 is good...
Rick
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: