Am I reading this right, or am I missing something?

I took a look at the Smith Chart, and for impedances in this region
of the chart, series L and parallel C is not the way to get a match. In
my version, 80-j300, you need 4.7pf shunt C and .15uH series L. I have
no idea how Fig. 1C in the article managed to get a good match with the
single insulated wire up the middle of the radiator.

It's possible he isn't creating a full match (with an L network) in
this case. He might just be cancelling out the negative reactance,
using a combination of (series L from the stub, and a bit of series C
from the capacitive coupling between stub-feed and radiator), and not
bothering with a shunt component at all. This would, perhaps, result
in an 80+0j feedpoint impedance and about a 1.6 SWR at the feedpoint,
which would probably end up significantly lower at the other end of
the feedline due to feedline losses.

Or, there might be something stranger going on, with the stub giving a
bit of shunt C to ground (in the PL-259), some parallel L/C inside the
radiator, and six other bits of odd voodoo.

The author says that it ought to be possible to get down to below
1.5:1 on the repeater portion of the band... this suggests that the
design isn't one which "tries" to achieve a true 1:1 match. The WA-2
and similar 5/8-wave antennas using a tapped coil seem to be able to
get down arbitrarily close to 1:1 at their best.

Beats me. Almost makes me want to try building one just to measure it
out and see how well it can work. On the other hand, given the
comments by Cebik and others about the somewhat illusory nature of the
gain advantage of a 5/8-wave, I may just stick with J-poles and
quarterwave ground planes.

--
Dave Platt AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!