That is a mis-application of the equations...to wit, it is the mis-application of an exponential skin depth approximation formula to a construct where the approximation is no longer valid. To calculate properly, one must enlist the aid of "Bessels", which most people are incapable of manipulating. The exponential approximation is used outside of it's region of applicability simply out of ease of use...we were taught to use it in lieu of actual high level math.
The skin depth of a planar E/M wave normal to a flat conductive copper sheet is as per calculation. Note the key words "planar E/M wave" and "flat conductive copper sheet".
There is a significant difference between that scenario, and the scenario of current travelling within a conductor we call "wire". Within a wire, the magnetic field internal is a consequence of the transport current, NOT a consequence of a planar E/M wave driving normal to the conductive surface.. The effect we call "skinning" is a consequence of that transport current changing amplitude, and doing so within a conductive medium. That medium's reaction to the changing current is to create eddies which are responsible for altering the overall current density profile.
The exponential equations are off by a factor of at least 3 in the audio range for a wire of 1mm diameter. At 20 KHz in a 1 mm diameter wire, the central current density is about 80% of the surface density.
Primarily because it was born of a mis-application of both theory and test. The failings in his test setup are not "intuitively obvious to the most casual of observers"...it requires an understanding of problems associated with measurement of high slew rate currents in very low impedance circuits. I understand that topic rather well..
Which is indeed nice, I'm confident he deserved it for other contributions. But his Essex echo article is not one of them.
jn
Last edited:
For a treatment of skin effect in round wires see section 4.4 in chapter 4 of 'Fields and Waves in Communication Electronics' by Ramo, Whinnery and Van Duzer. Don't bother unless you are happy with Bessel functions. As jn says, the effect is smaller in round wires than the planar theory might suggest. Their figure 4.4a shows a 1mm copper wire has about 45% current density at the centre even at 100kHz. You have to have fatter wires (diameter > 4 'skin depths') before the planar approximation can be used reasonably well.
It is interesting to conjecture whether Hawksford's mistake would have been spotted by proper peer review. Given that many EEs seem unaware of the difference between planar and round geometry for skin effect it might still have got passed for publication.
It is interesting to conjecture whether Hawksford's mistake would have been spotted by proper peer review. Given that many EEs seem unaware of the difference between planar and round geometry for skin effect it might still have got passed for publication.
Is there any data on this?
With copper (not a ferromagnetic material) this is not an issue, is it?
Electrostriction on wire insulators would be an issue if the electric field would be above some hundred kV/m
If by “transport current” you mean the electric current flowing in the conductor (as measured with an ampere meter), isn’t this current a consequence of a planar (or a bit different) E/M wave driving along the conductive surfaces of say two parallel conductors or along the conductive surfaces within a coaxial cable?
George
No conjecture is required. It would not have survived peer review. In fact, I read some rather unkind words written about the whole thing in a different publication. Man, nasty ain't the word for it...
It would have required cherry picking reviewers with no e/m theory background. By default, that selection process would negate the term "peer".
As to "many EE's seem unaware"...Since there are many specialized disciplines within the category "EE", it is unfair to use such a broad paintbrush. Circuit theory, switchmode design, PCB layout, semiconductor physics, test engineer, digital, analog...
The same could be said of physicists...condensed matter, high energy, nuclear, heck, even dark matter...it's all physics after all, right??
jn
Journal editors don't always pick the most appropriate reviewers. They sometimes interpret 'peer' to mean 'someone working in an academic department with the same name on the door'. Not necessarily someone senior, with lots of knowledge and experience. They sometimes just pick someone who has recently published a paper on a similar topic - by that method I have reviewed a few papers.jneutron said:
I have a Masters and Doctorate in EE, but it was only fairly recently that I realised that geometry has such a big effect. It is there in one of my Master's textbooks (as mentioned above), but it certainly was not part of the course. An EM expert would know it, of course, but not necessarily a run-of-the-mill communications engineer or audio engineer - yet these are the people who an editor would regard as 'peers'. Hence my comment was not meant to be critical of EEs. Critical of editors, maybe!
I've reviewed only for physica, so cannot comment on many of the other journal picks. Although I suspect you may be accurate.
I recall that after John Atkinson published Hawksford's '85 article, JA duplicated the test. The results did not repeat.
jn
I had a private communication with someone we all know and hate 
He wrote: “… (we) need to think through the implications of believing audiophile legends; the question I always ask myself is, "If this belief is true, would (fill in the blank- computers, automobiles, airplanes) still work?”
I read publications and sites for radio amateurs. Well, I scratch my head on what are we dealing with in audio electronics sites!
George
He wrote: “… (we) need to think through the implications of believing audiophile legends; the question I always ask myself is, "If this belief is true, would (fill in the blank- computers, automobiles, airplanes) still work?”
I read publications and sites for radio amateurs. Well, I scratch my head on what are we dealing with in audio electronics sites!
George
I interrupted listening to Bastianini and Scotto for you
My preference goes as this:
F,G,A,D,E, (H,B,C).
The rating (I listened to them twice) was through the piano, violin, female voice.
Just for the ridicule of it, I will have a look later at your previous post for the flac files explanation.
George
Last edited:
What I had downloaded and listened to were 8 files. What are the rest then (D,G,H)?
George
http://www.youtube.com/watch?NR=1&v=mYpwQsHb2_o&feature=fvwp
mud in your ear - muddy waters - YouTube
Last edited:
Hey, which would you rather have in the morning - a cup of coffee or a cup of copper?
The experiment has been interesting so far because it seems that there is a slight preference for the original file. That's good, as it shows that maybe this DA-AD loop isn't transparent. After that, no real pattern emerges. Banana, potato, coffee, mud or copper - no one material dominates.
The experiment has been interesting so far because it seems that there is a slight preference for the original file. That's good, as it shows that maybe this DA-AD loop isn't transparent. After that, no real pattern emerges. Banana, potato, coffee, mud or copper - no one material dominates.
Unfortunately not. This is a giveaway, in the solo piano portion, between the 2 sections of playing - comparing original, F, with copper, B - the highlighted section, 11.00 - 12:30 secs, has been amplified 50dB:
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.