Michael Coslo wrote in
:

....
Are you sure, Owen? I read it as the bottom part of the j-pole. I made
a j-pole once, the top was copper, and the support was thick walled
PVC. Obviously that wasn't an active part. The "u" shaped portion was
all above the support leg.

The antenna matched up well with moving the clips back and forth.
Still works okay. I don't look at them as any better or worse thatn a
1/4 wave gp, or vertical dipole, both of which I had before. Did not
measure for feedline radiation, but I have people at home who let me
know of such problems.

Mike,

Even if you put a J Pole on an insulating support, or build it with
flexible wire inside a rigid PVC tube, you can't pretend the the common
mode current path on the outside of the coax feedline does not exist.

If you take extreme care with symmetry of the J Pole, then common mode
current is lower, and with some further decoupling might be very
acceptable. I doubt that common constructions achieve that goal,
enquirers here and elsewhere often complain of a sensitivity of VSWR of
proximity of their body and other things to the feedline... a sign of
feedline common mode current.

If by a vertical dipole, you mean a coaxial dipole (though I note people
are now using 'coaxial dipole' to refer to the Double Bazooka), they are
also notoriously bad for feedline decoupling and need further sleeves
(Bazooka Balun) or radials to effectively decouple the feedline.

The J Pole (and many other end fed antennas such as the Ringo) have a
matching arrangement that makes VSWR very sensitive to the antenna's
local environment. One of the most common questions is "I tuned this up
real good in the workshop, and when I put it in the attic, the VSWR is
3:1".

Sure, the attic is a challenging place to put an antenna, but a
narrowband matching network exacerbates the problem.

Why is the attic so bad? You have to ask is the roof conductive or
variabile permittivity (under any weather) (steel, shingles, concrete
tiles etc), does it have sarking or other conductive insulation products
(such as foil aircell blankets) under it, or on the ceiling, does the
ceiling have foil backed insulation on it, what other conductors are in
the roof space (water, gas, flues, HVAC ducts, structural steel, framing
braces, wiring...etc). What works for one installation might not work
for another because the installer has no idea of the installation.

An experienced eye sees things that an inexperience eye doesn't.

A well executed ground plane antenna is an antenna that a person with
little knowledge and experience can implement with a high level of
confidence that it is reasonably efficient and effective.

I didn't really want to can the J Pole, but to say to the OP if he
REALLY MUST put an antenna in the roof space, he ought consider a folded
dipole with half wave coax balun and dresss the feedline away
horizontally for a wavelength or two for minimal effects on pattern
(though that is not well controlled in the roof space).

BTW, a question that reveals the understanding of a J Pole by its
devotees is "should the braid go to one side or the other". All three
answers (yes, three) are offered in similar proportion, but I have never
seen anyone support them with measurement.

It seems to me that the preferred method from purely an electrical point
of view is none of those three. If you establish the tapping point,
drill a hole in either tube at that point, and another at the bottom of
the U, then pass the coax up through the bottom hole inside the tube
(bonding the shield to that point), exit through the other hole bonding
the coax shield to the tube and connecting the inner conductor to the
opposite side, you have built an integral balun which helps to reduce
common mode current. A further ferrite balun below the tube somewhat
using say #61 would give improved suppression at 144MHz.

Owen