this coaxial collinear antenna was built to test for my own consumption whether or not this antenna method actually works.
the
guiding principle is that we have multiple 1/4 wave elements fed in
series , the radiation from each adds in phase to produce a large vertically polarised omni-directional antenna with substantial gain.
My problem in thinking about this antenna is that the simple theory is all well and good in principle. The practice is more concerning, especially when you consider the phasing relationships.
The
problem lies in the differant wavelengths of the free space outer
element compared with the fact that the inner coaxial element ( it
is still a transmission line !) the wavelength can be 60 to 80 percent
differant. what happens to the phasing model as seen from a distant place, like the station you are trying to talk to ?
The literature says that to achieve the correct phasing you cut the 1/4 wave radiating elements for the phase shift as seen by the the transmission line.
This makes the radiating elements (outer coaxial sheaths) the wrong length ! That is a
non resonant length presenting a high SWR to each coaxial element...all
in series ! Furthermore, the phasing as seen by a remote observer becomes skewed. SWR matters at UHF, it is turning precious watts into British Thermal Units, not radiation !
This antenna was an attempt to seperate hype from reality. My conclusion is that the hype does not live up to the reality ! The
gain measured in practice , at 70cms, was less then just using my 2
meter (146Mhz) J Pole antenna on 70, which is just plain wrong!
So
what was wrong with my collinear ? There are TEN radiating
elements here, including the stick at the top and the brass sleeve decoupling assembly
below. The sleeve was cut to resonate with the outer braid,
accounting for the dilectric constant of the PVC outer insulation. This
is the low loss method. I have seen in the literature the use of
coaxial ferrite sleeves to eliminate coaxial outer currents, but please, these
sleeves eliminate SWR but at the cost of turning to useless heat all
those carefully generated RF watts and precious recieved picowatts. Thats
why I used a classical 1/4 wave decoupling sleeve. A better
decoupling method is to use a quarter wave groundplane or 1/4
wavelength radius circle (not a big ask at 70cms) however I could not solve the mechanical problem of mounting it to its pvc pipe radome.
Humpth! I will not make a coaxial collinear again.
However,
supposing I make my own AIR SPACED coaxial assemblies. The inner and
outer phasing relationships would be always correct and
maybe the hype and reality might line up for once. I am debating
with myself if I use copper or aluminium coaxial elements. I dont know
which would be easier to produce. I can easily braze aluminum but
brazing copper needs something better than a propane torch.
Watch this space !
The
coaxial cable was quality low loss,1/2inch foam di-electric
Belden 50ohm Ethernet Cable. A foam dilectric cable was chosen as its
internal wavelength would be a closer match to the free space
wavelength.
