|
Copied in full, my reply to /audio/tweaks/messages/160.html : *************************
jj, I know that you seem to consider such levels (1%) as immaterial for audio,. perhaps your codec work has displaced your sense of what is relevant, maybe I'm the one who's all wet. It is my opinion that signal distortion that may occur as far down as -80 to -90 dB may be relevant to SOTA audio playback. If skin effect eddy currents cause a 1% level of effect, then that is only -40 dB down. It would also be likely affected by the homogenity (or lack of it) within the wire, and therefore be less than predictable and wholly
linear in it's effects. Solder junctions, and other such discontinuites come to mind here. Also, there is no doubt that the propensity for the current to flow toward the outer surface of the wire will magnifiy any surface discontinuities or disruptions (ala tinning or platings). If there is a net difference
between the surface and the center of a wire/wire bundle, then it is possible that effects that range from -40 down to -80 dB or so ARE occuring. To me, that would be reason enough to look into it, or consider it in a cable design. Jon Risch
************************************** Now I don't consider writing that "....maybe I'm the one who's all wet." and that " It is my opinion that......" qualifies as a flame or a poke at jj that I did not also take at myself, or that I did not consider that what I was stating was anything more than my opinions. I am quite aware of the difference between loss and distortion, and carefully worded my post: { If skin effect eddy currents cause a 1% level of effect, then that is only -40 dB down. ] An AC signal that is 1% of the original, and inverted, will cause a 0.1 to 0.9 dB drop in voltage amplitude. [ It would also be likely affected by the homogenity (or lack of it) within the wire, and therefore be less than predictable and wholly linear in it's effects. ] My point here, that I did not spell out, was that if the eddy current was at a 1% loss level, and was not totally linear in it's nature, then it would be possible for it to create a signal distortion. This would, of course, be X amount of dB down from -40 dB, and if the amoung of eddy currrent distortion was 40 or 50 dB down from the eddy current s themselves, not an inconceivable amount, then the total amount of distortion is in the range of -80 to -90 dB down depending on which set of numbers you pick. Again, not out of the range of potential audibility. Considering that there have been no studies (that I know of) that address the effect that skin effect would have on a complex musical signal, one with many frequencies present simultaneously, there is no telling exactly how it might affect audibility, based strictly on scientific conjecture and suppositions. Past posts of yours have given the impression that you were unconcerned with distortion of even higher amounts than 40 dB down, but that series of postings ended with bad feelings on your part also, so I can not be sure Your statement that the loss of 1% would only constitute a 0.04 dB amplitude drop is correct for POWER, but for the voltage in an interconnect, it would correspond to a 0.09 dB drop, or almost a tenth of dB. If you don't care to discuss the subject, then don't post a response, that is what I thought had happened. Now you come back posting a tantrum, and claiming to be insulted by my post. I do not have a mean bone in my body, and try to be very patient in my web postings, it doesn't help when you are imagining insult and intended injury when there is none intended. If I was that unclear, I should try to be more precise, and for any idea that I was trying to really slamm you, I apologize. If you would care to discuss the issue further, then I welcome any input or thoughts you have on the subject. Jon Risch
|
