Originally Posted by Robert Macy

On Jul 16, 10:58*am, Roy Lewallen wrote:
I'm sorry, this response contains some misleading advice.

Robert Macy wrote:

Just saw this thread and some very good suggestions.

One thought. *Unless your close to the TV Transmitters, stay away from
using wire as the antenna. *There is a good reason antennas are made
of tubes, not wire.

The reason usually is physical strength and rigidity. The larger
diameter also increases bandwidth, but often this isn't necessary to
proper operation.

*At high frequency, like television transmission
frequencies, the current creates a repulsion field that pushes the
current away from the center of the conductor. *In other words, all
the current travels on the outside surface of the wire. *Look up the
term, "skin depth". *At frequencies as low as 20MHz, more than 99% of
the current will be within 3 mils of the surface. *The only easy way
to lower the losses in the antenna is to use large diameter
conductors, but since the inside of the conductor carries no current,
you don't need metal there, so it is ok to use hollow tubes.

This is true. However, in almost all cases the loss caused by using
wire, even very small wire is still negligible. Exceptions are antennas
which are very short in terms of wavelength, particularly at low
frequencies. As frequency increases, the length of an antenna of equal
performance decreases in direct proportion. However, the loss decreases
only as the square root of frequency. So antennas of the same wavelength
size become proportionally less lossy at higher frequencies.

Antenna manufacturers save themselves money by lowering material costs
and shipping weight. *They use hollow tubes.

Another important reason for using hollow tubes is structural weight.

*If you don't care about
weight or material cost, go ahead and use solid rods, 1/4 inch, or
even 3/8, but stay away from 18 Awg, way too small.

18 AWG wire won't result in appreciable loss for nearly any antenna.

One other thing, nature abhors sharp edges, that's why bubbles are
round, so don't use square tubes or sharp bar stock either.

Square stock is slightly lossier than round, but the loss will be
negligible when unsing any practical size.

*Use
rounded tubes or rods. *Even smoothing and polishing the surface
lowers the resistance.

Polishing won't make any detectable difference.

When you're done, passivate the surface of the
conductors to prevent corrosion over time. [meaning: paint the
antenna] *Over time, corrosion will deteriorate your antenna's
performance.

It depends on the type of corrosion. But it would have to be severe
before becoming so bad as to cause an appreciable reduction in
performance. Aluminum, tin, and some other metals passivate themselves
by forming a hard insulating oxide layer on the outside. Unless you're
in a maritime climate, copper won't deteriorate in a way that matters,
either. Insulated wire is an easy way to prevent corrosion in an
unfavorable climate.

Rounded surfaces also means make your connections smooth
with nice transitions. *As in, "if it looks good, it works good."

If only that were true! But unfortunately it isn't.

You
can use aluminum if the lengths are continuous and/or you make
connections using constant mechanical pressure, like a "lots of teeth"
star washer that has bitten down through the insulating oxide layer
held with a bolt.

This can cause more problems than it solves, if the bolt and washer are
the wrong metal such as steel. A good book on Yagi antenna construction
will tell you about techniques for working with aluminum.

All in all, it seems a lot easier to buy a fringe field antenna and
put that in your attic. *But if you do it yourself, hope you're
successful, document what you built, and share it here.

With that I agree.

Roy Lewallen, W7EL

My advice was NOT misleading. Yours was a very poor choice of word.
Misleading means the suggested effort would result in moving away from
an optimum solution. ALL of my advice leads to better solutions and
is therefore "not misleading" and I stand behind my suggestions.

You are very correct on catching the lack of thoroughness addressing
'mixed metal' contacts. Yes, a lot of electrolytic action happens at
the junction of dissimilar materials. One must be extremely careful
when making such contacts.

Had you criticized my comments by suggesting that many of the efforts
involved will not yield noticeable improvement [especially to a
novice], I would have accepted that. After working for years in low
noise, high performance systems these techniques have become de
rigueur for initial construction. The OP probably would not notice
improvements except in the most fringe of conditions.

Yes, the diameter of the rod will broaden bandwidth, but elements
having ratios on the order of 80:1, the effect on bandwidth won't be
very noticeable. Length variations and spacing will have more
impact.

Regarding skin depth of a conductor: Always keep in mind that the
skin depth equation is based upon the assumption of PLANAR wave. The
equation is extremely simple and easily memorized as the square root
of 2 divided by three terms:

skin depth(in meters) = sqrt( 2/(p*o*w) )
where p = magnetic permeability
o = conductivity
w = frequency in radians

for copper, p = 4 pi 10-7
o = 58 MS/m
w = 2*pi*f, with f in Hz

results are in meters, so I suggest using an Excel spreadsheet
formula.

skin depth of copper at 80MHz is 0.3 mils! 99% of the current is in
less than 1 mil of the conductor.

Using finite element analyses [femm 4.2] techniques it is easy to
calculate the impedance of a conductor as long as the dimensions stay
below 1/10 of wavelength. For 80MHz that would be 1.2 feet. At
80MHz, 18 Awg copper wire is approx 220 milliohms per foot and 3/8
inch aluminum tubing is approx 36 milliohms per foot. Neither of
these impedances would have much impact to the signal coming from the
377 ohm source impedance of free space.

Normally we would have predicted the decrease in impedance by applying
the ratio of the increased perimeter reduced by the less conductive
aluminum. The ratio of perimeter is 0.375/.04 or approx 9.4, but
aluminum is not as conductive as copper so the conductivity ratio is
25/58 for a total change in resistance by approx 0.375/.04*(25/58) =
4.04 improvement. Finite element analyses calculates the improvement
to be more than 50% higher than that. [It's caused by the small
radius of the wire.] Plotting the current density down into the
conductors shows what happened. The 'effective' skin depth in the 18
Awg wire is about 50% less than in the aluminum, all due to the
reduced radius of the outside of the conductor. As I said, nature
hates sharp edges.

Again, as the element 'taps' into free space the 377 ohms of free
space predicts a difference of less than 0.005 dB on the signal. So
there will indeed be an extremely small effect from using 18 Awg wire
or 3/8 tubing on the received signal strength.

If the antenna were to be much smaller than wavelengths and
capacitance were added to resonate the elements, THEN the impedance
dfference would become noticeable and affect how much signal is
available to the receiver.

Regarding bar stock? picture the current concentrating at each of the
four corner edges, with the current not being uniformly distributed
around the perimeter. That would almost be equivalent to having 4
parallel small diameter wires mounted on 3/8 inch centers! Really
wasted the metal.

Regarding the importance of smoothness: My outside antenna became
badly pitted from atmosphere, even with aluminum developing an oxide
layer, it still corroded. The roughness lowered the gain of the
antenna enough to notice it on the reception from fringe stations.
Years ago [and at higher frequencies] we plated metal in our
resonators with silver *and* polished the silver to get the impedance
down. We measured huge differences in impedance [measured as improved
Q] as we polished the surface down to mirror finishes.

Whoever suggested the ARRL Antenna handbook is right. Great book.

Regards,
Robert