ark@rabbit.UUCP (Andrew Koenig) (03/12/84)
I had thought that audio frequencies were way too low for the skin effect to make any measurable difference in speaker wire. I knew that skin effect was important at radio frequencies, but that at lower frequencies it would be insignifant. Perhaps not. I have just learned for the first time what the numbers actually are. Apparently, the skin depth is 0.001 inch at 1 megaherz, and goes as 1/sqrt(frequency). Thus at 20 kHz the skin depth is only 0.007 inches, which is pretty close to the diameter of a strand in ordinary stranded wire. It is therefore conceivable that insulating the individual strands would make a measurable difference in the behavior of speaker wire at 20 kHz and lower frequencies.
dave@rocksvax.UUCP (Dave Sewhuk) (03/14/84)
Some of us here looked up your info on skin effect and indeed it
appears to be within reason, however look at this from another angle,
using the numbers from our trusty Electronics Engineers handbook, we
found that that @ 20Khz the depth will be about 20 mils. Doubling this
to get the diameter we find that this is about 18 gauge wire. So until
you get down to 18 gauge you are not even close to are making any loss.
The skin effect is defined as the point where current density at depth
mentioned is 1/e (37%) of the current flowing on the surface of the
conductor. It is not a digital effect, but describes the exponential
decreasing AC current you see as you head toward the center of the
conductor.
>From the table of "largest permissible wire diameter for skin effect
ratio of n" for a frequency of 100 Khz to keep the AC resistance within
1% (ratio 1.01 AC/DC) of the DC resistance we find that 27 gauge is the
number.
Seeing that we usually want to keep the DC resistance down we use
larger gauge wires usually made of smaller gauge strands, you only
should worry if the strands of wire are larger than noted above.
In a bundle you will find that each strand is really poorly coupled
to its neighbor due to oxidation, so that seperately insulating each strand
should not be required.
--
Dave
Arpa: Sewhuk.HENR@PARC-MAXC.ARPA
uucp: {allegra, rochester, ritcv, ritvp, amd70, sunybcs}!rocksvax!dave