On Tue, 8 Jun 2004 17:05:53 -0400, "Jack Painter"
wrote:

"Gary Schafer" wrote

Oh boy! I just got back from vacation and am just now reading this
stuff.

Jack, Bruce and the others are entirely right. I once had a hard time
figuring out why RF would not flow on the inside of a tube too. It
would seem logical that it would do as you say but it doesn't.

Look up "wave guide beyond cutoff". That will answer your question
about why rf dose not flow on the inside of a tube.

It will flow on the inside for only a very short distance from the
opening. Then it gets canceled. This is how many signal generator
attenuater work.
They use a tube of 6 or so inches long with a sliding probe inside fed
from one end. On the other open end is a fixed pickup probe. When the
movable probe is close to the fixed probe on the other end, maximum
signal coupling is obtained. As the other probe is moved away inside
the tube the signal becomes highly attenuated.

It is operating as a wave guide that is much too small for the
frequency involved. If the tube diameter was made large enough to be a
quarter wave length in diameter then the rf would propagate through
it. But that would be in a different mode than the skin effect
conduction being discussed.

By the way did you know that skin effect even comes into play in 60 hz
distribution systems?

Regards
Gary

Hi Gary, welcome back, and thanks for your replies.

Right principles, wrong application. Trying to apply high power microwave
principles (3-15 gHz) to low power 2-30 mHz) is not the same.

Sorry Jack but you are wrong. It has nothing to do with microwave
frequencies. A wave guide beyond cutoff is the mode that the tube is
operating in and it simply tells you that the frequency is too low for
the given size tube to propagate through. The energy inside the tube
gets shorted out. Many 2-30 mhz signal generators use that type
attenuator.

Now at 100 mHz
and below, while there would still a small but measurable difference of skin
effect at high transmit power, it ain't much and has nothing to do with low
power 2-30 mHz where a thin walled copper tube has ZERO measurable
difference in skin effect to a copper strap of even slightly smaller gage.

It has everything to do with it. Skin effect is ever present in all
conductors at ALL frequencies. Note my reference to 60 hz power
transmission where it is also important.

That has been my never paid attention to point all along, that skin effect
involves the entire cross section of thin material, and copper tubing is
more than thin enough to carry current in it's entire (that means from outer
to inner surface) cross section. That's exactly why copper tube is used so
much in AM broadcast components.

That is a contradiction to your point. You say that current flows
entirely through the walls of copper tubing and then say that is why
it is used in AM broadcast components. If that were true then they
would not use copper tubing but instead they would use solid copper
rod for better conduction.

The reason copper tubing is used is that there is no current of any
significance past a certain depth and to use solid rod would be a
waste of copper.

This is not even related to waveguides
which must by design AVOID all skin effect which causes great resistance and
heating at the current and velocites involved in microwave transmission.

Well, microwave transmissions don't travel any faster than HF
transmissions. But you might note that most wave guide inner surfaces
are silver plated to reduce skin losses.


As we eventually got around to research rather than blindly arguing
positions of opinion, then the participants hopefully learned something.
I've learned that applying the math from formulas for skin effect in
conductors of known ohmic value and used with a known frequency can
determine the wall thickness of a conductor which has full cross sectional
current on it. Guess what? The original poster's question about using copper
tubing remains answered. A 1" copper tube has more surface area and carries
just as much low power RF on it's entire cross section as a 1" wide piece of
copper strap that is nearly the same gage.


While skin effect is a gradient and not an absolute barrier, there is
current that flows at all levels in a conductor. Even on the inner
surface of your copper tube. But the amount of current there is so
small that it is immeasurable. It decreases exponentially.

One skin depth is defined as the depth at which the current has
dropped to about .37 times the current at the surface. (If you notice,
this is the same decay rate that a capacitor has when it charges or
discharges.) When you go that same distance (deeper) again the
remaining current will again drop to .37 times the current that it was
at the first skin depth.

So you can see that the current never reaches zero as you go deeper
but it only takes a few skin depths to decrease the current to a very
small value which is insignificant.

..0058" is the skin depth in copper at 200 khz. Skin depth decreases by
10 for each 100 times increase in frequency. So at 20 mhz the skin
depth would decrease by 100 from that. It gets pretty thin!


Skin effect is the reason coax cable works as it does. None of the RF
on the inside of the cable appears on the outside of the cable. Other
than leakage between strands of the shield of the cable. Those wire
strands on coax cable are pretty thin. Much thinner than your copper
pipe. Hard line has no leakage.

Regards
Gary


Best,

Jack Painter
Virginia Beach Va