Roy,

Thank you. It is a quick experiment to build a test coax coil and measure the Q.
That should produce enough evidence to test a counterpoise.

In the mean time the research to build a coil out of copper tubing continues. So
far the only alternative I can conceive is to make a wooden form and wrap the
coil on the outside.

Thanks again - Dan

Roy Lewallen wrote:
Mike Speed wrote:

Two factors. One is surface roughness. The other is caused by the
current having to continually move from one group of wires to another as
it travels.


Interesting, but how do you know the current is moving as you say?


Skin effect is well known. On a good conductor at high frequencies,
current is concentrated very near the surface. When a bundle of wires
ducks under another in the direction of current flow, the current has to
migrate to the outside again, which means it has to move from one
conductor to another. There's no question that it happens -- what's a
bit harder to pin down is just how much loss typically results.

I'm just now doing some research on how significant these
effects are, but so far I've found out they're very noticeable.


Ok. I'm curious: What equipment are you using for the research?


Books, and to a lesser extent the web. Information about this is
scattered among a number of sources. Quite a few discuss surface
roughness in a general way, but there's a particularly good explanation,
analysis, and something of a quantitative treatment in Johnson &
Graham's _High-Speed Signal Propagation: Advanced Black Magic_. The
effect of weaving is harder to track down -- most authors simply assume
coax shield conductivity loss to be negligible, and don't deal with
woven conductors in any other context. But it really isn't, if you're
interested in good accuracy. And of course when the braided conductor is
the primary conductor, it becomes much more important. I know Tom, W8JI
has done some measurements on braided vs solid strap, and I'll be asking
him for more information before long. I do know that he found a very
significant difference, and I have a great deal of respect for his
experience, measurements, and opinions.

But hey, you don't have to believe me. Make up some coils and measure
their Q -- it's not hard at all. Then stick them outside for a while and
measure them again.


What would be a good way to measure Q?


The way I do it is by resonating the inductor with a parallel air
variable capacitor. It's important to keep it away from just about
everything. I couple in and out with a very small (typically 1 pF at HF)
capacitor, and make sure that the impedances of the source and detector
are either very high or quite low (say 50 ohms) to minimize loss. I use
a signal generator for the source and a scope for the detector. Using a
frequency counter connected to the signal generator, I measure the
resonant frequency and -3dB points. The Q is the ratio of the center
frequency to the 3dB bandwidth. For convenience, I made a 3dB pad I can
switch in an out of the signal generator. With this, you don't even need
a linear detector, and a diode and meter would do. My measurements have
been within about 5 - 10% of readings with a good HP Q meter on the few
occasions when I've compared them. That's close enough for my purposes.

Or do like most amateurs do -- make the coils,
discover that you can talk to far away places "barefoot", and declare
that they "work".


Uugghh, don't I know.


Many people have worked the world with 1 watt, knowing that's what they
were running. A lot more have worked the world with 1 watt, thinking
they were running 100. Ignorance is bliss.

Roy Lewallen, W7EL