Jeff says:
> How can I model the increase in resistivity of a coil or stripline at
high
>frequencies due to skin effect?
Skin depth is given by the eqn:
d = (2.6/sq.rt. f)K
K is one for copper, f is in MHz, and the depth is in thousandths of an
inch ('mil')
K is given by sq. rt.[(1/u) p/pc] u is the relative permeability, p is
the resisitivity, pc is the resistivity of copper.
One skin depth is where the current has decreased to 1/e or approx 63%
of the surface value. It's normally assumed that at a depth of greater than 5
skin depths, the current is negligible.
Which is why 1 oz copper PC board doesn't make a good shield at HF.
So I'd suggest modelling with a conductor that is 10 skin depths thick for
a coil, then plug in the resisitivity of copper, and you'll get a
value per unit length for the inductance. For a
stripline, don't forget both conductors.
Have fun!
>1. Some hi-fi purist advertising promotes "oxygen free copper" as
having
>better properties for propagating audio. I alway assumed that was the
>fanciest possible way to say "not corroded". What do you (or the list)
>think?
Personally, I believe it's BS.
>2. Sometimes people talk about the the crystal structure of the copper
in
>cables. Seems like BS to me. Any possible effects?
Only on your wallet! Seems like BS to me, too!
Goes with the 'only' way to do audio is with triodes in class A, without
feedback, using transformers handwound with silver wire. And replace the triode
every 300 hundred hours. No electrolytics in the power supply, otherwise you get
'electrolytic sound'. And the pins on the mains connector must be gold plated to
avoid non-linearities.
73
Peter G3RZP
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|