I was trying to increase the overall performance of the J-pole, in this design. The diagram provided is a more simplified version of the one I did. These are a few of the modifications I came up with. I added a cap on the top end of the PVC. Mounted a so-239 to a split piece of copper tubing, that took the place of the #14 copper wire. And, I also added a short aluminum mast that fit into the lower end of the PVC.  I mounted the antenna to a 10′ antenna mast and a small tripod on the roof. I tried to add some type of a ground plain but everything I did made the antenna perform poorly. After all my efforts the end result was an antenna that out performs the 1/2-wave colinear copper version, with only one exception. The working model is somewhat narrow banded and still requires more experimentation. Some of the elements must be a little long.
Article originally available at http://home.comcast.net/~buck0/5-8thx2j.htm
Hi Simone, shouldn’t the phasing-stub, in the centre of the radiator, be 1/8 not 3/8 wavelength?
73! Tim
No Tim, the phasing stub needs to be 3/8 of a wave length because you need to bring the lower and upper elements into phase, that being 1 wavelength. The bigger issue is the mis-match of the 1/4 wave tuned stub feeding the 5/8 length element. J-poles are high impedance, end-fed 1/2 wave RESONANT antennas, and that’s why you can feed them with a 1/4 wave transmission line. 5/8 antennas are non-resonant antennas that have a significant out-of-phase current/voltage reflection, due to excess capacitance of the design; that’s why you always see a coil at the antenna feed-point. the J-pole transmission line design cannot get around that issue, and therefore, a transmission line matching device is not a good way to feed a 5/8 antenna. This antenna would be much better off being a double 1/2 wave antenna collinear antenna, with a coil phasing the upper and lower radiators. The coil would designed to be less than a 1/2 wave in length, because the inductance of the coil will also add to the phasing angle. This can be designed through experimenting with coils of various total wire lengths, and their calculated inductance, and adding the phase angle produced by that coil at a given frequency and using online inductance phasing angle calculators, so you get a total of 180 degrees of phase difference between the two ends of the coil. Keep in mind velocity factor of wire in the atmosphere, you can discount all wire lengths by about 4%.