Correct, and I should have said that. The field is down 95% by 3 and 98% at 4 skin depths.
I can’t argue with this because, as I said, I haven’t looked at this stuff for a long, long time. However, the text by Ott that I quoted says that for a conducting shield 1 skin depth thick, the "absorption loss is 9 dB. Doubling the thickness of the shield doubles the loss in dB." This indicates that the energy in the field that penetrates the conductor is being dissipated inside the shield somehow.
True. And even before those frequencies, the circuit model gets dicey because the components can’t be considered as lumped elements having on the characteristic of resistance or capacitance or inductance. Components first begin to act as resonant circuits and then with increased frequency have to be regarded as distributed elements in which case the circuit approach is awfully cumbersome. Although with computer modeling it might not be that bad.
Things are beginning to fall apart around 1 gigaHz for ordinary circuit elements as I remember it. Elements in integrated circuits can be made small enough that the frequency can be higher before you run into trouble but I have no idea what it is.
There’s no real practical significance to my last post as far as circuits go. 10[sup]18[/sup] Hz means a wavelength of 3x10[sup]-10[/sup] m, i.e. x-ray type wavelengths.
After thinking about this one for a while, I still can’t work out what it means. If you come back to this thread, Ring:
In a perfect dielectric, the E and B fields are 90[sup]o[/sup] out of phase, and so no energy can be dissipated, but that doesn’t mean that energy cannot be transported.
“Evanescent waves” aren’t something I’ve come across regarding field penetration in conductors, but rather total internal reflection in dielectrics.
I make it 3 x 10[sup]-11[/sup] m, which is gamma rays (10[sup]-10[/sup] - >10[sup]-15[/sup]).
Mostly, however, I’m curious about what property of conductors this phenomenon is related to. I’ve not come across this in any of the literature, since, as you say it’s meaningless for practical circuit design. So, what is it for?
What do you want me to say? I know how to use a calculator, so do you. 3x10[sup]8[/sup] ms[sup]-1[/sup]/1x10[sup]18[/sup] s[sup]-1[/sup]=3x10[sup]-10[/sup] m.
The distinction between x-rays and gamma rays is somewhat artificial. X-rays come from electron transition, gamma rays come from nuclear interactions. The frequencies overlap.
What’s it for? It demonstrates the boundary between conductor and dielectric. It explains why gamma rays penetrate metals.
And it’s probably me. After checking textbooks and searching the Web I can’t seem to find any globally agreed upon definition of what an evanescent wave even is.
Roy McCammon is a guru at IEEE.org and here’s what he says:
(Please notice he uses the same scientific term I did (slosh))
I can back up Roy McCammon. I did a quick look through several of my EM books. They all use evanescent only when the fields are non-lossy, with E and H 90 degrees out of phase. Attenuation is used when the fields are propagating and decaying. Using this definition, the fields in a good conductor are attenuating, but not evanescent.
One source of misconception is the fact that the imbalanced charges on conductors can only exist on the conductor surface. E-fields and circuit voltages are created by this net-charge or “charge-imbalance,” and it drives electric currents within wires. Some people wrongly conclude that since all the charge is on the wire surface, all the current must also be on the surface, since current is just charge-flow. Wrong.
In wires, electric current is a flow of cancelled-out charge. The population of electrons inside the wire is mixed in with a population of protons of the metal atoms, so the net charge within the metal is zero. Yet the negative charges can flow along through the positive ones. Even though the NET charge is zero, the wires still contain an enormous amount of movable charge. If the electrons flow through the protons (or vice versa) then we have a large current even though we still have zero net charge.
A net charge does exist on the surface of wires. Yet this charge is negligable when talking about the flowing charges in a current. The surface charge is on the order of picoCoulombs, while the electron population with a wire is on the order of tens of Coulombs. The surface charge is only important in generating e-fields and causing voltage. Here’s a short paper on this stuff:
Electrostatics and circuits, R. Chabey, B Sherwood (PDF)
You’re not alone! I spent lots of hobby time re-teaching myself electronics from the ground up, rooting out errors and misconceptions, etc. Electricity is fouled up because it’s taught wrong in grade school, and even PhD people still have those wrong concepts stuck in their heads. We’d have to perform some intentional “un-learning” to root them out of there. Most people never do this. See: Electricty Misconceptions Spread By K6 Textbooks
Here’s a personal favorite: a great website at MIT that shows us the EM fields around a dipole antenna as the antenna emits radio waves. (Large MPEG download, but worth it!)
The name of the stuff that flows within wires is CHARGE. The word “current” means “charge flow.” What flows in rivers, water? Or “current?” What flows in wires? Charge? Or “current?” To help defeat this error, just call them “currents” instead of “current.” “Current” sounds like the name of a substance, while “Currents” are like motions.
This stuff might seem like a minor quibble, but I’ve found that it has major consequences. It leads to sloppy thinking on the part of textbook authors, and they end up putting many other errors in texts on intro electrical physics.
Dunno about you, bbeatty, but I find it conceptually useful to talk about things like “a current of 100 A flowing from a to b”, even though it’s AC and the charge isn’t flowing anywhere.
And apologies to the others for my earlier, sloppy use of the symbol “B” in place of “H”.