On Fri, 21 Nov 2008 18:24:20 -0800, Roy Lewallen
wrote:

Flat conductors aren't as attractive as they look at first glance. The
problem is the same proximity effect mentioned earlier in the posting.
Current is distributed evenly around a round conductor (assuming the
perimeter is a very small fraction of a wavelength), but not along a
flat strip. Because of proximity effect, the current is much more
concentrated near the edges than at the middle. The result is that the
resistance is considerably higher than for a wire with the same surface
area. In figuring an "equivalent diameter" of a thin flat strip in order
to get the same L and C properties, the rule is that a strip is
equivalent to a wire whose diameter is half the strip width. This means
that a strip of width w or total "circumference" 2 * w is equivalent to
a wire with a circumference of pi * w / 2 ~ 1.6 w, in so far as L and C
go. Since the same phenomenon affects the inductance and resistance,
this would also be a good working rule for estimating the relative R of
a strip or wire.

Roy Lewallen, W7EL

Thanks. I think you just explained the cause of a problem I fought in
about 1980. I had "designed" a 930MHz yagi antenna for a utility
telemetry system. In order to cut system costs, I decided to build
the antenna from stamped 0.062" aluminum. My initial dimensions were
stolen from a Scala yagi which used approximately 0.500" diameter
round rods for elements. I reasoned that to obtain the same
bandwidth, I would need to use the same circumference as the rod. That
made the initial prototypes elements 0.8" wide. After some tweaking,
the antenna tuned to the correct center frequency, but the 2:1 VSWR
bandwidth was much less than the original Scala antenna.

So, I increased the width of the stamped elements (with aluminum duct
tape) until the bandwidth improved. I landed at 1.25" or 2.5 times
the width of the rod elements, somewhat larger than the recommended
2.0 times the rod diameter.

However, when I added a coined stiffener groove to the stamped "boom"
and elements, the bandwidth increased again, to much more than
necessary. After the usual all night cut-n-try session, I landed on
2.0 times the width of the rod elements, with the coined stiffeners,
which apparently increased the effective diameter of the elements.

Coining the "boom" also wrecked all the element tuning since it
increases their effective end to end length by the depth of the
coining. I had a hell of a time dealing with the sheet metal vendor,
trying to control the stiffener dimensions. It seems that aluminum
stretches when coined, often in a rather unpredictable manner. I
eventually gave up and went to 0.125" sheet aluminum and no
stiffeners. Unfortunately, only a handful of prototypes were made and
shipped, so I have no clue as to how well (or badly) they worked in
the field.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558